diff options
Diffstat (limited to 'old/54889-0.txt')
| -rw-r--r-- | old/54889-0.txt | 31656 |
1 files changed, 0 insertions, 31656 deletions
diff --git a/old/54889-0.txt b/old/54889-0.txt deleted file mode 100644 index 061f39d..0000000 --- a/old/54889-0.txt +++ /dev/null @@ -1,31656 +0,0 @@ -The Project Gutenberg EBook of Observations of a Naturalist in the Pacific -Between 1896 and 1899, v. 2, by H. B. Guppy - -This eBook is for the use of anyone anywhere at no cost and with -almost no restrictions whatsoever. You may copy it, give it away or -re-use it under the terms of the Project Gutenberg License included -with this eBook or online at www.gutenberg.org/license - - -Title: Observations of a Naturalist in the Pacific Between 1896 and 1899, v. 2 - Plant-Dispersal - -Author: H. B. Guppy - -Release Date: June 11, 2017 [EBook #54889] - -Language: English - -Character set encoding: UTF-8 - -*** START OF THIS PROJECT GUTENBERG EBOOK OBSERVATIONS OF A NATURALIST *** - - - - -Produced by Greg Bergquist, Sonya Schermann and the Online -Distributed Proofreading Team at http://www.pgdp.net (This -file was produced from images generously made available -by The Internet Archive/American Libraries.) - - - - - - - - - - Transcriber’s Note - - -When italics were used in the original book, the corresponding text has -been surrounded by _underscores_; bold or larger characters (such as the -large numbers in the index) have been surrounded by =equal signs=. -Superscripted characters are preceded by ^. Mixed fractions have been -displayed with a hyphen between whole number and fraction for clarity. -When a fractional number has been used in a range (i.e. one and a half -to two and a half), the dash separating the two numbers has been spaced -(i.e. 1-1/2 - 2-1/2). - -Some corrections have been made to the printed text. These are listed in -a second transcriber’s note at the end of the text. - - - - - OBSERVATIONS OF A NATURALIST IN THE - - PACIFIC BETWEEN 1896 AND 1899 - - - - -[Illustration] - - [_Frontispiece._ - -[Illustration: Selala (Vanua Levu, Fiji) (1/3).] - -[Illustration: Fruits with seedlings nearly ready to fall from the tree -of Rhizophora mangle (shortest) and Rhizophora mucronata (longest). From -Vanua Levu, Fiji. (1/4 of the true length.)] - -[Illustration: Rhizophora mangle (Vanua Levu, Fiji) (1/4).] - -[Illustration: Rhizophora mucronata (Vanua Levu, Fiji) (1/5).] - - - - - OBSERVATIONS OF - A NATURALIST IN - THE PACIFIC BETWEEN - 1896 AND 1899 - - BY - H. B. GUPPY, M.B., F.R.S.E. - - VOLUME II - _PLANT-DISPERSAL_ - - - - - London - MACMILLAN AND CO., LIMITED - NEW YORK: THE MACMILLAN COMPANY - 1906 - - _All rights reserved_ - - - - - RICHARD CLAY AND SONS, LIMITED, - - BREAD STREET HILL, E.C., AND - BUNGAY, SUFFOLK. - - - - - Dedication - - TO THOSE NUMEROUS PERSONS TO WHOM I WAS INDEBTED FOR - GREAT KINDNESS AND ASSISTANCE DURING MY - SOJOURN IN HAWAII AND FIJI - - - - - PREFACE - - -ALTHOUGH this volume contains a great amount of original material, I am -largely indebted to the labours of my predecessors for its present form; -and a scheme that at first was limited only to my own observations in -the Pacific has gradually extended itself to the general subject of -plant-dispersal. The farther I proceeded in my work the more I realised -that the floras of the Pacific islands are of most interest in their -connections, and that the problems affecting them are problems -concerning the whole plant-world. Deprived of the writings of Seemann, -Hillebrand, Drake del Castillo, and other botanists, several of whom -have lived and died in the midst of their studies of these floras, and -without the aid of the works of Hemsley and Schimper, generalisers who -have mainly cleared the way for the systematic study of -plant-distribution and plant-dispersal, it would not have been possible -for me to accomplish such an undertaking. - -My interest in plant-dispersal dates back to 1884, when, whilst surgeon -of H.M.S. _Lark_, in the Solomon Islands, I made some observations on -the stocking of a coral island with its plants, which were published in -the _Report on the Botany of the “Challenger” Expedition_. In 1888 I -followed up the same line of investigation during a sojourn of three -months on Keeling Atoll, and during a journey along the coasts of West -Java. But realising that as yet I had barely touched the fringe of a -great subject, and that several years of study would be required before -one could venture even to appreciate the nature of the problems involved -and much less to weigh results, I took advantage of the circumstances of -my life to make, between the years 1890 and 1896, a prolonged -investigation of the plants of the British flora, mainly from the -standpoint of dispersal by water. This involved the study of the -seed-drift of ponds and rivers and of the plants supplying it, a study -which brought me into close relation with aquatic and sub-aquatic -plants. This line of investigation led me into contact with many other -aspects of plant-life; and as time went on my field of interest extended -to the plants of dry stations and to the bird as an agent in -plant-dispersal. Only a few of these results have been published, as in -the journals of the Linnean Society and of the Royal Physical Society of -Edinburgh as well as in the pages of _Science Gossip_. They lie for the -most part still within my note-books, and fitly so, since I regarded -such studies chiefly as a preparation for the investigation of the -general question of plant-dispersal. - -When again, in October, 1896, I found myself once more in the Pacific, -the subject was taken up again with zeal; but my larger experience had -only increased my diffidence, and the unknown looked so overwhelming -that I settled down for the next three years content with merely making -experiments and recording observations. Here again the main problem was -attacked through the study of seed-buoyancy, and gradually it led me to -the systematic study of the mangroves and of the beach-plants, whilst my -inland excursions brought me into familiarity with the plants of the -interior. My geological exploration of the island of Vanua Levu, in -Fiji, greatly assisted me by giving a method to my botanical examination -of the island. - -Whilst working out my geological collections in England, in the years -1900-1902, I devoted an hour or two daily to the elaboration of my -botanical notes and to a consideration of the problems concerned. During -a winter in Sicily I took up again the subject of the beach-plants; and -after the publication of the volume on the geology of Vanua Levu I was -able to accomplish a plan, for years in my dreams, of visiting the -eastern shores of the Pacific. During a period of three months from -December, 1903, to March, 1904, I examined the littoral flora of the -west side of South America at various localities between Southern Chile -and Ecuador; and finally completed this investigation by comparing the -shore-plants on the Pacific and Atlantic coasts of the isthmus of -Panama. Returning to England with a fresh collection of data, I passed -many months in elaborating and arranging all my notes, waiting vainly -for a clue to guide me in framing a scheme by which I could bring the -results of many years of work into some connected form. At last I -decided once again to take the floating seed as my clue, and without any -prearranged plan I allowed the work to evolve itself. Now that it is -finished, I can see some obvious defects; but if any other plan had been -adopted I scarcely think that I should have been more successful in -piecing together in a single argument materials resulting from so many -years of research and relating to so many aspects of plant-life. - -Yet the final object of a naturalist would be but a sorry one, if his -aim were only to write a treatise and append his name to it. His -personal faith lies behind all his work; and no one can pursue a long -line of study of the world around him without rising from his task with -some convictions gained and some convictions lost. - -As far as the observation of Nature’s processes at present in operation -can guide us, the world presents itself to us only as a differentiating -world. We can perceive, it is true, a progressive arrangement of types -of organisms from the lowest to the highest, and we can perceive a -development of varieties of the several types; but the only process -evident to our observation is that concerned with the production of -varieties of the type. Nature does not enlighten us as to the mode of -development of the type itself. We can, for instance, detect in actual -operation the process by which the different kinds of bats or the -different kinds of men have been developed; but there is no principle in -Nature evident to our senses that is concerned with type-creation. -Though we can supply it by hypothesis, we cannot discover it in fact. On -the other hand, the evidence of differentiation is abundant on all sides -of us, both in the organic and in the inorganic worlds. The history of -the globe has ever proceeded from the uniform to the complex; and in the -closing chapter of this book an endeavour is made to connect the -differentiation of plant and bird with the differentiation of the -conditions of existence on the earth. But this leaves no room for the -development of new types of organisms; and so far as observation of the -processes of Nature at present working around us can guide us, each type -might well be regarded as eternal. We can never hope to arrive at an -explanation of the progressive development of types by studying the -differentiating process; and since the last is alone cognisable for us, -evolution, as it is usually termed, becomes an article of our faith, and -of faith only. - -In illustration of this argument, let me take the case of the races of -men. We see mankind in our own day illustrating the law of -differentiation all over the globe, as far as physical characters are -concerned. Just as the ornithologist would postulate a generalised type -in tracing the origin of various allied groups of birds, so the -anthropologist, guided by his observation of the changes now offered by -man in different regions, would postulate a generalised original type as -the parent-stock of mankind. Observation of the processes of change now -in operation by no means leads us to infer that such a generalised type -was an anthropoid ape, or even simian in character. In so doing we -should be forming a conclusion not warranted by the observation of -existing agencies of change, and we should be confusing the two distinct -processes of evolution and differentiation, or rather of progressive and -divergent evolution, of which the last alone comes within our field of -cognition. The study of variation can do no more than enable us to -ascertain the mode of development of different kinds, we will say, of -birds or of men. The origin of the type lies outside our observation. -“Given the type, to explain its origin”: this is the problem we can -never solve, and Nature aids us nothing by the study of her ways. On the -other hand, there is the subsidiary problem.... “Given a type, to -explain its varieties” ...; and here Nature’s processes are apparent to -us in a thousand different shapes. - -It might seem that the presumptive evidence connecting man in his origin -with the monkeys is so strong that, supposing his simian descent were -regarded as a crime, a jury would without hesitation pronounce his -guilt; but until some observer of the processes followed by Nature can -bridge over the gap that divides man from the ape, until indeed he can -offer a legitimate illustration of how it is accomplished in similar -cases in our own day, the gap remains. Those who have read the recent -work of Prof. Metchnikoff on the Nature of Man will properly regard his -chapter on the simian origin of man as a brilliant argument advanced by -a most competent authority. Yet he fails to complete his case by -bridging over this gap, and can only appeal to the results of the now -famous researches of De Vries concerning the mutations of the evening -primrose (Œnothera). It is probable, he says, that man owes his origin -to a similar phenomenon (English edition, p. 57). Several objections -could be raised against this illustration from the plant-world, the most -important of them lying in the circumstance that these mutations could -only be urged as instances of the sudden development of new species of -the evening primrose type. They merely illustrate the process of -differentiation from a given type, and by no means represent the process -of progressive evolution from a simian to a man. - -However, look where we may—and this is the great lesson I have learned -from my researches in the Pacific islands—Nature does not present to our -observation any process in operation by which a new type of organism is -produced. The processes involved lie hidden from our view. The channels -by which impressions from the outside world reach us are comparatively -few; and although it seems likely that the future development of man -will be mainly concerned with the acquirement of additional -sense-channels, no newly acquired sense will enable him to be at once an -actor in and a spectator of the great drama presented in the organic -world. That a creature should be able to get at the back of its own -existence, or, in other words, to penetrate the secret of its own -creation, is unthinkable. Outside the limited field of observation that -immediately surrounds us extends the region where reason alone can guide -us, and beyond lies the realm where reason fails and faith begins. - - H. B. GUPPY. - -_November 8th, 1905._ - - - - - LIST OF SOME OF THE PRINCIPAL AUTHORITIES QUOTED IN THIS VOLUME, - WITH AN ENUMERATION OF THE AUTHOR’S BOTANICAL PAPERS - - -BURKILL, I. H., “The Flora of Vavau, one of the Tonga Islands,” Journal - of the Linnean Society, vol. xxxv., Botany, 1901. - -CHEESEMAN, T. F., “The Flora of Rarotonga,” Transactions of the Linnean - Society, 2nd Ser., Botany, vol. vi., part 6, 1903. - -DRAKE DEL CASTILLO, E., “Flore de la Polynésie Française,” Paris, 1893. - - “Remarques sur la Flore de la Polynésie” (Mémoire couronné par - l’Académie des Sciences), Paris, 1890. - -EGGERS, BARON H. VON, “Die Manglares in Ecuador,” Botanisches - Centralblatt, No. 41, 1892. - - “Das Küstengebiet von Ecuador,” Deutsche Geographische Blätter, heft - 4, band 17, Bremen, 1894. - -EKSTAM, O., “Einige blütenbiologische Beobachtungen auf Novaja Semlja,” - Tromso Museums Aarshefter, 18, 1895. - - “Einige blütenbiologische Beobachtungen auf Spitzbergen,” Tromso - Museums Aarshefter, 20, 1897. - -GUPPY, H. B., “The Dispersal of Plants as illustrated by the Flora of - Keeling Atoll,” Journal of the Victoria Institute, London, 1889. - - “The Polynesians and their Plant-Names,” Journal of the Victoria - Institute, London, 1896. - - “The River Thames as an Agent in Plant-Dispersal,” Journal of the - Linnean Society, Botany, vol. xxix., 1891-93. - - “River-Temperature,” part iii., Proceedings of the Royal Physical - Society of Edinburgh, 1896. (The first two parts deal principally with - the temperature of ponds and rivers, whilst in the last part the - thermal conditions are discussed especially in connection with the - life of aquatic plants.) - - “On the Postponement of the Germination of the Seeds of Aquatic - Plants,” Proceedings of the Royal Physical Society of Edinburgh, 1897. - - “On the Temperature of Springs as especially illustrated by the - Wandle and other Tributaries of the Thames.” (This paper, of which I - have no copy, was published in the Journal of the Royal Meteorological - Society, about 1895. It throws light on the thermal conditions of - plants in springs.) - - “Water-Plants and their Ways,” _Science Gossip_, Sept., Oct., Nov., - 1894. (The various modes of dispersal of land as well as water plants - are here dealt with, their thermal conditions are discussed, and in - the November number are given the results of four years’ observations - on the life-history and life-conditions of Ceratophyllum demersum.) - - “Caddis-Worms and Duckweed,” _Science Gossip_, March, 1895. (A short - note.) - - “Stations of Plants and Buoyancy of Seeds,” _Science Gossip_, April - and May, 1895. - - “Irregularity of some Cotyledons,” _Science Gossip_, September, - 1895. - - “Plants of the Black Pond, Oxshott,” _Science Gossip_, October, - 1895. - - “On the Habits of Lemna minor, L. gibba, and L. polyrrhiza,” Journal - of the Linnean Society, Botany, vol. xxx. (This paper contains the - results of three years’ systematic observations of these plants.) - - “The Distribution of Aquatic Plants and Animals,” _The Scottish - Geographical Magazine_, January, 1893. - -HEMSLEY, W. B., “Report on the Botany of the _Challenger_ Expedition,” - 1885. - - “The Flora of the Tonga Islands,” Journal of the Linnean Society, - Botany, vol. xxx. - -HILLEBRAND, W., “Flora of the Hawaiian Islands,” Heidelberg, 1888. - -HORNE, J., “A Year in Fiji,” London, 1881. - -KOLPIN RAVN, F., “Om Flydeevnen hos Froene af vore Vand-og Sumpplanter,” - Botanisk Tidsskrift, 19 bind., 2 hefte, Kjobenhavn, 1894 (“On the - Floating Capacity of the Seeds of Aquatic and Marsh Plants”). (A - _résumé_ in French is appended to the paper.) - -MARTINS, CH., “Expériences sur la Persistance de la Vitalité des Graines - flottant à la Surface de la Mer,” Bull. Soc. Botanique de France tome - iv., 1857. - -NADEAUD, J., “Enumération des Plantes indigènes de l’Ile de Tahiti,” - Paris, 1873. - -PENZIG, O., “Die Fortschritte der Flora des Krakatau,” Annales du Jardin - Botanique de Buitenzorg, 2 ser., tome 3, Leide, 1902. - -PERKINS, R. C. L., “Fauna Hawaiiensis,” vol. i., part iv. (Vertebrata) - Cambridge University Press, 1903. - -REINECKE, F., “Die Flora der Samoa-Inseln,” Engler’s “Botanische - Jahrbücher,” band xxv., heft v., Leipzig, 1898. - -SCHIMPER, A. F. W., “Die indo-malayische Strandflora,” Jena, 1891. - -SEEMANN, B., “Flora Vitiensis,” London, 1865-73. - -SERNANDER, R., “Den Scandinaviska Vegetationens Spridnings-biologi,” - Upsala, 1901. - -THURET, G., “Expériences sur des Graines de diverses Espèces plongées - dans de l’eau de Mer,” Archives des Sciences (Phys. et Nat.) de la - Bibliothèque Universelle, tome 47, Geneva, 1873. - -TREUB, M., “Notice sur la nouvelle Flore de Krakatau,” Annales du Jardin - Botanique de Buitenzorg, 1888. - - -_Note._—Amongst the works quoted which are not specially particularised - in the text are Scott Elliot’s “Nature Studies,” 1902, and Beal’s - “Seed Dispersal,” Boston, 1900. - - - - - CONTENTS - - - PREFACE _Pages_ - vii—x - - LIST OF SOME OF THE PRINCIPAL AUTHORITIES QUOTED, _Pages_ - WITH AN ENUMERATION OF THE AUTHOR’S BOTANICAL xiii—xv - PAPERS - - LIST OF ILLUSTRATIONS _Page_ - xxvii - - ADDITIONS AND CORRECTIONS _Page_ - xxviii - - - CHAPTER I - - INTRODUCTION - -The study of insular floras.—Their investigation in this work from the - standpoint of dispersal.—The significance of plant-distribution in the - Pacific.—The problems connected with the mountain-flora of Hawaii.—The - persistence of dispersing agencies at the coast, their partial - suspension on the mountain-top, their more or less complete suspension - in the forest, and the effect on the endemic character of plants.—The - connection between the endemism of birds and plants.—The relative - antiquity of plants of the coast, forest, and mountain-top.—The - genetic relation between coast and inland species of the same - genus.—The ethics of plant-dispersal.—Evolution takes no heed of modes - of dispersal.—The seed-stage is the price of Adaptation. - - _Pages_ 1-11 - - - CHAPTER II - - THE FLORAS OF THE PACIFIC ISLANDS FROM THE STANDPOINT OF DISPERSAL - BY CURRENTS - -The initial experiment.—The proportion of littoral plants.—The two great - principles of buoyancy.—The investigations of Professor Schimper.—The - investigations of the author.—The great sorting process of the - ages.—Preliminary results of the inquiry into the buoyancy of seeds - and fruits. - - _Pages_ 12-22 - - - CHAPTER III - - THE LESSON OF THE BRITISH FLORA - -Results of observations on the buoyancy of over 300 British plants.—The - small proportion of plants with buoyant seeds or seedvessels.—Their - station by the water-side.—The great sifting experiment of the - ages.—Summary. - - _Pages_ 23-30 - - - CHAPTER IV - - THE LESSON OF THE BRITISH FLORA (_continued_) - -The choice of station of the water-side plant possessing buoyant seeds - or seedvessels.—Determined by its fitness or unfitness for living in - physiologically dry stations.—In the internal organisation of a plant - lies the first determining influence of station.—The grouping of the - British strand-plants.—Whilst the Xerophyte with buoyant seed or fruit - finds its station at the coast, the Hygrophyte similarly endowed makes - its home at the river or pond side.—The grouping of the plants of the - river and the pond.—Summary. - - _Pages_ 31-39 - - - CHAPTER V - - THE FIJIAN STRAND-FLORA - -The inland extension of the beach-plants.—The grouping of the - coast-plants.—Their modes of dispersal.—The zone of change.—Summary. - - _Pages_ 40-46 - - - CHAPTER VI - - THE TAHITIAN STRAND-FLORA - - (_From materials supplied mainly by the work of Drake del Castillo_) - -Lacks the mangroves and their associated plants.—Possesses mainly the - plants of the coral beach.—Predominant agency of the currents.—Inland - extension of shore-plants.—Summary - - _Pages_ 47-50 - - - CHAPTER VII - - THE HAWAIIAN STRAND-FLORA - -Its poverty.—Its negative features.—Their explanation.—The subordinate - part taken by the currents.—The Oregon drift.—The inland extension of - the beach-plants.—Summary - - _Pages_ 51-60 - - - CHAPTER VIII - - THE LITTORAL PLANTS AND THE CURRENTS OF THE PACIFIC - -The working value of the currents as plant-dispersers.—The relation - between the currents and the distribution of shore-plants.—The clue - afforded by the American plants.—Two regions of tropical shore-plants, - the American and the Asiatic.—America, the home of the cosmopolitan - tropical shore-plants that are dispersed by the currents.—Hawaii and - the currents.—Summary - - _Pages_ 61-75 - - - CHAPTER IX - - THE GERMINATION OF FLOATING SEEDS - -Germination in the floating seed-drift of tropical estuaries.—A strain - of vivipary.—Abortive germination of seeds in warm seas.—A barrier to - plant dispersal.—The borderland of vivipary.—Summary - - _Pages_ 76-87 - - - CHAPTER X - - THE RELATION OF THE BUOYANCY OF SEEDS AND SEEDVESSELS TO THE DENSITY - OF SEA-WATER - -The general principles concerned.—The subject assumes a statistical - character.—Seeds and seedvessels are as a rule either much heavier - than sea-water or much lighter than fresh water.—The present littoral - plants with buoyant seeds or seedvessels could be equally well - dispersed by currents in oceans of fresh water.—Seed-buoyancy has no - relation either in the present or in the past to the density of the - sea.—Though an accidental attribute, the specific weight of seeds has - had a profound influence on plant-distribution.—Summary - - _Pages_ 88-98 - - - CHAPTER XI - - ADAPTATION AND MEANS OF DISPERSAL - -Nature has never concerned herself directly with providing means of - dispersal.—Fleshy fruits not made to be eaten.—Nor “sticky” seeds to - adhere to plumage.—Nor prickly fruits to entangle themselves in fur - and feathers.—The dispersal of seeds a blind result of the struggle - between the intruding Evolutionary power and the controlling influence - of Adaptation. - - _Pages_ 99-103 - - - CHAPTER XII - - THE CAUSES OF THE BUOYANCY OF SEEDS AND FRUITS OF LITTORAL PLANTS, - WITH ESPECIAL REFERENCE TO THOSE OF THE PACIFIC ISLANDS - -The classification of buoyant seeds and fruits.—The first group, where - the cavity of the seed or seedvessel is incompletely filled.—The - second group, where the kernel is buoyant.—The third group, where - there is air-bearing tissue in the seed-tests or fruit-coats.—The - buoyant seeds and seedvessels of the littoral plants of the British - flora.—Summary - - _Pages_ 104-118 - - - CHAPTER XIII - - ADAPTATION AND SEED-BUOYANCY - -The question of the operation of Natural Selection.—Are there two - principles at work?—The presence of buoyant tissue in the seed-tests - and fruit-coats of inland plants, both wild and cultivated.—Useless - buoyancy.—The buoyancy of seeds and fruits is not concerned with - Adaptation.—Summary. - - _Pages_ 119-129 - - - CHAPTER XIV - - THE RELATION BETWEEN LITTORAL AND INLAND PLANTS - -Professor Schimper’s views.—Great antiquity of the - mangrove-formation.—Problem mainly concerned with the - derivation of inland from littoral plants.—Grouping of - the genera possessing both coast and inland - species.—Scævola.—Morinda.—Calophyllum.—Colubrina.—Tacca.—Vigna.—Premna - - _Pages_ 130-139 - - - CHAPTER XV - - THE RELATION BETWEEN LITTORAL AND INLAND PLANTS (_continued_) - -Inland species of a genus developed from littoral species originally - brought by the currents but no longer existing in the - group.—Illustrated by the Leguminous genera Erythrina, Canavalia, - Mezoneuron, and Sophora, and by the Apocynaceous genus Ochrosia.—The - Hawaiian difficulty. - - _Pages_ 140-154 - - - CHAPTER XVI - - THE RELATION BETWEEN LITTORAL AND INLAND PLANTS (_continued_) - -The Fijian difficulty.—Inland species of a genus possessing - fruits not known to have any means of dispersal through - agencies now at work in the Pacific.—Pandanus.—Its remarkable - distribution in oceanic groups.—To be attributed perhaps - to extinct Columbæ or extinct Struthious - birds.—Barringtonia.—Guettarda.—Eugenia.—Drymispermum.—Acacia - laurifolia.—Conclusions to be drawn from the discussion.—Summary of - Chapters XIV., XV., XVI. - - _Pages_ 155-169 - - - CHAPTER XVII - - THE STORIES OF AFZELIA BIJUGA, ENTADA SCANDENS, AND CÆSALPINIA - BONDUCELLA - -Afzelia bijuga.—The African home of the genus.—The double station of - Afzelia bijuga, inland and at the coast.—The nature of the buoyancy - of its seeds.—Summary relating to Afzelia bijuga.—Entada - scandens.—Its station and distribution.—Darwin’s opinion of the - plant.—The dispersal of its seeds by the currents.—Summary relating - to the plant.—Cæsalpinia bonducella and C. bonduc.—Their station and - distribution.—Their characters in various Pacific groups.—The - parents of inland species.—Their dispersal by the currents.—The - germination of their seeds.—A dream of vivipary.—The causes of the - seed-buoyancy.—Summary of results - - _Pages_ 170-197 - - - CHAPTER XVIII - - THE ENIGMAS OF THE LEGUMINOSÆ OF THE PACIFIC ISLANDS - -Leguminosæ predominate in tropical littoral floras.—The anomalies of - their distribution in the Pacific islands.—They conform to no one rule - of dispersal or of distribution.—Strangers to their stations.—The - American home of most of the Leguminous littoral plants.—Summary - - _Pages_ 198-203 - - - CHAPTER XIX - - THE INLAND PLANTS OF THE PACIFIC ISLANDS - - PRELIMINARY COMPARISON OF THE PHYSICAL CONDITIONS OF HAWAII, - FIJI, AND TAHITI - -Introductory remarks.—The tranquil working of the winds and currents - contrasted with the revolutionary influence of the bird.—The Hawaiian, - Fijian, and Tahitian groups.—Their surface-areas and elevations.—Their - climates.—The mountain climate of Hawaii.—The rainfall of the three - groups.—Summary - - _Pages_ 204-219 - - - CHAPTER XX - - THE ERAS IN THE FLORAL HISTORY OF THE PACIFIC ISLANDS - - THE AGE OF FERNS - -The eras in the plant-stocking.—The age of ferns and lycopods.—The - relative proportion of vascular cryptogams in Hawaii, Fiji, and - Tahiti.—The large number of peculiar species in Hawaii.—The mountain - ferns of Hawaii.—The origin of peculiar species.—Dr. Hillebrand’s - views.—Their origin connected not with greater variety of climate in - Hawaii, but with isolation.—Summary - - _Pages_ 220-230 - - - CHAPTER XXI - - THE ERAS OF THE FLOWERING PLANTS - - THE ERA OF THE ENDEMIC GENERA - - THE AGE OF COMPOSITÆ - -The islands of the tropical Pacific as the homes of new genera and new - species.—The significance of a large endemic element.—Synopsis of the - eras.—The era of endemic genera.—The endemic genera of - Compositæ.—Their affinities and mode of dispersal.—The mystery of the - suspension of the dispersing agencies.—Mr. Bentham’s views.—The - remnant of an ancient Composite flora in the tropical Pacific.—The - dispersion of the Compositæ antedates the emergence of the - island-groups of the Fijian region at the close of the Tertiary - period.—Summary - - _Pages_ 231-249 - - - CHAPTER XXII - - THE ERA OF THE ENDEMIC GENERA (_continued_) - - THE AGE OF THE TREE-LOBELIAS - -The distribution of the arborescent Lobeliaceæ.—On the upper flanks of - Ruwenzori.—The Lobeliaceæ of the Hawaiian Islands.—The Lobeliaceæ of - the Tahitian or East Polynesian region.—The capacities for - dispersal.—The explanation of the absence of the early Lobeliaceæ from - West Polynesia.—The other Hawaiian endemic genera.—The Fijian endemic - genera.—Summary - - _Pages_ 250-267 - - - CHAPTER XXIII - - THE ERA OF THE NON-ENDEMIC GENERA OF FLOWERING PLANTS - - THE MOUNTAIN FLORAS OF THE PACIFIC ISLANDS AS ILLUSTRATED BY - THE NON-ENDEMIC GENERA - -The mountain-flora of Hawaii.—A third of it derived from high southern - latitudes.—An American element.—Compared with Tahiti and - Fiji.—Capacities for dispersal of the genera possessing only endemic - species.—Acæna, Lagenophora, Plantago, Artemisia, Silene, Vaccinium, - &c.—Capacities for dispersal of the genera possessing non-endemic - species.—Cyathodes, Santalum, Carex, Rhynchospora.—Fragaria chilensis, - Drosera longifolia, Nertera depressa, Luzula campestris.—Summary. - - _Pages_ 268-288 - - - CHAPTER XXIV - - THE ERA OF THE NON-ENDEMIC GENERA OF FLOWERING PLANTS (_continued_) - - THE MOUNTAIN-FLORAS OF THE TAHITIAN AND FIJIAN REGIONS - -The mountain-flora of the Tahitian region, as illustrated by the - non-endemic genera.—Derived chiefly from high southern - latitudes.—Weinmannia, Coprosma, Vaccinium, Astelia, Coriaria, - Cyathodes, Nertera depressa, Luzula campestris.—The mountain flora of - Rarotonga.—The mountain-flora of the Fijian region, as illustrated by - the non-endemic genera.—Weinmannia, Lagenophora, Coprosma, Astelia, - Vaccinium, Nertera depressa.—The Fijian Coniferæ.—Dammara, Podocarpus, - Dacrydium.—Not belonging to the present era of dispersal.—The age of - dispersal of the Coniferæ in the Pacific.—Earlier than the age of - Compositæ and Lobeliaceæ.—The first in the Mesozoic period.—The last - in the Tertiary period.—Summary - - _Pages_ 289-306 - - - CHAPTER XXV - - THE ERA OF THE NON-ENDEMIC GENERA OF FLOWERING PLANTS (_continued_) - -THE AGE OF THE MALAYAN PLANTS AS REPRESENTED IN THE LOW-LEVEL FLORA OF - HAWAII AND IN THE BULK OF THE FLORAS OF THE FIJIAN AND TAHITIAN - REGIONS - - THE AGE OF WIDE DISPERSAL OVER THE TROPICAL PACIFIC - -The widely dispersed genera which possess only peculiar species in - Hawaii.—Pittosporum.—Reynoldsia.—Gardenia.—Psychotria.—Cyrtandra.—Freycinetia.—Sapindus.—Phyllanthus.—Pritchardia.—Summary. - - _Pages_ 307-332 - - - CHAPTER XXVI - - THE ERA OF THE NON-ENDEMIC GENERA OF FLOWERING PLANTS (_continued_) - - THE AGE OF MALAYAN PLANTS (_continued_) - - THE AGE OF WIDE DISPERSAL OVER THE TROPICAL PACIFIC (_continued_) - -The widely dispersed genera that are as a rule not entirely - represented by endemic species in any - archipelago.—Elæocarpus.—Dodonæa.—Metrosideros.—Alyxia.—Alphitonia.—Pisonia.—Wikstrœmia.—Peperomia.—Eugenia.—Gossypium.—The - last stage in the general dispersal of plants of the Malayan - era as illustrated by the widely-dispersed genera having - as a rule no peculiar - species.—Rhus.—Osteomeles.—Plectronia.—Boerhaavia.—Polygonum.—Pipturus.—Dianella.—Summary. - - _Pages_ 333-358 - - - CHAPTER XXVII - - THE ERA OF THE NON-ENDEMIC GENERA OF FLOWERING PLANTS (_continued_) - - THE AGE OF MALAYAN PLANTS (_continued_) - - THE AGE OF LOCAL DISPERSAL - -Synopsis of the Chapter given on page 359 - - _Pages_ 359-410 - - - CHAPTER XXVIII - - THE POLYNESIAN AND HIS PLANTS - -Identity of the problems presented by the indigenous plants and the - peoples of the Pacific islands.—The food-plants of the Polynesians and - the pre-Polynesians.—Their weeds.—The aboriginal weeds.—The white - man’s weeds.—Weeds follow the cultivator but are distributed by - birds.—The general dispersion of weeds antedates the appearance of the - Polynesian in the Pacific.—Weeds of little value to the - ethnologist.—Aleurites moluccana.—Inocarpus edulis, Gyrocarpus - Jacquini, Serianthes myriadenia, Leucæna Forsteri, Mussænda frondosa, - Luffa insularum.—Summary - - _Pages_ 411-428 - - - CHAPTER XXIX - - BEACH AND RIVER DRIFT - -In the south of England.—On the coast of Scandinavia.—In the - Mediterranean.—Southern Chile.—Very little effective dispersal by - currents in temperate latitudes.—Cakile maritima.—In tropical - regions.—River drift.—River and beach drift of Fiji.—Musa Ensete.—The - coco-nut.—River and beach drift of Hawaii.—Comparison of the beach - drift of the Old and New Worlds.—Summary - - _Pages_ 429-439 - - - CHAPTER XXX - - THE VIVIPAROUS MANGROVES OF FIJI - - RHIZOPHORA AND BRUGUIERA - -_Rhizophora._—Represented by Rhizophora mucronata, Rhizophora - mangle, and the Selala, a seedless intermediate form.—Their - mode of association and characters.—The relation of the - Selala.—Polyembryony.—The history of the plant between the - fertilisation of the ovule and the detachment of the seedling.—Absence - of a rest period.—Mode of detachment of the seedling.—Capacity - for dispersal by the currents.—_Bruguiera._—The mode of - dispersal.—Peculiar method of fertilisation.—Length of period between - fertilisation and the detachment of the seedling.—Mode of detachment - of the seedling.—Summary - - _Pages_ 440-467 - - - CHAPTER XXXI - - A CHAPTER ON VIVIPARY - -The significance of vivipary.—The scale of germinative capacity.—A lost - habit with many inland plants.—The views of Goebel.–-The shrinking in - the course of ages of tropical swamp areas.—The variation in the - structures concerned with vivipary.—Abnormal vivipary.—Summary. - - _Pages_ 468-473 - - - CHAPTER XXXII - - THE WEST COAST OF SOUTH AMERICA - -The littoral floras of the West Coast of South America.—The Convolvulus - soldanella zone of Southern Chile.—The plantless or desert zone of - Northern Chile.—The Sesuvium zone of Peru.—The Mangrove zone of - Ecuador and Colombia.—The two varieties of Rhizophora mangle, the - “mangle chico” and the “mangle grande.”—The floating vegetable drift - of the Guayaquil River.—The Humboldt current and the climate of the - West Coast of South America.—The advance northward of the arid - climatic conditions of the Peruvian sea-border.—The retreat of the - mangroves.—Evidence of ancient coral reefs on the coast of Peru.—The - shore plants and stranded seed-drift of the Panama Isthmus.—Summary. - - _Pages_ 474-501 - - - CHAPTER XXXIII - - SEED-DISPERSAL AND GEOLOGICAL TIME - -The shifting of the source of Polynesian plants from the New to the Old - World.—The floral history of Polynesia stated in terms of geological - time.—The suspension of the agencies of dispersal in later - periods.—Parallel differentiation in the course of ages of climate, - bird, and plant.—New Zealand.—Insects and bats as agents in - plant-dispersal.—The effective agency of sea-birds in other - regions.—The observations of Ekstam.—The Spitzbergen controversy.—The - efficacy of ducks as distributors of aquatic plants.—Summary - - _Pages_ 502-514 - - - CHAPTER XXXIV - - GENERAL ARGUMENT AND CONCLUSION _Pages_ 515-523 - - - APPENDIX _Pages_ 525-605 - - - - - LIST OF ILLUSTRATIONS - - PLATE. - - The Fijian species of Rhizophora _Frontispiece._ - - - FIGURES. - - TO FACE PAGE - - Diagrams illustrating some of the causes of 111 - seed-buoyancy - - Figures illustrating the development of the seed 452-453 - and the germinating process of Rhizophora and - Bruguiera - - Diagrams illustrating the structure of the growing 574 - seeds of Barringtonia - - Diagram illustrating the prevailing 585 - cloud-formations of Mauna Loa - - - MAPS. - - Oceania 12 - - The Ocean Currents 61 - - Trade routes of the Pacific Ocean (intended to 66 - illustrate the distances traversed by floating - seeds in that ocean) - - The West Coast of South America 474 - - Rough plan of the Gulf of Guayaquil 484 - - - - - ADDITIONS AND CORRECTIONS - - - Page 5 and subsequent pages. _For_ Ipomea _read_ Ipomœa. - - Page 68. _For_ Hippomanes _read_ Hippomane. - - Page 68. _For_ Conocarpus erecta _read_ Conocarpus erectus. - - Page 122. Sir W. Buller includes the fruits of the Puriri tree (Vitex - littoralis, according to Kirk) amongst the food of the New Zealand - fruit-pigeons. - - Page 177. _For_ Entata, in the head-line, _read_ Entada. - - Page 266. The fruits of Oncocarpus vitiensis have been found in the - crop of a Fijian fruit-pigeon (Carpophaga latrans). _See_ Hemsley’s - _Bot. Chall. Exped._, Introd., 46, and iv. 308; also Newton’s - _Dictionary of Birds_, p. 724. - - Page 368. Sernander (p. 185) observes that the fruits of Naias marina - have little or no floating power. - - Page 416. For the first eight lines read as follows:—“Of these, 22 - occur in Continental regions on both sides of the Pacific; 12 are - found in the Old World alone; one is peculiarly American, and two are - confined to the Australian and Polynesian regions. A few of these can - be regarded as exclusively American in their origin, though the bulk - of them hail evidently in the first place from the Old World. But - from the circumstance that all or most of the other species of the - genus concerned are confined to America, it may legitimately be - inferred that Waltheria americana, Ageratum conyzoides, and Physalis - angulata are American-born species. Teucrium inflatum is a peculiar - instance of an American weed collected in Polynesia before apparently - it had been recorded from the Old World.” - - Page 438. _For_ Conocarpus erecta _read_ Conocarpus erectus. - - Page 417. Add after Cardiospermum halicacabum.... “Its seeds, as my - experiments show, possess little or no capacity for dispersal by - currents, since they sink at once or within a few days, even after - drying for months.” - - Page 455. Omit the reference to figure 6 in the centre of the page. - - Page 498. _For_ Hippomanes _read_ Hippomane. - - Page 508. Amongst my Solomon Island collections identified at Kew were - the fruits of a species of Litsea from the crop of a fruit-pigeon - (Hemsley’s _Bot. Chall. Exped._, IV. 295.) - - Page 533. _For_ Commelyne _read_ Commelina. - - Page 539. At foot of page, _for_ Thames sea-drift, _read_ Thames - seed-drift. - - Page 581. _For_ Crambe maritimum _read_ Crambe maritima. - - Page 618. Under Mascarene Islands add Myoporum to the plants linking - them to the Pacific Islands. - ------------------------------------------------------------------------- - - - - - OBSERVATIONS OF A NATURALIST - IN THE PACIFIC - - - - - CHAPTER I - - INTRODUCTION - -The study of insular floras.—Their investigation in this work from the - standpoint of dispersal.—The significance of plant-distribution in the - Pacific.—The problems connected with the mountain-flora of Hawaii.—The - persistence of dispersing agencies at the coast, their partial - suspension on the mountain-top, their more or less complete suspension - in the forest, and the effect on the endemic character of plants.—The - connection between the endemism of birds and plants.—The relative - antiquity of plants of the coast, forest, and mountain-top.—The - genetic relation between coast and inland species of the same - genus.—The ethics of plant-dispersal.—Evolution takes no heed of modes - of dispersal.—The seed-stage is the price of Adaptation. - - -TO proceed from the general to the special is the only method of dealing -with insular floras. A broad and comprehensive grasp of -plant-distribution, such as is only acquired by a life-time of research -aided by travel and the handling of large collections, is a necessary -foundation for the study; yet in the nature of things such -qualifications can be possessed by but a few. To direct an inquiry in -the opposite direction, and endeavour to attack the problem of -continental floras through the insular floras would result merely in the -investigation of a few of the many questions connected with -plant-distribution. - -The panoramic sketch of the surveyor on the mountain-top aids him in a -thousand ways when after months of tedious labour he plots the details -in his chart. Without such a panoramic view of the plant-world in his -mind’s eye, an observer like myself can only look for guidance to the -writings of those who have generalised on the foundations of a far -broader experience, such as those of Bentham, De Candolle, Gray, Hooker, -Schimper, and others. - -It would be quite possible for a botanist possessing a profound general -acquaintance with the plant-world to dispense altogether with actual -observation and experiment on modes of dispersal. It would be quite -possible for him to arrive at conclusions, which, even if they did not -always come into line with results of observation and experiment, we -should be compelled to prefer. It is only from his more elevated -position that a general can follow the course of a battle; whilst the -private with his experience confined to a limited area of the field of -conflict may form the most erroneous ideas of the progress of the fight. -So it is with observers whose employment it is to struggle with the -details and secondary principles of plant-distribution, and so it is -with the generaliser who has already roughly mapped out the principal -features of the main problem. - -When Mr. Bentham in 1869, remarking on the paucity of species common to -tropical Asia and America, characterised them either as plants wholly or -partially maritime and spread by the currents, or as weeds dispersed by -cultivation over the warm regions of the globe, he mentioned amongst the -plants in the former category, Gyrocarpus jacquini. This tree presents -one of the mysteries connected with the Pacific islands; and I don’t -imagine that this eminent botanist could have known anything except -inferentially as regards the mode of dispersal of its fruits. Yet -experiment shows how well founded the inference was, whilst behind it -lay a life-time of botanical research. - -The author thus approaches the subject of the floras of the Pacific -islands rather as a plotter of detail than as a delineator of great -designs. However much we may study the means of dispersal, we have -behind them the great facts of distribution, serving like the main -stations of a trigonometrical survey, and with these we have to make our -lesser facts and observations square. One is conscious all the time that -much of what seems new in one’s researches has already been foreseen by -the generaliser, and that one can do little else than assist in -confirming some of his results. This is all that I can lay claim to in -this work. - -The floras of the islands and coasts of the tropical Pacific are here -regarded entirely from the standpoint of plant-dispersal. The fruits and -seeds rather than the flowers have been the subject of my -investigations; and although there is much to please the eye in the -flora of a Pacific island, it was always with a sense of disappointment -that I turned away from some pretty flowering plant that failed to -present me with its seed. Amongst the wonders of the plant-world rank -the Tree Lobelias of the Hawaiian Islands; yet their greatest charm to -me lay not so much in their giant-flowers and their arborescent habit as -in the mystery surrounding the home of their birth and their mode of -arrival in these islands. When I first stood under the shade of the -lofty Dammara vitiensis, the Kauri Pine of Fiji, all my interest lay in -its cones lying on the ground; and I remember how eagerly I handled my -first specimen, and how anxiously I watched its behaviour when -experimenting on its capacity for different modes of transport. When a -strange plant presented itself on a beach, my first care was to -ascertain the fitness of its fruits or seeds for transport by the -currents; and all inland plants with fruits likely to attract -frugivorous birds were at once invested with a special interest for me. - -The mangrove swamps were always great places of interest, and months of -my sojourn in the Pacific must have been passed in exploring their -creeks and in examining their vegetation. Botanists usually avoid these -regions; but the observation of the germination of the Rhizophora fruits -on the trees and the inquiries connected with their methods of -distribution over the oceans were pursuits so engrossing that I ignored -the numerous discomforts connected with the exploration of these gloomy -regions. The magnificent mangrove forests of the Ecuador coast of the -Pacific will live longest in my memory, though the risks were -considerably greater and the discomfort of existence extreme. But the -mangrove swamps present us with glimpses into the conditions of plant -life during the warmer epochs of the earth’s history, when perhaps the -seed-stage was largely dispensed with, whilst an atmosphere, laden with -moisture and screening off much of the sun’s light, enveloped most of -the circumference of the globe. - -The plant world viewed only from the standpoint of dispersal may lack -much that is pleasing to the eye, though it abounds with small and great -problems fascinating to the reason. Matters of great moment are here -involved, and in the case of the Pacific islands they concern not only -the source of the oceanic floras, but the story of the islands -themselves; whilst behind these there rise up questions of yet deeper -import, questions that are bound up with the beginnings of genera and -species, and with other mysteries of life on the earth. The distribution -of plants presents something more than a problem of means of dispersal, -or a problem of station, or a problem of plant migration connected with -climatic changes. It is something a great deal more than all three, -since it is indissolubly connected with a past, of which unfortunately -we know very little. Let us take it to be a question of means of -dispersal, and then in imagination transporting ourselves to the -Scandinavian coast, let us gather up the stranded West Indian beans of -Cæsalpinia, Mucuna, and Entada, that have been drifted there for ages by -the Gulf Stream, and lie in some cases semifossilised in the adjacent -peat-bog. Was ever dispersal so utterly purposeless as this? Yet here -lies a principle of plant-dispersal that is fundamental. We see it in -the thistle-seed floating seaward in the wind. Nature never intended its -pappus for such an end. It was formed for quite another purpose, yet it -aids largely the dispersion of the plant. What can be more significant -than that? - -Or let us take it to be a matter of station. Given time and the -recurrence of the same conditions, with others I once imagined that we -could explain most things in plant-distribution, whether of plants at -the coast or of plants inland, whether of plants of the alpine peaks or -of plants of the plains, or of plants of the river or of the pond. Time, -it was held, had long since discounted the means of dispersal, and -distribution became merely an affair of station. But the supplanting of -many indigenous species of a flora by introduced species is a common -story in the plant-world; and such a view needs no further discussion -here. Nor is distribution only concerned with plant-migration. Any -theory of the origin of alpine floras on tropical mountains will have to -explain the presence of the temperate genera, Geranium and Sanicula, not -alone on the summits of the mountains of Equatorial Africa and -Madagascar, but on the uplands of Hawaii in mid-Pacific, where also are -found Ranunculus, Vaccinium, Fragaria chilensis (the Chilian -strawberry), and Drosera longifolia. - -Taking genera of different stations each in their turn, and following up -the clues thus afforded, it would be possible to find support for all -the reputable views relating to plant-distribution. The wide range of -aquatic plants under conditions that completely change the character of -the terrestrial vegetation, such, for instance, as Myriophyllum and -Ceratophyllum, might be plausibly attributed to the relative uniformity -of the conditions of aquatic life both in time as well as space. The -occurrence of Vaccinium on mountain-tops over most of the world, even on -the highlands of Samoa, Tahiti, and Hawaii in the Pacific Ocean, would -be rightly regarded as evidence of active dispersal of the seeds through -the agency of birds from one mountain-summit to another, whether in -mid-ocean or in the centre of a continent. The prevalence of the same -beach-plants over most of the globe in the same climatic zones would -point unmistakably to the predominant agency of currents. But with many -plant-genera, some of which range the world, whilst others again may be -restricted to a single group of islands in the Pacific, there is often -no question either of means of dispersal, or of station, or of -plant-migration, and problems of a very different nature are opened up. - -When we leave the beach and the mountain-top, the river and the pond, -all the troubles of distribution begin; and since but a small proportion -of plants in a typical flora belong to these stations, it follows that -difficulties will dog our steps with the large majority of the plants. -The agencies of dispersal now working around us, the current, the wind, -the insect, the bird, and the bat, will explain many of the features of -littoral and alpine floras and of the vegetation of ponds and rivers. -Here we have in so many cases wide-ranging genera with the means of -dispersal ready to hand. We can connect the wide range of Vaccinium with -the wide range of birds of the grouse and other families that feed on -the berries. We can associate the great areas of aquatic or sub-aquatic -genera, like Potamogeton and Sparganium, with the migratory habits of -the ducks in the stomachs of which we find their seeds. We can connect -the great ranges of beach plants like Ipomœa pes capræ in the tropics, -and Convolvulus soldanella in the temperate regions with the currents, -and the almost cosmopolitan range of many ferns and lycopods with the -winds and other agencies. - -When, however, we enter the forests we find genera that are often much -more restricted in their areas, and species that are yet more limited in -their range. There is very little dispersal going on here. The birds are -strange. Their distribution is usually very local. They look lazily down -at us from the branches, as they disgorge the seeds and stones of the -fruits they have eaten, which cover the ground around. We can almost -fancy that they say:—“Our work is done. We rest from the toil of our -ancestors. They carried seeds to far-distant Hawaii, Tahiti, and Savaii. -Our work is done.” And as we walk through those noiseless forests, where -the machinery of species-making is ever in silent motion, we become -aware that we are treading one of Nature’s great workshops for the -manufacture of species and genera. Outside the forest all is bustle and -hurry. We are in the streets, or rather in the distributing areas of the -plant-world. We hear the noise of the breaker, the roar of the gale, the -cry of the sea-gull, the flapping of a myriad pairs of wings of some -migrating host overhead, and we know that the current, the wind, and the -bird are actively at work; but their operations are confined mainly to -the beach, the mountain-top, the river, and the pond. - -Let us take a well-wooded Pacific island several thousand feet in -height. We find on its beaches the same littoral plants that we have -seen before on the tropical shores of Malaya, of Asia, of Africa, and of -America. We find in its ponds and rivers the same species of -water-plants, such as Ceratophyllum demersum, Ruppia maritima, and Naias -marina, that are familiar to us in the cool and tepid waters of much of -the globe. On its level summit, if it remains within the clouds we find -in the boggy ground, where Sphagnum thrives, genera that are represented -in Fuegia, New Zealand, and the Antarctic islands, such as Acæna, -Lagenophora, and Astelia, and the world-ranging Drosera longifolia. In -other elevated localities we find Ranunculus, Geranium, Sanicula, -Artemisia, Vaccinium, and Plantago, chiefly genera of the temperate -regions of the northern hemisphere; whilst there are also found Gunnera, -Nertera, and Uncinia, all hailing from the south and belonging to the -Antarctic flora characterising all the land-area around the globe in the -latitude of New Zealand and Fuegia. The Hawaiian species of Nertera and -of Uncinia occur also in New Zealand, and the first-named is found also -in Tristan da Cunha and in South America. In the Hawaiian uplands there -is also to be seen Deyeuxia, a genus of grasses found in the Tibetan -highlands and in the Bolivian Andes at elevations of 16,000 to 19,000 -feet; and the same species that exists in Australia may be found in the -mountains of Hawaii. Here also, both in Hawaii and Tahiti, occurs Luzula -campestris. - -In making the foregoing remarks on the alpine plants of a Pacific -island, I have had Hawaii in my mind, but we find the elements of a -similar widely-distributed mountain-flora in the less lofty peaks of -Tahiti and Samoa, and traces even in Fiji, where the mountains, however, -have only a moderate elevation. But the point I wish to lay stress on is -the cosmopolitan yet temperate character of the mountain-flora of an -island lying in the midst of the tropical Pacific. As he shifts his -station on this mountain-summit, the observer might at different times -imagine himself in the Sierra Nevada of California, on a Mexican -tableland, on a peak of the Andes, or in the lowlands of Fuegia. Other -plants that I have not mentioned, such as Coprosma, would bring back to -him New Zealand. He might even be on a mountain-top in Central Africa, -or on a Madagascar plateau; whilst in the boggy region of an elevated -Hawaiian tableland he would meet with not only the physical conditions, -but also several of the plants found on the higher levels of Tristan da -Cunha. - -It is, however, to be noted that although these mountain-tops in the -mid-Pacific have been stocked with genera from the four quarters of the -compass, the species as a rule are restricted to that particular -archipelago. Whilst the beach and the river in most cases possess plants -that have very wide ranges over the earth, a good proportion of the -species on the mountain-summit are not found elsewhere. This implies a -partial suspension of the means of dispersal on the mountain-top, whilst -the currents and waterfowl are still actively distributing the seeds of -the littoral tree and of the aquatic plant. We here get a foreshadowing -of another great principle, or of another line along which Nature has -worked in stocking these islands of the Pacific with their plants, a -subject concerning which much will be said in later pages. - -Hitherto, we have dealt only with a small proportion of the flora, and -with but a small portion of the area of the island. We have yet to deal -with the intermediate region between the sea-border and the summit of -the island, or, in other words, with the forested mountain slopes. This -is the home of many of the peculiar species and peculiar genera, both of -plants and birds; and it is with this zone that we shall be mainly -concerned when we come to contrast the floras of the several -archipelagoes of the tropical Pacific. Here the agencies of dispersal -have, to a large extent, ceased to act; and the question will arise as -to the connection between the endemic character of the plants and the -endemic character of the birds. We shall have to ask why this island, -after receiving so many plants, ceased to be centres of dispersal to -other regions. It is possible that these seeds or fruits have lost their -capacity for dispersal; but only a few instances of this change present -themselves. Rather it may be supposed that the birds that originally -brought the seeds to the island came to stay; and this at once suggests -another query as to the cause of the change of habit. I am alluding here -not to the plants with minute seeds, such as Sagina and Orchis, which -Mr. Wallace, in his _Darwinism_, regards as capable of being transported -by strong winds over a thousand miles of sea; but to those numerous -plants found in the Fijian, Tahitian, and Hawaiian forests, where the -seeds and “stones” are large and heavy, measuring often as much as a -quarter of an inch (6 mm.), and sometimes nearly an inch (25 mm.) in -size. The reader will be surprised to learn how little “size” has -determined the distribution of seeds and fruits in the Pacific. He will -have to appeal to the habits of pebble-swallowing of the Dodo, the -Solitaire, the Goura pigeon, the Nicobar pigeon, &c., if he desires to -find a parallel in the habits of birds. - -It is here assumed that the reader is already acquainted with the -principles involved in a discussion of island-floras, principles clearly -laid down in the writings of Hooker, Wallace, Hemsley, and others. As a -general rule in an island or in a group of islands where there are a -large number of plants not found elsewhere, there is also a large -endemic element in the avifauna, and where none of the plants are -peculiar, endemic birds are either few or wanting. As an example of the -first we may mention Hawaii, and Iceland affords an instance of the -second. But there is no hard and fast rule connecting the endemic -character of the plants and birds of an island with its distance from -other regions. Even the small group of Fernando Noronha, lying only some -200 miles off the coast of Brazil, possesses its peculiar birds and its -peculiar plants; and we can there witness the singular spectacle, as -described by Mr. Ridley, of an endemic bird, a frugivorous dove, engaged -in scattering the seeds of endemic plants over the little group. This is -the only fruit-eating bird in the islands, remarks the same botanist in -the _Journal of the Linnean Society_ (vol. 27, 1891); and “when one sees -the number of endemic species with edible fruits, one is tempted to -wonder if it were possible that they were all introduced by this single -species of dove, or whether other frugivorous birds may not at times -have wandered to these shores.” This inter-island dispersal in a -particular group of peculiar plants by peculiar birds is a common -spectacle in the Pacific. The contrast between the large number of -plant-genera possessing fruits that would be dispersed by frugivorous -birds and the poverty of fruit-eating birds in the avifauna is well -displayed in Hawaii. - -The island of St. Helena would seem to offer an exception to the rule -that endemic birds and endemic plants go together, since, though its -flora possesses a very large endemic element, there are scarcely any -endemic or even indigenous birds recorded from the island. We can never -know, however, how much of the original fauna disappeared with the -destruction of the forests. It would nevertheless appear that but few of -the genera possessing peculiar species of plants were adapted for -dispersal by frugivorous birds. The lesson to be learned from this -island concerns the Compositæ, often arboreous, that constitute the -principal feature of its flora. St. Helena retains almost more than any -other island evidence of the age of Compositæ which has left its impress -on many insular floras; and when we discuss the original modes of -dispersal of the endemic Hawaiian genera of the same order we shall look -to the flora of this Atlantic island for assistance in the matter. To -the age of Compositæ belong the beginnings of several insular floras. - -To return to the main line of our argument, it would seem that in a -Pacific island there is a constant relation between free means of -dispersal and the preservation of specific characters. The ocean-current -and the aquatic bird are in our own time actively engaged in dispersing -the seeds of shore-plants and water-plants, and we see the same species -ranging over the world. On the other hand on the mountain-top the -agencies of dispersal are beginning to fail, and as a result many a -mountain has some of its species restricted to its higher regions. In -the forest zone there has been a more or less complete suspension of the -activity of the dispersing agencies, and new genera are formed whilst -peculiar species abound. Free means of communication with other regions -restrains but does not arrest the differentiating process that is ever -in progress throughout the organic world. Isolation within certain -limits gives it play. - -It is in this connection interesting to reflect that during the -differentiation of the inland flora the littoral plants have lagged -behind or have remained relatively unchanged. The currents have been -working without a break throughout the ages; and the cosmopolitan -Ipomœa, that now creeps over the sand of the beach, or the wide-ranging -Rhizophora, that forms the mangroves of the coast-swamp, must have -witnessed the arrival of the ancestors of several of the endemic inland -genera. The swamp-plants of the littoral flora are probably older, -however, than the beach-plants which have been recruited from time to -time in one region or another of the tropics from the inland flora. Yet -as a body the littoral plants have lagged far behind the inland flora. -We might thus expect that in a Pacific island, excluding the -wind-distributed plants, such as the ferns and the lycopods, the most -ancient types of the plants would be found at the coast, the most modern -in the forests, whilst the plants of the mountain-summit would represent -an intermediate age. - -But true as this may be, the composition of a strand-flora is a very -complex one. Although, as Prof. Schimper remarks, the mangrove formation -is more isolated than the beach formation, and affords evidence of a -much earlier separation, the beach-plants as a body are anything but -homogeneous in their character. Their physiognomy varies to some extent -with the alteration in the characters of the inland flora, changes to -which the mangrove formation makes a very slow response. Yet amongst the -plants of the beach we find strangely assorted forms that are as ancient -denizens of the coast as the mangroves themselves. Take, for instance, -Salsola Kali, that thrives alike on a beach in Chile, on the sea-shore -in Devonshire, and in the salt-marshes of the interior of Tibet. Then, -again, there is a type of littoral plant, of which Armeria vulgaris and -Plantago maritima may be taken as examples, which is equally at home on -the beach and on the tops of inland mountains. We might in a sense apply -the wrecker’s motto, - - “What the sea sends and the land lends,” - -to the history of a littoral flora. Yet on the other hand the inland -flora in its turn receives a few recruits from the littoral flora; and -it is the relation between the inland and coast species of the same -genus that offers one of the most fascinating studies in the botany of -the Pacific Islands. - -This introductory chapter may be concluded with a few remarks on what -may be termed “the ethics of plant-dispersal.” Not that this is in any -way a suitable phrase, but it best expresses my sense of the lack of -propriety in some things connected with this subject. It is odd, for -instance, that we speak of the dispersal of plants and animals in the -same breath, as if the process was in both cases identical. Seeing that -from this point of view we judge a plant only by its seeds and fruits, -it is apparent that we are following quite a different method than that -which we employ in the study of the dispersal of animals. Whilst the -zoologist classifies the units of dispersal, the botanist does nothing -of the kind; and the two systems of classification are at the outset -fundamentally distinct. The student of plant-dispersal thus often finds -himself placed in an awkward dilemma. For him a family is a collection -of allied genera having similar seeds or fruits and fitted often for the -same mode of dispersal. A family like Sterculiaceæ, possessing such a -variety of seeds and fruits suitable for very different modes of -dispersal, is from his standpoint a collection of dissimilar units. -Genera like Commersonia, Waltheria, Kleinhovia, Sterculia, and -Heritiera, that he so often meets with in the Pacific Islands, have in -these respects frequently very little in common; and yet one of the -earliest determining influences in plant-life must have lain in the -capacity for dispersal. - -Yet chance seems to reign in the processes of plant-dispersal ever going -on around us. In the floating seed, in the achene with its light pappus -blown before the gale, in the prickly mericarp entangled in the plumage -of a bird, in the “stone” of the drupe disgorged or ejected by the -pigeon, in the small grain that becomes adhesive in the rain, in the -tiny rush-seed enclosed in the dried pond-mud on the legs of some -migratory bird, in all these we see the agencies of dispersal making use -of qualities and of structures that were developed in quite another -connection and for quite another purpose. That such characters have been -so to speak appropriated by these agencies is a pure accident in a -plant’s life-history. If the evolutionary force had been in operation -here, it would have selected some common ground to work on. There would -have been some uniformity in its methods, whereas the modes of dispersal -are infinite. The qualities and characters that happen to be connected -with dispersal belong to a plant’s development in a particular -environment. They can never have been adapted to another set of -conditions that lie quite outside that environment. There is a relation -of a kind between the specific weight of wood and the density of water, -and this, in a sense, sums up the connection between a seed and its -distributing agencies. - -Evolution has never concerned itself directly with means of dispersal. -Evolution and Adaptation represent the dual forces that rule the organic -world, the first an intruding force, the last a passive power -representing the laws governing the inorganic world. To these laws the -intruding power has often been compelled to bend, and it has had to pay -its price, and sometimes it has succumbed, and sometimes it has turned -its defeat into a victory. Nature, so watchful over the young plant, as -represented by the seed, is finally compelled to let it go, and -dispersal begins where evolution ends, or rather when the evolutionary -power fails. The seed-stage itself is the price of adaptation. The death -of the individual may also be regarded from the same standpoint. It -represents a defeat of the evolutionary force, which, however, has been -retrieved by the gift of reproductive power. - - - - - CHAPTER II - - THE FLORAS OF THE PACIFIC ISLANDS FROM THE STANDPOINT - OF DISPERSAL BY CURRENTS - -The initial experiment.—The proportion of littoral plants.—The two great - principles of buoyancy.—The investigations of Professor Schimper.—The - investigations of the author.—The great sorting process of the - ages.—Preliminary results of the inquiry into the buoyancy of seeds - and fruits. - - -IN the previous introductory chapter some of the numerous questions -affecting insular floras were briefly referred to. I will now ask my -reader, if he has had the patience to read it, to consign that chapter -for the time at least into oblivion, and to proceed with me to our -Pacific island with the intention of investigating its flora from the -standpoint of dispersal. We will together take up the subject _de novo_, -after banishing from our minds all preconceptions that we may have -possessed. - -After having been over the island gathering specimens of all the seeds -and fruits, we return to our abode on the beach. But we are puzzled -where to begin. The problem presents itself as a tangled skein, and our -difficulty is to find an “end” that we can follow along with some -chances of success. In our trouble we look around us; and at that moment -we see a number of floating seeds and fruits carried by the current past -the beach. This presents us with a clue and our investigation begins. - -[Illustration: - - OCEANIA - John Bartholomew & Co., Edin^r. -] - -We place all our seeds and fruits in a bucket of sea-water and notice -that many of them sink at once. In a few days we look again and observe -that many more are at the bottom of the bucket, only a small percentage -remaining afloat. We then remark to our surprise that nearly all of the -floating seeds and fruits belong to coast plants, those of the inland -plants, which indeed make up the great bulk of the flora, having, as a -rule, little or no buoyancy. After a lapse of weeks and months the seeds -and fruits of the coast plants are found to be still afloat. In the -results of this experiment we see the work of the ages. There has been, -in fact, a great sorting process, during which Nature has “located” the -plants with buoyant seeds or seed-vessels at the sea-coast, placing the -others inland. This is the clue that we shall follow up during many -chapters of this book; and having in this manner introduced the reader -to the subject, I will now refer to the general results of my -investigations in this direction in the Pacific Islands. - -In Fiji there are about eighty littoral plants out of a total of at -least 900 species of indigenous flowering plants, that is to say about -nine or ten per cent. (Note 1), the littoral grasses and the sedges -being with one or two exceptions excluded. These shore plants belong to -the sandy beach and to the coast swamp, and most of them are distributed -over the tropical shores of the Indian and Pacific Oceans, whilst not a -few occur on the coasts of tropical America. They form the -characteristic plants of the coral atoll, and many of them have long -been known to be dispersed by the currents. From the list given in Note -2 it will be seen that these eighty species belong to about seventy -genera. Nearly all of them (95 per cent.) possess seeds or seed-vessels -that float at first in sea-water; whilst three-fourths of them (75 per -cent.) will float unharmed for two months and usually much more, and -several of them will be found afloat after a year or more, being still -capable of reproducing the plant (Note 3). - -The prevalence in the Fijian strand-flora of Leguminosæ, which are -included in my list under the divisions Papilionaceæ, Cæsalpinieæ, and -Mimoseæ, is very significant. They make up about 29 per cent. of the -total. Excluding weeds and a few other introduced plants, there are some -fifty species known from the Fijian Islands, and of these almost half -belong to the littoral flora, which as we have seen constitutes only a -fraction (one-tenth) of the whole flora. If we regard the genera, we -find that out of some thirty Leguminous genera twenty are littoral and -in most cases exclusively so. This conspicuous feature in the -constitution of the strand-flora is of prime importance as concerns the -question of adaptation to dispersal by currents, since nearly all the -Leguminosæ with buoyant seeds offer themselves as defiant exceptions to -any such law. - -I will now contrast the Fijian inland flora with that of the coast from -the point of view of the buoyancy of the seed or fruit, according as it -presented itself for possible dispersal by currents. Rather over a -hundred plants were experimented upon (Note 4). After excluding some -introduced plants there remain some ninety species belonging to about -sixty genera, and of these quite 75 per cent. sank at once or in a few -days. I may add that all kinds of fruits are here represented, the -capsule, the achene, the coccus, the berry, the drupe, &c. Of the -buoyant residue few possess seeds or fruits that will float uninjured -for any length of time. Not many gave indications directly in opposition -to the principle that whilst the seeds or fruits of shore-plants -generally float, those of inland plants usually sink, since as pointed -out in Note 5 most of the difficulties are removed during the subsequent -developments of the principle discussed in the later pages of this work -or are to be explained on other grounds stated in the note. - -We pass now from Fiji as typical in its flora of the Western Pacific to -Tahiti as representing in its flora the more strictly oceanic groups of -Eastern Polynesia. In the Tahitian region, which is taken as including -in a general sense the Society Islands, the Marquesas, and the Paumotus, -there are only between 50 and 60 littoral plants, excluding the -occasional additions from the inland flora. As indicated by the letter T -preceding the species in the list of Fijian shore plants, nearly all are -to be found in Fiji, and the few not yet recorded from that group, which -I have referred to in the remarks following the list, will probably be -found there by some subsequent investigator. In Tahiti also between 75 -and 80 per cent. of the strand plants have seeds or seedvessels that -float for months; and here also Leguminosæ predominate, forming about 30 -per cent. of the total. A conspicuous negative feature in the Tahitian -strand-flora is concerned with the absence of the mangroves and their -numerous associated plants, which together form the mangrove formation -in Fiji. This remarkable character in the distribution of shore plants -in the Pacific is discussed in Chapter VI. - -Not having visited Tahiti, I can only deal inferentially with the inland -plants, as in the case of the strand-flora. Here also the plants are in -the mass Fijian in a generic and often in a specific sense, and there is -no reason to believe that the principle involving the non-buoyancy of -the seeds or fruits of inland plants does not as a rule apply to Tahiti -as well as to Fiji. - -The Hawaiian Islands, standing alone in the North Pacific, form a floral -region in themselves, a region that is the equivalent not of one group -in the South Pacific, such as that of Fiji or of Tahiti, but of the -whole area comprising all the groups extending from Fiji to the Paumotu -Archipelago. Lying as it does mainly outside the zone of influence of -the regular currents that would bring the seeds of tropical plants to -its shores, Hawaii possesses a strand-flora that is meagre in the -extreme. Not only does it lack the mangrove formation so characteristic -of Fiji, but it lacks also many of the plants of the beach formation -that are found both in Fiji and in Tahiti, plants that give a peculiar -beauty to the reef-girt beaches all over the South Pacific. Its poverty -is sufficiently indicated in the number of its species, thirty in all, -barely more than half of the number found in Tahiti, and not much over a -third of those occurring in Fiji. Though coral reefs with their -accompanying beaches of calcareous sand are relatively scanty, the -characteristic littoral plants have not been numerous enough to hold -their own against intruders from the inland flora, and endemic species -have taken a permanent place amongst the strand plants. The Hawaiian -strand-flora has thus quite a facies of its own, and it will be found -discussed in Chapter VII., whilst a list of the plants is given in Note -28. It will thus not be a matter for surprise that the littoral flora of -Hawaii follows the principle of buoyancy only in a modified degree. It -is true that about two-thirds of the species of the present beach flora -possess seeds or seed-vessels that float for months; but since there are -reasons for believing that several of them are of aboriginal -introduction, this proportion is reduced to a third. In the list of the -Fijian shore plants given in Note 2, those occurring also in Hawaii are -preceded by H. - -When we look to the Hawaiian inland flora for indications respecting the -principle of the non-buoyancy of the seeds or seed-vessels of inland -plants, we find that so far as it has been there tested this principle -receives fresh support from the plants growing on the slopes of the -Hawaiian mountains. Although the author was only able to sample the -inland flora, we have in the list given in Note 6 all kinds of plants, -from the forest-tree to the herb, and most varieties of fruits. -Excluding a few introduced plants, there are in this list about fifty -species of indigenous plants belonging to about forty genera. Of these -plants quite 80 per cent. possess seeds or fruits that sink either at -once or in a week or two. Of the “buoyant” residue very few have seeds -or fruits that will float for months. These apparent exceptions to the -principle are in great part capable of being explained on the grounds -referred to in Note 5 in connection with the Fijian inland plants; and I -have alluded to them in Note 7. - -The littoral flora of Fiji is essentially Malayan and Asiatic, and for -our purpose is eminently typical. Its plants are found far and wide on -the tropical coasts of the Old World, and sometimes also in the New -World. In more than half the species we are concerned with the dispersal -by currents of more or less dry indehiscent fruits that range usually in -size from a marble to a cricket-ball, as illustrated by those of -Hernandia peltata and Barringtonia speciosa, whilst with most of the -rest the currents distribute large seeds, several of which are -Leguminous, as in the case of Mucuna, Cæsalpinia, and Entada, with -others of the Convolvulus type, as in the instance of Ipomœa pes capræ. -It is remarkable that in selecting plants with buoyant seeds or -seed-vessels for a station at the coast Nature has generally ignored -those with very small seeds. When such small seeded plants, as Sesuvium -portulacastrum, occur on the beach, the seeds have as a rule no -buoyancy. Pemphis acidula is, however, an exception; but its case is a -very rare one. It will be established in the next chapter that the -non-buoyancy of small seeds is generally true also of plants growing by -the river or by the pond. - -The point at which we have arrived in our inquiry concerning the general -collection of seeds and seed-vessels that we placed in sea-water is that -_the plants with buoyant seeds or seed-vessels have been for the most -part “located” at the coast_. But if we look a little more closely at -the sunken and floating seeds, we find that in the same genus there are -species with seeds or seed-vessels that sink and species with those that -float. We look again and then perceive that the same general principle -is true of different species of the same genus growing inland and at the -coast. We learn now that as a rule when a genus possesses both littoral -and inland species, the seeds or fruits of the former float in sea-water -for a long time, whilst those of the latter have little or no floating -power. But we have yet to examine the structure of the coverings of the -buoyant seed or fruit; and we shall then discover that the different -behaviour in water is often associated with corresponding structural -differences of a striking character. The structural causes of buoyancy -are dealt with in Chapter XII.; and we will now content ourselves with -enunciating the second principle that _in a genus comprising both coast -and inland species, only the coast species possess buoyant seeds or -seed-vessels_. - -The important principle above indicated was not altogether new to me, as -is shown in the next chapter. But it was new in the case of the floras -of the Pacific Islands. When it first presented itself in Hawaii I was -engaged in trying to find a connection between the inland and littoral -species of Scævola; and its discovery led me to form a plan worthy -almost of Don Quixote, namely, to cultivate the beach species of Ipomœa, -Scævola, and Vitex in the interior with the hope of finding them -converted into inland species when I returned to Hawaii after a lapse of -years. Little matters often determine a career, and for a while my -future movements and probably the remainder of my life were largely -centred around my interests in the well-being of Scævola Kœnigii. The -scheme was actually undertaken, and I had fixed on a little plot of land -at the foot of the mountains rising behind Punaluu in Kau. The -transaction was on the point of completion when the owner changed his -mind and the plan fell through. Subsequent observation and reflection -have led me to believe that in most cases no connection exists between -the littoral and inland species of a genus; and I have dwelt on this -incident merely to show the importance that I rightly attached to this -distinction, whilst misinterpreting its meaning. - -But to return to my own investigations. Had I indeed read more carefully -Professor Schimper’s work on the Malayan strand-flora, this subject -would have been found discussed by an observer far abler than myself, -though from a very different standpoint, that of Adaptation and Natural -Selection. He points out (pp. 179-182) that with a number of these -tropical genera possessing both littoral and inland species, such as -Barringtonia, Calophyllum, Clerodendron, Cordia, Guettarda, and -Terminalia, greater buoyancy of the fruits of the shore species is -associated with certain structural characters in the fruit-coverings, -whilst with the inland species, where the floating power of the fruits -is either much diminished or entirely absent, these structural -characters are either less developed or lacking altogether. - -The question of structure and the debateable matters concerned with it -are treated at some length in Chapters XII. and XIII., and Professor -Schimper’s views are there given. I will content myself with remarking -that the genus Terminalia was especially studied by him in this respect. -He tested the buoyancy of the fruits of ten species, and found that the -flotation period varied from nothing to 126 days and more. By far the -best “floaters” were the fruits of Terminalia Katappa, the only littoral -species tested, all the others being inland species with less buoyant -fruits, and diminished ranges, some of the fruits sinking at once, -whilst the others sank usually in a few days or in a few weeks. It was -also ascertained that, although the buoyant tissue in the fruit-coats -varied in amount generally with the floating-powers, it was rarely -absent altogether in the inland species, a very significant conclusion, -as will subsequently be pointed out. - -Several other striking examples of this principle came under my notice -in the Pacific, and perhaps the most significant is that of Scævola, a -genus of the Goodeniaceæ, confined mainly to Australia and the Pacific -islands, but possessing also a littoral species, S. Kœnigii, that is -found on tropical beaches all round the globe. It is associated in both -Hawaii and Fiji with inland species, none of which are common to the two -archipelagoes, and in the case of the Hawaiian species not found outside -the group. All the species have fleshy drupes, both coast and inland -plants, the “stone” in the littoral species possessing a thick covering -of buoyant tissue, which is absent or but slightly developed in the -inland species. The fruits of the shore species float for many months; -whilst those of the inland species experimented on by me (S. -Chamissoniana and S. Gaudichaudii in Hawaii, and S. floribunda in Fiji) -sank at once or within a few hours. Here we are only concerned with the -difference of buoyancy between inland and littoral species. The several -other questions involved concerning this genus will be dealt with later -on in this work. - -The genus Morinda offers another good example of this principle. It -includes one widely-spread littoral species (M. citrifolia), found not -only in all the Pacific archipelagoes, but also over much of the -tropics. It is associated in all the large groups with one or more -inland species, some of which are endemic and others more generally -distributed. The littoral species displays in its pyrenes a singular -air-cavity, the nature of which is discussed in Chapter XII., which -endows them with great floating powers. This cavity is not found in -inland species, and the pyrenes have in consequence no floating power -(see Note 8). - -Calophyllum Inophyllum, an Old-World littoral tree, spread far and wide -over the Pacific islands, has very buoyant fruits. In the groups of the -South Pacific it is associated with inland species that are commonly -found in the forests, namely, C. spectabile and C. Burmanni, the fruits -of both of which, according to my observations in Fiji, have limited -floating powers, sinking after periods varying from a few days to four -weeks, and lacking in great part the buoyant coverings of the littoral -species. Professor Schimper obtained similar results with inland species -from other regions (Note 9). - -The fruits of the two Fijian coast trees, Barringtonia speciosa and B. -racemosa, possess great floating powers; whilst those of an undescribed -species that I found in the mountains of Vanua Levu sink at once. -Another Fijian inland species (B. edulis, Seem.) that is often planted, -has fruits that float heavily for about a month. This difference in -buoyant powers is also associated with characteristic differences in the -structure of the fruits. It would be interesting to learn what floating -capacity belongs to those of the Samoan endemic species (B. samoensis, -Gray). Professor Schimper’s observations on the genus in the Malayan -region point in the same direction, but more than one difficulty awaits -its solution in the re-examination of the genus. He says, however, that -B. excelsa, Bl., a Malayan species, sometimes cultivated and growing -both inland and at the coast, has fruits that floated for one hundred -days after drying (p. 173). - -A striking instance of this principle is afforded in the case of the two -Fijian species of Tacca, the wide-ranging littoral species, T. -pinnatifida, where the seeds float for several months, and the inland -species, T. maculata, Seem., found also in Australia and Samoa, where -the seeds sink at once or in a few days. The seeds of the shore plant -owe their buoyancy to the spongy tissue in their coverings, which is -either absent or much less developed in those of the inland species. -This point might also be determined for the new Samoan inland species -described by Reinecke, the German botanist, as T. samoensis. - -Another good illustration is afforded by the two species of Premna of -the South Pacific, though here the buoyancy of the “stone” of a drupe is -concerned. With P. taitensis or P. integrifolia, a small littoral tree -or shrub, these stones possess great floating-power, and are often found -in the floating seed-drift of the Fijian estuaries and in the stranded -drift of the beaches. In the case of Premna serratifolia, an inland tree -of moderate size, the stones have as a rule little or no buoyancy. As -shown in Note 32, where this genus is discussed in detail, the buoyancy -is mainly due to empty seed-cavities. - -Other instances might be given in illustration of this principle; but it -will have been noticed that already many of the familiar trees and -shrubs of a tropical beach have been mentioned in this connection either -by Professor Schimper or by myself. There are other genera that afford -similar indications but in a less direct fashion. - -For instance, there are three widely spread Leguminous beach plants of -the Pacific, Erythrina indica, Canavalia obtusifolia, and Sophora -tomentosa, none of which are found in Hawaii; but in that group the -genus is represented in each case by an inland species, Erythrina -monosperma, Canavalia galeata, and Sophora chrysophylla, the last two -species being peculiar to those islands. The seeds of the three littoral -species will float for a long time in sea-water, whilst those of the -three Hawaiian inland species have no buoyancy. I may say that some very -interesting questions relating to the origin of these inland species are -here raised. They will be discussed in a later chapter (Chap. XV.). - -There are a number of plants belonging to the Convolvulaceæ in these -islands that behave in an irregular way in flotation experiments ; but -their inconstant behaviour can in most cases be explained in accordance -with the principle that in the same genus the shore species have buoyant -seeds and the inland species non-buoyant seeds. Thus, whilst the seeds -of the littoral species, Ipomœa pes capræ, I. grandiflora (Lam.), and I. -glaberrima (Boj.), can float for long periods, and those of the inland -species, I. pentaphylla, I. tuberculata, and I. Batatas (Sweet Potato), -have no buoyancy, the seeds of other inland species, I. insularis -(Steud.), I. bona nox (L.), and I. turpethum (R. Br.), are inconstant in -their behaviour. The three last-named species are, however, to be found -also flourishing at times at and near the coast, and the varying -floating powers of their seeds may probably be connected with their -varying stations. This is indeed suggested by the case of Argyreia -tiliæfolia in Hawaii, in which in my experiments the seeds of plants -growing at the coast floated, sometimes for months, whilst those from -inland plants sank. - -This behaviour of the Convolvulaceæ becomes yet more intelligible, and -more in accordance with the principle, when we reflect that the cause of -buoyancy is not concerned with the seed-coats or with the nucleus, -neither of which are able to float, but with the air-spaces left by the -incomplete filling-up of the seed-cavity by the crumpled embryo. The -extent to which the seed-cavity is filled up varies not only between -different genera and between different species of the same genus, but -also amongst individuals of the same species. Even the seeds of Ipomœa -pes capræ, amongst the most typical of floating seeds, display this -variation, and they show it also in their floating power, since about a -third of the seeds usually sink during the first month or two of the -flotation experiments. We can thus explain also why in the case of -Ipomœa insularis seeds from Fiji floated for months, whilst those from -Hawaii had no floating power. - -The seeds of the different species of Hibiscus also appear to behave -very irregularly; but even here most of the difficulties can be removed, -when we come to consider a further extension of the principle. Thus, -whilst the seeds of Hibiscus tiliaceus, a wide-ranging littoral tree -known to be dispersed by the currents, float for a long time, those of -H. Youngianus (Gaud.), an endemic Hawaiian species, and of two -wide-ranging species, H. diversifolius (Jacq.) and H. Abelmoschus (L.), -also float for some time. The Hawaiian plant, however, grows in wet -places; and this applies also to H. diversifolius which grows in swamps -at and near the coast. The extension of the principle to water-side -plants generally, which is discussed in the next chapter, will explain -the difficulties connected with these two species. But we have in H. -Abelmoschus a remarkable exception to any rule of buoyancy, since it -grows in dry situations, is often cultivated, and yet possesses a -special layer of buoyant tissue in the seed-coats to which the floating -power is due. The seeds of Hibiscus esculentus (L.), the widely spread -cultivated plant of the tropics, have no buoyancy. - -Some curious indications are supplied by Cæsalpinia, a Leguminous genus, -containing two wide-ranging shore species. Speaking generally the rule -applies; and I found in Fiji that whilst the seeds of the two littoral -plants (C. Bonducella and C. Bonduc) were as a rule buoyant, those of an -inland mountain species sank. But it is very remarkable that although -the seeds of C. Bonducella have long been known to be transported by the -currents, and are often stranded by the Gulf Stream on the coast of -Scandinavia, when it grows in Hawaii, where it is as a rule an inland -plant, the seeds lose their buoyancy. This is quite in accordance with -the general principle; but I must refer the reader for a general -treatment of this genus to Chapter XVII. There also will be found the -instance of another Fijian littoral plant, Afzelia bijuga, a common -littoral tree with buoyant seeds which also lose their buoyancy when the -tree grows inland. A similar instance is afforded by Kleinhovia Hospita, -the seeds of which seem to lose their buoyancy in inland stations. Not -all littoral plants, however, lose the floating power of the seeds when -grown away from the coast. The seeds of Ipomœa pes capræ retain it in -spite of the change of station. This point is dealt with in Chapter XIII -and in Note 44. - -In concluding this general sketch of the first results obtained by -testing the buoyancy in sea-water of a collection of seeds and fruits -from a mountainous Pacific island, such as we find in Fiji, I must -remind the reader that the subject has only been lightly treated. -Enough, however, has been said to illustrate the character of the -sorting-process by which in the course of ages the plants with buoyant -seeds or seedvessels have been gathered at the coast. This is -indicated:— - -(1) By the far greater proportion of species with buoyant seeds and -seedvessels amongst the shore plants than among the inland plants. - -(2) By the circumstance that almost all the seeds or fruits that float -unharmed for long periods belong to shore plants. - -(3) By the fact that when a genus has both inland and littoral species, -the seeds or fruits of the coast species as a rule float for a long -time, whilst those of the inland species either sink at once or float -only for a short period. - -These results, therefore, justify our dividing the flora of our island -into two groups, the one including the plants with buoyant seeds or -fruits and comprising most of the littoral plants, the other including -the plants with non-buoyant seeds or fruits, a group which contains -almost all the inland plants and indeed nine-tenths of the flora. This -classification is a very crude one; but it enables us at once to assign -a value to the agency of currents in stocking a Pacific island with its -plants. Yet this is but the initial step in an inquiry that branches off -in a thousand different ways, even if restricted to the littoral plants. -There are a host of difficulties connected with the history of the -strand-flora of such an island which can only be properly gauged when -viewed from various standpoints. - - - - - CHAPTER III - - THE LESSON OF THE BRITISH FLORA - -Results of observations on the buoyancy of over 300 British plants.—The - small proportion of plants with buoyant seeds or seedvessels.—Their - station by the water-side.—The great sifting experiment of the - ages.—Summary. - - -THE singular relation between station and seed-buoyancy that exists in -an island of the tropical Pacific, such for instance as Vanua Levu, -Tahiti, or Hawaii, would lose much of its significance if it stood alone -in the economy of plant-life. It must be true not only of tropical -floras generally, but of those of the temperate regions; and there can -be little doubt that it prevails all over the world. Displayed to us at -first in a Pacific island, it acquires a new significance when we study -it in the light of numerous observations made in Europe. It exhibits -itself then as part of a far wider method pursued by Nature in -determining the stations of plants. It is not only at the coast, but -also at the river-bank and at the lake-side that Nature “locates” the -plant with the buoyant seed or seedvessel. This relation is indeed as -well exhibited in inland districts as it is at the coast. - -In this connection I have the results of my own investigations on the -buoyancy of the seeds and fruits of British plants and on the -composition of the seed-drift of ponds and rivers, which were carried on -in the years 1890-96. Some of them were published in a short paper on -the seed-drift of the Thames, read before the Linnean Society of London -in June, 1892, and in the columns of _Science Gossip_ for April, May, -and October, 1895; but the mass of the observations remain in my -notebooks. Nor do my observations of the period since elapsed lead me to -alter the position then adopted. I have since pursued the same line of -inquiry in Hawaii, Fiji, on the Pacific coast of South America, and in -Sicily, and with the same results. - -Since the elaboration of my notes was begun in 1900, Dr. Sernander, the -Swedish botanist, has published (1901) his work in Swedish on the -Dispersal-biology of the Scandinavian plant-world, in which the -seed-drift of river, pond, and sea is exhaustively treated. Although -this author has dealt with plant-dispersal from a somewhat different -standpoint, I have perused his pages with the keenest interest and with -great profit, having gone over much of the same ground with respect to -the seed-drift of ponds and rivers. Yet the introductory remarks to my -paper in _Science Gossip_ in 1895 are as apposite now as they were then, -and the reader will, I trust, pardon my reproducing them. - -“By following up the path of inquiry that is concerned with the -flotation of seeds and seedvessels, we are guided into other fields of -research that give promise of interesting discoveries in connection with -plant-life. We are led in the first place to consider the question of -utility, and to ask whether the buoyancy of the seed or fruit has been a -matter of moment in the history of the species. Nature is ever engaged -in telling off the plants to their various stations. She places the -yellow iris at the river’s side and assigns to the blue iris its home in -a shady wood. Under her direction the common alder thrives at the -water’s edge, whilst its fellow species live on the mountain slope. -These and similar operations are carried on daily around us, and we know -but little of the wherefore and the how. We are induced, therefore, to -inquire whether by pursuing the line of investigation above indicated we -may be able to get a glimpse at the methods adopted by Nature in -selecting stations for plants.” - -I possess the results, which are given in Note 10, of buoyancy -experiments and observations on the seeds and seedvessels of about 320 -British flowering plants belonging to about 65 families. Of these about -260 are included in my own results, the data for the rest being obtained -from the writings of Darwin, Martins, Thuret, Kolpin Ravn, and -Sernander. In the great proportion of cases, 240, or 75 per cent., -sinking took place at once or within a week; whilst 80, or 25 per cent., -floated for a longer period, usually a month or more; and about 60, or -nearly 20 per cent., floated for several months. It is to this last -small group that belong the seeds or seedvessels that float through the -winter in our ponds and rivers. - -If the grasses had been properly represented, the grains of which -possess as a rule but little buoyancy, except through air-bubbles -temporarily entangled in the glumes, the proportion of seeds and fruits -that sink at once or in a few days would probably have been about 80 per -cent. Then again, since the plants from stations where buoyant seeds and -seedvessels are most frequently found—that is at the river-side, the -pond-margin, and the sea-coast—are much more completely represented in -these experiments than those from other stations, it would seem that -even 80 per cent. is too low a figure. Even if the 80 plants with the -buoyant seeds or seedvessels included all the species thus -characterised, which they certainly do not, we should obtain an estimate -for the British flora (rather over 1,200 species of flowering plants) of -about 93 per cent. with non-buoyant seeds or fruits. This is, of course, -too high. It is, however, very probable that the proportion of plants -with non-buoyant seeds or seedvessels for the whole British flora is -about 90 per cent. - -This proportion of plants with non-buoyant seeds or seedvessels, that is -to say, of those that sink at once or within a week, is also -approximately correct for the flora of one of the larger islands of the -tropical Pacific. The data at my disposal only enable me in the cases of -Fiji and Hawaii to fix it at between 95 and 85 per cent., or on an -average 90 per cent. With the floras of continental regions the -proportion would doubtless be markedly higher. That seeds and -seedvessels as a rule possess but little buoyancy was a sound conclusion -of Darwin, and one, as he remarked, that is in accordance with the -common experience of gardeners. Thuret, after experimenting on the -buoyancy in sea-water of the seed or seedvessels of 251 species of -plants, belonging to 77 families and to various regions, found that -scarcely two per cent. had any powers of flotation, all the rest sinking -at once or in a few days, a result that led De Candolle in a note to -this memoir to reiterate his opinion regarding the inefficacy of -currents as plant distributors. Thuret, however, did not select many of -his plants from stations where buoyancy is most frequently exhibited, -and his estimate errs, therefore, in imputing too little buoyancy to -seeds in general. The power of seeds and fruits to germinate after -prolonged flotation in sea-water has long been well established, and it -is often illustrated in this work, so that there is no need to dwell -upon it here. (See Note 11.) - -Of the 240 species of British plants where sinking took place at once or -within a week, in about 50 per cent. the plants had dry indehiscent -fruits, such as we find in the genus Ranunculus and in the Umbelliferæ, -the Compositæ, and the Labiatæ; whilst in about a third the plants had -dehiscent fruits with small seeds, such as are characteristic of the -Cruciferæ, the Caryophyllaceæ, and the Juncaceæ. Plants with large -seeds, such as those of Nuphar luteum and Convolvulus arvensis, make up -only six per cent. of those of the non-buoyant group, the remainder -comprising plants with berries, such as Solanum, and others with -miscellaneous fruits. - -Of the 80 plants where the seeds or fruits floated more than a week, -usually for several weeks, and often for months, 70 per cent. possessed -dry, indehiscent fruits, such as those of Hydrocotyle vulgaris, Bidens -cernua, Lycopus europæus, Carex, &c., whilst only 6 or 7 per cent. had -dehiscent fruits with small seeds, such as we find in Lysimachia and -Menyanthes, the remainder being generally characterised by large seeds, -such as those of Convolvulus sepium, C. soldanella, Iris pseudacorus, -Calla palustris, &c. It would thus appear that, in so far as buoyancy is -concerned, Nature has for the most part ignored the small seed and has -confined herself mainly to the dry indehiscent fruit. We have already -seen that this is also true of the same great sorting-process in the -tropical islands of the Pacific, and it doubtless applies all over the -world. - -We have now to learn the significance of this distinction amongst -British plants between those with and those without buoyant seeds or -seedvessels. When we regard the stations of these 80 plants of the -buoyant group we find that about 70 per cent. of them are placed by the -river, or the pond, or the sea, the fresh-water stations much -predominating. But if we include the plants of the moist meadows -adjoining the rivers, such as Ranunculus repens, Rhinanthus crista -galli, some Cyperaceæ, &c., the buoyant fruits or seeds of which are -regularly swept into the stream in the time of flood, we shall raise the -proportion possessing a water-side station to 80 per cent. On the other -hand, about two-thirds of the 240 plants of the non-buoyant group, which -are enumerated in Note 10, live away from the water-side; but the -proportion of plants with a relatively dry station would be considerably -higher than this figure for the whole flora, since my investigations -were especially directed towards plants frequenting wet stations, and -the number of them is excessive in the list. - -Supposing, however, that our materials were restricted to the 260 plants -tested by myself, we should obtain highly instructive results, since in -a general sense the floating powers of their seeds or fruits were tested -to the finish. We place them, let us say, in a bucket of water, and -after six months we find that in not more than forty plants are the -seeds or seedvessels still afloat. These forty plants, excluding two or -three littoral plants, are nearly all plants of the borders and vicinity -of rivers and ponds. (They are indicated in the list given in Note 10 by -the numbers vi. and xii., the last being those where the flotation -experiment was prolonged to a year and over.) - -It would thus seem—I am now quoting mainly from my paper in _Science -Gossip_ for May, 1895—that there are gathered at the margins of rivers -and ponds, as well as at the sea-border, most of the British plants that -could be assisted in the distribution of their seeds by the agency of -water. This great sifting experiment has been the work of the ages, and -we here get a glimpse at Nature in the act of selecting a station. But -the curious character of the sorting process becomes yet more apparent -when we discover that the buoyancy of the seeds or fruits of species of -the same genus may become a matter of station. - -We will first take the four British species of Stachys (arvensis, -betonica, sylvatica, and palustris). Of these the fruits of S. palustris -alone possess any buoyancy, being able to float for weeks. It is the -only species that finds its characteristic home at the water-side; and -as observed by Sernander its reproductive shoots occur in the -Scandinavian fresh-water drift. - -Galium illustrates the same principle. Whilst in my experiments the -fruits of G. aparine and of another species growing in a dry station -displayed little or no floating power, those of G. palustre, which alone -grows at the water-side and in wet situations, have great buoyancy. As -my observations show, they float unharmed through the winter in our -ponds and rivers, and, according to Sernander, are often found in the -Baltic sea-drift. (See Note 12.) - -The achenes of Potentilla afford another example. Those of P. -tormentilla and of another species from dry situations have but little -floating power. On the other hand, those of P. comarum float -indefinitely. The last also came under my notice in the floating drift -of ponds in February; and we learn from Sernander that they occur in the -fresh-water and salt-water drift of Scandinavia. - -As a further instance, I will take the two British species of Iris. The -familiar river-side Iris pseudacorus has seeds that float unharmed in -our ponds and rivers from the autumn to the spring, and often for a year -or more. On the other hand, the seeds of Iris fœtidissima, which has its -home in the shady wood, sink at once even after drying for months. - -The nature of the sorting-process is especially well shown in some of -the families, as for instance with the Labiatæ. Let the reader put on -one side the four species with buoyant fruits, namely, Lycopus europæus, -Mentha aquatica, Scutellaria galericulata, and Stachys palustris, and on -the other side all the species with non-buoyant fruits, such as Salvia -verbenaca, Thymus sp., Calamintha officinalis, Nepeta glechoma, N. -cataria, Prunella vulgaris, Stachys arvensis, S. betonica, S. sylvatica, -Galeopsis tetrahit, Ballota nigra, Lamium purpureum, L. album, Teucrium -scorodonia, and Ajuga reptans, and he will at once perceive that he has -separated the regular water-side plants from those growing in drier -stations. - -If he does the same with the Umbelliferæ he will find that when he is -separating Hydrocotyle vulgaris, Cicuta virosa, Œnanthe crocata, and -Angelica sylvestris from Æthusa cynapium, Pastinaca sativa, and -Chærophyllum sylvestre, on account of their buoyant fruits, he is also -distinguishing them on account of their stations. On the other hand, -there are apparently weighty exceptions to this rule in the non-buoyancy -of the fruits of the three British species of Apium (graveolens, -nodiflorum, inundatum), which grow in streams and marshes. Or, again, if -we look at the sea-coast representatives of the family, we find that -whilst the fruits of the Samphire (Crithmum maritimum) float buoyantly -for months, those of Eryngium maritimum seemingly set the law at -defiance, and all sink in less than a week or ten days, even after -months of drying. To regard these as exceptions, however, is to miss the -essential point of the principle concerned. It is not thereby implied -that all water-side plants, whether by the sea or by the river or by the -pond, have buoyant fruits or seeds, but that nearly all plants with such -fruits or seeds have been gathered at the water-side. It will be shown -in the next chapter that several other influences go to determine the -station of a plant on a beach or by a river. This is true of the -Compositæ, which, if we except our two species of Bidens (cernua and -tripartita), come under the play of other determining causes, as -indicated by the little or no buoyancy displayed by the fruits of Aster -tripolium, Senecio aquaticus, and Carduus palustris. - -Within the limits of a genus we can, however, point to other examples of -this principle. Take, for instance, Convolvulus arvensis, the common -weed of our fields. Its seeds, whether fresh or dried for months, have -no buoyancy. On the other hand, those of Convolvulus soldanella float -unharmed in sea-water for half a year and more. Its seeds have come -frequently under my notice among the stranded drift of the Devonshire -beaches, and also on the coasts of Chile; whilst Sernander includes them -amongst the drift of the beaches on the Norwegian coasts. It is -remarkable that Convolvulus sepium, which accompanies C. soldanella over -much of its great range, has seeds that are sometimes able to float -unharmed for long periods, even for years (Notes 13, 41, 49). Though not -strictly a water-side plant, it grows commonly over other plants on the -banks of the Thames; and when it fruits its seeds occur typically in the -floating drift of that river. According to Gray, it is almost a -river-side plant in the United States, where it is found “especially on -the moist banks of streams.” Not all the seeds of C. sepium, however, -are buoyant; and in its varying behaviour in this respect it resembles -the inland species of Ipomœa, which are referred to in the previous -chapter. - -The British species of Euphorbia also seem to behave in accordance with -the principle that when a genus has littoral and inland species, the -first-named alone possesses buoyant fruits or seeds. Thus, whilst the -sound fruits of E. helioscopia and of another species found commonly as -a garden weed are non-buoyant, those of E. paralias, the familiar -beach-plant, float for several weeks, and are to be noticed among the -stranded drift of the coasts frequented by this plant. (See Note 90 for -later results.) - -The structural characters connected with the buoyancy of the seeds or -seedvessels of some of the British plants are dealt with in Chapter XII. -Here it may be remarked that this capacity is often associated, as with -the Pacific island plants, with a “buoyant” tissue, that is either -absent or less developed in the case of the non-buoyant group. - -Enough has now been said to show in a general fashion how Nature through -the agency of buoyant seeds and fruits has affected the stations of -plants of the British flora. Allowing this line of inquiry to develop -itself as the work proceeds, we will here pause and close the chapter -with a reference to some of the principal points that have been brought -into prominence. - -(_a_) The proportion of flowering plants of the British flora that -possess buoyant seeds or seedvessels is very small, probably not more -than 10 per cent. - -(_b_) In so far as buoyancy is concerned, Nature has for the most part -ignored the dehiscent fruit with small seeds, such as we see in the -Cruciferæ and the Caryophyllaceæ, and has chiefly endowed with floating -power the dry indehiscent fruit, such as we see in the Umbelliferæ and -in the Labiatæ. - -(_c_) In the great sorting-process that has been in operation through -the ages, nearly all the plants with buoyant seeds or seedvessels have -been located at the water-side, principally by ponds and rivers, but -also on the sea-beach. On the other hand, the great majority of the -plants with seeds or seedvessels that sink have found a home in drier -stations. - -(_d_) The character of the operation is well displayed in certain genera -possessing species of the water-side and species of drier situations, -and in the case of genera having both coast and inland species. In both -instances the species by the water-side possesses buoyant seeds or -fruits, whilst that of the station in a drier locality or removed from -the coast has seeds or fruits that sink. - -(_e_) Yet it is necessary to remember that the principle involved is not -that all water-side plants have buoyant seeds or fruits, but merely that -plants thus endowed gather at the water-side. There are many plants with -non-buoyant seeds or fruits on our beaches and beside our ponds and -rivers. - -(_f_) We have now learned from the British flora that the “locating” of -plants with buoyant fruits or seeds on the beaches of the tropical -islands of the Pacific, and indeed of tropical regions generally, is but -a part of a much wider principle by which plants thus endowed are placed -at the water-side, whether by a river or a pond or by the sea. - -(_g_) It is with this distinction between a fresh-water and a salt-water -station that we shall be occupied in the next chapter; and it is of -great interest, since it leads us to discover that the wider principle -is in its turn part of a far larger scheme. - - * * * * * - -_Note._—It must be clearly understood that by water-side plants the true -aquatic plants, such as the Water-lilies, the Myriophylls, the -Potamogetons, &c., are not implied. It will be seen from the list in -Note 10 that in most cases the seeds or fruits of aquatic plants have -little or no floating power. This is true, for instance, of Ranunculus -aquatilis, Nymphæa, Nuphar, Myriophyllum, Ceratophyllum, Callitriche, -Naias, Zannichellia, Ruppia, and half the Potamogetons. - - - - - CHAPTER IV - - THE LESSON OF THE BRITISH FLORA (_continued_) - -The choice of station of the water-side plant possessing buoyant seeds - or seedvessels.—Determined by its fitness or unfitness for living in - physiologically dry stations.—In the internal organisation of a plant - lies the first determining influence of station.—The grouping of the - British strand-plants.—Whilst the Xerophyte with buoyant seed or fruit - finds its station at the coast, the Hygrophyte similarly endowed makes - its home at the river or pond side.—The grouping of the plants of the - river and the pond.—Summary. - - -BY following up the clue supplied by the floating seed, we have arrived -at the conclusion with respect to the British flora that plants with -buoyant seeds or fruits gather at the water-side. But we have yet to -inquire why some of these plants are “located” at the sea-coast and -others on the borders of ponds and rivers. Mere buoyancy aided by chance -has not determined the choice. There are definite principles at work in -the economy of plant-life that make the selection for each plant. - -Rivers in all parts of the world carry to the sea in great abundance the -seeds and fruits of the plants that are stationed at their borders; and -such seed-drift is found in quantity washed up on the beaches in the -vicinity of the estuary. One finds, for instance, on such beaches in the -South of England the stranded fruits and seeds of Bidens cernua, Alnus -glutinosa, Sparganium ramosum, Iris pseudacorus, &c., mingled with those -of true beach plants like Cakile maritima, Convolvulus soldanella, -Euphorbia paralias, &c. Yet we would be much surprised if either the -Bidens or the Alder or the Sparganium were to establish itself on the -sandy beach, even though they have had through the ages innumerable -opportunities of doing so. We thus see that mere buoyancy of fruit or -seed cannot determine a station on a sea-beach, and that some other -factor makes the choice. The nature of this factor I will now endeavour -to explain; but in so doing it will be necessary to employ a few -technical terms, which it is not easy to dispense with altogether. - -It may be doubted whether Professor Schimper could have conferred a -greater benefit on the student of plant-distribution than in his clear -delineation of the connection between the habit or organisation of a -plant and its station. Nature has imposed an important structural -distinction between plants that have been endowed with the means of -checking excessive transpiration or water-loss in stations where there -is risk of drought, as in deserts and in similar arid localities, and -those that live in stations where such safeguards are not needed. Hence -arises the distinction between Xerophytes on the one hand, and -Hygrophytes on the other. This contrast is shown not only in minute -structural features, but also, as my readers are aware, in the external -characters, as in hairiness, succulency, a leathery cuticle, the -occurrence of thorns, and in several other characters of the plants of -the steppe and the desert. This important subject is dealt with by -Professor Schimper in his recent work on Plant-Geography; but it was -from his earlier work on the Indo-Malayan strand-flora that I learned -this valuable lesson in plant-distribution. - -It has been ascertained, however, that a safeguard against excessive -water-loss by transpiration is not only needed by plants living in arid -localities, but also by those placed at the coast. Both the shore plant -and the plant of the steppe and the desert present the same xerophilous -organisation, provision against excessive transpiration being also -required by the beach plant to prevent the injury of the green cells -from the accumulation of salt in the tissues. It would thus appear that -plants of the Hygrophytes that possess buoyant seeds or fruits are -gathered at the borders of ponds and rivers, whilst those of the -Xerophytes that are similarly endowed find their station on the -sea-shore. This important distinction penetrates very deeply into the -conditions defining the stations of plants. The connection between the -plant of the coast and the plant of the steppe or the desert is -strikingly shown on those occasions when the beach plants extend inland -over parched and arid plains, such as occurs for instance in North -Africa, and in the larger islands of Fiji, as described in Chapter V. - -The causes of the buoyancy of fruits and seeds, as pointed out in -Chapter XII, are so various, that it appears at first sight impossible -to connect them with the xerophilous or hygrophilous organisation of a -plant, or, in other words, with any structural characters associated -with particular stations; yet behind all lies the general principle -that, given a plant of the buoyant group, if it is a Xerophyte it finds -its way to the coast, and if a Hygrophyte it makes its home by ponds and -rivers. In the case of a tropical littoral flora, such as we find in a -Pacific island, the large proportion of plants with buoyant fruits or -seeds gives so much prominence to the subject of their distribution by -currents that the question of “station” is often masked. On the other -hand, in the shore-flora of a temperate region like that of Great -Britain, the plants with buoyant seeds or fruits are in the minority, -and the question of “station” is the first to obtrude itself. - -In establishing the principle that most of the plants with buoyant seeds -or fruits have been gathered at the water-side, it was never implied -that all the plants by the river or by the pond or at the coast are thus -characterised. There is much to learn from the circumstance that whilst -nearly all plants with buoyant seeds or fruits are placed at the -water-side, not all water-side plants have buoyant seeds or fruits. In -the first place, it is to be inferred in the light of what has been said -above that the first determining principle in the selection of a station -is concerned not with the buoyancy of the seeds or fruits, but with the -xerophytic or hygrophytic organisation of a plant. In other words, it is -the fitness or the unfitness of a plant for living in situations where -the loss of water by transpiration requires to be checked that primarily -determines the station at the coast. We thus see in the internal -organisation of the plant the primary determining influence on station. -Buoyancy of seed or fruit comes subsequently into play, the Xerophyte -and the Hygrophyte, thus endowed, ultimately finding their way, the -first to the beach, the second to the bank of the river or to the margin -of the lake or pond. - -In the next place, when we regard the composition of the British -coast-flora, and examine the distribution of the plants in other -situations than on the beach, we obtain some interesting results. There -is first a group of plants, including such as Armeria vulgaris, -Artemisia maritima, Cochlearia officinalis, Erodium maritimum, -Matricaria inodora, Plantago coronopus, Polycarpon tetraphyllum, -Raphanus maritimus, Spergularia rubra, Silene maritima (see Note 15), -and others, all of which occur not only at the coast and on the adjacent -hill-slopes, but also often far inland, and sometimes at considerable -elevations in mountainous districts, as in Central Europe. It is on this -occurrence of certain shore-plants in alpine regions that Prof. Schimper -lays much stress in his memoir on the Indo-Malayan strand-flora (p. 28), -and in his later work on Plant Geography (Engl. edit., p. 716), when -pointing out that here temperature does not play a determining part, and -that in both stations, whether on the sandy beach or on the -mountain-top, the same xerophilous organisation is needed to obviate the -risk of impeded water-supply. He quotes in this connection the -observation of Battandier that many alpine species from the Atlas -Mountains occur on the Algerian beaches, but not in intervening regions. -Mr. Druce, in his discussion of the British species of Sea-Thrifts and -Sea-Lavenders (Armeria, Statice), brought the subject of the occurrence -of maritime plants on mountain summits again to the front; but he did -not advance any general explanation, and seems to regard it as the -result, as it doubtless is, of the recurrence of suitable stations -(_Jour. Linn. Soc. Bot._, Dec. 1900). - -Very few of these plants have any capacity for dispersal by currents, a -subject dealt with in Note 16. Several of them have dehiscent, -small-seeded fruits which, as pointed out in the previous chapter, -hardly ever come into the buoyant category. I have experimented on the -greater number of them, and in only one species, Matricaria inodora -(var. maritima), do the results indicate a capacity for dispersal over -wide tracts of sea. - -If we look again at a list of British shore-plants, we find another -group of plants frequenting salt marshes and muddy shores, and found -also often far inland, as in the saline plains of Central Asia. Here we -have such plants as Aster tripolium, Glaux maritima, Plantago maritima, -Salicornia herbacea, Salsola kali, Samolus valerandi, Scirpus maritimus, -Suæda fruticosa, S. maritima, Triglochin maritimum, T. palustre, &c. It -becomes in this connection a subject of peculiar interest to the student -of plant-distribution when he reads in Mr. Hemsley’s paper on the flora -of Tibet (_Jour. Linn. Soc. Bot._, vol. 35) that amongst the British -shore-plants above-named the two species of Triglochin and the same -species of Glaux and Salsola occur in the salt marshes of the Tibetan -uplands at elevations of 15,000 to 16,000 feet, Scirpus maritimus also -being found in the swamps of the lower levels. We have the same thing, -affecting much the same plants, illustrated in America. Thus we learn -from Asa Gray that Salicornia herbacea, Scirpus maritimus, Triglochin -maritimum, &c., which are common in salt marshes on the coast of the -United States, occur also in the interior of the continent in the -vicinity of salt-springs. - -Facts of this sort are well known, and I merely refer to them here in -order to emphasise the importance of this little group of British -littoral plants, those of the salt marsh. Their very wide distribution -is connected with the frequent recurrence of suitable conditions, not -only in space, but what seems of greater import, also in time. One can -scarcely doubt when the Saltwort (Salsola kali) is seen on the -Devonshire coast, on a beach in Chile, and in the elevated regions of -Central Asia that here a very ancient type of plant finds its still more -ancient conditions of existence. In the capacity which most of the -plants of the salt marsh possess of germinating in sea-water, this group -of littoral plants is sharply distinguished, as far as my observations -show, from the other groups of British shore-plants. For instance, in my -experiments the seeds of Aster tripolium, Salicornia herbacea, and -Triglochin maritimum germinated freely in sea-water, whilst those of -Spergularia rubra, Cakile maritima, Convolvulus soldanella and others -failed to do so (see Note 19). It will also be noticed with respect to -this group of littoral plants that, except in the case of Scirpus -maritimus, the seeds or fruits have little or no floating power, the -exception offered by Salsola kali being not very striking. This feature -is brought out in the Table given in Note 10; but some of the details of -my observations are given in Note 17. - -There yet remains a third group of the British shore-plants, namely, -that comprising the plants that rarely stray far from the beach and -often possess seeds or seedvessels that will float for months. Here we -have such species as Arenaria (Honckeneya) peploides, Beta maritima, -Cakile maritima, Crambe maritima, Crithmum maritimum, Convolvulus -soldanella, Eryngium maritimum, Euphorbia paralias, Glaucium luteum, -Lathyrus maritimus, Polygonum maritimum, &c. The seeds or seedvessels of -quite half of these species will float for months unharmed in sea-water, -but in a few, as with Cakile maritima and Eryngium maritimum, they float -for only a week or two, whilst in others again like Glaucium luteum they -have no buoyancy. (Some details of the buoyancy experiments on these -plants are given in Note 18; and the long list in Note 10 may be first -consulted.) - -It is not necessary to enter here into more detail with respect to -British shore-plants. Enough has been said to disclose cleavage-lines in -what might have appeared as a homogeneous plant-formation. We can thus -discern the elements of at least three groups amongst the plants of our -beaches, each group bearing the impress of an independent history:— - -(_a_) The plants of the beach and of the inland plain or of the distant -mountain peak, excluding those of the salt marshes. Armeria vulgaris, -Silene maritima, and Spergularia rubra may be taken as examples. The -currents here as a rule take little or no part in their dispersal. - -(_b_) The “saline” group, including the plants of the saline plains and -the salt marshes of the interior of continents. Of these Glaux maritima, -Salsola kali, and Triglochin maritimum are examples. The capacity of -germinating in sea-water is a distinguishing character of most of the -plants; and but few of them possess seeds or seedvessels that are -markedly buoyant. - -(_c_) The true beach plants that rarely stray far from the beach, of -which Arenaria peploides, Cakile maritima, and Convolvulus soldanella -are examples. Many of them have buoyant seeds or fruits capable of -dispersion over wide areas through the agency of the currents. - -The reader will be able to extend this subject for himself if he is so -inclined, but we have gone far enough together to learn that the plants -with buoyant seeds or fruits are in the minority on our beaches, -scarcely a third of the total being fitted for dispersal by the currents -over broad tracts of sea. The British strand-flora thus differs -strikingly from the littoral flora of a Pacific island, or indeed of any -ordinary tropical coast, and in this respect it is to be regarded as -typical of the temperate regions. It has been remarked before that on a -beach in the tropics we would expect to find that quite three-fourths of -the plants are provided with buoyant fruits or seeds distributed far and -wide over the tropical seas by the currents. - -We pass on now to briefly discuss from the same standpoint the British -plants that find their homes on the borders of rivers and ponds. It is -here that the hygrophytes with buoyant seeds or fruits gather together, -just as the xerophytes with similar seeds or fruits collect on the -beaches. We have seen before that only a portion of the beach plants -belong to the buoyant group, and the same applies to the plants at the -edges of rivers and ponds. The plant-formation is no more homogeneous -there than it is in the case of the strand-flora. Let us see if we can -discern some lines of division there also, or in other words let us -endeavour to connect the absence or presence of floating power in the -fruits and seeds with some variations in the placing of the plants. We -still pursue the clue to the study of the complicated problems connected -with plant-stations by taking the floating seed as our guide. - -We will carry ourselves in thought to the Thames-side between Teddington -and Twickenham at the end of August, 1892. The river is at the -high-water level, and we see flourishing at the margins, sometimes a -little above the water and sometimes a little within its reach, -Ranunculus repens, R. sceleratus, Spiræa ulmaria, Lycopus europæus, -Scutellaria galericulata, different species of Rumex, Alnus glutinosa, -Iris pseudacorus, Sparganium ramosum, and different species of Carex, -with several other plants, all contributing their seeds or fruits to the -drift that floats in the river from the autumn to the spring. - -But besides these plants there are a number more or less submerged in -the stream, including Nasturtium amphibium, N. sylvestre, Stellaria -aquatica, Myosotis palustris, and Veronica beccabunga; and as the water -falls other plants still more submerged come into view on the exposed -flats, such as Nasturtium officinale, Apium nodiflorum, and Polygonum -hydropiper. None of these plants are represented by their seeds or -fruits in the floating river-drift. Several of them possess dry -dehiscent fruits with small seeds, such as Nature ignores in the matter -of buoyancy, and the small fruits of Myosotis, Apium, and Polygonum have -little or no floating power. - -We have thus here a clear dividing line between the plants with buoyant -seeds or fruits that were more or less exposed above the high-water -level, and those that were more or less submerged at that state of the -tide. That which occurs in the Lower Thames twice in the day within the -reach of the tide represents what happens in the higher part of the -river during the seasonal floods, but in the last case the effects -cannot be so readily distinguished. We thus perceive that the buoyant -seed or fruit is as a rule only characteristic of the plants of the -river-side that grow more or less exposed above the water, whilst those -plants liable to periodic submergence have seeds or fruits that sink. - -In this connection it is of especial interest to observe that as a -general rule the truly aquatic plants of English rivers contribute -little or nothing to the floating seed-drift. I pointed this out several -years ago, in my paper on the Thames, as an agent in plant-dispersal, -and it has been already noticed in this work (page 30). We look in vain -amongst the floating winter drift of our rivers for the seeds or fruits -of Ranunculus aquatilis, Nuphar luteum, Nymphæa alba, and of the species -of Myriophyllum, Limnanthemum, Callitriche, Ceratophyllum, Zannichellia, -and of several of the Potamogetons, all of which give character in -summer to the aquatic vegetation of the river. In their place we find -only the seeds and fruits of the plants growing on the banks. - -There is, however, another small group of river plants, which in their -structure and habits and in the behaviour of their floating fruits come -between the true aquatics and the plants of the river-banks. They belong -mostly to the Alisma family, and Alisma plantago and Sagittaria -sagittifolia may here be specially mentioned. Their fruits display great -variation in their floating power; and on this point M. Kolpin-Ravn, -writing to me in 1895, made the following interesting suggestion, that -since these plants approach true aquatics in structure they may be also -regarded as approaching them in the inconstancy of the buoyant capacity -of their fruits, those of aquatics having typically little or no -floating power. - -Seed-buoyancy, however, does not play quite such an important part in -the plant-economy of a river as the examination of the floating drift -would lead one to expect. Only a portion of the bank-plants have buoyant -seeds or fruits, whilst amongst the true aquatics, the semi-aquatics, -and the plants periodically submerged, the rule of non-buoyancy -prevails. And, indeed, when we look at all the possible stations for the -plants of the British flora, we discover that seed-buoyancy can rarely -be connected with station. It is, however, in those few stations that -plants with buoyant seeds have mainly gathered. There it is, probably, -that the remnants of a past floral age find a refuge, since it would -seem likely that the tendency has been in the course of geological time -for the development of dry stations for plants at the expense of the wet -stations. - -The following is a summary of some of the points discussed in this -chapter:— - -(1) In the case of the strand-flora of a Pacific island, and indeed in -that of an ordinary tropical region, the large proportion of plants with -buoyant seeds or fruits tends to mask all other issues, and we are -seemingly only concerned with dispersal by currents. - -(2) But in the British strand-flora where plants with buoyant seeds and -fruits are in a minority, constituting less than a third of the total, -it is seen that the issue is primarily an affair of station, an -inference that may be applied generally to temperate regions. - -(3) All British shore-plants may be regarded as owning certain -characters in common which may be collectively designated the -xerophilous habit, and we may extend this view to other temperate -strand-floras. - -(4) But this xerophilous habit is also characteristic of inland plants -in certain localities, as of those of the steppe, the desert, the rocky -mountain-top, and of other exposed situations, in all of which checks to -the loss of water by transpiration are required. Whilst the risks of -drought are thus guarded against in the case of plants stationed in arid -localities, the risk of injury to the plant from the accumulation of -salt in the tissues is obviated in the instance of the plants of the -coast. - -(5) On the other side we have the hygrophilous habit characteristic of -plants living under conditions where checks to transpiration are -relatively little needed. All the plants of the margins of rivers and -ponds belong here, and indeed all plants living under moist conditions. - -(6) This distinction between the xerophilous and hygrophilous habits -penetrates deeply into all questions connected with stations, and lies -behind all matters relating to the buoyancy of seeds or fruits. It is -the fitness or unfitness of a plant for living in dry situations that -primarily determines the station. If a xerophilous plant has a buoyant -seed or seedvessel it finds its way ultimately to the coast; if it is -hygrophilous and its seeds or fruits can float, then it is finally -established on the side of a pond or river. - -(7) The composite character of the British strand-flora is to be -explained on the above principles. We have in the first place the plants -confined to the sandy beach, many of which possessing buoyant seeds or -fruits are dispersed by the currents. Next come the plants of the sandy -beach which are found also far inland in open plains and on -mountain-tops; and afterwards come the plants of the salt-marsh and -mud-flats of the coast, which appear again in the saline plains and -swamps in the interior of the continents. - -(8) The plant-formation of the river’s border displays also lines of -division, and is by no means homogeneous; and indeed other factors -besides those connected with seed-buoyancy have here been in operation. - -(9) In only a few of the possible stations of British plants can a -direct connection be traced with seed-buoyancy. Yet it is at these few -stations, such as at the coast and by the pond or river, that the plants -with buoyant seeds and fruits have mainly gathered. - -(10) The plants now frequenting wet stations may often be regarded as -the remains of an age when moist conditions for plant-life prevailed. - - - - - CHAPTER V - - THE FIJIAN STRAND-FLORA - -The inland extension of the beach plants.—The grouping of the coast - plants.—Their modes of dispersal.—The zone of change.—Summary. - - -HAVING learned from the British flora the real significance of the -buoyant seed or fruit in a littoral flora, we will now return to the -Pacific and proceed to deal with the composition and general character -of the strand-plants. - -Speaking of the Malayan strand-plants, Professor Schimper remarks (pp. -11, 12) that both in outward appearance and in anatomical structure they -are xerophilous in character, whether in the case of those of the -mangrove-swamp or in those of the beach. Since the tropical shore-flora -of the Pacific islands is essentially Malayan, the identity usually -extending to the species, the same conclusion may be applied to its -character. The xerophilous habit may show itself externally in a variety -of ways, as in hairiness, leaf-structure, a leathery cuticle, -succulency, &c. - -From this xerophilous habit of the Pacific strand-flora we should expect -to find that many of the plants stray far from the coast, wherever the -suitable conditions for their type of organisation occur, whether in the -inland plain or on the mountain-top. This is indeed the case; but in -dealing with this subject it will be necessary to discuss in some -general detail the littoral floras of the Fijian, Hawaiian, and Tahitian -groups in succession. - - - THE FIJIAN STRAND-FLORA - - THE INLAND EXTENSION OF THE BEACH PLANTS - -Viewed from the old standpoint of “station,” where one would distinguish -sharply between the coast and the inland plants, the Fijian strand-flora -exhibits a number of inconsistencies, all at first sight extremely -puzzling. When, however, we regard their xerophilous character and -reflect that this habit, and not mere fitness for growing at the coast, -is the primary determining factor of their station, much that is strange -appears normal and plain. - -Let me refer in this connection to the impression that the distribution -of the Fijian shore-plants made on Mr. Horne, the director of the -Botanic Gardens of Mauritius, who spent a year in the botanical -investigation of the group about a quarter of a century ago. In his -account of the group (pp. 59, 60) he says that several of “what are -known as sea-shore plants” are found far in the interior of the larger -islands; and amongst others he names such characteristic beach plants as -Cerbera Odollam, Hibiscus tiliaceus, Ipomœa pes capræ, and Pandanus -odoratissimus. On the other hand, he remarks that several species of -inland plants occur at the coast, and that several plants growing on the -mountain-tops are found near the sea. This apparent confusion of station -he seems to attribute to the circumstance that the mountains of Fiji are -not high enough for the development of an alpine flora. But such a view -could not be held now, since the effect of an alpine flora would be the -introduction of further elements of confusion in the occasional -occurrence of some of the alpine plants on the sea-coast, as we find in -Hawaii. - -Yet this apparent mingling of the littoral and inland floras in Fiji -becomes intelligible when we perceive that the seeming confusion of -station is mainly restricted to the xerophilous plants of the arid -inland plains and of the bare mountain-tops. The rank humid forests that -cover so much of the interior of the islands, and the luxuriant -vegetation of the mountain-gorges, are not here concerned. Such a -mingling occurs it is true under certain conditions; but in the general -physiognomy of the flora the distinction between the shore and inland -plants holds good. The same shore plants that are distributed far and -wide over the Pacific here present themselves; and although some of them -extend far inland, where the scantily-vegetated plains descend to the -coast, this does not deprive them of the right of being still regarded -as littoral plants. - -Still, when we look at a fairly complete list of the shore-plants of -Fiji, numbering in all about eighty, we perceive that about two-thirds -of them also occur inland, either in Fiji or in some other tropical -region; and if we reflect that many of the residue are plants of the -mangroves that would not be found inland except under estuarine -conditions, it becomes evident that with this reservation there are very -few littoral plants in Fiji that do not at times leave the coast. - -Cæsalpinia Bonducella may be taken as a type of those shore-plants that -stray far away from the coast, even into the interior of continents, -since in India it reaches the Himalayas. Although Terminalia Katappa and -Calophyllum Inophyllum often owe their existence inland in different -parts of the tropics to man’s agency, this cannot be said of most -others, as Cassytha filiformis, Casuarina equisetifolia, Cycas -circinalis, Ipomœa pes capræ, Pandanus odoratissimus, Premna tahitensis, -Tacca pinnatifida, Tephrosia piscatoria, Vitex trifolia, &c., when they -occupy the extensive inland plains that slope to the coasts on the lee -sides of the large islands of Fiji. Plants, like Hibiscus tiliaceus, are -found in a Pacific island almost as frequently away from the beach as on -the beach itself; and this is true of most other regions of the tropics -where it occurs. - -Other plants that appear to be altogether confined to the sandy beach in -Fiji, break away on rare occasions from their usual station and appear -on the bare rocky summits of hills near the coast, even though the -hill-slopes are densely wooded. On such bare hilltops in Vanua Levu, -varying from 500 to 1,100 feet in elevation, one is surprised at times -to find shore creepers and climbers like Canavalia obtusifolia and -Derris uliginosa associated with other beach-plants more frequently -found inland, such as Tephrosia piscatoria and Vitex trifolia, and in -the company of climbing species of Morinda and of small trees of Fagræa -Berteriana. When the “talasinga” (sun-burnt) districts, as the Fijians -term the plains on the north sides of the islands, extend a long -distance from the coast into the heart of the island, they carry with -them their peculiar vegetation and the intruding beach-plants up to -considerable elevations above the sea. We then find familiar -beach-plants like Cerbera Odollam and Ipomœa pes capræ growing far -inland at heights of 1,000 feet and over above the sea. (See Notes 20 -and 21.) - -One is never quite sure of the behaviour of shore-plants in Fiji when -the “talasinga” plains lie behind the beach, since even Scævola Kœnigii, -usually a steadfast beach-plant, occurs at times some miles inland. (See -Notes 20 and 55.) There are, however, a few that never came under my -notice inland, such as Pemphis acidula, Triumfetta procumbens, and -Tournefortia argentea. The extension of sea-coast plants for any -distance inland depends a good deal on the occurrence of -scantily-vegetated plains, or of scrub-covered, rolling country at the -back of the beaches; and doubtless that which I have described in the -case of Fiji is to be found in other tropical coast-regions. Professor -Schimper informed me by letter that he had noticed a similar inland -extension of the shore-plants in the Seychelles.... I have only here -touched on this subject. In Notes 20 and 21 the reader will find further -details of the inland extension of the beach-plants, and in Note 22 is -given a general account of the “talasinga” plains, in which the -wandering beach-plants mingle with the peculiar vegetation of the plains -themselves. Covered with reeds and bracken, and dotted over with clumps -of Casuarinas and Acacias, with the Cycad and Pandanus distributed -irregularly over their surfaces, such level districts possess, as -remarked by Seemann, a South Australian look. - - - THE GROUPING OF THE FIJIAN LITTORAL PLANTS. - -The littoral plants readily divide themselves into three principal -groups as concerning their station, namely: - -(_a_) The “beach-formation,” typically exhibited on the whitish -calcareous beaches of reef-bound coasts. - -(_b_) The “mangrove-formation,” found at intervals all along the coasts, -but most fully developed at the estuaries, and for the most part -occupying flats regularly overflown by the tide. - -(_c_) The “intermediate formation,” comprising the plants of the tracts -between the beach and the mangrove-swamp and at the borders of the -swamps. - -This grouping does not differ materially from that adopted by Professor -Schimper in the instance of the Indo-Malayan strand-flora. (See Note -23.) - -To the beach-formation belong, amongst the trees and shrubs, -Barringtonia speciosa, Calophyllum Inophyllum, Guettarda speciosa, -Pemphis acidula, Scævola Kœnigii, Tournefortia argentea, &c., and -amongst the creepers and procumbent plants, Canavalia obtusifolia, -Ipomœa pes capræ, Triumfetta procumbens, &c. To the mangrove-formation -belong the Asiatic and the American species of Rhizophora, and species -of Bruguiera, Carapa, Lumnitzera, &c. Amongst the trees that gather -around the borders of the mangrove-swamp, constituting the intermediate -formation, occur Barringtonia racemosa, Excæcaria Agallocha, Heritiera -littoralis, Hibiscus tiliaceus, and several other species, all of them -being equally at home on the sandy beach, at the border of a -mangrove-swamp, and on the banks of an estuary. The climbers, such as -Entada scandens, Mucuna gigantea, Derris uliginosa, &c., belong more to -the mangrove and to the intermediate formations than they do to that of -the beach. Referring the reader to the more complete lists given in Note -24, I may remark that it is not always possible to distinguish sharply -between the three formations, since some of the plants belong to two, -and a tree like Cerbera Odollam may, in different localities, be -referred to all three formations. The general distinction, however, -prevails in the physiognomy of the coast-flora. - -The mangrove-formation comprises, it may be pointed out, many plants -other than mangroves, plants that find a home in the mangrove-swamps of -Fiji, either within their limits or at their borders. It presents, -indeed, a world in itself. When the mangroves establish themselves in a -new locality they carry along with them a host of hangers-on, both -plants and animals, that only find a home under the favourable -conditions of a mangrove-swamp. Thus, the absence of the -mangrove-formation from a Pacific island deprives its littoral flora of -many very striking features. For this reason the Tahitian shore-flora -must seem to a botanist coming from Fiji comparatively tame and -monotonous; whilst that of Hawaii, for this and for other reasons to be -subsequently mentioned, is still less interesting, and scarcely even -gives a character to the coasts. - -We are now, therefore, prepared to learn that a large number of the -plants other than true mangroves, that thrive in or around the Fijian -mangrove-swamp, are not to be found in those Polynesian islands where -true mangroves do not exist; and that a law of association here exists. -Many of the plants of the intermediate formation are so closely bound up -with the mangroves in their life-conditions that they are not to be -found where the mangroves are absent, even though their seeds or fruits -are pre-eminently fitted for dispersal by the currents. The influence of -“station” here rules supreme. This matter will be treated more in detail -when discussing the Tahitian and Hawaiian strand-floras in Chapters VI. -and VII. - - - THE MODES OF DISPERSAL OF THE FIJIAN STRAND-PLANTS. - -The predominant influence of the currents having been already -established, there remains for consideration the distribution of the -floating capacity of the seeds or fruits among the different formations. -One can say that almost without exception the seeds or fruits or -seedlings of the mangrove and intermediate formations float for long -periods. In the case of some of the true mangroves, as in Rhizophora and -in Bruguiera, where germination takes place on the tree, it is the -seedling that floats, whilst in others, as in Carapa and Lumnitzera, it -is the seedvessel that floats. The plants with non-buoyant seeds or -fruits that belong to the littoral flora are all confined to the beach -formation, but they do not form more than a sixth of the total. Almost -all the “good floaters” of the beach-plants are widely spread over the -shores of the Pacific and of much of the tropics, and include such -familiar species as Barringtonia speciosa, Cæsalpinia Bonducella, -Terminalia Katappa, and many others mentioned in the lists of Notes 2 -and 24. - -When, however, we come to the dozen or so of beach-plants that possess -seeds or fruits with little or no floating power, we find that several -of them have a limited distribution in the Pacific, such as Acacia -laurifolia, Drymispermum Burnettianum, Eugenia Richii, &c., whilst -others, such as Casuarina equisetifolia, Tephrosia piscatoria, -Triumfetta procumbens, and Wikstrœmia fœtida, are widely spread. This -small non-buoyant group of the beach-plants has a nondescript -appearance, and it is here that the inland flora is most likely to make -its influence felt by additions to the number. It is here indeed that -the littoral floras of the tropics mostly differ, the accessions from -the inland flora varying in each region. It is in fact the zone of -change. - -A number of these plants, such as the species of Drymispermum, Eugenia, -and Wikstrœmia, have probably been dispersed by frugivorous birds; -whilst others, like Triumfetta procumbens, possess fruits that might -have been transported in birds’ plumage. From the frequency with which -Tephrosia piscatoria is associated on hilltops in Fiji with Fagræa -Berteriana and climbing species of Morinda that are well suited for -dispersal by frugivorous birds, it seems likely that it is also -distributed by birds fond of a drier diet. It is possible that the -Polynesians, who much value the wood of Casuarina equisetifolia, have -often assisted in dispersing the tree. - -The following is a summary of the contents of the chapter. - -(1) The extension inland of the Fijian strand-flora is to be attributed -to the xerophilous organisation of the plants, and to the exceptionally -favourable conditions that are offered to such plants on the plains, and -in other scantily vegetated localities, lying usually on the drier sides -of the larger islands. - -(2) Excluding the mangroves and the plants associated with them in the -coast-swamps, there are few littoral plants of the islands of the -tropical Pacific that do not extend inland in one region or another. - -(3) The Fijian shore-plants can be rudely arranged in three groups, -those of the mangrove-swamp, those of the sandy beach, and those of the -intermediate districts, the last including those plants that occur -typically at the borders of a mangrove-swamp, though some of them can -thrive equally well on a beach. - -(4) There is a law of association connecting many plants with a -mangrove-swamp in such a manner that when the true mangroves are not -represented in a Polynesian group, as in Tahiti or in Hawaii, the plants -in question are also absent, notwithstanding that in many cases, such as -those of Clerodendron inerme and Heritiera littoralis, they possess -seeds or seedvessels of great floating power. - -(5) The fruits or seeds or seedlings, as the case may be, of the plants -of the mangrove-swamp and of the bordering districts float almost -without exception for long periods. This is true also of five-sixths of -the beach-plants, whilst the remainder owe their dispersal chiefly to -birds. - -(6) The small non-buoyant group of the beach-plants represents that -portion of the strand-flora that is most likely to be recruited from the -inland flora. It is here that exists the zone of change; and it is in -this respect that the littoral floras of the tropics differ principally -amongst each other, the recruits from inland varying naturally with the -floras of different regions. - - * * * * * - -Though it does not come within my plan to discuss the littoral floras -of the adjacent smaller groups of Tonga and Samoa, it may be remarked -that they reflect most of the principal features of the strand-flora -of Fiji. In particular it may be observed that they possess the -mangrove-formation, but to a more limited extent. Both own the -mangrove genera Rhizophora and Bruguiera, whilst Carapa is also found -in Tonga. The intermediate formation is represented in Tonga by -Clerodendron inerme, Excæcaria Agallocha, and Heritiera littoralis; -whilst in Samoa we find, besides the first-named species, Barringtonia -racemosa and Scirpodendron costatum. In both the beach-formation is -well represented. - - - - - CHAPTER VI - - THE TAHITIAN STRAND-FLORA - - (_From materials supplied mainly by the work of Drake del Castillo_) - -Lacks the mangroves and their associated plants.—Possesses mainly the - plants of the coral beach.—Predominant agency of the currents.—Inland - extension of shore-plants.—Summary. - - -JUST as the littoral plants of Fiji may be regarded as typical of -Western Polynesia, so the strand-flora of Tahiti, or, rather, of the -Tahitian Islands, may be considered as representing Eastern Polynesia. -We have thus the Tahitian area, comprising generally the Cook and -Austral Groups, the Society Islands, the Paumotus, and also the -Marquesas, as contrasted with the Fijian area, including the -neighbouring Samoan and Tongan groups. For the sake of brevity the terms -Fiji and Tahiti are often used as equivalents of the entire areas (see -Note 25). - -The littoral flora of this part of the Pacific lacks the mangroves and -most of the plants that are associated in the Fijian region with a -mangrove-swamp, either at its borders or within its interior. Thus we -miss here the true mangroves of the genera Rhizophora, Bruguiera, -Carapa, and Lumnitzera, as well as the accompanying trees and shrubs, -such as Barringtonia racemosa, Excæcaria Agallocha, and Heritiera -littoralis. The climbers and straggling plants that are so -characteristic of the borders of the mangrove-creeks in Fiji proper are -also wanting, such as Clerodendron inerme, Derris uliginosa, and Smythea -pacifica; and we do not find in the Tahitian region the Giant-Sedge -(Scirpodendron costatum) that is so common in the mangrove-swamps of -Fiji, and occurs also in Samoa. - -It is not at first sight easy to account for the absence from Tahiti of -the mangrove-formation and of so many of the plants that grow at the -borders of a mangrove-swamp in Fiji. Their absence can scarcely be due -to the want of suitable stations, as is indicated by the common -occurrence in the Tahitian coast-marshes of Chrysodium aureum, the Great -Swamp-fern, that not only abounds in the mangrove belts of Fiji, Tonga, -and Samoa, but is associated with mangrove-swamps over much of the -tropical zone. Nor can it be said that the currents are ineffective, or -that the seeds or fruits of the missing plants possess, as a rule, -insufficient floating powers. Most of the plants of the Tahitian beaches -hail, like those of Fiji, from Malaya, and have been brought through the -agency of the currents; and many of the absent littoral plants that have -the same home, such as Heritiera littoralis and Clerodendron inerme, -have fruits or seeds just as capable of floating unharmed over the same -extent of ocean. It is not any defect in floating-power that has -prevented the establishment of two such plants in the Tahitian area. -Entada scandens, which in some parts of the world is a typical climber -of the mangrove-formation, and in other places thrives well in the -absence of mangrove-swamps, has only been recorded from Rarotonga in -this region by botanists, but I believe Wyatt Gill refers to its -occurrence in Mangaia in one of his books. - -On the other hand, it is likely that the floating seedlings of -Rhizophora and Bruguiera, which represent the only means of dispersal by -the currents at the service of these mangroves, would not arrive at -Tahiti in a condition favourable for the establishment of the plants. My -observations, which are described in Chapter XXX., go to show that, -though the seedlings will float uninjured in still sea-water for months, -they will not withstand prolonged sea-buffeting. These two genera of -mangroves, it is most important to remember, supply the pioneers and the -principal components of a mangrove-swamp in the Western Pacific. Where -they fail to establish themselves, the requisite conditions for the -large number of plants and animals that find their home in and around a -mangrove-swamp would not be provided. We thus perceive that the absence -from the Tahitian coast flora of several plants that are associated in -Fiji with the mangrove-swamps depends on a law of association, which has -already been referred to in the preceding chapter, and is not concerned -with incapacity for dispersal by currents (see Note 26). - -Whilst the Tahitian coast flora does not, therefore, possess the plants -of the mangrove-swamp and its vicinity, it includes most of the typical -beach-trees of the coral islands and reef-fronted coasts of other parts -of the South Pacific. Thus we find here on the sandy beaches -Barringtonia speciosa, Calophyllum Inophyllum, Cerbera Odollam, -Hernandia peltata, Guettarda speciosa, and numerous other plants that -are indicated by the letter T in the list of Fijian littoral plants -given in Note 2. The total number of Tahitian shore-plants is thus -considerably less than that of Fiji (there are about 55 in Tahiti and -about 80 in Fiji); but in its turn, as will subsequently be shown, it is -much larger than that of Hawaii, where the number is about 30. - -Quite three-fourths of the strand-flora of this region have buoyant -seeds or seedvessels capable of floating for long periods; and there is -no difficulty in assigning by far the greater share in the stocking of -the beaches with their plants to the agency of the currents. The -currents in their operations have indeed carried the fruits or seeds of -many of these plants across the South Pacific as far as the islands -extend, namely, to Ducie Island and to Easter Island. There are few more -significant proofs of the efficacy of the currents in distributing -plants over the Pacific than the discovery, by Mr. Arundel, of -Barringtonia speciosa in Ducie Island in association with Tournefortia -argentea (Challenger, Botany, III. 116). - -The residue of the Tahitian coast flora possessing fruits or seeds that -are unsuited for dispersal by currents includes such plants as -Heliotropium anomalum, Triumfetta procumbens, Tephrosia piscatoria, -Wikstrœmia fœtida, &c. The small nucules of the first-named are perhaps -dispersed by granivorous birds; the fruits of Triumfetta are probably -transported in birds’ plumage; those of Wikstrœmia are distributed by -frugivorous birds; and the seeds of Tephrosia may be dispersed like -those of Heliotropium. - -The recruits or intruders from the inland flora do not appear to be -numerous or to give any special character to the shore flora. (See Note -27.) - -From not having a personal acquaintance with this region it is not -possible for me to discuss the extension of the shore-plants inland -except in a general way. From the pages of the work of Drake del -Castillo we can, however, infer that several plants such as Cassytha -filiformis, Cerbera Odollam, Colubrina asiatica, Hernandia peltata, -Morinda citrifolia, and Pandanus odoratissimus have extended inland to -the mouths of the Tahitian valleys, and have ascended the lower slopes -of the hills that lie near the coast. Others, like Cæsalpinia Bonduc, -Gyrocarpus Jacquini, and Ochrosia parviflora, have climbed far up the -mountain-sides to elevations of from 2,000 to 2,400 feet above the sea. -It is also evident from Mr. Cheeseman’s memoir on the Rarotongan flora -that coast plants also stray inland in that island. In an island like -Rarotonga, where a sorry substitute for a mangrove-swamp exists in the -form of a few coastal muddy places occupied by Vitex trifolia and -Sesuvium Portulacastrum, Entada scandens takes to the hills; and thus it -is that in this island it is most abundant in the interior, climbing to -the tops of the highest trees and “covering acres of the forest with a -dense canopy of green.” - - - _Summary of the Chapter._ - -(1) The Tahitian region possesses most of the plants that frequent the -sandy beaches of the Pacific islands. - -(2) But it lacks the mangroves and the associated plants of the -mangrove-swamp. - -(3) It also wants many of the plants that grow in the vicinity of such -swamps. - -(4) But since the plants last-mentioned often possess fruits or seeds -capable of being carried great distances by the currents, their absence -is to be attributed to the necessary conditions being lacking on account -of the failure of the mangroves. - -(5) Most of the beach plants, however, owe their existence in this -region to the transport of their buoyant fruits or seeds by the -currents. - -(6) The negative features of the Tahitian strand-flora are mostly to be -connected with the absence of Rhizophora and Bruguiera, the pioneers of -the mangrove-swamp; and their absence is, in turn, to be attributed to -the inability of their floating seedlings to reach this region in a fit -condition for establishing themselves. - - - - - CHAPTER VII - - THE HAWAIIAN STRAND-FLORA - -Its poverty.—Its negative features.—Their explanation.—The subordinate - part taken by the currents.—The Oregon drift.—The inland extension of - the beach plants.—Summary. - - -COMPARED with the rich strand-flora of Fiji, that of Hawaii presents but -a sorry aspect. In the number of species (30) it does not amount to -half; whilst it lacks the great mangrove-formation and the luxuriant -vegetation accompanying it that gives so much character to the shores -and estuaries of Fiji. Strangely enough, it is also deprived of most of -the familiar trees that, whether in foliage, in flower, or in fruit, -form the chief attraction of the sandy beaches of the Pacific islands. - -Neither the mangroves, therefore, nor the plants of the intermediate -formation, are to be found in Hawaii; and when we reflect that the -absentees from the beach formation include most of the trees, under the -shade of which the visitor to the Pacific islands can nearly always find -protection from the fierce rays of a tropical sun, it cannot be a matter -of surprise that this littoral flora has such a poverty-stricken -appearance. We look in vain for such shady beach trees as Barringtonia -speciosa, Terminalia Katappa, and Hernandia peltata; and we are lucky if -we find some small trees under which we can obtain a scanty shade. - -I have been speaking, of course, of the indigenous shore-plants, those -that have arrived at these islands without the assistance of man. Yet it -must be added that the existing littoral flora does include some of the -missing indigenous trees, though rarely in any number. There is, -however, scarcely one of them that is regarded by Dr. Hillebrand as -having formed part of the original flora. That botanist would indeed rob -the present beach flora, scanty as it is, of most of its conspicuous -plants, as far as their claims to be considered indigenous are -concerned. Dr. Hillebrand indeed includes Calophyllum Inophyllum, -Hibiscus tiliaceus, Thespesia populnea, Morinda citrifolia, Cordia -subcordata, and Pandanus odoratissimus in the present Hawaiian flora, -and nearly all of them are to be found at times at the coast as well as -inland; but he regards all, excepting the last-named, as having been -introduced by the aborigines. I was not inclined at first to go quite so -far as Dr. Hillebrand in this direction; but he carefully considered the -case of each individual plant, and, remembering his sojourn of twenty -years in the islands, his authority cannot be lightly put aside. In the -list of Hawaiian strand-plants given in Note 28 there are several -species not always littoral in the group, but typically littoral in -other tropical regions. One species, Ipomœa glaberrima, Boj., has not -been recorded before from these islands. - -A strong reason in favour of the contention of this botanist is that all -the trees above-named are useful in some way to the natives; and, -indeed, when we look at the works dealing with the floras of the islands -of the South Pacific, we observe that in almost all the groups one or -other of these six trees bears the reputation of having been introduced -by the aborigines. All of them in their turn lose their fame as truly -indigenous plants in some group or other. The occurrence of two or three -useless South Pacific beach trees, that are known to be dispersed by the -currents, in the indigenous strand-flora of Hawaii, would go far to -invalidate Dr. Hillebrand’s argument, since the six trees in dispute are -also known to be dispersed by the currents. But such trees are not to be -found; and we look in vain for trees like Cerbera Odollam, Guettarda -speciosa, Gyrocarpus Jacquini, and Hernandia peltata, that are spread -far and wide over the beaches of the South Pacific. - -It is also of interest to notice how trees like Morinda citrifolia and -Terminalia Katappa, concerning the non-indigenous character of which -there can be but little doubt, are in our own day acquiring a littoral -station. The second is not even regarded by Dr. Hillebrand as having -been introduced by the natives, but is referred by him to the European -epoch. After having been extensively planted, it is now, as I found, -becoming a littoral tree on the coast of Oahu, and supplies its buoyant -fruits in a regular way to the beach drift. Its native name of Kamani is -merely that of Calophyllum Inophyllum. All the six trees in dispute are -known in Hawaii by the names by which they are distinguished far and -wide over the South Pacific, a fact of which the reader may satisfy -himself by referring to my paper on Polynesian plant-names. The -Hawaiians, when their ancestors abode in the South Pacific, must have -been well acquainted with one or other of the prevailing names of -Terminalia Katappa (Talie, Tara, &c.); but it had lapsed in the memory -of the race when the Europeans introduced the tree into Hawaii. - -It may be added in this connection that Dr. Hillebrand weakens his -argument by regarding Pandanus odoratissimus as of pre-aboriginal origin -or as truly indigenous. Like the other six trees in question, its fruits -are known to be capable of dispersal far and wide by the currents; and -if this species of Pandanus is indigenous, we are obliged to assume that -its fruits were first brought by the currents. That being so, we cannot -exclude the probability of the currents having been also effective with -several of the other plants regarded by Hillebrand as of aboriginal -introduction, more especially those with large fruits like Calophyllum -Inophyllum, and Cordia subcordata, where the alternative agency of -frugivorous birds would be impracticable, at least over a wide extent of -ocean. Pandanus odoratissimus is, as I venture to think, a tree that was -introduced ages since by the aborigines. Next to the Coco palm, few -trees have been more utilised by island-peoples, more particularly -perhaps in the ruder stages of their history. - -This point has been discussed at some length, because on the correctness -of Dr. Hillebrand’s view depends the explanation to be subsequently -given of the origin of the shore-flora of Hawaii. Though differing in -some details, my observations on the Hawaiian coast plants, which are -given in Note 29, tend to strengthen his contention. - -I now return to the consideration of some of the negative features of -the Hawaiian strand-flora, and will allude first to the absence of the -mangroves and of the numerous other plants that live in and around a -mangrove-swamp. This cannot be connected with a total absence of -suitable stations. Although it is true that there are but few large -rivers and but few suitable localities, yet such localities exist. The -shores of Hilo Bay might readily have been the home of a mangrove-swamp; -and one can point to different places on the coast of Oahu, such, for -instance, as Pearl Harbour, which in Fiji would have been occupied by a -luxuriant growth of mangroves. The same argument applies to the missing -beach trees, such as Barringtonia speciosa, Hernandia peltata, Guettarda -speciosa, &c., that adorn the beaches of many a coral island or of many -a coral-bound coast in the South Pacific. Although in a large island -like Hawaii with its lava-bound coasts but few white calcareous beaches -exist where we might expect to find such a flora, yet such beaches occur -wherever the scanty coral reefs are found off the coast; and it is just -in those localities, as is pointed out in the account of my observations -in Note 29, that the “plantes madréporiques” of the French botanists, -the plants of the coral atoll and of the reef-girt coast, make their -best endeavours to establish themselves. In other islands like Oahu, -where coral reefs are more developed, calcareous beaches are more -frequent, and there the few “madreporic” plants of Hawaii make a home. - -Nor can the deficiencies in the Hawaiian strand-flora be connected with -climatic conditions. That its meagre character cannot be so explained is -indicated by the manner in which the Indo-Malayan shore-plants have -pushed their way northward on the western side of the Pacific to the -Liukiu and Bonin Islands. Here in latitude 26-27° N. we find several -Fijian littoral trees and shrubs, such as Hernandia peltata, Pemphis -acidula, Pongamia glabra, Sophora tomentosa, Terminalia Katappa, -Tournefortia argentea, &c., that do not occur in Hawaii, although this -group is some degrees nearer the equator, namely, in latitude 19-22° N. -They are accompanied by the mangroves (Rhizophora, Bruguiera, &c.) in -strength as far as South Liukiu in latitude 25° N.; but we learn from -Dr. Warburg that the mangroves thin off further north, though they reach -to South Japan, where Döderlein found in latitude 32° N. solitary -examples of Rhizophora mucronata. These interesting facts of -distribution, which are taken from Schimper’s work on the Indo-Malayan -shore-plants (pp. 85, 90), show us that we can scarcely look to climatic -conditions for the explanation of the absence of mangroves and of many -other tropical littoral plants from Hawaii. We form the same opinion -when we regard the extension northward of the mangrove-formation on the -American coasts of the North Pacific Ocean. According to the account of -Dr. Seemann given in the “Botany of the Voyage of H.M.S. _Herald_,” the -mangroves with the coco-nut palm, and many other littoral plants common -on the western shores of tropical America, reach their northern limit a -little north of Mazatlan within the mouth of the Gulf of California in -latitude 24° 38ʹ N. The parallel of 25° N. latitude, as indicated in -Drude’s Atlas, probably represents the extreme northern limit, which is -thus five or six degrees north of the latitude of the large island of -Hawaii. - -Neither can the explanation be found in the deficient floating powers of -the seeds or seedvessels of many of the “absentees.” Those of -Barringtonia speciosa, Guettarda speciosa, Heritiera littoralis, the two -species of Terminalia, &c., possess great buoyant powers equal to, and -probably often exceeding, those of the plants that, like Ipomœa pes -capræ, have succeeded in establishing themselves in Hawaii. One has only -to look at the lists giving the results of flotation experiments in -Notes 2 and 3, in order to realise that there are very few of the -“absentee” littoral plants, the non-existence of which in Hawaii could -be attributed to deficient floating powers of the fruit or seed. Being -able to float unharmed for months, and in several cases even for years, -the seeds or fruits of the shore-plants unrepresented on the Hawaiian -beaches have been carried far and wide by the currents over the tropical -Pacific even to Ducie and Easter Islands, that is, as far as the islands -extend. - -The only plants about which one could express a doubt concerning their -ability to reach Hawaii through the agency of the currents, and to -establish themselves there, are the true mangroves of the genera -Rhizophora and Bruguiera. Since germination takes place on the tree, it -is only through the floating seedlings that they could reach these -islands; but, as shown in Chapter XXX., it is doubtful whether the -seedlings would be in a fit condition for reproducing the plant after -such a long oceanic voyage. If they had been as successful in -establishing themselves in Hawaii as they have been in the Liukiu -Islands, which lie in latitude a few degrees farther north, these two -species through their reclaiming agency would alone have prepared the -way for the whole mangrove formation. We have seen in the preceding -chapter that the absence of the mangrove formation from Tahiti appears -to be mainly due to the failure of the pioneer species of Rhizophora and -Bruguiera to establish themselves there. This evidently also applies to -Hawaii, the cause of their exclusion being connected neither with -climate nor with station, but as in Tahiti with the general unfitness of -the floating mangrove seedlings for crossing broad tracts of ocean -without injury to the growing plantlet. - -With regard, however, to the bulk of the “absentee” littoral plants, -those of the beach-formation, no such incapacity on the part of the -buoyant seed or fruit can be accepted. These plants, which have reached -Tahiti in numbers, have in the mass failed to reach Hawaii. It will, -therefore, be of interest to glance at the character of the fruits of -the “absentee” trees, which a traveller fresh from a visit to the coral -islands and reef-girt coasts of the South Pacific sadly misses on the -Hawaiian beaches. We notice in the first place that the absent trees, -such as Barringtonia speciosa, Cerbera Odollam, Guettarda speciosa, -Heritiera littoralis, Terminalia Katappa, &c., have large fruits which -could only have been carried to Hawaii by the currents, the agency of -birds being quite out of the question. On the other hand, almost all the -littoral plants of Hawaii, whether trees, shrubs, or herbs, which are -regarded as truly indigenous by Mann, Hillebrand, and other Hawaiian -botanists, have only small fruits or seeds available for dispersal, from -which the agency of birds cannot, on the point of size, be excluded. -Amongst these shore plants possessing buoyant seeds or fruits are -Cassytha filiformis, Colubrina asiatica, Ipomœa pes capræ, Scævola -Kœnigii, Vigna lutea, and Vitex trifolia; whilst amongst the plants with -non-buoyant fruits or seeds are to be reckoned Heliotropium anomalum, H. -curassavicum, Tephrosia piscatoria, Tribulus cistoides, &c. The seeds or -seedvessels of the plants of the buoyant group possess great floating -powers; and it seems at first sight scarcely credible that the currents -which have failed to establish Barringtonia speciosa, Guettarda -speciosa, and the other trees that through this agency have often found -a home on the remotest islands of the Pacific, should have succeeded in -the instances of plants like Scævola Kœnigii and Vitex trifolia. - -It would indeed almost seem that in nearly all cases where it would be -impossible in point of size for a bird to transport the fruit or seed of -a shore-plant to Hawaii, such a plant is not to be found in the -strand-flora of that group, even though it is well adapted for dispersal -by the currents. Many of the littoral trees missing from the Hawaiian -coast-flora, having large buoyant fruits, come into this category; and -grave suspicion is thus apparently cast on the agency of the currents in -the case of the plants with small fruits and seeds that really compose -the strand-flora, even when their capacity for sea-transport has been -well established by observation and experiment. The efficacy of the -currents would thus seem to be called into question for the whole -littoral flora of Hawaii. - -If, however, we were to adopt such a sweeping conclusion we should be -led into an error. It is pointed out in the following chapter that -nearly all these large-fruited beach trees that are found far and wide -over the South Pacific, but are absent from Hawaii, do not occur as -indigenous plants in America. If, therefore, the fruits of such Old -World littoral trees as Barringtonia speciosa, Cerbera Odollam, -Guettarda speciosa, Ochrosia parviflora, Terminalia Katappa, &c., that -could be dispersed only by the currents, have failed to reach Hawaii, it -is essential to remember that they have also failed to reach America. -This suggests that Hawaii may have received some of its littoral plants -from America through the agency of the currents; and it is shown in the -following chapter that, as a rule, when a South Pacific plant with -buoyant fruits or seeds is not found in America, it is equally absent -from Hawaii. The question thus acquires quite a different aspect, and we -shall accordingly have to regard tropical America in the next chapter as -a possible centre of diffusion of littoral plants over the globe, a -centre possibly as important as that connected with the tropics of the -Old World. - -Although, however, the currents have played a part in stocking the -Hawaiian beaches with their plants, their share in the work has been -unimportant, and the number of plants concerned is limited. If we take -away the seven or eight littoral plants introduced by the aborigines, as -well as the three endemic species as indicated in the list in Note 28, -and then remove from the residue the plants with small fruits or seeds -possessing little or no buoyancy, there remain only the following eight -species, the presence of which in Hawaii might be attributed to the -currents, namely, Cæsalpinia Bonducella, Cassytha filiformis, Colubrina -asiatica, Ipomœa glaberrima, Ipomœa pes capræ, Scævola Kœnigii, Vigna -lutea, and Vitex trifolia. Of these plants, three species, those of -Cassytha, Scævola, and Vitex, possess fruits that would be likely to -attract frugivorous birds, and are in some cases known to be dispersed -by them (see Chapter XIII.); so that we are not in these instances -restricted to the agency of the currents. With the other five the -currents offer the readiest explanation, but, as is indicated in the -cases of Cæsalpinia Bonducella and Ipomœa glaberrima (Chapter XVII.), it -is quite possible that birds have occasionally intervened. Altogether we -may infer that in stocking the Hawaiian beaches with their littoral -plants the currents have taken a subordinate part. - -Coming to the Hawaiian littoral plants having seeds or fruits that have -no floating power, we find that they present a motley group. It has been -already remarked that this is the group of shore plants that derives -most recruits from the inland flora, and that it is in this group that -the differences between the shore-floras of tropical regions find their -expression. Yet a very odd collection of plants is here exhibited. -Sometimes the beach-flora is composed in great part of these plants; and -a sorry spectacle is presented by a beach possessing such plants as -Gossypium tomentosum, Heliotropium anomalum and H. curassavicum, -Lipochæta integrifolia, Tephrosia piscatoria, Tribulus cistoides, &c. -Yet to the student of plant-distribution such a motley collection would -be full of suggestiveness. From the circumstance that species of -Cuscuta, Jacquemontia, and Lipochæta, that are peculiar to the Hawaiian -Islands, have made their homes on the beach, he would infer that since -Nature has been compelled to borrow from the endemic inland flora, there -has been some difficulty in stocking the beaches with their plants. The -occurrence of endemic species amongst the strand-plants would be viewed -by him as especially indicating incapacity on the part of the ocean -currents. - -Yet in the quantities of drift timber, showing evidence of many months -and probably even of years of ocean-transport, to be seen stranded on -the weather coasts of these islands, the observer discerns undoubted -evidence of the efficacy of the ocean currents. But what he finds are -huge stranded pine logs of “red-cedar” and “white-cedar” from the -north-west coasts of America. He may search the drift for days together, -as I have done, and discover no tropical fruits or seeds except such as -could be supplied by the present Hawaiian flora. The subject of this -drift is especially discussed in Note 30; and it need only be mentioned -here that it is not improbable that, as shown in the next chapter, some -drift may reach Hawaii from tropical America under exceptional -conditions, and that its presence is masked by the Oregon drift. - -The agency of the drifting log in carrying small seeds in its crevices -would be effectual in the instance of plants from the temperate coasts -of North America. For example, the nutlets of Heliotropium curassavicum, -which have no buoyancy, might easily be washed, together with sand, into -the cracks of a pine log stranded temporarily on the Oregon coast where -this plant occurs. The _modus operandi_ was brought home to me when -examining the drift brought down by the Chancay River on the coast of -Peru. Here I found this species of Heliotropium growing on the margin of -a swamp near some stranded logs, that would probably be carried out to -sea when the river was next in flood. - -It is probable, I may add, that the seeds or fruits of some of the -plants of the non-buoyant group of the Hawaiian littoral flora may be -dispersed in birds’ plumage. For instance, the spiny fruits of Tribulus -cistoides sink in sea-water; but they are well suited for entangling -themselves in birds’ feathers. - -It is possible that the hairy seeds of Gossypium tomentosum may have -been thus distributed; but there is much that is enigmatical about this -plant (see Chapter XXVI). - -THE INLAND EXTENSION OF THE BEACH PLANTS OF HAWAII.—When we regard the -inland extension of littoral plants in Hawaii, we get fresh indications -of the meagreness of the strand-flora. Several of the species, as -Cæsalpinia Bonducella, Cassytha filiformis, Tephrosia piscatoria, &c., -show themselves only occasionally on the sandy beaches, though they are -common enough on the old scantily vegetated lava-flows near the coast -and are often found miles inland. Indeed, Dr. Hillebrand not -infrequently in describing the station only gives prominence to the -situation of the plants away from the beaches, and places most of them -on the old lava plains that extend inland from the coast. It is only by -a detailed examination of extensive coast lines in these islands that I -have succeeded in preserving to a small degree their reputation as beach -plants. A few of them behave somewhat strangely in their inland station. -Thus, the seeds of Cæsalpinia Bonducella obtained from various -localities showed no buoyancy in my experiments; and had I not found a -solitary buoyant seed in the stranded drift I should have inferred that -this was a rule without exception. - -It is to be remarked that whilst some plants like Scævola Koenigii -occasionally stray a few hundred yards inland on the surface of the old -lava-flows, others like Ipomœa pes capræ and Vitex trifolia, that are -spread far and wide over the inland plains of Fiji, are confined in -Hawaii to the beaches and their immediate vicinity. Some of the plants -like Hibiscus tiliaceus, Morinda citrifolia, and Pandanus odoratissimus, -that are regarded as having been introduced by the aborigines, behave -exactly like indigenous plants in the inland plains; but this is not -necessarily an indication of an indigenous plant in this group, since -the Cactus (Opuntia Tuna) and the Castor-Oil Plant (Ricinus communis) -have spread all over the drier lower regions of the islands, whilst -Aleurites moluccana, the Candle-Nut Tree, which has no means of reaching -these islands without man’s agency, now forms entire woods on the -mountain slopes, usurping the place often of the original forests.... -Further details relating to this subject are given in Note 31. - -The principal points in the foregoing discussion of the strand-flora of -Hawaii may be thus summed up:— - -(1) The indigenous, that is, the pre-aboriginal, strand-flora of this -group lacks not only the mangroves and their associated plants, but also -most of the characteristic beach-trees of the South Pacific, which are -known to owe their wide distribution in tropical regions to the -currents. - -(2) The meagreness of the littoral flora is intensified by the tendency -of some of the plants to extend inland and to desert the coasts, and by -the occurrence on the beaches of peculiar species not found outside the -Hawaiian Islands. - -(3) The absence of the mangrove formation and of so many of the typical -beach trees of the Pacific cannot be attributed either to the lack of -suitable stations, or to climatic conditions, or to deficient floating -power of the seed or fruit. - -(4) As in the case of Tahiti, the mangroves and their associated plants -are lacking because the floating seedlings of Rhizophora and Bruguiera, -the pioneer plants of a mangrove-swamp, have failed to reach Hawaii in a -fit condition for establishing themselves. The numerous plants that -accompany a mangrove-swamp have thus been unable to find a home, though -the buoyant powers of their fruits or seeds are often great. - -(5) With the missing beach-trees, however, which possess fruits that can -float for years unharmed in sea-water, no such incapacity is suggested. -Most of them have large fruits, which could only reach Hawaii through -the currents. This absence from the Hawaiian indigenous strand-plants of -most, if not all, of the large-fruited species, where on account of size -the agency of birds is absolutely excluded, is very remarkable; and it -at first seems to throw grave suspicion on the efficacy of the currents -for the whole strand-flora. - -(6) It is, however, to be noticed that these large-fruited beach trees -have not only failed to reach Hawaii but have also failed to reach -America. The question thus acquires quite a different aspect, and -America becomes the possible source of most of the Hawaiian plants with -buoyant seeds or fruits. - -(7) This subject is discussed in the next chapter; but it is here shown -that at their best the currents have taken but a secondary part in -stocking the Hawaiian beaches with their plants, since many of the -plants have non-buoyant seeds or fruits. - -(8) The drift stranded on the shores of the Hawaiian Islands is composed -of logs from the north-west coast of North America. No drift from the -south has been discovered; but it is not unlikely that future -investigators will find some seed-drift from tropical America. - -[Illustration: - - THE WORLD - SHOWING - OCEAN CURRENTS - - John Bartholomew & Co., Edin^r. -] - - - - - CHAPTER VIII - - THE LITTORAL PLANTS AND THE CURRENTS OF THE PACIFIC - -The working value of the currents as plant-dispersers.—The relation - between the currents and the distribution of shore-plants.—The clue - afforded by the American plants.—Two regions of tropical shore-plants, - the American and the Asiatic.—America, the home of the cosmopolitan - tropical shore-plants that are dispersed by the currents.—Hawaii and - the currents.—Summary. - - -ACTIVE as the currents are in dispersing seeds and fruits over the -Pacific, it should be remembered that those plants that owe their -distribution to this agency are only shore-plants, and not, indeed, all -the shore-plants, but only those with buoyant seeds or fruits. Even the -coral atoll owes a great deal to the agency of the fruit-pigeon and of -other birds; for instance, their species of Ficus, Eugenia, and Pisonia. -In order, therefore, not to form an exaggerated notion of the efficacy -of the currents, it will be necessary to obtain some numerical idea of -what they have really accomplished in transporting seeds and seedvessels -over the oceans in a state fit for successful germination on the shores -upon which they are stranded. It is requisite to make this proviso, -because in some cases the currents work to no purpose. Thus, the empty -nuts of Aleurites moluccana are carried far and wide over the Indian and -Pacific Oceans, and are stranded on the beaches of the various islands, -as I have found myself in the cases of Keeling Atoll, Java, and Fiji. -The Coco-de-Mer, or the Double Coco-nut Palm, is another apt instance. -Though its fruits have been carried far and wide over the Indian Ocean, -the species is restricted to the Seychelles. So also the acorns of -various species of Quercus are widely but ineffectually distributed by -the currents both in temperate and tropical regions. (This subject of -useless dispersal is dealt with in Chapter XIII.) - -It is essential to bear in mind at the outset that for their inland -plants the Pacific islands can draw on the floras of a relatively large -portion of the globe. Such plants, having as a rule fruits or seeds that -sink in sea-water, or are incapable of floating for long periods, could -only have arrived at these islands, where man’s interference is -excluded, through the agencies of winds and birds, assisted by other -lesser agencies, as those of bats, insects, &c. On the other hand, for -their littoral plants, which are for the most part dispersed by the -currents, the source of supply is very restricted. The shore-plants with -buoyant seeds or fruits of the islands of the tropical Pacific, that are -here dealt with, number only about seventy, and it is not likely that -this number will be greatly increased, since, whatever may be the -deficiencies in our acquaintance with the inland floras of these -islands, we have a fairly complete knowledge of the strictly littoral -plants. - -I do not suppose, indeed, that the number of such plants with seeds or -fruits capable of being transported unharmed over wide tracts of sea -would much exceed 100 for the whole Indo-Pacific region from India to -Tahiti. Professor Schimper gives a list containing 117 tropical plants -distributed far and wide over the shores of this region, and made up of -species dispersed by currents, birds, and man. Taking a liberal -estimate, not over two-thirds of the plants mentioned in this list are -dispersed by currents. Then, again, if the flora of a coral atoll, like -that of Diego Garcia or of the Keeling Islands, is taken as affording an -index of the work of the currents, the number of plants dispersed by the -currents would appear to be indeed restricted, since in either case -their indigenous flowering plants, including those of both the buoyant -and non-buoyant groups, do not exceed fifty. - -About twenty years ago, Mr. Hemsley, who, in his work on the botany of -the _Challenger_ Expedition, prepared the way for the investigation of -this subject, made a list of not less than 120 plants, almost all -tropical, that are “certainly or probably dispersed” by the currents -(Introd. _Chall._ Bot., p. 42). This is admittedly only a preliminary -list, and as the result of recent investigations some plants have to be -omitted and others to be added; but I doubt whether, numerically, it is -far below the mark. The relative efficacy of the currents seems to have -been first systematically discussed by De Candolle in his _Géographie -Botanique_, which was published in 1855. Data were then very scanty, and -out of a list of nearly 100 inter-tropical species (Old World plants -found in the New World and New World plants found in the Old World) he -designates nine only as exclusively dispersed by the currents. Even this -list, in one respect, needs correction (see Note 33); but it is of -interest to note that this eminent botanist from the first never looked -upon the agency of the currents as a very important factor in -plant-dispersal; and, finding in the specially directed and carefully -performed experiments of Thuret confirmation of his views, he reiterated -his opinion in a note to that author’s paper in 1873 (cited in Chapter -III.). - -However, De Candolle was quite right in minimising the effect of -currents on the distribution of plants. His extensive survey of the -plant-world from the standpoint of dispersal gave him that sense of -proportion in assigning values to dispersing agents which enabled him to -feel his way almost intuitively, even where exact data were often -lacking. It is, however, a little disappointing to find such a slight -treatment of the subject in Kerner’s great work on the Natural History -of Plants, though one can scarcely controvert his opinion that the -dispersion of plants, as a whole, is not appreciably affected by this -process. Numerically speaking, this is in the main correct; yet it is -here that the genius of Schimper led him to recognise and to mark out a -line of investigation, fruitful in important results, in connection with -the weighty question of “Adaptation.” If the author of this work has -been able to add a little to our acquaintance with this subject, he owes -much to the inspiration he received from Schimper’s memoir on the -Indo-Malayan Strand-Flora. - -Still, it must be admitted that the effectual operations of the currents -as plant-dispersers are limited to the shore-plants with buoyant seeds -or fruits. If we were to include in our list the shore-plants of -temperate regions that possess seeds or fruits capable of floating in -sea-water for long periods, and of afterwards germinating, the total for -the whole world would not, I imagine, reach 200. We cannot here concern -ourselves with those purely river-side plants that contribute their -buoyant seeds and seed-vessels to river-drift, since there is no -evidence indicating that river-side plants are effectively dispersed by -the currents unless they also frequent the estuary and the coast-swamp; -and in that case they come under the head of littoral plants. The total -for the whole British flora would probably not far exceed a dozen, and -nearly all of them are very widely dispersed. - -The working value of the currents as plant-dispersers in the Pacific can -be rudely estimated by the number of littoral plants with buoyant seeds -or fruits that occur in the various groups. Most of these plants hail -from the Indo-Malayan region. Speaking generally of the extension -eastward of the Indo-Malayan strand-plants over the Pacific, Prof. -Schimper (page 195) remarks that they become fewer and fewer in number -as they extend farther from their original home, their number shrinking -to a very few in the most remote groups of the Marquesas and the -Hawaiian Islands. This is well illustrated in the following numerical -results that I have prepared. Of the whole number, some seventy in all, -of the littoral plants of the tropical Pacific with buoyant seeds or -fruits, Fiji possesses about sixty-five, Tahiti about forty, and Hawaii -only about sixteen. As shown, however, in Chapter VII., some of the -Hawaiian littoral trees that are useful to the aborigines were probably -introduced by them. The number actually introduced through the currents -into Hawaii in all likelihood therefore does not exceed ten. There is a -method in this diminution in numbers, as the plants migrate eastward and -northward over the Pacific, which has been described in detail in the -preceding chapter. The efficacy of the currents as plant-dispersers in -the tropical Pacific therefore diminishes as we proceed eastward. - -In the South Pacific the littoral plants preserve their Old World origin -as far as the Polynesian archipelagoes extend eastward across to -Pitcairn, Elizabeth, and Ducie Islands, where we find in one or other of -them such characteristic Indo-Malayan beach trees as Barringtonia -speciosa, Cerbera Odollam, Guettarda speciosa, Hernandia peltata, and -Tournefortia argentea (see Note 34). In the more distant Easter Island -there is a suspicion, for the first time, of immigration from South -America in the presence of Sophora tetraptera. In the islands relatively -close to the American continent, as in Juan Fernandez and in the -Galapagos group, the Indo-Malayan strand-plants are no longer -represented. - -We come now to consider the relation between the distribution of the -shore-plants and the currents. It is quite legitimate to discuss the -currents of the Pacific from the botanist’s point of view, that is to -say, from the standpoint of the distribution of littoral plants with -buoyant seeds or fruits. For ages the buoyant seeds and fruits of the -strand-plants of the tropical Pacific have been drifting over that -ocean, and we have the results now before us in the dispersal of the -species to which they belong. There is no necessity to endeavour to make -the distribution of such littoral plants square with the arrangement of -the currents as shown in a chart. The usual result of such a comparison -has been to lead the investigator, whether an anthropologist, a -zoologist, or a botanist, to find his facts at variance with the course -of the prevailing currents. Man, animals, and plants have entered the -Pacific from the west, whilst the most available currents are from the -east; and one may be perhaps permitted the solecism that the Pacific -islands have apparently been stocked with their shore-plants, with their -aborigines, and with much of their fauna by currents running in the -wrong direction. These Pacific islands could only have had a direct -communication with the Old World, from which they have mainly derived -their shore-plants, by the currents; but since both the aborigines and -the plants have forced their way across the ocean to the Tahitian region -in the teeth of the regular currents, indicated as such in the chart, we -are compelled to assume that they have availed themselves either of the -Equatorial Counter-Current or of the occasional easterly drift currents -that mark the prevalence of westerly winds during the short season of -the year when the easterly trade-winds do not prevail. - -The Equatorial Counter-Current hypothesis would involve a preliminary -crossing of the whole breadth of the Pacific Ocean, that is to say, a -voyage of some 8,000 miles, before the drifting seed doubled back to the -Polynesian Islands. The other view is a much more probable one, as is -sufficiently indicated by the following extract from the “Admiralty -Sailing Directions for the Pacific Islands” (II., p. 25, 1900).... “In -the western part of the Pacific these trades ... are frequently -interrupted by winds which blow from west or north-west, especially -during the months of January, February, and March, when the north-west -monsoon of the Indian Ocean extends out in the Pacific as far as the -Samoa Islands.” In various works on this region one may find reference -to canoes blown off the shore during this season and carried some -hundreds of miles to the eastward. A ship can then sometimes sail with a -fair wind from the southern end of the Solomon Group to the Fijis; and -as we learn from Mariner, the crocodile may be at such times carried -away from the Solomon Islands and stranded in Fiji. Mr. Hedley, in his -exceedingly interesting paper on a zoogeographic scheme for the -mid-Pacific (_Proc. Linn. Soc. N.S.W._, 1899), gives many details of -this nature; but there is no space to deal further with the matter here. - -After all, the botanist must take his cue from the drifting seed and the -distribution of the plant. He finds the seed floating in the open sea as -well as stranded on the beach. He then discovers the plant growing on -the beaches, and by experiment he tests the floating capacity of the -fruit or seed. Finally he ascertains the home of the plant. He does this -for all the littoral plants with buoyant seeds or fruits, and he forms -his own conclusions of the efficacy of the currents independently of the -current-chart, remembering that he has in Time an important factor that -the geographer does not possess in dealing with the currents. The effect -of time has often been to obscure the differential results of the -operations of the currents in the case of those species that, like -Barringtonia speciosa, are almost universally distributed in the islands -of the Pacific. It is obvious that such plants cannot aid us much in the -matter of ascertaining the track followed by the drifting seed in -entering this ocean. But if we find a littoral plant with buoyant seed -or fruit that has only partially performed the traverse we shall possess -in the interrupted operation an important piece of evidence. - -Several years ago, in my paper on Polynesian plant-names, read before -the Victoria Institute, I developed this argument when endeavouring to -find in the floating seed a clue to the route pursued by the Polynesians -in entering the Pacific. Since that time my acquaintance with these -islands and their plants has been considerably extended; but no -important modification of the principal argument is now needed. It was -then pointed out that in Nipa fruticans, the swamp-palm of the Malayan -Islands and of tropical south-eastern Asia, we have a plant well fitted -for the purpose and one well known to be dispersed by the currents over -small tracts of ocean. The Nipa Palm has attempted to enter Polynesia -from the Malayan region by two routes, namely, by Melanesia and by -Micronesia. Along the first route it has in the course of ages reached -the Solomon Islands, where I found it in 1884. Along the second route it -has extended its range to Ualan at the eastern end of the Caroline -Group, where it was observed by Kittlitz many years ago, as indicated in -the narrative of his voyage (_Reise nach russische America, nach -Mikronesien_, etc., 1858, ii. 35), and in Dr. Seemann’s English edition -of the same author’s _Vierundzwanzig Vegetationsansichten ... des -stillen Oceans_. - -The question now arises as to which of these two routes was taken by the -drifting seed. In my paper I adopted the view that the shore plants -reached Fiji and Samoa by Micronesia, that is to say, by the Caroline, -Marshall, and Gilbert Groups. This is the route which, as mentioned by -Mr. Hedley in the paper above quoted, Mr. Woodford prefers for some of -the Lepidoptera; and it is the one that is favoured by Mr. Wiglesworth -for the birds, since in his memoir entitled _Aves Polynesiæ_ he remarks -that certain indications tend to show that the Pelew Islands have served -as a sort of bridge for the spread of species from Indo-Austro-Malaya -right across the Pacific. Though I still think that the beach trees, -most of which would find a home on the numerous coral atolls of the -Marshall, Gilbert, and Ellice Groups, often followed that track, yet I -am now inclined to consider that the mangroves and their associates, -plants which find their most suitable home in the estuaries of large -elevated islands, like those of the Solomon Group, in all probability -reached Fiji in the mass by the Melanesian route. - -[Illustration: - - TRADE ROUTES - - OF THE - - PACIFIC OCEAN - - (_On Mercator's projection_) - - John Bartholomew & Co., Edin^r. -] - -Although the Old World has supplied to the Pacific islands most of their -littoral plants that are dispersed by the currents, that is to say, the -plants with buoyant seeds or seed vessels, yet there is an appreciable -American element, and it is with the plants occurring in the New World -that we are now concerned. The total number of the littoral plants of -these islands that possess buoyant seeds or fruits is, according to the -lists given under Note 35, about seventy. Of these about forty-five are -exclusively Old World species, sixteen occur in both the Old and New -Worlds, three are exclusively American, and six are Polynesian. - -The question we have now to ask ourselves is whether the shore plants -common to both the Old World and America have their homes in America, or -whether they have been derived from the other hemisphere. With one or -two exceptions, as in the cases of the Australian genera Dodonæa, -Scævola, and Cassytha, which, as shown in a later page in this chapter, -present no great difficulty, there does not seem to be any serious -objection, as far as the numerical distribution of the species is -concerned, in regarding America as a possible home of the genus. It is -not often we shall come upon such a striking instance of the principle -that where the species are most numerous there is the home of the genus, -as in the instance of Cocos. The Coco-nut palm has been carried around -the world through the agencies of man and the currents, whilst the home -of the genus is in America. - -Now assuming that in having to choose between the Old World and the New -World as the home of most of the genera in the list we selected the -latter, we have to ask ourselves in what degree this would be consistent -with the place America holds with regard to the distribution of tropical -shore-plants dispersed by the currents and with reference to the -arrangement of the currents. If we except the African continent, there -is no part of the world that bears such a definite relation to the -currents as America, and with an ordinary chart of these regions their -arrangement is to be understood at a glance. Yet strange to say, as far -as the distribution of tropical littoral plants is concerned, America -holds a position that the present system of the currents on its coasts -will not altogether explain. Within the lifetime of the species of -mangroves and other plants of the coast swamps that are found on both -the Pacific and Atlantic coasts of tropical America the two continents -of this name have been united by the emergence of the Isthmus of Panama. - -Few things are more significant in plant-distribution than the -arrangement of the tropical littoral plants with buoyant seeds or -fruits, a subject that is discussed with some detail by Professor -Schimper in his work on the Indo-Malayan strand-flora (page 190). These -plants group themselves into four sections:— - -(_a_) Those of the Pacific and Atlantic coasts of tropical America -(including the West Indies) and of the West Coast of Africa. They -include mostly plants of the mangrove-swamps and their vicinity, such as -Anona paludosa, Avicennia tomentosa, A. nitida, Conocarpus erectus, -Laguncularia racemosa, Rhizophora mangle, etc. - -(_b_) Those of the Old World excluding the African West Coast and -extending from the East Coast of Africa eastward to the Pacific islands. -This is much the largest group and comprises many of the plants named in -the list given in Note 35 under Old World species. One may cite as -examples of plants ranging almost all over this area, Barringtonia -speciosa, B. racemosa, Bruguiera gymnorhiza (in its most comprehensive -sense), Carapa moluccensis, Derris uliginosa, Guettarda speciosa, -Hernandia peltata, Heritiera littoralis, Pemphis acidula, Rhizophora -mucronata, etc. Plants of the mangrove-swamp and of the beach are, -therefore, here included. - -(_c_) Those occurring all around the tropics and including many of the -plants mentioned under Note 35 as Pacific island shore-plants found also -in America. Most of them belong to the Leguminosæ, and there may here be -mentioned Canavalia obtusifolia, Cæsalpinia Bonducella, Entada scandens, -Gyrocarpus jacquini, Ipomœa pes capræ, Sophora tomentosa, and Vigna -lutea. - -(_d_) Those confined to a portion of the two great regions, such as Nipa -fruticans in the Old World, and the Manchineel (Hippomane mancinella) to -tropical America. - -It is to be noted that the ubiquitous species do not include any of the -mangroves. Each of the two regions has its own species, none being -common to both the American and Asiatic regions, although, as is shown -in Chapter XXX., the American species of Rhizophora is now seemingly -breaking its bounds and intruding into the Pacific islands. On the other -hand, some of the mangrove genera, Avicennia, Carapa, and Rhizophora, -are found all round the globe, whilst others are restricted to one or -other of the two regions, Bruguiera, Lumnitzera, and Sonneratia, for -instance, to the Old World region, and Laguncularia to the American and -West African region. - -For convenience we may designate the two great regions of tropical -strand-plants, with buoyant seeds or fruits, the American and the -Asiatic regions, remembering that the first includes both coasts of -America as well as the African West Coast, whilst the second extends -from the East Coast of Africa to Polynesia. Excluding the ubiquitous -species, these two regions are well distinguished from each other. If we -look at the chart of the currents we perceive the reason of the American -region including the West African Coast, and we see why none of the -indigenous plants of this region occur on the African East Coast. So -also with the Asiatic region, a glance at the chart will show that all -the portions of its area are in connection with each other directly or -indirectly through the currents, and that only time is required for the -transport of buoyant seeds over most of the region. - -Hitherto I have mainly followed Professor Schimper in this matter; but -since my visit to Ecuador and the Panama Isthmus some further -considerations have presented themselves to me. If the reader will look -again at the map of the currents, he will observe that there is little -reason for supposing that the Asiatic region can lend its littoral -plants to the American region. On the other hand there are greater -facilities, as far as currents are concerned, for America supplying the -Asiatic region, namely by means of the great equatorial currents that -course westward across the Pacific to the tropics of the Old World. - -It would therefore seem that the American region can receive nothing by -the currents from the Asiatic region. If accordingly it gives but gets -nothing back, we are compelled to assign an origin in the American -region to all littoral plants dispersed by the currents that are found -in the tropics around the globe. This is what we have already regarded -on other grounds as possible for nearly all the littoral plants of the -tropical Pacific with buoyant seeds or seedvessels that are found in -America. These plants are practically the same as those distributed -around the tropical zone which are enumerated in the list given under -Note 35, _b_. With their home in America, by crossing the Pacific they -would ultimately arrive at the East African coast, where their course -westward would terminate; whilst commencing their journey from the east -side of the American continent they would reach the West African coast; -and their distribution around the tropics of the world would be -explained. There follow from these considerations the corollaries that a -tropical strand-plant dispersed by the currents which has its birthplace -in Asia could never reach the American region, and that American -strand-plants are for the most part native-born, excepting those, if -there are any, that hail originally from the African West Coast. - -It is necessary in passing to explain the similarity of shore plants on -the Pacific and Atlantic coasts of Tropical America. For the mangroves -and their accompanying plants inter-communication between the two coasts -is now impossible; and a communication between the two oceans must be -postulated within the lives of the existing species. For the plants like -Entada scandens and Ipomœa pes capræ, which occur inland as well as at -the coast, it is easy to show that in the case of the Panama Isthmus, -their seeds could be readily carried into the Atlantic and Pacific -Oceans by rivers draining the opposite slopes of the same “divide,” so -that the dispersal of the same species from a common centre into two -oceans may be seen in operation in our own day. My observations on this -subject are given in Chapter XXXII., to which the reader is referred. - -I have now gone far enough to indicate the place that America holds with -regard to the distribution of tropical shore-plants dispersed by the -currents and with regard to the currents. There is every probability, as -I venture to think I have shown, that the Pacific islands have derived -most of their ubiquitous shore-plants with buoyant seeds or fruits from -America. But one of the results of our discussion of America in this -double aspect was that excepting in the case of the African West Coast -it gives but does not receive plants from the Old World. We apply this -test, with perhaps a little hesitation, to the shore-plants of the -Pacific islands that are dispersed by the currents; and we find, as will -be seen below, that it is responded to in a remarkable manner. - -It has been observed in the previous chapter that scarcely any of the -large-fruited beach-plants of the South Pacific islands, that could only -have been dispersed by the currents, have reached Hawaii. We do not find -amongst the truly indigenous coast flora of this group any of the -following trees: Barringtonia speciosa, Calophyllum Inophyllum, Cerbera -Odollam, Guettarda speciosa, Hernandia peltata, Ochrosia parviflora, -Pongamia glabra, Terminalia Katappa, Terminalia littoralis, &c. It was -also noted that the currents had not only failed to establish these -plants in Hawaii, but that they had also failed to establish them in -America, the suggestion being that the Hawaiian Islands had been, in -part at least, stocked by the currents from America. That the -Indo-Malayan strand-plants in their extension eastward over the Pacific -should have failed to reach America, is a result we might have expected -from the arrangement of the currents. Yet mingled with them we have -plants like Ipomœa pes capræ, Canavalia obtusifolia, and Sophora -tomentosa, that also occur in America. Since, however, their seeds are -not better adapted for accomplishing the passage across the Pacific from -the Old World to America than the equally buoyant fruits of the -above-named littoral trees that have failed, the presumption arises that -their home is in America, and that they have performed the easier -passage across the Pacific westward from America to the Old World. - -The exclusion of so many characteristic shore-trees from America that -range often over the whole tropical region from the African East Coast -to the islands of the Central Pacific, is not a matter of seed or -fruit-buoyancy, but a matter concerned with the home of the species, and -with the arrangement of the currents. Those shore-plants of this region -that occur also in America have their home in that continent, and have -subsequently been carried across the Pacific by the currents westward to -the Asiatic shores. - -The only exceptions, that I can recall, to the rule that America does -not receive shore-plants dispersed by the currents from the Old World, -are presented by the three Australian genera, Dodonæa, Scævola, and -Cassytha, of which widely spread littoral species occur in America, -namely, Scævola Lobelia, Dodonæa viscosa, and Cassytha filiformis. They -offer, however, but little difficulty, since, as pointed out in other -parts of this work, Dodonæa viscosa has probably been in part dispersed -by man, whilst the other two species are as well fitted for dispersal by -birds as by currents. The occurrence therefore of these species in -America does not necessarily raise the question of the currents. - -The same exclusive principle is illustrated in the scanty littoral flora -of Hawaii. Deprived, like America, of the characteristic large-fruited -beach-trees of the South Pacific, species that could only have reached -it through the agency of the currents, it is scarcely to be expected -that it would have received its few littoral plants with buoyant seeds -from the source which has failed it in the cases of the numerous -absentees. It is to America therefore that we look for the source of its -littoral plants as far as the agency of the currents is concerned. - -The Hawaiian Islands contain about twelve plants, named in the list -given in Note 36, that possess seeds or fruits known to be dispersed by -the currents, and capable, as experiments indicate, of floating in -sea-water for prolonged periods. Not all of them are at present littoral -in their station in this group; but their claim to be considered such in -other regions is established in the Note above mentioned. Of these -plants, seven at least are found in America, five in the Old World also, -and two exclusively in America. This proportion of American plants is -far greater than that characterising the whole littoral flora of the -Pacific islands dispersed by currents, where out of some seventy species -only nineteen are found in America (see Note 35). As far as the -distribution of the plants is concerned, it is therefore quite possible -that Hawaii has received most of its plants that are dispersed by the -currents from tropical America. - -We will now consider how such a possibility is in accordance with the -arrangement of the currents in the North Pacific. If we look at the -Quarterly Current Charts for this ocean published by the British -Admiralty we notice that all through the year the Hawaiian Group lies -more or less within the area of currents flowing from the West Coast of -America, the Northern Equatorial Currents as they are collectively -named. Except in the winter months these currents come from the N.E. and -E.N.E., and bring drift from the coasts of British Columbia, Oregon, and -Northern California. It is then that they pile up huge pine logs on the -shores of the Hawaiian Islands, as I have described in Chapter VII. and -in Note 30; and, according to Dr. Hillebrand, they transport this drift -timber much farther south to the shores of the Marshall and Caroline -Groups. One might cite other facts illustrative of the working of these -currents, such as one finds in the pages of Fornander and other authors; -but this would scarcely come within the province of this work. I may -here remark that when in Honolulu I was informed that a bell-buoy which -had got adrift on the Californian coast was subsequently washed up on -the coasts of Kauai. It is stated in Findlay’s “North Pacific Directory” -(1886, p. 1068), that a junk carrying nine hands that had been blown off -the south coast of Japan in a typhoon, anchored, after ten or eleven -months at sea, in December, 1832, near Waialea in Oahu, the view taken -of its course being that after drifting along in the Japan Current it -came within the range of the south-west current that carries pine timber -to Hawaii from the West Coast of America. - -The portion of the Northern Equatorial Current that strikes the Hawaiian -Group during the greater part of the year is no doubt a south-westerly -deflection of the Japan Current from the American West Coast; and it -would be impossible to find any tropical drift mingled with the pine -logs stranded on the islands during that period. However, in the winter -months, centering in January, the Japan Current flows down the West -Coast of America to about the latitude of Cape Corrientes on the coast -of Mexico, before being deflected westward. Here it meets with a portion -of the Peruvian Current, and both flow westward, the united stream -striking probably only the southernmost islands of the Hawaiian Group. -It is at this season alone that there would be any likelihood of drift -from tropical America being stranded on the Hawaiian beaches, and it is -quite possible that at such a time the Northern Equatorial Current may -carry intermingled in its stream pine logs from Oregon and seed-drift -from Panama. - -I am not inclined to attach any value except in the Western Pacific to -the agency of the Equatorial Counter-Current in transporting seeds and -fruits over the Pacific. It presents seemingly the only opportunity of -the transportal of the seeds and fruits of Asiatic littoral plants to -America; but if at all effective in this way, it would have endowed the -littoral flora of the western shores of tropical America with many of -the trees so characteristic of the coral islands of the Pacific. In this -sense, it has failed completely as an effective agency in -plant-dispersal; and judging by results we may, I think, dismiss it from -our consideration. However, Dr. Hillebrand (p. xv.) assumes that during -the prevalence of south-westerly gales in winter in the Hawaiian -Islands, the Equatorial Counter-Current would be pushed northward so as -to mingle to the east of the group with the North Equatorial Current. In -this manner it is supposed that seed-drift brought direct from the -Asiatic side of the Pacific would be stranded on these islands. This -appears to me to be most improbable, since some ten or twelve degrees of -latitude usually intervene between the Hawaiian Group and the Equatorial -Counter-Current (_see_ Admiralty Sailing Directions, Pacific Islands, -1900, II., 31, and the Quarterly Current Charts; also Encyclopædia -Britannica, vol. 18, p. 118). - -The most serious objection from the botanist’s standpoint against such a -view as that of Dr. Hillebrand is the absence from Hawaii of most of the -shore-plants that we should expect the currents to have brought from the -Old World. It is also evident that as far as the currents are concerned -the Hawaiian Islands are far more likely to receive littoral plants from -America than from the Old World. Though no tropical drift has yet been -found stranded on the coasts of these islands, yet it is not unlikely -that future investigators may find some seed-drift from Central America -on the most southerly coasts of the group, as on the south-east shores -of the large island of Hawaii. It would only be stranded in the winter -months and then probably in small quantities. - - - _Summary of the Chapter._ - -(_a_) Since the effective operations of the currents are limited to the -shore-plants with buoyant seeds or fruits, such plants forming but a -small proportion of any flora, it must be acknowledged that, numerically -speaking, the results of the dispersing-agency of the currents on -plant-distribution in general are but slight. - -(_b_) Yet the importance of the subject is by no means to be measured by -a numerical scale of results, a line of inquiry being here opened up -leading to fields of investigation full of promise for the student of -plant-distribution. - -(_c_) Whilst dealing with the relation between the distribution of -shore-plants and the arrangement of the currents, it is quite legitimate -to discuss the currents of the Pacific from the point of view of the -botanist, who, after all, must take his cue from the drifting seed and -the resulting distribution of the plant. - -(_d_) The shore-plants of the Pacific islands that are dispersed by the -currents being mainly Indo-Malayan in origin, it follows that they have -extended eastward over the Pacific to the Tahitian islands against the -stream of the South Equatorial Current and against the trade-wind. It -is, however, shown that they could have availed themselves of the -interval between January and March when the North-west Monsoon reaches -the Pacific. - -(_e_) It is claimed that whilst the mangroves and their associated -plants have for the most part entered the Pacific by the Melanesian -route through the Solomon Islands, the beach-plants have also followed -the route through Micronesia by the Caroline, Marshall, and Ellice -Groups. - -(_f_) A small number of the strand-plants of the Pacific islands that -are dispersed by currents occur in America as well as in the Old World; -and questions of prime importance arise when we have to decide whether -their home is in the Old World or in the New World. - -(_g_) Good reasons are given for regarding them as chiefly of American -origin; and it is shown that America with regard to the arrangement of -the currents stands in the singular relation of being a disperser but -not a recipient of shore-plants. - -(_h_) It is pointed out that the tropical shore-plants that are -distributed by currents belong to two great regions which are the effect -of the present arrangement of the currents, viz., the American including -the West Coast of Africa, and the Asiatic comprising the remainder of -the tropical zone. Each region has its own plants, and those that occur -in both, being in fact distributed all round the tropics, are regarded, -according to the principle above stated, as having their home in the -American region. - -(_i_) The occurrence of the same strand species on the Pacific and -Atlantic coasts of tropical America is regarded as indicating that the -arrangement of the existing species of its shore-plants, more -particularly of the mangroves, antedates the emergence of the Panama -Isthmus. This hypothesis is not needed for the coast plants like Entada -scandens that occur inland, since we can now observe their seeds being -carried down into the Atlantic and Pacific Oceans by rivers draining the -opposite slopes of the same “divide” in the Panama Isthmus. - -(_j_) It is shown that the currents of the Pacific have failed to -establish the numerous beach-trees (possessing buoyant fruits) of the -Pacific islands, not only in the Hawaiian Group, but also on the coast -of America; and it is therefore argued that we should expect the -Hawaiian Group to have received through the currents its shore-plants -with buoyant seeds or fruits from the tropical west coasts of America. - -(_k_) In support of this contention it is pointed out that most of the -Hawaiian strand-plants that are dispersed by the currents are found in -America, and some indeed in America to the exclusion of the Old World. - -(_l_) The arrangement of the currents in the North Pacific also favours -the view that the Hawaiian Islands are more likely to receive plants by -the agency of the currents from America than from the Asiatic side of -the Pacific. - - - CHAPTER IX - - THE GERMINATION OF FLOATING SEEDS - -Germination in the floating seed-drift of tropical estuaries.—A strain - of vivipary.—Abortive germination of seeds in warm seas.—A barrier to - plant dispersal.—The borderland of vivipary.— Summary. - - -THE tendency of the floating seed or fruit to germinate in the estuaries -of tropical rivers is especially characteristic of the plants of the -mangrove-swamps and of their borders. In the Fijian rivers, and -particularly in the estuary of the Rewa, where the river-water is -usually mixed with that of the sea, there are frequently to be found in -a state of germination floating fruits of Barringtonia racemosa, Carapa -obovata, Clerodendron inerme, Derris uliginosa, Smythea pacifica, &c.; -whilst the floating fruits of more characteristic beach-trees like -Barringtonia speciosa and Cerbera Odollam, that grow also on the sides -of the estuaries, were never noticed in this condition. That this -tendency should be restricted to the plants of the mangrove-formation -and is not to be observed in the beach-trees is a singular fact. There -is, however, an intermediate group of littoral plants mostly belonging -to genera of the Leguminosæ and Convolvulaceæ, such as Mucuna and -Ipomœa, where germination of the floating seed is apt to begin but ends -abortively, and results in the sinking and death of the seed. The -subject of the germination of seeds in the floating drift of tropical -estuaries presents itself, therefore, in three aspects:— - -(1) As concerning the plants of the mangrove-formation, where, excluding -the viviparous species (when germination takes place on the plant), -germination is frequent in the water: - -(2) As concerning the beach-trees where it is rare or absent altogether: - -(3) As concerning certain Leguminous and Convolvulaceous littoral plants -where germination is not infrequent but always abortive. - -Dealing first with the plants of the mangrove-formation, it may be -remarked that the same tendency of the floating fruits or seeds to -germinate, which is above noticed in the case of the estuaries of Fiji, -came under my observation in the floating drift of the estuary of the -Guayaquil River in Ecuador, the germinating fruits and seeds being -carried far out to sea. The seeds of Anona paludosa, which float in -quantities in the river-drift, were often found germinating; and the -same may be said of the fruits of Laguncularia racemosa and of the -“joints” of Salicornia peruviana which abound in the creeks of the -mangrove-delta and are carried out to sea in the germinating condition. - -It might be expected that this readiness to germinate in the brackish -water of estuaries would prove to be a formidable obstacle to the -dispersal of these plants over wide tracts of ocean. The exposed -portions of the seedling might be deemed ill-suited to withstand, -without injury, the “wear-and-tear” of transport by currents over long -distances, even when not affected by the sea-water; and it might be -thought that they would be often nibbled off by fish or destroyed by -other aquatic animals. Only the specially organised seedlings produced -by a viviparous process on the tree, such as those of Rhizophora and -Bruguiera, might be regarded as able to survive the effects of prolonged -immersion in the oceanic currents. - -Observation, indeed, shows that such seedlings are exposed to and suffer -from these perils; yet it is evident from the distribution of the -species that, whether in the germinating condition or not, the seeds and -fruits of Anona paludosa and Laguncularia racemosa have been carried by -the currents from America to the West Coast of Africa. The seedlings of -Avicennia and of Rhizophora mangle have also performed the same -trans-Atlantic voyage. Those of both these mangroves are to be observed -floating off the coasts and in the estuaries of both coasts of America. -The seedlings of Avicennia are particularly abundant in the -mangrove-creeks of the delta of the Guayaquil River; and I observed them -in a healthy condition, ten to twenty miles out at sea, floating -together with those of the Rhizophora. Since, as in the case of -Rhizophora, germination occurs normally on the plant, Avicennia can only -be dispersed by its floating seedlings. Yet it is noteworthy that -although Avicennia seedlings appear, to a marked degree, less fitted for -ocean transport than those of Rhizophora and Bruguiera, the species have -a much wider distribution. Avicennia officinalis has a cosmopolitan -distribution in the tropics and beyond, occurring as it does on the -Atlantic and Pacific coasts of America, on both coasts of Africa, over -Asia and Australia, as well as in New Caledonia and New Zealand, but not -in Polynesia (_Bot. Chall. Exped._, III., 178).... I have now gone far -enough to show that the tendency displayed by the seeds and fruits of -several of the plants of the mangrove-formation to germinate either on -the tree or in the floating drift of estuaries has not affected the -general distribution of the species in its main outlines. Few fruits are -found more often in a germinating condition in the floating drift of the -Rewa River in Fiji than those of Barringtonia racemosa, yet the species -ranges from the African East Coast eastward to Polynesia. Seedlings as -well as seeds or fruits, whether or not in a germinating condition, are, -therefore, able in such cases to disperse the species. - -This readiness of the floating fruits of plants of the mangrove -formation (excluding the viviparous species) to germinate in the -estuaries is, I am inclined to think, due in the main to the strain of -vivipary that runs through nearly all the plants of the mangrove-swamp -and of its borders. It would, indeed, appear that the viviparous habit -(the capacity of germinating on the plant) which finds its extreme -development in Rhizophora and Bruguiera of the Fijian swamps is -represented in its earliest stage in the readiness of the floating -fruits of Barringtonia racemosa, Carapa obovata, &c., to germinate in -the Fijian estuaries, and as remarked in Note 37 there is a suspicion of -vivipary in the instances of both the species just named. Intermediate -cases, as that of Laguncularia in the Ecuador swamps, occur in other -regions with species where germination only takes place at times on the -plant. This subject is, however, generally discussed in Chapter XXX. and -need not be further dealt with here. - -A predisposing cause of the germination of floating seeds and fruits in -tropical estuaries would seem to be afforded by the super-heating of the -water of the estuary. This came under my notice both in the Rewa River -in Fiji and in the Guayaquil River in Ecuador, where the water of the -estuary is often noticed to be some degrees warmer than that of the sea -outside, and of the water from the river above the estuary. (See Note -38.) - -We come now to the subject of the non-germination in tropical estuaries -of the floating fruits of the beach-trees, such as Barringtonia speciosa -and Cerbera Odollam, that in the Pacific islands may contribute to -river-drift. Such trees may grow on the banks of the estuary, and their -fruits would thus readily fall into the water; but in the Rewa estuary -in Fiji it was evident that the fruits and seeds of beach-plants, such -as Scævola Koenigii, are also brought in by the tide. The seeds of -Morinda citrifolia were often noticed in the Rewa drift together with -the fruits of Heritiera littoralis, which is both a beach and a swamp -plant, but never in a germinating condition. The same remark applies -also to the fruits of beach trees found afloat in the sea between the -islands, such as Cordia subcordata, Guettarda speciosa, and Terminalia. -It is possible that a few of these plants, as in the case of -Barringtonia speciosa, display traces in the structure of their fruits -of a lost viviparous habit. (See Note 50.) It is pointed out in -discussing Guettarda that germination is much more easily induced than -one would expect in the case of fruits with such a hard ligneous -putamen. - -An interesting subject is presented in the abortive germination of the -floating seeds of many plants of the Leguminosæ and Convolvulaceæ both -at sea and in a tropical estuary. My conclusions on this matter are -based partly on observations made in Fiji, but mainly on the results of -numbers of experiments, this being unavoidable, since the abortive -germination causes the sinking of the seed. The principal determining -cause of the germination in water of one of these floating seeds is -evidently to be sought in the temperature of the water, it being -immaterial for the earliest stage of germination, as many of my -experiments indicate, whether the seed or fruit is afloat in the sea or -in the river. In these flotation experiments, when conducted under warm -conditions with sea-water, the earliest signs of germination were -frequently displayed in the softening, swelling, and sinking of the -seed. If the swelling seed is taken out in time and planted after a -preliminary soaking in fresh water, the germinating process is at once -resumed and is often successfully and rapidly completed; but if the seed -is allowed to remain in the vessel after it has absorbed sea-water the -vitality of the embryo is destroyed and the seed decays. - -That many seeds would fail from this cause to cross an ocean my -experiments repeatedly demonstrated. Nor does the appearance of a seed -afford any indication of its probable failure to cross an ocean. Some -seeds of Mucuna, as far as their hard coverings could guide one, would -seem to be quite secure from such a risk. The stony seeds, for instance, -of M. urens D.C. look as if they might safely be transported by the -currents round and round the globe; and De Candolle very rightly placed -this species in his scanty list of plants dispersed by currents. Yet few -seeds are more treacherous when their buoyancy in sea-water is tested in -a warm place, as in a hot-house. They may take up water, swell, and sink -in a week, or they may float unharmed for a year. - -The seeds most exposed to this risk are those of the Leguminous giant -climbers, the lianes of the coast and inland forests of the islands of -the tropical Pacific. They belong to the genera Mucuna, Strongylodon, -&c.; and thus several of the plants that constitute for the student of -plant-dispersal the enigmas of the Pacific are here included. The seeds -of Mucuna are especially liable when afloat in sea-water under warm -conditions to display the early signs of germination, swelling up and -sinking to the bottom of the vessel, a process, however, soon arrested -and followed by the death of the embryo unless the seed is removed in -time. Yet the seeds of this genus are notably long “floaters.” Those of -an American species, variously designated as Mucuna pruriens D.C. and M. -urens D.C., have long been known to be washed ashore together with the -seeds of Entada scandens on the western shores of Europe, and -particularly on the Scandinavian coast, where they form regular -constituents of what the Scandinavian botanists correctly term the -Gulf-stream Drift. - -Mucuna urens D.C. occurs with other American shore-plants that are -dispersed by the currents on the African West Coast; and there is no -reason to doubt that its seeds perform the trans-Atlantic voyage. It is -found in Polynesia, in Hawaii, in the Marquesas, and according to -Reinecke also in Samoa; and probably it occurs in other groups. The -specific determinations of the genus, however, need thorough -overhauling, so that it is not possible to deal more than in general -terms with the distribution of a species. The distribution of Mucuna -urens in the Pacific is, however, irregular, and no doubt this is to be -connected with the uncertain behaviour of its seeds when transported by -tropical currents. The seeds would, I venture to think, often sink -through abortive germination in the warm areas of equatorial seas. - -When in Hawaii I kept ten of the seeds of this species (M. urens D.C.) -in sea-water for four and a half months, none of them sinking in that -period, the temperature of the water rarely reaching over 80°F., the -average daily temperature being 76-77°. However, when four years -afterwards in England I placed five of the seeds obtained at the same -time in sea-water under conditions where the water-temperature ranged -for the first few weeks between 75° and 90°, three of them began to -swell within ten days, and on removal at once germinated healthily. The -remaining two were afloat at the end of twelve months, and when planted -one of them germinated a month afterwards. - -Having experimented on the seeds of about half a dozen different species -of Mucuna in sea-water, all with buoyant qualities, it is possible for -me to lay down the general rule for the buoyant seeds of the genus that -sinking is the result of an attempt at germination, which, as before -observed, proves abortive unless the seed is removed in time. It is -obvious that the gardener wishing to raise plants of this genus without -delay might profitably adopt the method of keeping them afloat in water -at a temperature of 80-90° F. until they begin to swell, which may -happen in some cases in a few days. Sea-water seems to produce the most -rapid results. - -When on Keeling Atoll in the Indian Ocean I collected, amongst the -stranded seed-drift brought by the currents to those islands, the seeds -of five or six species of Mucuna, two of which were identified at Kew as -M. macrocarpa, Wall., and M. gigantea D.C. (see my paper on the -dispersal of plants at Keeling Atoll). No plant of this genus appears up -to that time to have been recorded from the Keeling Islands, so that at -all events most if not all of the seeds had been brought by the currents -from the Indian Archipelago, some 700 miles away. It may be added that -amongst the drift gathered by me on the south coast of Java the seeds of -three species of Mucuna were identified at Kew, including the two -above-named species from Keeling Atoll. - -These current-borne seeds of the Keeling beaches had probably performed -an ocean journey of a thousand miles, since the route could scarcely -have been direct. Yet their behaviour when placed eighteen months after -in sea-water in a hothouse in England was most erratic. Of three seeds -of Mucuna gigantea all swelled and sank within eight days. Two seeds of -M. macrocarpa sank after floating from sixty to a hundred days; whilst -of two seeds of another species both remained afloat after a year. In a -sea-water experiment in England on five Hawaiian seeds of M. gigantea, -under the conditions referred to in the Mucuna urens experiment, one -sank within ten days, whilst three of them were afloat after twelve -months, one of them subsequently germinating. This species, it may be -remarked, is widely distributed as a coast plant over tropical Asia, -Australia, and in Polynesia. It seems to take the place in the Old World -which Mucuna urens takes in America, and it is curious that they meet in -Polynesia, being sometimes associated as in Hawaii. In the chapter on my -observations in Ecuador and in Panama it is remarked that Mucuna seeds -are frequent constituents of river, sea, and stranded drift. I, -therefore, have enjoyed the opportunity of observing the behaviour of -the seeds of this genus in a variety of localities, namely, in the -Keeling Islands, in West Java, in Fiji, Hawaii, and tropical America; -and this may be pleaded as an excuse for entering into so much detail -respecting them. - -The large seeds of Strongylodon lucidum (S. ruber), a Leguminous liane -that ranks with the species of Mucuna amongst the huge climbers of the -forest of the Pacific islands, behaved in a similar way in my flotation -experiments in sea-water. Though, as shown in Note 3, these seeds can -float for a year and retain their germinating power, some of them -brought their buoyant capacity prematurely to an end by an abortive -attempt at germination. These black rounded seeds form a common object -amongst the river seed-drift stranded on some of the Fijian beaches in -the vicinity of estuaries. They are so hard and durable that they are -mounted in brooches in Honolulu. Yet these pebble-like seeds will -sometimes begin to swell in a few days in sea-water. Out of five seeds -placed in sea-water in England under warm conditions (the water -temperature for the first few weeks ranging between 75° and 90° F.), one -swelled and sank within ten days, another did so after two months, -whilst the other three were afloat after twelve months, and one of them -subsequently germinated. There is some disagreement amongst botanists as -to the limits of the specific characters of the plants of this genus -(see Note 39); but the plan seemingly most in accord with the -fundamental principles regulating plant-distribution in this region of -the Pacific is to regard the forms found in Hawaii, Tahiti, and Fiji, as -referable to one species. In addition to the Polynesian forms there are -only two or three species, found in the Philippines, Madagascar, and -Ceylon, and it is with the species from the last-named locality that the -Polynesian species is by some identified. - -The seeds of several other Leguminous climbers would probably act in a -similar way, for instance, those of Entada scandens; but the seeds of -this plant experimented on by me were too few to enable an opinion to be -formed. Of four seeds of Dioclea violacea from Fiji that were subjected -to the same experiment as those of Strongylodon lucidum, all floated in -sea-water after a year, with the exception of one that did not swell and -sink until after ten months. On the other hand, in my experiment in Fiji -on the fresh seeds of Canavalia obtusifolia, a plant found on tropical -beaches all round the globe, seventy per cent. sank in the first six or -seven weeks, swelling and displaying the first signs of germination, but -quite ten per cent. were afloat after three months. - -My experiments on the foregoing and other littoral species of the -Leguminosæ merely indicate that under the ordinary temperature of -tropical currents a portion of the seeds will probably sink owing to -abortive attempts at germination. It is likely that if in the -experiments in England a constant temperature of 85° to 90° F. had been -sustained throughout, most if not all of the seeds would have swelled -and sunk within a month or two. The temperature of the experiments in -Fiji and Hawaii did not exceed that of many tropical currents; but there -are areas of superheating in equatorial seas, which I think would prove -insurmountable barriers in the path of most drifting Leguminous seeds, a -subject to which further reference will be made. - -Coming to the Convolvulaceæ, my experiments show that the buoyant seeds -often lose their floating powers from the same cause. Those of Ipomœa -pes capræ may be taken as an example. I was surprised to find when -experimenting on the buoyancy in sea-water of these seeds in Fiji and -Hawaii that a considerable proportion, about a third, sank in the first -two months, swelling and sinking to the bottom. That this swelling -represented the early stage of germination was well brought out in -parallel experiments in fresh water and sea-water made in England on the -buoyant seeds of the British littoral species, Convolvulus soldanella. A -good proportion of the seeds in the first part of the experiment -absorbed water, swelled, and sank, those in fresh water proceeding at -once to germinate healthily at the bottom, whilst those that sank in -sea-water merely decayed. Of the survivors about fifty per cent. in -either case floated after six months. It may be added that the seeds of -other tropical littoral species, such as those of Ipomœa glaberrima and -I. grandiflora, behaved in the same way. - -It would appear from my experiments, and it is a result that we should -expect, that buoyant seeds of the Leguminosæ and Convolvulaceæ would -often float for much longer periods under cool than under warm -conditions. There must be areas of high temperature in mid-ocean that -would prove much more fatal to the chances of a drifting tropical seed -than the icy waters of a Polar current. In my paper on Keeling Atoll I -have described how I procured the germination of a seed of Ipomœa -grandiflora, Lam., after a year’s flotation in sea-water in London, -which included a period of three weeks when the water temperature was at -or about 32° F. These seeds from this point of view would be exposed to -much more risk of sinking through abortive attempts at germination when -drifting across some parts of the Pacific Ocean. It would appear from -the Admiralty Chart of Surface-Temperatures, published in 1884, that -such an area with a surface-temperature of 83° to 86° throughout the -year extends north and east of New Guinea well into the Pacific, -reaching in the first half of the year as far east as the Tahitian -region. It would seem highly probable that the immersion of Leguminous -or Convolvulaceous seeds for many months in these tepid waters would in -most if not in all cases induce incipient germination which would lead -to the sinking of the seed. There are, however, exceptional cases, as -that of Cæsalpinia bonducella, which, as my experiments recorded in -Chapter XVII. indicate, appear to be quite proof against any conditions -of temperature such as are likely to be found in tropical seas in the -present day. - -There are a few general considerations arising out of the foregoing -observations to which reference may now be made. The study of the -behaviour of the floating seed or fruit often carries us, as I have -before implied, to the borderland of vivipary. When from a canoe on a -Fijian river we lift up the germinating fruit of Barringtonia racemosa -from amongst the drift floating past in the stream and pull down from -the branches overhead the seedling a foot in length of Rhizophora, we -hold in our hands the two extremes of the series of vivipary. With many -of the plants of the mangrove-formation there is a fine adjustment with -respect to the germinating capacity of the seed, or in other words a -delicate balancing of organisation on one side and of physical -conditions on the other. A slight disturbance of the equilibrium would -produce great results in plant distribution. Thus, an elevation of the -temperature of the sea-water in the tropics to 90° F. would, I -apprehend, produce the abortive germination of nearly every floating -seed and fruit in equatorial seas, even of those of the beach-trees like -Barringtonia speciosa and Terminalia littoralis that are regarded as -proof against such risks under existing conditions where the -surface-temperatures would average 78° to 80°. - -There would thus be a barrier to the dispersal of plants by currents as -effective as that of a frozen ocean. In the warm, humid climates of the -early geological ages, seed-transport by currents may have been often -impossible, since the seeds that did not begin to germinate on the -plants of the swamps would probably do so in the tepid water of the sea. -Viviparous plants would, however, be placed at no greater disadvantage -than they are at present, since the genera Rhizophora, Avicennia, and -others are now only dispersed by the floating seedlings. But such an -increase of temperature at the present time would mean the death in the -current of the floating seeds and fruits of nearly all non-viviparous -shore-plants. As a rule every Leguminous and Convolvulaceous seed would -swell up and go to the bottom; whilst fruits like those of Barringtonia -racemosa and Carapa obovata, that often germinate afloat in tropical -estuaries, would invariably do so under the changed conditions, and the -seedlings not being adapted for ocean transport would perish. - -Yet we know that with the seeds of many inland plants temperature has -seemingly very little to do with starting the process of germination. We -are familiar with the fact that the seeds of many plants that fail to -germinate in the summer of their production habitually germinate under -apparently less favourable conditions of temperature in the following -spring. This is attributed by botanists to the immaturity of the seed on -first falling from the plant, a further period of maturation being -necessary before, under any conditions, germination is possible. - -We see this also well illustrated in the floating seeds and fruits of -the Thames drift. Most of them fail to germinate in the drift at the end -of the summer and the beginning of autumn, and defer the process until -the following spring, when they germinate freely in the water under much -cooler conditions than those which they experienced in the early part of -their flotation in the drift. There are, however, exceptions to this -rule. Plants like Caltha palustris, for instance, are rarely represented -in the spring seed-drift of ponds and rivers, because most of the fruits -or seeds germinated soon after falling into the water in the previous -summer. - -In most of my sea-water experiments in England the immersion had a very -marked influence, not in causing premature germination and destroying -the germinating capacity, as often happens with the floating seeds of -Convolvulaceæ and Leguminosæ, especially in the tropics, but in -postponing without injury to the seed the process of reproducing the -plant. Such seeds or fruits when placed in fresh water after many months -of flotation in sea-water germinated very freely in a few days, whilst -those left in the sea-water under precisely the same conditions remained -unchanged. This is true of many of the seeds and fruits found in the -Thames drift, such as those of Ranunculus repens, Lycopus europæus, -Rumex, &c. A striking instance was also afforded by the seeds of -Arenaria (Honckeneya) peploides, where seeds transferred directly to -fresh water, after many months flotation in sea-water, germinated in a -few days; whilst those left in the sea-water remained unchanged. This -subject is discussed at length in Note 19, and needs no further mention -here. - -If the seeds of many plants in Great Britain postpone through immaturity -their germination to the following or even to the second spring, it goes -without saying that this does not exclude temperature as the ultimate -determining factor in germination. The immaturity of seeds adds another -link to the series of the germination-range in plants. This range begins -with the plants where germination takes place on the tree and the -seedlings hang suspended from the branches, as in the typical mangroves -Rhizophora and Bruguiera. Here, as is shown in Chapter XXX., there is -evidently no period of repose between the completion of the maturation -of the seed and the commencement of germination. The range ends with the -detachment of immature seeds which ripen apart from the parent plant, -and may postpone the germinating process for months and often for years. -All intermediate stages exist between these two extremes. Thus the -seedling may at once detach itself from the parent as in Avicennia, or -the germinating process on the plant may be limited to the protrusion of -the radicle as in Laguncularia, or the seeds may be quite mature and -ready to germinate as soon as they fall to the ground, as we find with -many small seeded plants. All the stages, of which only a few are here -indicated, are full of suggestiveness for the student of plant-life. - -This subject is dealt with from other standpoints in Chapter XXX., but -the reader will now see more clearly what was meant when I said that the -study of the behaviour of the floating seed leads us to the borderland -of vivipary. In this range of the germinating process we may possess an -epitome of the history of the climatic conditions of plant-life from an -early era in the world’s story, beginning with those ages when perhaps -under the uniform conditions that then prevailed, all plants were more -or less coast-plants and more or less viviparous, and coming down to the -present era when with an extensive and varied land-surface there is -great variety both in climate and in the range of germination. The -mangrove-swamp and its viviparous trees would thus represent from this -point of view a condition of things once more or less universal on the -globe. - - - _Summary of the Chapter._ - -(_a_) The tendency of the floating seed or fruit to germinate in the -brackish water of tropical estuaries is especially characteristic of the -plants of the mangrove-swamp and their vicinity; but with those of the -beach trees that occur in the river-drift it is rarely if at all to be -observed. - -(_b_) From the wide distribution of plants of the mangrove-formation it -is evident that this readiness of the floating seed or fruit to -germinate is not prejudicial to the dispersal of the species. - -(_c_) It may perhaps be in the main attributed to a strain of vivipary -running through all the plants of the mangrove-formation, which finds -its extreme development in the viviparous species, where germination -takes place on the tree. But it is probably favoured by the superheating -of the waters of tropical estuaries. - -(_d_) In the case of the buoyant seeds of several climbers and creepers -of the Leguminosæ and Convolvulaceæ, more or less littoral in their -station, it is shown that in warm water, whether fresh or salt, a good -proportion are apt to sink through incipient germination, which results -when the experiment is made in sea-water in the death of the embryo. - -(_e_) Though in tropical currents of ordinary temperature a good number -of such floating seeds would escape this risk, it is argued that there -are certain warm areas in the tropical seas that would prove much more -fatal to the chances of these drifting Leguminous and Convolvulaceous -seeds than the icy waters of a polar current. It is thus held that these -seeds often sink in mid-ocean in tropical latitudes through abortive -germination. - -(_f_) The study of the behaviour of the floating seed or fruit leads us -to the borderland of vivipary. In the scale of the germinative capacity -of plants it is possible to arrange a continuous series that commencing -with the mangroves, where germination takes place on the tree, ends with -those numerous inland plants where seeds are liberated in an immature -condition. - -(_g_) It is suggested that the viviparous habit may have been the rule -under the uniform climatic conditions of early geological periods and -that with the differentiation of climates that marked the emergence and -extension of the continental areas the viviparous habit has been lost, -except in those regions of the mangrove-swamps which to some extent -retain the climatic conditions once general over the globe. With -differentiation of climate the true seed-stage with its varying -rest-periods has been developed. - - - CHAPTER X - -THE RELATION OF THE BUOYANCY OF SEEDS AND SEEDVESSELS TO THE DENSITY OF - SEA-WATER - -The general principles concerned.—The subject assumes a statistical - character.—Seeds and seedvessels are as a rule either much heavier - than sea-water or much lighter than fresh water.—The present littoral - plants with buoyant seeds or seedvessels could be equally well - dispersed by currents in oceans of fresh water.—Seed-buoyancy has no - relation either in the present or in the past to the density of the - sea.—Though an accidental attribute, the specific weight of seeds has - had a profound influence on plant-distribution.—Summary. - - -TO find amongst the results of my numerous experiments examples -illustrating the influence of density on flotation has not been so easy -as I at first imagined. Excluding all adventitious causes of buoyancy, a -matter discussed in Note 40, it may be inferred that the great majority -of seeds and fruits sink both in fresh water and sea-water. Of those -that are buoyant many float indefinitely in both waters, whilst in a -very few cases, where the floating power is derived from an outer fleshy -covering, as with the fruits of Potamogeton natans, the fruits float a -much shorter time in sea-water than in fresh water, on account of the -injurious effect of the salt upon their coats. - -Experiments have to be specially directed towards this subject. It would -be useless to experiment in fresh water at one time and in sea-water a -month later. Nor would it answer to employ seeds and fruits from -different localities, since variations in this way sometimes occur. It -is necessary that the experiments should be made on seeds or fruits -collected at the same time and place, and that they should be -simultaneous and carried on under the same conditions. As the discussion -proceeds, the reader will perceive that many interesting points are -opened up, and that such an investigation, instead of being, as the -title of this chapter might suggest, an abstruse and disconnected -inquiry, is of considerable importance in relation to the dispersal of -plants through the agency of currents. - -Guided by the results of my experiments in this direction I will proceed -to lay down certain general principles:— - -(A) In the first place it may be accepted as a general rule that _seeds -or seedvessels that sink in fresh water sink also in sea-water_, the -difference in density between the two being rarely a factor of any -importance in determining buoyancy. The great majority of seeds and -fruits come under this category, since, as is pointed out in Chapter -VIII., only a small proportion of the whole, say a tenth, possess -floating power. We might cite, as illustrative of this principle in -temperate regions, almost all the 240 species included in the -non-buoyant group of the British plants experimented on (see Chapter -III. and Note 10). As a general rule this is true alike of the small -seeds of the Cruciferæ and Scrophulariaceæ, of the nutlets of the -Labiatæ and Boragineæ, of the genus Scirpus, and of the dust-like seeds -of Juncus. The results of my experiments on the plants of the tropical -Pacific are no doubt typical of other tropical regions; and if I wished -to quote instances, I should have to enumerate not only most of the -plants without buoyant seeds or fruits that are mentioned in the Fijian -and Hawaiian lists given under Notes 2, 4, and 6, but also to appeal to -tropical regions generally. - -(B) One can carry the principle above-named yet further and say that not -only as a rule do seeds or fruits that sink in fresh water sink also in -sea-water, but that so far as tested _many of them sink in water of much -greater density than that of ordinary sea-water_ (1·026). Thus, for -instance, the seeds of Nuphar luteum, Scrophularia aquatica, and -Stellaria aquatica, the nutlets of Polygonum persicaria, and the achenes -of Aster tripolium sank in sea-water the density of which had been -raised to 1·050, the limit of the experiment. The minute seeds of Juncus -communis and J. glaucus and the larger seeds of Luzula campestris, even -after drying for six months, sank in salt water having a density of -1·075. It would, however, seem probable that for most of these small -seeds and seedvessels a density of 1·100 would prove to be the critical -point. If this is so, then most of those that sink in sea-water would -float in the dense water (1·160) of the Dead Sea. - -However, my investigations have only gone a small way in this direction; -and perhaps some of my readers will pursue the inquiry. I will take the -case of the nutlets of Scirpus palustris. They sink in fresh water and -in sea-water, or may float in rare cases for a day or two. Out of 100 of -these seed-like fruits, 25 floated in salt water of a density of 1·075, -13 in water of 1·050, 7 in sea-water (1·025), and 3 in fresh water, -(1·000). It would thus appear that the proportion of buoyant nutlets is -doubled with every increase of ·025 of the density scale. At this rate -of increase they would all float in salt water of a density of 1·125, -which may be regarded as the suitable medium for the flotation of the -fruits of this Scirpus.... The seeds of Glaucium luteum, the Sea-Poppy, -have no buoyancy either in fresh water or in sea-water even after -prolonged drying. They all sank in water of a density of 1·050, but 18 -per cent. floated when the density was raised to 1·075. At the rate of -increase noticed in the case of Scirpus palustris, all the seeds would -float in water of a density of 1·130-1·140.... The acorns of the Common -Oak (Quercus robur) have usually but little buoyancy unless they have -been long drying. After soaking in fresh water for half an hour 100 -mature fruits, without the cupule, that had been kept a fortnight, I -found that only 2 floated in fresh water, 6 in sea-water (1·025), and 18 -in water of 1·050. At this rate of increase all would float in water -having a density of 1·080-1·090. - -(C) There is also another general rule, and it is this:—_Seeds or fruits -that float for a long time in sea-water usually float almost as long in -fresh water_. Here belong the greater number of buoyant seeds and -fruits, those only able to float for a few weeks being comparatively -few. Now with the long-floating seeds and fruits, those for instance -that float in the drift of English rivers from the autumn to the spring, -or those that are transported by currents over the tropical zone, there -is, as a rule, but a slight difference between their flotation periods -in fresh water and sea-water. If one of them sinks after floating for -several months in fresh water, it will sink in sea-water a few days -after. Fruits of Scævola Kœnigii, pyrenes of Morinda citrifolia, and -seeds of Thespesia populnea, Ipomœa grandiflora, Cæsalpinia bonducella, -and of different species of Mucuna, that had been kept afloat for a year -in sea-water, floated just as buoyantly in fresh water at the close; and -in those cases where any sank during the course of the experiment, it -was ascertained that they were able to float in fresh water almost to -the end. - -That many of the seeds and fruits of tropical littoral plants that are -known to be dispersed by the ocean-currents will float well in fresh -water is shown in the constant occurrence in the floating drift of -Fijian estuaries, where the water may be quite fresh or brackish, of the -seeds and fruits of plants like Cerbera odollam, Clerodendron inerme, -Entada scandens, Heritiera littoralis, Ipomœa pes capræ, Morinda -citrifolia, Mucuna, Vigna lutea, &c. In the same way I noticed afloat in -the Guayaquil River in Ecuador, when the water was quite fresh, seeds -and fruits characteristic of the sea-drift, such as those of Anona -paludosa (seeds), Entada scandens, Ipomœa, Mucuna, Vigna, &c.; and when -we supplement observation with experiment, as for instance in the case -of Anona paludosa, we find that they will float equally long in fresh -and sea-water. - -The same rule prevails with most of the buoyant seeds and seedvessels of -plants of the British flora—seeds and fruits, as I may remind the -reader, that are mostly to be found in river and pond drift. I am not -able to distinguish any difference of importance in the results of the -separate fresh-water and sea-water experiments. Thus with the seeds or -seedvessels of Bidens cernua, several species of Carex, Galium palustre, -Iris pseudacorus, Lycopus europæus, Ranunculus repens, and numerous -others, the difference after a flotation of many months was but slight. -If the results of the separate experiments were to be compared, there -would be at least ninety afloat in fresh water for every hundred afloat -in sea-water; and if at the end of a sea-water experiment, whether -occupying three, six, or twelve months, the seed or fruits were to be -placed in fresh water, quite nine-tenths and sometimes more would remain -afloat. A striking illustration of the principle that the excess in -density of sea-water, as compared with fresh water, adds but little to -the floating capacity of seeds is to be found in the results given in -Note 41 of simultaneous experiments made some years since by Mr. Millett -and myself at Marazion and in London on the seeds of Convolvulus -soldanella. - -(D) In their relation, therefore, to the density of fresh water and -sea-water, most seeds and seedvessels may be placed in _two principal -classes, the first including quite four-fifths of the total, where they -are much heavier than sea-water, and the second comprising most of the -remainder, where they are much lighter than fresh water_. - -(E) It would be surprising, however, if there were not some seeds or -seedvessels that come between these two extreme groups; some, indeed, -that have a specific weight approximating to that of fresh water, or to -that of sea-water, or fluctuating between them, and presenting such -evidence of a fine adjustment that the observer, forgetting that they -are members of a series, might be apt to regard them as specially -adaptive in their origin. It will thus be seen that this subject is -gradually assuming a statistical character; and in truth we shall -ultimately recognise here the play of the laws of numbers. - -As an example of the plants where the specific weight of the seeds or -fruits is near that of fresh water, Alisma plantago may be taken. In the -course of an experiment, by lowering the density of the water from 1·025 -to 1·020, I sent a shower of floating carpels to the bottom. The results -vary considerably, as one might expect; but, generally, during the first -few days of an experiment about twice as many (sometimes in all as much -as 80 per cent.) sank in fresh water as in sea-water, a few only -floating in either water for long periods.... The seeds of Arenaria -peploides present an example where the specific weight is between that -of fresh water and of sea-water. For the purposes of dispersal they may -be considered as heavier than fresh water and lighter than sea-water. -The details are given in Note 18; but it may be remarked here that -plants possessing seeds or fruits that sink in fresh water and float in -sea-water are very rare. As indicated below, this is what we might look -for on statistical grounds. - -Plants whose seeds or fruits are not much lighter than sea-water are -exceptional. In such cases the effect of increased density of the water -is to extend the period of flotation. Thus, in my experiments on the -nutlets of Scirpus maritimus, the majority of the fruits floated in -fresh water only eight to ten days; whilst in ordinary sea-water they -floated in most cases two to three weeks; but when the density was -raised to 1·050, the greater number of them were afloat after two -months. In a few plants, as with Spiræa ulmaria, the effect of the -difference in density between fresh and sea-water was not to extend the -period of flotation, but to increase the number that floated for a given -period, the extreme limit of the buoyancy of the carpels in either water -with this species being about three weeks. - -Amongst tropical plants, as illustrated by those of the Pacific islands, -cases also came under my notice where the mean specific weight of the -seed is somewhere between those of fresh water and sea-water. The seeds -of Afzelia bijuga, an inland as well as a littoral tree in Fiji, offer -an interesting example. If we place 100 seeds of a littoral tree in -sea-water, we find that on the average about 70 float. If then we lower -the density gradually, some of the seeds begin to sink at once; and on -the removal of the survivors to fresh water, about 47 will remain -afloat. The results may thus be stated:—Out of 100 littoral seeds, 30 -are specifically heavier than sea-water (1·025); 23 are between -sea-water and fresh water in specific weight; whilst 47 are lighter than -fresh water (1·000). When, however, we take 100 seeds of inland trees, -we find that on the average 87 are heavier than sea-water, 5 are in -weight between sea-water and fresh water, and 8 are lighter than fresh -water. The significance of these figures becomes evident when we arrange -them in curves. The combined result for littoral and inland seeds is -given in the diagram below; and we see there, what is also indicated -with the separate curves that we are dealing with a double series, one -concerned with seeds lighter than fresh water, and the other with seeds -heavier than sea-water. The reader can himself supply the separate -curves for the littoral and inland seeds. The point, however, to notice -is that if a botanist with a statistical bent were to make a -miscellaneous collection of the seeds of the Vesi (Afzelia bijuga) in -one of the Fijian islands, in order to test their buoyancy, he would -obtain such a result as is given in this diagram. Two varieties of the -tree would be at once indicated, and further research would indicate -that these varieties were connected with littoral and inland stations. -This subject is further dealt with in Chapter XVII. - - +------------------------------------------------------------------------------+ - |Combined results for 200 seeds of Afzelia bijuga (100 littoral; 100 inland). | - +--------------------------------+------------------------+--------------------+ - | | | | | - |Percentage.| Heavier than sea- | Between sea-water and | Lighter than fresh | - | | water, or +1·025. | fresh water in weight. | water, or -1·000. | - +-----------+--------------------+------------------------+--------------------+ - | | | | | - | 100 | | | | - | | | | | - | +--------------------+------------------------+--------------------+ - | | | | | - | 80 | | | | - | | | | | - | +--------------------+------------------------+--------------------+ - | | | | | - | 60 | | | | - | | | | | - | +--------------------+------------------------+--------------------+ - | | . | | | - | 40 | . | | | - | | . | | | - | +--------------------+------------------------+--------------------+ - | | . | | | - | 20 | .| | . | - | | |. | . | - | +--------------------+------------------------+--------------------+ - | | | . . | | - | 0 | | . | | - | +--------------------+------------------------+--------------------+ - | | | | | - +-----------+--------------------+------------------------+--------------------+ - -It might seem strange that the seeds of Entada scandens should come into -the category of seeds with a specific weight near that of fresh water; -yet my observations in Fiji indicate that such is the case. In the -discussion of this plant in Chapter XVII. it is pointed out that, as a -rule, not more than a fourth will float in a river when they are first -freed from the pod, and not more than fifty per cent. will float in the -sea. Those that float, however, in either water will usually float -indefinitely. The seeds also of Mucuna gigantea D.C. are not very much -lighter than fresh water. Out of six seeds that floated in sea-water -buoyantly, five floated in fresh water, but heavily. - -It is of interest to notice in this connection that the -mangrove-seedlings produced by germination on the tree, as in the case -of Rhizophora and Bruguiera, have a mean specific weight somewhere -between fresh water and sea-water. This is often illustrated in a -curious way, when the seedling has not been prematurely detached from -the tree. Thus in the sea off the coast of tropical America, as well as -amongst the Fijian Islands, the seedlings of Rhizophora mangle are as a -rule to be observed floating horizontally; whilst in the fresh or -brackish water of the estuaries of these regions they assume a more or -less vertical position, only the plumular portion protruding above the -water. This is also true of the seedlings of Rhizophora mucronata, the -Asiatic mangrove, and of Bruguiera rheedii. This subject is discussed in -detail in Chapter XXX.; but it may be here remarked that a good -proportion of Rhizophora seedlings, when detached in the mature -condition from the tree, have no buoyancy, between 20 and 50 per cent. -going to the bottom when they fall into a river, and between 5 and 10 -per cent. when they drop into the sea. The navigator might often obtain -an indication of the density of the sea-surface when approaching the -mouth of a large river by observing the floating Rhizophora seedlings (a -foot long) which are carried out to sea in numbers. If he sees them from -the deck of his ship floating horizontally he will infer that the -surface-water is mainly sea-water. In ordinary fresh water when they -float vertically he would not be able to distinguish them from floating -seeds or fruits. - -It has only been possible to treat this subject in an illustrative -manner. More details might have been given; but I have gone far enough -to bring the following points into relief and to justify one in drawing -the conclusions to be now stated. - -_Most seeds and seedvessels in respect of their floating powers tend to -gather around two centres or means and to form two groups, the sinking -group and the buoyant group._ - -_In the sinking or non-buoyant group, which includes 80 per cent. of the -whole, the mean specific weight is considerably greater than that of -sea-water (1·026), which would require its density to be raised to 1·100 -in order to serve as a floating medium for many of them._ - -_In the buoyant group the mean specific weight is much lighter than that -of fresh water (1·000); and from this it is to be inferred that in -oceans of fresh water the same fruits and seeds in the mass would be -distributed by the currents that are transported by them at the present -day. Even though it arose from an ocean of fresh water, the coral island -would receive the same littoral plants through the agency of the -currents that it receives under its existing conditions._ - -The number of plants with seeds or fruits between fresh water and -sea-water in specific weight is very small, probably not over 2 per -cent. of the total. Most seeds or fruits that sink in fresh water sink -also in sea-water, and most that float in sea-water float also in fresh -water. _Nature has thus created a wide gap between the sinking and the -floating seed; and nearly all of the work of the present currents in -plant-dispersal might have been effected, so far as the density is -concerned, in fresh water._ She has not arranged seeds and seedvessels -in what the statistician would term “a good series.” As indicated in the -diagram below, there are two series that meet in the neutral region -where the density is between fresh water and sea-water, but with -culminating points placed on the one side far above the density of -sea-water and on the other far below that of fresh water. - - +-------------------------------------------------------------------------------+ - | Relation of the specific weight of seeds and fruits to the density of | - | fresh and sea-water. | - +--------------+--------------------+-----------------------+-------------------+ - | Percentage. | Heavier than sea- |Between fresh and sea- | Lighter than fresh | - | | water, or +1·026. | water, 1·000-1·026. | water, or -1·000. | - +-------------+--------------------+-----------------------+--------------------+ - | | | | | - | 100 | | | | - | | | | | - | +--------------------+-----------------------+--------------------+ - | | | | | - | 80 | | | | - | | | | | - | +--------------------+-----------------------+--------------------+ - | | . | | | - | 60 | . | | | - | | . | | | - | +--------------------+-----------------------+--------------------+ - | | . | | | - | 40 | . | | | - | | .| | | - | +--------------------+-----------------------+--------------------+ - | | | . | | - | 20 | | . | | - | | | . | | - | +--------------------+-----------------------+--------------------+ - | | | . | . | - | 0 | | . .| | - | | | . | | - | +--------------------+-----------------------+--------------------+ - | | | | | - +-------------+--------------------+-----------------------+--------------------+ - -I do not, therefore, think that the buoyancy of seeds and fruits has had -any relation either in the present or in the past to the density of the -sea. Nor is it to be supposed that any slight variations in density in -the course of ages would have materially affected the dispersal of -plants by currents. It is to be inferred that the physicist and the -geologist would be prepared to grant only small variations, such as a -change from 1·020 to 1·025. It will be gathered from what has been said -before that changes of this nature would have a very slight influence on -the buoyancy of seeds and fruits, since the plants they would affect -would be very few. The change that the student of plant-dispersal would -require to produce any marked alteration in distribution would be in -amount alarming to the physicist. - -Whether or not the oceans have been getting fresher or salter in the -course of ages (see Note 42), we will be moderate in our demands, and -will listen to the physicist when he argues that a diminishing density, -for instance, from 1·035 to 1·025, in the course of ages might explain -some of the peculiar features in the present isolation of insular -floras. Many seeds, he would contend, that could float across an ocean -having a density of 1·035 would be unable to accomplish it when the -density fell to 1·025. It has, however, been remarked that the critical -point of density for the flotation of seeds or fruits that sink under -present conditions is probably about 1·100. Cases of such a fine -adjustment to the density of sea-water are too few to endow this -argument with any weight. Or it might be suggested that with a gradual -increase in density in the lapse of ages seeds might float now that sank -before, or they might float for a longer period. Such a change, however, -would not have much effect, since nearly all the seeds and seedvessels -that sink in our rivers sink also in our seas, and a much greater -increase of density is required to make any difference. - -Yet, although we might term the sinking of a seed or fruit an accidental -attribute of certain plants, just as we might regard the floating of a -log as an accidental attribute of a pine, since in either case the -specific weight might have been acquired without any direct relation to -the density of water, still the sinking of the seed or fruit signifies a -profound distinction not only, as is stated below, in plant -distribution, but, as we shall see later on, in plant-development. -Especially striking, says Prof. Schimper (p. 153), is the dependence -between an over-sea area of distribution and a station at the coast in -the case of species of the same genus of which some belong to the -littoral and some to the inland flora. In the first place, as has been -often remarked in these pages, we have a wide distribution generally -associated with considerable buoyancy of the seeds or fruits. In the -second case the areas are usually very restricted and there is little or -no buoyancy. The better fitted a seed or fruit is for dispersal by -currents the greater, therefore, is the area of the plant. Whether such -an important relationship depends on an accidental attribute of the seed -or fruit is the question that immediately presents itself. But it is -obvious that in raising such a question we touch on a very vital point -in adaptation, since if attributes developed in one connection have a -profound influence in another we may have to rearrange some of our -fundamental notions of the inner workings of Nature. - -Let us, therefore, look a little closer into this matter, and turn again -to the Pacific islands. The present state of things may be thus tersely -described. Whilst the shore-plants dispersed by the currents have -remained relatively the same, changes of all kinds, from the production -of a variety and of a species to the development of a genus, have taken -place in the inland floras. Now, let us imagine that all this is altered -and that every seed or fruit is buoyant. There would then be but little -distinction between the strand and inland floras, since they would be in -a constant state of interchange, and most species would be widely -distributed. A relatively monotonous aspect would belong to all insular -floras, and indeed to much of the plant-world, since isolation, one of -the principal conditions for the origin of new species and new genera, -would often not exist. - -On the other hand, let us suppose that all seeds and fruits were -non-buoyant. The agency of birds would then be alone available for -stocking new islands with most of their plants. The conditions of -isolation would be intensified. There would be no widely-ranging -strand-flora, since every island and every stretch of continental -sea-board would possess its own littoral plants that could only reflect -the peculiarities of the inland flora. The only determining factor -between coast and inland plants would be the presence or absence of the -capacity or organisation for occupying a station on the sea-shore. - -We have now proceeded far enough to disclose the far-reaching influence -on plant-distribution and on plant-development that the relation between -the specific weight of seeds and fruits and the density of sea-water -must possess. Yet it has been shown that when such a relation is viewed -statistically it has an accidental aspect. We will accordingly devote -the next few chapters to the discussion of the buoyancy of seeds and -fruits from the structural standpoint. - - - _Summary of the Chapter._ - -(_a_) The great majority of seeds and seedvessels (quite 80 per cent.) -are much heavier than sea-water, but a noticeable proportion are -considerably lighter than fresh water, whilst those with a specific -weight near that of fresh water or of sea-water are very few. - -(_b_) The buoyancy of seed and fruit has no direct relation to the -density of sea-water, and even if the ocean was deprived of all its -dissolved salts, the agency of the dispersal of plants by currents would -not be materially affected. - -(_c_) Small changes in sea-density, such as the physicist would allow, -would, therefore, have no appreciable influence on the operations of the -currents as plant dispersers; and only great changes in density, such as -are presented by the waters of the Dead Sea, would add materially to the -number of floating seeds and fruits. - -(_d_) Although the specific weight of seeds and fruits in its relation -to sea-density may be regarded as an accidental attribute, their -non-buoyancy in the great majority of plants has had a far-reaching -influence not only on plant-distribution, but on plant-development. The -plant-world would be transformed if all seeds and fruits floated in -sea-water. - -(_e_) If the floating seed or fruit displays a quality that, so far as -the density of the sea is concerned, has been developed in quite another -connection, we have next to inquire whether the structure of such -buoyant seeds and fruits also affords evidence of non-adaptation. - - - - - CHAPTER XI - - ADAPTATION AND MEANS OF DISPERSAL - -Nature has never concerned herself directly with providing means of - dispersal.—Fleshy fruits not made to be eaten.—Nor “sticky” seeds to - adhere to plumage.—Nor prickly fruits to entangle themselves in fur - and feathers.—The dispersal of seeds a blind result of the struggle - between the intruding Evolutionary power and the controlling influence - of Adaptation. - - -BEFORE entering into a discussion of the causes of the floating powers -of seeds, it is necessary that I should state my general position on the -relation between capacities of dispersal in the organic world and the -question of adaptation. Adaptation runs through all the organic and -inorganic worlds, and we cannot conceive an universe without it. The -naturalist who looks only for the end in the purpose served makes but a -partially legitimate use of the phrase. On the other hand, it has been -improperly appropriated by those who hold to the theory of Natural -Selection, as indicating the result of small fortuitous variations that -have chanced to be of service to the species in the struggle for -existence. There is no question here of any end in view. Nature is -represented as working blindly, and the result of such “fortuitous -variation” is termed an adaptation. We cannot, however, pick and choose -only adaptations that are very evident in their character. We must -include everything in the organic world as an adaptation, whether -apparent or not, that is in direct relation with the organism’s -conditions of existence. It is not conceivable that an organism can be -adapted to conditions outside its environment, and yet many so-called -adaptations are of this character. - -Nature—and I here confess my belief in a determining agency working -above and through all living and dead matter, but largely controlled and -checked by the laws of the physical world—Nature, as I apprehend, has -never concerned herself directly with providing means of dispersal -either for plants or animals. With regard to plants, she makes no direct -provision for the distribution of their fruits or seeds. If she had done -so, she would have employed some uniformity in her methods, as in the -instance of the means of reproduction; whereas the modes of dispersal -are almost infinite in their variety. When I say that Nature makes no -direct provision for the dispersal of plants and animals, I mean not in -the sense that a bird is adapted for an aerial life, or an aquatic plant -for a more or less submerged existence. That a bird is often able to -distribute its kind over a great area is the “accident” of its -conditions of existence. In a similar way the wide distribution of the -“ticks” that they carry round the world is due to the parasitical habits -of these insects, habits that have been acquired without any view to -their mode of dispersal by birds. - -Similarly it cannot be said of seeds or fruits that are transported by -birds, whether adhering to their plumage by means of hooks or hairs, or -through some viscid excretion, or inclosed in soil adhering to the feet -or legs, or carried in the stomach and intestines, that Nature has made -any special provision for their dispersal. The dispersing agencies take -advantage of certain capacities or characters of a seed or fruit that -have been developed in the plant for quite other reasons and in -conformity with quite other principles. There may be mentioned as -examples the mucosity of seeds, the fleshiness of fruits, the occurrence -of hairs and prickles, &c. Yet as far as their connection with dispersal -is concerned, such capacities and characters are blind results in the -history of the plant’s development, the dispersing agencies making use -of what was not intended for them. - -“Adaptation to definite life-purposes,” as Sachs terms it (_Physiology -of Plants_, 1887, p. 122), is seen everywhere; but it is adaptation -restricted to the organism’s conditions of existence. It is not -conceivable, as I have said, that an organism can be adapted to -conditions outside its environment. If there is such a seeming -adaptation, it is but a blind result, the accidental outcome of -collision or contact between two sets of conditions. If we represent a -number of these sets of conditions by several circles gradually -increasing in size until they encroach on each other, we find that the -circles lose their form and acquire a polygonal shape. All characters -seemingly connected with modes of dispersal have only this indirect -relation to such agencies; and their utility in these respects is an -accident in the plant’s life. They have not been acquired in connection -with the dispersing medium, but are the products of the laws of growth -and heredity, guided by a determining agency, and acting within the -organism’s conditions of existence. It is within these narrow limits -that all evident adaptations lie. In matters outside the conditions of -the development of seeds and fruits, the evolutionary or determining -principle “lets them go.” Detached from the plant, they come in contact -with conditions for which they were never created. The predominant power -in Nature, that brings to a successful issue the development of an -organism, has its limitations, and this is one of them, the evolutionary -or determining influence being ever checked and hampered by the laws of -the inorganic world. - -I can only refer briefly to some of the reasons that have led me to -apply this view of the duality of forces in Nature to the subject of -plant-dispersal. The principles of evolution and adaptation rule the -world except in matters of dispersal. Take, for instance, the fleshy -fruits which the gardener often makes more attractive to birds than they -are in the wild condition. The result is certainly to increase their -facilities for dispersal by birds; but such a result was as little -intended by man as it was by Nature when species of Cornus, Ficus, -Prunus, Viburnum, and other genera matured their drupes, berries, and -fleshy fruits in the Cretaceous epoch. - -Children are now taught in several excellent little books on -“Nature-Study” that fleshy fruits are specially adapted to be eaten by -animals to secure the distribution of the seeds. We read in one book -that plants produce these fruits “on purpose to be eaten,” in another -that they are “intended to be eaten,” and in a third that the -seed-coverings are adaptations, all with the ulterior object of -distribution by frugivorous animals. I must be pardoned if I venture to -express my dissent from these statements, more especially since they are -made by authors from whom it might be thought almost impertinent for me -to differ. Yet authority can be claimed for holding the opposite view. - -When the botanist speaks of “useless secretions” in a plant, he is -alluding amongst other things to the sugar and organic acids of fruits. -“How and why all these substances originate is,” as Professor Sachs -observed in the work before quoted, “not known.” It is, however, -suggested by Dr. Kerner, in his _Natural History of Plants_ (Engl. edit. -i, 460-462), that such secretions, though useless to plants, may exist -for the purpose of alluring animals to assist in seed-dispersal. There -are some botanists, it may be remarked, that would reject such a view of -the nature of fruits. Dr. Stapf in his memoir on the flora of Kinabalu -observes in this connection that the fact that a fruit is fleshy and -attractive to birds is “no proof that it is really devoured by them, and -still less that it is dispersed by them.” Neither in fleshy fruits, nor -in minute seeds, nor in seeds capable of being transported by the wind -does he regard the general object of the particular character as -primarily to act as a means of dispersion. - -The same plea is made for the mucosity of seeds like those of Capsella -and Plantago (see Note 43), or for the “stickiness” of other seeds and -fruits like those of Pisonia, qualities that favour adherence to passing -objects. This is the reason, we are told, why seeds are “sticky.” Such -secretions I infer are often materials lost to the plant; and being in -that sense excretory we are not called on to supply a use for them. They -can, therefore, not be regarded as having any teleological significance, -since adaptation arises only from the requirements of the plant’s -conditions of existence. If they are serviceable in assisting the -distribution of seeds, such an event can only be described as an -accident in the plant’s life arising from chance contact with another -environment. - -The appendages of seeds and fruits, such as hooks and hairs, that render -them liable to adhere to fur or feathers, are also regarded as special -adaptations to this end. Without entering into the physiological -significance of hairs and prickles generally, concerning which, as many -of my readers will know, much might be said not in favour of such a -view, I would refer to cases like that of Cæsalpinia Bonducella, where -the large prickly pods could not possibly be intended to aid the plant’s -dispersal, whilst the leaf-branches are also prickly, and the seeds are -well known to be distributed by the currents. There are other cases like -that of Bidens cernua where the achenes, by reason of their barbed -bristles, and on account of a layer of “buoyant tissue” in the -fruit-coats, are dispersed both by birds and by water. We may fitly ask -to which capacity the theory of adaptation should be applied. Spiny -fruits may be sometimes so large, as in the instance of Trapa natans, -that the question of adaptation to dispersal cannot be raised. - -The great variety of the modes of dispersal of seeds is in itself an -indication that the dispersing agencies avail themselves in a hap-hazard -fashion of characters and capacities that have been developed in other -connections. Seeds and fruits, having developed certain characters under -a particular set of life-conditions, on being detached from the parent -plant are brought into contact with conditions quite outside their -original environment. Qualities and capacities are then brought into -play which have no connection with the life-history of the plant. The -care with which the mother plant guards the maturing seeds, and the -protection of the environment, are at a certain period withdrawn, and -the seeds are left to take their chance under strange conditions. It -would be idle to see anything purposeful in the waste that results. -Rather we would see in it the effect of one of the numerous limitations -of the determining or evolutionary power in Nature. Such a power has to -adapt its workings to the laws of the physical world, checked here, -frustrated there, at times, as in this particular case, losing all -control, but in the end prevailing. - -My general position may be thus summarised. As concerning the -distribution of fruits and seeds, the dispersing agencies take advantage -of characters and capacities that were never intended for them, -characters and qualities indeed that are often only brought out in -relation to another environment. Thus no question of adaptation as -regards means of dispersal can arise, since such capacities for -dispersal have no connection with the plant’s life-history. That seeds -are dispersed at all is a blind result of the ever-continued struggle -between the opposing forces of evolution and adaptation; that is to say, -between the determining power that lies behind organic life and the -physical conditions to which it has to adapt its ends. - - - - - CHAPTER XII - - THE CAUSES OF THE BUOYANCY OF SEEDS AND FRUITS OF LITTORAL PLANTS WITH - ESPECIAL REFERENCE TO THOSE OF THE PACIFIC ISLANDS - -The classification of buoyant seeds and fruits.—The first group, where - the cavity of the seed or seedvessel is incompletely filled.—The - second group, where the kernel is buoyant.—The third group, where - there is air-bearing tissue in the seed-tests or fruit-coats.—The - buoyant seeds and seedvessels of the littoral plants of the British - flora.—Summary. - - -IN the following pages I have adopted in its main features the -classification of buoyant seeds and fruits employed by Professor -Schimper in his work on the strand-flora of the Indo-Malayan region. The -causes of buoyancy, as he points out, are very various, but they can be -arranged in a few categories; each category, however, usually admitting -great variety within its limits. It is this want of uniformity that -first attracts our attention when we come to study the structure of -seeds and fruits from the standpoint of their buoyancy. Whilst in the -Pacific I went over most of the field traversed by Professor Schimper in -Malaya (the majority of littoral plants of these regions being common to -both), and as a result I have added not a few plants to his original -groups. - -It will be seen from the following synopsis that there are three -principal groups. The first group includes those seeds and fruits where -the buoyancy is derived from unfilled space in the seed or fruit cavity. -The second group comprises those seeds or fruits where the floating -power is due to the buoyant kernel or nucleus. The third group includes -those where the buoyancy arises from the existence of air-bearing tissue -in the coverings of the seed or fruit. - -The first two groups I will term the mechanical or non-adaptive groups, -not only on account of the structure inducing the buoyancy, but because, -as Professor Schimper remarks, the same structure often occurs with -inland fruits and seeds possessing little or no floating power. In many -of these cases, as he points out, the question of adaptation to -dispersal by ocean currents cannot, therefore, be raised. The third -group may be named the adaptation group, because it is on these examples -of buoyant seeds and fruits that this investigator chiefly based his -contention that in the main the structures concerned with buoyancy -represent adaptations to dispersal by currents effected through the -agency of Natural Selection. It is accordingly to this group that -Professor Schimper especially directed his attention, and the result of -his observations made in the home of the plants and of his -investigations in the laboratory has been the elucidation of many -difficult points in the structure of their fruits and seeds. To the two -“mechanical” groups he did not pay the same attention; and as their -examination came more within the limits of my own capacity as an -inquirer I have worked them out with some detail, the subdivisions of -the first group being my own as well as much of the material. - - -_Synopsis of the buoyant fruits and seeds of littoral plants of the -tropical Pacific classified according to the cause of buoyancy._ (The -authorities are indicated by the initial letter, S = Schimper, G = -Guppy. Details are given under some of the species in latter part of -volume.) - -FIRST GROUP.—The floating power is derived from unoccupied space in the - cavity of the seed or fruit, no part of the seed or fruit as a rule - possessing independent floating power. - - SUB-GROUP I., where the seed is concerned. - - SECTION I. The seeds have little or no albumen, and neither the tests - nor the seed-contents have any buoyancy. The cotyledons are generally - large, foliaceous, and crumpled or folded, or otherwise arranged, so - that the seed-cavity is incompletely filled. - - S. G. Hibiscus tiliaceus. - G. Hibiscus diversifolius. - S. G. Thespesia populnea. - S. Suriana maritima. - G. Kleinhovia hospita, _variable_. - S. G. Colubrina asiatica. - S. Dodonæa viscosa. - G. Argyreia tiliæfolia, _variable_. - G. Ipomœa bona nox, _variable_. - G. Ipomœa glaberrima, Boj. - S. G. Ipomœa grandiflora. - S. G. Ipomœa pes capræ. - G. Ipomœa turpethum, _variable_. - G. Cassytha filiformis. - S. Euphorbia atoto. - -_Notes._—The species marked “variable” have seeds that sometimes sink -and sometimes float. With the exception of Kleinhovia they are only at -times littoral in station. - -The plants of the British flora are represented by Convolvulus -soldanella and C. sepium, the last being “variable” and not a littoral -species. - - SECTION II. All the seeds belong to the Leguminosæ. Neither the tests - nor the seed-contents have any buoyancy, the floating power arising - from a large central cavity produced by the bending outward of the - cotyledons during the final shrinking stage of the maturation of the - seed. - - S. Mucuna (generically). - G. Mucuna urens D.C. (Hawaii). - G. Mucuna, species of. - S. G. Vigna lutea. - S. G. Cæsalpinia bonducella. - G. Cæsalpinia bonduc. - G. Entada scandens. - - SUB-GROUP II., where the fruit is concerned. - - SECTION III. The seed only partially fills the fruit-cavity, and as a - rule is not buoyant. The fruit shell, usually woody, may be also - buoyant. - - S. G. Heritiera littoralis. - G. Smythea pacifica. - G. Dalbergia monosperma. - S. G. Derris uliginosa. - S. G. Pongamia glabra. - G. Desmodium umbellatum. - G. Gyrocarpus jacquini. - - SECTION IV. The floating power is derived from empty seed-cavities, - where owing to abortion of the ovule or some similar cause the seed - is not developed. - - S. G. Morinda citrifolia. - G. Premna tahitensis. - -_Note._—Professor Schimper, in the case of Morinda citrifolia, holds the -view that we have here a special adaptation to dispersal by currents. - -SECOND GROUP.—Here the floating power is due mainly or entirely to - buoyant kernels. In the case of seeds the tests are non-buoyant; but - with “stones” the floating capacity may be aided by a layer of - air-bearing tissue inside the shell. - - SECTION I. Non-Leguminous. - - S. G. Ximenia americana (drupe). - S. G. Calophyllum inophyllum (drupe). - -_Note._—Professor Schimper would place these two plants in the second -section of the third group on account of the layer of air-bearing tissue -inside the shell of the “stone”; but they are assigned to this section, -since the floating power is mainly due to the buoyant kernel. - -Arenaria (Honckeneya) peploides, a British beach plant, belongs here. - - SECTION II. Leguminous seeds. - - G. Dioclea. - G. Strongylodon lucidum. - S. Canavalia (generic). - G. Canavalia sericea. - S. G. Canavalia obtusifolia. - S. Erythrina (generic). - S. G. Erythrina indica. - P. Erythrina ovalifolia (Penzig). - S. G. Sophora tomentosa. - G. Afzelia bijuga. - G. Lathyrus? - -THIRD GROUP.—The floating power is due to the presence of air-bearing - tissue in the seed-tests or fruit-coats. - - SECTION I. The buoyant tissue occurs at the outside or forms the - periphery of the seed or fruit. Unless otherwise indicated the fruit - is implied in the list below. - - S. G. Carapa moluccensis (seed). - S. G. Carapa obovata (seed). - G. Inocarpus edulis. - G. Serianthes myriadenia. - G. Parinarium laurinum. - S. G. Barringtonia speciosa. - G. Barringtonia racemosa. - S. G. Pemphis acidula (seed). - S. Terminalia (generic). - S. G. Terminalia katappa. - G. Terminalia litorea. - S. Lumnitzera (generic). - S. G. Lumnitzera coccinea. - S. G. Guettarda speciosa. - G. Wedelia strigulosa. - S. G. Scævola Kœnigii. - S. G. Cerbera Odollam. - G. Ochrosia parviflora. - S. G. Cordia subcordata. - S. G. Tournefortia argentea. - S. G. Clerodendron inerme. - G. Vitex trifolia. - G. Vitex trifolia, var. unifoliolata. - G. Tacca pinnatifida (seed). - S. Nipa fruticans. - S. Cocos nucifera. - G. Scirpodendron costatum. - -_Additions of shore-plants from Malaya and tropical America mostly given -in Schimper’s work on the Indo-Malayan strand-flora._ - - S. Cynometra cauliflora. - S. Conocarpus erectus. - S. G. Laguncularia racemosa. - S. Lumnitzera racemosa. - S. Sonneratia (seed). - S. Barringtonia excelsa. - S. Scyphiphora hydrophyllacea. - S. Wollastonia glabrata. - G. Hippomane mancinella. - -_Note._—Here belong a species of Vitex, probably V. agnus castus, the -fruits of which occur in the stranded drift of the Sicilian beaches, and -also the British littoral shore-plants, Cakile maritima, Crithmum -maritimum, Matricaria inodora, and Scirpus maritimus. - - SECTION II. The buoyant tissue forms a layer inside the hard test of a - seed or inside the shell of the “stone” of a drupaceous fruit, and to - this cause the floating power is mainly or entirely due. - - G. Mucuna gigantea (seed). - S. Hernandia peltata. - S. Excæcaria agallocha. - S. Cycas circinalis. - S. Pandanus odoratissimus. - G. Anona paludosa (seed) of tropical America. - -_Note._—I have followed Schimper in respect to Pandanus, but it might be -by some placed in the first section of this group. - -Here belongs Euphorbia paralias, a British littoral plant, the buoyant -seeds of which occur in the stranded seed-drift of English and -Mediterranean beaches. - - -In the following general discussion of the groups, reference will be -made only to the plants best illustrating the different varieties of -structure connected with buoyancy; whilst mention of the other plants -will in some cases be found in other parts of this volume, as shown in -the Index; and the matter is discussed at some length in not a few of -the species. - - - THE FIRST GROUP. - -Of the first group, where the floating power is due to the unoccupied -space in the cavity of the seed or fruit, the Convolvulaceæ offer the -most typical examples. Here as a rule the crumpled embryo fills the -seed-cavity more or less incompletely; and it is on the relative size of -the unoccupied space that the sinking or floating of the seed depends. -In those plants where the seed sinks the seed-cavity may be almost -filled, as in Ipomœa tuberculata, or densely packed, as in Ipomœa -pentaphylla, and in species of Cuscuta. When the seed floats, as with -Ipomœa pes capræ, I. glaberrima, &c., the unoccupied space is relatively -large; and when, as with I. bona nox and I. turpethum, the behaviour of -the seeds is irregular, some floating, and others sinking, a -corresponding variation exists in the extent to which the seed-cavity is -filled. This applies also to the irregular behaviour of the seeds of -Ipomœa peltata and of Argyreia tiliæfolia. A singular instance is -afforded by the seeds of Ipomœa insularis, collected by me in Fiji and -Hawaii. Those from Fiji were incompletely filled, and consequently -buoyant. Those from Hawaii were more densely packed and sank.... The -three British species of Convolvulus illustrate the same principle, -namely, C. arvensis, with non-buoyant seeds; C. soldanella, with buoyant -seeds; and C. sepium, with seeds irregular in behaviour. - -In the case of plants of the Convolvulaceæ, possessing buoyant seeds, -there is always evidence of marked shrinking of the seed-contents before -the final setting and hardening of the seed-coats. The embryo often -appears shrivelled and dried up, and is almost brittle, so that large -spaces are produced in the seed-cavity. If we partly divide such a seed -and place it in water, the embryo absorbs water rapidly, and within an -hour is soft, healthy-looking, and much swollen, the interspaces being -filled with a jelly-like mucilage. It is therefore evident that absolute -impermeability of the seed-coats is essential for the successful -transport by sea-currents of the floating seed; and we can only suppose -that the shrinking of the seed-contents takes place before the final -setting of the tests. That with the buoyant seeds the coats are quite -waterproof was illustrated in many of my experiments where, after a -period of flotation covering several months, and sometimes a year or -more, the seed-contents were still quite dry and shrunken. The limit of -buoyancy, as I have shown in Chapter IX., depends on an attempt at -germination on the part of the floating seed, which then absorbs water, -softens, swells, and sinks. - -It is, therefore, not a matter of surprise that non-buoyant seeds of the -Convolvulaceæ do not gain floating power after prolonged drying of many -months. It is also to be expected that, as we find in Fiji, when a -characteristic shore-species with buoyant seeds like Ipomœa pes capræ -extends far inland, the seeds retain their floating powers. -Seed-buoyancy of this description is, on the face of it, purely -mechanical. - - - EXPLANATION OF THE DIAGRAMS ILLUSTRATING THE CAUSES OF SEED-BUOYANCY - -1. _Entada scandens_ (natural size): (_a_), the shell; (_b_), the - kernel; (_c_), the intercotyledonary cavity. The shell consists of - three coats—an outer and an inner hard chitinous coat, and an - intermediate layer of brown cellular tissue containing little or no - air. The buoyancy is due entirely to the central cavity, neither the - seed-tests nor the seed contents possessing any floating power (see - page 181). - -2. _Mucuna urens_, from Hawaii (natural size). The kernel (_b_) sinks, - and the shell has no floating power except where it possesses (under - the raphe) a layer of dark brown, air-bearing, spongy tissue (_a_). - This, however, is not sufficiently developed to endow the seed with - buoyancy, which is due to the intercotyledonary cavity (_c_). (see - page 111). - -3., 4. _Mucuna gigantea_, from Fiji (natural size). The kernel (_b_) - sinks, and the seed owes its floating power entirely to the existence - in the shell (_a_) of a layer of brown, spongy, air-bearing tissue - which is mostly developed at the circumference and is almost wanting - at the flat sides of the seed (see page 115). - -5., 6. _Dioclea_ (_violacea?_), from Fiji (natural size). Here the - kernel (_b_) is buoyant and endows the seed with floating power. - Though the shell (_a_) possesses a thick layer of reddish-brown - cellular tissue, this tissue contains but little air and aids the - floating power but slightly (see page 113). - -7. _Strongylodon lucidum_, from Fiji (natural size). The floating power - is due entirely to the buoyant kernel (_b_). There is a very scanty - amount of loose brown tissue (_a_) under the raphe; but it has no - appreciable effect on the buoyancy (see page 113). - -8., 9., 10. _Cæsalpinia bonducella_ and _C. bonduc_, from Fiji (natural - size). Neither the seed-tests (_a_) nor the kernel (_b_) have any - floating power in themselves, the buoyancy being connected with a - large internal cavity (_c_), which normally is intercotyledonary, as - in Fig. 8 (C. bonducella). With both plants, but more especially with - C. bonduc (Figs. 9 and 10), there may be a lateral cavity (_d_), or - the kernel may be loose in the shell (Fig. 10), but this does not - necessarily imply buoyancy (see page 194). - -11., 12. _Arenaria peploides_ (enlarged: seeds 4 mm. in size). Here the - curved embryo (_a_) sinks, and the spongy air-bearing albumen (_b_) - gives buoyancy to the seed (see page 116). - -13. _Euphorbia paralias_ (enlarged: seeds 3 mm. in size). The kernel - (_b_) sinks, and the seed owes its buoyancy to a layer of air-bearing - tissue (_a_) in the shell (see page 116). - -14. _Morinda citrifolia_ (enlarged pyrene 7 mm. long). The floating - power is due to the bladder-like air cavity (_a_). The seed (_b_) - proper is enclosed in the woody tissue behind the bladder (see page - 112). - -15. _Cucurbita_ (seed enlarged), from the Valparaiso beach-drift (see - page 125). The kernel (_b_) has no buoyancy. The shell (_a_) is formed - of two layers of air-bearing tissue, the outer composed of prismatic - cells and the inner of a spongy vacuola-material. - - [_To face page 111._ - -[Illustration: Diagrams illustrating some of the causes of -seed-buoyancy.] - -Another type of the buoyant seeds of the first group is presented by -several species of Leguminosæ, as with Entada scandens, some species of -Mucuna, and Cæsalpinia bonducella. As with the Convolvulaceous seed, the -embryo sinks and the seed-shell has no buoyancy; but here the floating -power is due to the existence of a more or less symmetrical long central -cavity produced by the arching or bending outwards of the large -cotyledons which lie usually in close contact with the seed-shell. This -arching outward of the cotyledons depends on a shrinking process in the -setting or final stage of the maturation of the seed. The stages of the -process may be traced in the immature seeds, which are much larger and -in some cases twice the size of the mature seed. In this immature -condition the seed-coats are soft, and the flabby fleshy and thick -cotyledons fill up the seed-cavity. As the hardening and setting process -continues, the cotyledons diminish in size, become firmer, and gradually -bend outward, leaving a central cavity. This arching outwards is no -doubt in part the result of the contraction of the seed-tests during the -shrinking process. Considerable variation prevails in the results, and -where the cavity is very small the seed sinks. Further details relating -to this subject will be given in my treatment of some of the plants, and -especially under Cæsalpinia. But it may be here remarked with reference -to Hawaiian seeds of Mucuna urens D.C., that although they are strictly -referable to this group, they display beneath the hard test, on the side -beneath the raphe, a scanty layer of dark spongy air-bearing tissue -which is sufficiently buoyant to float up detached portions of the test, -but does not of itself give buoyancy to the seed. The significance of -this structure will be subsequently pointed out. The seed owes its -floating power to the large central cavity, but this layer of spongy -tissue adds to its buoyancy. - -The section where the buoyancy of the fruit is connected with unoccupied -space in the fruit-cavity is extremely heterogeneous in its composition. -Every fruit has a method of its own, and the great variety of causes of -buoyancy of a mechanical character is here exemplified. For instance, -with Gyrocarpus jacquini and Cassytha filiformis the cause of buoyancy -is in the main the same as that described in the case of the -Convolvulaceæ. The origin of the floating power of the pods of Derris -uliginosa is two-fold. In the first place the seed or seeds but partly -fill the pod, and in the second place the seed is able to float of -itself by reason of its possessing, as in the seeds of Entada scandens, -a large central cavity produced by the arching out of the cotyledons -during the final stage of maturation. A double cause is also to be -assigned to the buoyancy of the fruits of Heritiera littoralis and of -Smythea pacifica, where, in addition to the unoccupied space produced by -the shrinking of the seed, the fruit-case itself floats, though nothing -but a mechanical explanation is to be given of the floating of empty -ligneous fruits. - -One of the most suggestive types of buoyancy belonging to the first -group is presented by those cases, which are, however, not very -frequent, where the floating power is to be attributed to empty -seed-cavities produced by the abortion of the ovule or failure of the -development of the seed. A significant instance of this is afforded by -the fruits of Premna taitensis, a coast plant. The buoyant “stone” of -the drupe, which is often found afloat in the Rewa estuary in Fiji, is -4-locular, each cell containing normally one seed, but as a rule only -one cavity contains a mature seed, the three other cavities becoming -more or less empty through the failure of their seeds. It can be proved -that neither the seeds nor the substance of the “stone” are buoyant, and -that the “stone” owes its capacity of floating for months to the empty -cavities arising from the failure in development of three out of the -four seeds. In Fiji we see the rivers distributing these small fruits, -and we find the “stones” stranded on the beaches and floating in the -currents amongst the islands; and there can be no doubt that this is one -of the effective modes of dispersal of the species; yet, if there was -ever a case of accidental buoyancy concerned with dispersal by currents, -we have it here. Further details are given in Note 32. - -It is probably also to the abortion of the ovule, or to the failure of -the seed, that the remarkable air-cavity (see Note 8) to which the -pyrenes of Morinda citrifolia owe their floating power, is to be -attributed. To this structure Professor Schimper (pp. 165, 183, 200) -attaches considerable importance as an example of special adaptation to -dispersal by currents through the influence of Natural Selection. He -suggests, however, that possibly its morphological significance may be -found in its being a peculiarly modified seed-chamber. The case of -Premna taitensis above cited indicates that the latter view is the most -probable. The subject awaits a careful microscopical study of the -seed-development of the genus Morinda since, as elsewhere remarked, the -non-buoyant pyrenes of inland species have not such an air-chamber. An -outline sketch of a pyrene of Morinda citrifolia is given in the -preceding plate. A good figure of it occurs in Schimper’s _Plant -Geography_, p. 28. A very suggestive instance of this nature is -described under Brackenridgea in Note 46 and in Chapter XIII. - - - THE SECOND GROUP. - -Here are included those seeds and stone-fruits that possess buoyant -kernels. Professor Schimper points out that since this is a feature both -with inland as well as coast plants such a character cannot be viewed as -an adaptation to dispersal by currents. The plants concerned belong -mostly to the Leguminosæ, and we find here some of the most widely -spread of strand species, such as Canavalia obtusifolia and Sophora -tomentosa, as well as some of the giant climbers of the coast forests -belonging to the genera Dioclea and Strongylodon. The kernels when -divested of their coverings float buoyantly, but they soon absorb water -and sink usually in a day or two, a circumstance indicating that it is -to the impervious coverings that they indirectly owe their capacity to -keep the seed or fruit afloat. It is noteworthy that seeds of -Strongylodon lucidum from Fiji display beneath the raphe a trace of an -internal layer of loose cellular tissue which, however, has no -appreciable effect on the buoyancy; whilst with seeds of Dioclea -(violacea?) from the same locality there is a thick layer of loose -tissue which aids the floating power of the kernel but is not of itself -sufficiently aeriferous to buoy up the seed. - -This leads one to refer to two other plants belonging to this group, -Calophyllum inophyllum (Guttiferæ) and Ximenia americana (Olacineæ), -where, though the floating power is mainly due to the buoyant kernel, it -is also aided by a layer of air-bearing tissue inside the hard shell of -the “stone” of the drupe. Professor Schimper places these fruits in the -third or adaptive group on account of the layer of buoyant tissue, but -it would be more correct to class them according to the predominant -cause of their buoyancy. It can be shown that with a non-buoyant kernel -the “stone” no longer floats. This double cause of the floating power -renders an explanation very difficult, since it would seem indefensible -to give conflicting interpretations of their nature. With Ximenia -americana there is another great difficulty. Its drupes are known to be -dispersed by fruit-pigeons (_Introd. Chall. Bot._ p. 46); and judging -from the rare occurrence of the “stones” in the drift there is good -reason to believe that bird agency in the Western Pacific is predominant -in the dispersal of the plant. It is by such test cases as this that we -must put to the proof the reality or non-reality of the influence of -adaptation on seed-buoyancy. - - - THE THIRD GROUP. - -We have here those plants where the floating-power is entirely or mainly -due to an air-bearing tissue in the seed-tests or fruit-coats. Several -of the fruits are figured in Schimper’s _Indo-malayische Strand-flora_, -and one or two are figured in the English edition of his work on -_Plant-Geography_, p. 29. - -In the first section, where the buoyant tissue occurs at the outside or -forms the periphery of the seed or fruit, are included several of the -most familiar of the littoral trees and shrubs of the Pacific islands, -such as Barringtonia speciosa, Cerbera Odollam, Guettarda speciosa, -Pemphis acidula, Scævola Kœnigii, Terminalia katappa, and several others -named in the synopsis. I cannot enter into detail here, but the reader -will find fuller particulars of each plant in most cases in Professor -Schimper’s work, and in some instances in my separate discussion of the -plants concerned. In nearly all cases we are concerned here with the -fruits, and only in a few cases with the seeds, as with Carapa and -Pemphis acidula. - -This investigator observes that to this sub-group belong the fruits and -seeds usually described in systematic works as provided with corky or -suberous coverings; but he points out (p. 167) that the resemblance is -nearly always quite superficial, and is limited to colour and -consistence, suberous tissue occurring in only a few cases, as in the -fruit-coats of Clerodendron inerme. The buoyant tissues, he remarks, are -often more or less ligneous, and in those cases where there is no lignin -reaction they resist the action of sulphuric acid much more effectively -than pure cellulose; whilst in their physical characters, as well as in -their behaviour with reagents, they differ just as much from ordinary -cork. Thus, they are but little elastic and often easily crumble away; -whilst in large fruits, like those of Cerbera and Terminalia, they would -soon be stripped off entirely when subjected to the “wear-and-tear” of -transport by currents, if they were not traversed by numbers of stout, -tough fibres which hold the materials together. Where the buoyant -tissues are firmer, as with Clerodendron inerme and Cordia subcordata, -the fibrous framework is scanty or absent, whilst very small seeds or -fruits, like those of Tournefortia argentea and Pemphis acidula, where -the “wear-and-tear” would be comparatively slight, often possess no -protecting fibres in the buoyant tissues. - -In one or two fruits, like those of Cerbera Odollam, these tissues -display large intercellular spaces; but in the majority of cases such -spaces are insignificant in size or absent altogether. Speaking -generally, however, there is, as Professor Schimper observes, great -similarity in the structure of the buoyant tissue in the coverings of -these fruits and seeds. The cell-walls are thin or only slightly -thickened, and detached air-bearing portions of the tissue will float -for many weeks. The great floating capacity of these fruits and seeds is -stated by this investigator to be entirely due to the tenacity with -which the air is retained in the covering tissues. It is, however, -noteworthy that in the case of Scævola Kœnigii the fruits are just as -well suited for dispersal by frugivorous birds as by the currents, a -significant circumstance discussed in the next chapter. - -The second section contains those plants where the buoyant tissue occurs -inside the hard shell of the fruit or seed, such as is found, for -example, in Anona paludosa, Mucuna gigantea, Hernandia peltata, Cycas -circinalis, &c. Professor Schimper here includes Calophyllum inophyllum -and Ximenia americana; but I have before remarked that the buoyancy of -their fruits is mainly due to their buoyant kernels. This aeriferous -tissue forms a layer between the seed or nucleus and the hard outer -shell. It is described by the above-named authority as soft or friable -and dark brown. The cells contain air and may be closely arranged or -separated by small interspaces, their walls being neither woody nor -suberous. - - - _The structure of the buoyant seeds and seedvessels of the littoral - plants of the British flora._ - -The littoral plants with floating seeds or fruits form but a section of -the strand-plants of the British flora, scarcely a third, as is pointed -out in Chapter IV., of the total number. Though small in number they -exhibit great variety in structure; and notwithstanding that as far as -they have been examined they may all be referred to one or other of the -groups and sections of the classification adopted in the synopsis for -the plants of the Pacific islands, nearly every plant presents in the -structure of its seeds or seedvessels a type of buoyant structure -different from the others. - -The first group is represented by the seeds of Convolvulus soldanella, -which owe their floating power to the incomplete filling of the -seed-cavity. The second group, where the buoyancy arises from the -buoyancy of the kernel or nucleus, is illustrated by the seeds of -Arenaria (Honckeneya) peploides, but in a fashion quite unique. The test -is thin but impervious, and has no buoyancy; the curved embryo also -sinks; and the floating power arises from the air contained in the loose -spongy albumen, around which the embryo is coiled (see figure). A more -normal component of the second group is represented in some Leguminous -seeds, perhaps of Lathyrus maritimus, that occur regularly amongst the -stranded seed-drift of the north coast of Devon. Here the kernel of the -seed is buoyant. The seeds of Euphorbia paralias are indebted for their -floating capacity to a layer of spongy tissue containing large -air-spaces placed between the kernel and the chitinous outer test, -neither of which possess any floating power (see figure). They thus -belong to the second section of the third group. - -The fruits of Cakile maritima, Crithmum maritimum, Matricaria inodora, -and Scirpus maritimus, all belong to the first subdivision of the third -group where the air-bearing tissue exists in the peripheral coverings, -the seed or nucleus in all cases sinking. With Cakile maritima there is -a light spongy outer case of aeriferous tissue, which, however, soon -loses the epidermis, a circumstance that probably explains the limited -period of flotation of about a week. The walls of the mericarp of -Crithmum maritimum are composed of spongy cellular air-bearing tissue -with a persistent epidermis, and the floating powers of the fruits are -consequently great. The achenes of Matricaria inodora have beneath the -epidermis a layer of buoyant tissue, and their structure is similar to -that found with the buoyant achenes of littoral species of Wedelia, -plants of the same order of Compositæ that are found on the Pacific -islands. The cause of the floating power of the fruits of Scirpus -maritimus lies entirely, according to Kolpin Ravn, in the air-bearing -cells of the epidermis. The reader will find the results of my -experiments on the buoyancy of the seeds in Notes 16, 17, and 18. - - - _Summary of the Chapter._ - -(1) Following the main lines of Schimper’s classification of those of -the Indo-Malayan region which possesses for the most part the same -species, the buoyant seeds and fruits of the littoral plants of the -Pacific islands are classed in three groups: the _first_ where the -cavity of the seed or fruit is incompletely filled, the floating power -arising from the empty space; the _second_ where the buoyancy is derived -from the buoyant nucleus or kernel; and the _third_ where it arises from -air-bearing tissues in the coats of the seed or fruit. - -(2) The first and second groups, in which the question of adaptation to -distribution by currents through the agency of Natural Selection is not -raised, since the same structural characters are found in seeds and -fruits of inland plants not dispersed by the currents, are termed the -mechanical or non-adaptive groups. The third is distinguished as the -adaptive group, because it is here that Schimper finds evidence in -favour of the Selection Theory. - -(3) The first group is best represented by the Convolvulaceous and the -Leguminous types. In the former, which is well illustrated by Ipomœa pes -capræ, the seed-cavity is imperfectly filled by the crumpled embryo, the -result of the shrinking process during the final setting of the seed. In -the latter, which is exemplified by Entada scandens and Cæsalpinia -bonducella, the seed displays a large central cavity produced by the -arching outward of the cotyledons during the shrinking process -accompanying the last stage of the maturation of the seed. As an -instance of fruits belonging to the group, those of Heritiera littoralis -may be cited. An uncommon type is presented in the “stones” of the -drupes of Premna taitensis, and in the pyrenes of Morinda citrifolia, -where the buoyancy arises from empty seed-cavities resulting from the -failure of some of the seeds. - -(4) The second group with buoyant kernels includes mostly widespread -Leguminous species, such as Canavalia obtusifolia and Sophora tomentosa. - -(5) The third or “adaptive” group comprises many of the characteristic -littoral trees and shrubs of the Pacific islands, such as Barringtonia -speciosa, Guettarda speciosa, Terminalia katappa, Tournefortia argentea, -&c., that contain in their fruit-coverings a buoyant cork-like material -often bound together by fibres, but which proves on examination to -resemble cork only in appearance. In another type, illustrated by the -fruits of Cycas circinalis and the seeds of Anona paludosa, the buoyant -tissue forms a layer inside the shell of the seed or “stone.” - -(6) Some fruits like those of Ximenia americana and Calophyllum -inophyllum illustrate both the so-called mechanical and adaptive -principles in their structure; whilst with the first-named species they -are as well adapted for dispersal by frugivorous birds and are known to -be a favourite food of fruit-pigeons. The same difficulty arises with -the fruits of some other characteristic littoral plants, as with Scævola -Kœnigii, the drupes of which are equally well fitted for dispersal by -birds and currents. - -(7) The same general principles have been at work in determining the -structures concerned with the buoyancy of the fruits and seeds of -British littoral plants. Although the species are few in number they -exhibit in this respect great variety, eight species illustrating six or -seven types of buoyant structure. - - - - - CHAPTER XIII - - ADAPTATION AND SEED-BUOYANCY - -The question of the operation of Natural Selection.—Are there two - principles at work?—The presence of buoyant tissue in the seed-tests - and fruit-coats of inland plants, both wild and cultivated.—Useless - buoyancy.—The buoyancy of seeds and fruits is not concerned with - adaptation.—Summary. - - -WHEN we speak of a certain structure as an adaptation to dispersal by -currents through the agency of Natural Selection, it is necessary at the -outset to be quite clear as to what is implied. Professor Schimper, who -brought his great and varied knowledge of many other phases of -plant-life to bear on this subject, is careful to clear the ground of -preliminary erroneous conceptions in such a perspicuous and impartial -manner that we cannot do better than follow his guidance. There are, he -observes (p. 178), many mechanisms or contrivances in plants, which, -though they seem to have arisen with a fixed purpose, can in no wise be -regarded as having been developed for that end, since they were produced -in quite a different connection and have merely acquired a new or -supplementary function, of which they are the cause and not the effect. - -This is very much the position that I have taken up for the whole -subject of the relation between plants and their dispersing agencies, -and it will be found discussed in Chapter XI. It involves, as I venture -to think, a dominant principle in the organic world, which it is one of -the objects of this work to emphasise, namely, that Nature in dispersing -plants habitually makes use of structures and capacities that were -originally developed in quite another connection. Behind this change of -function, this new purpose, lies the secret of the organic world. There -is for me no more pregnant fact in plant-life than the thistle-seed -blown before the wind, or the seed of our sea-convolvulus floating in -the sea. It proves to my mind that the evolutionary power in nature is -checked and hampered by conditions not of its own creation, and that two -opposing forces are ever at work, the one creating and the other -limiting the creative power, the actual mode of dispersal being but a -blind and accidental result of the struggle. - -The question of the operation of Natural Selection is not raised, as -Professor Schimper indicates, until we consider whether the new function -has had any bettering influence on the structure or mechanism with which -it has come to be concerned. If such a modification is thus brought -about it might be legitimately claimed as a result of this agency, and -the term “adaptation” could be used. But if there is no evident change -produced, we should be compelled to assign very subordinate limits to -the capacity of Natural Selection; and in the instance of buoyant fruits -and seeds it would be restricted to determining a plant’s station by the -water-side and in increasing its area. It is only in the first case that -we could speak of them as adaptations in the meaning attached to this -term in the language of the Selection Theory. It would at first sight -seem easy to ascertain whether the characters of fruits and seeds, to -which the buoyancy is due, are adaptations in this sense of the word; -but in reality it is far from being so. We can, however, proceed with -unanimity up to a certain stage in the argument; but there agreement -ends. - -It has been before established that in the Pacific islands, and indeed -in the tropics generally, the plants with buoyant seeds or seedvessels -are mainly stationed at the coast. It has also already been shown that -this littoral station is often associated with a special buoyant-tissue -in the coverings of the seed or fruit; and it will now be pointed out -that this tissue is, as a rule, absent or but scantily developed in the -case of inland species of the same genus. Of great importance, remarks -Professor Schimper (p. 179), in relation to the Selection Theory and the -development of adaptations, is the comparison of the fruits and seeds of -strand-plants with those of allied inland species; and he finds here -evidence in support of the Darwinian view. He takes the cases of the -genera Terminalia and Calophyllum, which contain both inland and -littoral species; and he shows that although the same buoyant-tissue -occurs in the fruit-coats of inland species, it is there much -diminished, and in consequence the floating powers are considerably -lessened or lost altogether (see Chapter II.). It is not pretended that -this tissue has had any connection in its origin with dispersal by -currents, but merely that its greater development in the shore species -is an adaptation to this mode of transport. - -Further testimony is adduced by this investigator (p. 182) in supporting -his view in the fruits of the genera Barringtonia, Clerodendron, Cordia, -and Guettarda, where the buoyant tissues extensively developed in the -coast species are either non-existent or only represented by a trace in -the inland species of the same genus, a difference in structure -associated with the loss or great diminution of the floating capacity of -the fruits concerned. I have been able to establish other examples in -the cases of the genera Scævola and Tacca, which will be found referred -to in Chapter II. - -Professor Schimper (p. 200) points to the circumstance that the -“adaptations” in these fruits all belong to the diagnostic marks of the -genera and the species, and contends that these plants abundantly prove -the erroneous nature of the contention that Natural Selection could have -played no part in the elimination of the strand-flora. My own contention -is that Natural Selection has played such a part, but that in doing so -it has merely availed itself of characters previously existing, without -originating, modifying, or improving them in any way. The foregoing -evidence might with equal fitness be employed to show, as pointed out in -Chapter II., that in the course of ages there has been a great sorting -process by which, excluding the mangroves, plants of the xerophilous -habit possessing buoyant seeds and fruits have been sorted out and -placed at the coast. Direct evidence does not lead us farther than to -the establishment of a littoral station for plants thus endowed. The -problem whether the characters of their fruits and seeds that are -concerned with buoyancy may be regarded as adaptive in the Darwinian -sense lies beyond the reach of direct testimony. We can, however, -approach it from the outside by several directions, and from some of -these we will now proceed to deal with it. - -There is first the singular circumstance that in Fiji, when the littoral -plants with buoyant seeds or fruits leave the beach and extend far -inland, they, as a rule, retain their floating powers and, of course, -their buoyant structures. I found this to be true of Cassytha -filiformis, Cerbera Odollam, Ipomœa pes capræ, Morinda citrifolia, -Scævola Kœnigii, and one or two other plants mentioned in Note 44, where -this subject is discussed. My experiments on these plants indicated that -their fruits or seeds floated equally long, whether obtained from coast -or from inland plants. This, at first sight, appears to present a -serious objection to the adaptation theory; but it was not so regarded -by Professor Schimper, who in a letter to me, dated March 8th, 1900, -observed that he did not see “why littoral plants growing inland should -lose their adaptations to littoral life, especially if those adaptations -are not conflicting with the conditions of life beyond the littoral -zone, and if the competition does not require special adaptations.” - -My view, however, is that any process of adaptation is unnecessary. All -these plants, it is contended, were originally inland plants that -acquired the buoyant qualities of their seeds and fruits in the inland -stations, and ultimately found a station at the coast through the -sorting process above referred to. In the case of plants like Ipomœa pes -capræ and Cassytha filiformis this would be conceded, since they belong -to the acknowledged non-adaptive groups discussed in the preceding -chapter. It is only to some of these plants, such as Scævola Kœnigii and -Cerbera Odollam, that the adaptation view of Professor Schimper is -applied; and the question arises whether we are justified in making such -a distinction, or, in other words, whether it is antecedently probable -that two independent principles have been at work in determining the -fitness of seeds and fruits for dispersal by the currents. - -The plants for which the influence of adaptation through Natural -Selection is claimed belong, as stated in Chapter XII., almost entirely -to the third group. It is admitted that with the other two groups the -utmost that any sorting or selecting process would effect would be to -determine a station at the coast and to extend the area of distribution. -The numerical aspect of the question therefore acquires some importance; -and the reader’s attention is accordingly directed to the results -tabulated in Note 45, where it is shown (assuming for the time that -there is no difference of opinion about the adaptive significance of the -seeds and fruits concerned) that the plants of the third or adaptive -group make up only about half the total. It would therefore appear that -if the agencies of Natural Selection have been at work here either in -bettering or in developing buoyant structures, half of the shore-plants -with buoyant seeds or fruits have not come within their influence. - -But the subject takes another aspect when we reflect that in some -buoyant fruits, as with Ximenia americana and Calophyllum inophyllum, -the two principles would seem to have been at work. Whilst from this -standpoint Natural Selection is regarded as having either developed or -increased in amount the layer of buoyant tissue in the fruit-coats, the -buoyant kernels are not viewed as adaptive in their origin. In the case -of Ximenia americana the dispersing agency of frugivorous birds adds -another factor, since, as before stated, its drupes are known to be -dispersed by fruit-pigeons. In the cases of Scævola Kœnigii and of Vitex -trifolia, two plants belonging to the adaptive group, Professor Schimper -(pp. 156, 188) admits also the dispersing agency of frugivorous birds, -and he claims it for Morinda citrifolia, in the pyrenes of which he also -detects a special adaptation to dispersal by currents. It may be added -that, as he also points out, fruits of the non-adaptive group of -littoral plants, such as Premna integrifolia (P. taitensis) and Cassytha -filiformis, would sometimes also attract birds. In fact, those of the -last-named have been found in the crops of pigeons (_Introd. Chall. -Bot._, p. 46). - -Looking at all these littoral plants with fruits that are equally fitted -for dispersal by birds and by currents, we may now ask, Where does the -general principle of adaptation to dispersal lie? Whatever view we -adopt, we must apply the same view to all, whether it be a question of -dispersal by birds or by currents. We cannot choose between two sets of -principles determining the buoyancy of seeds and fruits any more than we -can regard a fleshy drupe and a buoyant seed as illustrating different -principles regulating the dispersal of plants. Nature works with -uniformity in these matters, and if the Natural Selection theory is held -to explain one case it ought to account for all. Yet nobody would go so -far as this; and this view of dispersal is on many grounds antecedently -improbable. These difficulties disappear if we assume that in all cases -the dispersing agencies have without modification made use of characters -and capacities that were developed, as we now see them, in quite other -connections and under quite other conditions. - -It will now be necessary to look a little closer into the subject of the -buoyant tissue, to the existence of which in their coats about half of -the littoral plants concerned owe the floating power of their fruits or -seeds. In the first place, it is to be remarked that in the case of some -of the seeds of the plants of the non-adaptive groups it is also -represented to a small degree in the seed-coats, although, as with -Strongylodon lucidum and Mucuna urens, it is not present in sufficient -amount to float the seed. In the next place, it should be noted that -with some genera possessing, like Terminalia, both inland and coast -species it is to be found alike in the fruit-coverings of inland and of -littoral plants, though in a less degree in the case of the fruits of -inland species, the floating power of which is proportionately -diminished. There are, however, a few cases where this buoyant tissue is -developed in inland species which belong to genera or subgenera that -have no littoral species. This is what we would expect, if Natural -Selection has merely concerned itself with placing plants of xerophilous -habit possessing buoyant seeds or fruits at the coast. Under such -conditions we would now and then expect to find an inland plant -possessing buoyant fruits or seeds of this description that has never -been able to establish itself at the coast. - -A good instance is afforded by Pritchardia Gaudichaudii, a fan palm -peculiar to Hawaii, the drupes of which float for several weeks and have -a covering of spongy buoyant tissue (see Chapter XXV.). The seeds of -Hibiscus Abelmoschus, a species distinguished subgenerically from the -littoral Hibiscus tiliaceus, offer another example. They float for -months, and owe their buoyancy to a layer of air-bearing tissue between -the kernel and the test, in this respect differing from the seeds of the -littoral species, where the floating power is due to unoccupied space in -the seed-cavity. The buoyancy of the seeds of Hibiscus Abelmoschus thus -offers another example of ineffectual floating power, since it is not a -littoral plant, is often cultivated, and has accompanied aboriginal man -over much of the tropical zone. - -A singular instance of the dispersal by currents of an inland plant that -occurs both wild and cultivated in tropical America, the West Indies, -and on the West Coast of Africa, is afforded by Spondias lutea, Linn., -which is referred to at the end of Chapter XXXII. Its “stones,” which -are provided with a cork-like covering much as we find with those of -Cordia subcordata and Guettarda speciosa, possess great buoyancy, and -are found in the river and beach drift of those regions with the seeds -in a sound condition. - -A very remarkable case of ineffectual buoyancy is presented by the -seedvessels of Brackenridgea, which have been found floating in the -drift off the coast of New Guinea. They owe their floating power to -closed cavities which would seem to arise from the failure of one of the -seeds or from the abortion of an ovule. But, according to Beccari, their -fleshy coverings would aid their dispersal by frugivorous birds; and -since the species are all much localised and are rarely littoral in -their habit, it is very probable that birds have mainly effected the -dispersal of the genus (see Note 46). It has, however, been shown in the -previous chapter that Premna taitensis and Morinda citrifolia owe their -dispersal by currents to similar cavities in the seeds or “stones.” - -Amongst the inland plants possessing seeds or fruits that are dispersed -by the currents without aiding the distribution of the species may be -recognised types of both the adaptive and non-adaptive groups. A -singular instance is afforded by the large seeds almost an inch long of -a huge pumpkin (Cucurbita) which, in sound condition, form one of the -commonest constituents of the beach drift on the coast of Chile from -Valparaiso northward to Iquique. The fruit is commonly eaten by the -lower classes. The seeds, which are very buoyant, contain a kernel that -does not float, the buoyancy being due to the water-tight coats which, -as shown in the plate in Chapter XII., possess well developed -air-bearing tissues. It may here be observed that Martins refers to the -germination of seeds of Cucurbita pepo after 45 and 93 days’ flotation -in sea-water. - -One sometimes finds buoyant tissue developed in the seeds of -bottle-gourds, where it can serve no useful purpose of dispersal. Thus -small bottle-gourds, seemingly of the genus Cucurbita rather than of -Lagenaria, are to be commonly found afloat in the Guayaquil River and -stranded on the Ecuador beaches. They will float for many months, and -contain the seeds dried up into a small loose compacted mass in their -interior. These seeds, which contain a layer of spongy air-bearing -tissue in their coverings, will in several cases float for months. Some -that I had been keeping two months afloat in sea-water germinated -freely. It is shown in Note 47 that bottle-gourds containing sound seeds -are dispersed far and wide by the currents. In some species the seeds -are buoyant, and in others they sink in sea-water; but the gourds -themselves will float for probably a year or more, and the floating -capacity of the seeds when it exists is too insignificant to affect the -fruit’s buoyancy. - -Other instances of the useless buoyancy of fruits of inland plants are -afforded by different species of Citrus. In the floating drift of the -Fijian rivers the fruits of the wild and indigenous Shaddock (C. -decumana) and of an inedible Orange, also wild and indigenous (C. -vulgaris?), are at certain times to be found, the latter often in -numbers. The first-named floats four to five weeks in sea-water, and the -last-named nearly two months, and both are to be observed floating out -at sea between the islands. The fruits of the Tahitian Orange, a variety -of C. aurantium, floated in sea-water between three and four weeks. The -seeds of these and other species of Citrus sank in from a few hours to a -day or two. The buoyancy of the fruit depends on the rind—the thicker -the rind the greater the floating power. This was not only shown in the -length of the period of flotation, but also in the buoyant behaviour of -the fruit. With the Tahitian Orange, where the rind is relatively thin, -the fruits floated heavily in sea-water and only protruded slightly -above the surface. With the Shaddock and with the other indigenous -species of Citrus, the fruits floated lightly and protruded half-way out -of the water. - -There is nothing trivial in these examples of buoyant fruits. That they -have at times aided in the dispersal of the genus, with man’s assistance -in planting the seeds of the stranded fruits, I cannot doubt; but -unaided by man such buoyant capacities would be useless for purposes of -effective dispersal by currents. Between the two genera Terminalia and -Citrus there is this great distinction, that the former is more or less -halophilous, some of its species being at home on the sea-beaches, -whilst the latter, as Schimper would term it, is salt-shy, and includes -no halophytes or plants of the sea-shore amongst its species. The only -effect of buoyancy of the fruits on the distribution of the species of -Citrus would be to place them by the side of the river and the pond. -This has evidently been its result in the case of the Shaddock in Fiji, -where, as Seemann remarks, it often thickly lines the banks of the -rivers. - -As also indicating that the buoyancy of the seed or fruit would never, -apart from the halophilous habit, endow an inland plant with a littoral -station, the examples of the Oak (Quercus robur) and of the Hazel -(Corylus avellana) may be taken. As shown in Note 48, these fruits -acquire floating power by drying, on account of the space formed by the -shrinking of the kernel. They occur commonly in beach drift, but rarely -in a sound condition; yet experiment has proved that they will sometimes -germinate after prolonged sea-water flotation. The fruits of other -species of Quercus are also transported in tropical regions by the -currents, but never, as far as I could learn, effectively. The Amentaceæ -as an order are “salt-shy,” and with only a few exceptions shun the -sea-beach. - -In the great sorting-process, by which xerophytic plants with buoyant -seeds or fruits have been placed at the coast, and hygrophytic plants -with similar fruits or seeds have been stationed at the riverside or by -ponds and lakes, one might expect to find that other influences may have -at times been in conflict with the selecting operation here indicated. -To this cause may probably be attributed the cases of “useless buoyancy” -above referred to. Here we find in some inland plants fruits and seeds -with buoyant tissues in their coverings that in the case of littoral -plants would have been regarded as the result of adaptation to dispersal -by currents. Such cases go to emphasize the conclusion already indicated -that these tissues could not have been developed through the agency of -Natural Selection. But the great objection against the application of -the Darwinian view to the general subject of the buoyancy of the seeds -and fruits of littoral plants lies in the circumstance that quite half -of the plants concerned are admitted to be outside the scope of the -theory, and that for these another explanation has to be found. I think -we may fairly claim that in a matter which finally resolves itself into -a question of buoyancy one explanation should cover all. We have thus to -decide whether to regard as adaptations to dispersal by currents the -structures of the buoyant seeds and fruits of littoral plants; or -whether to hold the view that as far as dispersal by currents is -concerned such structures are purely accidental, and that Nature has -never directly concerned herself in the matter at all. The first -explanation lies under the disadvantage above alluded to, and it remains -to be learned whether the second view could be made to cover all cases -of dispersal by currents. Further investigation on many points is yet -required; but, apart from the evidence against Natural Selection as the -principal agency that has been produced in this chapter, a powerful -argument in favour of the view that the buoyancy of seeds and fruits is -not concerned with adaptation is, that as a rule the floating capacity -of the seed or fruit has no direct relation with the density of -sea-water. Generally speaking, as shown in Chapter X., these seeds and -fruits are much more buoyant than they need to be, that is to say, if -they owe their floating power to adaptation to dispersal by currents. -This is quite in accordance with the argument developed in Chapter XI. -with regard to the general question of plant-distribution, that -dispersing agencies make use of characters and capacities of seeds and -fruits that were never intended for them. - - - _Summary of the Chapter._ - -(1) There are many mechanisms or contrivances in plants that now serve a -purpose for which they were not originally developed. - -(2) Of this nature, it is contended, is the relation between fruits and -seeds and the agencies of dispersal. - -(3) If, however, the structure or mechanism is made more effective by -the new function, such a modification may be regarded as an “adaptation” -in the language of the theory of Natural Selection. - -(4) It is held by Professor Schimper that the structures connected with -the buoyancy of the fruits or seeds of several tropical littoral plants -are, in the above sense, adaptations; and he points to several genera -where the buoyant tissues in the coverings of the fruits or seeds of the -coast species are scantily represented or absent in the inland species -of the same genus, a difference corresponding with the loss or -diminution of the floating powers. - -(5) This contrast in structure and in floating capacity between the -fruits or seeds of inland and coast species of the same genus is beyond -dispute, and the author adduces fresh data in support of it. - -(6) But he contends that it is not proved that the relatively great -development of buoyant tissues in the case of littoral plants is the -effect of adaptation; and that if the selecting process had been -confined to sorting out the xerophilous plants with buoyant seeds or -fruits and to placing them at the coast, the same contrast would have -been produced. - -(7) In support of this contention he points out that when such littoral -plants extend inland the floating capacity and the buoyant tissues are -as a rule retained; and that in those exceptional cases where inland -plants possess buoyant fruits or seeds these tissues are sometimes well -developed under conditions in which they could never aid the plant’s -dispersal. - -(8) But the most serious objection against the adaptation view is that -admittedly only about half of the shore-plants with buoyant fruits or -seeds come within its scope. Therefore a second explanation has to be -framed for the other plants concerned. - -(9) As showing the difficulties raised by regarding some of the -structures connected with buoyancy as “adaptive” and others as -“accidental,” it is pointed out that some fruits possess the two kinds -of structure. It is also shown that in several cases fruits endowed with -buoyant tissues are just as well adapted for dispersal by frugivorous -birds; and the instance of Ximenia americana is cited where a drupaceous -fruit, known to be dispersed by fruit-pigeons, possesses also in its -“stone” both the “adaptive” and “non-adaptive” types of “buoyant -structures.” - -(10) It is urged that whatever is the relation between the buoyancy of -the seeds and fruits of shore-plants and dispersal by currents, there -has been a uniform principle affecting all. - -(11) The weight of evidence is regarded as adverse to the Natural -Selection theory, an inference which is consistent with the conclusion -arrived at in Chapter X. that there is no direct relation between the -density of sea-water and the buoyancy of seeds and fruits, the floating -capacities being as a rule far greater than the adaptation view would -explain. Nature, it is held, has never made any provision for dispersal -by currents, the buoyancy of seeds and fruits being, as concerns the -currents, a purely accidental quality. - - - - - CHAPTER XIV - - THE RELATION BETWEEN LITTORAL AND INLAND PLANTS - -Professor Schimper’s views.—Great antiquity of the - mangrove-formation.—Problem mainly concerned with the - derivation of inland from littoral plants.—Grouping of - the genera possessing both coast and inland - species.—Scævola.—Morinda.—Calophyllum.—Colubrina.—Tacca.—Vigna.—Premna. - - -IN discussing the relation between the littoral and inland floras in the -Pacific it will be at first necessary to pick up some of the threads of -the various lines of investigation dealt with in the previous portion of -this work. Apart from considerations connected with the genetic history -of the plants concerned, when we come to inquire into the sources of any -individual strand-flora, whether in the temperate or in the tropical -regions, we arrive at the rough and ready inference that it is composed -of “what the sea sends and the land lends.” But it has been already -shown that the relative proportion of the current-borne and in -consequence widely dispersed plants in a strand-flora varies greatly in -different regions. Thus in the Pacific islands, as typified by those of -Fiji, about 90 per cent. have buoyant seeds or seedvessels originally -brought from distant localities; and in the tropics, as a rule, the -average would probably be never under 75 per cent. On the other hand, in -a temperate region the plants derived from inland would be most -predominant, making up probably some three-fourths of the whole, whilst -the proportion of current-dispersed plants hailing from distant places -would be relatively few. - -It is on this account that there is such uniformity in the general -composition of the strand-flora over a large part of the tropics, since -current-dispersed plants are widely spread. But in the temperate regions -we find a great contrast in this respect. There are, it is true, a few -current-borne plants that one meets everywhere. For instance, -Convolvulus soldanella is to be gathered on English beaches and on those -of New Zealand and of the coast of Chile. But these littoral plants with -buoyant fruits hardly give a feature to the strand-flora. A multitude of -intruders, either characteristic of the inland flora of the region or -confined only to the seaboard of that part of the world, also make their -home on the beach and frequently endow a beach-flora with its leading -features. The possible associations of plants on a beach in a temperate -region are thus very great; and I have already discussed this in part in -Chapter IV. as concerning the British shore-flora. One has only to look -at a work like that of Dr. Willkomm on the vegetation of the strand and -steppe-regions of the Iberian peninsula to realise how the few littoral -plants familiar to the English eye cut but a sorry figure amongst the -numbers of strange intruders from the arid regions inland. So again, as -I found on the Chilian beaches, Convolvulus soldanella finds odd -associates amongst the species of Nolana and Franseria that are peculiar -to the coasts of that part of the globe (see Chapter XXXII.); and -different grotesque American forms of the Cactaceæ with a -Mesembryanthemum and a host of strange-looking plants descend from the -arid slopes of the hills behind to keep company with the far-travelled -English beach-plant (see Note 49). Or again, a glance at the pages of -Professor Schimper’s great work on _Plant-Geography_ will bring the same -fact home in a still more varied fashion. - -Yet on tropical coasts the intruding inland element is also -distinguishable, though it may influence only to a small degree the -general character of the strand-flora. Dividing it, as we have described -in Chapter V., into the plants of the sandy beach and of the -mangrove-swamp, we find in the mangroves the most stable element and in -the beach-plants those most liable to change. Professor Schimper -observes that whilst the physiognomy of the beach-flora varies to some -extent with the alterations in the inland flora, the mangrove-formation -makes but a slow response to such changes. As he points out in his work -on the Indo-Malayan Strand-Flora (p. 199), seeds and seedvessels are -being continually brought down to the sea-coast through the agencies of -rivers, winds, and birds; and in this manner, in the course of ages, the -beach-flora is recruited from the inland plants. But for the mangroves -such additions to their numbers are rarely possible. Whilst the same -genera are often shared by both the beach and inland floras, we have in -the mangrove-formation families, sub-families, and genera almost -peculiar to itself, and including plants, like those of the Rhizophoreæ, -that in their characters betray but little kinship with others and give -but little indication of their descent. The mangroves have remained -through the ages as something apart from other coast-plants, isolated -both in their history and in their characters, and especially -distinguished by their “adaptations” to their surroundings. - -Such is the line of argument followed by this eminent German botanist in -his account of the development of a tropical strand-flora. In various -parts of this work I have ventured to suggest that the mangroves may be -the remnant of an ancient flora widely distributed over the lower levels -and coastal regions of the globe in an age when vivipary (meaning, -thereby, germination on the plant) was the rule rather than the -exception. At such a period, as I imagine, the climatic conditions of -the earth were much more uniform than they are at present, at least in -the lower levels; and a warm atmosphere, charged with aqueous vapour and -heavy with mist and cloud, enveloped a large portion of the globe. The -mangroves, it may be remarked, are by no means universally distributed -on tropical coasts in our own time. (Professor Schimper describes their -distribution in his _Indo-malayische Strand-Flora_, pp. 85, 86, and in -the English edition of his _Plant-Geography_, p. 409.) They are not -found on rainless coasts even when under the Line, except where there -happen to be large estuaries; but where a rank and luxuriant inland -flora betokens a high degree of humidity, there they thrive. This is -well illustrated on the rainless shores of tropical Peru, a locality -described in Chapter XXXII. of this work. - -Yet if, as it is here contended, the mangroves form a remnant of a once -widely spread viviparous flora, it might be expected that the -beach-plants of that age would have been also viviparous, and that with -their present descendants, as well as with some of the inland plants -allied to them, we ought to find in the anomalous structure of the seed -some indication of the lost viviparous habit. This appears to be the -case, as described in Note 50, with the Barringtoniæ, a tribe that has -supplied some of the most characteristic beach-trees, and also with some -genera of the Guttiferæ. Perhaps, indeed, when the seeds of several -other littoral beach-trees come to be examined, for instance, Guettarda, -analogous structures may be found. - -Although the beach-flora of the tropics is less stable in its -composition than the mangrove-formation, it is not to be assumed that in -the Pacific region or in the tropics generally it is at all modern in -its character. Though in the main, no doubt, more recent than the -mangroves, since it is likely that in early geological periods the swamp -rather than the sandy beach formed the predominant feature of the -sea-border throughout the tropics, yet it bears in several respects the -impress of a high antiquity. There are few beach plants in the tropical -Pacific that are not found over the tropics of a large portion of the -globe, a circumstance that would in itself warrant our assigning a great -age to the beach-flora; and it is highly probable that some at least of -the beach plants of the Pacific that occur on the east and west coasts -of tropical America are, for reasons given in Chapter XXXII., older than -the barrier now interposed by Central America between the Atlantic and -Pacific oceans. There are, it is true, a few species, like Acacia -laurifolia and Drymispermum Burnettianum, which, on account of their -restriction to the beaches of the Western Pacific and their lack of -capacity for dispersal by currents, may be regarded as local -productions; but for the great majority, ranging as they do over much of -the tropics, it is not possible to determine when and where they assumed -their littoral habits. That except in a few instances their home in some -bygone age lay outside the Pacific can scarcely be doubted. - -It is therefore to be expected that in a discussion of the relation -between the strand and inland floras in the Pacific islands the problem -will be mainly concerned with the possible derivation of inland from -littoral plants. In such a discussion the relation between the beach and -inland species of the same genus becomes a subject of great interest. It -is a subject that had a peculiar fascination for Professor Schimper, who -refers to it more than once in his pages; and though never able to take -it up, he viewed it as a very promising field of inquiry. The question -has been frequently alluded to in this work; and it is especially dealt -with in one connection in Chapter II. It is there shown that whilst, as -a general rule, the seeds or seedvessels of the coast species possess -great floating power, those of the inland species of the same genus have -little or none, and that both may have independent modes of dispersal, -the first by currents, and the last through frugivorous birds. - -A close connection between the beach and inland floras is apparently -displayed in the circumstance that quite a third of the genera of the -Pacific insular floras containing littoral species (some 70 in all, -excluding the mangroves) possess in this region also inland species. But -the further examination of this interesting group of genera, which are -enumerated in the list below, goes to show that the connection between -the inland and coast species of a genus is by no means always so close, -or of such a character, as one might have expected. It will not be -possible, however, to do much more than indicate in this chapter the -results of this inquiry; but the details will usually be found either in -the separate discussion of the genus or in other parts of this work. For -convenience of treatment these genera may be grouped in the following -sections. - - - _Grouping of the Plant-Genera of the Islands of the Tropical Pacific - that possess both Littoral and Inland Species._ - -Section I. Where the littoral and inland species are most probably of -independent origin, both possessing their own means of dispersal; -Calophyllum, Hibiscus, Colubrina, Morinda, Scævola, Cordia, Ipomœa, -Vitex, Tacca, Casuarina. - -Section II. Where the littoral species have probably given rise to -inland species, and both still exist in the group of islands: Vigna, -Premna. - -Section III. Where inland species have been probably developed from -littoral species no longer existing in the group: Canavalia, Erythrina, -Sophora, Ochrosia. - -Section IV. Where the littoral and inland species are evidently of -independent origin, and there is no means of accounting for the -existence of the inland species by agencies of dispersal at present in -operation: Barringtonia, Pandanus. - -Section V. Where in the same genus some inland species are derivatives -of the coast species and others are of independent origin: Guettarda. - -Section VI. Where the coast species, having little or no capacity for -dispersal by currents, are regarded as derived from the inland species -in one group of islands and as afterwards distributed to those in the -vicinity: Eugenia, Drymispermum, Acacia. - - - SECTION I - -This group, which includes those genera where the coast and inland -species are regarded as of independent origin, both possessing their own -means of dispersal, contains about half of the total number of genera -here concerned. We will first deal with the genera Calophyllum, Morinda, -and Scævola, where the littoral species have buoyant fruits or seeds -that are dispersed by currents, whilst the inland species have more or -less non-buoyant fleshy fruits that could only be dispersed by -frugivorous birds. Here the inland and coast species could have arrived -independently at the island, and we are not called upon either on this -ground or by reason of affinity of characters to connect the one with -the other. - -The genus Scævola is very typical of its kind and has been already in -part discussed in Chapter II. The wide-ranging shore-species, S. -Kœnigii, that is distributed over the Pacific may sometimes, as in -Hawaii, be accompanied by numerous inland species, all endemic, seven of -them being enumerated by Hillebrand; or, as in Fiji and Tonga, there may -be associated with it a solitary inland species, S. floribunda (see Note -51); or, as in Tahiti, it may exist by itself. On the other hand, as in -the Kermadec Islands, a single inland peculiar species may alone -represent the genus. The inland species have fleshy drupes which, as far -as examined, have no floating power and possess no buoyant tissues in -their coverings; and their independent dispersal by birds cannot be -doubted. The endemic character of most of the inland species of the -Pacific islands is most probably due to the suspension of the -transporting agency of frugivorous birds, just as the wide range of the -solitary littoral species may be attributed to the uninterrupted agency -of the currents. There is nothing in the description of the endemic -species given in Hillebrand’s _Hawaiian Flora_ to indicate any especial -genetic connection between the inland species and the beach plant, S. -Kœnigii; and the occurrence of a solitary inland peculiar species in the -Kermadec Islands clearly proves an origin independent of any littoral -plant. - -Morinda is another critical genus in this discussion. Besides the -widespread littoral species (M. citrifolia) that is distributed by the -currents and is also dispersed by man, there are in the Pacific islands -a number of inland species, mostly climbers and denizens of the forests. -In the _Index Kewensis_ six are accredited to Fiji and five to New -Caledonia. Hillebrand gives a peculiar Hawaiian species, and there is a -widespread species (M. Forsteri) that ranges over the South Pacific from -New Caledonia to the Marquesas and the Paumotu Islands. Since, as -indicated in Chapter II. and in Note 8, the pyrenes of the fruits of the -inland species are not dispersed by the currents and could readily be -transported by frugivorous birds, we are not called upon to connect them -in their origin with M. citrifolia, the wide-ranging species of tropical -beaches. - -The fact of the dispersal of certain inland species of the genus over -large areas of the tropics, such as in the case of Morinda umbellata -through tropical Asia and Malaya, and M. Forsteri in the Pacific, is -indeed sufficient proof that these inland plants are independent of any -littoral species in the Pacific and possess their own means of -distribution. Though the genus, comprising at least forty species, is -mainly confined to the Old World, there are a few species in America; -but M. citrifolia, the familiar beach species of the Old World and the -Pacific, is not indigenous there, and, as far as I can gather, all the -American species belong inland. Facts of distribution of this nature -negative the possibility that the Pacific islands have received their -inland species of Morinda through the intervention of the far-ranging -littoral plant. - -As respecting Calophyllum, which is represented all over the tropical -South Pacific by the wide-ranging C. inophyllum and by a tree of the -inland forests found also in Malaya and in Ceylon (C. spectabile), there -are, apart from questions of affinity, grave objections against the -derivation of the same inland species from the coast species all over -this area. The fruits of the two inland species of Fiji, C. spectabile -and C. burmanni, have sappy outer coverings and are quite suited for -dispersal by fruit-pigeons. As observed in Chapter II. and Note 9, they -have limited floating capacities and their dispersal by birds is -necessary to explain their distribution. Since the timber is greatly -valued by the Polynesians, it is not unlikely, however, that those -islanders have assisted in the distribution of the inland species. It is -not possible to do more than touch on this subject here; but it may be -inferred that the history of Calophyllum in the Pacific has not been one -that would warrant our regarding the inland trees as derivatives of a -coast species. - -There are other genera of this section where, for reasons of a different -character, there is no cause for assuming that the inland species are -derived from the coast species, or _vice versâ_. Thus, in Fiji, -Casuarina equisetifolia, a widely distributed species of the Old World, -occurs at the coast and in the scantily wooded plains behind; while C. -nodiflora, a New Caledonian species, finds its home in the lower -forests. There are many endemic species in Australia and New Caledonia; -and we are not called on to connect together these two species in Fiji. -In the same way we are not under any obligation in the case of the -numerous inland species of Ipomœa of the Pacific islands to connect them -with the coast species. They are all widely ranging species, and their -seeds have been carried to the islands, each in its own fashion. So -again with the inland species of Hibiscus found in the Polynesian -islands and often cultivated, we cannot either from the point of view of -dispersal or of affinity connect them with the far-ranging littoral -species, H. tiliaceus, which belongs to a section of the genus distinct -from those sections to which the inland species belong. - -In a similar way there is no ground for supposing that Cordia aspera, an -inland species confined to Fiji, Tonga, and Samoa, is derived from C. -subcordata, the widely distributed littoral species of the Pacific and -of the Old World, since they belong to different sections of the genus. -But, apart from any question of affinity, the drupes of inland species -of Cordia are known to be well suited for dispersal by frugivorous -birds, though, unlike the littoral species above named, not adapted for -transportal by the currents. The genus Vitex, which is represented by a -wide-ranging littoral species in the Pacific (V. trifolia), appears to -be associated with inland species only in Fiji, where one or two, -seemingly endemic, occur. But there is nothing in Dr. Seemann’s -description of V. vitiensis, one of these species, that at all suggests -its derivation from the strand species, a very variable plant that often -extends far inland into the plains, adopting a different habit of growth -in those localities. It is known that Vitex fruits can be dispersed both -by birds and by currents. This genus is more fully discussed in a later -chapter. - -Of the genus Colubrina there seem to be only two Pacific species -known—one the widely distributed shore-plant, C. asiatica, a straggling -shrub with alternate leaves found in all the Pacific groups and on the -beaches of much of the tropics of the Old World; the other a tree, C. -oppositifolia, with opposite leaves, that is peculiar to the Hawaiian -islands, where it frequents the open-wooded and scrubby inland -districts. The seeds of the shore-plant float unharmed for many months, -whilst the fruits of the inland plant, which differ in some important -respects (see Note 52), would float only for a week or two. The strand -species is also quite at home inland in many parts of the world; and -there is nothing from the standpoint of affinity to indicate that in -Hawaii it has given birth to an inland species so divergent in habit and -in character. There is of course the difficulty of explaining how a -plant like C. oppositifolia, with such a dry, unattractive fruit, could -be indebted to birds for its original introduction into the group; but -the same difficulty arises with a host of Hawaiian plants. It is, -however, evident from its distribution over the islands of this -archipelago that it possesses or has possessed some means of -inter-island dispersal, and since it is not of much service to the -aborigines we must look therefore to the bird. - -In the instance of the genus Tacca there is in Fiji an inland species, -T. maculata, associated with a wide-ranging beach species, T. -pinnatifida, which also grows inland. The first-named is recorded from -the north coast of Australia and from Samoa, and though, unlike the -beach plant, its seeds are unfitted for dispersal by currents (see -Chapter II.), they might be distributed by birds. Dr. Reinecke describes -another inland species from Samoa, T. samoensis. The beach plant, T. -pinnatifida, grows so typically (sometimes side by side with T. -maculata) in the inland plains of Fiji that one would not be justified, -apart from questions of affinity, in regarding it as the parent form of -inland species in the Pacific islands. - -For food and other purposes Tacca pinnatifida is or was much valued by -the Pacific islanders, and it grows so abundantly that cultivation is -rarely practised. That the Polynesians have aided the currents in the -distribution of the plant there can be no doubt, and this is -particularly indicated by its occurrence in Hawaii. The genus contains -ten or a dozen species, of which at least three are peculiar to America; -but T. pinnatifida, the characteristic shore-plant of the Old World, and -according to Schimper the only one that can be so designated, is not -found in America, where, as far as I can gather, there is no -widely-spread beach species dispersed by the currents from which the -peculiar species could have been derived. In the case of the Pacific -species, however, it should be noted that I am not endeavouring to prove -the improbability of the inland species having been derived from the -coast species in other regions, as in Australia, but that my point is to -show there is no reason to suppose that this has taken place in the -Pacific. There is no difficulty in attributing the dispersal of inland -species to birds; and we are therefore not called on to connect them -with the beach plants. - - - SECTION II - -This division includes those genera where the littoral species has -apparently given rise to one or more inland species and both still exist -in the same group of islands. Two genera alone, Vigna and Premna, come -into this category. The first-named seems to present a good case for the -derivation of an inland from a coast species in Hawaii. Besides Vigna -lutea, the beach species, which is found not only all over the Pacific -islands but on the tropical beaches of the Old World, there are in -Hawaii two endemic species (V. sandwicensis and V. oahuensis) that occur -in the mountains, usually at elevations of from 1,500 to 5,000 feet; but -I do not find any more inland species recorded from the other Polynesian -archipelagoes. It may at first be noted that Vigna lutea, which in some -parts of the world strays inland, displays considerable variety in its -littoral station in the Pacific. Thus, in Hawaii, I found it sometimes -on the sandy beach, sometimes on a rocky shore, and sometimes on the -edge of old lava-cliffs overlooking the sea. In Fiji, though usually a -trailer on the beach, it may become a climber hanging from the trees -bordering the creeks in the mangrove-swamps. Though Hillebrand makes no -mention of forms intermediate between coast and inland species in -Hawaii, I found in one locality at the coast some specimens of Vigna -lutea displaying the twisted pods and two callosities on the standard -that are characteristic of V. sandwicensis, one of the inland species. -The seeds of Vigna lutea float in sea-water unharmed for months, and -they are to be found in the stranded drift of the Hawaiian and Fijian -beaches, and floating in the drift of the Fijian rivers. I was unable to -obtain the mature seeds of the inland species, and it has therefore yet -to be determined whether they follow the rule in the loss of buoyancy. -It may be added that a plant of Vigna lutea raised in Hawaii from seed -displayed some small tubers of the size of a pea on its roots. - -The case for Premna is stated in Note 32. In this genus, as with Vigna, -the final test of experiment is needed; but the data at my disposal -point to the probability that an inland species has here been derived -from a littoral plant. - -The summary of this chapter is given at the end of Chapter XVI. - - - - - CHAPTER XV - - THE RELATION BETWEEN LITTORAL AND INLAND PLANTS - (_continued_) - -Inland species of a genus developed from littoral species originally - brought by the currents but no longer existing in the - group.—Illustrated by the Leguminous genera, Erythrina, Canavalia, - Mezoneuron, and Sophora, and by the Apocynaceous genus, Ochrosia.—The - Hawaiian difficulty. - - - SECTION III - -HERE we have three genera of the Leguminosæ, namely, Erythrina, -Canavalia, and Sophora, and one Apocynaceous genus, Ochrosia, in which -it is considered that inland species have been probably developed from -littoral species no longer found in the group. In this case the shore -species, possessing buoyant seeds or fruits that are known to be -dispersed by the currents, is absent from the particular group in which -the inland species occurs; and since the last-named displays no capacity -for distribution by currents, or seemingly by birds, we are driven to -infer that it was originally derived from a coast species, brought by -the currents, that has since disappeared. - -Hawaii is the only region concerned here; and these four genera may be -said to well illustrate the particular “Hawaiian difficulty.” If this -explanation of the origin of the inland species is legitimate, then it -offers us a mode of explaining still more perplexing cases in the -Hawaiian flora, such as those relating to the endemic species of -Mezoneuron (Leguminosæ) and to Hillebrand’s Vallesia (Apocynaceæ), where -there is apparently no littoral species known from any region. - -Dealing with the three Leguminous genera, it is at first to be remarked -that the great floating powers of the seeds of the littoral species are -in all three cases to be attributed to the buoyant kernel; whilst on -account of the non-buoyancy of the kernel the seeds of all the inland -species possess no floating power. Some very interesting points are -raised in each of the three genera, and I will first deal with the genus -Erythrina. - - - ERYTHRINA. - -If we look over the Pacific islands in search of a critical locality for -the investigation of the genetic relation between the littoral and coast -species of Erythrina, we discover it, as far as I can gather, only in -one group. In Fiji, Tonga, and Samoa we find only the littoral species; -in Hawaii there is only an inland species; whilst in Tahiti occur both -the littoral and the inland species—E. indica, the wide-ranging -shore-tree of the South Pacific, and E. monosperma, the inland tree of -Hawaii—the last found nowhere else in Polynesia, and confined to the -Pacific. In Tahiti there are no other species, and it is between these -two species that the connection, if it exists, is to be sought. (Further -details relating to the genus are given in Note 53. In this place only -the facts bearing on the argument will be discussed.) - -The buoyant seeds of Erythrina indica are well known to be dispersed by -the currents; whilst those of E. monosperma, as obtained from Hawaii, -have no floating power and sink at once, or in a day or so, even after -drying for two years. In Tahiti the first-named species is a -characteristic plant of the beach, whilst the last grows there in the -valleys and on the mountains up to elevations of 700 to 800 metres. We -have now to inquire whether there is any decided affinity between the -two species, and whether the divergent characters of the inland species -can be connected with its station. With regard to the first query we may -quote in reply the observation of Drake del Castillo, that as concerning -the foliage and the inflorescence E. monosperma is very nearly related -to E. indica, differing only from it in the more hairy calyx, in the -more permanently tomentose and much shorter pod, and in the paucity of -seeds (one or two in number). - -We will now see whether it is possible to connect these differences in -character with differences of station. Neither Nadeaud nor Drake del -Castillo give precise descriptions of the station of Erythrina -monosperma in Tahiti; but Nadeaud and Lepine remark that it grows on -precipices as well as in the valleys on the north or dry side of the -island; and we may infer that it affects exposed dry rocky stations. In -Hawaii, according to Hillebrand, it is found on the dry rocky hills and -plains of all the islands up to 1,000 feet. I was particularly -interested in this tree whilst in the group, and found it in the large -islands of Maui and Hawaii thriving in rocky arid districts of little -rainfall, accompanied by Cactus opuntia, Ricinus communis, and -Cæsalpinia bonducella. It is often to be observed on scantily vegetated -lava-flows, a solitary tree growing here and there out of a crack in the -old lava, or it may dot the rocky slopes of some barren declivity. I -found it in the dry gulches behind Lahaina at elevations of 800 to 1,200 -feet above the sea, growing amongst huge blocks of stone in clumps of -ten or twelve trees. When one contrasts the inland station of E. -monosperma with that of E. indica on the beach where the atmosphere is -more humid and the conditions more suited for plant-growth, it appears -probable that the differences between these two species may be largely -connected with station, especially as regards hairiness and the -diminished size of the pods. - -Assuming, therefore, that Erythrina monosperma is but the inland form of -E. indica and that the differences between the two species are mainly an -affair of station, we have next to account for the occurrence of the -inland species in Hawaii without the littoral species. The agency of -currents in explanation of the existence of E. monosperma in Hawaii is -at once excluded, since the pods dehisce on the tree, and the seeds, as -already remarked, have no floating power. Nor does it seem likely that -beans half an inch (13 mm.) long could be transported unharmed in a -bird’s stomach over the two thousand miles of sea that intervene between -Tahiti and Hawaii. Yet one cannot doubt that the pyrenes and “stones” of -genera like Coprosma, Nertera, Cyathodes, and Osteomeles have been -carried by frugivorous birds to Hawaii. But a bean is somewhat different -from the crustaceous pyrene of Coprosma or the hard “stone” of -Cyathodes; and although, as indicated by the occurrence of an endemic -species of Erythrina in Fernando Noronha, birds may carry large beans -unharmed over a couple of hundred miles of sea, one hesitates to -conclude that they could effect this when the tract of ocean to be -traversed is ten times as great. There are again reasons for believing -that the seeds of Erythrina monosperma are particularly ill-suited for -dispersal by birds, since, notwithstanding their hardness, they soon -absorb water through the micropylar opening; and they germinated so -readily in my experiments that the digestive juices in a bird’s stomach -would probably soon find access and destroy the kernel. It is, however, -known from the observations of the Messrs. Layard in New Caledonia that -a small crow and different species of parrots feed on the seeds of -Erythrina, and they may aid in the local dispersal (_Ibis_, vol. 6, -1882). - -To admit man’s agency in carrying to Hawaii the seeds of a tree which is -only useful in supplying him with light wood for his outriggers and his -fishing-net floats would compel us to place in the same category a great -number of plants in some way useful to him which are recognised as -indigenous. The Polynesian ransacks the vegetable world for his wants, -and carries with him in his migrations only his food-plants and the -seeds of his sacred trees. - -There remains then the possibility that the parent species, Erythrina -indica, was once in Hawaii but has since disappeared. In order to -establish this, it will be requisite to show not only that the -extinction of a shore-plant is probable, but also to explain why the new -species has selected such arid inland localities for its stations, to -account for the loss of buoyancy of the seeds, and, if possible, to give -an instance of the production of a new species of Erythrina in a small -isolated oceanic island. - -A study of the special circumstances of Hawaii leads one to conclude -that a shore-tree may become extinct in one of two ways. It may be -exterminated by insect pests, or it may be forced inland through -unsuitable coast-conditions and there be lost in the resulting new -species. One characteristic shore-tree, Cordia subcordata, has indeed -been almost exterminated by insects, and even Erythrina monosperma is -now from the same cause on its road to extinction (see Note 53); but -there is no indication of their leaving modified descendants behind that -are pest-proof. The most probable view then is that the littoral tree, -having been driven inland through the unsuitability of the -coast-conditions, such as lack of beaches or want of moisture, has there -become modified. This is what has really happened, as I have shown, with -Cæsalpinia bonducella in Hawaii. As indicated in Chapter XVII., this -characteristic beach-plant has here been driven off the beach. There -would thus be no difficulty in assigning a reason why a littoral tree -like Erythrina indica should select arid localities when it extends -inland, since, as is pointed out in Chapter IV. and in other parts of -this work, the plants of the beach and of the arid inland district -possess the same xerophilous habit. - -With regard to the loss of buoyancy of the seeds in the case of -Erythrina monosperma, it may be remarked that this is precisely what has -happened with the seeds of Cæsalpinia bonducella, its usual associate on -the old lava-wastes in Hawaii, and with an inland species of Cæsalpinia -in Fiji. It is argued that the same thing has occurred with the inland -Hawaiian species of Canavalia and Sophora, as shown in later pages of -this chapter. It has certainly happened with the inland form of Afzelia -bijuga in Fiji, a tree dealt with in Chapter XVII. These are all -Leguminous genera; and in all of them, with the exception of Cæsalpinia, -where the floating power arises from a central cavity in the seed, the -seeds of the littoral species possess, like Erythrina indica, buoyant -kernels. Whilst most littoral plants with buoyant seeds or fruits retain -the floating capacity of the seed or fruit when they extend inland, the -Leguminosæ often offer exceptions to the rule. - -That inland endemic species of Erythrina can be developed in isolated -islands is illustrated by the existence in Fernando Noronha, some two -hundred miles from the coast of Brazil, of a peculiar species, E. -aurantiaca, described by Mr. Ridley. Here also is found an inland -species of Guettarda peculiar to the locality; but in neither genus does -the littoral species occur. - -Many difficulties will yet have to be explained before it can be finally -established that Erythrina monosperma has been derived from E. indica or -some similar shore species that was originally dispersed by the -currents; but we are almost driven towards such a view, since it is hard -to believe that the beans were carried to Hawaii by birds over some two -thousand miles of sea. Observers in other regions where littoral and -inland species of the genus occur may perhaps devote their attention to -the relation between the two; and if they are able to supplement -observation and experiment by a microscopical investigation, some -interesting results would be obtained. For instance, I would suggest -that in Queensland a thorough examination of the littoral E. indica and -the inland E. vespertilio might be undertaken; or perhaps there may be -some other littoral form. - -With the two other Leguminous genera, Canavalia and Sophora, to be -immediately discussed, we have for the most part the same questions -raised. Both possess wide-ranging current-dispersed littoral species in -other parts of the Pacific, but only endemic inland species with -non-buoyant seeds in Hawaii. The pivot of the discussion will be here -also the impracticability of these inland species ever having reached -the Hawaiian Islands through the agency of the currents, and the great -difficulty in believing that their beans were carried unharmed by birds -over half the breadth of the Pacific Ocean. If we reject alike the -current, the bird, and the parentage of a lost littoral species, we must -fall back on the continental hypothesis, against which in the case of -Hawaii the evidence is overwhelming. - - - CANAVALIA. - -This genus is represented in the tropical islands of the South Pacific -from Fiji to Tahiti by three littoral species, none of which have been -found in Hawaii, where only an endemic inland species exists. Reference -will alone be made here to such facts as bear on the probable history of -the mysterious Hawaiian species, additional particulars being given in -Note 54. The littoral species, Canavalia obtusifolia (D.C.), C. sericea -(Gray), and C. ensiformis (D.C.), have buoyant seeds and are dispersed -by the currents; whilst the inland Hawaiian species, C. galeata (Gaud.), -a forest climber peculiar to that group, has non-buoyant seeds. We thus -have repeated the problem of Erythrina monosperma. The absence of the -littoral species from Hawaii can scarcely be attributed to the failure -of the currents, since Ipomœa pes capræ, which accompanies C. -obtusifolia as a beach-creeper all round the tropical globe, is present -on the Hawaiian beaches. Nor can it arise from lack of floating-power on -the part of the seeds, since experiment indicates that the seeds of C. -obtusifolia will float for months unharmed in sea-water. Nor can it be -ascribed to climatic conditions, since this tropical shore species -extends into cooler latitudes than those of the Hawaiian Islands, being -found in the Kermadec Group and in the Bermudas, which are subtropical -both in position and as regards much of their vegetation. The reason -perhaps we may never learn from the plants themselves, though it may be -possible to obtain some light on the problem from outside sources. - -Canavalia galeata differs much in its habits, as well as in some of its -characters, from the existing littoral species of regions outside the -Hawaiian Group. It is a stout climber ascending the forest trees to a -considerable height, though, as is indicated in Note 54, the shore -species sometimes display a tendency in the same direction. It is -described by Hillebrand as occurring “on all islands, in forests up to -2,000 feet.” Like those of the inland species of Erythrina (E. -monosperma), its seeds sink in sea-water even after being kept for four -years, nor could the pods be utilised for dispersal by the currents, -since they float, when unopened, only for four or five days. Here also, -as with Erythrina, the seeds of the inland species no longer possess the -buoyant kernels to which the floating capacity of the seeds of the coast -species is due. Though we have to exclude the currents, we can scarcely -in its case appeal to bird-agency when we wish to account for the -transportal of the original seeds to Hawaii, as that would imply that -birds can carry beans nearly an inch, or 2 to 2.5 centimetres, in length -unharmed in their stomachs over a tract of ocean some 1,500 or 2,000 -miles across. We should have to learn much that is unexpected of the -modes of dispersal of the Leguminosæ before we could accept such an -hypothesis. - -Canavalia galeata indeed presents to the student of dispersal one of the -enigmas of the Hawaiian flora; and it should be noted that the mystery -of its distribution is concerned not only with the means of transportal -of the seeds of the original species to the group, but also with its -present dispersal among the islands. It is, however, suggestive that Dr. -Hillebrand mentions two varieties, one of them found on Kauai, with -somewhat smaller seeds; so that some inter-island differentiation is -evidently in progress. No attempt is made here to connect this inland -species directly with the absent beach-plants. That is a matter for the -systematist; but we are not tied down to existing shore-plants in -finding an ancestor, since the common parent of the littoral and inland -species may have been a shore-plant dispersed by the currents. - - - MEZONEURON. - -Another closely parallel instance, offering, from the standpoint of -dispersal, the same difficulties presented by Canavalia galeata, is to -be found in Mezoneuron kauaiensis (Hillebr.), a tall inland shrub also -peculiar to the group and belonging alike to the Leguminosæ. The -difficulties are so nearly identical that the same explanation will have -to cover both; but it is significant that with Mezoneuron there is no -littoral species to which we can appeal to extricate us from the -difficulty. Yet the genus is related to Cæsalpinia, and the species was -first described by Mann as C. kauaiensis, so that it may have once -possessed a littoral species that has ceased to exist as such. When we -come to discuss Cæsalpinia and Afzelia (Chapter XVII.) we shall obtain -from those genera many suggestions as to the probable past of Canavalia -galeata and Mezoneuron kauaiensis, two of the greatest riddles presented -by the Leguminosæ of Hawaii. - -The flat seeds of this species of Mezoneuron measure about an inch (2·5 -cm.), and seem most unsuitable for dispersal by birds over a wide extent -of ocean. Nor can we appeal to the currents, since my experiments in -Hawaii show that the seeds have no buoyancy and that the pods only float -for a week in sea-water. Dr. Hillebrand records this shrub from Kauai, -Oahu, and Maui; I found it also on the lower slopes of Hualalai in -Hawaii and therefore the same question of inter-island dispersal here -presents itself that was connected with Canavalia galeata, since we have -also to explain the transport of the seeds between islands 70 to 150 -miles apart. The critical point in the history of these two enigmatic -inland plants of the Hawaiian Islands was doubtless concerned with the -loss of buoyancy of the seeds of the original littoral plant. It will -subsequently be shown that this is what is now in actual operation with -Cæsalpinia and Afzelia in different parts of the Pacific. - - - SOPHORA. - -In this genus, as in Erythrina and Canavalia, we have a littoral -species, Sophora tomentosa, that ranges over the tropical beaches of the -globe, including most of the islands of the Pacific, but does not occur -in Hawaii, where the genus is represented by an endemic inland species, -S. chrysophylla. Here also we find the shore-species with seeds capable -of floating for months on account of their buoyant kernels, and the -inland species with seeds that sink even after years of drying (see Note -56). Unless other inland species of Sophora have recently been described -from the tropical Pacific, the Hawaiian species is the only one of its -kind known from this region. - -But the problem wears a different aspect in the case of this genus, -since the endemic inland species of Hawaii is a tree of the mountains -where a temperate climate prevails, whilst Sophora tomentosa is a shrub -of the tropical beach that only at times extends into subtropical -latitudes. The Mamani tree, as the Hawaiians name S. chrysophylla, -extends up to 9,000 or 10,000 feet above the sea, forming, with Myoporum -sandwicense and one or two other trees and shrubs, the highest belt of -the forest in the larger islands. It is in the open woodland between -6,000 and 7,000 feet that it is most at home, and here it attains a -height of 20 to 30 feet. It descends in places to as low as 2,000 feet -above sea-level; but here is living under uncongenial conditions, and, -like Myoporum sandwicense, becomes dwarfed and shrubby. The climatic -conditions under which S. chrysophylla thrives in the Hawaiian mountains -are therefore those of the temperate zone. From the data given in -Chapter XIX., the mean annual temperature at an elevation of 6,000 to -7,000 feet would probably be about 55°, the average temperature of New -Zealand. - -We must therefore look to the temperate and not to the tropical zone for -the home of the parent species of Sophora chrysophylla; and if it was -originally derived from a shore-plant dispersed by the currents, the -widespread S. tomentosa could scarcely have been the species concerned. -But this strand-plant is disqualified for another potent reason, since -it belongs to a different section of the genus. Whilst S. tomentosa -belongs to the section possessing smooth pods, S. chrysophylla is -referred to the section Edwardsia having four-winged pods, which -comprises about ten species found in Chile and Peru, Hawaii, New -Zealand, Further India, and the Isle of Bourbon. What strange principle -in distribution, we may fitly ask, has linked together in this odd -fashion the continents of the Old and New World and the islands of the -Indian and Pacific oceans? - -Yet, discredited as Sophora tomentosa is as a possible parent of the -Hawaiian mountain species, it may yet afford us a clue. It is -significant that the distribution of this wide-ranging beach-shrub in -the tropics of the southern hemisphere is almost coterminous with that -of Sophora tetraptera, a species widely spread in the south temperate -zone from Chile to New Zealand and extending towards the tropics as far -as Juan Fernandez in lat. 33° S. and to Easter Island in lat. 27° S. -Though not strictly a beach-plant, S. tetraptera is a plant of the -sea-border; and it is remarkable, but not surprising, how in New -Zealand, one of its principal homes, its behaviour in respect of its -vertical distribution presents a great contrast to that of S. -chrysophylla in the tropical latitudes of Hawaii. We have seen that, in -Hawaii, S. chrysophylla, which thrives as a tree 20 to 30 feet high in -the mountains, becomes shrubby when it descends to the lower levels. In -New Zealand, S. tetraptera is, as we learn from Kirk, a prostrate shrub -in the mountains, whilst in the lower elevations towards the sea it -becomes a tree 30 and even 50 feet in height. It can scarcely be doubted -that, if we exchanged the habitats of these Hawaiian and New Zealand -species, each would to a great extent take up the other’s station and -the other’s habit. - -The whole problem of the dispersal of Sophora was brought immediately to -my notice at Corral, in latitude 40° S. on the coast of Chile. Here a -small tree of the section Edwardsia was growing in fruit on the lower -slopes of the hills, becoming bushy when descending to the beach. -Specimens of its four-winged pods have been identified at the Kew Museum -as those of Sophora tetraptera; and, as far as the pod is concerned, I -cannot distinguish between my specimens of the Hawaiian S. chrysophylla -and the Chilian species. Subsequently I found the buoyant seeds of the -same plant amongst the stranded beach-drift at Bahia San Vincente, -nearly 200 miles further north. This led to my experimenting on the -capacity of the plant for dispersal by the currents, and as a result it -was ascertained (see Note 56) that whilst, as in the case of S. -chrysophylla, the pods floated only one or two weeks, the seeds on -account of their buoyant kernels floated for several months in -sea-water, retaining their power of germination. The Chilian plant thus -differs significantly in its capacity for dispersal by currents from the -Hawaiian species, the seeds of which sink in sea-water even after years -of drying. - -The Mamani tree in Hawaii had always been an object of great interest to -me. I was attracted by the mystery surrounding its origin and had long -suspected that the clue was to be found in the non-buoyancy of its seeds -and in the absence of a littoral species of the genus. When in Fiji it -was to the littoral Sophora tomentosa that I looked in vain for a -solution of the riddle, and seven years afterwards on the coast of Chile -a solution of this enigma of the Hawaiian mountains presented itself in -the form of an argument somewhat in this shape. - -On account of the elevated station of the Mamani tree (S. chrysophylla) -in Hawaii it is to be inferred that the original species was a plant of -the temperate regions or of the uplands of some tropical mountains. If -it has had its origin in some shore-plant dispersed by the currents, -that species can only now be found on the coasts of extra-tropical -regions. Such a maritime plant had buoyant seeds; and plants of this -type are presented by Sophora tetraptera and its allied species that are -at home in the cool latitudes of the southern hemisphere, as in Chile -and New Zealand. No difficulty, as I argued, could be connected with the -loss of buoyancy of the seeds of the Hawaiian mountain species, since it -follows the general principle (laid down in Chapter II.) that in the -same genus coast species have buoyant seeds or fruits, and inland -species those that sink; and in support of this view it was recalled -that this is what happens to the seeds of Cæsalpinia bonducella and -Afzelia bijuga when the plants extend inland in the Pacific islands. It -was held, in short, that the original form of Sophora chrysophylla in -Hawaii was a coast plant with buoyant seeds, and therefore indebted for -its presence to the currents. Hailing from an extra-tropical region, it -abandoned the beach and found suitable conditions of existence in the -mountains, where it underwent specific differentiation. Such was the -explanation that presented itself to me on a Chilian beach. - -The first objection that offers itself against this view is that Sophora -chrysophylla is one of several species characterising the antarctic -element of the mountain flora of Hawaii, and that many of these plants, -such as those of the genera Astelia, Coprosma, Gunnera, Myoporum, &c., -could only have reached these islands through the agency of frugivorous -birds (see Chapter XXIII.). There is, therefore, something to be said -for this mode of dispersal; but though one can understand how hard seeds -and the “stones” and crustaceous pyrenes of fleshy fruits might be -transported unharmed in a bird’s stomach half-way across the Pacific -Ocean to the distant group of Hawaii, it is difficult to understand how -Leguminous seeds, except in such cases as Tephrosia piscatoria, could be -ejected unharmed by a bird after an ocean passage of some 1,500 or 2,000 -miles. Yet evidence pointing to such a possibility is not lacking. It -was pointed out by W. O. Focke (_Nat. schaft. Ver. zur Bremen_, -Abhandl., Band 5, 1876) that for many Leguminosæ we are driven to the -agency of birds in order to explain their dispersal. In this connection -he mentions the case of a pigeon killed by some beast of prey that he -found in his garden in the early winter. In the following spring he -noticed numerous seedlings of Vicia faba sprouting up from amongst the -feathers that alone remained of the bird. In this observation he -detected the normal method of the dispersal of the Leguminosæ by birds, -the seeds not being ejected by the bird but being set free by its death. -It is well known that Darwin had this idea in his mind when he conducted -his experiments on the dispersal of seeds; and reference may here be -made to one that is recorded in _More Letters of Charles Darwin_ (i., -436). Out of a number of seeds left in the stomach of an eagle for -eighteen hours, the majority were killed; but amongst the few that -germinated afterwards was a seed of clover (Trifolium). If such a bird -had carried a Sophora seed to Hawaii, this would have involved a -continuous flight of, on the average, 100 miles per hour for a period of -fifteen to twenty hours. This would just come within the limitations -laid down by Gätke as regards length and velocity of flight—a subject -discussed in Chapter XXXIII. - -We will now turn to the Sophora seeds themselves for evidence of their -capacity of surviving the perils of such a journey. The seeds of Sophora -chrysophylla, which are about a quarter of an inch (6 to 7 mm.) in -length, possess unusually hard coverings for the order, and in that -respect appear fitted for dispersal by animals. Indeed, in the large -island of Hawaii wild pigs and sheep feed on the pods, and no doubt aid -in the distribution of the plant over the island through the germination -of ejected undigested seeds. But since the species is found on most of -the larger islands, it is apparent that to birds we must look for the -explanation of its inter-island dispersal. Mr. Wilson, in his _Aves -Hawaienses_, remarks that one of the Hawaiian finches (Loxioides) feeds -on the seeds of this tree, which probably, he adds, also serve as the -food of Chloridops kona, another big finch; and it is to be inferred -from the observations of Mr. Perkins, quoted by Mr. Evans in his book on -Birds, that the Drepanididæ, a family peculiar to Hawaii, are in the -habit of splitting the pods of trees like Acacia koa and Sophora -chrysophylla to obtain the seeds. It would, however, seem that the -agency of birds confined to these islands does not carry us very far -when we wish to explain the original transport of the seeds over a -breadth of ocean of some 1,500 miles and more. Yet we know that this -must have happened with some of the Hawaiian plants, such as Osteomeles -anthyllidifolia and Nertera depressa, that are not confined to these -islands and possess fruits that would attract frugivorous birds. But -whether it has occurred with the dry beans of the Hawaiian species of -Sophora is another matter. - -On the whole I am inclined to the view, bearing in mind the general -indications of the Leguminosæ in the Pacific, that S. chrysophylla -originally reached Hawaii as a littoral plant through the agency of the -currents. Many points still need investigation; but it may be pointed -out that South America probably received Sophora tetraptera from New -Zealand by the West Wind Drift Current. - - - OCHROSIA (Apocyneæ). - -This genus seems to offer the strongest testimony in support of the -derivation of an inland species from a strand-plant. The drupes are so -large, the minimum size of the “stone” being 1-1/2 or 2 inches (37 to 50 -mm.), and so dry and unattractive for birds, that any other agency but -that of the currents appears to be out of the question. Indeed their dry -appearance would suggest to my readers that only birds of the habits of -the ostrich would venture on such a diet. It is, however, worth noting -that whilst in the Keeling Islands I learned that a cassowary that had -been kept on the atoll was a very efficient distributor of the seeds of -Ochrosia parviflora, scattering the undigested stones everywhere, and -causing the young trees to become so numerous that they had to be -destroyed. A similar habit of the cassowary in the Aru Islands is -recorded by Beccari, where the dry fruits of a palm, 2-1/2 inches -across, are swallowed by these birds and the seeds dispersed. -Cassowaries are active agents in dissemination, for they swallow every -kind of pulpy fruit, and convey them long distances undigested; they are -also excellent swimmers and traverse considerable expanses of water -(Beccari, quoted in _Chall. Bot._, iv., 297, 313). - -Modern ornithologists would probably not object to our appealing to the -former volant habits of the cassowary and its allies even across a wide -tract of sea; but, excepting in New Zealand and its vicinity, such birds -are not at our disposal in the island groups of the open Pacific. There -is a possibility that the extinct Columbæ and other exterminated birds -of the Mascarene Islands might account for some anomalies in their -floras; and in Chapter XVI. reference is made to the fact that these -islands possess more endemic species of Pandanus than any other oceanic -groups, a genus possessing drupes that in the case of inland species -seem unfit for any mode of dispersal with which we are familiar. In the -islands of the tropical Pacific, however, it is not possible to find -such a way out of the difficulty, since, as shown in Chapter XXXIII., -the birds are lacking. - -The genus, according to the _Index Kewensis_, includes about ten species -distributed over the islands of the Indian Ocean, and found also in -Malaya, Australia, and throughout the Pacific. It is essentially an -insular genus, and two at least of the species are wide-ranging littoral -trees, one, Ochrosia borbonica, mainly distributed over the islands of -the Indian Ocean and of Malaya, and the other, O. parviflora, chiefly of -the islands of the Pacific. It will be out of place to deal here in any -detail with this interesting genus, and my remarks will be confined to -such matters as concern the origin of the inland species of the Hawaiian -Islands, species that are peculiar to that group. Some confusion has -prevailed amongst different authors in the determination of the limits -of the various species, and to avoid this I have mainly followed -Schumann in his monograph on the order (Engler’s _Naturl. Pflanz. Fam._, -Theil 4, Abth. 2, 1895), as indicated in Note 57. - -Besides the littoral species Ochrosia parviflora, Hensl., that ranges -over most of the archipelagoes of the Pacific from the Solomon Islands -to Tahiti, but is not found in Hawaii, we have in the Pacific, O. -elliptica, Lab., of New Caledonia and Fiji; another species of New -Guinea and the Ladrones; and one or two inland species of Hawaii. -Ochrosia parviflora was familiar to me on Keeling Atoll, in the coral -islets of the Solomon Group, and on the islets and coasts of certain -parts of Fiji. Its fruits, which are dispersed by the currents, were -found amongst the stranded drift of the Keeling and Fijian beaches. -Although usually a coast-tree in Fiji, it came under my notice in one -locality growing inland; and it is a very suggestive circumstance in -connection with the inland species of Hawaii, that in Tahiti this tree -is only described by the French botanists as growing in the mountains at -elevations of 700 to 800 metres above the sea, it having for some reason -abandoned the beach. The process which we thus see in operation in -Tahiti is completed in Hawaii, and we there find a peculiar inland -species far away in the interior of the islands which is placed by -Schumann in the same section of the genus with the littoral O. -parviflora, that is not, however, found in the group. It may be remarked -that Gray describes only one species from Hawaii, O. sandwicensis, but -Schumann makes two species of it—one, O. compta, Sch., peculiar to the -group and referred to the same section as O. parviflora; the other, the -original species of Gray, which he considers as probably a variety of O. -borbonica. These determinations of the German botanist, who had no -theory to serve, are especially interesting. It is with the littoral -trees now missing from the Hawaiian beaches that he compares the inland -species of the group, trees now chiefly characteristic the one of the -Indian Ocean and the other of the South Pacific; and we can scarcely -doubt that originally one littoral tree ranged over both oceans. - -Hillebrand describes Ochrosia sandwicensis of Gray as a shrub or small -tree, 6 to 12 feet in height, growing in the open woods of the lower and -middle regions on all the islands. Its dry ellipsoid fruit is two inches -(5 cm.) long, and possesses a thin suberose covering on one side and a -very thick woody endocarp, one-quarter to one-third of an inch (6 to 8 -mm.) in depth. The other species which he characterises as a variety is -not so generally distributed in the group. We have to explain not only -how the original species reached the group, but also how they have been -distributed over the islands. The currents could scarcely have -transported the fruits as we now see them. Those of O. sandwicensis have -only a trace of a buoyant covering, and, judging from some fruits that I -examined, they could possess little or no floating power. Even the most -enthusiastic advocate of dispersal by birds must pause here; and there -remains the view, supported by evidence of a striking character, that -the inland Hawaiian species are derived from littoral species that, -having been originally brought by the currents, like O. parviflora in -Fiji, abandoned the beach and took to the mountains, where they have -become differentiated. - -It is probable that the lesson of Ochrosia in Hawaii can be applied to -one or two of the other Hawaiian “difficulties,” and that plants that -now set at defiance all the attempts of the student of dispersal to -explain their occurrence in this group may have commenced their -existence in these islands as littoral species brought originally by the -currents and afterwards driven off the beach. One of the greatest -enigmas of the Hawaiian flora is connected with another small -Apocynaceous tree peculiar to the group and described by Hillebrand as -Vallesia macrocarpa and by other Hawaiian botanists as a species of -Ochrosia. Schumann, however, places it in a new genus, Pteralyxia, near -to Alyxia, a genus already in the islands. However this may be, its dry -drupaceous fruits two inches (5 cm.) in length, and its pyrenes almost -as long, could never have been transported as such by the birds of our -own time; and if they could have been carried in the stomach of a bird -given to the dietetic humours of the cassowary, such birds in their -trans-oceanic passages would have left some trace behind in the groups -of the mid-Pacific. In our perplexity we read again the lesson of -Ochrosia. - - - _Summary of Chapter_ (see end of Chapter XVI.). - - - - - CHAPTER XVI - - THE RELATION BETWEEN LITTORAL AND INLAND PLANTS - (_continued_) - -The Fijian difficulty.—Inland species of a genus possessing - fruits not known to have any means of dispersal through - agencies now at work in the Pacific.—Pandanus.—Its remarkable - distribution in oceanic groups.—To be attributed perhaps - to extinct Columbæ or extinct Struthious - birds.—Barringtonia.—Guettarda.—Eugenia.—Drymispermum.—Acacia - laurifolia.—Conclusions to be drawn from the discussion.—Summary of - chapters XIV, XV, XVI. - - - SECTION IV - -HERE we deal with two genera, Pandanus and Barringtonia, where inland -endemic species occur in the same group with the wide-ranging coast -species, but possess fruits concerning which it is either difficult or -almost impossible to suggest a mode of dispersal by existing agencies. -This section is especially concerned with Fiji, and represents the -peculiar “Fijian difficulty” that is illustrated by other genera as—for -instance, the Coniferous genus Dammara—which are not in any sense -littoral. Further investigation is, however, requisite in the case of -Barringtonia, and to a less degree with Pandanus; and I can only here -point to the general indications of the data at my disposal. We have in -these genera to assume either that the inland species are derived from -the coast species, or that the seeds were brought by one of the extinct -birds of the Western Pacific, by a megapode or by one of the Columbæ, or -by some Struthious bird like the moa or the cassowary, or, if these two -assumptions fail, that there has been a continental connection through -the islands to the westward with the mainland beyond. - - - PANDANUS. - -I take this genus first because the recent monograph on the Pandanaceæ -by Dr. Warburg (Engler’s _Das Pflanzenreich_, 1900) enables me to tread -on relatively safe ground in making my deductions. The three genera of -the order, Freycinetia, Pandanus, and Sararanga, each tell their own -story; and in each and all of them I have taken an especial interest -from the standpoint of their dispersal. Freycinetia is fully discussed -in Chapter XXV., and presents no difficulties respecting its dispersal. -In the discovery of Sararanga the author has had a share. It was first -established by Mr. Hemsley from specimens sent by me to Kew in 1885; and -it has received from the botanist the name given to it by the natives of -the islands of Bougainville Straits in the Solomon Group, where I first -collected it. It contains only one species and was also discovered by -Dr. Beccari, the celebrated Italian botanist, in Jobie Island, New -Guinea. From the other two genera of the order, Pandanus and -Freycinetia, it stands quite apart; and it apparently presents us with a -relic of some ancient flora on the western borders of the Pacific. Its -fleshy drupes (one-half to three-quarters of an inch in size) inclosing -several small osseous pyrenes seem suited for dispersal by birds; and it -is not at first sight easy to understand why its distribution should be -so limited, unless this is connected with its diœcious habit (see -Guppy’s _Solomon Islands_, p. 302; _Journ. Linn. Soc. Bot._ vol. xxx.; -and Warburg’s monograph). - -It is, however, with the genus Pandanus that we are here especially -concerned. If the advocate of the previous continental connections of -Fiji and the groups around were to look for evidence in support of his -views, he apparently could not do better than take this genus. Whilst P. -odoratissimus, the littoral species of tropical Asia and Malaya, is -found on the coasts of almost all the Pacific islands from Fiji to -Tahiti and northward to Hawaii, it is only in the archipelagoes of the -Western Pacific, namely, in Fiji and Samoa, that inland endemic species -have been found. (Such species occur also in the more western islands -not dealt with here—New Caledonia, Solomon Islands, &c.) Not even in -Hawaii, with all its botanical evidence of antiquity, has an inland -endemic species been found, although the coast species extends miles -inland, and for nearly 2,000 feet up the mountain slopes. When, however, -we turn to Fiji and Samoa, we find in each group two endemic inland -species. To endeavour to connect the inland species of Fiji and Samoa -with the widespread littoral Pandanus odoratissimus, that owes its -dispersal largely to the currents, is out of the question, at least for -the student of plant-dispersal, since they belong to different sections -of the genus, and in their characters are often far removed (see Note -58). - -As regards the agency of birds, it is of course possible that -fruit-pigeons that can disperse the “stones” of Canarium and Elæocarpus -could transport the smaller drupes of Pandanus to oceanic islands like -the Fijis, Samoa, and the Mascarene Islands; and in Note 58 reference is -made to the size of the drupes of the endemic species of Pandanus in -those groups. But my difficulty is that I have not come upon any record -of birds eating these fruits; and I should imagine that amongst living -birds only those like the cassowary and its kin would prefer such a kind -of diet; whilst the only pigeon that could have ever attempted it must -have been able to swallow pebbles like the dodo. It is remarkable that -the Mascarene Islands, the home of the extinct Columbæ, possess more -endemic species of Pandanus than any other groups. - -Dr. Warburg points out that, with the exception of some three or four -species dispersed by the currents (P. dubius, P. leram, P. polycephalus, -P. odoratissimus), almost all the species (156 in number) are very -restricted in their areas. When we look at his table of the distribution -of the genus we notice that, excepting the islands of the Hawaiian and -Tahitian regions, nearly all the elevated or mountainous islands of the -tropical and subtropical latitudes of the Indian and Pacific oceans have -their peculiar species, whether in the case of Mauritius, Rodriguez, -Réunion, and the Seychelles in the one ocean, or of Lord Howe Island, -New Caledonia, Fiji, and Samoa in the other. The student here hesitates -even to raise the question of present plant-dispersal in the face of -such evidence of isolation all over the area of the genus. He is almost -inclined to evade the issue and to place the matter beside that of the -dying or extinct Columbæ that have been found in some of these islands, -as in Mauritius, Rodriguez, Réunion, and Samoa. - -For reasons above given in the instance of Fiji and Samoa, it would seem -futile to attempt to connect in their origin the inland with the coast -species; and it may be inferred that, excepting the few dispersed by the -currents, the species are in the main inland in their stations. Those -peculiar to Fiji, for instance, occur in the swampy forests of the lower -regions of the interior, as well as high up towards the mountain -summits. When traversing the Fijian forests I often used to speculate on -the modes of dispersal of the plants familiar to me; but the sight of a -strange Pandanus usually brought my speculations to a close. Many of the -enigmas of insular floras would be solved if we could interpret aright -the 156 species of Pandanus that are enumerated and described by Dr. -Warburg in his monograph. Observers like myself obtain little peeps into -the conditions of existence of these interesting plants; and the -travelled botanist, who becomes a systematist in his later years, -attains to a far more extensive view, yet even he can only penetrate the -mystery for a little way. - -It is doubtful whether Pandanus odoratissimus, the shore-tree of the -tropical beaches of the islands of the Pacific and Indian oceans, of -Australia, Malaya, and Southern Asia, can aid us much in any one -locality, since its distribution has no doubt been often assisted by -man. Yet it is probable that the currents have played a predominant part -in its dispersal. Its fruits occur commonly in beach-drift, both in the -Indian and Pacific oceans, and are often incrusted with serpulæ, -polyzoa, and cirripedes. At certain seasons the currents bring them to -Keeling Atoll in abundance. When, however, we come to inquire why it is -that this beach species is the only representative of the genus in -Hawaii and Tahiti, we are met with the possibility of its having been -introduced by the aborigines. The tree is almost as useful to a -Polynesian as the coco-nut palm, and it has been cultivated by him in -some of the atoll-groups, as in the Marshall and in the Radack -archipelagoes. In Chapter VII. good reasons are advanced for regarding -it as an aboriginal introduction into Hawaii. When, therefore, we learn -that in the group just named it extends from the sea-coast to nearly -2,000 feet above the sea, that in Samoa it may at times be found at a -similar elevation though usually restricted to the sea-border, and that -in the same way in Tahiti and in Fiji it may leave the coast-region and -extend into the heart of the islands, we are not inclined to look for -any marked differentiation in its character. This indeed appears to be -the case. Numerous varieties in different regions are referred to by Dr. -Warburg; but the only important one in the Pacific islands here -mentioned is a cultivated form from the Marshall Group. A variety from -Hawaii is distinguished chiefly by the smaller size of its drupes. - -Assuming, therefore, that the inland species are as a rule not derived -from littoral species originally brought by the currents, and that no -birds of our own time are in the habit of carrying the drupes of -Pandanus to oceanic islands, in order to explain the distribution of -such species we have to choose between the possibility of the agency of -extinct Columbæ and birds similar in their habits and the alternative of -a continental connection. Dr. Warburg, who says but little of the mode -of dispersal of Pandanus drupes, regards the genus as having now two -centres, one in the East African islands (Madagascar, the Mascarenes, -and the Seychelles), and the other in Papuasia (New Guinea, extending -doubtless to New Caledonia). My readers will recall to their minds that -zoologists have at times felt bound to postulate a continent in both the -centres of the genus Pandanus. There is the well-known Lemuria of the -Indian Ocean, and then we have in the Western Pacific Forbes’ Antipodea -and Hedley’s Melanesian Plateau. - -Before, however, we accept the indications of the distribution of -Pandanus as favouring a continental hypothesis for either area it is -essential to exclude the agency of the extinct Aves. In this connection -it is of prime importance to notice that the Mascarene Islands are -remarkable, when contrasted with all other oceanic islands, not only for -the predominance of peculiar species of Pandanus, but also as having -been the home of extinct Columbæ like the dodo and the solitaire. The -dodo’s habit of swallowing pebbles of the size of a nutmeg -(_Encyclopædia Britannica_, vii., 322), and the solitaire’s inclination -for swallowing stones as large as a hen’s egg (_Birds_, by A. H. Evans, -p. 331), doubtless represent, as explained below, a capacity for the -dispersal of large fruits and seeds that would be regarded as -“impossible” for distribution by birds now. It is quite possible that at -some time the ancestors of these birds possessed the powers of flight -now owned by the Nicobar pigeon, in the gizzard of which, in the Solomon -Islands, I found quartz pebbles half an inch across (_Solomon Islands_, -p. 324). In the work just quoted I refer on page 325 to the observation -of Messrs. Chalmers and Gill that the Goura pigeon of New Guinea usually -carries a good-sized pebble in its gizzard. We do not, however, seem to -possess any record of extinct Columbæ in the tropical islands of the -Western Pacific. The nearly extinct Didunculus of Samoa apparently -prefers berries and soft fruits. Dr. Reinecke says that it especially -favours the berries of Cananga odorata, the seeds of which are not over -a third of an inch (8 mm.) in length. - -It would appear from Mr. Hamilton’s note in the _Transactions and -Proceedings of the New Zealand Institute_ (vol. 24) that the extinct -Struthious birds of New Zealand, as in the case of the moa, carried -crop-stones sometimes as large as a pigeon’s egg. These pebbles are, of -course, swallowed by birds to enable them to crush the hard seeds, and -“stones” of fleshy fruits, on which they feed. In the Solomon Islands I -noticed that the Nicobar pigeon was able in this way to crack the seeds -of Adenanthera pavonina, which for their fracture require a blow with a -hammer. The implication is that the extinct Columbæ were able to -transport to oceanic groups seeds and “stones” which no existing pigeon -could now carry over a tract of ocean. I am inclined to extend this view -also to extinct Struthious birds, and to suppose that they were able, -like the cassowary (see page 152), to fly across tracts of sea in ages -gone by. Though such an agency would come under discussion in connection -with the floras of New Zealand and Madagascar, we have no evidence to -show that birds of this family ever reached the tropical islands of the -open Pacific. - -The Megapodidæ of the Western Pacific are a family of birds that suggest -themselves in this connection. Their distribution corresponds with that -of Pandanus in the Western Pacific, excepting the littoral species; and -like Pandanus the Megapodes have “differentiated” in every group. The -limited powers of flight possessed by existing species would unfit them -for crossing wide tracts of sea; but the parent form or forms of all -these species must have been able to traverse broad tracts of ocean. -These birds subsist on fallen fruits, seeds, &c.; but I have no data -relating to them as seed-dispersers. - -It is evident from the endemic character of most of the species of -Pandanus in oceanic islands that, except with a few widely-spread -littoral species, the dispersal of the genus has been for ages -suspended. Whether the explanation is to be found in the isolation and -differentiation of the extinct Columbæ of the Mascarene Islands, where -the endemic species of Pandanus are most numerous, has yet to be -established. It seems to offer the only way out of the difficulty, -unless we accept the old view concerned with the continent of Lemuria. - - - BARRINGTONIA. - -There are two littoral species of this genus in the Pacific, B. speciosa -and B. racemosa, both widely spread over the Old World, but only the -first is generally distributed over the Polynesian region reaching east -to Ducie Island, whilst the second does not extend east of Fiji and -Samoa. With the exception of one or two inland species in Fiji and Samoa -no inland species have been recorded from the groups of the open -Pacific, and the genus is not represented at all in Hawaii. If it were -not for a suspicion that the aborigines may have aided in the -distribution of the inland species, the advocate of the previous -continental connections of the islands of the Western Pacific would -receive from their occurrence in these islands considerable support for -his views. The fruits of the inland Fijian species are large, the -smallest being three inches in length; and the agency of birds seems to -be out of the question. - -The fruits of the littoral species possess dry buoyant husks that enable -them to be carried by the currents over wide tracts of ocean. Those of -the Fijian inland species display only a trace of these buoyant -coverings and the floating power is much diminished or absent -altogether. These inland species are two or three in number. One of -them, described as a new species by Seemann under the name of B. edulis, -has edible kernels and is sometimes cultivated. A species that I found -growing in the plantations of the Solomon Islanders in Bougainville -Straits may be near the Fijian tree just named (_Solomon Islands_, pp. -85, 297). Its kernels are edible; and I may add that the Solomon -Islanders cultivate other species with edible fruits. We cannot, -therefore, exclude the agency of the aborigines in the distribution of -the inland species of this genus. Horne found an undescribed species in -Fiji, which may be that which I found on the slopes of Mount Seatura in -Vanua Levu, as described in Note 50; and it is quite possible that it -was originally a cultivated tree, though not necessarily within the -memory of the later generations of the aborigines. - -This retrocession to the wild state of cultivated plants and the -resulting production of apparently new species is a point on which Dr. -Beccari lays considerable stress in the English edition of his book on -the Great Forests of Borneo. He takes the case of Nephelium and other -fruit-trees and shows how in old clearings, long since abandoned, they -have undergone singular alteration in characters. For these reasons, -therefore, Barringtonia can scarcely be regarded as offering in its -inland species unequivocal evidence of a previous continental condition -of the islands of the Western Pacific. Nor, as shown in Note 50, should -we be justified in establishing a genetic connection between the inland -and coast species; but a great deal of research is needed before we can -handle the numerous interesting problems connected with the genus; and -indeed it cannot be said that the specific limits of the inland -Polynesian trees have been definitely determined, or the species -themselves diagnosed. - - - SECTION V. - -In this section are included those genera where within the same genus -some inland species have been derived from the coast species whilst -others have been originally brought by birds. Guettarda alone belongs -here. In this genus we find, as is so frequently the case, a littoral -tree (G. speciosa) widely spread in the Old World and ranging over the -whole tropical Pacific as far east as Pitcairn and Elizabeth islands, -but absent from Hawaii. Here also as with Pandanus it is only in the -Western Pacific that we find inland endemic species so distinct in -character from the littoral tree that they may be regarded as of -independent origin. - -Since, however, there is an inland form of the coast species in Tahiti -(Guettarda speciosa, var. tahitensis) which, according to Drake del -Castillo, is distinguished only by its more rounded leaves and by the -more marked pubescence of the under leaf-surfaces, we evidently have -there an inland species in process of development from the littoral -species. This inland tree is found at elevations as great as 600 metres -or almost 2,000 feet above the sea; and indeed if we follow Nadeaud the -specific differentiation is complete. However, there is no doubt raised -as to its close affinity to the beach tree; and we are almost compelled -for another reason to regard it as a derivative of the shore species, -because, as pointed out in Chapter XXVII., there are very few inland -plants in the Tahitian flora possessing fruits as large as those of -Guettarda that owe their presence in those islands to frugivorous birds. - -Of the two inland species of the genus found in Fiji, G. inconspicua and -G. vitiensis, it may at once be said that, as indicated in Dr. Seemann’s -work, their characters are far from suggesting any connection in origin -with G. speciosa, the shore-species, the inland and littoral plants -belonging to different sections of the genus. In their case we can only -look to the frugivorous bird for the explanation of their existence in -the group. The fruits would be probably small; and in this connection it -is to be noted that Mr. H. N. Ridley in his paper on the flora of -Fernando Noronha evidently looks to birds to account for the presence of -a species of Guettarda on the island, a species not found elsewhere. - -But another inland Fijian form of Guettarda found by me in Vanua Levu at -elevations of 1,000 to 1,400 feet above the sea, and dubbed by the -natives with the name of the littoral tree (Mbua-mbua), corresponds in -its close relation to G. speciosa with the inland Tahitian form of that -tree, and is to all appearance a derivative of it. It is chiefly -distinguished by its thinner, more hairy leaves, which taper at each end -and are not subcordate at the base as is often the case with the leaves -of G. speciosa. The coverings of the fruit are less fibrous and the -putamen is not so deeply notched or grooved. The difference also extends -to the buoyancy of the fruits in accordance with the principle laid down -in Chapter II. Whilst those of G. speciosa float for many months and are -of common occurrence amongst the stranded drift of tropical beaches, as -for instance in the Keeling Islands, in the Solomon Group, and in Fiji, -those of the inland species float only for a few weeks, their softer -coverings decaying more rapidly in sea-water. - -We seem therefore to have had two principles at work in Fiji in -determining the origin of the inland species of Guettarda. Whilst in one -case the inland species is so sharply distinguished from the coast -species as to require the independent agency of frugivorous birds to -explain its presence, in the other the inland form, as in the instance -also of the Tahitian variety, is so much akin to it that the probability -of derivation from it is very great. - - - SECTION VI. - -In this section are contained genera possessing littoral species -restricted to the Western Pacific islands, and dispersed by birds, but -having little or no capacity for dispersal by the currents. They are -regarded as derived from the inland species of the genus in the western -part of the Pacific, and as distributed from thence over the islands in -that part of the ocean. We are here only concerned with Fiji, Tonga, and -Samoa and the neighbouring islands. The genera Eugenia, Drymispermum, -and Acacia are here comprised. - -The genus Eugenia, though essentially inland in its station, is apt to -lend species to the beach-flora in different parts of the tropics. Such -species, being dispersed by frugivorous birds and other animals, and -possessing but slight capacity for distribution by the currents, are -usually restricted in their areas. Thus, Schimper (p. 118) names two or -three species, including E. javanica, as amongst the Indo-Malayan -strand-flora. Ridley notices that E. grandis is a common sea-shore tree -in the Malay peninsula; and the author observed two littoral trees of -the genus in the islands of Bougainville Straits in the Solomon Group, -the fruits of one of them that flourished in the interior of the coral -islets being found in the crops of fruit-pigeons. So also in Fiji, some -of the inland species, as E. rariflora, appear at times amongst the -strand vegetation and in the coral islets. There is, however, one Fijian -species found also in Samoa and Tonga that is a characteristic beach -tree, namely E. richii (Gray), and it is more or less confined to that -station. The fruits will float a fortnight in sea-water, which is nearly -twice as long as most other Eugenia fruits will float; and it is quite -possible that the currents may assist the pigeons in distributing the -species. This genus is dealt with more in detail in Chapter XXVI. - -The genus Drymispermum (Thymeleaceæ) comprises in the Western Pacific a -number of species, of which two range over the groups of Fiji, Tonga, -and Samoa, whilst some four or more are peculiar to Fiji. All are inland -plants with the exception of D. Burnettianum, a characteristic littoral -shrub of these three groups. Its bright red drupes float only from five -to ten days, even after some weeks of drying; and like those of the -inland species they are well suited for dispersal by fruit-pigeons. This -beach-plant may be regarded as probably an intruder in the strand-flora -from the interior of one of the islands of the Western Pacific, whence -birds, perhaps assisted a little by currents, have carried it to the -neighbouring groups. - -The very remarkable coast tree, Acacia laurifolia, alone represents its -genus in the littoral flora of the Pacific islands. It is confined to -the Western Pacific, having been found in New Caledonia, the New -Hebrides, Fiji, Tonga, and Samoa; but it is doubtful whether it is truly -indigenous in all these localities. Thus, in Samoa, though restricted to -the coast districts, as we learn from Reinecke it seldom flowers, and -according to that botanist it was probably introduced through -cultivation. It is, however, evidently regarded by the Samoans as a tree -of their group, as is shown in a curious legend, given by Dr. George -Turner in his latest book on those islands, which I have quoted in my -book on the Solomon Islands, p. 287. Both in Fiji and Samoa it bears the -name “tatangia” or “tatania,” whilst its hard wood was employed for -various purposes, the leaves being used as spoons. The tree flowers and -seeds freely on the Fijian beaches. The pods dry up on the plant, and do -not dehisce, but are apt to break across between the seeds into -article-like portions, the seeds being ultimately liberated by the decay -of the pod or its fragments. The seeds either sink at once or in the -course of a day or two; whilst the pods or their fragments float at -first in sea-water, but all are at the bottom in a week or less. With -its absence of any apparent means of dispersal this small tree presents -quite an anomaly in the strand-floras of the Western Pacific, and can -only be regarded as a loan from the inland flora, though probably of a -very ancient date, and perhaps going back like Acacia koa, the -forest-tree of Hawaii to some early epoch in the history of these -islands. - - - _The conclusions to be drawn from the discussion of the relations -between the littoral and inland species of the same genus in the Pacific - islands._ (Chapters XIV., XV., XVI.) - -In ten of the twenty-two genera here dealt with (Calophyllum, Hibiscus, -Colubrina, Morinda, Scævola, Cordia, Ipomœa, Vitex, Tacca, Casuarina) -the shore and inland species have their own independent modes of -dispersal, usually by currents in the case of coast plants, and by birds -in that of inland plants; and the relations between the two are not such -as to suggest a derivation of one from the other. - -In six genera the inland species are regarded as derived from the -littoral species. In two of them, as in Vigna and Premna, where the -coast and inland species occur in the same group of islands and are -connected by intermediate forms, there is direct evidence in favour of -this conclusion; but such a development of inland species need not have -taken place in every group, since in the instance of Premna it has -apparently occurred only in the Western Pacific, and the inland and -coast species have extended independently to the eastern groups through -the agencies of birds and currents.... In the other four genera -(Canavalia, Erythrina, Sophora, Ochrosia) we have presented the -so-called “Hawaiian difficulty,” that group being alone concerned. -Although these genera have no littoral species in Hawaii, they have -inland species in those islands, which are in three genera endemic. -Since these inland species have non-buoyant seeds or seedvessels, the -transport of which by birds half-way across the Pacific Ocean is in the -case of the first three genera unlikely and in the last impossible, it -is assumed that they are all derived from original coast species with -buoyant seeds or fruits, such as are widely distributed over the Pacific -but are not now existing in Hawaii. This assumption, in the instance of -the Leguminosæ, to which the first three genera belong, derives support -from the singular fact in the distribution of the order pointed out by -Mr. Hemsley, that it is wanting in many oceanic islands where there is -no littoral flora. - -In one genus, Guettarda, the inland species are regarded as having been -sometimes developed independently of the coast species, and as at other -times derived from it, both principles having been at work in Fiji and -only the last in Tahiti. - -In two genera, Pandanus and Barringtonia, which represent the “Fijian -difficulty,” there is no reason on grounds of affinity to connect the -inland with the coast species; and since the agency of existing birds is -improbable in the first genus and out of the question in the second, -whilst the operation of the currents is excluded for the inland species -of both genera, it is assumed that we must either appeal to the agency -of extinct birds, such as those of the Mascarene Islands, or we must -fall back on the hypothesis of a continental connection. In the instance -of Barringtonia it is also possible that some of the inland species may -have been derived from species spread through cultivation. - -Lastly, in three genera (Eugenia, Drymispermum, Acacia) the coast -species are viewed as derivatives of the inland flora in the Western -Pacific, not necessarily in Fiji, but it may be in New Caledonia or in -one of the other large groups. In this case the coast species of all -three genera are either unfitted for dispersal by currents, or display -the capacity only in a small degree. - -We thus see that in only seven of these twenty-two genera, containing -both littoral and inland species in the Pacific islands, can it be -argued from the standpoint of dispersal that the inland species are or -may have been derived from the shore species; and in most instances the -evidence is largely presumptive in its character. In three genera the -reverse has been the case, and here the coast has borrowed from the -inland flora. In twelve, or more than half of the genera, the shore and -inland species have been evidently independent in their origin. It is -accordingly apparent that in the Pacific the strand flora has lent more -to the inland flora than it has borrowed from it; but with a large -proportion of these coast genera no interchange has taken place. -Two-thirds of the genera of the beach-plants have no inland species, and -in their case the question of such a connection cannot be raised. With -the remaining genera such a relation can be suggested in only two-fifths -of the cases, or in about one-seventh of the total number of beach -genera. Where a connection can be traced, it points more frequently to -the derivation of the inland from the shore plant. Taking all the -evidence together, the beach flora presents itself in the Pacific as -practically independent of the inland flora as regards its origin. It -has received in these regions but few recruits from inland. It has -yielded, except in Hawaii, but few recruits to the inland flora. In this -ocean it bears the stamp of a high antiquity, though in the mass no -doubt of more recent origin than the mangrove flora. - -Yet, as I have remarked in different parts of this work, even with the -beach genera possessing no inland species, considerable variety is -displayed in the behaviour of the strand species. Thus, whilst some, -like Pemphis acidula, Tournefortia argentea, and Triumfetta procumbens, -rarely if ever leave the beach, others, like Heritiera littoralis and -Excæcaria agallocha, find a home on the borders of the mangrove swamps, -and one or two extend inland and take their place in the forests, either -as trees (Afzelia bijuga) or as giant climbers (Entada scandens). Others -again, like Cassytha filiformis, Cerbera Odollam, and Cycas circinalis, -with a number of other beach-plants, may invade the interior of the -island wherever arid plains or exposed scantily wooded districts offer -conditions conformable to the xerophytic habit of the beach-plants. - -It will thus be perceived that although the inland and coast floras of -an island are in the mass distinct, the line of separation is by no -means always well defined. Beach-plants are something more than -salt-lovers in their ways. They are in the first place xerophilous, or, -in other words, they will be equally at home in exposed situations away -from the coast where the soil is dry and the rainfall scanty. Whenever -these conditions are presented by the districts backing the coast, as we -find for instance in the plains on the lee or dry sides of many a -Pacific island, the shore-plants will often leave the beach and travel -far inland. - - - _Summary of Chapters XIV., XV., XVI._ - -(1) Though littoral floras are as a rule chiefly made up of two sets of -plants, one brought through the agency of the currents from regions -outside, and the other derived from the inland flora of the region -concerned, the proportion of the two varies much amongst temperate and -tropical strand-floras, the current-borne plants forming the majority in -the tropics, and those from the inland flora of the region prevailing in -the temperate zone. - -(2) There is, therefore, far greater uniformity as a rule amongst -tropical strand-floras than in the temperate zone, since in temperate -latitudes the prevailing constituents of the strand flora vary with the -inland flora of every region, whilst in the tropics the predominant -plants are those ranging far and wide on the shores of the warm regions -of the globe. - -(3) Regarding the tropical strand-flora as comprising two formations, -that of the beach and that of the mangrove swamp, the last, which is the -older of the two, may, it is suggested, be viewed as the remnant of an -ancient flora widely spread over the lower levels and coastal regions of -the globe, during an age when, in a warm atmosphere charged with watery -vapour and heavy with mist and cloud, vivipary or germination on the -plant was not the exception but the rule. - -(4) But it is contended that even in the beach formation some of the -plants may date back to this age of vivipary, as is indicated by the -anomalous seed-structures of some of the genera, such as Barringtonia, -which seem to indicate a lost viviparous habit. - -(5) Since the beach formation of the islands of the tropical Pacific is -largely formed of plants ranging over great areas in the tropics, there -is no reason to expect that it owes much to recruits from the inland -floras of this region. The discussion, therefore, of the relation -between the littoral and inland floras is mainly concerned with the -possible origin of inland from coast plants in these islands. - -(6) Yet there are numerous cases of genera possessing both coast and -inland species that are of peculiar interest in determining the true -relation between the beach and inland floras. - -(7) As the result of a detailed discussion of these genera, the -conclusion is formed that the beach and inland floras have been in the -main developed on independent lines, the beach flora receiving from the -inland flora but few recruits, and except in Hawaii yielding but few -plants to the inland flora. Only a third of the genera of the beach -flora have also inland species, and in only a few of these genera, or -about a seventh of the whole beach flora, can any question of a -connection between coast and inland species of the same genus be raised. - -(8) Two special difficulties arise in this discussion. The first is the -“Hawaiian difficulty,” which is more particularly concerned with genera -of the orders Leguminosæ and Apocynaceæ. Here are genera which possess -both inland and littoral species, but only the first occur in Hawaii. In -the absence of any likely means of dispersal, whether by currents or by -birds, it is assumed that the inland species are derived from shore -plants, originally brought by the currents, that have since disappeared, -a view supported by the fact that Leguminosæ are wanting in oceanic -islands where there is no littoral flora. The second is the “Fijian -difficulty” which is best represented by Pandanus. From our inability to -regard the inland species as derivatives of the coast species, or to -supply them with a means of dispersal, we are compelled to regard them -either as having been a part of the original continental flora of Fiji -or as owing their existence there to the agency of extinct birds having -the habits of the Nicobar pigeon and of the extinct Columbæ of the -Mascarene Islands. Since the Mascarene Islands are noted not only for -their extinct Columbæ but also for their number of peculiar species of -Pandanus, the implication seems to lie against the continental view. The -subject, however, awaits further investigation. In the Western Pacific -the possible agency of the parent forms of the existing species of -Megapodidæ is worthy of attention. Like the Columbæ and Pandanus in the -Mascarene Islands, the Megapodes and Pandanus have “differentiated” -together in the Western Pacific. - -(9) The general view of the independent origin of the beach and inland -floras of the Pacific islands is supported by the large number of genera -in the strand flora that only possess littoral species. - -(10) Such shore species, together with other strand plants, sometimes -extend into the interior of an island, but only as a rule where the -requisite conditions for a plant of xerophilous habit exist. - -(11) Shore plants, it is pointed out, are xerophytes first and -halophytes afterwards; and under certain conditions the purely -xerophilous inclination prevails and the plants travel far inland. - - - - - CHAPTER XVII - - THE STORIES OF AFZELIA BIJUGA, ENTADA SCANDENS, AND CÆSALPINIA - BONDUCELLA - -Afzelia bijuga.—The African home of the genus.—The double station of - Afzelia bijuga, inland and at the coast.—The nature of the buoyancy - of its seeds.—Summary relating to Afzelia bijuga.—Entada - scandens.—Its station and distribution.—Darwin’s opinion of the - plant.—The dispersal of its seeds by the currents.—Summary relating - to the plant.—Cæsalpinia bonducella and C. bonduc.—Their station and - distribution.—Their characters in various Pacific groups.—The - parents of inland species.—Their dispersal by the currents.—The - germination of their seeds.—A dream of vivipary.—The causes of the - seed-buoyancy.—Summary of results. - - -IN this chapter we have a study of Leguminous strand plants that are of -great interest. It can be safely said that the student of -plant-dispersal in the Pacific will be brought into contact with the -problems here involved wherever he goes. - - - AFZELIA BIJUGA (Gray). - -This Old World tree, which belongs to the sub-family Cæsalpiniæ, is of -great interest to the student of plant-dispersal. It is one of that -large group of Indo-Malayan plants that extend into the Western Pacific, -and give the prevailing character to the floras of such archipelagoes as -that of Fiji. It is a large tree yielding a valuable timber used by the -Fijians and Samoans for many purposes, such as for canoes, house-posts, -clubs, kava bowls, &c., but it has not been recorded from the Tahitian -region, and is unknown from Hawaii. In the fact of its being a littoral -as well as an inland tree, it possesses a peculiar interest from the -standpoint of plant-dispersal, and especially since this difference in -station is associated with a difference in buoyancy, the seeds of the -inland trees usually sinking, whilst those of the coast trees usually -float, and often for a period of months. - -A glance at the distribution of the genus will enable us to appreciate -some of the points that will be touched upon in the following -discussion; and it may be here remarked that the explanation of the -distribution of these Leguminous trees will go far to make clear some of -the most difficult points in plant-geography. Of the eleven species -enumerated in the _Index Kewensis_, five belong to tropical Africa, -occurring on both the east and west coasts as well as in the interior, -three are confined to the mainland of tropical Asia, and two are -peculiar to Malaya. In the last place we have the wide-ranging Afzelia -bijuga, which, if it does not actually occur on the east coast of -Africa, is found at all events in Madagascar and in the Seychelles, and -is to be followed by the way of the Chagos Archipelago to the Malayan -Islands and Queensland, and eastward to Fiji and Samoa. - -The most suggestive feature in the distribution of the genus is to be -seen in the frequent station of the species by rivers. We learn from -Oliver’s _Flora of Tropical Africa_ that these trees find a home along -river-courses on both sides of the continent, as on the banks of the -Congo, the Niger, the rivers of Senegambia, and the Zambesi, the Zambesi -species being found also on the shores of Lake Nyassa. Since tropical -Africa possesses about half of the species, it would seem highly -probable that it is the home of the genus, and that from the -rain-forests in the heart of the continent rivers flowing east and west -have borne the buoyant seeds of the wandering species to the coasts of -the Atlantic and Pacific Oceans. The operation that I witnessed on a -miniature scale in the case of a _species_ of Entada (E. scandens) in -the Isthmus of Panama, as described in a later page of this chapter, has -been in progress through the ages with the _genus_ of Afzelia in the -breadth of the African continent. According to the principle illustrated -by Afzelia bijuga in the forests of Fiji, the seeds of the African -forest-trees would, as a rule, possess no floating power; but now and -then in the lapse of long periods of time buoyancy in some species would -be developed, and such species would ultimately, through their buoyant -seeds, find their station along the lower courses of the rivers. - -To sustain this view it is not necessary that continuous rain-forests -should now clothe the elevated regions in the interior of tropical -Africa; but it is requisite that there should be sometimes a generic -similarity between the plants of the East African and West African -rain-forests; and it is evident that this is the case. Pechuel-Lösche, -as quoted by Schimper (_Plant-Geography_, p. 299), describes the -rain-forest on the Loango coast as covering the mountain ranges and as -extending to the river-plains. In such a locality the operation would be -rapid. In advancing this hypothesis I am referring to the possibility, -however, of such an operation having effected the distribution of -Afzelia in tropical Africa in the past rather than in the present. I -would suggest that botanists in other habitats of the genus, as for -instance in Queensland, might put it to the test of observation and -experiment. - -The interest that attaches itself to the story of the genus in its -African home may be extended to the species that forms its outpost in -the Pacific, and we shall see there a littoral species that doubtless -had its home in the interior of a continent endeavouring, with a -considerable measure of success, to become again an inland plant. Horne -(p. 112), who was familiar with Afzelia bijuga at the two extremes of -its range, namely, in the Mascarene Islands and in Fiji, speaks of it as -characteristic of the shores of tropical regions; and Schimper, who -includes it in the Indo-Malayan strand-flora, implies that it is more or -less exclusively confined to the coast and its immediate vicinity (pages -121, 191-2). In the Seychelles, according to Mr. Button, this tree -attains gigantic dimensions on the sandy flats. Still larger trees occur -in the coral islands of the Chagos Archipelago; but in the atoll of -Diego Garcia, as we learn from Mr. Bourne, it is almost extinct only -some four or five trees existing there about twenty years ago, the -increase of the tree being prevented through the destruction of the -fallen seeds by the rats (_Journ. Linn. Soc. Bot._, vol. 22, 1887). - -Afzelia bijuga may, therefore, be safely regarded as a littoral tree. We -shall now see the importance of this conclusion when we come to consider -its station in the Pacific islands, where it grows both inland and at -the coast, and we have to decide to which station we must assign the -priority. Speaking of its occurrence in Fiji, Dr. Seemann says it is -“common in the forests all over Viti,” but makes no allusion to it as a -littoral tree either in Fiji or elsewhere. On the other hand, Mr. Horne -(p. 112) describes it as “generally growing on the shore or sandy -beaches, and in rocky clefts, and by the sides of streams in the -interior of Viti Levu and Vanua Levu.” It was on or near the coast in -Fiji that the present writer was most familiar with this tree, sometimes -bordering the sandy beach, at other times growing behind the -mangrove-belt, or again thriving in the half sandy and half swampy soil -of some low islet off the mouth of the Rewa. Especially is it to be -found on those parts of the coast where the hill-slopes descend rapidly -to the beach, or where some lofty spur from the mountains of the -interior reaches the shore. It is also not uncommon on the banks of -rivers both in their lower and upper courses. But it is as a forest-tree -of the interior that it is most valued by both the white men and the -natives on account of the superior quality of its timber in that -station. There, far removed from stream or river, the Vesi, as the -Fijians name Afzelia bijuga, takes its place amongst the lofty -forest-trees, such as the Ndamanu (Calophyllum), the Ndakua (Dammara), -and the Wathi-wathi (Sterculia). It is not often that one finds a tree -in these islands that, like the Vesi, is able to make its home in almost -any station, excepting, however, the “talasinga” or “sun-burnt” regions -of the plains. Wherever tall trees grow gregariously in Vanua Levu, one -will probably find Afzelia bijuga, whether beside a sandy beach, or -bordering a swamp, or on a river’s bank, or on some rocky declivity, or -on the great forest-clad mountain-slopes and plateaux of the interior. -No doubt the same diversity of station is displayed in Samoa, where, -according to Dr. Reinecke, the tree is most frequent in the -“coast-bush.” - -From the variety in station it might be expected that corresponding -variations in character would be found. There are differences, such as -in the quality of the timber and in the size of the seeds between coast -and inland trees; but the most important distinction in connection with -the study of the dispersal of the species is to be found in the -circumstance that whilst the seeds of the coast trees are, as a rule, -buoyant, and often float for months, those of the inland trees usually -sink, even after being kept for three or four years. I made a -considerable number of experiments on the buoyancy of the seeds of this -tree in Vanua Levu, and found that with the coast trees, as a rule, -either all the seeds or the majority of them floated in sea-water, -whilst with the inland trees either all of them or the majority of them -sank. The buoyant seeds are able in most cases to float for a long time. -Thus, in one experiment half were afloat after two months, and in -another half were afloat after five months. It is probable that several -of the exceptions, where inland seeds float, will prove to be connected -with an inland station by a river. (I experimented on eight sets of -seeds of coast trees from eight different localities, and found 70 to be -the mean percentage of buoyant seeds. In the same way, four sets of -seeds from four different inland localities gave 13 as the mean -percentage of buoyant seeds.) - -As in the case of Entada scandens, there is a rather fine adjustment -between the mean specific weight of seeds and the density of water. If -we place a number of the buoyant seeds in sea-water and begin to lower -the density, some of the seeds will at once commence to float heavily -and afterwards sink; and when the density has been lowered to -approximately that of fresh water, usually about a third will be found -at the bottom of the vessel. Out of 100 coast seeds, 70 will, as a rule, -float in the sea and about 47 in the river; whilst of the same number of -inland seeds, 13 on the average will float in sea-water and 8 or 9 in -fresh water. The bearing of facts of this kind is especially discussed -in Chapter X. - -Coming to the causes of the floating-power of the seeds, we find that -with the buoyant seeds the kernel floats, whilst with the non-buoyant -seeds it sinks, the seed-tests in neither case possessing any -floating-power. In this respect, therefore, the seeds of Afzelia bijuga -belong, with the seeds of some other Leguminous littoral plants of the -Pacific islands, such as Canavalia obtusifolia, Erythrina indica, and -Sophora tomentosa, to the second section of the second non-adaptive -group of buoyant seeds (page 107). But though we can in a measure -explain the cause of the buoyancy, we are still ignorant of the manner -in which the difference in the buoyant behaviour of coast and inland -seeds has been brought about. It is possible that this may be connected -with another difference between the coast and inland seeds, the latter -being markedly smaller, and it is noteworthy that in my experiments the -smaller seeds were generally those that sank. (Whilst the inland seeds -averaged between 8/10 and 1 inch, or 20 to 25 mm., in greatest diameter, -12 to 16 being required to make an ounce, the coast seeds measured 1 to -1-2/10 inch, or 25 to 30 mm., and only 10 or 11 were needed to weigh an -ounce.) - -There can be no question that the seeds are at times transported by the -currents over wide tracts of sea, and this no doubt explains the -occurrence of Afzelia bijuga in oceanic islands. They may be usually -seen lying free in numbers on the ground beneath the tree or else still -inclosed in the fallen dehiscing and decaying pods; and they might be -swept sometimes into the sea or washed down into an adjacent stream. -They thus came under my notice amongst the stranded beach drift at the -mouths of estuaries in Fiji. But it is remarkable that the seeds have -not apparently been recorded from the beach drift of other tropical -regions. Penzig does not note them amongst the seeds stranded on the -shores of Krakatoa. They did not occur amongst my collections from the -beaches of Keeling Atoll or of the south coast of Java; nor does -Schimper mention them amongst the drift of the Java Sea. In the _Botany -of the “Challenger” Expedition_ the species is not even referred to in -any connection. Although, however, the capacity of these seeds for -dispersal by currents is for the first time established by me, their -fitness in this respect was surmised by Schimper (p. 191), when he -placed the species in his list of tropical shore plants evidently -distributed by the currents. - -It will thus be gathered that we have yet much to learn in this matter; -and I would recommend any resident in the tropics to take up this -subject. When indeed we remember the fine adjustment existing between -the specific weight of the seeds and the density of water, and recall -the unknown factor determining the difference in buoyancy between the -kernels of coast and inland seeds, we can understand how under -particular conditions in certain portions of its range the seeds of -Afzelia bijuga may perhaps never possess any floating power. It would -seem, in fact, that the seeds are much more buoyant in the Western -Pacific than they are in the Java Sea; or it may be that the tree is -much less frequent; or that the stranded seeds are soon destroyed by -crabs, such as is the fate of much seed-drift on the Keeling beaches; or -lastly that, as in Diego Garcia, rats in destroying the fallen seeds are -bringing about the extermination of the species. - - - _Summary relating to Afzelia bijuga._ - -(1) Assuming that the genus has its home in the African continent, and -that the species have frequently a riverside station, it is argued that -the distribution of the genus on both sides of that continent can only -be explained by its dispersal by rivers from a centre in the interior. - -(2) Afzelia bijuga, a widely distributed shore tree of tropical Asia, -occurs in Fiji both at the coast and in the inland forests. - -(3) This double station is associated _inter alia_ with a different -buoyant behaviour of the seeds, those of the coast trees usually -floating for long periods, whilst those from inland generally sink. - -(4) There can be no doubt that this widely ranging littoral tree has -been dispersed by the currents; but the specific weight of the coast -seeds is on the average but slightly less than that of sea-water; and it -is to this fine adjustment, always liable to be disturbed by variations -in the environment, that the irregularities in the distribution of the -species are to be attributed. - - - ENTADA SCANDENS (Benth.). - -The story of Entada scandens, a plant familiar to many of my readers -under the name of the Queensland Bean, is a story of three continents, -Africa, Asia, and America. From the point of view of its dispersal two -features at once attract attention in the case of this giant-climber; in -the first place its wide distribution over the tropics of the Old and -New Worlds, and in the second place the great capacity of its large -seeds, often two inches across, for dispersal by the currents. But -before discussing these matters it will be necessary to glance at the -distribution of the genus, since much light will thereby be thrown on -some of the numerous difficult points affecting this extremely -interesting tropical plant. Of the thirteen species enumerated in the -_Index Kewensis_, seven are African, three are American, one is Burmese, -one hails from Madagascar, and, lastly, there is the world-ranging -Entada scandens, concerning whose home botanists are not agreed. Most of -the species would seem to be inland plants, whilst Entada scandens -thrives both inland and at the coast. Africa would thus appear to be, as -with Afzelia, the principal home of the genus, but with America as a -subsidiary centre. - -In many points Entada scandens presents a parallel to Cæsalpinia -bonducella, another Leguminous tropical plant which occurs also at the -coast and inland. But since they both owe their wide distribution to -their littoral station, it will be as coast plants that they will be -most properly considered in this and the following chapter. Yet if the -student were to regard the distribution of these two plants in a -continental region as in India, where they extend inland to the -Himalayas, he might fail to discern their true station. To accurately -gauge the matter of their station, it is necessary for him to look at -the plants as they occur in the islands of the Pacific. There he will -first see the stranding of the seeds on a shore by the currents, then -their germination and their development into giant-climbers over the -littoral trees or into straggling bushes on the beach; and afterwards he -will observe the plants of both species extending inland, and in these -three stages he will learn their history in the Pacific; but a history, -it may be observed, that in this region represents their efforts to -return to an inland station, such as they once possessed in their -birthplace in some distant region of the globe. - -Dealing first with the station of Entada scandens, it may be remarked, -as Dr. Seemann points out, that in Fiji it is most characteristic of the -mangrove-formation. But it also occurs amongst the trees at the back of -the mangrove swamp, on the beaches, on the banks of the estuaries, and -at the edge of the inland forests where they border on the plains. -Sometimes in the company of Derris uliginosa it grows not as a climber, -but as a prostrate plant on the sandy beaches; and here, not being able -to assume its normal habit of a climber, it does not seed. It is to be -found at times far inland in open-wooded districts. Thus in Vanua Levu I -found it growing in the Mbua district four miles inland, and 1,400 feet -above the sea. Reinecke speaks of it in Samoa only in connection with -the “urwald,” or primeval forest. Cheeseman describes it as most -abundant in the interior of Rarotonga, covering the trees with a -wide-spreading canopy of green. In the Malayan region Schimper refers to -it as a plant of the beach-tree formation. In Ecuador and on the Panama -Isthmus it grows not only at the coast, but also on the hill-slopes in -the rear of the mangrove-belt. - -With reference to the distribution of the plant, it may be remarked -that, although it is found all round the tropics and possesses great -capacity for dispersal by currents, there are certain difficulties in -explaining its wide area and in accounting for its very peculiar -distribution in the Pacific islands. It was doubtless in allusion to -some of these difficulties that Mr. Darwin, in a letter to Sir Joseph -Hooker, remarked: “Entada is a beast” (_More Letters_, &c., i, 93). -There is at first the question of the identity of the species in the Old -and New Worlds. It is here assumed that it is the same in both -hemispheres; but it must not be forgotten that the identity is “not -beyond doubt” (_Bot. Chall. Exped._ iv, 147). - -Then there is the difficulty connected with its occurrence on both -coasts of tropical America. In this respect it is at one with some other -littoral plants, like Ipomœa pes capræ, as well as with the plants of -the mangrove formation, as is pointed out in Chapter VIII. Whilst with -the mangroves it is necessary to assume that they antedate the land -connection between North and South America, this is not requisite in the -case of Entada scandens, since it grows in the interior of the Panama -Isthmus, and rivers on the north and south sides now carry its seeds -seaward from the same “divide” to the Atlantic and Pacific Oceans, as -described in Chapter XXXII. - -But, as I have also shown in Chapter VIII, America forms with the West -Coast of Africa a region characterised by the same tropical littoral -flora. This region, on account of the arrangement of the currents, -stands in a very peculiar relation with the Asiatic region, which -comprises the rest of the tropics, and to a great extent possesses its -own peculiar strand-flora. There are a few littoral plants, like Entada -scandens, Canavalia obtusifolia, Sophora tomentosa, and Ipomœa pes capræ -that occur in both areas; but the large majority are confined to one or -other of them, either to the American region, including the African West -Coast, or to the Old World region, which includes the African East -Coast. The American region gives to the Old World, but it can receive -nothing in return. For this reason, it is argued, we are compelled to -regard most, if not all, of the cosmopolitan tropical shore plants that -are dispersed by the currents, such as those above named, as having -their home in the American region. Entada scandens would, therefore, -from this standpoint have its home in America. - -Then, again, there is the difficulty connected with the distribution of -this plant on both sides of tropical Africa. Though Oliver in his _Flora -of Tropical Africa_ mentions this species only in connection with the -West Coast, he says it is probably widely spread in that continent, and -he refers to a pod in the Kew Museum indistinctly labelled “Lake Ngami.” -I have not come upon any reference to its being a littoral plant on the -East Coast, but since numerous littoral plants of tropical Asia are -found on that coast its occurrence there or in the East African islands -would be expected. However, as the genus has a centre in America, and as -this species is regarded as of American birth, we are not called upon to -employ the argument used in assigning to a non-American genus like -Afzelia an African home. Since the African West Coast belongs to the -American region of tropical shore plants dispersed by the currents, the -presence of Entada scandens on that coast of Africa can be readily -explained, whilst if it has reached the Malayan Archipelago from America -by way of the Pacific, it would, by extending like many other Malayan -coast-plants along the shores of the Indian Ocean, almost complete its -circuit of the globe. It is in this fashion, I believe, that the other -littoral plants,, like Cæsalpinia bonducella, Canavalia obtusifolia, and -Ipomœa pes capræ, that are found all round the tropics, have performed -the circuit of the globe with America as their home. - -One may remark in passing that the double home of the genus in America -and the Old World, though offering a serious difficulty in plant -geography, has no immediate bearing on the present mode of distribution -of Entada scandens. Questions relating to the distribution of tropical -shore-plants that are dispersed by the currents at first resolve -themselves into considerations of the arrangement of the currents. -Entada is not alone amongst the genera containing littoral species in -having a home both in the Old and in the New World. Carapa is another -instance, and additional cases might be cited. - -The next peculiarity in the geographical range of this species is -concerned with its irregular distribution in the archipelagoes of the -tropical Pacific. Notwithstanding its great capacity for dispersal by -the currents, although it occurs in all the groups of the Western -Pacific as well as in the Cook Islands, it has not been recorded from -the Society Islands, the Paumotus, the Marquesas, and Hawaii. Since, -however, its seeds have been gathered by Mr. Arundel on the beaches of -Flint Island, lying about six degrees north of Tahiti (_Bot. Chall._ iv, -302), it is not unlikely that it will be found growing in other parts of -Eastern Polynesia south of the equator. One might have looked for an -explanation of its rarity in Eastern Polynesia to the absence of -mangrove swamps, in which, as in Fiji, it is sometimes most at home; but -this is negatived by its abundance in Rarotonga, where mangrove swamps -do not exist. - -_The dispersal of Entada scandens by the currents._—This plant offers -one of the most conspicuous examples of the transport of seeds across -oceans through the agency of the currents. In the pages of many -botanical works, from the close of the 17th century onward, reference is -made to the transport of its beans (often in association with those of -Mucuna urens and Cæsalpinia bonducella) by the Gulf Stream or other -currents across the Atlantic to St. Helena, the Azores, the west coast -of Ireland, the Hebrides, the Orkney Islands, the coasts of Scandinavia, -and even as far north as Nova Zembla (see Hemsley’s _Bot. Chall. -Exped._; Sernander’s _Skand. Veg. Spridningsbiologi_, &c.). That the -seeds of Entada scandens retain their germinating capacity after this -ocean-transport has been demonstrated not only by the germination of -stranded seeds on the shores of St. Helena, but also by the germination -when sown at Kew of seeds drifted to the Azores, as well as by the -results obtained by Lindman, who procured the germination of the seeds -of this plant and of Mucuna urens that had been washed up on the -Scandinavian beaches (see Sernander, pp. 7, 390). - -One of the most interesting references to the conveyance by currents of -these seeds to the coasts of Europe is to be found in Dr. Sernander’s -recent work on the modes of dispersal of the Scandinavian flora, where -he sums up the results of Lindman’s investigations respecting the Gulf -Stream drift. The stranded seeds of Entada scandens, it appears, have -been found all along the Norwegian coast, but occur most frequently -north of the Söndmöre district. They have even been found in a -sub-fossil condition in the peat-bogs of Tjörn on the Bohuslän coast in -Sweden, having been originally stranded on a beach in that locality at -some distant, but post-glacial, epoch. Few phenomena in -plant-distribution are more suggestive than this ineffectual transport -through the ages of these large tropical beans to coasts within the -Arctic Circle. The seed, no longer under the care of the mother-plant, -becomes a waif, exposed to the pitiless laws of the physical world which -here prevail. It was not thus that the plant was reared, but it is in -this haphazard fashion that its seeds are spread. The philosopher could -unravel most of the tangled problems connected with present and past -plant-dispersal, if he could follow the clue supplied by this stranded -tropical seed on a Scandinavian beach. - -It is a far jump from the North Cape to the coral islands of the Pacific -and Indian Oceans; yet it is within the area covered by the drifting -Entada bean. The stranded seeds occur commonly on the Fijian beaches and -on other islands of the South Pacific; but I never found them in Hawaii. -They were gathered by me on the shores of Keeling Atoll in the Indian -Ocean, and on the south coast of Java. Penzig found on the Krakatoa -beaches, in 1897, not only the stranded seed but the established plant. -They came under my notice in numbers on the beaches of Ecuador and on -the Pacific and Atlantic coasts of the Panama Isthmus; and, as I -learned, they are equally common on the other parts of the coasts of -Central America. Not uncommonly these stranded seeds in various parts of -the world are to be found incrusted with polyzoa and tubicular annelids, -which afford proof of prolonged flotation in the sea. These seeds are -also to be frequently noticed floating in the drift of the tropical -estuaries. Thus they came under my observation afloat in numbers in the -Fijian estuaries, in the Guayaquil river, in the estuary of the Chagres -at Colon, and in the mouth of a river on the Panama side of the isthmus. - -The mode of liberation of the seeds is worthy of a passing remark. The -huge pods, often several feet in length, ultimately break up into -separate joints bearing the seeds. The joints may decay on the ground, -and the seeds are thus freed; or not infrequently in a mangrove-swamp -they fall at once into the water, and there they float, as may often be -observed in Fijian rivers, until their decay sets free the seed. - -The seeds of Entada scandens are often quoted, and justly so, as -striking examples of the dispersal of seeds by currents. Yet in few -plants could the nature or the structural cause of the buoyancy have so -little claim to be considered as adaptive in its character. Quite half, -and sometimes even the majority, of the seeds freshly liberated from the -plant have no buoyancy at all. The mean specific weight of the seed is -about that of sea-water, but markedly higher than that of fresh water; -whilst the principal determining cause of the buoyancy is, as shown -below, purely mechanical, and one that, whilst favouring the wide -distribution of the species, could not be improved by or come within the -scope of Natural Selection. - -From experiments made in Fiji and Ecuador, it appears that at least 50 -per cent., and often more than half, of the seeds when first liberated -from the pod have no buoyancy in sea-water. Of those that float in -sea-water, a proportion varying between one-third and one-half sink in -fresh water, so that in the case of plants growing on the banks of a -river only about one-fourth or one-third would be carried down to the -sea. So fine is the adjustment of the specific weight of these seeds to -the density of water, a subject discussed in its general bearings in -Chapter X, that if one gathers a number of drift seeds on a beach, let -it be in Fiji or in Ecuador, although, of course, all will float in the -sea, only one-half or two-thirds will float in the neighbouring -fresh-water stream. Those that float appear to be able to float almost -indefinitely. This is sufficiently established by the transport of the -seeds in a sound condition by the currents across the Atlantic, and by -such evidence as the stranding of seeds incrusted with polyzoa and -serpulæ on the beaches of Keeling Atoll. It has been also proved by the -following experiment. Several years since, I placed a seed in a vessel -of sea-water, where it still floated buoyantly in a perfectly sound -condition twelve months afterwards. - -With regard to the cause of the buoyancy, investigation shows that -neither the seed tests nor the seed contents have any floating power, -the buoyancy arising from a large central cavity produced by the -shrinking and bending outward of the cotyledons during the drying and -hardening of the maturing seed (see figure in Chapter XII). With the -seeds that sink, this cavity is, as a rule, reduced to small dimensions, -and may be represented only by a narrow slit. In some cases, however, -where the cotyledons are unusually thick and heavy, even a large central -cavity will not give floating power to the seed. There is an indication -in my experiments that seeds from inland plants that have matured their -pods in the forests sink in a much greater proportion than seeds of -coast plants, or of those growing on the banks of estuaries. This we -might expect, since in the shade of the forests the drying process that -accompanies the setting and final maturation of the seed would be less -complete and the intercotyledonary cavity smaller than with the seeds -matured in more exposed situations. This is a point, however, that -requires further investigation. - -It will thus be seen that in respect of buoyancy the seeds of Entada -scandens are to be referred to the mechanical or non-adaptive group of -buoyant seeds, described in Chapter XII, which comprises several other -Leguminous strand-plants, including Cæsalpinia bonducella. I especially -studied the various stages in the development of the buoyancy of seeds -in this mechanical group in the case of the species of Cæsalpinia just -named, and the description of the process as given under that plant will -apply to all. - - - _Summary relating to Entada scandens_ - -(1) This plant, which has been distributed by the currents over the -tropics of the globe, has its station in the mangrove swamp, on the -beach, by the estuary, and in the inland forest. - -(2) It is regarded as an American plant that has reached the shores of -the Indian Ocean by crossing the Pacific, and the coast of West Africa -by crossing the Atlantic. - -(3) Its occurrence on both coasts of America is attributed to its having -a focus of dispersal in the forests of Central America, from which its -seeds have been transported by the rivers to the shores of the Atlantic -and Pacific Oceans. - -(4) Its irregular distribution in the Pacific islands, to wit, its -absence from Hawaii and its rarity in the Tahitian region, is not to be -easily explained, but it is more than likely that it will be -subsequently recorded from other localities in Eastern Polynesia. - -(5) Although the seeds offer a striking example of dispersal by -currents, since they are to be found stranded on beaches over much of -the globe, from within the Arctic Circle to the Coral Sea, in few plants -could the character of the buoyancy and the structure connected with it -have so little claim to be considered as adaptive in their nature. At -least 50 per cent. of the seeds sink in sea-water, and the cause of the -buoyancy of the other seeds is only to be connected with the large size -of a cavity produced by the shrinking of the embryo within the seed -tests during maturation. - - - CÆSALPINIA - -This genus is represented in the tropics of both the Old and the New -World by some eighty species of trees, shrubs, and climbers, some of -which are noted for their dye-woods, and others for the beauty of their -flowers. In the Pacific islands the botanist is only concerned with -three widely distributed species, all more or less littoral in their -station, and in great part dispersed by the currents, namely, Cæsalpinia -nuga (Ait.), C. bonducella (Flem.), and C. bonduc (Roxb.). - -With Cæsalpinia nuga we have little to do, since, although widely -distributed in tropical Asia and the Malayan region, and reaching to -both New Guinea and North Australia, it has not apparently penetrated -into the Pacific further east than the Solomon and New Hebrides groups. -I found it growing on the coasts of the larger islands of the Solomon -group, but no observations were made on its mode of dispersal. However, -as its seeds were identified at Kew (_Bot. Chall. Exped._ iv, 311) -amongst my collections of stranded drift from those islands, it would -appear to be to some degree dispersed by the currents, though since it -does not extend far into the Pacific, its capacity for dispersal by this -agency would seem to be limited. Schimper includes it among the -strand-plants of the Indo-Malayan region. - -It is with the other two species, Cæsalpinia bonducella and C. bonduc -that we are especially interested. Their extremely hard, marble-like -seeds at once attract attention, and when pale in colour they look not -unlike quartz pebbles as they lie stranded on a beach. The prickly pods -and the recurved prickles of the leaf-branches often make these plants -provokingly evident to a stranger. Though usually to be characterised -when growing on a beach as straggling shrubs, they will often climb -trees when opportunities occur, and they then display themselves as -stout-stemmed climbers. I have seen one or other of them in the mangrove -swamps of Fiji ascending the Bruguiera trees to a height of 30 feet and -more, the stem quite bare below, but leafing and flowering in the -tree-branches above. - -From the standpoint of dispersal there are few more interesting plants -in the Pacific islands; but their discussion raises several difficult -questions, and it will be, therefore, requisite to treat them somewhat -in detail. With regard first to the diagnostic characters between the -species, it may be observed that, as a rule, they are sufficiently -evident, such, for instance, as the number, size, and form of the -leaflets, the presence or absence of foliaceous stipules, and the colour -of the seeds, though, as shown below, the seed-colour in the case of -Fijian plants does not always present a constant distinction. Yet as I -found in Fiji the difference between the two species is not in all cases -well pronounced, and intermediate forms occur, about which it is -sometimes difficult to decide to which of the two species they should be -assigned. - -Mr. Hemsley remarks (_Bot. Chall. Exped._ iii, 114, 145, 300) that the -two species have been often confused. I venture to think that this has -been in some cases due to the occurrence of these intermediate forms. -One has only to look at the different “distributions” given by botanists -for C. bonduc, as indicated below, in order to suspect that the cause of -confusion has been at times with the plants themselves. When in Fiji I -paid a good deal of attention to this subject, and the results of the -comparison of the foliage and seeds of the plants obtained from fourteen -different localities in Vanua Levu are given below. - -It will be seen in this table that I distinguish in Fiji three littoral -forms and one inland or mountain variety, which may perhaps be a -distinct species. Those of the strand include Cæsalpinia bonducella, C. -bonduc, and an intermediate form. C. bonduc is typically distinguished -by its large leaflets, by the absence of foliaceous stipules, and by its -pale yellow seeds; whilst C. bonducella is similarly characterised by -its small leaflets, its foliaceous stipules, and its lead-coloured or -darkish grey seeds. But in the first species the colour of the seeds may -often be yellow mixed with pale-grey, or almost white; whilst in the -second species the seeds may be stained with brownish-yellow patches. - -It seemed to me when examining fresh specimens in Hawaii and Fiji that -the ultimate colour of the seed is a good deal determined by the degree -of alteration of the original olive-green colour of the immature seed. -All gradations may be noticed from the olive-green of immaturity to the -yellow, pale grey, and dirty white hues of the mature seeds of -Cæsalpinia bonduc and to the lead or slate-colour of those of C. -bonducella. It almost appeared as if the changes might be compared to -the bleaching which a dark volcanic rock undergoes in the weathering -process through the hydration and removal of the iron oxides. - - - CÆSALPINIA IN FIJI, TAHITI, AND HAWAII. - - A = Locality. - B = Species. - C = Foliaceous stipules. - D = Pairs of pinnæ. - E = Leaflets—Pairs. - F - Leaflets—Length in inches. - G = Leaflets—Form. - H = Seeds—Size in tenths of inch. - I = Seeds—Colour - - +--------+---------------+--------+------+------+-------------+-------------------+-------------+----------------------+ - | A | B | C | D | E | F | G | 8 | 9 | - +--------+---------------+--------+------+------+-------------+-------------------+-------------+----------------------+ - | {| Bonducella |Present | 8-9 | 9-11 |1-1/4 - 1-1/2| Oblong, obtuse |6-1/2 - 7-1/2| Usually lead- | - | {| | | | | | mucronate: base | | colour with at | - | {| | | | | | rounded and in— | | times brownish- | - | {| | | | | | equilateral | | yellow patches. | - | {| | | | | | | | | - |Coast, {| Bonduc |Absent | 5-6 | 4-6 | 2-1/2 - 5 | Oblong, acuminate | 5-1/2 - 6 | Pale yellow | - |Fiji {| | | | | | mucronate, base | | | - | {| | | | | | rounded or |6-1/2 - 7-1/2| Pale grey, sometimes | - | {| | | | | | subcordate | | mixed with | - | {| | | | | | | | yellow. | - | {| | | | | | | | | - | {| Intermediate |Present | 7-8 | 7-9 | 2-3 | Oblong, obtuse | 6-7 | Lead-colour or | - | | | | | | | mucronate, rounded| | pale grey with | - | | | | | | | at base; upper | | brownish-yellow | - | | | | | | | leaflets may be | | patches | - | | | | | | | elliptical | | | - | | | | | | | | | | - |Inland, | Mountain |Present | 5-6 | 9-10 |1-1/2 - 2-3/4| Lanceolate with | 6 | Yellowish or pale | - |Fiji | species | | | | | long tapering | | grey or mixed. | - | | | | | | | aristate apex | | | - | | | | | | | and rounded | | | - | | | | | | | base | | | - | | | | | | | | | | - |Coast, | Bonducella |Present | | | 1/2 - 1-4/5| Oblong | | | - |Tahiti | | | | | | | | | - | | | | | | | | | | - |Inland, | Bonduc |Absent | | 5-6 | | Oblong | | | - |Tahiti | | | | | | | | | - | | | | | | | | | | - |Inland, | Bonducella | | 4-6 | 6-8 | 1-4/5 - 2 | Oblong, obtuse, | 6-7 | Lead-colour. | - |Hawaii | | | | | | not cordate at | | | - | | | | | | | base | | | - +--------+ --------------+--------+------+------+-------------+-------------------+-------------+----------------------+ - -_Note._—The characters of the Fijian plants are from my own - observations. Drake del Castillo is quoted for Tahiti, and Hillebrand - for Hawaii. Reinecke observes that the pods of C. bonducella in the - inland forests have no prickles. - -In Fiji all three coast forms may be found on the same beach, or they -may exist apart. The large-leaved species (C. bonduc) appears to be much -the most frequent in Vanua Levu; and the intermediate form is common -enough to disturb the serenity of the observer’s mind when he is anxious -to diagnose rather than to collect cumbersome specimens. The mountain -form, which came under my notice as a climber in the forest at an -elevation of 1,700 feet on the slopes of Koro-mbasanga in Vanua Levu, -acquires from the lanceolate shape of its leaflets quite a character of -its own, though it comes nearest to Cæsalpinia bonducella. Mountain -forms also occur, as indicated in a later page, in the forests of Samoa -and in Tahiti; but in the first-named group they are referred by -Reinecke to C. bonducella, and in Tahiti by Drake del Castillo to C. -bonduc. In the Samoan forests the inland plants possess pods deprived of -the prickles that are so characteristic of the beach plants. Before one -can pronounce definitely on the relation between the coast and inland -forms in any of the groups, a thorough investigation of the connections -between the two shore-species is needed. I am inclined to think that -they will prove to belong to a single dimorphic (or perhaps polymorphic) -species. - -_The distribution of Cæsalpinia bonducella and C. bonduc._—Botanists -agree in giving C. bonducella a distribution around the tropics of the -globe; but they are not at all unanimous with respect to the other -species. According to Mr. Hemsley this species is by no means so -universally dispersed as C. bonducella. It is unknown from Africa and -Australia; but it is generally characteristic of tropical Asia and the -Malay Archipelago. The same authority alludes to specimens in the Kew -Herbarium from Florida and the West Indies (_Bot. Chall._ iv, 300). -Drake del Castillo gives both species a range through the tropics, -whilst Schimper seems in doubt about the occurrence of C. bonduc in the -New World, and Mr. Burkill makes no allusion to its American habitat in -his paper on the Tongan flora. The cause of this confusion is doubtless -to be mainly attributed to the variation in characters of the plants, -and to the occurrence of intermediate forms. - -We should be scarcely consistent if we assumed that of two kindred -shore-species dispersed by the currents one had its home in America and -the other in the Old World. The same home must belong to both. According -to the principle laid down in Chapter VIII, and referred to under Entada -scandens, it is held that a strand-plant, with its home in Asia, on -account of the arrangement of the currents could never reach the -American continent, and that American shore-plants are for the most part -native-born except those hailing from the African West Coast, which, -however, lies within the American province of tropical strand-plants. -From this standpoint Cæsalpinia bonducella would be regarded as now -having its home in the New World, and since it is found on both the -Pacific and Atlantic coasts of that continent (as well as on both coasts -of Africa), it is assumed, as with Entada scandens, that it has reached -the African West Coast by crossing the Atlantic, and the African East -Coast by way of the Pacific and Indian Oceans. The genus, I may remark, -is distributed over the tropics of the eastern and western hemispheres. - -As regards the general distribution of the two species in the Pacific -islands, it would appear from the writings of Seemann, Hillebrand, -Hemsley, Drake del Castillo, Reinecke, Cheeseman, and Burkill that with -the exception of Hawaii and Samoa, where Cæsalpinia bonducella alone -occurs, and of Rarotonga where C. bonduc alone is found, they are -generally associated in the larger groups, as in Fiji, Tonga, Tahiti, -and the Marquesas. - -_The station of Cæsalpinia bonducella and C. bonduc._—Both the species -are to be regarded as littoral plants likely to stray inland. The -first-named is described in the _Botany of the “Challenger” Expedition_ -as essentially a sea-side plant, though flourishing equally well inland, -and in India extending to the Himalayas as far as Kumaon, and up to -elevations of 2,500 feet. Schimper speaks of both species as -characteristic of the Indo-Malayan strand-flora, and he quotes Kurz when -referring to C. bonduc as a constituent of the beach-jungle of Pegu. - -In the Pacific islands they are typically littoral in their station; but -they may extend inland, and in one or two groups they are only known in -their inland station. Dr. Seemann speaks of both species only in -connection with the beaches in Fiji, and alludes to Cæsalpinia -bonducella (p. 72) as sometimes climbing over the mangroves. In Vanua -Levu both came under my notice on the beaches, and in their immediate -vicinity, usually as straggling bushes, whilst at times they were to be -observed climbing the mangroves at the borders of the adjacent swamp. In -this island of the Fijis they do not, as a rule, stray far from the -beach, and strange to say are not to be ranked amongst those seashore -plants that frequent the “talasinga” regions or inland plains. Judging -from the mountain form found in the forests of Koro-mbasanga, if they -extend inland in Fiji they prefer the forests and become differentiated -in character. In Tahiti, as we learn from Nadeaud and Drake del -Castillo, C. bonducella occurs on the beach and extends inland to the -mouths of the valleys; whilst C. bonduc is only recorded from the -mountains at elevations of 600 to 700 metres (2,000 to 2,300 feet). -Jouan is quoted by Mr. Hemsley as remarking that C. bonduc is as common -in the Marquesas as brambles are in Europe (_Bot. Chall. Exped._ iii, -145). In Rarotonga, according to Cheeseman, C. bonduc is restricted to -the interior. In Samoa, as we are informed by Reinecke, C. bonducella is -frequent both in the coast districts and in the mountain-forests. In the -Samoan mountains the pods lose their prickles, and from this -circumstance, as well as from the extremely widespread distribution of -the species over the islands, the German botanist concludes that the -plant has been for ages established in the group. - -In Hawaii, Cæsalpinia bonducella, which alone occurs, rarely figures as -a beach plant; but it is found, as Hillebrand observes, in the lower -plains of all the islands. In the large island of Hawaii I found it not -on the scanty beaches of the coast, but on the partly vegetated surface -of the old lava-flows at distances varying usually between a hundred -yards and a mile from the sea, but extending at times a few miles -inland, and in one locality reaching an elevation of 2,000 feet above -the sea. It was mostly observed by me on the dry side of the island, -where, associated with Erythrina monosperma, the Cactus, and the -Castor-Oil plant, it thrives in very arid localities, where the rainfall -is only a few inches in the year. Farther inland, where the old -lava-surfaces were more vegetated, it was associated with such shrubs as -Osteomeles anthyllidifolia and Cyathodes tameiameiæ. Dr. Hillebrand, -writing of a generation and more ago, says that in his time the plant -was less common than formerly. - -_The Methods of Dispersal of Cæsalpinia bonducella and C. bonduc._—We -come now to the modes of dispersal of these plants; and in so doing we -have to choose between the agencies of birds and of currents. The seeds -of C. bonducella are on the average 7/10 of an inch (18 mm.) in -diameter, whilst those of C. bonduc are rather smaller (6/10 of an inch -or 15 mm.). As far as their size and character go, it would seem -scarcely likely that birds could transport these seeds across an ocean; -but our knowledge of the agency of birds is of a very imperfect nature. -Yet their occasional dispersal by birds is not improbable. When I was in -the Keeling Islands the residents informed me that the seeds of C. -bonducella are sometimes found in the stomachs of sea-birds, such as -frigate-birds and boobies. (See Note 59.) - -However, it has long been known that the seeds of one or both of these -species are carried great distances by the currents; but it is to be -gathered that the older botanists, in alluding to this fact, more -usually referred under the synonym of Guilandina bonduc to Cæsalpinia -bonducella. De Candolle, loth to attach much importance to the effective -transport of seeds by currents, was compelled to admit this species in -his scanty list of current-dispersed plants (see Note 33). For more than -two centuries it has been known that the seeds of C. bonducella are -carried in the Gulf Stream drift to the coast of Europe from the -American side of the Atlantic; and ever since they were recorded by -Sloane in 1696 as stranded in a fresh condition on the beaches of the -Orkney Islands, they have been found washed up on other localities, as -on the coasts of Ireland and of Scandinavia and on the shores of the -islands of the Western Atlantic. According to Robert Brown, a plant was -raised from a seed cast up on the west coast of Ireland; and with -respect to Scandinavia, Dr. Sernander informs us that the seeds of -Cæsalpinia bonducella, like those of Entada scandens and of Mucuna -urens, are of frequent occurrence amongst the “Gulf Stream products” -stranded on the Norwegian coasts. The seeds of this species are commonly -washed ashore at St. Helena, and there are specimens in the Kew Museum -that were stranded on Tristan da Cunha. (Those interested in the subject -will find it discussed by Mr. Hemsley in the _Botany of the “Challenger” -Expedition_, and also by Dr. Sernander in his recent work on -Scandinavia.) - -The seeds of Cæsalpinia bonducella have been also found stranded on -beaches in other parts of the world. Thus Prof. Schimper found them in -the beach-drift of the south coast of Java. Prof. Penzig noticed them -amongst the stranded seeds of the Krakatoa beaches; but it does not -appear that the plant had established itself up to the date of his visit -in 1897, or fourteen years after the great eruption. They have been -picked up on the other side of the Indian Ocean on the east shores of -Africa (_Bot. Chall. Exped._ iv, 300). They came frequently under my -notice stranded on the beaches of Keeling Atoll in the same ocean; and -seedlings sprouting from the seeds were sometimes to be seen growing -amongst the drift just above the high-tide level. The seeds of both C. -bonducella and C. bonduc have been found also on the shores of Jamaica. -Those of both species are not uncommon amongst the stranded drift of the -Fijian beaches; but notwithstanding a careful search I found only a -solitary seed of C. bonducella in the Hawaiian beach-drift, a -circumstance explained below as arising from the usual non-buoyancy of -Hawaiian seeds. - -That the seeds of Cæsalpinia bonducella stranded on the coasts of an -oceanic island are able to germinate and reproduce the plant is, of -course, established by the distribution of the species; and we have just -observed that the process was noticed by the author on Keeling Atoll -where the plant has found a home. It is to be noted that the plant -collected by Darwin in this atoll was identified by Prof. Henslow as C. -bonduc; but the plant observed by me was more like C. bonducella, and -the stranded seeds collected by me were referred at Kew to this species. -Some curious considerations arise from the fact that although, just as -in the Keeling Islands, the plants of C. bonducella have evidently -established themselves from drift seed in one locality in the Bermudas, -they do not seem to have done so either on the shores of Krakatoa, or of -St. Helena, where, although they are frequently washed ashore, Mr. -Melliss never met with an instance of germination (see _Bot. Chall. -Exped._ iv, 300, and Penzig). This is doubtless in part the result of -the destructive efforts of the crabs, which, as I have shown in my paper -on Keeling Atoll, nibble off the shoots of many germinating seeds in -beach drift. - -The readiness or non-readiness of seeds to germinate on a beach, and the -nature of the conditions essential for the process, are matters that are -directly concerned with their effective dispersal by currents. On -account of the stony character of the seeds of these two species, it -might be expected that germination would only take place under -exceptional conditions. It should, however, be observed that the fine -transverse striæ on their outer surface represent original fissures or -cracks in the epidermis of the soft immature seed; and as such may be -regarded as lines of weakness in the seed-tests. If a pod is opened -before the seeds are mature, we find the seeds about twice the size of -maturity, and so soft that they can be indented by the nail. The -transverse striæ that mark the mature seed are displayed as indistinct -cracks in the epidermis; and if the immature seed is exposed to the sun, -in a few hours these cracks gape widely, and the seed has the grooved -appearance of a top. If a pod opens prematurely on a plant, as sometimes -happens, the immature seeds will be noticed with the epidermis scaling -off. It is evident that the “setting” or the induration of the -seed-coats and the final great contraction of the seed take place in the -pod before dehiscence. From these remarks it would seem probable that -seeds lying exposed to the fierce rays of the sun on a tropical beach -would be liable to develop cracks along the old fissures, and that such -cracks by permitting the entrance of moisture would favour germination. - -My experiments show that high temperature under moist conditions will -not of itself induce germination or in any way affect the seed. Thus in -two sets of experiments, in 1890 and 1902, I failed to induce the -germination of seeds which, after floating a year in sea-water, were -kept in moist soil at a high temperature. In one case a temperature -varying from 80° to 110° F. was sustained for several weeks, and in the -other experiment a temperature of 80° to 90° was kept up for five -months. When, however, an incision was made into the epidermis, or the -seed-coats were partially penetrated with a file, the seeds swelled up -in a day or two, and in a few days began to germinate. - -The rapid transformation of the stone-like seed into a softened, -swollen, germinating mass ranks amongst the numerous little wonders of -the plant world. The seed, in fact, assumes again the appearance of -immaturity, and in so doing it suggests to us that the rest-stage -exemplified in the hard, pebble-like seed is but an adaptation to -general climatic conditions, and that in a region of great heat and -humidity, where there are no seasons, and where the sun’s rays are for -ever screened off by mist and cloud, it could be dispensed with -altogether. One of my Hawaiian dreams was to establish vivipary in -Cæsalpinia bonducella by subjecting the maturing pod on the plant to -very warm and humid conditions, my expectation being that the soft, -swollen seed would at once proceed to germinate in the pod, and that the -final process of setting, as indicated by the induration and contraction -of the coats, or in other words the rest-stage, would be done away with. -The dream, however, bore some fruits in enlarging my standpoint in the -matter of vivipary, and I have referred to the subject in Chapter XXXI. - -The seed-shell, about 1·5 mm. in thickness, consists of three coats: the -outer skin very tough and waterproof; the inner skin seemingly -permeable; and the intermediate layer of hard prismatic tissue, the -“prismenschicht” of Schimper (p. 164). This middle layer absorbs water -rapidly and in large quantity, so that if a fragment of the shell is -placed in water it will be found after a day’s soaking to be three times -as thick as it was in the dry state. If one files a seed, or makes a -small incision, so as to expose the middle layer without piercing the -inner coat, and then places it in water, it will be noticed that the -middle layer at once begins to absorb water; and within a couple of days -the whole seed will swell and attain the size it possessed in the -so-called immature condition. During the process the outer skin -stretches, usually without rupturing; and all three coats, previously so -hard that a heavy blow with a hammer is required to break the seed, -become in a day or two soft enough to be easily cut with a knife. The -seeds thus treated swell in two days to three times their original size -and increase their weight fourfold. Water finds its way to the nucleus -or embryo partly through the dilated inner opening of the micropylar -passage and partly through the inner skin. The nucleus then swells up -into a fleshy mass, filling the seed-cavity, and in two or three days -more germination begins. - -I pass now to the discussion of the buoyancy of the seeds. Considering -that both species occur in oceanic islands, and that the currents are -active agents in transporting the seeds, their behaviour under -experiment appears at first sight to be full of anomalies. Thus, it was -ascertained at Kew (_Bot. Chall. Exped._ iv., 301), both with -comparatively fresh and with older seeds, that those of Cæsalpinia -bonducella floated in salt water, whilst those of C. bonduc sank; but in -the record given of the experiment no mention is made of the original -station of the parent plants; and it will be shown later on that the -station of the plant, whether at the coast or inland, has an important -determining influence on the buoyancy. - -In Fiji I found that almost without exception the seeds of littoral -plants of Cæsalpinia bonducella floated both in sea-water and in fresh -water. On the other hand, in Hawaii the seeds of this species, obtained -from three typical localities removed inland from the beach, sank -without exception, even after drying for several months; and the only -buoyant seed noted in these islands was a solitary seed collected from -the beach drift. In Hawaii, however, as before remarked, the species is -not strictly a littoral plant, occurring as it does in the lower levels, -but not necessarily in the vicinity of the coast. In the case of seeds -of littoral plants of C. bonduc in Fiji, I found that sometimes all -floated in sea-water and sometimes only a portion of them, whilst their -specific weight was on the whole rather greater than that of the seeds -of the other species. Thus, in one experiment half the seeds floated in -sea-water and a quarter in fresh water, whilst with seeds from another -locality 90 per cent. of the seeds floated in sea-water and 80 per cent. -in fresh-water; and in a third set of seeds all floated in both waters. - -The above experiments on Fijian seeds all relate to littoral plants. In -the instance, however, of the inland species from the mountains of Vanua -Levu, all the seeds sank in sea-water, even after being kept for five -years. If we follow the indications of these several experiments we -shall find that Cæsalpinia presents another illustration of the general -principle established in Chapter II that the seeds of inland plants sink -and those of coast plants float. - -My data, therefore, show that with the seeds of Cæsalpinia buoyancy goes -with station and not necessarily with species. It is probable, -therefore, that with the two widespread species, C. bonducella and C. -bonduc, varying results will be obtained with seeds from different -localities, whether insular or continental, according to the original -station. The typically buoyant seeds of the former species may, as we -have seen in Hawaii, lose their floating powers when they grow inland; -and the seeds of an inland species from the mountains of Fiji sink at -once. It is essential in interpreting the results of experiments on the -seeds of these plants to be acquainted with the stations; and in this -respect those of the Tahitian plants may be regarded as probable test -cases. We have seen that in Tahiti, C. bonduc is an inland plant, and C. -bonducella usually a beach plant; and I have no doubt that experiments -in that island on the seeds of these two species from the particular -stations just referred to will give results in agreement with the -principle here laid down. - -With reference to the duration of the floating powers of these seeds it -may be observed that a seed of Cæsalpinia bonducella, originally found -stranded on the beaches of Keeling Atoll, floated after a year in -sea-water as buoyantly as at the commencement of the experiment. Seeds -of Fijian littoral plants of both C. bonducella and C. bonduc floated in -my experiments after two and a half years’ immersion in sea-water, -showing no change whatever. Some of the seeds removed at the end of the -first year were filed and placed in soil, when they germinated -healthily. In Chapter IX it is pointed out that some buoyant seeds of -other Leguminous plants, such as Mucuna urens, would be apt to germinate -abortively and to sink in crossing the more heated areas of tropical -seas. The seeds of Cæsalpinia, judging from my experiments and -observations noted on page 84, seem to be quite proof against such -risks. This was well brought out in an experiment where seeds of the two -species of Cæsalpinia were kept afloat for two and a half years in a -vessel of sea-water together with seeds of Mucuna and Strongylodon. None -of the Cæsalpinia seeds attempted to germinate in the sea-water; but -with the other genera some of the seeds began to germinate, and sank in -the course of the first warm season, when the water-temperature ranged -from 75 to 90° Fahr. - -The seeds develop their buoyancy during the great contraction that, as -before described, marks the final setting of the seed-coats and the -ultimate maturation, as it is termed, of the seed. During this shrinking -process the kernel also shrinks within the seed-tests, and cavities are -thus produced within the seed-shell, on the relative size of which -depends the buoyancy of the seed, neither the seed-shell nor the kernel -possessing independent floating-power. These cavities, as illustrated in -the figures given in Chapter XII, are of two kinds. That usually -produced, being the one that mainly determines the buoyancy, is a large -central hollow caused by the arching outwards of the cotyledons during -the shrinking process, such as is found also in the seeds of Entada -scandens, Mucuna urens, and some other Leguminous littoral plants. With -such seeds the kernel never rattles when the seed is shaken, since the -cotyledons lie in close contact with the seed-shell. The other kind of -cavity is produced between the seed-shell and the kernel by the general -or partial shrinking of the kernel away from the shell, the cotyledons -remaining in apposition, as shown in the figures. When the shrinking -away from the shell is general, the kernel lies loose within the shell, -and the seed rattles when shaken. When the shrinking is partial the -cavity is on one side of the seed and the kernel is fixed. - -Professor Schimper (p. 164) remarks that the buoyant seeds of Cæsalpinia -bonducella all rattle when shaken, and that it is to the incomplete -filling of the seed-cavity, thus indicated by the loose kernels, that -the buoyancy of the seed is due. The rattling of the kernel was, -however, quite exceptional in the seeds handled by me, even in the case -of originally buoyant seeds kept for five years. Seeds with loose -kernels were, in fact, more frequent with non-buoyant seeds than with -those that floated. Thus in Fiji I found that whilst with the buoyant -seeds 17 to 20 per cent. had loose kernels, with non-buoyant seeds the -proportion was as much as 60 per cent. - -The normal cause of buoyancy is, therefore, a large intercotyledonary -cavity with the cotyledons lying in close contact with the seed-shell; -but the two kinds of cavity may sometimes be combined. Out of a number -of buoyant seeds of Cæsalpinia bonducella examined by me, 80 per cent. -owed their buoyancy solely to a large central cavity (4 to 5 mm. -across). In 6 per cent. it was due solely to the shrinking of the kernel -away from the seed-shell; whilst in 14 per cent. it was to be attributed -partly to a reduced central cavity (2 to 3 mm. wide), and partly to a -space outside the kernel. The only difference noted in the structure of -the buoyant seeds of C. bonduc was that the two kinds of cavities were -more often combined. - -The reason of the absence of floating power was clearly indicated in the -non-buoyant Hawaiian seeds, where there was no central cavity, or it was -represented by a narrow slit. The solitary buoyant seed found in the -beach drift had a typical large central cavity. With the non-buoyant -seeds of the inland species of the mountains of Vanua Levu it was -ascertained that two-thirds had loose kernels with the cotyledons -closely appressed. In the others there was a lateral cavity outside the -kernel, the central cavity being only represented by a slit, a hair’s -width in breadth. In the non-buoyant seeds of C. bonduc, the central -cavity was only 2 to 3 mm. wide, and the lateral cavities were small. - -Respecting the influence of “station” in producing the differences in -buoyancy, it cannot be said to be connected with the maturation of the -seeds of inland plants under more humid conditions than those which -prevail at the coast. In Fiji some of the littoral plants with buoyant -seeds grow on the mangrove-trees in the shade and humidity of the -swamps; whilst in Hawaii the inland plants of Cæsalpinia bonducella with -their non-buoyant seeds thrive in exposed arid situations in districts -of little rainfall, such as on scantily vegetated lava-flows. With -non-buoyant seeds, where there is little or no cavity, the cotyledons -are always thicker and moister than in the case of the seeds that float. -Though associated with differences in station, as implied in the terms -“coast” and “inland,” the cause of the difference in buoyancy is not -connected with different degrees of humidity, but with some other cause -or causes acting on the spot which, while they favour the drying of the -kernel in coast plants before the seed-coats finally set, impede it in -the inland plants. That the seed does not subsequently acquire floating -power, even after years of drying, was shown in several of my -experiments. - -The light, unopened prickly pods of both species float buoyantly, even -when the inclosed seeds have no floating power. In an experiment on -Cæsalpinia bonduc in Fiji the pods remained afloat after a month in -sea-water. With those of C. bonducella in Hawaii I found that they -floated for several weeks, and in one case a pod was afloat after three -months. The pods dehisce on the plant; but they sometimes do not open -sufficiently to allow the seeds to fall out. The pods, however, have to -be torn off from the plant, and are not likely to occur in the drift. -Indeed, they never came under my notice in any locality in the drift, -and as an effective aid to dispersal they must be disregarded. The -buoyancy of the seeds and their well established distribution by -currents render unnecessary an appeal to the floating pod. - -The following is a summary of the foregoing remarks on Cæsalpinia -bonducella and C. bonduc. - -(1) The two species in Fiji are not always sharply distinguished, since -intermediate forms occur, and here probably lies the explanation of the -confusion that has sometimes occurred in diagnosing the species. - -(2) Both are typical littoral plants, distributed over most of the -tropical zone, and occurring in company in most of the Pacific -archipelagoes; but they at times extend far inland. - -(3) Though it is not unlikely that sea-birds may have aided in their -dispersal, the oceanic currents have been the great agencies in their -dispersal, as is indicated by the frequent transport of seeds in the -Gulf Stream drift across the Atlantic, and by their occurrence in beach -drift in various parts of the world. - -(4) Having regard to the present arrangement of the currents and the -distribution of the two species, reasons are given for the belief that -their original birthplace was in the interior of the American continent. - -(5) Notwithstanding the stony hardness of the seeds, when a notch is -made in the outer skin a seed rapidly takes up water, and in a few days -it becomes a soft and much swollen germinating mass. The author is -inclined to think that this was the original condition of the seed, and -that the rest-stage is an adaptation to secular differentiation of -climate in later epochs. - -(6) Unlike the seeds of other Leguminous littoral plants, those of -Cæsalpinia are not likely to germinate abortively when floating in warm -tropical seas, a risk that restricts the distribution of several -littoral species. - -(7) As tested by experiment, the seeds of both species are often able to -float unharmed for years; but on the other hand seeds not infrequently -have no floating power. - -(8) Observation, however, shows that buoyancy goes with station, and -that the general rule here applies that the seeds of coast plants float -and those of inland plants sink. - -(9) The nature of the influence of “station” on the seed-buoyancy is -obscure; but it is evidently not connected with the usual differences -between coast and inland localities, such as those concerned with -exposure or shade, dryness of soil, relative humidity, and similar -contrasts. - -(10) The buoyancy of the seed is developed during the final shrinking -process associated with its maturation, a large cavity between the -cotyledons being usually produced. - - -------------- - -_Note._—Since most of the principal conclusions of this work are -involved in my especial study of the littoral species of Afzelia, -Cæsalpinia, and Entada, the reader is advised, if he wishes to form an -opinion of the author’s method of investigation, to read this chapter -carefully through. With most other shore-plants, though in not a few -cases studied with the same detail, the exigencies of space have often -limited me to the employment of the general results in the appropriate -chapters without entering into details. Should he desire to test any -view of his own relating to plant-dispersal, he could not do better than -begin with the materials here provided. - - - - - CHAPTER XVIII - - THE ENIGMAS OF THE LEGUMINOSÆ OF THE PACIFIC ISLANDS - -Leguminosæ predominate in tropical littoral floras.—The anomalies of - their distribution in the Pacific islands.—They conform to no one rule - of dispersal or of distribution.—Strangers to their stations.—The - American home of most of the Leguminous littoral plants.—Summary. - - -IT is my intention here to gather up some of the “ends” of the great -tangle presented by the Leguminosæ in the Pacific. When we look at the -indigenous phanerogamic floras of Fiji, Samoa, Tahiti, and Hawaii we -find that the Leguminosæ form 5 or 6 per cent. of the total in each of -the three first-named groups, and only about 2·5 per cent. in Hawaii. -The paucity of Leguminosæ in oceanic floras was long ago pointed out by -Sir Joseph Hooker, whose work forms the foundation of much of our -knowledge of insular plant-life. This is emphasised by Mr. Hemsley in -his volume on the _Botany of the “Challenger” Expedition_ (Introd. p. -25), where he makes the very significant remark that the Leguminosæ are -wanting in a large number of oceanic islands where there is no truly -littoral flora. The islands, however, here more especially referred to, -are those of the southern Atlantic and Indian oceans, such as St. -Helena, Tristan da Cunha, and Amsterdam. It is especially true of New -Zealand, where the Leguminosæ barely make 2 per cent. of the total. Of -the Polynesian islands, as he points out, it is not so correct; and, in -fact, the proportion found in the Fijian, Samoan, and Tahitian floras, -respectively, is much the same as that which characterises the British -flora, namely, 5 to 6 per cent. - -When we come to explain the paucity of the Leguminosæ in the Hawaiian -flora we bring to light the singular principle that _Leguminosæ are far -more characteristic of the littoral flora than of the inland flora of a -Pacific island._ About half of the Leguminosæ of Fiji and Tahiti are -coast plants; and about 30 per cent. of the littoral plants of the -islands of the tropical Pacific belong to this order. Since, therefore, -Hawaii possesses much fewer shore-plants (30) than does Tahiti (55) or -Fiji (80), the paucity of its Leguminous plants is readily accounted -for. - -We have next to notice a principle, which is, in fact, deducible from -the first, namely, that _buoyant seeds are much more characteristic of -the Pacific Leguminosæ than of any other order_. Three-fourths of the -species have buoyant seeds, and, in fact, about a third of the littoral -Polynesian plants with buoyant seeds or fruits belong to this order. - -It may, therefore, be inferred that _the Leguminosæ owe their presence -in the islands of the tropical Pacific mainly to the currents_. - -From Mr. Hemsley’s conclusion that the Leguminosæ are wanting in a large -number of islands where there is no truly littoral flora, the -presumptions arise that _when inland species exist that possess no -capacity for dispersal by currents they are to be regarded as -derivatives from the littoral flora, and that they owe their origin to a -strand-plant possessing buoyant seeds originally brought by the -currents_. It has been shown in the case of Afzelia bijuga and of -Cæsalpinia that when Leguminous shore-plants extend inland the seeds -often lose their buoyancy, and it is probable that divergence in other -characters may occur, leading, as in the mountains of Fiji, to the -development of a new species of Cæsalpinia. It is urged that by a -continuation of the same process the inland species, Erythrina -monosperma, has been developed in Tahiti and Hawaii, and the inland -species, Canavalia galeata and Sophora chrysophylla, have been produced -in the last-named group. All these species have non-buoyant seeds, and -in all three cases there is no littoral species in Hawaii, it being -assumed that the parent strand-plant has been driven inland from the -beach. _It is not necessary that the littoral species should be now -represented in the flora._ - -It is remarkable that _in almost all cases the cause of buoyancy is of -the non-adaptive or mechanical kind, due either to cavities formed by -the shrinking of the seed-nucleus during the setting of the seed or to -the light specific weight of the kernel_. There is but little to show -that the buoyancy of the seeds of Leguminosæ is anything but an -adventitious character of the seed, as far as its relation to dispersal -by currents is concerned. Although this capacity has been the great -factor in the wide distribution of the species, yet it is evident that -Nature here takes advantage of a quality that could never by its aid -become a specific distinction. The upshot of the selecting process would -be the dispersal by the currents of nearly empty seeds or seeds that -have lost their germinating capacity. - -The distribution of the Leguminosæ in the Pacific islands, and indeed of -tropical islands generally, is often full of inconsistencies. This is -the only order that sets at nought most of the principles established -for the other plants of the sea-coast, and that defies the application -of the laws of plant-dispersal now most in evidence. Take, for instance, -the inexplicable affinity of Acacia koa, the well-known Koa tree of the -Hawaiian forests, to Acacia heterophylla, a tree restricted to the -Mascarene islands of Mauritius and Bourbon. Mr. Bentham, who placed them -in the same group with three or four Australian species, even doubted -whether the difference between the Hawaiian and Mascarene species -amounted to specific rank. These two closely related Acacia trees of -far-separated islands of the Indian and Pacific Oceans represent -outliers of the great formation of phyllodineous Acacias that have their -home in Australia (_Introd. Chall. Bot._ p. 26). As far as I can gather -Acacia seeds have no known means of dispersal. Not even when the tree -has a littoral station, as in the case of Acacia laurifolia in Fiji, -have the seeds or pods any capacity worth speaking of for dispersal by -currents. We must appeal to the birds; but to what birds we may ask, -unless it be to the extinct Columbæ and their kin, or to the Megapodes. -Some of the other Hawaiian difficulties connected with the inland -Leguminosæ are repeated in the Mascarene Islands. Thus, Bourbon, like -Hawaii, has its inland species of Sophora of the section Edwardsia. - -In their irregular distribution the Leguminosæ of the Pacific islands -are often a source of perplexity to the student of plant-dispersal. -Take, for example, the inland Erythrina, E. monosperma, of Hawaii, -Tahiti, and perhaps New Caledonia. Then look at the singular -distribution of the Sophoras of the Edwardsia section in Chile and Peru, -Hawaii, New Zealand, Further India, and Bourbon. The botanist, again, -finds a climber like Strongylodon in the forests of Fiji, Tahiti, and -Hawaii, and he picks up the seeds on the beaches of those islands and -notices that they float unharmed for many months in the sea, yet when he -pays heed to the distribution of the genus he finds that it only -comprises four or five species, and that it occurs outside the Pacific -only in the Philippines, Ceylon, and Madagascar. The extraordinary -distribution of Entada scandens in the Pacific islands has been before -alluded to in these pages. Here we have a plant, the seeds of which are -known to be transported unharmed by currents all round the tropics. Yet -it is absent from Hawaii and from almost all of the islands of Eastern -Polynesia. In many cases an endeavour has been made in this work to -explain these difficulties. But the order in the Pacific teems with such -difficulties. We may ask with astonishment why it is that the genera, -and sometimes even the separate species, of the Leguminosæ seem so often -to follow in each case a principle of their own. - -Plants of this order in the Pacific conform to no one rule of dispersal -or distribution, whether we regard a species, a genus, or the whole -order. Take, for instance, the presence in Hawaii of Canavalia galeata, -a plant that, as we know it now, could not possibly have reached there -through the agency of the currents, and the absence from the same group -of Entada scandens that could have been readily transported there by the -currents from America. Or, if we take the whole order and look at the -structures connected with the buoyancy of the seeds, we find two types -of structure and the elements of a third. Then, again, whilst most -littoral plants with buoyant seeds retain the buoyancy of their seeds -when they extend inland, Leguminous shore-plants, like Afzelia bijuga -and Cæsalpinia bonducella, when they extend inland in Fiji and Hawaii, -lose in great part or entirely the floating power of their seeds. - -Furthermore, most strand-plants, being typically xerophilous in -character, when they extend inland shun the forests and prefer the dry -soil and sparsely vegetated surface of the open plain; but the -Leguminous genera and species (Mucuna, Afzelia, Entada, &c.) when they -leave the coast take to the forests, growing usually as stout lianes, -but sometimes as tall trees. Here again the Leguminosæ seem to follow a -principle of their own. As far as I know, this is the only order in the -Pacific possessing forest-trees which, as in the case of Afzelia bijuga -in Fiji, are equally at home in the woods of the interior and of the -coast. - -Indeed, judging from Professor Schimper’s observations, the littoral -Leguminosæ of the tropics often display a physiological constitution -that seems in some respects out of touch with their surroundings. They -may, as in Sophora tomentosa and in Canavalia, present the xerophytic -character of strand-plants, but frequently they are not halophilous or -“salt-loving,” like other plants associated with them on the same -shore-station. They are often shy of salt in their tissues, though able -to thrive in salt-rich localities. That capacity which strand-plants -usually possess of storing up chlorides in their tissues, and especially -in their leaves, without injury to themselves, is but slightly possessed -by such characteristic shore-plants as Canavalia, Pongamia glabra, and -Sophora tomentosa. This capacity, which, as Professor Schimper -indicates, goes to determine whether or not plants are capable of living -in salt-rich localities, has often no determining influence with the -Leguminosæ. (See Note 60.) - -Though the plants of this order form such a large element in the -strand-flora of the Pacific islands and of the tropics generally, they -seem in other respects, besides those just referred to, to act as if -they were strangers to the station. Look, for instance, at the readiness -of the floating beans of Mucuna, Strongylodon, &c., to germinate, as -shown in Chapter IX, in the tepid waters of the warmer areas of the -tropical oceans. This is a great deal more than a disturbing factor of -distribution. It is significant also of the plants being out of touch -with their dispersing agencies. - -One may notice in conclusion the fact brought out in Chapter VIII that -nearly all the littoral plants dispersed by the currents that are common -to the Old and the New Worlds belong to the Leguminosæ. This is held to -indicate that their home is in America, since that continent distributes -but does not receive tropical littoral plants dispersed by currents. - - - _Summary._ - -The Leguminosæ are far more characteristic of the littoral flora than of -the inland flora of the Pacific islands; and since the greater number of -them have buoyant seeds, it follows that this order mainly owes its -presence in this region to the currents. - -As it has been shown that in a large number of islands where there is no -littoral flora the Leguminosæ are wanting, the presumption arises that -when, as in Hawaii, inland species occur which at present have no -capacity for dispersal by currents, they have been derived from -strand-plants originally brought by the currents, even though such shore -species no longer belong to the flora. - -As far as its relation to dispersal by currents is concerned, the -buoyancy of the seeds of Leguminosæ is merely an adventitious character, -and the structure connected with it has no specific value. - -Plants of this order in the Pacific are a source of much perplexity and -conform to no one rule of dispersal, whether as regards their -disconnected distribution, their means of dispersal, the structural -cause of buoyancy, the loss of buoyancy of inland species, and in other -particulars. Even in their physiological constitution they are often at -variance with the bulk of littoral plants when they grow on the -sea-shore, since typical beach-plants of the order, though thriving in -salt-rich localities, are shy of salt in their tissues. - -It is probable that whilst the Pacific islands have derived most of -their littoral plants that are dispersed by currents from the tropics of -the Old World, they have received most of their strand Leguminosæ from -America. - - - - - CHAPTER XIX - - THE INLAND PLANTS OF THE PACIFIC ISLANDS - - PRELIMINARY COMPARISON OF THE PHYSICAL CONDITIONS OF HAWAII, FIJI, AND - TAHITI - -Introductory remarks.—The tranquil working of the winds and currents - contrasted with the revolutionary influence of the bird.—The Hawaiian, - Fijian, and Tahitian groups.—Their surface-areas and elevations.—Their - climates.—The mountain climate of Hawaii.—The rainfall of the three - groups.—Summary. - - - INTRODUCTORY REMARKS. - -I WILL carry my readers back to that moment when we began to investigate -together the composition of the floras of the islands of the tropical -Pacific from the standpoint of dispersal. It will be remembered that -after collecting all the fruits and seeds of a particular island we -placed them in sea-water, and that some nine-tenths of them went to the -bottom at once or in a few days. We found, speaking generally, that the -buoyant seeds and fruits belonged to coast plants, whilst those at the -bottom of the vessel proved to be obtained from inland plants. Since -that period we have been occupied in following up the clue supplied by -the floating seeds and fruits. In their company we have travelled far -beyond the Pacific islands. We have not only seen their fellows in other -parts of the tropics, both on the coral atoll and on the continental -coast, but we have met their representatives on the beaches of Europe -and of temperate South America. We have followed them in their ocean -traverses round most of the tropical zone, and on the way we have -naturally interested ourselves in the question of the currents. We have -weighed these seeds and fruits and have compared their specific weight -with that of sea-water. We have cut them up and carefully examined them, -and under their guidance we have explored the mangrove-swamps both of -Polynesia and of Ecuador, and have penetrated the mysterious _cul de -sac_ of vivipary. Having formed our opinion of them, we now bid the -subject farewell, and stand once more on the same Pacific beach where, -it seems so long ago, our investigations began. - -For the seed and fruits lying at the bottom of the sea-water we have to -appeal to other agencies than to that of the currents if we wish to -inquire into their means of arriving at this island. In imagination we -leave the reef-lined shores for the interior, and exchange the -exhilarating surroundings of a coral beach, where “the sky is always -blue and the wind is always true,” for the arid conditions of an inland -plain, or for the humid conditions of the forest, where the rain is -incessant and the cloud-cap and mist seemingly eternal. When we look at -the motley collection of fruits and seeds obtained in such localities, -we are at a loss to know where to take up the clue. After vainly -endeavouring to obtain some inspiration as to the manner of commencing -the inquiry, we do what all good naturalists in the Pacific islands do -from force of habit when they meet with difficulties of any kind—we sit -down and light our pipes. Then come a flood of old memories and old -trains of thought that came to us years before on some mountain-top or -in a shady gorge or on some river-bank, in regions Pacific and -non-Pacific, and by degrees our ideas shape themselves and we begin to -think the matter over in an orderly fashion. - -When the winds first brought the spores of ferns to this Pacific island, -the ocean currents brought the seeds and fruits of littoral plants, and -the birds transported the seeds and “stones” of various inland species. -All three agencies have been working side by side since the earliest -stage in its history. Yet it is only in the work of the wind and the -current that we find any indication of stability in the floral history -of the island. With the work of the bird it has been very different. -Since the first bird carried seeds to this locality all else has been -turmoil and change. Wave after wave of migrant plants has overrun the -interior of the island, and all have left their mark; but the great -distributing factor and disturbing agent has always been the bird. -Genera have been born and have disappeared, and in their place new -genera have arisen. Whole families even have participated in the -revolutions of the plant-world, and species have grown rankly in the -great confusion. Last of all came man with his cultivated plants and his -weeds, introducing new elements of change and discord into the island, -and often upsetting the floral economy altogether. The history of man’s -most troubled epoch would not be more full of catastrophes and great -events than the history of the plants of this Pacific island. Yet -through all these changes the winds and currents have been quietly -carrying on their work, bringing the same plants to beach and hillside -that they did before the age of unrest began. - -The monotonous character of an island flora that has been supplied by -the winds and currents can be readily imagined. For their variety the -floras of the Pacific islands are mainly indebted to the bird, the great -disturber of the peace of the plant world. We cannot attach too much -importance to the contrast in the results produced by these several -agencies in stocking a Pacific island with its plants. On the one hand -we have the tranquil working through the ages of the winds and currents. -On the other hand there has been the revolutionary influence of the -bird. One cannot doubt that many of the species of flowering plants now -growing on the beach and many of the ferns on the upper mountain-slopes -have witnessed changes within the forest-zone of the island, such as an -antediluvian might record if he had lived through the ages to the -present time. - -Now, what are these changes? How has the bird acted unconsciously such a -determining part? These are questions which I will endeavour in some way -to answer as one picks one’s path slowly through the various epochs in -the plant-history of these islands. We already are fairly well -acquainted with the beginnings of a flora either on a coral atoll or on -an ordinary tropical beach. What we have yet to learn is the subsequent -history of the flora. When Dr. Treub undertook, in 1886, his now -celebrated examination of the new flora of Krakatoa after the great -eruption, he commenced a series of observations which will no doubt be -prolonged into future centuries. Botanists a hundred and two hundred -years hence will complete a long chain of observations which will be -unique as a record of plant-colonisation; and science is deeply indebted -to Prof. Penzig for making, in 1897, the second examination of the new -flora. Though deprived of the valuable record that future generations -will possess for Krakatoa, we yet have at our disposal in the completed -process displayed by many a Pacific island a means of working backward -and in a sense completing the history. - -In order to attack this problem I have mainly confined myself to the -Fijian, Tahitian, and Hawaiian floras, taking the three archipelagoes -just named as the centres of the regions in which they occur. These -three groups lie near the three angles of the triangular area of the -Pacific over which the various archipelagoes are scattered. They are -thus geographically well placed for an inquiry into the subject of -plant-dispersal over this ocean, and each of their floras has been -investigated by botanists of various nationalities—American, Austrian, -British, French, German, and Italian. The Fijian area may be regarded as -including the adjacent Samoan and Tongan groups, though the individual -group or the whole area will always be in this work particularised. In -the same way Tahiti will be viewed as usually representative of the -larger islands of the surrounding groups of the Cook and Austral Islands -and of the Marquesas; and under the designation of the Tahitian area or -Tahitian region there will be generally included the Paumotu -archipelago. - - - COMPARISON OF THE AREAS AND ALTITUDES OF HAWAII, FIJI, AND TAHITI. - -Since differences in physical conditions have played an important part -in plant distribution in these groups—such, for instance, as in -determining the development of a mountain flora or in favouring the -relative abundance of particular types of plants—it is at first -essential to obtain a general idea, in the case of the larger islands of -each group, of their size and elevation, and of the more conspicuous -differences in their climates. - -Hawaii, the largest island of the Hawaiian archipelago, has an area of -4,210 square miles. All the other islands of the group are considerably -smaller—Maui, the second in size, having a surface of 760 square miles; -Oahu coming next; and after it Kauai, with an area of 590 square miles. -The area of Viti Levu, the largest island of the Fijis, is 4,112 square -miles, being thus closely similar to that of the island of Hawaii; Vanua -Levu, the second in size, is 2,433 square miles in extent; whilst the -other important islands of the group are much smaller, Taviuni, the -third in size, having an area of 218, and Kandavu an area of 125 square -miles. Tahiti, the largest and loftiest island of Eastern Polynesia, has -a surface of about 400 square miles; whilst most of the other elevated -islands of the groups around are considerably smaller. - -In respect of elevation above the sea, there is a great contrast between -the islands of these three regions. Taking the Hawaiian Group first, we -notice that the three principal mountains of the large island of Hawaii -rise in the cases of Mauna Kea and Mauna Loa to between 13,000 and -14,000 feet, and in that of Hualalai to rather over 8,000 feet. Situated -between these three mountains there is an extensive tableland or -plateau, known as the Cattle Plains, which is elevated between 4,000 and -6,000 feet, and has an area of not less than 200 square miles. At least -a third of the whole area of the island exceeds 4,000 feet in altitude. -In the eastern portion of Maui the huge mass of Haleakala rises to -rather over 10,000 feet; whilst Mount Eeka, in West Maui, rises in bulk -to some 6,000 feet. The island of Kauai, which is elevated between 5,000 -and 6,000 feet, possesses in its interior an elevated tableland 40 -square miles in extent and 4,000 feet in altitude. Oahu attains in Mount -Kaala a maximum elevation of 4,000 feet, but 3,000 feet is the limit of -the other peaks, and much of the island is low in elevation. - -On the other hand, in the two largest islands of Fiji, namely, Viti Levu -and Vanua Levu, we find in the first-named only two or three of the -highest mountain peaks rising to between 4,000 and 5,000 feet; whilst -the highest peak of Vanua Levu reaches only to about 3,500 feet. Amongst -the lesser islands, Taviuni just reaches the level of 4,000 feet, and -Kandavu, the next in height, about 2,750 feet. The area of the -land-surface in this group that is above a level of 4,000 feet is very -scanty, and for the botanist a negligible quantity, so that for purposes -of comparison the Fijian Islands, as far as elevation is concerned, -correspond to the lower levels of the Hawaiian Islands, that is, to the -areas below 4,000 feet. The same may be said of the Samoan Islands with -the exception of a limited area in the centre of Savaii, where a peak -rises to 5,400 feet above the sea. - -Coming to the Tahitian region, we find that Tahiti, the most elevated -island, attains an extreme height of about 7,300 feet; but from its -surface-configuration it is evident that not one-tenth of the area -exceeds 5,000 feet; yet since its total extent is about 400 square miles -there must be an elevated region of some 30 square miles in amount -comparable in some degree with the uplands of Hawaii. The Marquesas, -next in order in size and height, attain a maximum elevation of about -4,000 feet; whilst, amongst the Cook and Austral Groups, Rarotonga -reaches a height, according to Mr. Cheeseman, of 2,250 feet. Excepting -the limited elevated area of the uplands of Tahiti, there is nothing in -Eastern Polynesia corresponding to the higher levels of the Hawaiian -Islands over 4,000 feet. We formed the same conclusion for Fiji, and I -may add that it applies to the whole area of Fiji, Samoa, and Tonga, -since the solitary peak of Savaii in the second-named group, which -reaches 5,400 feet, alone represents a high-level area. The uplands of -Hawaii—that is to say, the elevated region between 4,000 or 5,000 feet -and 14,000 feet (strictly speaking 13,800 feet)—are therefore almost -unrepresented amongst the Oceanic groups of the South Pacific; and it is -only in the peak of Savaii and in the limited high levels of Tahiti that -we would expect to find their conditions reproduced. The great effect -that this contrast implies in determining differences between the floras -of the Hawaiian, Fijian, and Tahitian regions will become apparent as we -proceed in this discussion. - - - COMPARISON OF THE CLIMATES OF HAWAII, FIJI, AND TAHITI. - -Before comparing the climatic conditions in the three groups, it may -first be remarked that since they lie, roughly speaking, at not very -dissimilar distances north and south of the equator a great contrast is -not to be expected in so far as they agree in elevation. The mean -latitudes do not differ greatly, that of Hawaii being 20° to 21° N., and -those of Fiji and Tahiti both about 18° S. The climate of both groups is -tempered by the north-east trade in the one region and by the south-east -trade in the other. Still there is a difference in the temperature and -dryness of the air which noticeably distinguishes Hawaii from Fiji, and -to a less extent from Tahiti. The mean temperature of the Hawaiian -Islands would be 74° or 75°; whilst that of Tahiti is placed at 76° to -77°, and that of Fiji at 79°. But it is to be observed that to a person -residing in Fiji after a residence in Hawaii the climate is perceptibly -warmer, more humid, and more enervating. No doubt this is in part -connected with the greater dryness of the air in Hawaii, where the -average relative humidity at Honolulu is placed at 72 per cent., and it -must be much less on the Kona coast on the dry side of the largest -island. It is, however, probable that the Hawaiian climate was less dry -before the destruction of the forests, and that the contrast with the -Fijian climate was then less pronounced. - -The great distinguishing feature, however, of the Hawaiian Islands is to -be found in their mountain climate. This is not represented in Fiji, but -slightly in Samoa, and to a small extent in Tahiti; and I will now refer -more particularly to this important subject. - -In the uplands of the large island of Hawaii, on the tops of the lofty -mountains 10,000 to 14,000 feet above the sea, we have a mean -temperature only found far north. Snow lies often on these barren -summits in winter, more particularly on Mauna Kea, which thus derives -its native name of the White Mountain. The details of my meteorological -observations on Mauna Loa will be found in Note 61; and only some of the -general results will be referred to here. - -The mean temperature for the period of twenty-three days passed by me on -the summit of Mauna Loa (13,600 feet) between August 9th and 31st, 1897, -was 38·5° F. The mean temperature for a period of twenty days from -December 24th, 1840, to January 12th, 1841, during which Commodore -Wilkes and his party were making pendulum observations on the summit of -the same mountain, was approximately 33·5° (see Note 61). From these -results, which are tabulated below, it will be seen that the mean annual -temperature would be probably about 36°, which is scarcely comparable -with any continental climate, since only a difference of a few degrees -is indicated between the mean temperatures of August and of a similar -period in mid-winter. I may add that although it was in the summer month -of August, water froze inside my tent during twenty out of the -twenty-three nights passed on the top. We may, therefore, infer that the -temperature falls below the freezing point at night practically -throughout the year. It will be seen from the table that the mean annual -temperature for the summit of Mauna Loa, as here computed from the -observations of Commodore Wilkes and myself, comes very near to that -which might be estimated by employing Hann’s tables of variation in -temperature with altitude on tropical mountains (see Schimper’s -_Plant-Geography_, iv. 691). - - - WINTER AND SUMMER TEMPERATURES ON THE SUMMIT OF MAUNA LOA - (13,600 FEET), IN DEGREES FAHRENHEIT. - - +---------+---------------+--------------+-------+--------+--------+------------+ - |Observer.| Period. | Mean daily |Lowest.|Highest.|Mean for|Approximate | - | | | range. | | |period. |yearly mean.| - +---------+---------------+--------------+-------+--------+--------+------------+ - |Wilkes |Dec. 24, 1840- |17°-50°=33° | 13° | 55°? | 33·5° | } | - | |Jan. 12, 1841 | | | | | } | - | | | | | | | } 36° | - |Guppy |Aug. 9-31, 1897|23·2-53·8=30·6| 15 | 61·2 | 38·5 | } | - +---------+---------------+--------------+-------+--------+--------+------------+ - -Estimated mean annual temperature of the summit of Mauna Loa, taking -that of the coast at 75°, would be 34° if the rate of increase was the -same as on Mount Pangerango in Java (1° per 328 feet). - -The great daily range of temperature is one of the most striking -features of the climate of the summit of Mauna Loa. The extreme recorded -by me was 38·7°, whilst Wilkes registered as much as 42°. As on most -lofty mountains the dryness of the air, as indicated by the relative -humidity, was usually great. The average percentage during my stay -between 8 and 9 A.M. was 44, at midday 43, and between 5 and 6 P.M. 56. -This may be contrasted with 72, the average for the year at Honolulu. In -the tropics the mean for the year in the lower levels often rises to 80 -and over; and it can scarcely be doubted that the Hawaiian climate is -generally drier than it was before the destruction of the forests. The -lowest relative humidity recorded by me on the summit of Mauna Loa was -20 per cent. Junghuhn on the summits of two mountains in Java, 10,500 -and 11,500 feet in height, recorded percentages as low as 5 and 13. -Further details relating to this subject are given in Note 61. The -rainfall on the top of Mauna Loa is probably very slight. During my -sojourn rain was noted on six days, but on only two could it be -measured, and the total fall could not have amounted to over a third of -an inch. - -The mean annual temperature of the great forest-zone at the elevations -where it displays the greatest luxuriance of growth on the island of -Hawaii, that is, between 4,000 and 6,000 feet, would be estimated at 63° -and 57° F., if we take the rate of decrease before employed of about -three degrees per 1,000 feet. But remembering the heavy rainfall in this -region and the usual occurrence of a protecting belt of cloud during the -day, this might seem to be too high. According, however, to a table -given by Mr. Jared G. Smith in his annual report of the Hawaii -Agricultural Experiment Station for 1902, the average temperature at -4,000 feet would be 65°. I cannot help thinking this is excessive as an -average for the island. In the latter part of May, 1897, the mean -temperature during my sojourn of twelve days at elevations between 6,000 -and 6,700 feet around the slopes of Mauna Kea was 51·2°; whilst for -eight days in the first part of June in the same region the mean -temperature was 58·2° at an altitude of 4,000 to 4,300 feet. - -It is possible, as I have pointed out on a later page, to recognise in -the different zones of vegetation the floras of a variety of latitudes; -and these zones are to a large extent controlled by temperature as well -as by other conditions. Thus the Fijian would be amongst familiar -vegetation on the lower slopes of Mauna Kea, whilst the Maori would be -at home halfway up the mountain-slopes, and the African from the upper -forests of Kilima Njaro and Ruwenzori would find in the higher levels -much to remind him of his native land. - -The upper woods extend usually to 8,000 or 9,000 feet above the sea, and -vegetation of a scrubby character occurs as high generally as 10,000 or -11,000 feet. The highest regions present only a barren rocky waste. - - - THE RAINFALL. - -_The Hawaiian Islands._—Although on account of the extensive deforesting -of the Hawaiian Islands since their discovery the contrast between this -group and that of Fiji is now, as regards rainfall, somewhat emphasised, -it is almost certain that in early times the contrast was much less -marked. In the lower levels the natives and sandal wood traders in the -past, and the agriculturists in the present, have accomplished much in -this direction. Between 1,000 and 3,000 feet, whole forests were in my -time disappearing under fire and axe for the coffee plantations. Above -those levels up to the higher limits of the woods, cattle were -destroying the forests in a wholesale fashion; whilst foreign insects -were proving themselves almost as great enemies to the vegetation. I -remember an enterprising agriculturist explaining to me how he cleared -the land of forest around his station. A large tract having been fenced -in, the cattle were introduced. After destroying the undergrowth and the -young trees, the animals attacked the bark of the trees, and in a year -or two, without fire or axe, the land was cleared. The consequence of -this unchecked destruction of the forests was in my time becoming only -too evident. When I passed through Ookala, on the Hamakua coast, at the -end of May, 1897, there was a water famine. Water was sold at a quarter -of a dollar a bucket, and the allowance for a family was three oil-cans -a week. Stealing water was a crime and punished by the plantation -authorities by dismissal or a five-dollar fine. - -If we could look back for fifty or sixty years—I am now quoting from the -reports of Prof. Koebele and Dr. Stubbs—we should see large forests -where we now see barren slopes and plains. Originally forests covered -the upland plateaux and mountain slopes of all the islands. Now much of -the original forests has been removed, and large areas of naked soils -and bare rocks remain. The present forest area, writes Mr. Giffard, the -editor of the _Hawaiian Forester_ (August, 1904), is about 20 per cent. -of the islands, a small fraction of what it was a hundred years ago. It -is, however, very satisfactory to learn that American energy is now -combating this evil. Already in the January number of the same journal -is to be found a report by Mr. W. L. Hall, of the Bureau of Forestry, on -“The Forests of Hawaii”; and now, under the charge of Mr. Jared G. -Smith, institutions have been formed and experiment stations have been -established for “the intelligent and skilful cultivation of the soil.” -Hawaii owes much to the United States Department of Agriculture. May we -in England take the cue in the case of our own Crown colonies! - -Under these circumstances the comparison of the present rainfall of -Hawaii must be carried out with discrimination. But it may be at once -observed that to make a contrast in detail between the rainfalls of -these three groups is quite beyond the province of this work; and this -remark applies also to the other observations on the climatic -conditions. I can only treat the subject in an illustrative fashion in -connection with the general subject of their floras. - -Thanks to Professor Lyons, the Government meteorologist, the rainfall -has long been systematically investigated. It may be said to range -anywhere between 10 and 300 inches. As in most groups within the -trade-wind belts, there is a great contrast in the rainfall between the -weather and leeward sides of the islands, which is well exhibited in the -large island of Hawaii. Whilst in the Hilo district on the wet side of -the island the annual rainfall near the coast is about 120 inches, on -the Kona coast of the dry side of the island it may be anything between -20 and 50 inches and it may fall to less than 10. The effect of -elevation is, however, evident on both the weather and lee sides of the -island. Thus at a height of 1,650 feet in the Hilo district it is as -much as 180 inches, and at a greater elevation 210 inches. At a height -of about 1,600 feet at Kealakekua, on the dry side of the island the -average yearly rainfall, according to the results kindly supplied to me -by the Rev. S. H. Davis, was for the six years, 1891-6, 60 inches. On -the beach, as he says, it is “very much less,” probably not 30 inches. -Dr. Maxwell, in his report on “Irrigation in Hawaii,” mentions a -locality in Maui where the rainfall at the sea-shore was 28 inches, and -at a height of 2,800 feet up the mountain side as much as 179 inches. In -the region of the cloud-belt, which coincides with that of the -forest-zone on the slopes of the great mountains of Hawaii and extends -up from about 3,000 to 7,000 or 8,000 feet above the sea, the average -annual rainfall would probably be rarely under 200 inches, and in some -localities it might approach 300 inches. There are some particularly wet -mountains, and amongst these may be placed the high table-land of Kauai -(4,000 feet) and the flat summit of Mount Eeka (6,000 feet) in West -Maui. Here in a region almost of eternal mist we have developed a -special bog-flora. - -Hillebrand describes the flat top of Mount Eeka as “wrapt in a cloud of -mist nearly the whole year.” Whilst descending this mountain I was -overtaken by the darkness at a little under 5,000 feet above the sea. -Through the night there was a continuous soft rain, or rather a heavy -wet mist, and I passed it under conditions suggestive of living in a -sponge. Everything was reeking with moisture. The air was saturated with -it, and water dripped from every leaf and branch, whilst the ground on -which I stood was soft and yielding and soaked with water like a sponge. -The surface was cut up by numerous narrow water-channels ten to twenty -feet deep and only a couple of feet wide, their very existence almost -concealed by ferns, whilst torrents rushed along at the bottom and kept -up a strange music through the night. This was the longest night I have -ever experienced, as my standing-ground was very limited, and with a -water-channel a foot or two away on either side I had to keep on my legs -until the dawn. - -Above the cloud-belt, at elevations of 10,000 feet and over, the -rainfall is evidently very small. I have before remarked that during my -stay of twenty-three days (August 9-31) on the summit of Mauna Loa -(13,600 feet) the rain did not exceed one-third of an inch in amount. I -have by my side the report to the Weather Bureau, compiled by Prof. -Lyons, on the rainfall of this large island of Hawaii for the entire -month (August, 1897); and it enables one to make a comparison, in some -respects unique, of the distribution of the August rainfall on Mauna -Loa, from its base to its summit, where it occupies the breadth of the -island. Whilst on the east or wet side from the coast up to 1,500 feet -amounts ranging from 11 to 15 inches were measured, on the west or dry -side between one and two inches were registered at the coast, and 10 -inches at Kealakekua, about 1,600 feet above the sea. But the level of -maximum precipitation would lie much further up the mountain slopes on -either side, probably at an altitude of 4,000 or 5,000 feet, and here -the rainfall for the month could not have been less in either case than -20 inches. Above this line of greatest rainfall the amount of -atmospheric precipitation would become less and less until beyond the -upper forest zone above 10,000 feet to the summit (13,600 feet) the -quantity would be very small; and judging from my observations, that -covered three-fourths of the month, the rainfall on the top of the -mountain for August would not have far exceeded half an inch. - -The dry climate of the summits of Mauna Kea and Mauna Loa is reproduced -on the tops of the Java mountains and on the summits of the Owen Stanley -Range in New Guinea. Sir W. Macgregor found a fine and dry climate on -the top of the mountains last named, beyond the limits of the forests, -which extend to 12,000 feet above the sea. Below lay the cloud belt, a -zone of moss and fog, where at an elevation of 7,000 to 8,000 feet -everything was reeking with moisture (_Journ. Roy. Geogr. Soc._ 1890). -Observers at the coast often little imagine, when looking at a -cloud-concealed mountain peak, that although the cloud-belt from below -looks black and lowering and rain is falling heavily in the gloomy -forests, there is on the upper side a region of bright sunshine, and -that the peak stands out, unseen by them, above a sea of clouds -sparkling brilliantly in the sun and dazzling in their whiteness. It -will be seen from the table given in Note 61, that during my sojourn on -the summit of Mauna Loa the sky was cloudless or almost free from cloud -during nearly half the time. The mean cloudiness in the forenoon for -twenty-two days was 1·3 and for the afternoon 3·5, whilst the nights -were cloudless. - -_The Rainfall of Fiji._—The rainfall of Fiji is known to be very large. -In illustration I will take Vanua Levu, the second largest island, -partly because of my familiar acquaintance with it, and partly because I -have at my disposal measurements for both the lee and weather sides of -the island—the first dry and characterised by a scanty and peculiar -vegetation, the second humid and densely forested. At Davutu, near the -sea-level on the weather or wet side of the island, the average yearly -fall for a period of sixteen years up to 1898 was 160 inches (these -observations were made in the grounds of the manager’s house and I am -indebted to Mr. Barratt for allowing me to inspect them). The -mountainous backbone of the island, which has an elevation ranging -usually from 2,000 to 3,000 feet, is generally in the rain-clouds. -During the months I was occupied in examining the geology of these -mountains, it was a common experience to be drenched to the skin all day -long, and I cannot doubt that the annual rainfall in the higher levels -must often reach 300 inches. Those familiar with the “sun-burnt” lands -or “talasinga” plains that mainly form the north or lee side of the -island, would expect a great difference in rainfall as compared with the -south or weather side. There is a marked difference, it is true, but it -is far less than we might have looked for. At Delanasau on the north -coast, less than a hundred feet above the sea, the mean rainfall for -seven years (1871-77), according to the observations of Mr Holmes, was -113 inches, and the range 80 to 159 inches (see Horne’s _Year in Fiji_). -In discussing the origin of the arid-looking plains on the north or lee -side of the island in Note 22, I have shown that the explanation is to -be found not so much in the rainfall as in the dryness of the air as -indicated by the relative humidity. - -The rainfall varies greatly in and around Vanua Levu, but there is -little doubt that by far the greatest bulk of the rain is precipitated -on the upper weather slopes of the mountainous backbone of the island. -Taviuni, which lies off its weather coast, is probably the wettest among -the smaller islands of the group. In 1877, when 80 inches were recorded -by Mr. Holmes at Delanasau on the north side of Vanua Levu and 73 inches -at Levuka in the island of Ovalau, 251 inches were measured in Taviuni -at Ngara Walu 564 feet above the sea; and in 1875 the rainfall recorded -at Taviuni was 212 inches, and at Delanasau 126 inches (Horne). - -Fortunately, the Fijian islands have not been long enough occupied by -the whites to produce much effect on the rainfall through the -destruction of the forests. A significant warning, however, has been -given in the vicinity of Levuka. The woods of the hills around the town, -as we learn from Mr. Horne, were cut down to prevent them from affording -shelter to the unfriendly natives of the interior, the result being to -reduce the number of rainy days in a few years from 256 to 149 per -annum. - -_The Tahitian rainfall._—The annual rainfall of the coast districts of -Tahiti is placed at about 50 inches (_Encycl. Brit._ vol. 23); but, as -is observed by Nadeaud and Drake del Castillo, the rain-clouds gather -round the peaks, and the precipitation is much greater in the interior -than at the “littoral,” with a corresponding result in a striking -difference between the vegetation of the two regions. Probably, -therefore, the rainfall for the year on the wooded mountain slopes and -at the heads of valleys where the vegetation is most luxuriant would be -over 100, and perhaps as much as 150 inches in places. (The annual -rainfall in Rarotonga is, according to Cheeseman, about 90 inches.) - -It is evident that in the three groups of Hawaii, Fiji, and Tahiti, the -rainfall varies greatly with situation and with elevation; but the -contrast is much greater in Hawaii than in Fiji. Thus there would be -scarcely any place on the lee side of Vanua Levu where the average -annual fall would be less than 80 or 90 inches, except perhaps in the -Undu Promontory, whilst on the lava-bound coast of the west or lee side -of Hawaii, it may be reduced to 20 inches and less. There is no doubt -that this was to some extent the case in pre-European times, since Fiji -must have possessed for ages, on the northern sides of the larger -islands, its arid “talasinga” or “sun-burnt” plains; and in the island -of Hawaii there must have always been vast, scantily vegetated lava -fields at the sea-border. It is probable, however, that it is in the -older islands of the Hawaiian group, those where the volcanic forces -have been long extinct, that the rainfall has been chiefly affected by -deforestation. Speaking generally, in pre-European times the climatic -conditions of the lower levels of the group, that is below 4,000 feet, -which are alone comparable with Fiji, were less contrasted with the -climatic conditions of the Fijian islands than they are at present. By -reason of their great elevation, the Hawaiian islands present a mountain -climate not found in Fiji, and scantily represented in Tahiti. It is, -therefore, in the flora of the Hawaiian uplands that we should expect to -find the great distinguishing feature between that group and Fiji. - - - _Summary of the Chapter._ - -(1) Whilst the winds and the currents have been working tranquilly -through the ages, bringing always the same vascular cryptogams and -shore-plants to the Pacific islands, the bird has ever been a disturbing -factor in the inland flora, and changes often of a revolutionary -character have taken place from time to time within the forest-zone. - -(2) In the discussion of the inland plants of these islands, the Fijian, -Tahitian and Hawaiian areas are taken as centres of development and -dispersal, and as including the groups around. - -(3) On account of the contrast in physical conditions presented by these -archipelagoes, differences with which some of the most distinctive -features of the floras are to be connected, a comparison of the islands -from this standpoint is first necessary. - -(4) Since the largest islands of the Fijian and Hawaiian areas are from -five to ten times the size of Tahiti, the largest island of the Tahitian -region, we would expect to find in the two first-named groups a much -more varied flora. - -(5) There are three huge mountain-masses in the Hawaiian group which -rise to between 10,000 and 14,000 feet, and there is in the aggregate a -large area elevated more than 4,000 feet above the sea. These elevated -regions are almost unrepresented in the southern groups, the Fijian -islands being only comparable with the lower levels of the Hawaiian -islands below 4,000 feet, and the same is true of all the groups with -the exception of a limited area in Tahiti, where the mountains reach a -height of 7,300 feet, and of the solitary peak of Savaii in Samoa, which -attains an altitude of 5,400 feet. Thus the conditions for a high-level -or mountain flora which exist in Hawaii are not to be found in Fiji, but -slightly in Samoa, and to a limited extent in Tahiti. - -(6) From their position with regard to the equator and with reference to -the trade-winds a great contrast between the climates of these three -regions—the Fijian, the Tahitian, and the Hawaiian—is, as far as the -islands agree in elevation, not to be expected, and in fact does not -exist. The Fijian climate, however, is now warmer and more humid, and -the general rainfall is greater than in the case of Hawaii, but it is -probable that these differences were much less pronounced before the -destruction of the Hawaiian forests, which has been in progress since -the discovery of the group. - -(7) Anywhere around the coasts of the larger Fijian islands we might -expect an annual rainfall of not less than 80 or 100 inches. In the -Hawaiian group the rainfall at the coast may be anything between 10 and -100 inches, but is generally less than 50 inches. In Tahiti, at the -coast, it is 50 inches. In all cases the rainfall increases greatly with -elevation. In the Fijian mountains the rainfall probably varies between -200 and 300 inches. In the Hawaiian forest-zone it would range probably -between 100 and 200 inches, though this is probably exceeded in a few -localities. In the Tahitian uplands it would doubtless exceed 100 inches -and approach 150 inches. - -(8) Quite a different climate prevails on the lofty summits of Hawaii -13,000 to 14,000 feet above the sea. Here the snow lies in winter, and -the mean annual temperature is only a few degrees above the freezing -point, probably about 36° F. The difference between the mean summer and -winter temperatures is very small, and does not exceed five or six -degrees. Water freezes here during nearly every night of the year. The -daily variation of temperature is very large, the average being probably -about thirty degrees. Great dryness of the air prevails, the average -relative humidity in August, 1897, being about 43 per cent. There is but -little rain. The sun shines fiercely, and the sky is usually clear. - -(9) All Pacific climates are represented in the Hawaiian mountains, that -of Fiji on the lower slopes, that of New Zealand half way up, and that -of the Antarctic islands on the summits. - -(10) When contrasting the floras of Fiji, Tahiti, and Hawaii, it will be -necessary to restrict our comparison in the case of Hawaii to the lower -slopes below 4,000 or 5,000 feet; and we should expect the Hawaiian -mountain flora to be scantily represented in Tahiti, and scarcely at all -in Fiji and Samoa. - - - - - CHAPTER XX - - THE EPOCHS IN THE FLORAL HISTORY OF THE PACIFIC ISLANDS - - THE AGE OF FERNS - -The epochs in the plant-stocking.—The age of ferns and lycopods.—The - relative proportion of vascular cryptogams in Hawaii, Fiji, and - Tahiti.—The large number of peculiar species in Hawaii.—The mountain - ferns of Hawaii.—The origin of peculiar species.—Dr. Hillebrand’s - views.—Their origin connected not with greater variety of climate in - Hawaii, but with isolation.—Summary. - - - _Introductory Remarks_ - -IN the endeavour to follow the various stages in the floral history of -the Pacific islands from the standpoint of plant-dispersal, a method is -here adopted which is not often employed. The usual mode of making a -general description of a flora is not intended to bring out its genesis -in point of time. We describe the result of a long series of changes -dating back to some unknown period, much as one might describe the -present condition of a people without reference to their history; and -for obvious reasons rarely is an effort made to differentiate the epochs -of the stocking of the region with its plants. The difficulties -investing such a task in the case of a region situated within a -continental area would be almost insuperable. With the oceanic groups of -the Pacific such difficulties, though still very numerous, would at all -events be fewer in number and less formidable in appearance. - -Taking my cue from the well-known instance of Krakatoa, it is here -assumed that the earliest epoch is connected with the arrival of the -cryptogamic flora (ferns, mosses, lichens, &c.) through the agency of -the winds, and with the arrival of the littoral plants through the -agency of the currents. The next era is represented by the genera now -peculiar to each group, since it is implied that they have descended -from the earliest phanerogams that established themselves in the group. -The following epoch, which ends only with the arrival of man, is -characterised by the genera found outside the group; and here different -degrees of antiquity are indicated according as the genus is represented -wholly or in part by peculiar species, or contains only species found in -other regions. The following eight chapters will be devoted to the -development of the method here briefly indicated. - - - THE AGE OF FERNS. - -It was established by Dr. Treub in the case of Krakatoa that ferns and -algæ formed the earliest vegetation of this island after it had been -completely stripped of all its plants in the great eruption of 1883. It -is, therefore, but natural that the vascular cryptogams should first be -dealt with in any discussion relating to the historical aspects of these -floras. - -It has been before remarked that the epoch of ferns and lycopods, which -began with the earliest stage in the island’s floral history, may be -regarded as extending to our own day. It is thus implied that the -vascular cryptogams of those early times are yet brought there, and -that, alike with the littoral plants, these ferns and lycopods have -witnessed almost unchanged the great revolutions that have marked the -history of the inland flowering plants, more particularly those of the -forest flora. This, as I will show, is true in Hawaii, though only in a -partial sense in comparison with the other island-groups of Fiji and -Tahiti, since in Hawaii nearly half the ferns and lycopods are peculiar -to that group, whilst in Fiji and Tahiti not more than 8 or 9 per cent. -appear to be endemic. (Rarotonga, according to Cheeseman, possesses one -new species amongst its seventy-two ferns and lycopods, and probably in -this it is typical of the smaller elevated islands of Eastern -Polynesia.) - -The large proportion of peculiar Hawaiian species is the central fact in -the distribution of vascular cryptogams in the Hawaiian, Fijian, and -Tahitian archipelagoes, and indeed in the Pacific islands; and it is -around this fact that much of the following discussion will lie. (For -the data relating to the Tahitian region, I have almost exclusively -followed Drake del Castillo.) - -On looking at the table given below, it will be noticed that whilst -there are about the same number of species of ferns and lycopods in the -Tahitian and Hawaiian islands there are at least half as many again in -Fiji. When we reflect that the total areas of the Fijian and Hawaiian -groups are in each case about 7,000 square miles and that the extent of -the whole Tahitian region does not amount to 2,000 square miles, these -facts acquire a fresh significance. Ferns and lycopods might, therefore, -be expected to figure more largely in the Tahitian flora than in those -of Fiji and Hawaii; and this is indeed the case. When we examine the -relative proportion of the vascular cryptogams to the indigenous -flowering plants in each area we find that whilst in Hawaii they form -about 18 per cent. of the total flora and in Fiji not much more than -this (see Note 62), in Tahiti they constitute just a third. This excess -of vascular cryptogams is reflected in the flora of the outlying groups, -the proportion in Rarotonga being, according to Cheeseman, 30 per cent. -It is, therefore, evident that in comparison with the other groups -Tahiti possesses a marked preponderance in ferns and lycopods. In this -respect the Tahitian islands resemble those of Juan Fernandez, where -judging from the data relating to the indigenous flora given in -Hemsley’s _Botany of the Challenger Expedition_, the proportion of -vascular cryptogams amounts to between 30 and 38 per cent. - -But it has been already implied that the proportion of endemic species -of ferns and lycopods is from four to five times as large in Hawaii as -it is in Tahiti or Fiji. In Hawaii, therefore, there has been a -production of many new species, whilst in Fiji and Tahiti there has been -a great rush of immigrants. “Formative energy” in Hawaii (to adopt an -expression of Dr. Hillebrand) and “active colonisation” in Fiji and -Tahiti, such would appear to be the most conspicuous features in the -history of the vascular cryptogams of these three archipelagoes. - -In these floras it is, therefore, apparent that respecting the vascular -cryptogams the average number of species in a genus does not supply a -means of contrasting them. As indicated in the table, the fern and -lycopod floras of Fiji and Hawaii are similar in this respect. Yet in -each the average number of species to a genus has a separate -significance. A genus may acquire its species through immigration, or -they may arise from its formative energy within the particular area. The -first principle has been largely dominant in Fiji, the last in Hawaii, -and the resemblance between the average number of species in a genus in -these two groups is to a large extent accidental. Between the vascular -cryptogams of Fiji and Tahiti, however, such a comparison is legitimate; -and since the average formative energy is in these groups about the -same, the difference is to be attributed to a lessened number of -immigrants into the Tahitian area. - - - TABLE OF VASCULAR CRYPTOGAMS (FERNS AND LYCOPODS) IN THE GROUPS OF - TAHITI, HAWAII, AND FIJI. (See note 63.) - - +--------------------+------+------+----+ - | Group. |Tahiti|Hawaii|Fiji| - +--------------------+------+------+----+ - | Number of genera. | 38 | 29 | 40 | - +--------------------+------+------+----+ - | Number of species. | 154 | 155 |237 | - +--------------------+------+------+----+ - |Species to a genus. | 4·1 | 5·4 |5·9 | - +--------------------+------+------+----+ - | Number of | 13 | 70 | 20 | - | endemic species. | | | | - +--------------------+------+------+----+ - | Percentage of | 8 | 45 | 8 | - | endemic species. | | | | - +--------------------+------+------+----+ - |Percentage of ferns | | | | - | and lycopods among | 33 | 18 | 21 | - |the vascular plants.| | | | - +--------------------+------+------+----+ - -The results, so far mentioned, are in the main consistent with the -geographical position and the degree of isolation of these three areas. -From their proximity to the large continental islands of the Western -Pacific, the Fijian islands would have readily received a great number -of immigrants from the west, since the intervening sea is not over 500 -miles in breadth. They lie in the track of the main line of migration -into and across the South Pacific, a track which has been followed by -flowering plants and animals as well as by aboriginal man. Assuming that -the migration of the vascular cryptogams extended from Fiji eastward to -Tahiti, fewer of the immigrants would reach the last-named group. Fewer -still would reach the Hawaiian islands, which excluding the groups of -low coral islands to the southward are cut off on all sides, whether -from the Fiji-Samoan and Tahitian areas, from the coasts of North -America, or from the regions north and west, by a breadth of ocean that -is never less than 1,500 miles. - -That the main track of the ferns and lycopods across the South Pacific -to Tahiti has been eastward there can be little doubt. This is indicated -in the tables given by Drake del Castillo for Eastern Polynesia, and -also by an analysis I have prepared of the distributions that he gives -for the species of the Tahitian region (see Note 64). Out of the 154 -species there are only two that belong exclusively to the American side -of the Pacific; whilst 58 are derived exclusively from the Asiatic side, -and mainly from Indo-Malaya. The drift of the ferns and lycopods -eastward from Fiji is also brought out in the number of Tahitian species -common to Hawaii and Fiji. Of these about 76 per cent. are common to -Fiji or to the groups around, and only 30 per cent. occur in Hawaii. The -Tahitian species found in Hawaii occur also in Fiji with the exception -of two or three mountain species which have doubtless failed to find a -suitable elevation in Fiji. These two or three mountain ferns and -lycopods are probably the only vascular cryptogams possessed in common -by Hawaii and Tahiti to the exclusion of other groups. (See Note 64.) - -The prevailing Indo-Malayan origin of the ferns and lycopods of the -archipelagoes of the Fijian area (Fiji, Tonga, Samoa) is so well -established in the writings of Seemann, Baker, Hemsley, Christ, and -Burkill that there is no necessity to enter into details here. That the -stream of vascular cryptogams to Hawaii has proceeded mainly from the -Old World side of the Pacific is shown in the circumstance that of the -eighty and odd species found outside the group nearly half are from the -Asiatic side exclusively and only three from America alone, whilst about -a fourth occur in both continents, and a fourth are confined to -Polynesia. One point, says Dr. Hillebrand, comes out in strong relief, -and that is “the great number of ferns scattered over the long track -which leads from the Hawaiian islands through Polynesia and Malaysia to -the east coast of tropical Africa.” But he adds significantly that “it -cannot be inferred from this fact that all the species in question have -travelled eastward to find the terminus of their long migration on this -group, unless the principle be established, that the formative energy of -a species or genus be greatest at the circumference or farthest -extremity of its area” (p. 542). - -Though evidently prepared to admit the general eastward trend of plants -in the Pacific, Dr. Hillebrand (p. xxviii) puts forward in the case of -the ferns the startling view that originally spores of a few simple -species have been diffused over various countries and that they have -there evolved on parallel lines “predetermined by the structure of the -original immigrant” a series of higher forms, so that the same form -might have been produced in two widely distant localities, as, for -instance, in Ceylon and Hawaii. The editor, Mr. W. F. Hillebrand, gives -good reasons for his belief that this does not represent the matured -opinion of the author. It is, however, worth noting in this connection -that Dr. Karl Mueller has advanced a similar view with respect to the -lower orders of plants. (See a translation of his paper in _Trans. and -Proc. N. Z. Inst._ Vol. 25.) Bearing in mind the known capacity of ferns -for dispersal by the winds, an hypothesis of this kind, even if -established, seems scarcely needed in the study of fern-dispersal. - -It is probable that many of the ferns and lycopods reached Hawaii -directly and not through South Polynesia. The mountain-ferns of this -group could hardly have been received by that route, since, as is shown -below, they do not as a rule occur in that region. - -Some other interesting relations present themselves in connection with -the Hawaiian ferns and lycopods when we consider the distribution of its -non-endemic species in the other two groups of Fiji and Tahiti. Out of -these species, some eighty in all, not more than half are common to all -three groups, and about two dozen have not been found either in Fiji or -in Tahiti. Of these last quite half are mountain species in Hawaii, -having their station at elevations exceeding those of the highest -districts of Fiji and of the several islands of the Tahitian area, -excepting the limited region comprised in the uplands of Tahiti itself. - -A glance at the list, given in Note 65 of some of the mountain ferns of -Hawaii not recorded from Fiji and Tahiti will show that these species -are very widely distributed. Ferns and lycopods found in the Himalayas -and in the Andes meet on the higher slopes of the lofty mountains of -Hawaii and in no other of the less elevated island-groups of the open -Pacific. This distribution of the vascular cryptogams thus foreshadows a -principle that will come into prominence in the case of the flowering -plants, namely, that difference in elevation has been an important -factor in determining some of the contrasts between the Hawaiian, -Fijian, and Tahitian floras. The contrasts here implied are those -connected with the climatic conditions of station, since several plants -of temperate regions, such as Aspidium filix mas, Asplenium trichomanes, -Asplenium adiantum nigrum, &c., that are at home in the highlands of -Hawaii, do not occur in either Fiji or Tahiti. We can infer that widely -ranging ferns and lycopods have been dispersed over the oceanic groups -of the tropical Pacific with a fair degree of uniformity, and that any -marked contrasts in their distribution may be attributed to considerable -differences in the altitude of the islands. - -In appreciating such a conclusion, and in dealing with apparent -exceptions to the rule, the relation between the vertical range of a -species and its lateral distribution has to be considered. We find, for -instance, that whilst the Common Bracken (Pteris aquilina) is a mountain -plant in Hawaii, it occurs also in Fiji and Tahiti. Since, however, it -is found all over the temperate and tropical regions, and has a vertical -range in Hawaii of from 800 to 8,000 feet, any difficulty in this -respect is thus explained. Aspidium aculeatum, a characteristic fern of -temperate latitudes, seems at first to present a difficulty, which, -however, proves to be more apparent than real. Whilst it has been -recorded from Hawaii at heights of 6,000 to 9,000 feet, and from Tahiti -at 4,000 feet, it has also been found in Fiji and Samoa; but since it -was not collected by Seemann in Fiji, it can scarcely be common, and -Horne seems only to have obtained it from the tops of mountains in Vanua -Levu at an elevation of 1,800 feet. - -Up to this point the non-endemic ferns and lycopods have been chiefly -discussed. We will now briefly deal with the probable cause of the -relative preponderance of peculiar or endemic species in Hawaii as -contrasted with Fiji and Tahiti. In this respect the Hawaiian islands, -as remarked at the commencement of this chapter, come into sharp -contrast with the other two groups; but it would seem that the -differentiation has rarely acquired a generic value (see Note 66). In -this respect the age of ferns is markedly distinguished from the -succeeding era, the age of the arborescent Compositæ and of -Tree-Lobelias, to which a large number of peculiar genera belong. This, -according to my view, is to be attributed to the circumstance that -whilst the dispersion of spores by the wind is probably as active in our -own time as it was in the earliest stage of the floral history of -Hawaii, the dispersion of seeds by birds, to which the flowering plants -in the main originally owe their presence in this group, has been -greatly influenced by the various changes that have affected the -migration of birds over the Pacific, a subject discussed in later pages. - -Respecting the origin of the species of ferns and lycopods peculiar to -Hawaii, it is first of importance to quote the remarks of Dr. Hillebrand -on the subject. Speaking of the whole flora (p. xxv), but evidently with -the ferns more especially in his mind, he says:—“Nature here luxuriates -in formative energy. Is it because the islands offer a great range of -conditions of life? Or is it because the leading genera are in their age -of manhood, of greatest vigour? Or is it because the number of types -which here come into play is limited, and, therefore, the area offered -to their development comparatively great and varied?” It is deeply to be -regretted that sickness and death intervened before the author was able -to give to the world his matured views on the very important points here -raised. Yet they are much the same questions that man is ever putting to -the life around him. There is the same querulous note that we find in -all, the question that begins, the question that ends, and the reply -that never comes. - -“The evolution theory (writes Dr. Hillebrand, p. xxix) could hardly find -a more favourable field for observation than an isolated island-group in -mid-ocean, large enough to have produced a number of original forms, and -at the same time so diversified in conditions of temperature, humidity, -and atmospheric currents as to admit an extraordinary development in -nearly every direction of vegetable morphology, uninfluenced by -intercrossing with foreign elements.” Isolation thus admittedly offers -the preliminary determining or favouring conditions. This is directly -indicated by the fact that Hawaii possesses fewer genera of ferns and -lycopods than either Fiji or Tahiti, notwithstanding that it has the -same area as Fiji, and is in extent three or four times the size of the -whole Tahitian area. One effect of isolation in Hawaii has, therefore, -been greater room for the development of new forms. It has, however, -already been remarked that the islands of the Fijian area are much less -isolated than those of the Hawaiian group, and that in consequence the -free immigration possible in the one group has been checked in the -other. Fiji possesses in respect to vascular cryptogams at least half as -many species again as Hawaii, but Hawaii has three or four times the -number of peculiar species. Yet before this great contrast can be -ascribed to different degrees of isolation, it is necessary to exclude -another possible cause presented by the greater range of life-conditions -in Hawaii. It is possible that all the Hawaiian peculiar species may -belong to the higher levels, elevations, as before shown, not -represented in the Fijian islands, which correspond only to the lowlands -of Hawaii, that is, to levels below 4,000 feet. If this is the case, the -contrast between Fiji and Hawaii would be connected mainly with a -difference in life-conditions, and, however potent the isolating -influences might have been in Hawaii, they could hardly have been -concerned with this striking difference. - -In order to determine this point, I went carefully through the account -given by Hillebrand of the Hawaiian ferns and lycopods, noting the -altitudes there given, and making use of the maps and of my own local -knowledge of the islands of Oahu and Hawaii, where the elevation is -neither directly nor indirectly implied. As a result, I found that out -of sixty-six endemic species available for my purpose, forty-seven had -their stations at levels below 4,000 feet, that is in the region -corresponding to Fiji, and nineteen at elevations exceeding this height. -This, however, did not finally decide the question, since the proportion -of endemic species may be much smaller in the region below 4,000 feet -than in that above it. I, therefore, went over the ground again, and -found, as shown in the table below, that the percentages of peculiar -species amongst the total available for my use were not very far apart, -58 per cent. for the upper region and 43 per cent. for the lower region. - - - Distribution of the Hawaiian ferns and lycopods above and below 4,000 - feet. - - +------------------------+-------+--------+----------------+ - | |Number.|Endemic.| Percentage of | - | | | |endemic species.| - +------------------------+-------+--------+----------------+ - |Species below 4,000 feet| 110 | 47 | 43 | - +------------------------+-------+--------+----------------+ - |Species above 4,000 feet| 33 | 19 | 58 | - +------------------------+-------+--------+----------------+ - -From the above it would appear that although the process of -species-production in the Hawaiian islands has seemingly been rather -more active above than below 4,000 feet, if we were to compare the -entire vascular cryptogamic flora of Fiji with that of the corresponding -lower levels of the Hawaiian group we should obtain much the same -contrast in the proportion of peculiar species that we obtained when -comparing all the ferns and lycopods of both groups. In other words, if -we were to restrict our comparison with Fiji, and I may add Tahiti, to -that lower portion of Hawaii that corresponds in elevation, we should -not get results very different from those to be obtained by including -the Hawaiian upland regions as well. - -We are, I think, on these grounds justified in assuming that the -relatively great development of new species of ferns and lycopods in -Hawaii as contrasted with Fiji is not to be connected with the greater -elevation of those islands. The only thing that we have been able to -associate with the greater altitude of the Hawaiian Islands, and the -consequent greater range of climatic conditions, when contrasting the -Fijian and Hawaiian vascular cryptogams, is the occurrence of a number -of peculiar mountain species and of wide-ranging temperate species that -are found in the uplands of Hawaii, but not in the less elevated islands -of Fiji. - -On the whole, therefore, it is to be inferred that the greater display -of formative power among the ferns and lycopods of the Hawaiian Islands -is in great part to be associated with the isolation of this group as -compared with those of Fiji and Tahiti. The indications supplied by the -vascular cryptogams resemble in kind those we shall obtain from the -study of the flowering plants, but there is this important distinction. -In formative power, as shown in the development of new specific and -generic types, the Hawaiian vascular cryptogams are far exceeded by the -flowering plants where the proportion of endemic species amounts to 80 -per cent. We have no reason to believe that the winds, to which the -ferns and lycopods chiefly owe their dispersal, are less effective now -in carrying their spores than they were in the earliest era of the -floral history of Hawaii or in the intervening periods. In the course of -ages the winds have been more uniform in their action as -plant-dispersers even than the currents, and certainly far more than -birds. - -On the other hand, in the case of the Hawaiian flowering plants that -depend on the varying influence of the migrant bird, the agency of -dispersal has often been suspended altogether, and far greater -differentiation or departure from the original type has resulted, the -amount of change often reaching to the value of a generic distinction. -It is a question, however, whether the isolation of the Hawaiian Islands -is to be entirely connected with their mid-oceanic position. It will be -shown in Chapter XXXIII. that effects almost as great have been produced -in continental regions and in continental islands, and that the isolated -situation of Hawaii has not induced but has intensified these results. -In the later eras of plant-life a process of segregation has been ever -active throughout the tropical world whether in the case of an elevated -oceanic island or of a mid-continental mountain. - -The following are some of the principal points that have been emphasised -in the foregoing discussion of the ferns and lycopods of the Hawaiian, -Fijian, and Tahitian Islands:— - -(_a_) In all three groups the vascular cryptogams (ferns and lycopods) -have been largely supplied from the warmer regions of the Old World. But -whilst in the South Pacific the migration has been mainly from Fiji -eastward to Tahiti, it is probable that Hawaii in the North Pacific has -been in part independently stocked. - -(_b_) Whilst in Hawaii many peculiar species of ferns and lycopods have -been developed, in Fiji and Tahiti there have been comparatively few. - -(_c_) Whilst there has been more or less free immigration into Fiji and -Tahiti there has been comparative isolation in Hawaii. Though the areas -of the Fijian and Hawaiian archipelagoes are about the same, Fiji -possesses at least half as many species again as Hawaii; but Hawaii owns -three or four times the number of peculiar species. - -(_d_) Though the land-area of the Tahitian region does not exceed a -fourth part of that of Hawaii, it has the same number of species. The -Tahitian islands therefore display a predominance of ferns and lycopods. - -(_e_) The non-effective influence of the greater elevation of the -Hawaiian Islands on its preponderance of peculiar species is shown by -comparing all the ferns and lycopods of the Fijian and Tahitian Islands -with those of the corresponding lower levels of the Hawaiian Islands, -when we find much the same contrast exhibited in the number of peculiar -species. - -(_f_) Whilst a large proportion of the ferns and lycopods are common to -all three groups, Hawaii possesses a number of mountain species, widely -distributed in temperate regions and on the higher levels of mountainous -areas in the tropics, that are not found either in Fiji or in Tahiti. -Their absence from these two groups is due to the insufficient elevation -of the islands and to the non-existence there of extensive areas of any -altitude. - -(_g_) The agency of the winds in dispersing the spores of ferns and -lycopods has been relatively uniform through the ages when compared with -the varying agency of the migrant bird, to which the flowering plants -mainly owe their distribution. Thus it is that in the Pacific islands -the vascular cryptogams have experienced much less differentiation than -the flowering plants, though as a rule far older denizens of the -islands. Yet we cannot doubt that the same principle has been at work in -both cases, the difference arising in the instance of the flowering -plants from the interrupted and often suspended agency of birds in the -work of dispersal. - -(_h_) It is a question whether there is not something more concerned in -the isolation of the Hawaiian group than its mid-oceanic position, since -effects almost as great have been produced in continental regions. - - - - - CHAPTER XXI - - THE ERAS OF THE FLOWERING PLANTS - - THE AGE OF COMPOSITÆ. - -The islands of the tropical Pacific as the homes of new genera and new - species.—The significance of a large endemic element.—Synopsis of the - eras.—The era of endemic genera.—The endemic genera of - Compositæ.—Their affinities and mode of dispersal.—The mystery of the - suspension of the dispersing agencies.—Mr. Bentham’s views.—The - remnant of an ancient Composite flora in the tropical Pacific.—The - dispersion of the Compositæ antedates the emergence of the - island-groups of the Fijian region at the close of the Tertiary - period.—Summary. - - - _The Endemism of the Pacific Island Floras._ - -AS far as the production of new species is concerned, the Hawaiian group -presents the same contrast with the Fijian and Tahitian groups in -respect of the flowering plants that it does as regards the ferns and -lycopods. The proportion of endemic species, after excluding all -introduced plants, is in Hawaii 80 per cent., in Fiji about 50 per -cent., and in Tahiti 35 per cent. (see Table A). The same contrast is -also displayed in the number of peculiar genera. In Hawaii there are, -according to Dr. Hillebrand, 37 or 38, and in Fiji Dr. Seemann -discovered 16; whilst, as we learn from Drake del Castillo, there are -only 3 or 4 in the Tahitian Islands. (As will be pointed out later on, -these numbers for Fiji and Hawaii have to be reduced, but the general -inference to be drawn from them is not materially affected; see Table -B.) - -But if we look at the accompanying table (Table B) we notice that the -flora of Hawaii is sharply contrasted with those of Fiji and Tahiti not -only in the large proportion of endemic genera, but also in the large -number of non-endemic genera with peculiar species, and in the small -proportion of genera possessing no peculiar species. There is an endemic -element of greater or less degree in about 70 per cent. of the Hawaiian -genera, whilst in Fiji only about 53 per cent. and in Tahiti as few as -34 per cent. of the genera contain to a varying extent peculiar species. -Another feature brought out in this table is the relative poverty of -genera in the Hawaiian Islands. Fiji, though about the same size as -Hawaii, contains nearly half as many genera again, whilst the islands of -the Tahitian region, which in the aggregate amount to only one-third or -one-fourth of the area of the islands of Hawaii, possess nearly as many -genera. - - - TABLE A (FLOWERING PLANTS). - - _Proportions of Endemic Species in the Hawaiian, Fijian, and Tahitian - floras, with those - for Samoa, Tonga, and Rarotonga added._ - - +---------+------------------+----------------+----------------+ - | Groups. |Number of species.| Number of | Percentage of | - | | |endemic species.|endemic species.| - +---------+------------------+----------------+----------------+ - | Hawaii | 686 | 546 | 80 | - | | | | | - | Fiji | {S. 617 | {288 | {47 | - | | {H.1086 | {620 | {57 | - | | | | | - | Tahiti | 315 | 112 | 35 | - +---------+------------------+----------------+----------------+ - | Samoa | 326 | 110 | 34 | - | | | | | - | Tonga | 285 | 17 | 6 | - | | | | | - |Rarotonga| 140 | 17 | 12 | - | Island | | | | - +---------+------------------+----------------+----------------+ - -_Remarks._—The materials for this table have been obtained from the -works of Hillebrand for Hawaii, Seemann and Horne for Fiji, Drake del -Castillo for Tahiti, Reinecke for Samoa, Hemsley and Burkill for Tonga, -and Cheeseman for Rarotonga. The two estimates for Fiji are marked S. -for Seemann and H. for Horne, the last being a rough preliminary -computation made by Horne himself. - -The results given are only to be considered as approximations liable to -emendation, but as regards the proportion of endemic species in the -several groups they no doubt illustrate fairly well the relative degree -of endemism in the various archipelagoes. The results for Samoa, Tonga, -and Rarotonga are merely added in order to enable a comparison to be -made with sub-groups of a region and with solitary islands, the -Hawaiian, Fijian, and Tahitian groups being regarded as the three -principal centres of plant-life in the open Pacific. - -All plants introduced by the aborigines and the white man are excluded. -In so doing, I have mainly followed Seemann, a safe guide in all matters -relating to weeds and to cultivated plants. The flora of a Pacific -island thus treated undergoes serious diminution in its extent. In the -case of the Rarotonga flora, for example, which according to Cheeseman -includes about 260 flowering plants, the number of truly indigenous -plants, in the sense here implied, is only 140. Though this is an -extreme case, it will serve to illustrate the principle here followed. - - - TABLE B (FLOWERING PLANTS). - - _Comparison of the Hawaiian, Fijian, and Tahitian genera._ (_All genera - containing - introduced plants entirely are excluded._) - - +----------+------------------------------------------+---------+-----------+ - | | Non-endemic genera. | | | - | +------------+---------------+-------------+ | | - | Group. | No endemic | Some species | All species | Endemic | Total. | - | | species. | endemic, some | endemic. | genera. | | - | | | not. | | | | - +----------+------------+---------------+-------------+---------+-----------+ - |Hawaii | 70(31)| 30(13)| 95(43)| 28(13)| 223(100)| - | | | | | | | - |Fiji {| S. 150(47)| S. 74(23)| S. 87(27)| S. 10(3)|S. 321(100)| - | {| H. 162(47)| H. 80(23)| H. 94(27)| H. 10(3)|H. 346(100)| - | | | | | | | - |Tahiti | 125(66)| 21(11)| 40(21)| 4(2)| 190(100)| - |(Eastern | | | | | | - |Polynesia)| | | | | | - +----------+------------+---------------+-------------+---------+-----------+ - -_Remarks._—The figures in brackets are percentages. S. = Seemann, H. = -Horne and Seemann. - -In the construction of this table, Hillebrand, Seemann, and Drake del -Castillo have been mainly followed, except with regard to the endemic -genera for Hawaii and Fiji. In this respect the _Index Kewensis_ has -been largely consulted as well as Engler’s publications, as indicated in -the text. Hillebrand’s total of nearly forty Hawaiian peculiar genera -and Seemann’s total of sixteen for Fiji have thus been considerably -reduced. The two results given for Fiji are those of Seemann alone and -with Horne superadded. Horne discovered, according to Hemsley, no new -genera, but several genera from outside regions were added to the Fijian -flora. Taking them as twenty-five (two-thirds of his own computation), I -have apportioned them as in Seemann’s results. The Tahitian region here -includes Eastern Polynesia. - -It is necessary before proceeding further to obtain a correct idea of -the significance of a large endemic element in the phanerogamic flora of -a Pacific archipelago. We have therefore at the outset to inquire -whether it is indicative of isolation or of antiquity. If the number of -peculiar genera is to be regarded as the test of the relative antiquity -of different Pacific floras and, by implication, of the islands to which -they belong, these three groups, as shown in Table B, would arrange -themselves in the following order, namely, Hawaii, Fiji, Tahiti. This -test might be reliable if the several groups were in the same condition -of isolation. Since, however, as we have previously seen, the Fijian -Islands still enjoy a fairly free communication with the islands -westward, whilst the Hawaiian group is largely cut off, it is apparent -that the tendency to generic differentiation in Fiji might have been -often swamped by immigration, and that Fiji with its much smaller number -of endemic genera may even be older than Hawaii. This objection does not -apply quite as forcibly to a comparison between Hawaii and Tahiti, yet -for reasons before given it may be regarded as sufficient to negative -any inferences concerned with relative antiquity. - -On account, therefore, of the great differences in the degree of -isolation of these three groups, we cannot be guided in our estimation -of the relative antiquity of their floras by their number of peculiar -genera. With the evidence at our disposal we are compelled to accept the -view, which indeed a single glance at a map would suggest, that the -number or proportion of endemic genera is to be connected with the -degree of isolation. Whether a parallelism can be traced in the original -stocking of these groups with their earliest flowering-plants is a -matter that can only be elucidated by a further analysis of the peculiar -genera. - - - SYNOPSIS OF THE ERAS OF THE FLOWERING PLANTS IN THE TROPICAL PACIFIC. - -A. _The Era of the Endemic Genera._—Mostly American in their affinities. -Represented particularly by Compositæ and Lobeliaceæ. - - B. _The Era of Non-Endemic Genera._ - - (1) The mountain genera, either cosmopolitan in temperate latitudes or - derived from the New Zealand or the Antarctic flora. Mostly - represented in Hawaii. - - (2) The genera forming the low-level flora of Hawaii below 4,000 or - 5,000 feet and composing almost the entire floras of the Fijian and - Tahitian regions. Predominantly Indo-Malayan. - - (_a_) The age of general dispersal over the tropical Pacific, the - genera with only peculiar species being first treated, and afterwards - those possessing a non-endemic element. - - (_b_) The age of local dispersal over the tropical Pacific. - - - THE FIRST ERA OF THE FLOWERING PLANTS, BEING THE AGE OF THE ENDEMIC - GENERA. - -With the above preliminary remarks I pass on to the next stage in the -history of the stocking of these islands with their plants. The age of -the ferns and lycopods is left behind, and it is assumed that the next -era is mainly indicated by those genera of phanerogams that are now -peculiar to their respective groups. In this connection by far the most -interesting of the three regions, the Hawaiian, the Tahitian or East -Polynesian, and the Fijian, is that of Hawaii, which, as before -observed, is distinguished from the groups of the Fijian and Tahitian -regions, or, in other words, from all the oceanic archipelagoes of the -tropical Pacific, by its large number of endemic genera. - -Peculiar genera of shrubby and arborescent Compositæ and of arborescent -Lobeliaceæ form the most striking characteristics of the endemic genera, -and therefore of the ancient flora of Hawaii. It is in this connection -of singular interest to remark that of the three endemic genera of the -Tahitian flora one is an arborescent genus of the Compositæ, and the -other two are shrubby genera of the Lobeliaceæ. There are, therefore, -indications here of an ancient insular flora of the Pacific, -characterised mainly by the prevalence of Compositæ and Lobeliaceæ. It -is, however, remarkable that not only are no endemic genera of these -orders known from Fiji or from the adjacent groups of Samoa and Tonga, -but that the Lobeliaceæ are not represented at all, whilst amongst the -Fijian Compositæ, with the exception of Lagenophora, the genera display -no endemic element as far as the data at my disposal indicate. - -The problem we are brought face to face with is clearly stated by Mr. -Hemsley in the _Introduction to the Botany of the Challenger Expedition_ -(p. 68). “In Polynesia as elsewhere,” he remarks, “the Compositæ more -particularly are perplexing to the botanical geographer, for although -they have their greatest affinities in America, as well as the -sub-arboreous Lobeliaceæ, so numerous in the Sandwich Islands, yet the -bulk of the vegetation seems to have been derived from the -Australo-Asiatic region.” - -In attempting to approach this problem I do so from the standpoint of -dispersal. There are so many intricate questions bound up with the -systematic position of these genera that in dealing with them the -student of plant-distribution would require the capacities and -opportunities of the eminent botanist who dealt with the distribution of -ten thousand species of Compositæ. On such ground, therefore, and only -under the guidance of others, I will lightly tread. - - - THE ENDEMIC GENERA OF COMPOSITÆ. - -On account of their endemic character the peculiar genera of Compositæ -are regarded as belonging to the oldest era of the flowering plants of -the island-groups lying in the tropical latitudes of the open Pacific. -This is the view of Bentham, but it is, of course, the opinion that most -botanists would arrive at with the facts before them. With the exception -of the solitary Tahitian genus Fitchia, they are all restricted to the -Hawaiian Islands, and nearly all are either shrubby or arborescent, the -greatest height of 25 to 30 feet being attained in the Tahitian genus -and in Hesperomannia of Hawaii. - -Nine Hawaiian genera are included in this era, though, strictly -speaking, we ought only to concern ourselves with the six genera, Remya, -Argyroxiphium, Wilkesia, Dubautia, Raillardia, and Hesperomannia, since -the other three, Tetramolopium, Lipochæta, and Campylotheca, are only on -the borderland of generic distinction. It is, however, necessary that we -should include these three genera in our treatment of the Hawaiian -endemic genera, more especially because they appear to have been the -last arrivals of the early Compositæ. They still display, as shown -below, a very suggestive connection with the land of their birth, a -circumstance that is of much importance in finally determining the -source of the other strictly endemic genera, where the links with their -original homes have been in most cases largely severed. - -It would, however, be quite out of place here to enter into any details -into the affinities of these Hawaiian genera of Compositæ, and I will -limit myself here to such general conclusions as may be derived from the -pages of Bentham, Hillebrand, Hemsley, and other writers, and such as -are in accordance with the facts of distribution given in the _Index -Kewensis_. Most ancient of all are the genera Remya, Argyroxiphium, -Wilkesia, and Hesperomannia, which, although belonging to tribes that -only occur on the American continent, as in the Mexican region, stand -quite isolated, and, as Dr. Hillebrand remarks, probably belong to the -oldest denizens of the Hawaiian Islands. It is noteworthy that these -four ancient genera only contain two species apiece, a circumstance that -favours their priority in point of age. - -The American affinities, however, are not always of the character that -we might have expected. Thus, it was remarked by Mr. Bentham that -although the tribe Mutisiaceæ attains a great development in South -America, and especially in Chile, its only representative in the Pacific -islands is the very rare arboreous Hesperomannia of Hawaii. - -Rather less isolated in character, and we would presume therefore of -somewhat less antiquity, are the two closely allied genera of Raillardia -and Dubautia, which have a close relative in Raillardella of the Sierra -Nevada in California. Then we come to the three genera, Tetramolopium, -Lipochæta, and Campylotheca, that, being still in touch with the world -outside, may be regarded as the latest arrivals of the early genera of -the Compositæ. Tetramolopium, concerning which botanists were unable to -agree, would seem, according to the _Index Kewensis_, to possess Mexican -and Ecuadorian as well as Hawaiian species. Lipochæta, nearly related to -other American genera, contains a dozen species, of which eleven are -found only in Hawaii, whilst the twelfth occurs, according to the _Index -Kewensis_, in California, and, according to Dr. Hillebrand, in the -Galapagos group. Of the generic value of Campylotheca there seems a -doubt, and its distinctness is scarcely recognised in the _Index -Kewensis_. It is, however, closely allied to Coreopsis, an American -genus represented, according to Drake del Castillo, in the Marquesas. - -In the Tahitian region, that is to say in Eastern Polynesia, the genus -Fitchia alone belongs to the early age of the Compositæ, so -characteristic of Hawaii. Indications of the former widespread range of -the genus over this region of the South Pacific are afforded by its -being now represented by two species in Tahiti and by one species in -Rarotonga, localities nearly 700 miles apart. It was thus regarded by -Bentham, who saw in it a solitary remnant of the ancient South Pacific -flora. Like the Hawaiian genera, as shown below, it is often restricted -to the higher levels. Botanists differ about its affinities, and a -discussion of the subject will be found on pages 20 and 66 of the -_Introduction to the Botany of the Challenger Expedition_. - -The restriction of these ancient genera of the Polynesian Compositæ to -the upland regions is of some interest. “The preponderance of Compositæ -among the high-level plants obtains almost throughout the world.” This -observation was made by Mr. Hemsley in connection with the flora of the -highlands of Tibet (_Journ. Linn. Soc. Bot._ vol. 35, 1902), where the -Compositæ constitute about 19 per cent. of the flowering plants; and I -may remark in passing that, according to Mr. Ball, one of the most -conspicuous elements in point of frequency in the higher flora of the -Great Atlas is presented by the Compositæ which make up between 12 and -13 per cent. of the whole flora (Hooker and Ball’s _Marocco and the -Great Atlas_). This feature of alpine floras is brought into great -prominence in Schimper’s recent book on Plant Geography. - -Some of the most lasting reminiscences that the naturalist will bear -away with him from the highlands of Hawaii are connected with the -Compositæ. Those who have ascended the mountains of Mauna Kea and Mauna -Loa, will remember that amongst the last plants occurring above the -forest zone, and scattered about on the ancient lava fields at -elevations exceeding 10,000 feet above the sea, are species of -Raillardia and the beautiful “Ahinahina” (Argyroxiphium). It is, -however, in the open, scantily wooded region, elevated 6,000 to 9,000 -feet, and lying between the true forest zone below and the bare lava -slopes above, that the shrubby and arborescent Compositæ of the large -island of Hawaii are most at home. Such regions, as Hillebrand well -describes (p. xxiv), are characterised by stunted trees, chiefly -Sophora, Cyathodes, Myoporum, and others, associated with arborescent -Raillardiæ of the order of Compositæ. Between them luxuriate other -shrubby Compositæ of the genera Raillardia, Dubautia, Campylotheca, and -Artemisia, together with Strawberries, Raspberries, and species of -Vaccinium. - -Botanists have not given us much account of the associates of the -interesting genus Fitchia on the uplands of Tahiti. We learn, however, -from Nadeaud that in his time these Composite trees and shrubs were -spread over the higher region of the island of Tahiti above 800 and -1,000 metres. Cheeseman, to whom we are indebted for the discovery and -the description of the Rarotongan species, tells us that this tree, -which attains a height of 25 feet in the sheltered valleys, and is much -dwarfed on the exposed ridges and hill-tops, often forms the greater -part of the forest above 500 feet, and reaches the highest peaks of the -island (2,250 feet). - -In discussing the probable mode of dispersal of these early Composite -plants of the Pacific we shall be treading on somewhat debatable ground. -We will, however, point out that the mere possession of structures that -could be utilised for dispersal of the seeds is not the only important -question here involved. If we could demonstrate that all these genera -possess exceptional capacities for distribution over the ocean, we -should prove too much, since the process has been in the main suspended -for ages. If, on the other side, it could be shown that their fruits are -not at all suited for such dispersal, we should prove too little, since -the ancestors of these genera must have been transported to these -islands in some fashion or other. This clearly indicates that other -important factors have also come into play in determining the -distribution of the early Compositæ of the Pacific islands. - -It was long ago pointed out by De Candolle that the possession of a -pappus does not, as a rule, increase the area of a Composite plant, -although as regards hooks and barbed appendages, such as occur in -Bidens, the greater areas of the plants thus provided may be, as he -thought, in some measure explained. Even in respect to hooks and barbs -it would be easy to point to cases where, as Bentham remarks, unusual -powers of adherence are by no means indicative of wide dispersal in all -cases. In any event it will be also incumbent on us to explain why these -genera no longer possess facilities for distribution. This suspension of -the means of dispersal is not, however, peculiar to the age of the -endemic genera of the Pacific islands. It is a character but in a less -degree of the succeeding age, the age of genera found outside the group, -but represented within it by endemic species; and from this we may -suspect that we have had in operation in the Pacific an influence, -far-reaching both in time and space, to which the agencies of dispersal -have been compelled to adapt themselves, an influence which has acted as -a distributor of the distributing agencies. - -Coming to the fitness for dispersal of the achenes of the early -Composite genera of the Pacific islands, it will be assumed that they -have been, as a general rule, transported in birds’ plumage. The fruits -are usually 2·5 to 12 millimetres (1/16 to 1/2 inch) in length, and are -provided either with a pappus of soft or stiff bristles, or with awns or -teeth, but these appendages vary much in size in the different genera -and in different species of the same genus. The instance of Lipochæta is -especially significant as indicating the alterations which the -appendages of the achene may have undergone in the cases of other -genera. With most species there are usually two or three teeth or short -awns, but in some species these are obsolete, and in others they are -long and stout. - -Bearing these facts in mind we should hesitate to rely too much on the -present condition of the achenes in the other genera as an indication of -the fitness for dispersal of the fruits of their ancestors. In one -genus, Campylotheca, which may be regarded as among the youngest of the -genera, the achenes are provided with barbed or hooked awns which cause -them to adhere as tenaciously to one’s clothes as in the case of those -of Bidens, an allied genus. In Fitchia, the Tahitian genus, which may be -looked upon as one of the oldest of the Pacific genera of Compositæ, the -achene is furnished with two long awns or setæ, which, as Drake del -Castillo observes, recall those of Bidens. The achenes of the other -Hawaiian genera, as regards their fitness for dispersal in plumage, may -be said to give less definite indications. In some, as in Dubautia and -Raillardia, there is a typical pappus of ten to twenty long hair-like -bristles. In others again, as in Wilkesia and Argyroxiphium, the pappus -is much reduced, and in some species of Lipochæta it is, as above -remarked, quite obsolete. - -The chances of the achenes of the parent plants having in some cases -been originally transported to the islands in the plumage of birds would -be increased by a bird making its nest of the plant-materials or amongst -the plants themselves, or by its pecking at the fruit-heads. In our own -time different species of the grouse family on the slopes of the -Californian and Columbian mountains make their nests on the ground under -the shade of Artemisia bushes and find a portion of their sustenance in -their fruits. Artemisias also form one of the features of the vegetation -of the Hawaiian uplands; but since they present only specific -differentiation they are referred to a later era. Yet it will be on the -slopes of the Rocky Mountains and of the Californian Sierra Nevada, -amongst the “sage-brush” and the grouse, that we may have to stand when -we look in thought across the Pacific towards far distant Hawaii and ask -ourselves whence came its tree-like Raillardias, its shrubby Dubautias, -its tall Wilkesias, and the silvery Ahinahinas (Argyroxiphium). - -It is possible that in some genera the achenes have, or had, a means of -adhering to plumage through a “sticky” secretion, such as is sometimes -found with Lagenophora, an Hawaiian genus of the next era, and also with -the weed-plant Adenostemma viscosum; but this is a point that has not -yet been investigated. Nor can we altogether exclude the chance of the -achenes having in some cases been transported unharmed to Hawaii in a -bird’s stomach. The possibility of this has been above implied in the -case of Artemisia; and it is pointed out in Chapter XXXIII. that pigeons -in Hawaii feed sometimes on the achenes of Compositæ. The Hawaiian goose -(Bernicla sandwicensis) lives, according to Mr. Dole, on Sonchus asper, -an introduced plant, as well as on berries (Wilson’s _Aves -Hawaiiensis_). There are numerous references of this nature in books -about birds, and it should always be remembered that birds in pecking at -the fruit-heads scatter the seeds on their feathers. (See Note 67.) - -From the foregoing remarks it may, I think, be inferred that the achenes -of the ancestors of the original Composite genera of the Pacific islands -were in all probability not unfitted for transport by birds, more -especially in their plumage. Some of my readers, however, may express a -doubt as to whether birds likely to disperse seeds would be found in any -numbers at the great heights where some of the continental Compositæ -occur. But it is well known that birds of the grouse and partridge -family frequent high levels in continental regions over much of the -globe. Arborescent Compositæ are found at heights of 10,000 to 14,000 -feet on the mountains of Central Africa; and it should be noticed that -Sir Harry Johnston observed “francolins” on the slopes of Ruwenzori up -to 13,000 feet (_Uganda Protectorate_, vol. 1; _Trans. Linn. Soc. Bot._, -Ser. II. vol. 2). Sir Martin Conway in the Bolivian Andes found geese, -ducks, gulls, snipe, &c., numerous in suitable places up to 17,000 feet -(_Journ. Roy. Geogr. Soc._, 1899); whilst geese and teal were noticed by -Sir Joseph Hooker and others at elevations of 17,000 feet in the -mountains of Tibet (Hooker’s _Himalayan Journals_; _Journ. Linn. Soc. -Bot._, vol. 35, p. 147). These are all birds, as shown in Chapter -XXXIII., that are likely to disperse plants, and probably none more -effectually than the goose, of which Hawaii possesses a particular -variety or species. It may be remarked that geese, ducks, gulls, and -other birds use Cotula plumosa in Kerguelen for making their nests (Dr. -Kidder quoted by Mr. Dixon in his book on Birds’ Nests). - -Sea-birds were probably the principal agents in carrying the achenes of -the early genera of the Compositæ to Hawaii. Dr. Hillebrand attached -importance to the tropic-bird (Phaethon) in the distribution of species -(Introd., p. 30); and since these birds breed at the crater of Kilauea -in Hawaii, 4,000 feet above the sea, and also high up in Tahiti -(Moseley), its agency is not unlikely, I am inclined to think, however, -that birds like the petrels and puffins, that in nesting burrow in the -ground, choosing places where the vegetation is thickest, and where they -would be likely to get seeds on their feathers, would be more efficient -agents. This is the view expressed by Prof. Moseley in Wallace’s _Island -Life_, p. 250. He considered that albatrosses, petrels, and puffins have -played a great part in the distribution of plants, and to some degree -especially account for the otherwise difficult fact that widely distant -islands in tropical seas have similar mountain plants. Birds, he says, -that in high latitudes, as at Tristan da Cunha and Kerguelen, often -burrow near the sea-level, in the tropics choose the mountains for their -nesting-place; and he refers to a puffin that nests on the top of one of -the high mountains of Viti Levu at an altitude of 4,000 feet, to a -petrel nesting among ferns at Tahiti at an elevation of 4,400 feet, and -to another petrel breeding in like manner in the high mountains of -Jamaica at a height of several thousand feet above the sea. He gives -point to these interesting remarks, which might be supplemented by data -from other parts of the world, by observing that it is not necessary -that the same species should now cover the range of the plants -concerned. The ancestor of the species might have carried the seeds, and -the range of the genus is alone sufficient. It may be added that, as I -have shown in Chapter XXXIII., sea-birds have been far more active -agents in the distribution of plants than many people might imagine. The -more recent observations of Ekstam in Spitzbergen have thrown -considerable light on this subject. - -Having in the first place formed the opinion that the achenes of the -early Hawaiian Compositæ are suited for dispersal by birds, and then -shown that sea-birds were probably the principal agents, we are met with -the curious difficulty that in the case of the early Hawaiian genera of -Compositæ the complete suspension for ages of the means of dispersal is -involved in the circumstances that these genera are confined to the -Hawaiian group. We can attribute to the agency of existing sea-birds the -occurrence of the genus Lagenophora in the uplands of Hawaii, on the -mountain-tops of Fiji, and in Australia and New Zealand; but the agency -of birds as at present in operation does not assist us except indirectly -in the case of the genera restricted to Hawaii or to Tahiti. Is it -possible, we may inquire, to penetrate this mystery? Why, we may ask -with Mr. Hemsley, has the agency ceased acting, and why have its -operations been confined to the conveyance of seeds _to_ the islands and -not _from_ the islands as well (_Intr. Bot. Chall. Exped._, p. 66)? I -need scarcely add that the same question presents itself with all the -other peculiar genera of these islands, and in fact with endemic genera -all over the world. What can be stranger, it may be remarked, than the -limited distribution of the Pandanaceous genus Sararanga in the Western -Pacific, although suited for dispersal by frugivorous birds. This is -not, indeed, a special difficulty connected with oceanic islands; it -applies to the whole plant-world; yet it is possible that, as it is -exhibited by the Compositæ in these islands, we may be in a better -position to grapple with the problem. But before doing so it will be -requisite to look a little closer at these early Hawaiian genera of the -Compositæ. - -The distribution within the archipelago of the genera and species of the -early Compositæ of Hawaii is worthy of notice from the light it throws, -not only on the relative antiquity of the genera, but also on the -subsequent conditions of isolation. Of the nine genera here referred to -five are distributed over most of the islands of the group. These -include all the genera possessing a number of species, namely, -Tetramolopium with seven species, Lipochæta with eleven, Campylotheca -with twelve, Dubautia with six, and Raillardia with twelve species. Of -the four genera remaining all have only two species, and are restricted -to two or three islands, Remya and Wilkesia being in both cases found in -Kauai and Maui, whilst Argyroxiphium is confined to the adjacent islands -of Maui and Hawaii, and Hesperomannia to those of Oahu, Lanai, and Maui. -These four genera that are restricted to only two or three islands are -the same before referred to as regarded by Hillebrand as the oldest, -partly on account of their isolated generic position, and partly because -in each case they only possess two species. - -Although the early Hawaiian Compositæ were evidently originally -transported to most of the islands of the group, it is noteworthy that -their subsequent isolation from the rest of the world has in the later -ages been repeated within the limits of the archipelago. Of the 56 -species, all of which are now endemic, 28, or just half, as shown in the -table on the following page, are confined to a single island. Of the -remainder, almost all are restricted to two or three adjacent islands. -Hillebrand gives only a solitary species, Lipochæta connata, as -occurring in all the islands. This suspension, to a great extent, of the -means of dispersal between the islands is also strikingly illustrated by -the Lobeliaceæ. - -We have only to mention the flora of Fiji and those of the adjacent -groups of Samoa and Tonga to exclude them from any share in the early -era of the Compositæ in the Pacific. The prevailing adventitious -character of the Fijian Compositæ is indicated in the fact that the -species of the majority of the genera are included by Seemann in his -list of Fijian weeds. There are only one or two Fijian Compositæ, such -as the mountain species of Lagenophora and the littoral species of -Wedelia, that merit the special attention of the student of dispersal. -So also with Samoa, Reinecke enumerates eight species, of which six are -weeds either of aboriginal or of European introduction, the others being -the littoral Wedelia above alluded to, and a species of Blumea found -also in Fiji. - - - DISTRIBUTION OF THE ENDEMIC GENERA OF COMPOSITÆ IN THE HAWAIIAN ISLANDS. - - +--------------+-----------------------------------------------------+-------+ - | | Distribution of the Species. | | - | +---------+----------+----------+----------+----------+ | - | Genus. | One | Two | Three | Four | | | - | | island. | islands. | islands. | islands. | General. | Total.| - +--------------+---------+----------+----------+----------+----------+-------+ - |Remya | 2 | — | — | — | — | 2 | - |Tetramolopium | 1 | 4 | 2 | — | — | 7 | - |Lipochæta | 3 | 4 | 3 | — | 1 | 11 | - |Campylotheca | 5 | 4 | 3 | — | — | 12 | - |Argyroxiphium | 1 | 1 | — | — | — | 2 | - |Wilkesia | 2 | — | — | — | — | 2 | - |Dubautia | 4 | — | 2 | — | — | 6 | - |Raillardia | 9 | 1 | — | 2 | — | 12 | - |Hesperomannia | 1 | 1 | — | — | — | 2 | - +--------------+---------+----------+----------+----------+----------+-------+ - | | 28 | 15 | 10 | 2 | 1 | 56 | - +--------------+---------+----------+----------+----------+----------+-------+ - -We have now, I venture to think, gone far to establish the existence of -an early “Composite” flora with mainly American affinities in the -Pacific islands, an ancient flora of which only the remnants now occur -in the uplands of Hawaii, Tahiti, and Rarotonga. That the achenes were -originally transported in birds’ plumage is, as we have seen, probable; -but we are still quite in the dark as to the causes of the subsequent -suspension of the means of dispersal and of the resulting period of -isolation, during which the original immigrant plants acquired their -endemic characters. In our uncertainty, therefore, we will look to Fiji -in the hope that in the absence of the early Compositæ from that group -we may find a clue that will enable us to divest this problem of some of -its difficulties. - -It might be at first considered that since these peculiar genera of -Compositæ occur in the higher levels of Hawaii and Tahiti their absence -from Fiji might be connected with the relatively low altitude of those -islands, a character that is concerned with the exclusion from the -Fijian flora of many Hawaiian and Tahitian mountain plants (see Chapters -XXIII. and XXIV.). But this view is at once negatived by the fact that -Fitchia thrives in Rarotonga, an island which does not far exceed 2,000 -feet in elevation. It is negatived also by the extensive development of -shrubby and arborescent Compositæ in the Galapagos Islands, on the -equator, in St. Helena in 16° South latitude, and in other tropical -islands, which are less than, or do not exceed, the Fijian Islands in -their altitude. - -During the age of the Compositæ it is reasonable to suppose that the -dispersal was general over the Pacific. The absence of genera indicating -this era from the islands of the Fijian region, that is, from Fiji, -Tonga, and Samoa, would become intelligible if these groups were -submerged during this age of the general dispersal of the order over -this ocean. In my volume on the geology of Vanua Levu in Fiji, I have -shown that these island-groups of the Western Pacific emerged from the -sea towards the close of the Tertiary period, a conclusion that would -enable us to assign the age of the general dispersal of the Compositæ -over the tropical Pacific to an earlier portion of the same period. - -In order, however, to make further progress in the discussion of this -difficult problem we are obliged to approach it from the outside. We -must in fact regard these genera from the standpoint of their position -as members of the vast and ancient order of the Compositæ. It is now -more than thirty years since Mr. Bentham completed his remarkable memoir -on the classification, history, and geographical distribution of the -Compositæ (_Journal Linnean Society, Botany_, London, Vol. 13, 1873). -Like De Candolle, when dealing with the facts of distribution, he -handled thousands of species, and as a result he drew certain inferences -which are of prime importance to students of plant-dispersal. In his -time the order included nearly 10,000 known species, and although this -number has since no doubt been considerably increased, it is not likely -that his main conclusions, in so far as they are free from purely -hypothetical considerations, will be materially affected by the later -discoveries. - -Accepting the antiquity of the order, and regarding it as probably -dating far back in geological time, he observes that the evidence points -to a very wide dispersion of its original stock at an early period. -Africa, West America, and possibly Australia, possessed the order at the -earliest recognisable stage. There must have existed, he contends, at -this early period some means of reciprocal interchange of races between -these regions. Then followed a stoppage of communication, or a -suspension of the means of dispersal, between the tropical regions of -the Old and New Worlds; but long after communication was broken off in -the warmer regions, it still existed, as he holds, between the alpine -heights in those regions and also between the high northern latitudes of -both hemispheres. Referring particularly to the Hawaiian Group, he -considers that the large endemic element among the Compositæ indicates -that the ancient connection, whether with America or with Australasia, -has been so long severed as not to have left a single unmodified common -form. Fitchia, the Tahitian genus, as we have already remarked, is -regarded as the only remnant of an ancient Composite flora in the -tropical islands of the South Pacific. - -In the light of these reflections it will be interesting to glance at -the general distribution of the shrubby and arborescent or woody -Compositæ. Mr. Hemsley, having generally discussed the subject, arrived -at the conclusion that, “although they form so large a proportion of the -floras of St. Helena, Juan Fernandez, the Sandwich Islands, and some -other islands, they are not specially insular.” There are scores of -them, he goes on to say, in South America, Africa, Madagascar, India, -Australia, and New Zealand from twenty to forty feet high, and more -truly arboreous than the insular ones; whilst nearly every sub-order has -its arboreous representatives. He was, however, unable to form any -definite opinion of the method of distribution of the woody Compositæ. -Taking those of St. Helena and Juan Fernandez, he observes that they are -not more closely allied to the Compositæ of the nearest continents than -they are to those of more distant regions. The occurrence of arboreous -Compositæ, belonging in each case to different tribes, in so many remote -oceanic islands, coupled with the distribution of the genera to which -they bear the greatest affinity, seems, he observes, to indicate that -they are the remains of very ancient types (_Introd. Bot. Chall. -Exped._, pp. 19-24, 66, 68; also Parts ii. p. 61, and iii. p. 23). - -The further discussion of this subject would lead us into a wide field -of inquiry, quite beyond the scope of this work. There is, however, an -inference that I think we may legitimately draw from geological evidence -in this region. With respect to the antiquity of the woody Compositæ of -the Pacific as illustrated by the endemic genera, both Mr. Bentham and -Mr. Hemsley view them as belonging to ancient types. Mr. Wallace, in his -_Island Life_, a book that becomes more and more indispensable for the -student of dispersal as years progress, dwells on the importance of -these ancient Compositæ in the floral history of the Pacific islands. We -may look upon the Hawaiian Compositæ, he remarks, as representing the -most ancient portion of the existing flora, carrying us back to a very -remote period when the facilities for communication with America were -greater than they are now. The date of this period of oceanic dispersal -of the Compositæ we can now approximately determine, since these plants -are absent from the Fijian region, an area of submergence during the -Tertiary era. Before the island-groups of the Fijian region had emerged -towards the close of the Tertiary period the achenes of the early -Compositæ had been dispersed far and wide over the tropical Pacific. - -But this is not all that we can infer from the convergence of these -independent lines of botanical and geological investigation. Mr. Bentham -observes that the tribes of the Compositæ had acquired the essential -characters now employed in classification before the dispersion of the -order over the Pacific. Since this general dispersion took place, as we -hold, during the Tertiary submergence of the island-groups of West -Polynesia (Fiji, Tonga, Samoa), it follows that the birth of the tribes -of the Compositæ antedates that period. If this interesting order could -supply us with a “datum-mark” in the history of the Pacific floras, it -would be stated in terms of the development of specific and generic -characters, but not of those of a tribe. - - - _Summary of Chapter._ - -(1) The Hawaiian Islands present the same contrast with the Fijian and -Tahitian groups as regards the development of new species in the case of -the flowering plants that they offer in the case of the vascular -cryptogams (ferns and lycopods). But the contrast is intensified, and it -is further emphasised as respecting the flowering plants by the -evolution of a large number of endemic genera. - -(2) This great preponderance of peculiar species and genera in Hawaii is -not to be connected with the relative antiquity of the group but with -its degree of isolation. - -(3) The earliest stage of the flowering plants of the islands of Hawaii -and of Eastern Polynesia (the Tahitian region) is indicated by the -endemic genera, particularly those of the Compositæ and Lobeliaceæ. Such -genera are numerous in Hawaii, and occur also in the Tahitian region, as -in Tahiti and Rarotonga; but do not exist in the groups of the Fijian -region (Fiji, Tonga, and Samoa). - -(4) The endemic genera of the Hawaiian Compositæ are mainly American in -their affinities. The relationship of the solitary Tahitian genus -(Fitchia) is still a subject of discussion. - -(5) In the Hawaiian Islands, as well as in Tahiti and Rarotonga, the -plants of the endemic genera of Compositæ are, as a rule, arborescent or -shrubby; and in the first two localities they are mainly restricted to -the higher levels. - -(6) In discussing the mode of dispersal of the achenes of the original -genera we have also to explain why the process of dispersal has been in -the main suspended. - -(7) It is shown that the achenes of these early Compositæ were in all -probability suited for dispersal in birds’ plumage. - -(8) Yet the isolating influence that cut off these genera from the -outside world has, in later ages, been active within the limits of the -Hawaiian archipelago, with the result that half the species are not -found in more than a single island. Inter-island dispersal has, -therefore, been also largely suspended. - -(9) The absence of endemic genera of Compositæ from Fiji, Tonga, and -Samoa cannot be attributed to unsuitable climatic conditions connected -with the relatively low elevation of those islands as contrasted with -those of Hawaii, since a species of Fitchia abounds in Rarotonga, which -is not far over 2,000 feet in elevation. Shrubby and arborescent -Compositæ of peculiar types also occur in the Galapagos and other -tropical islands not more elevated than the Fijis. - -(10) These endemic genera are the remains of an ancient Composite flora -in the islands of the tropical Pacific, and ages have elapsed since the -severance of their connections with regions outside. - -(11) According to Mr. Bentham the Compositæ were distributed over -Africa, West America, and possibly Australia, at an early period, but -subsequent to the differentiation of the tribes of the order. Some means -of reciprocal interchange of races between these regions then existed. -Then followed a suspension of the means of dispersal between the -tropical regions of the Old and New Worlds except between the alpine -heights of those latitudes. - -(12) It is inferred by the author of this volume that the general -dispersion of the early Compositæ over the Pacific took place during the -Tertiary submergence of the island-groups of the Fijian region (Fiji, -Tonga, and Samoa), and that their absence from that region may be thus -explained. At the time of this general dispersion, as above pointed out, -the tribes of the Compositæ had been already differentiated. - - - - - CHAPTER XXII - - THE ERA OF THE ENDEMIC GENERA (_continued_) - - THE COMPOSITÆ AND LOBELIACEÆ (_continued_) - - THE AGE OF THE TREE-LOBELIAS - -The distribution of the arborescent Lobeliaceæ.—On the upper flanks of - Ruwenzori.—The Lobeliaceæ of the Hawaiian Islands.—The Lobeliaceæ of - the Tahitian or East Polynesian region.—The capacities for - dispersal.—The explanation of the absence of the early Lobeliaceæ from - West Polynesia.—The other Hawaiian endemic genera.—The Fijian endemic - genera.—Summary. - - -THE Lobeliaceæ rank with the Compositæ in the prominence of their -position in the early Pacific floras. Though absent, as far as is known, -from Fiji, they are represented in Hawaii by 58 species, all endemic and -belonging to six genera, of which five are not found elsewhere. All -possess, as Hillebrand remarks, a woody stem, by far the greater number -being either tall shrubs, 5 or 6 feet high, or small trees, 10 to 20 -feet or more in height. In the East Polynesian or Tahitian region, the -order is represented by two genera containing in all five known species -and restricted to those islands. One genus is common to the islands of -Tahiti and Rarotonga, and the other is confined to Raiatea. The species -may be shrubby or arborescent. - -It was for some time considered that the oceanic archipelagoes of the -Pacific were the exclusive centres of these singular arborescent -Lobeliaceæ (I am here quoting Baillon in his _Natural History of -Plants_). And indeed this idea would receive some support from the -circumstance that Dr. Hillebrand, in his work on Hawaii, says little or -nothing about the affinities or general relations of plants which he -enthusiastically termed “the pride of our flora.” His death in 1886 -deprived his work of its crowning piece, a discussion of “the -interesting questions of the origin and development of the Hawaiian -flora” (see the Editor’s Introduction, p. ix.). In no group of plants is -this want more keenly felt than with the Lobeliaceæ. Yet in his time the -explorations had yet to be made that could set the student of -plant-distribution on the road to investigate this problem. - -It was true, no doubt, that types analogous to those of the Hawaiian -Lobeliaceæ were known from the American and African continents. Thus -Oliver in his _Flora of Tropical Africa_, published in 1877, gives an -account of the species of Lobelia then known from the mountains of this -region. The genus was, however, not entirely confined to mountainous -districts, but it would almost seem that most of the high mountains of -Equatorial Africa had their peculiar species, some of them being -tree-like and others shrubby. Two mountain species were recorded from -Abyssinia, one of them from an elevation of 11,000 to 13,000 feet and -growing to a height of 12 to 15 feet, the other from an altitude of -about 8,000 feet; another, Lobelia Deckenii, attaining a height of 4 -feet, was recorded from the uplands of Kilimanjaro, 12,000 to 13,000 -feet above the sea, and yet another from the mountains of Fernando Po, -at an altitude of 9,000 feet. So again, in the case of the American -continent, Hemsley, writing in 1885 (_Intr. Bot. Chall. Exped._, p. 32), -speaks of arborescent species of the American genera Centropogon, -Siphocampylus, &c.; and Baillon in his _Natural History of Plants_ -(Engl. edit. viii. 350) refers to the similar Tupas and Haynaldias from -South America. But what the student of plant-distribution looked for was -not merely the occurrence of “tree-lobelias” in other parts of the -world, but also the reproduction of these wonderful plants under the -same conditions and on the same scale as those familiar to him on the -Hawaiian mountains. He has accordingly had to wait for the results of -the more recent explorations of the mountains of Central Africa in order -to obtain his wish. - -On the upper flanks of Ruwenzori, Kilimanjaro, and Kenya, at elevations -of 9,000 to 13,000 feet and reaching to the snow-line, there flourish in -boggy portions of the forest arborescent Lobeliaceæ that attain a height -of 15 or 20 feet. They have the habit sometimes of a Dracæna and -sometimes of an Aloe, and do not exhibit the branching trunks so -characteristic of the Hawaiian genus of Clermontia. They all belong, -however, to the genus Lobelia, and thus do not display the extensive -differentiation of the endemic genera of Hawaii. Nor, apparently, has -there been the same degree of formative energy in the development of -species, since only about half a dozen species are hitherto known. We -find, however, produced on these lofty mountains of Equatorial Africa -the same climatic conditions under which the arborescent Lobeliaceæ -flourish in Hawaii, namely, the very humid atmosphere, the heavy -rainfall, and the mild temperature; and if there are important contrasts -in their character and in the amount of differentiation which they have -undergone in the two regions, the one a continental and the other an -insular region, it will be from such contrasts that some of the most -interesting results of this comparison of a mountain of Central Africa -with an island of the open Pacific will be ultimately derived (see Sir -H. Johnston’s _Uganda Protectorate_, 1902, and _Kilimanjaro Expedition_, -1886; also _Trans. Linn. Soc. Bot._, ser. 2, vol. 2, p. 341.) - - - THE LOBELIACEÆ OF THE HAWAIIAN AND OF THE EAST POLYNESIAN OR TAHITIAN - ISLANDS.[1] - - HAWAIIAN ISLANDS. - - +-----------+-----+-------------+-------------+--------------------------------------+ - |Genus. | N s | Distribution| Distribution| Height of | Nature of Station. | - | | o p | of genus. | in the | plant. +-------------+-----------+ - | | . e | | group. | | Elevation. | Station. | - | | c | | | +-------------+-----------+ - | | o i | | | | | | - | | f e | | | | | | - | | s | | | | | | - | | . | | | | | | - +-----------+-----+-------------+-------------+------------+-------------+-----------+ - |Brighamia | 1 | Endemic. | Molokai, | 5 to 12 | Islands not | Steep | - | | | | Niihau. | feet. | exceeding | palis or | - | | | | | | 3,500 feet. | mountain | - | | | | | | | gaps. | - | | | | | | | | - |Lobelia | 5 | Non-endemic.| General. | 4 to 6 | 2,000 to | Bridges, | - | | | | | feet. | 6,000 feet. | gulches | - | | | | | | | and woods.| - | | | | | | | | - |Clermontia | 11 | Endemic. | General. | Usually 10 | 2,000 to | Open | - | | | | | to | 6,000 feet. | woods. | - | | | | | 20 feet.[2]| | | - | | | | | | | | - |Rollandia | 6 | Endemic. | Oahu. | Usually 4 | Higher parts| Woods. | - | | | | | to 6 feet, | of Oahu, | | - | | | | | one species| which is | | - | | | | | 10 to | 4,000 feet | | - | | | | | 15 feet. | high. | | - | | | | | | | | - |Delissea | 7 | Endemic. | General. | 5 | 1,000 to | Woods and | - | | | | | to 10 feet.| 5,000 feet. | gulches. | - | | | | | | | | - |Cyanea | 28 | Endemic. | General. | Usually | 1,000 to | Woods, | - | | | | | 6 to | 5,000 feet. | ravines, | - | | | | | 15 feet.[3]| | gulches. | - +-----------+-----+-------------+-------------+------------+-------------+-----------+ - - - EAST POLYNESIAN OR TAHITIAN ISLANDS. - - +-----------+-----+-------------+-------------+------------+-------------+-----------+ - |Sclerotheca| 4 | Endemic in |{Tahiti, | 6 to 25 | 1,500 to | Humid | - | | | E.Polynesia.|{Rarotonga. | feet. | 3,000 feet. | wooded | - | | | | | | | slopes. | - | | | | | | | | - |Apetahia. | 1 | Endemic. | Raiatea. | 3 to 6 | In the | | - | | | | | feet. | mountains. | | - | | | | | | Elevation of| | - | | | | | | island 3,400| | - | | | | | | feet. | | - +-----------+-----+-------------+-------------+------------+-------------+-----------+ -Footnote 1: - - The materials are nearly all derived from the works of Hillebrand and - Drake del Castillo. Some of those relating to the elevations in Hawaii - are supplemented from my notes. All the genera are endemic except - Lobelia, of which all the species are apparently endemic, excepting - perhaps one, which, according to Hillebrand, resembles greatly a - species from the Liukiu Islands. - -Footnote 2: - - The range of the heights of different species of Clermontia is from 5 - or 6 feet for shrubs to 25 feet for trees. - -Footnote 3: - - The heights attained by different species of Cyanea range from 3 or 4 - feet to between 30 and 40 feet, thus:— - - In 8 species 3 to 6 feet. - In 9 species 6 to 10 feet. - In 7 species 10 to 15 feet. - In 3 species 15 to 25 feet. - In 1 species 30 to 40 feet. - - - THE LOBELIACEÆ OF THE HAWAIIAN ISLANDS. - -Having thus prepared the way, I will proceed to the discussion of the -Hawaiian Lobeliaceæ, dealing first with their “station.” Their vertical -distribution is well illustrated in the large and lofty island of -Hawaii. Whilst the woody Compositæ, as before described, are most at -home on the open-wooded and often scantily-forested slopes between 5,000 -and 9,000 feet, the Lobeliaceæ are most characteristic of the middle or -true forest zone that extends from 2,000 or 3,000 feet to between 5,000 -and 6,000 feet above the sea. This lies within the region of clouds and -mists, and it is here that the rain-belt or area of greatest rainfall is -situated, the annual amount averaging probably 150 to 200 inches. It is -in such humid conditions that, as Hillebrand observes, trees and jungle -are developed in greatest luxuriance; and it is here that “the -Lobeliaceæ exhibit their most striking forms.” The traveller, as he -ascends the mountains, finds the Tree-Lobelias in the region of mist and -rain-cloud; and he is lucky if he escapes the usual downpour and -encounters only a fine drizzling rain. - -The mild climate of this region is indicated by a mean annual -temperature ranging probably with elevation from 65° to 55° F. It is -secure from the frosts of the upper slopes of the mountain; whilst at -the same time it is above the regions of tropical heat. There is, -however, no doubt that when the forests extended to the coasts, as they -occasionally do now on the north side of Hawaii, the Lobeliaceæ occurred -much lower down than they do at present, though still only attaining -their greatest development in size and number in the higher levels. -Thus, at rare intervals, I noticed in the forests of Hamakua and Kohala, -where they descended to the coasts, species of Clermontia at an -elevation of only 500 or 600 feet above the sea. - -Probably in no part of the Hawaiian Islands are the conditions under -which the “Tree-Lobelias” thrive better illustrated than on the higher -slopes of Mount Eeka, a bulky mountain mass about 6,000 feet in height, -forming the western portion of Maui. Its flat top, as Hillebrand -observes, is wrapped in a cloud of mist nearly the whole year. On the -boggy surface of the summit, where Acæna exigua gives a tussocky -appearance, and Sphagnum or bog-moss abounds, flourish Cyperaceæ, -Lycopods, and Selaginellæ; and here Drosera longifolia and a peculiar -species of marsh violet (Viola mauiensis) find a home. The upper slopes, -down to 4,000 feet, present similar moist conditions, and here in an -open-wooded district, associated with Cyrtandræ, Marattias, and true -Tree-Ferns, the ground being covered with Lycopods, the “Tree-Lobelias” -abound. I noted four kinds within two hundred yards. Of the humidity of -the upper slopes of Mount Eeka I have a very vivid recollection, and my -experience of passing a night on that mountain is described in Chapter -XIX. - -The Lobeliaceæ, as Hillebrand remarks, occur invariably as isolated -individuals. I was often struck, however, with the preference the genera -showed for particular localities. Thus, Clermontia is well represented -on the western slopes of Mount Eeka, Delissea on the northern slopes of -Hualalai (3,800 to 4,500 feet), Cyanea on the Hamakua slopes of Mauna -Kea (2,300 to 4,100), and Lobelia on the southern slopes of Mauna Loa -behind Punaluu (2,000 to 3,500 feet). - -To the student of geographical distribution the Hawaiian Lobeliaceæ are -of especial interest. Mr. Hemsley observes that they have their greatest -affinities in America (_Intr. Bot. Chall. Exped._, p. 68). M. Drake del -Castillo, in his “Mémoire couronné par l’Académie des Sciences” (Paris, -1890), remarks that these plants connect Hawaii with America just as the -Goodeniaceæ link the same group with Australia. This is what we might -have expected since the centre of the order is in America, principally -in the Mexican and Andine regions (Drake del Castillo, _Flore Polyn. -Franc._, xi.). - -Though five out of the six genera are endemic, the sixth, that of -Lobelia, has a world-wide distribution. Here then, we have a genus that -belongs strictly to the next or second stage of the plant-stocking of -the Hawaiian Group, namely, when the non-endemic genera now containing -endemic species were introduced. As with the Composite genera, -Campylotheca and Lipochæta, Lobelia marks the beginning of the new or -the close of the old era. It is, however, necessary to point out that -many of the conditions favouring luxuriant and rank vegetable growth are -pre-eminently represented in the zone of the Lobeliaceæ. In these -soft-stemmed plants with their copious milky sap and large fleshy -flowers, sometimes two or three inches long, the very redundancy of -growth would tend both to exaggerate and to disguise the generic -distinctions. To the ordinary observer these “Tree-Lobelias” call up -vague notions of a flora of a bygone age, and by their _bizarre_ -appearance he might with some excuse be led to give play to his -imagination when describing them; but the systematic botanist, seeing -through their disguise, frames rather more prosaic notions of their -antiquity and degree of differentiation. According to my view, the first -Hawaiian Lobeliaceæ occupied open, exposed localities such as are held -by the decadent genus Brighamia now, and acquired their monstrous form -in the humid forests of a later age. (See Perkins in Note 80.) - -In his monograph on the Campanulaceæ (Engler’s _Nat. Pflanz. Fam._, teil -4, abth. 5, 1894), S. Schönland, speaking of the sub-family Lobelioideæ, -places the seven endemic Hawaiian and Tahitian genera in a group by -themselves. Though, as he observes, the Hawaiian tree-forms appear at -first sight to constitute a natural group, they cannot be sharply -distinguished from other forms, and even in habit come near some Indian -and Abyssinian types of Lobelia. In their treatment, he says, they -should all go together, and he does not approve of the endeavours of -some botanists to isolate one of them (Brighamia) from the rest and to -connect it with the Australian genus Isotoma. - -It is also to be noted that whilst four of the Hawaiian genera are more -or less dispersed over the group, one (Brighamia) with only one species -is confined to the islands of Molokai and Niihau, the double habitat -being suggestive of its approaching extinction. Another (Rollandia) with -six species is restricted to the island Oahu. Cyanea, which possesses -twenty-eight out of the total of fifty-eight species, may, from the -point of view of its formative energy, be regarded as in its prime. It -is thus apparent that, as with the Compositæ, the early Lobeliaceous -immigrants were not all contemporaneous arrivals. - - - DISTRIBUTION OF THE LOBELIACEÆ IN THE HAWAIIAN ISLANDS.[4] - -Column headings: - - A: Brighamia. - B: Lobelia. - C: Clermontia. - D: Rollandia. - E: Delissea. - F: Cyanea. - G: Total. - - +----------------------------------+----+----+----+----+----+----+----+ - | Hawaiian Lobeliaceæ. | A | B | C | D | E | F | G | - +----------------------------------+----+----+----+----+----+----+----+ - | Species confined to one island | — | — | 6 | 6 | 4 | 22 | 38 | - | Species confined to two islands | 1 | 2 | 2 | — | 2 | 5 | 12 | - | Species confined to three islands| — | 1 | 2 | — | 1 | 1 | 5 | - | Species generally distributed, | | | | | | | | - | but still endemic | — | 2 | 1 | — | — | — | 3 | - | +----+----+----+----+----+----+----+ - | | 1 | 5 | 11 | 6 | 7 | 28 | 58 | - +----------------------------------+----+----+----+----+----+----+----+ -Footnote 4: - - All the species are endemic. - -Another interesting fact of distribution, brought out by an analysis of -Hillebrand’s materials and illustrated in the subjoined table, is that -out of the fifty-eight Hawaiian species, all of which are endemic, -thirty-eight, or 66 per cent., are recorded from only one island. In -most of the other cases they are recorded from two or three islands, -usually adjacent, like Maui and Molokai; and except in the instance of -two species of Lobelia and one species of Clermontia they never range -over the length of the group. - -These facts speak eloquently of the suspension to a great extent of the -agencies of dispersal in recent times within the group. Some corrections -of the figures will be rendered necessary by future investigations, but -the main conclusion will not be materially affected. Such facts are -paralleled in the distribution of the Hawaiian insects, mollusca, &c.; -but these matters need only be mentioned here. We might, indeed, have -expected, apart from other considerations, that the isolation of the -Hawaiian Lobeliaceæ from their kindred in other parts of the world would -not have been reproduced within the group itself. This, however, is not -the case; and we now see that not only have they been deprived for ages -of their means of distribution over the Pacific, but that even within -the archipelago their transportal from island to island has been largely -suspended. We have before arrived at similar conclusions with regard to -the early Compositæ, when we saw that about half the species were not -found in more than one island. It is therefore evident that the same -great principle regulating the operations of the distributing agencies -has influenced to a similar extent both the Compositæ and the Lobeliaceæ -of the Hawaiian Group. - - - THE LOBELIACEÆ OF THE TAHITIAN OR EAST POLYNESIAN REGION. - -The order is represented in this region by two endemic genera, -Sclerotheca of Tahiti and Rarotonga, and Apetahia of Raiatea. These -islands are, however, not sufficiently large for the extensive -development of the arborescent Lobeliaceæ, such as we find in Hawaii. -The species in both genera are either arborescent or shrubby; but I do -not gather that they give any character to the floras of these islands. -According to the data given by Drake del Castillo for one of the two -peculiar species of Sclerotheca occurring in Tahiti, these plants grow -on the humid wooded slopes of the mountains at elevations of 2,000 to -3,000 feet. Whilst in one species the plants attain a height of 10 to 25 -feet, in the other they do not exceed 10 feet. Rarotonga possesses a -peculiar species of Sclerotheca, 4 to 6 feet high, which was discovered -by Cheeseman growing plentifully on the upper slopes of the highest -mountain of the island at altitudes of 1,500 to 2,200 feet. The same -botanist also came upon a second species of the genus on another -mountain in Rarotonga at elevations of 1,000 to 1,500 feet, but it was -rare and has not yet been described. The other genus, Apetahia, has only -been recorded from Raiatea, where it is represented by a solitary -species (6 feet high) growing, according to Nadeaud, in the mountains of -that island. - -It is apparent that the dispersal of these genera of the Lobeliaceæ -amongst the groups of Eastern Polynesia ceased long ago. From the -circumstance that Sclerotheca exists in Tahiti and in Rarotonga, which -are about 650 miles apart, it may be inferred either that the genus was -introduced into this region from outside, or else, which is perhaps more -probable, that it was developed in Tahiti whence it was transported to -Rarotonga. Hemsley speaks of this Tahitian genus as seemingly marking a -former wide extension of the Hawaiian arborescent type of the Lobeliaceæ -(_Introd. Bot. Chall. Exped._, p. 68). This is the view that will be -adopted in this chapter, and it is precisely the view advocated by -Bentham and followed here, in the case of the early Compositæ of the -Pacific. - -With regard to the absence of these arborescent Lobeliaceæ from the -island-groups of the Western Pacific, and notably from Fiji and Samoa, -where no members of the order seem to occur, it is probable that, as in -the case of the similar distribution of the early Compositæ described in -the preceding chapter, this is to be attributed to the fact that the -Western Pacific archipelagoes were more or less submerged during the -general dispersion of the Compositæ and Lobeliaceæ over the Pacific in -the earliest age of the floral history of these islands. The occurrence -of the early Compositæ and Lobeliaceæ in Rarotonga, which is almost -half-way between Tahiti and Tonga on the outskirts of the Fijian region, -sufficiently indicates that they are not lacking in that region from -inability to reach there in the past. During the age of general -dispersal of these two orders over the Pacific, probably only a few -rocky islets, tenanted perhaps by Conifers, marked the situation in the -Tertiary period of the present archipelagoes of Fiji and Samoa. - -One may note in passing the general absence of these arborescent types -of the Lobeliaceæ from Malaya, since they do not seem to have been -recorded either from the Owen Stanley Range in New Guinea or from -Kinabalu in North Borneo, the highest mountain in the Malayan Islands, -or from the mountains of Java. - -The consideration of the occurrence of these plants in other tropical or -subtropical oceanic islands need not detain us long, since, with the -exception of the solitary Lobelia scævolifolia of St. Helena, they seem -rarely to be found. This species, which is endemic, is a shrub growing -on the upper slopes and summit of the island at elevations of 2,000 to -2,700 feet (_Introd. Bot. Chall. Exped._, p. 40, and Part ii. pp. 54, -76). - -There are two herbaceous species of Lobelia in Juan Fernandez, of which -one only, according to Hemsley, could be regarded as indigenous. This is -a showy Chilian and Peruvian species (Lobelia tupa) noticed by Bertero -as very common in 1829 (_Bot. Chall. Exped._, Part iii.). Since, -however, it would belong to the present age of plant-dispersal in the -Pacific, it does not require further mention here; and indeed it would -almost appear, when we bear in mind the geographical position and the -history of this island since its discovery in 1563, that even as a truly -indigenous plant it is not above suspicion. Lobelias of this type are -now amongst the commonest plants of the coast regions of northern Chile, -where I noticed some as much as 9 or 10 feet high. - -_On the Capacities of Dispersal of the Lobeliaceæ of the Pacific._—Of -actual observations, with the exception of the instance of birds pecking -at the capsules of our garden Lobelias, I have come upon few that bear -directly on this point. When writing of the flora of the Kermadec Group, -many years ago, Sir Joseph Hooker referred (_Journ. Linn. Soc. Bot._, -vol. i.) to the minute seeds of Lobelia as not adapted for transport -unless their minuteness and number fit them for it; but since he -associates in this connection the tiny seeds of Metrosideros, which is -now represented by a species found all over the Pacific, it would seem -that the difficulty in the case of Lobelia is not connected so much with -the nature as with the suspension of these means of distribution during -the later stages of the plant-stocking of the oceanic islands of the -tropical Pacific. It will be gathered from the following remarks that -the descendants of the early Pacific Lobeliaceæ are probably as well -fitted for dispersal as their ancestors, and that the break in the -communication is the ultimate subject for inquiry. - -The fruits of the Hawaiian endemic genera are in four out of five cases -baccate, with usually fleshy or pulpy contents. Such berries, which are -generally yellow, but sometimes bluish in colour, vary in size from -about half an inch in Rollandia and Delissea to an inch in Cyanea, and -not infrequently to more than an inch in Clermontia. The fruits of -Lobelia and Brighamia are capsular and dehiscent. With regard to the two -genera of the Society Islands and Rarotonga, the fruits of Sclerotheca -are hard-walled capsules, opening by two pores; whilst those of Apetahia -are seemingly dry and indehiscent. I do not imagine, therefore, that the -character of the fruit has determined to any important degree the -distribution of these plants. - -Nor is there reason to suppose that the fruits have acquired their -baccate character in Hawaii, and that they were originally dry and -capsular. Both types of fruit are found among the arborescent Lobeliaceæ -of America, with which the Hawaiian genera have their affinities. -Centropogon, for instance, which occurs in Central America and in the -warm parts of North and South America, has, according to Baillon, a -somewhat fleshy berry. It is noteworthy that a similar question is -raised with respect to Cyrtandra as to the relation between fleshy -fruits in the Pacific islands and dry or capsular fruits in the -continental home of the genus (see Chapter XXV.). - -The berries of the Tree-Lobelias would attract birds. We learn from Mr. -Perkins that one of the Hawaiian Drepanids, the Ou, is very partial to -the berries of some of the Tree-Lobelias and especially those of -Clermontia, the seeds passing unharmed in the droppings. The mode of -dispersal of the seeds of the dry-capsular fruits is not so apparent; -but the fruits could scarcely be less inviting to birds than the dry -capsules of Metrosideros, the small seeds of which have in some way or -other been carried to almost every island-group of the Pacific. I have -beside me the dark brown, smooth crustaceous seeds of a species of -Clermontia. They measure 1/42 of an inch or 0·6 of a millimetre, and -about 500 go to a grain. Mr. Wallace, in his book on Darwinism, -advocates the paramount influence of winds over birds for carrying small -seeds, like those of Orchis and Sagina, over tracts of ocean a thousand -miles across. I am, however, not inclined to think that, except as -regards the spores of cryptogams, winds have done very much for Hawaii. -For small seeds we can appeal not only to the agency of birds and bats -but also to insects (see Chapter XXXIII.). - -Observations of this kind, however, merely indicate that these early -Lobeliaceæ possessed the same capacities for dispersal that in the -succeeding stages of the plant-stocking of the Pacific islands have -belonged to Metrosideros, Cyrtandra, Ophiorrhiza, Freycinetia, and many -other small-seeded genera. They go no way to explain why the same -agencies which transported the minute seeds in a later age could not -have been available for continuing the dispersal of the early -Lobeliaceæ. To find an explanation we are compelled to go behind the -mere capacities for dispersal and to appeal to the general laws of -distribution in so far as our facts enable us to interpret them. - -We have seen that the two principal components of the early Pacific -flora, the Compositæ and the Lobeliaceæ, have American affinities. The -plants of the later ages are mainly Old World in their connections. -Though containing often endemic species in the various groups, the -genera occur also outside each group. The stream of migration that came -from America during the early age of the Compositæ and the Lobeliaceæ, -when the islands of the Western Pacific were more or less submerged, was -during the later ages (after these islands had re-emerged) suspended or -diverted, giving place to a stream that brought plants in numbers from -tropical Asia, Malaya, and Australia. The general dispersion of the -Compositæ and Lobeliaceæ took place during the Tertiary submergence of -the islands of the Western Pacific, including the island-groups of Fiji, -Samoa, and Tonga. The migration from the west, mainly Indo-Malayan in -character, occurred after the re-emergence of those archipelagoes. Thus -we get to understand how genera like those of the early Lobeliaceæ and -Cyrtandra, which possess, as regards the minute size of their seeds, -closely similar capacities for dispersal, have such different -distributions, the first confined to Hawaii and Tahiti and American in -their affinities, the second widely spread over the Pacific with its -home in Malaya. - -We have yet to inquire whether this suspension of the means of transport -in the later ages of the Pacific Lobeliaceæ is confined to the tropics -or whether it extends to the colder latitudes in the southern -hemisphere. The indications of the Lobeliaceæ of the “antarctic flora” -go to establish that the dispersal of the order is still, or was very -recently, in operation in these high latitudes. It is well illustrated, -among other plants, by Lobelia anceps, which is found in extra-tropical -South America, Australia and New Zealand, and South Africa. This, -indeed, recalls Bentham’s view concerning the Compositæ, that whilst -communication was broken off in the tropics, it was kept up in higher -latitudes. - -Here ends, therefore, our consideration of the Tree-Lobelias of the -Pacific islands; but as it is not quite complete without a discussion of -the remaining endemic genera of other orders than the Compositæ and -Lobeliaceæ which also belong to the same early age of the Pacific -floras, I will proceed at once to their consideration. - - - THE HAWAIIAN ENDEMIC GENERA EXCEPTING THOSE OF THE COMPOSITÆ AND - LOBELIACEÆ. - -It will not be possible for me to do more than point out a few general -indications that can legitimately be drawn from these genera. The -subject bristles with difficulties for the systematist; but on one point -there can be but little danger of going astray, namely, in imputing to -them a high antiquity in the floral history of Hawaii. This can be said -of all of them, whether or not the generic distinction adopted in Dr. -Hillebrand’s work is always adopted by botanists. It is therefore in -this general sense that they may be regarded as belonging to the early -age of the Hawaiian flora. - -Although the genera of Compositæ and Lobeliaceæ are prominent amongst -the representatives of the original flora of the Hawaiian Islands, -forming about two-fifths of the whole, the genera of other orders are by -no means inconspicuous, and their variety is shown in the fact that -though twenty-three in number they belong to twelve orders. It is -possible to divide these genera into two groups—one the older and -perhaps more or less contemporaneous with the Lobeliaceæ and Compositæ, -the affinities when apparent being American; the other the more recent -and marking the close of the first era of the plant-stocking of these -islands, the affinities being all with the Old World, and especially -with Malaysia. This grouping is indicated in the list subjoined; and it -may be here remarked that whilst shrubs, undershrubs, and perennial -herbs of the Caryophyllaceæ, Labiatæ, and Urticaceæ form the features of -the earlier group, trees of the Rubiaceæ and Araliaceæ are the most -conspicuous members of the later group. At the close of the earliest era -known to us of the floral history of the Hawaiian Islands we observe the -commencement of those forests that now throughout Polynesia as well as -in Hawaii betray their Asiatic origin. - -In making this distinction I am proceeding on the assumption that the -stream of migration, at first chiefly American in its source, came -ultimately in the main from the Asiatic side of the Pacific. The change -commenced, as I hold, in the latter portion of the first era of -plant-stocking, an era characterised by the arrival of those early -plants that are now represented by the endemic genera of the -archipelago. The genera of this early period that belong neither to the -Compositæ nor to the Lobeliaceæ are, as above observed, arranged by me -in two groups, one regarded as contemporaneous with, the other as of -later origin than, the genera of these two orders. To the first belong -the shrubby, highly differentiated genera of the Caryophyllaceæ, -Schiedea and Alsinidendron, and the Labiate genera, similarly -differentiated, of Phyllostegia and Stenogyne. To the second belong the -Rubiaceous genera Kadua, Gouldia, Bobea, and Straussia, the Araliads -Cheirodendron, Pterotropia, and Triplasandra, and the Loganiaceous -Labordea. - -In the earlier group the fruits are dry in half the genera, and in such -cases granivorous birds probably were usually the transporting agents. -Only in one case (Nothocestrum) is the fruit a berry, and in the other -cases we have fruits like the fleshy nucules of Phyllostegia and -Stenogyne which would probably attract birds. In the later group -two-thirds or three-fourths of the genera have moist fruits such as -would be eaten by frugivorous birds. Of these most are drupes, -possessing not a single stone, but two or more pyrenes. This is the -first appearance of the drupe in the plant-history of the archipelago. -The Rubiaceous type of drupe inclosing two or more pyrenes plays a very -conspicuous part in the distribution of plants over the Pacific in the -succeeding eras. - -I would here lay stress on an important characteristic of all the fruits -of the endemic genera of the Hawaiian Islands. There are no “impossible” -fruits of this era in Hawaii, such as we occasionally find in the -succeeding eras. I mean by this term, fruits that defy the efforts of -the student of distribution to explain their transport in their present -condition. The discovery of a new inland genus possessing dry -indehiscent fruits three or four inches long, or even of a single -species of the coniferous Dammara, would play havoc with all our views -respecting the stocking of these islands with their plants. The finding -here of a large marsupial would scarcely produce more astonishment. The -fruits indeed of this early era are very modest in their size, the dry -indehiscent fruits and the stone-fruits rarely exceeding half an inch -(12 mm.) in size. - -There is another interesting point which is connected with the -deterioration of some of the fruits in their capacity for dispersal. -Some of the species of Phyllostegia, and a few also of the Araliads, as -well as those of Nototrichium, are ill fitted for dispersal by birds -now, the coverings of the seeds being not sufficiently hard to protect -them from injury in a bird’s stomach. At the same time there are in some -cases other species of the same genera that are better suited for this -mode of transport. The effect of dispersal by frugivorous birds is that -only the hard-coated seeds propagate the plant in a new locality. When, -however, as has occurred in the Hawaiian Islands, bird-agency largely -ceases to act, this selective influence is removed (see Note 68). - - -ENDEMIC HAWAIIAN GENERA, EXCLUDING THOSE OF THE COMPOSITÆ AND - LOBELIACEÆ, AS GIVEN IN HILLEBRAND’S “FLORA OF THE HAWAIIAN ISLANDS.” - - [Those preceded by * are not usually regarded now by botanists as - endemic, though they nearly - take that rank.] - - THE EARLIER GROUP. - - +-----------------+---------------+---------+------------+-----------------+-----------------------------+ - | Genus. | Order. |Number of|Character. | Fruit. | Affinities. | - | | | species.| | | | - +-----------------+---------------+---------+------------+-----------------+-----------------------------+ - |Isodendrion | Violaceæ. | 3 |Shrubs. |Capsule. |American (H). | - | | | | | | | - |Schiedea |Caryophyllaceæ.| 17 |Undershrubs,|Capsule. {|Near Colobanthus of the | - | | | | &c. | {| Antarctic islands, temperate| - |Alsinidendron |Caryophyllaceæ.| 1 |Undershrubs.|Capsule, with {| South America, | - | | | | | fleshy calyx. {| and Australia (C). | - |Platydesma | Rutaceæ. | 4 |Small trees |Capsule. | — | - | | | | or shrubs. | | | - |Hillebrandia | Begoniaceæ. | 1 |Herbs. |Capsule. | — | - |Nothocestrum | Solanaceæ. | 4 |Small trees.|Berry. |South American (H). | - |*Haplostachys | Labiatæ. | 3 |Herbs. |Dry nucules. |Regarded by Gray as a | - | | | | | | section of Phyllostegia. | - |*Phyllostegia | Labiatæ. | 16 |Undershrubs.|Fleshy nucules. {|Belong to the tribe Prasiæ, | - | | | | | {| which is mostly Asiatic. | - |Stenogyne | Labiatæ. | 17 |Trailers or |Fleshy nucules. {| Two other species of | - | | | | climbers. | {| Phyllostegia recorded | - | | | | | {| from Tahiti and Paumotu | - | | | | | {| Islands. | - |Charpentiera | Amarantaceæ | 2 |Trees. |Utricle. |American (H). | - |Touchardia | Urticaceæ. | 1 |Shrubs. |Achene with | — | - | | | | | fleshy perigone.| | - |Neraudia | Urticaceæ. | 2 |Shrubs. |Achene with |Allied to Bœhmeria, a | - | | | | | fleshy perigone.| genus of Old and New | - | | Worlds. | - | THE LATER GROUP. | - | | | | | | | - |*Pelea | Rutaceæ. | 20 |Trees. |Capsular. |Belongs to Melicope, an | - | | | | | | Old World genus (IK). | - |Broussaisia | Saxifragaceæ. | 2 |Small trees.|Berry. |Malayan (H). | - |*Cheirodendron | Araliaceæ. | 2 |Trees. |Drupe. |Referred to Panax, an | - | | | | | | Old World genus (IK). | - |*Pterotropia | Araliaceæ. | 3 |Trees. |Drupe. {|Malayan (H). | - | | | | | {| Pterotropia referred to | - |Triplasandra | Araliaceæ. | 4 |Trees or |Drupe. {| Heptapleurum of Old | - | | | | shrubs. | {| World (IK). | - |Kadua | Rubiaceæ. | 16 |Shrubs, &c. |Capsular |Approaches both Asiatic | - | | | | | | and American types (C). | - |Gouldia | Rubiaceæ. | 5 |Small trees |Drupaceous |American (C). | - | | | | or shrubs. | berry. | | - |*Bobea | Rubiaceæ. | 5 |Small trees.|Drupe. |Malayan (H). Genus | - | | | | | | also in Malaya (IK). | - |Straussia | Rubiaceæ. | 5 |Trees. |Drupe. |Near Psychotria, a genus | - | | | | | | of Asia and America (H). | - |Labordea | Loganiaceæ. | 9 |Small trees |Capsule with |Malayan (H). | - | | | | or shrubs. | pulp. | | - |*Nototrichium | Amarantaceæ. | 3 |Trees or |Utricle. |Referred to the Australian | - | | | | shrubs. | | Ptilotus (IK). | - +-----------------+---------------+---------+------------+-----------------+-----------------------------+ - - (H) = Hillebrand’s _Flora of the Hawaiian Islands_. - (C) = Drake del Castillo’s _Remarques sur la Flore de la Polynésie_. - (IK) = _Index Kewensis._ - -NOTE.—Probably Schumann’s genus, Pteralyxia, should be placed in the -later group (see p. 154). - -Another feature of interest is to be found in the distribution within -the archipelago of the species of the peculiar genera. As in the case of -the Compositæ and Lobeliaceæ, but few of the species are generally -distributed, most being restricted to one island or to two or three -adjacent islands. The suspension of the dispersal among the islands is, -however, not so marked as with the species of the two orders just named. - -NOTE.—Some further remarks on some of these genera are given in Note 68. - - - THE ENDEMIC GENERA OF THE FIJIAN ISLANDS. - -The interest that is associated with the endemic genera of Hawaii fails -to attach itself to those of Fiji. For this there are several reasons. -In the first place, our acquaintance with the Fijian flora is much less -complete. In the next place, the group holds a much less isolated -position, and the history of an endemic genus may have a significance -quite different from that connected with it in Hawaii. Fiji also lacks, -on account of its submergence in the Tertiary period, those highly -interesting genera of the Compositæ and Lobeliaceæ that form the chief -feature in the early history of the flowering plants of Hawaii. Then, -again, on account of our imperfect knowledge of the floras of the -neighbouring groups of continental islands to the westward, the New -Hebrides, Santa Cruz, and Solomon Groups, we can never feel quite -confident that any particular genus is really peculiar to the Fijian -archipelago. This is well brought out in the later history of the genera -designated by Dr. Seemann in his _Flora Vitiensis_ as peculiar to Fiji. - -Of the sixteen genera enumerated by Dr. Seemann, and given in the table -below, only about half now retain their character of being restricted to -Fiji. Nor does it seem likely that future investigations will increase -this number, since, judging from a remark made by Mr. Hemsley in his -paper on the botany of the Tongan Group, explorations subsequent to -those of Dr. Seemann, more especially those of Mr. Horne, have not -apparently added a single new endemic genus to the Fijian flora. It will -be seen from the list that at least four of the sixteen genera have -since been found in the Malayan region, and in one case (Smythea -pacifica) the same species occurs in both regions; whilst a fifth genus -(Haplopetalon) has been recorded from New Caledonia. - -There are, however, some peculiarities about the Fijian endemic genera -that will attract our attention from the standpoint of dispersal. One -remarkable feature is the paucity of species. Almost all the genera are -monotypic, that is to say, they are only known by a single species. -Amongst the twenty-eight Hawaiian genera that are strictly endemic, only -four or five are monotypic, and they are mostly regarded by Hillebrand -as worn-out, decadent types found in only one or two islands. In Hawaii -there are on the average six species to each endemic genus; and it is -thus apparent that in the display of formative energy Nature has worked -on very different lines in these two groups. Since the nine Fijian -endemic genera belong to nearly as many different orders, the -composition of this endemic generic flora is by no means homogeneous. It -is, I venture to think, such a motley collection as one might expect in -a region that has been exposed to wave after wave of migration from the -west, with no lofty mountains, as in Hawaii, to afford a refuge against -extinction. It by no means follows that all these endemic genera have -been produced in Fiji. Some of them may represent genera that have -become extinct in the large continental groups to the westward. - - - SEEMANN’S SIXTEEN FIJIAN ENDEMIC GENERA. - - +--------------+------------------+---------+-----------+-----------+-------------------+ - | Genus. | Order. |Number of| Character.| Fruit. | Affinities or | - | | |species. | | | other localities. | - +--------------+------------------+---------+-----------+-----------+-------------------+ - |Richella |Anonaceæ. | 1 |Tree. |Baccate(?).|Indian in type (C).| - | | | | | | | - |Trimenia |Ternstrœmiaceæ. | 1 |Tree. |Unknown. | | - | | | | | | | - |Pimia |Sterculiaceæ. | 1 |Tree. |Small |Related to | - | | | | |spinose | Australian genera | - | | | | |capsule. | (S). | - | | | | | | | - |Græffea |Tiliaceæ. | 1 |Tree. |Unknown. |Near Trichospermum,| - | | | | | | a Fijian and | - | | | | | | Malayan genus (S).| - | | | | | | | - |Thacombauia |Humiriaceæ. | 1 |Shrub. |Drupe. |Order mainly South | - | | | | | | American. | - | | | | | | | - |Amarouria |Simarubeæ. | 1 |Tree. |Dry drupe. |Near Soulamea, a | - | | | | | | Malayan genus (S).| - | | | | | | | - |*Smythea |Rhamneæ. | 1 |Straggling |Capsule. |Also in Burma, New | - | | | |shrub. | | Guinea, and Malaya| - | | | | | | (IK), (Sc). | - | | | | | | | - |*Oncocarpus |Anacardiaceæ. | 2(H) |Tree. |Drupe. |Also in New Guinea | - | | | | | | (IK). | - | | | | | | | - |*Haplopetalon |Rhizophoreæ. | 2 |Shrub. |Unknown. |Also in New | - | | | | | | Caledonia (IK). | - | | | | | | | - |*Nesopanax |Plerandreæ. | 1 |Tree. |Drupe. |=Plerandra (IK). | - | | | | | | | - |Bakeria |Plerandreæ. | 1 |Tree. |Drupe. | | - | | | | | | | - |Pelagodendron |Rubiaceæ. | 1 |Shrub. |Berry. | | - | | | | | | | - |*Paphia |Ericaceæ. | 1 |Shrub. |Berry. |=Agapetes, a | - | | | | | | Malayan genus | - | | | | | | (IK). | - | | | | | | | - |*Carruthersia |Apocyneæ. | 2(H) |Climber. |Berry. |Also in | - | | | | | | Philippines (IK). | - | | | | | | | - |*Couthovia |Loganiaceæ. | 2 |Tree. |Drupe. |Also in Kaiser | - | | | | | | Wilhelmsland, | - | | | | | | New Guinea (So). | - | | | | | | | - |Canthiopsis |Loganiaceæ. | 1 |Shrub. |Drupe. | | - +--------------+------------------+---------+-----------+-----------+-------------------+ - -Those genera marked * have since been found outside the group. - -The authorities are thus indicated: (C)=Drake del Castillo; (H)=Horne; -(IK)=_Index Kewensis_ (S)=Seemann; (Sc)=Schimper; (So)=Solereder in -Engler’s _Nat. Pflanz. Fam._ - -The fact that several of them are fitted for dispersal by frugivorous -birds is very suggestive of the lack of means of transport in later -times. In the instance of Couthovia corynocarpa the drupes are known to -be the food of fruit-pigeons at the present time (Seemann), whilst this -is also true of Oncocarpus vitiensis, though this genus has since been -found in New Guinea. Since, as will be pointed out in a later chapter, -birds must still be fairly active in carrying seeds to Fiji from regions -westward, it would seem that genera only become peculiar to Fiji when -they fail at their source, and it is indeed doubtful whether any of the -Fijian peculiar genera are home productions. One may instance in this -connection the genus Pimia, the fruits of which are especially well -suited for attachment to a bird’s plumage, yet it is only known from -Fiji. - -It should be here observed that no peculiar generic types have been -recorded from the adjacent Tongan Group, and scarcely any from Samoa. -Except perhaps with the Palmaceæ, no peculiar genera seem to be -mentioned in Dr. Reinecke’s memoir on Samoa. - - - _Summary._ - -(1) The Lobeliaceæ, like the Compositæ, take a prominent place in the -early Pacific flora, being represented, more particularly in Hawaii but -also in the East Polynesian or Tahitian region, by endemic genera of -tall shrubby and tree-like species. - -(2) Tree-Lobelias occur in other parts of the world, as in South America -and tropical Africa; but it is especially on the higher slopes of the -mountains of Equatorial Africa that they attain a development comparable -with that of Hawaii. - -(3) In Hawaii the Tree-Lobelias are most characteristic of the middle -forest-zone (3,000-6,000 feet), where the temperature is mild, the -rainfall heavy, and the atmosphere laden with humidity. - -(4) The affinities of these endemic genera of the Lobeliaceæ are mainly -American; but their generic distinctions have been both exaggerated and -disguised by redundant growth. - -(5) From the distribution of the genera and species within the Hawaiian -Group it is evident that, as with the early Compositæ, the original -Lobeliaceous immigrants were not all contemporaneous arrivals. Some of -the genera are on the point of extinction, whilst others are in their -prime. - -(6) The absence of the Lobeliaceæ from the groups of the Fijian area -(Fiji, Tonga, Samoa) is probably to be connected, as in the case of the -absence of the early Compositæ, with the circumstance that the general -distribution of these two orders over the tropical Pacific occurred -during the Tertiary submergence of these archipelagoes. - -(7) These endemic genera of the Lobeliaceæ possess the same facilities -for dispersal that are owned by other genera with minute seeds, such as -Cyrtandra, &c., that are dispersed over the Pacific; but in the case of -the Lobeliaceæ the agencies of dispersal have been for ages suspended. - -(8) This suspension is to be associated with the diverting of the main -stream of migration from its source in America, during the early age of -the Lobeliaceæ and Compositæ, to a source on the Asiatic side of the -Pacific. - -(9) The Hawaiian endemic genera other than those of the Compositæ and -Lobeliaceæ arrange themselves in two groups—an earlier group containing -highly differentiated Caryophyllaceæ and Labiatæ, and belonging to the -age of the Compositæ and Lobeliaceæ; and a later group, characterised by -Rubiaceæ and Araliaceæ, which marks the close of the first era, as well -as the change in the main source of the plants from America to the Old -World, the beginning of the Hawaiian forests, the appearance of the -Rubiaceous drupe, and the first active intervention of frugivorous -birds. - -(10) Though there are no “difficult” or “impossible” fruits (fruits, the -dispersal of which is not easy to explain) amongst the forty and odd -endemic genera of Hawaii and Tahiti, it is noteworthy that in some cases -the fruits are seemingly little fitted for dispersal now, and that this -deterioration in capacity for dispersal is to be frequently associated -with more or less failure of the inter-island dispersal in the case of -Hawaii. - -(11) The interest associated with the Hawaiian endemic genera fails to -attach itself to those of Fiji, where genera only seem to have become -peculiar because they have failed at their sources in the regions to the -west. The endemic genera of the Compositæ and Lobeliaceæ are here -lacking, and this is true also of the neighbouring Samoan and Tongan -Groups, it being held that the age of the general dispersion of these -orders over the Pacific corresponded with the Tertiary submergence of -the archipelagoes of the Western Pacific. Those of Fiji, which do not -amount to ten in number, belong to nearly as many orders and present a -motley collection such as one might look for in a group much less -isolated than Hawaii and exposed to wave after wave of migration from -the west. - - - - - CHAPTER XXIII - - THE ERA OF THE NON-ENDEMIC GENERA OF FLOWERING PLANTS - - THE MOUNTAIN-FLORAS OF THE PACIFIC ISLANDS AS ILLUSTRATED BY THE - NON-ENDEMIC GENERA - -The mountain-flora of Hawaii.—A third of it derived from high southern - latitudes.—An American element.—Compared with Tahiti and - Fiji.—Capacities for dispersal of the genera possessing only endemic - species.—Acæna, Lagenophora, Plantago, Artemisia, Silene, Vaccinium, - &c.—Capacities for dispersal of the genera possessing non-endemic - species.—Cyathodes, Santalum, Carex, Rhynchospora.—Fragaria chilensis, - Drosera longifolia, Nertera depressa, Luzula campestris.—Summary. - - - THE AGE OF THE ENDEMIC GENERA OF FLOWERING PLANTS. - -WE are now entering an era distinguished from the preceding age of the -endemic genera, the age chiefly of the Compositæ and Lobeliaceæ, by the -fact that the extreme isolation that followed that era no longer -prevails. In a sense these island-floras are in touch again with the -world around, though the main stream of plant-migration now comes from -the south and from the west. Yet in a large number of cases, the amount -varying greatly in the different groups, it is evident that this stream -has not flowed continuously to the present day. The agencies of -dispersal are often no longer active; but the period of inactivity has -not been sufficiently prolonged to produce generic distinction, and the -differentiating energy has been restricted to the development of new -species. - -Yet within these limits the development of new forms, as indicated in -Table B on p. 233, has often been very great. Thus, nearly half the -Hawaiian genera that are non-endemic are composed entirely of species -not found outside the group; and in this sense they may be regarded as -cut off from the regions around. In Fiji and Tahiti only about a fourth -are in this manner isolated, the agencies of dispersal being still -effective with the majority of the genera. It is apparent, therefore, -that the same question concerning the cause of the failure of the means -of dispersal presents itself in this era as in the last, and most -markedly in the instance of Hawaii. - -The simplest and quickest plan for bringing into relief the prominent -features of this age is first to regard the genera from the standpoint -of the elevation of their stations. We have before remarked that in the -occurrence of extensive regions of great altitude the Hawaiian Islands -differ conspicuously from the groups of Tahiti and Fiji (and I may add -Samoa); and that they present conditions for the development of a -temperate mountain-flora that are not found at all in Fiji and are -barely represented in Tahiti. That the Hawaiian flora responds to this -contrast between the elevations of the three groups is well established; -and I will now proceed to refer more in detail to the subject. - - - THE MOUNTAIN-FLORAS OF THE PACIFIC ISLANDS. - -In the Hawaiian Islands there are at least 37 or 38 genera, making up -about 19 or 20 per cent. of those belonging to this era, that may be -designated mountain genera, nearly all of them being characterised as -appertaining exclusively or in the main to temperate regions, or as -frequenting mountain-tops in tropical latitudes. In Tahiti there are -only 7 or 8 of such genera, about 4 per cent. of the total for the era. -In Fiji, excluding the Conifers, there are only 4 or 5, or not 2 per -cent. of the whole. In Samoa, which may be included in the Fijian area, -there are 3, or about 2 per cent. of the total. These are results which -we might have expected from the varying altitudes of these groups, as -described in Chapter XIX. - -Few things give more pleasure to the botanist than his recognition in -some remote locality of plants long familiar to him in other regions. -This will often be his lot on the mountain summits of Hawaii. If he has -been a mountain-climber in many countries, he will there notice again -the genera Artemisia, Geranium, Plantago, Ranunculus, Rubus, Sanicula, -Vaccinium, and others that he has met perhaps either in the Rocky -Mountains or in the Andes or in Equatorial Africa or in the Himalayas. -If fresh from Chile he will find on these heights the familiar Gunnera -and the Chilian Strawberry (Fragaria chilensis). If he has been in New -Zealand and in the islands of the Southern Ocean he will find old -friends in the genera Acæna and Coprosma. He may handle once again -plants like Nertera depressa, that he gathered on Tristan da Cunha; and -on the boggy summits of some of the mountains he will find the -ubiquitous Sun-dew (Drosera longifolia). - -Within the limited area occupied by the peaks of Tahiti he will find -genera like Astelia and Coprosma that are at home in New Zealand or in -Antarctic America, and may even find, as in the cases of Coriaria -ruscifolia and Nertera depressa, the identical species that are at home -in those distant regions. Even on the summit of Rarotonga he will gather -a species of Vaccinium. In Fiji, here and there on some isolated -mountain-top he may come upon a remnant of this Antarctic flora, such as -a solitary species of Coprosma or Lagenophora, that will carry him back -for a moment to high southern latitudes; and in the highlands of Savaii, -in the neighbouring Samoan Group, he will find again Nertera depressa -and a species of Vaccinium. But that which will interest him most in -Fiji will be the tall conifers of the genera Dammara, Podocarpus, and -Dacrydium, which will bring to him memories perhaps of New Zealand and -southern Chile, of South Africa, and of the mountain-woods of Java and -of Southern Japan. - -Yet the influence of isolation has been at work amongst the -mountain-plants of all these groups. The agencies that have dispersed -over the tropical Pacific plants from the cold latitudes of the southern -hemisphere, and those that have borne the seeds of Plantago, Sanicula, -and Vaccinium from mountain-top to mountain-top, even though it be to a -peak in mid-ocean, are to a great extent inactive now. - - - THE MOUNTAIN-FLORA OF HAWAII AS ILLUSTRATED BY THE NON-ENDEMIC GENERA. - -Let us look in the first place at Hawaii, where the breaking off of -communication with the outside world is especially pronounced. Here, all -the species of two-thirds or more of the mountain-genera are confined to -that group. Only in a relatively small number of cases are the species -in touch with the regions outside. The mystery of disconnection that is -so evident in the instance of the peculiar or endemic mountain-genera of -the Compositæ and Lobeliaceæ and other orders is here again presented to -us, and once more in the upland regions 4,000 to 10,000 feet above the -sea. We will now endeavour to discover from an examination of the -present distribution of the isolated mountain-genera (those non-endemic -genera possessing only peculiar species) along what tracks they arrived -at the Hawaiian uplands, tracks, as indicated by the local distribution -of the species, that have been more or less abandoned since. - -_The Mountain Genera with only Endemic Species._—By referring to the -Table on the following page it will be observed that nearly a third of -these mountain genera have now their principal homes in the high -latitudes of the southern hemisphere. They are components of what -Forster and Hooker have termed the “Antarctic” flora, a collection of -plants that range round the globe in high southern latitudes, namely, -over Fuegia, New Zealand, southern Australia, South Africa, and the -islands of the Southern Ocean, the “Antarctic” islands, as they have -been termed. These genera are Acæna, Gunnera, Coprosma, Lagenophora, -Astelia, Oreobolus, and Uncinia. (It is necessary to observe that I am -entirely indebted to the Introduction to the _Botany of the “Challenger” -Expedition_ for my information on the “Antarctic” flora.) - -We are thus led to expect that some of the other mountain genera may -have been similarly derived from cool southern latitudes, even though -they may be scarcely included in the “Antarctic” flora. This is very -probably true of Myoporum and Exocarpus, two genera that are chiefly -centred in Australia. A species of Sophora (S. tetraptera) is now one of -the most widely dispersed of the plants of high southern latitudes, a -circumstance which at all events explains the capacity for transport -that the ancestor of the Hawaiian “Mamani” (S. chrysophylla) must have -originally possessed (see Chapter XV.). Kinship between the Hawaiian -species and southern forms has been found in the case of a few of the -widely ranging genera here represented. Thus Decaisne placed Plantago -princeps next to P. fernandeziana of Juan Fernandez; whilst according to -Hillebrand, Plantago pachyphylla resembles P. aucklandica from the -Auckland Islands. These resemblances are consistently associated with -the respective range in altitude of the Hawaiian plants, since Plantago -princeps occurs usually between 2,000 and 4,000 feet, and P. pachyphylla -between 6,000 and 8,000 feet, the species of greatest elevation being -related with the species of highest latitude. It is thus seen that these -endemic mountain genera with peculiar species have very evident -affinities with the plants of extra-tropical southern latitudes, and -especially with the “Antarctic” flora. This affinity will also be found, -as will subsequently be noticed, in the case of genera like Cyathodes -and Nertera, where there is still a specific connection with the outside -world. - - - THE MOUNTAIN-FLORA OF HAWAII, AS REPRESENTED BY THE NON-ENDEMIC - GENERA (Compiled from Hillebrand’s Flora). - - +----------------+------------+---------------+--------------+------------+ - | | | Distribution |Distribution | | - | | | outside | in Hawaii, | | - | | | Polynesia. | Fiji, and | | - | | | | Tahiti. | | - | | Usual |---------------|--------------+ | - | Genus |altitude of |Both Worlds. |Hawaii only. | | - | |station in | |Old World. | |Hawaii, Fiji| | - | | feet. | | |New World. | | |Hawaii, | Fruit. | - | | | | | |Antarctic| | | Tahiti | | - | | | | | | flora. | | | |All | | - | | | | | | |Austra-| | | | three | | - | | | | | | |lia and| | | | groups | | - | | | | | | |New | | | | | | - | | | | | | |Zealand| | | | | | - +----------------+------------+-+-+-+-+-------+-+-+-+--------+------------+ - | | | | | | | | | | | | | - | WITH ALL SPECIES ENDEMIC. | - | | | | | | | | | | | | | - |Ranunculus (2) |6,000- 7,000|+| | | | |+| | | |Achene. | - |Viola (5) |2,000- 6,000|+| | | | |+| | | |Capsule. | - |Silene (4) |2,000- 9,000|+| | | | |+| | | |Capsule. | - |Geranium (6) |5,000-10,000|+| | | | |+| | | |Carpels. | - |Vicia (1) |7,000- 8,000|+| | | | |+| | | |Pod. | - |Sophora (1) |5,000-10,000|+| | | | | | | | + |Pod. | - |Rubus (3) |4,000- 7,000|+| | | | | |+| | |Berry. | - |Acæna (1) |5,000- 6,000| | | |+| |+| | | |Spinose | - | | | | | | | | | | | | achene. | - |Gunnera (1) |3,000- 6,000| | | |+| |+| | | |Drupe. | - |Sanicula (1) |6,000- 8,000| | |+| | | | | | + |Prickly | - | | | | | | | | | | | | carpel. | - |Coprosma (9) |3,000- 9,000| | | | | + | | | | + |Drupe. | - |Lagenophora (1) | 6,000- | | | |+| |+| | | |Viscid | - | | | | | | | | | | | | achene. | - |Artemisia (2) |4,000- 8,000|+| | | | |+| | | |Achene. | - |Lobelia (5) |2,000- 6,000|+| | | | |+| | | |Capsule. | - |Vaccinium (2) |3,000- 8,000|+| | | | | | | | + |Berry. | - | | | | | | | | | | | Samoa | | - |Myoporum (1) |Coast-10,000| | | | | + |+| | | |Drupe. | - |Plantago (2) |2,000- 8,000|+| | | | |+| | | |Capsule. | - |Exocarpus (2) |3,000- 6,000| | | | | + |+| | | |Fleshy nut. | - |Sisyrinchium (1)|4,000- 7,000| | |+| | |+| | | |Capsule. | - |Astelia (2) |2,000- 6,000| | |+| | | | | | + |Berry. | - |Oreobolus (1) | 6,000 | | |+| | |+| | | |Toothed | - | | | | | | | | | | | | nutlet. | - |Uncinia (1) |3,000- 5,000| | | |+| |+| | | |Awned | - | | | | | | | | | | | | nutlet. | - |Agrostis (3) |4,000- 6,000|+| | | | |+| | | |Awned grain.| - |Deschampsia (3) |3,000- 6,000|+| | | | |+| | | |Awned grain.| - |Trisetum (1) |3,000- 5,000|+| | | | |+| | | |Awned grain.| - |Poa (2) | |+| | | | |+| | | |Grain. | - | | | | | | | | | | | | | - | WITH ENDEMIC AND NON-ENDEMIC SPECIES. | - | | | | | | | | | | | | | - |Cyathodes (2) |2,000-10,000| | | | | + | | |+| |Drupe. | - |Lysimachia (6) |Coast- 6,000|+| | | | |+| | | |Capsule. | - |Chenopodium (2) |Up to 7,000|+| | | | |+| | | |Seed-like. | - |Santalum (3) |Coast-10,000| |+| | | | | | | + |Drupe. | - |Carex (5) |2,000- 7,000|+| | | | | |+| | |Nutlet. | - |Rhynchospora (4)|Up to 10,000|+| | | | | | | | + |Nutlet. | - |Panicum (14) |Coast- 6,000|+| | | | | | | | + |Grain. | - |Deyeuxia (3) |Up to 10,000|+| | | | |+| | | |Awned grain.| - | | | | | | | | | | | | | - | WITH NO ENDEMIC SPECIES. | - | | | | | | | | | | | | | - |Fragaria | | | | | | | | | | | | - | chilensis |4,000- 6,000| | |+| | |+| | | |Fleshy. | - |Drosera | | | | | | | | | | | | - | longifolia | 4,000 |+| | | | |+| | | |Capsule. | - |Nertera depressa|2,500- 5,000| | | |+| | | | | + |Drupe. | - | | | | | | | | | | | Samoa | | - |Luzula | | | | | | | | | | | | - | campestris |3,000-10,000|+| | | | | | |+| |Capsule. | - +----------------+------------+-+-+-+-+-------+-+-+-+--------+------------+ - -It is evident that in one or two cases the connection between the -representatives of the “Antarctic” genera on the Hawaiian uplands and -those of high southern latitudes has only been recently broken off. Thus -with reference to the Hawaiian species of the Cyperaceous genus, -Uncinia, it may be observed that although Hillebrand regards it as a -distinct species, Hemsley (_Intr. Bot. Chall. Exped._, p. 31) remarks -that it is very near if not the same as a New Zealand species, an -affinity very significant of the source of the mountain plants of this -group that are derived from these southern latitudes. - -The next component to be recognised in these Hawaiian mountain genera -with peculiar species is a small special American element; and in this -connection Sanicula and Sisyrinchium may be especially mentioned. The -first is mainly North American, and particularly Californian; but there -are two solitary species found on the continents and in oceanic islands -such as the Azores. The continental species, Sanicula europæa, occurs -not only in Europe and Central Asia, but in South Africa, and at high -elevations on the mountains of Equatorial Africa and of Madagascar. It -is not, however, with this widely ranging species that Sanicula -sandwicensis is related, but with S. menziesii, a species from -California and Oregon (Hillebrand). Sisyrinchium is confined to -temperate and tropical America; but a singular and suggestive outlier of -the genus (S. bermudiana) is found in Bermuda. - -The mountain genera that are distributed on both sides of the Pacific -constitute about three-fifths of the total. So far as my scanty data -show, they seem to have reached Hawaii from the four quarters of the -compass. The probable southerly origin of Plantago has been already -indicated. Hillebrand notes the great resemblance between Lobelia -gaudichaudii and an undescribed species from the Liukiu Islands, lying -on the west side of the Pacific. It is likely, also, that the genus -Ranunculus reached Hawaii from the west, since one of the species, R. -mauiensis, resembles R. repens of the Old World (Hillebrand); whilst the -other, R. hawaiiensis, comes near R. sericeus of Mauritius (Drake del -Castillo). On the other hand, the genus Rubus may hail from an American -source, since, in the opinion of Gray, Rubus hawaiiensis, one of the -mountain raspberries, finds its nearest relative in R. spectabilis from -the north-west coast of America; and there are reasons for believing, as -will subsequently be shown, that the genus Artemisia has an American -source. It is also probable that some of these genera have reached -Hawaii from the north, since it is likely, as pointed out in a later -page, that the Carices of the Hawaiian uplands came originally from -north-eastern Asia. - -In the previous paragraphs the mountain genera have been considered with -especial reference to their distribution and source beyond the confines -of the Pacific. If we now briefly discuss them from the standpoint of -their distribution within the Pacific, or rather as concerning their -presence or absence in the Fijian and Tahitian groups, we shall see that -to a large extent Hawaii has received its mountain genera of this era -independently of the other Pacific groups. - - Mountain genera possessing only peculiar species, in Hawaii only 20 - Mountain genera possessing only peculiar species, in Hawaii and Fiji 2 - Mountain genera possessing only peculiar species, in Hawaii and Tahiti 0 - Mountain genera possessing only peculiar species, in all three groups 4 - — - 26 - -It is here shown that three-fourths of the genera of the Hawaiian -mountains in this era are not found either in Fiji or Tahiti. This, as -before pointed out, is mainly to be attributed to the greater elevation -of the Hawaiian Islands. Had there been an island 13,000 to 14,000 feet -in height in Fiji, we cannot think that any such contrast in the floras -would have existed. The temperate genera of the Hawaiian uplands would -have been largely represented in the Fijian flora. Yet although we do -not find such genera as Ranunculus, Geranium, Sanicula, Uncinia, &c., in -Fiji and Tahiti, a small number of the Hawaiian mountain genera have -obtained a scanty footing. This is what we might have expected. Thus, -Lagenophora has been found on the mountains of Vanua Levu, and Vaccinium -in Tahiti and Rarotonga; whilst Coprosma and Astelia occur on the tops -of some of the mountains in both regions. In Fiji their distribution -seems sporadic, as shown not in Lagenophora alone, but also by Astelia, -which has been found only on the summit of Kandavu. - -_The Capacities for Dispersal of the Hawaiian Non-endemic Mountain -Genera possessing only Peculiar Species._—As shown in the Table, seven, -or 27 per cent., of these genera have fleshy fruits that would attract -frugivorous birds. In three cases (Gunnera, Coprosma, Myoporum) they are -drupes, in three others (Rubus, Vaccinium, Astelia) they are berries, -and in one (Exocarpus) there is a nut with a fleshy perigone. It is -particularly interesting to notice that frugivorous birds, and I include -here granivorous birds that are known to be frugivorous at times, could -have transported seeds of the “Antarctic” flora to this group. We can -observe the process in operation in our own time within the limits of -the group. It has been long known, and we find it referred to in the -pages of Hillebrand’s work, that the wild mountain-goose (Bernicla -sandwicensis) feeds upon the fruits of Coprosma ernodeoides, and of -Vaccinium reticulatum, the famous “ohelo.” The fruits of the first are -known to the natives as “kukai neenee” (droppings of geese), and the -hard stones or pyrenes are very well suited for withstanding the risks -of the digestive process. I found a number of these pyrenes in the -stomach of a mountain-goose shot by my companion, Dr. Krämer, high up -the slopes of Mauna Loa. - -According to Mr. Perkins, Chloridops kona, a big Hawaiian finch, feeds -on the fruits of the bastard sandal-tree (Myoporum sandwicense). There -are no “impossible fruits” among the mountain genera of Hawaii, that is -to say, fruits so large that bird agency must be excluded. All of them -are practicable in point of size. Thus amongst the largest, the “stones” -of Gunnera would not exceed 1/5 of an inch (5 mm.), and those of -Myoporum scarcely 1/4 of an inch (6 mm.); whilst the nuts of Exocarpus -range in the Hawaiian species from 3/10 to 6/10 of an inch (7-15 mm.), -and the beans of Sophora chrysophylla do not at the most exceed 1/4 of -an inch (6 mm.). - -The principal feature, however, which these mountain genera exhibit from -the point of view of their dispersal is the number of plants possessing -seeds or fruits capable of adhering to plumage. Half of these genera are -thus characterised. Of these Sanicula and Acæna represent the ordinary -hooked fruits; whilst the fruits of the Grasses and Sedges, Agrostis, -Deschampsia, Trisetum, Poa, Oreobolus, and Uncinia, are enabled by means -of their awns or of their serrated beaks to attach themselves to -plumage, and the same may be said of the carpels of Geranium. The fruits -of Lagenophora and the seeds of Plantago display the capacity of -adhesiveness by means of a gummy secretion. - -One or two of these genera need further mention. I will first take -Acæna, which is spread all over the south temperate zone both on the -continents and on the islands. The Hawaiian species (A. exigua) forms -tussocky growths on the swampy summits of Mount Eeka in Maui, and in -Kauai, at an elevation of 6,000 feet above the sea. Numerous observers -refer to the probable mode of dispersal of the genus in the “Antarctic” -and neighbouring islands. Captain Carmichael, in the instance of Acæna -sanguisorbæ on Tristan da Cunha, observes that it overruns the low -ground. Its burr-like fruit, as he describes, “fixes itself on the -slightest touch into one’s clothes, and falling into a hundred pieces -covers one all over with an unseemly crust of prickly seeds not to be -got rid of without infinite labour” (_Trans. Linn. Soc._, xii. 483, -1818). Both Mr. Moseley (Wallace’s _Island Life_, p. 250) and Dr. Kidder -(_Bull. U.S. Nat. Mus._, 2) refer to the burrowing habits of the -Petrels, Puffins, and other sea-birds amongst the vegetation covering -the ground in Tristan da Cunha, Marion Island, Kerguelen, &c., in places -where Acæna, amongst other plants, thrives. Mr. Moseley remarks that the -fruits of this genus stick like burrs to feathers, and he looks to -sea-birds for the dispersal of this and similar plants over the ocean. -He especially notes that the Petrels and other seafowl burrow and breed -high up the mountain-slopes of tropical islands as in Tahiti, Viti Levu, -Hawaii, and Jamaica.... It should be noted in the case of the Hawaiian -endemic species that it has been found only on two mountain tops; and -that however active may be the dispersal of the genus in south temperate -latitudes now, the Hawaiian Islands lie outside the present area of -dispersal. - -The next mountain genus I will specially refer to is Lagenophora, one of -the Compositæ. The solitary Hawaiian endemic species, L. mauiensis, is -restricted to the summit of Mount Eeka, in Maui. In the mountains of -Vanua Levu, Fiji, another peculiar species, L. pickeringii, has been -found; and there is a species, L. petiolata, in the Kermadec Islands -(Hooker, in _Journ. Linn. Soc._, i. 127); but the genus is chiefly -characteristic of Australia, New Zealand, and temperate South America, -one species occurring both in Fuegia and Tristan da Cunha. The genus has -no pappus; but Hooker in the case of the Kermadec species considered -that the “viscid fruit” favoured its dispersal; and this may probably be -true of the genus. - -With regard to the capacity for dispersal of the seeds of Plantago, it -may be pointed out that the seeds of Plantago major, P. lanceolata, &c., -become coated with a mucilaginous material when wetted. In 1892, when -experimenting on these plants, I found that the wetted seeds adhered -firmly to a feather, so that it could be blown about without their -becoming detached. Species of Plantago are so characteristic of the -“alpine” floras of the summits of lofty mountains in the tropics, as in -Java and many other regions, that the mode of dispersal has always been -a subject of curiosity. I cannot myself doubt that this is the -explanation of the occurrence of the representatives of the genus that -now thrive as endemic species on the higher slopes of the Hawaiian -mountains. This method of dispersal for Plantago is recognised by recent -writers on the subject of seed-dispersal. (In a paper in _Science -Gossip_ for September, 1894, I dealt with the “mucous adhesiveness” of -such seeds as a factor in dispersal. The subject had previously been -discussed by Kerner in one of the earlier volumes of his -_Pflanzenleben_; and I have summed up some of the results in Note 43 of -the present volume.) My readers can readily ascertain by a simple -experiment that a bird pecking the fruit-spikes in wet weather would -often carry away some of the sticky seeds in its plumage. Several years -ago, when I was endeavouring to examine the condition of these seeds in -the droppings of a canary, my efforts were defeated by the bird itself, -since, in spite of all my care, some seeds and capsules were always -carried by the bird on its feathers into the clean cage reserved for the -experiment. - -The plants of these mountain genera possessing dry seeds or fruits -neither very large nor very minute and suitable for bird-food are -Ranunculus, Viola, Vicia, Sophora, Artemisia, Sisyrinchium, six in all, -or 24 per cent. of the total. On the probable method of transport of the -ancestors of these endemic species the following remarks may be made. -With regard to Ranunculus, some authors like C. M. Weed -(_Seed-Travellers_, p. 48, Boston, 1899) perceive in the curved or -hooked beaks of the achenes a means of attaching the fruit to plumage. -This no doubt applies to some species, and it is advocated by Ekstam for -some of the plants of the Nova Zembla flora. There are others to which -this explanation would not be applicable, and the achenes of the -Hawaiian species do not appear to be specially fitted for this mode of -transport. I have found the achenes of Ranunculus frequently in the -stomachs of birds in England, in partridges frequently, and in wild -ducks at times. Those of certain species that possess buoyancy are -common in the floating seed-drift of rivers, as of the Thames (_Journ. -Linn. Soc. Bot._, xxix. 333), and they would probably in this way be -often swallowed by waterfowl. - -I have but few data directly relating to the dispersal of seeds of Viola -by birds. From the frequent occurrence of species in alpine floras, as -in the Caucasus, the Great Atlas, in the mountains of Equatorial Africa, -in Madagascar, &c., it may be inferred that birds transport the seeds -between the higher levels of many continental ranges in tropical regions -and to the mountain-slopes of neighbouring large islands. Viola -abyssinica, for instance, which occurs in Madagascar, is spread over the -elevated mountain ranges of tropical Africa. With regard to the five -Hawaiian species, it may be remarked that three of them are bog species -and two occur in dry situations. The first are most characteristic of -the mountains, one species occurring on the summit of Mount Eeka, 6,000 -feet above the sea. Judging from the stations alone, at least two -species were originally introduced into the Hawaiian Group. - -Viola seeds, as indicated by my experiments on the different British -species, including Viola palustris, are not buoyant, and there is no -possibility of the seeds being picked up by birds in floating drift. -There is, however, a possible means of dispersal in birds’ plumage by -means of the mucosity of the seeds of some species. Thus, although this -is not exhibited, as shown by my experiments, by Viola canina and V. -palustris, it is well displayed by the Field-Pansy (V. tricolor). I -found that the seeds of this species, after lying a little time in -water, were thickly covered with mucus, and that they adhered to a -feather, on drying, as firmly as if gummed. This did not, however, come -under my notice in the case of the seeds of one of the Hawaiian species, -V. chamissoniana, examined by me. One sometimes observes Viola canina in -England growing in places, as in the crevices and on the tops of old -walls, where its seeds could have only been carried by birds. In some -cases the propellent force of the seed ejected by the contracting valves -of the capsule would explain queer stations. In its power of -seed-expulsion, Viola chamissoniana, the common Hawaiian species, is -just as active as our British species. - -With regard to the Leguminous genus Vicia we have the observation of -Focke on the dispersal of its seeds by pigeons, as described before on -page 150. - -Sophora chrysophylla, the “Mamani” of the natives and one of the most -familiar of the trees of the Hawaiian mountains, is discussed at length -in Chapter XV., where the difficulty of supposing that its seeds could -be transported unharmed in a bird’s stomach half-way across the Pacific -is pointed out; and it is suggested that it was more probably derived -from a littoral species brought by the currents. However, the point is a -debatable one, and the seeds of the “Mamani” can scarcely be regarded as -“impossible” from the standpoint of dispersal. - -With reference to the possibilities of dispersal of the achenes of -Artemisia, some very suggestive indications are to be obtained from a -paper by Mr. D. Douglas on the North American Tetraonidæ published in -the _Transactions of the Linnæan Society_ for 1833. The “Cock of the -Plains” (Tetrao urophasianus), as we here learn, makes its nest on the -ground under the shade of Artemisia bushes, and lives on the foliage and -fruits of these and other plants. This bird is plentiful in Columbia and -North California, and another allied species is mentioned which lives on -the same sort of food. Later authors refer to these and other birds of -the same family as living chiefly on the Sage-brush (Artemisia -tridentata), a plant prevailing over great regions of the plains as well -as on the slopes of the Sierra Nevada and of the Rocky Mountains. -According to Dr. Sernander (page 228), birds when feeding on the fruits -of Artemisia vulgaris in the district of Upsala scatter them about and -thus aid in its dispersal. Artemisia achenes, since they have neither -pappus nor other appendages, nor any special adhesiveness when wetted, -depend largely on their small size and light weight to aid them in -dispersal. (Those of A. absinthium measure a millimetre in length, or -1/25 of an inch, whilst those of A. vulgaris measure 1·8 mm., or 1/14 of -an inch.) Driven as we are to look to bird-dispersal for the means of -transport of Artemisia achenes, it is interesting to find a possible -source of the Hawaiian endemic species on the nearest American mainland, -even though it is some 2,000 miles away. It is assumed that they would -be ordinarily carried in adherent soil or entangled in the feathers, and -on rare occasions in the bird’s stomach. - -The small seeds of Sisyrinchium possess no means of adherence to -plumage. They are crustaceous, and in cases where the stomach and -intestines of a bird are well filled with other food they are quite -capable of resisting injury. The solitary Hawaiian species has, -according to Hillebrand, a range in altitude from 3,500 to 7,000 feet. I -found this pretty herb most abundant on the “cattle-plains” of Hawaii -between 5,000 and 6,000 feet, where it is evidently in part dispersed by -the cattle and other animals. The seeds are very small, being about a -millimetre in size, and when dried nearly 100 go to a grain (0·65 -decigramme). They might thus also be transported in mud on birds’ feet. - -For the mode of dispersal of the minute seeds of Lobelia, the last of -the mountain genera to be specially noticed, I must refer the reader to -the remarks on this subject in Chapter XXII. They would probably be -carried in soil adhering to the legs or feet of a bird. - -There are one or two interesting points relating to the temperate genus -Silene, which is represented on these mountains. The four Hawaiian -species show a great range in altitude. Thus, whilst S. struthioloides -finds its home in Hawaii and Maui at elevations of 5,000 to 9,000 feet, -another species (S. lanceolata) thrives equally at elevations of 5,000 -or 6,000 feet on the central plateau of Hawaii and at heights only of -300 to 500 feet above the sea. Although I have not yet come upon any -direct reference to the mode of dispersal of the small seeds of this -genus, there is little doubt that their rough tuberculated surfaces -would favour their attachment to plumage. A very significant -observation, however, is made by Jens Holmboe in a paper on littoral -plants in the interior of Norway. He refers to the occurrence in no -small quantity of Silene maritima on the top of “Linnekleppen,” 331 -metres high, one of the highest peaks of Smaalenene, and distant about -29 kilometres from the nearest coast (_Strandplanter i det indre af -Norge_, “Naturen,” Bergen, 1899). Sernander (p. 405), commenting on this -observation, remarks that since bare hill-tops are frequented by birds, -such an agency in this instance is not impossible. - -I will conclude these remarks on the non-endemic Hawaiian mountain -genera possessing only peculiar species, with a few observations on the -genus Vaccinium in the Pacific. This genus is known to be distributed -over the northern hemisphere and to occur on the uplands of tropical -mountains, as, for instance, on the summits of the Java mountains and on -the high levels of the Equatorial Andes at altitudes even of 15,000 to -16,000 feet. There are apparently only some four or five species known -from the Pacific islands, from Hawaii, the Marquesas, Tahiti, Rarotonga, -Samoa, and the New Hebrides, and it would almost seem that these can be -reduced to one or two species. Although not yet recorded from Fiji, the -probability of the genus being represented on some of the mountains is -pointed out by Seemann. Of these Pacific forms a single species, V. -cereum, is spread over the East Polynesian region including the -Marquesas, Tahiti, and Rarotonga; and, according to Hillebrand, V. -reticulatum, one of the two endemic Hawaiian species, is nearly related -to it. Even the New Hebrides species (V. macgillivrayi) resembles it, -according to Seemann, in general appearance. That there has been a -single Pacific polymorphous species is, as shown below, not impossible; -but Reinecke, in describing in 1898 the Samoan species, V. antipodum, -was under the impression that it was the only species known from the -southern hemisphere, and says nothing of its affinity to other Pacific -plants. - -A few words on the station and habit of Vaccinium in the Pacific islands -may be here of interest. In Hawaii there are, according to Hillebrand, -two species, a high-level form, V. reticulatum, occurring at elevations -of 4,000 to 8,000 feet, and a low-level form, V. penduliflorum, ranging -between 1,000 and 4,000 feet. I may, however, remark that the last -species occasionally came under my notice at elevations of 6,000 to -7,000 feet. This species exhibits much variation, and Gray, Wawra, and -other botanists have evidently not been always able to distinguish -between the two species in their varying forms. It is not only -distinguished from the high-level species by its lower station, but also -by its epiphytic habit, a circumstance that, as pointed out below, may -explain some of the differences, since such a habit is bound up with the -difference in station. It seems, therefore, safer to regard them as -station forms of one species which is closely allied to V. cereum, the -species of the South Pacific, an inference which, if well founded, would -make highly probable the view that there has been a single polymorphous -Pacific species.... In Tahiti, as we learn from Nadeaud, V. cereum -occurs on the mountain-tops at altitudes exceeding 800 metres (2,600 -feet). In Rarotonga, according to Cheeseman, it is found on the summits -of most of the higher hills extending almost to the summit of the -island, 2,250 feet above the sea. The Samoan species, V. antipodum of -Reinecke, which that botanist considers as probably one with V. whitmei, -a Polynesian (Samoa?) species originally described by Baron F. von -Müller, grows in the central mountains of Savaii at an elevation of -1,500 metres (4,920 feet). - -These Pacific species of Vaccinium, as on tropical mountains of the -continents, occasionally assume an epiphytic habit, and it is here, as -above observed, that lies one of the distinctions between the Hawaiian -species. V. penduliflorum, the low-level form, occurs typically in the -forests, where, according to Hillebrand, it grows on the trunks of old -trees. The trees, however, may be quite in their prime, and I have -observed it growing in the fork of the trunk of an Olapa tree -(Cheirodendron gaudichaudii). It is in this connection of significance -to notice that a variety found in open glades and on grassy slopes is -described by Hillebrand as terrestrial in habit. The other high-level -form, V. reticulatum, grows gregariously on open ground, and is -typically terrestrial in its habit. The Samoan species, as we learn from -Reinecke, grows on trees, as on the branches of Gardenia. The epiphytic -habit of species of Vaccinium is especially discussed by Schimper in the -case of plants growing on the Java mountains. He there shows -(_Plant-Geography_, i. 14) that species which are epiphytes in the -virgin forest become terrestrial plants in the treeless alpine region. -This interchange of station, which is exhibited by several other plants, -including orchids and ferns, is connected with their xerophilous -characteristics, and is given by Schimper as an example of the -interchange of physiologically dry habitats. - -Of the mode of dispersal of Vaccinium by frugivorous birds, much has -been written and much will be familiar to my readers. The berries of V. -reticulatum are known to be the principal food of the Hawaiian -mountain-goose. But probably birds of the grouse family have been the -chief agents in distributing the genus over the continents. I have -frequently found the fruits in the stomachs of the Black Cock (Tetrao -tetrix), the Scotch Grouse (Lagopus scoticus), and the Capercailzie -(Tetrao urogallus); but the same story comes from all over the northern -hemisphere. The Willow Grouse (Lagopus albus), which travels round the -globe, is known to feed on them. Hesselman in Sweden and Ekstam in Nova -Zembla have especially investigated the dispersal of Vaccinium by Tetrao -tetrix and Lagopus (see Sernander, pp. 6, 226); and according to Mr. -Douglas and others the different species of Tetrao that frequent the -subalpine regions of the Rocky Mountains and the uplands of Columbia and -North California subsist on Vaccinium fruits. This family is not now -represented in the Hawaiian avifauna; but it is noteworthy, as indicated -by the differentiation of the Pacific species of Vaccinium, that -dispersal of the genus is there almost suspended except within the -region of Eastern Polynesia. It is probable that numerous other birds, -except the Hawaiian goose, aided the original dispersal. - -_The Mountain Genera with both Endemic and Non-endemic Species._—I pass -on now to consider those Hawaiian mountain genera that possess species -some of which are confined to the group, whilst others occur in regions -outside the islands. They are not many, as may be seen from the table -before given, and but few of them are entirely restricted to the high -levels, a range in altitude that may be frequently associated with great -lateral extension of the genus over different latitudes. Here the agents -of dispersal have through some species in each genus preserved a -connection with the outer world, though it may be restricted to the -limits of the Pacific islands. - -Cyathodes tameiameiæ, an Epacridaceous species found also in the uplands -of Tahiti, occurs, according to Hillebrand, on all the Hawaiian Islands, -from 1,800 feet up to the limit of vegetation 10,000 feet and over above -the sea. I found it, however, at even lower levels. On the Puna coast of -Hawaii, associated with Metrosideros polymorpha, Osteomeles -anthyllidifolia, and other inland plants, it descends on the surface of -ancient lava-flows to the coast wherever the bolder spurs reach the -sea-border. The other species, C. imbricata, is more exclusively -confined to the greater altitudes. It is endemic, and may possibly be a -station form of the other species. - -The six species of Lysimachia are found at different elevations, one -near the sea-shore, others at altitudes of 2,000 to 3,000 feet, and -others again at elevations of 6,000 feet. Chenopodium sandwicheum occurs -at all elevations from near the coast to the high inland plains of -Hawaii and to the upper slopes of Mauna Kea, that is to say, up to -altitudes of 6,000 or 7,000 feet. Hillebrand observes that it is a low -decumbent plant at the coast, and may become arborescent with a height -of 12 to 15 feet in the upper forests of Mauna Kea. - -The species of Santalum (sandal-wood trees) also display great vertical -range in these islands. Though S. freycinetianum, which is also a -Tahitian species, is most at home in the forests 2,000 to 4,000 feet -above the sea, it has, as Hillebrand informs us, a dwarfed form that -extends far up the mountain slopes of Mauna Loa and Hualalai to -elevations of 7,000 or 8,000 feet, and another dwarfed shrubby variety -that grows only near the sea-shore. Another species, S. haleakalæ, -occurs as a tall shrub on Haleakala at elevations of 8,000 to 10,000 -feet. Among the sedges, most of those of the genera Carex and -Rhynchospora are found at altitudes of between 3,000 and 7,000 feet, and -two grasses of the genus Deyeuxia occur at elevations of 6,000 to 8,000 -feet. - -Amongst these Hawaiian mountain genera with both endemic and non-endemic -species there are no plants possessing fruits which from their size -could be with difficulty regarded as dispersed by birds. The mode of -dispersal of these plants is in some cases indeed not far to seek. Thus -in the stomach of an Hawaiian goose (Bernicla sandwicensis), shot by my -companion Dr. Krämer on the slopes of Mauna Loa, I found a number of the -“stones” of Cyathodes tameiameiæ, the plant being abundant in fruit in -the immediate vicinity. It is highly probable that the seeds of Santalum -have been carried over the Pacific by frugivorous birds. We learn from -Dr. Brandis that Santalum album in India is mainly spread through the -agency of birds (_Bot. Chall. Exped._, iii. 13). The drupes of the -Pacific species, S. freycinetianum, that occurs alike in Hawaii, the -Marquesas, and Tahiti (Drake del Castillo), measure about half an inch. -There can be little doubt that with this tree, as with the species of -Cyathodes above mentioned, which also links together Tahiti and Hawaii, -there has been up to recent times an interchange by means of frugivorous -birds between these two regions, some 2,000 miles apart. - -The small seeds of the capsular fruits of Lysimachia could be -transported in birds’ plumage or in dried soil attached to their feet or -feathers. The seed-like fruits of Chenopodium were probably dispersed by -some granivorous bird, much as nowadays our partridges carry about in -their stomachs the similar fruits of Atriplex. The long-awned fruits of -Deyeuxia were, it is likely, transported in birds’ plumage, and -doubtless also those of Panicum; whilst the nutlets of Carex and -Rhynchospora might have been carried about in a similar fashion. - -The distribution of the non-endemic species of these Hawaiian mountain -genera may perhaps aid us in determining the original source of the -genus as well as in confirming the conclusions formed concerning the -other mountain genera that only possess species restricted to the group. -Lysimachia, Chenopodium, Carex, Rhynchospora, Deyeuxia, and Panicum are -found in both the Old and New Worlds. Since Hillebrand remarks that one -of the six species of Lysimachia (L. spathulata) occurs in Japan and in -the Liukiu, Bonin, and Marianne groups, we have here a valuable -indication of the route followed by a genus that has not been recorded -from the oceanic groups of the South Pacific. - -The capricious distribution of the genus Carex in the Pacific is -remarkable, and it is noticed by Hemsley in the Introduction to the -_Botany of the “Challenger” Expedition_. No species have been recorded -from Tahiti, the Marquesas, and Rarotonga, but three Fijian species are -mentioned by Hemsley, and there is another in Samoa. Of the five -Hawaiian species given by Hillebrand, two are endemic. Of the rest, C. -wahuensis (oahuensis), Meyer, occurs also in Korea and Japan, whilst C. -brunnea, Thunb., is found in Japan and Australia, and the third, C. -propinqua, Nees., occurs all round the border of the Pacific Ocean, from -Kamschatka through Alaska south to the Straits of Magellan. These three -species all possess a home in common in north-east Asia, and probably -there lies the source of the Hawaiian species of Carex—a conclusion -which would help to explain the irregular distribution of the genus -amongst the South Pacific groups. - -The genus Rhynchospora occurs alike in the Hawaiian, Tahitian, and -Fijian islands; but the groups in the North and South Pacific seem to -have been independently supplied with the original species, since R. -aurea, a widely spread tropical species, ranging the South Pacific from -New Caledonia to Tahiti, has not been recorded from Hawaii. A connection -between Hawaii and the Australian region seems to be indicated by a -species of Deyeuxia (D. forsteri) that is found also in Easter Island, -Australia, and New Zealand, and by the presence of the Australian and -New Zealand genus Cyathodes in Hawaii, though the existence of a species -common to both Tahiti and Hawaii goes to show that the route followed by -the genus lay through Eastern Polynesia. It is also not unlikely that -the genus Santalum reached Hawaii through Eastern Polynesia, since two -forms found in Hawaii and Tahiti are closely allied, and are, in fact, -regarded by Drake del Castillo as the same species. The genus occurs in -tropical Asia, Australia, and New Zealand. - -Looking at the indications above given, I should be inclined to think -that the genera Lysimachia and Carex reached the Hawaiian mountains from -temperate Asia or the islands off its Pacific coast, and that Cyathodes, -Santalum, and Deyeuxia hail from the Australian or New Zealand region by -way of Eastern Polynesia. - -_The Mountain Genera possessing no Endemic Species._—The few remaining -mountain plants of Hawaii to be considered are solitary, widely ranging -species of genera that here possess no peculiar species. Such may be -regarded as belonging to the latest age of the indigenous plants. They -still keep up, or kept up until recently, the connection with the world -outside Hawaii, and among them one may name here Fragaria chilensis, -Drosera longifolia, Nertera depressa, and Luzula campestris. - -Fragaria chilensis, the Chilian strawberry, flourishes at elevations of -between 4,000 and 6,000 feet on the Hawaiian mountains. Its fruits, -according to Hillebrand and other authors, are much appreciated by the -wild goose of the islands. This plant ranges in America from Chile north -to Alaska; and Drake del Castillo is doubtless on safe ground when he -assumes that a congener of this bird originally brought the species from -the nearest part of the American continent, namely from California -(_Remarques_, &c., p. 8). In this connection it should be remembered -that one of the endemic mountain-raspberries of Hawaii (Rubus -hawaiiensis) finds its nearest relative, according to Gray, in Rubus -spectabilis, a species from the north-west coast of America. - -The species of Sun-dew, Drosera longifolia, hitherto found only on the -marshy tableland of Kauai at an elevation of 4,000 feet above the sea, -occurs both in Asia and North America. Its minute fusiform seeds are -very light in weight, and might readily become entangled in a bird’s -plumage, or they could be carried in adherent dried mud. - -Luzula campestris, which grows on the high mountains of the Hawaiian -group from 3,000 feet upward, is also found in Tahiti. It is widely -distributed in cool latitudes, and there is no special indication of its -source. Its seeds are especially well suited for adhering to birds’ -feathers. When experimenting on these seeds in 1893 I ascertained that -whether freshly gathered or kept for more than a year they became on -wetting coated with mucus, and adhered firmly to a feather on drying. -There are many ways in which the “sticky” seeds in wet weather might -fasten themselves to a bird’s plumage. The plant-materials might be -used, for instance, for making nests. The Sea Eagle (Aquila albicilla), -as we learn from Mr. Napier (_Lakes and Rivers_), uses materials derived -from Luzula sylvatica in the construction of its nest. - -Nertera depressa, a creeping Rubiaceous plant, with red, fleshy drupes -containing two coriaceous pyrenes, is found in all the Hawaiian Islands -at elevations of 2,500 to 5,000 feet, and it grows on the mountains of -Tahiti at altitudes over 3,000 feet. The genus is widely diffused over -the southern hemisphere. This particular species is characteristic of -the Antarctic flora, being found all round the south temperate zone -(excepting South Africa) in New Zealand, Fuegia, the Falkland Islands, -and Tristan da Cunha, and extending up the Andes to Mexico, occurring -also on the summits of Malayan mountains at elevations of 9,000 to -10,500 feet above the sea, as on Pangerango in West Java (Schimper), and -on Kinabalu in North Borneo (Stapf). Captain Carmichael, who resided on -Tristan da Cunha in the early part of last century, states (_Trans. -Linn. Soc._, xii. 483) that its drupes are eaten by a species of thrush -and by a bunting. Professor Moseley, who visited the island in the -_Challenger_ many years after, remarks that its fruits are “the -favourite food of the remarkable endemic thrush, Nesocichla eremita,” -the bunting being Emberiza brasiliensis (_Bot. Chall. Exped._, ii. 141). -It would seem most likely that the Hawaiian Islands received this -representative of the Antarctic flora through the Tahitian Islands, as -in the case of the species of Cyathodes common to both these groups. - -Looking at the indications of these four widely ranging plants, the -Chilian strawberry (Fragaria chilensis), the Sun-dew (Drosera -longifolia), Nertera depressa, and Luzula campestris, it may be inferred -that with the exception of Nertera they all reached Hawaii from either -the Asiatic or American sides of the North Pacific, the last route being -evident in the case of the strawberry. Nertera depressa was probably -derived from southern latitudes. - - - _Summary._ - -(1) The second era of the flowering plants of the Pacific islands is -indicated by the non-endemic genera. Here also the isolating influences -have been generally active, and the work of dispersal is in some regions -largely suspended. Thus in Hawaii nearly half the non-endemic genera -possess only species that are restricted to the group, whilst in Fiji -and Tahiti about a fourth are thus isolated. - -(2) The contrast in the elevations of the islands of the Hawaiian, -Tahitian, and Fijian regions is reflected in the development of an -extensive mountain-flora in Hawaii, in its scanty development in Tahiti, -and, excluding the Fijian conifers, in a mere remnant in Fiji and Samoa. - -(3) The influence of isolation has been very active in the Hawaiian -mountains, since about two-thirds of the genera contain only species -confined to the group, and are thus disconnected from the world outside. - -(4) Amongst these disconnected Hawaiian mountain genera, Antarctic or -New Zealand genera, like Acæna, Gunnera, Coprosma, and Lagenophora, -constitute nearly a third. The American element, represented, for -instance, by Sanicula and Sisyrinchium, is small; whilst the genera -found on both sides of the Pacific form more than one-half of the total, -and include genera like Ranunculus, Viola, Rubus, Artemisia, Vaccinium, -and Plantago, that often represent the flora of the temperate zone on -the summits of tropical mountains. Three-fourths of these genera are not -found either in Fiji or in Tahiti. - -(5) The proportion of the disconnected Hawaiian mountain genera -possessing seeds or seedvessels suited for dispersal in a bird’s plumage -is very large, quite half belonging to this category; whilst only about -a fourth have fruits that would be dispersed by frugivorous birds. - -(6) The Hawaiian mountain genera that still remain in touch with the -external world through species found outside the islands whilst other -species are confined to the group, present a later stage in the -plant-stocking. Their widely ranging species, which would be dispersed -either by frugivorous birds, as with Santalum and Cyathodes, or in -birds’ plumage, as with Lysimachia, Carex, and Deyeuxia, seem to -indicate that the main lines of migration for these genera have been -from temperate Asia and from the Australian and New Zealand region, the -last by way of Eastern Polynesia. - -(7) The latest stage of the Hawaiian mountain-flora is exemplified by -those genera that are only represented in the group by a solitary -widely-ranging species, such as Fragaria chilensis, Nertera depressa, -Drosera longifolia, and Luzula campestris. It is our own age; and birds -are shown to be actual agents in the dispersal of the two first-named -species and to be probable agents with the two other species. The two -last-named species probably reached Hawaii from one or other side of the -North Pacific; whilst Fragaria chilensis doubtless hails from the -adjacent part of the American continent, and Nertera depressa from high -southern latitudes by way of Tahiti. - - - - - CHAPTER XXIV - - THE MOUNTAIN-FLORAS OF THE TAHITIAN AND FIJIAN REGIONS - -The mountain-flora of the Tahitian region, as illustrated by the - non-endemic genera.—Derived chiefly from high southern - latitudes.—Weinmannia, Coprosma, Vaccinium, Astelia, Coriaria, - Cyathodes, Nertera depressa, Luzula campestris.—The mountain flora of - Rarotonga.—The mountain-flora of the Fijian region, as illustrated by - the non-endemic genera.—Weinmannia, Lagenophora, Coprosma, Astelia, - Vaccinium, Nertera depressa.—The Fijian Coniferæ.—Dammara, Podocarpus, - Dacrydium.—Not belonging to the present era of dispersal.—The age of - dispersal of the Coniferæ in the Pacific.—Earlier than the age of - Compositæ and Lobeliaceæ.—The first in the Mesozoic period.—The last - in the Tertiary period.—Summary. - - - THE MOUNTAIN-FLORA OF THE TAHITIAN REGION AS ILLUSTRATED BY THE - NON-ENDEMIC GENERA - -THIS floral region of the Pacific corresponds with the limits of Eastern -Polynesia, and includes not only the Tahitian group proper, but also the -Cook, Austral, Paumotuan, and Marquesan groups. It is only, however, in -Tahiti, the peaks of which rise to over 7,000 feet above the sea, that -we should expect to find such a mountain-flora, since the islands of the -other groups are much lower, the highest of them in the Marquesan group -barely exceeding 4,000 feet. Yet even in Tahiti it is not possible to -speak of a mountain-flora in the sense that we attach to it in Hawaii. -The elevated area of its interior is, as described in Chapter XIX., -relatively very small; whilst, as Drake del Castillo points out, the -conditions presented by the steep mountain-slopes rarely afford a hold -for trees of any size, ferns often predominating in the higher levels. -Still, we can observe the traces of such a flora, and it is in this -sense only that the term “mountain-genera” is used in relation with this -group. - - - _Mountain-Genera of the Tahitian or East Polynesian Region._ - - Weinmannia, Saxifragaceæ, from New Zealand. } - Coprosma, Rubiaceæ, from New Zealand. }all species - Vaccinium, Vacciniaceæ, from the northern hemisphere. }endemic. - Astelia, Liliaceæ, from New Zealand. } - - Coriaria, Coriariaceæ, from New Zealand }some - Cyathodes, Epacridaceæ, from New Zealand }species - }endemic - - Nertera depressa, Rubiaceæ, a species of the Antarctic flora. - Luzula campestris, Juncaceæ, from the northern hemisphere. - -The Tahitian non-endemic mountain-genera, though scanty in number, are -of considerable interest to the student of plant-dispersal. Among those -possessing only species that are confined to Eastern Polynesia, genera -that would be regarded as belonging to a past era of dispersal, -Weinmannia, Coprosma, Vaccinium, and Astelia may be mentioned. - -Weinmannia, a Saxifragaceous genus of trees and shrubs, not represented -in Hawaii, but recorded from almost all the elevated oceanic groups of -the tropical South Pacific, as well as from the New Hebrides and New -Caledonia, has its home in South America, more particularly in the -Andes, and also occurs in New Zealand, Tasmania, and the Mascarene -Islands. One can scarcely doubt that, as in the case of Coprosma, the -Pacific Islands derived their species originally from high southern -latitudes, as from New Zealand, the absence of the genus from Hawaii -negativing an American origin. Of the two Tahitian species, one is -peculiar to Tahiti, whilst the other, W. parviflora, which is -conspicuous on the mountain-crests at elevations of 3,000 feet and over, -occurs also in the Marquesas. Another species grows in abundance in the -interior of Rarotonga. Samoa possesses two species, one of which, W. -affinis, occurs in Fiji, and the other, W. samoensis, which frequents -the mountains at elevations of 1,500 to 3,300 feet, is seemingly -endemic. Fiji possesses four or five species of Weinmannia occurring at -all altitudes up to 2,000 feet, of which some are evidently peculiar. -The capsular fruits of this genus contain hairy seeds that would -probably become entangled in a bird’s plumage. Dispersal by birds is -distinctedly indicated in the curious observation of Dr. Reinecke in the -case of the Samoan peculiar species. The seeds, he says, appear to -germinate by preference on the bark of other trees, young plants growing -epiphytically being of frequent occurrence. - -There is some evidence that the species of Weinmannia, about ten in all, -found in the tropical islands of the open Pacific are derived from one -or two polymorphous species. As we learn from Mr. Cheeseman, the -Rarotongan species, W. rarotongensis, has considerable affinity to -several closely allied Polynesian species, and its nearest allies are a -Fijian and Samoan species, W. vitiensis and W. samoensis. Possibly, he -remarks, fuller materials may lead to the union of several of these -forms under one species. - -The interesting New Zealand genus Coprosma, which we have noticed in -Hawaii, occurs also in the Tahitian region and Fiji; and it will be -further discussed under the last-named locality. The genus Vaccinium has -been previously dealt with in Chapter XXIII. - -The Liliaceous genus Astelia may be considered as representing, like -Coprosma, the Antarctic or New Zealand flora in the higher levels -(usually) of the islands of the tropical Pacific, where it grows both on -trees and on the ground. The genus, according to Hemsley, is chiefly at -home in New Zealand, but is also found in Fuegia and in South-east -Australia. It is represented in Hawaii, Tahiti, Samoa, and Fiji. In -Hawaii there are two peculiar species ranging between 2,000 to 6,000 -feet in elevation. The solitary Tahitian species, A. nadeaudi, is found -in the central mountains of Tahiti, reaching to the crests of Mount -Aorai, which attains a height of 6,700 feet. Fiji and Samoa possess a -species in common, A. montana, which is only recorded by Seemann, from -the summit of Kandavu, 2,750 feet above the sea; whilst in Samoa it -frequents, according to Reinecke, moist coast districts. The fruits of -Astelia are berries with crustaceous seeds that would be dispersed by -frugivorous birds. - -Amongst the Tahitian mountain-genera that possess species ranging far -beyond this region as well as species confined to the group may be -mentioned Coriaria and Cyathodes. It is to their non-endemic species -that we look for further clues as to the general lines of migration by -which the mountain-genera that only possess peculiar species reached -this group. The evidence afforded by Coriaria is of some importance. The -genus has not been recorded from Hawaii, and, so far as the collections -of Seemann and Home show, not from Fiji. It is found in the -Mediterranean region, the Himalayas, Japan, New Zealand, and Antarctic -America, including Chile; and there are two particular species, C. -ruscifolia and C. thymifolia, that occur in both cases in New Zealand -and the adjacent islands and in South America (Introd. _Chall. Bot._ p. -53). The first of these, which is very common in Chile, exists also in -Tahiti on the crest of Aorai, 6,700 feet above the sea. Drake del -Castillo also describes a peculiar Tahitian species, C. vescoi, of which -the altitude is not given. Here one is in doubt whether Tahiti derived -its wide-ranging species from New Zealand or from Chile; but in the New -Zealand home of Coprosma, another Tahitian mountain-genus, we are -afforded the clue. The fruits of Coriaria possess fleshy cocci that -attract birds, though it would seem that the seeds of plants of this -genus are poisonous for man. Among the numerous fruits that form the -diet of the New Zealand fruit-pigeon (Carpophaga novæ zealandiæ) are -included, as we learn from Sir W. Buller in his _Birds of New Zealand_, -those of the “tupakihi” or “tutu” shrub, which Kirk identifies with C. -ruscifolia, the species that also occurs on the summit of Tahiti. - -The Australian and New Zealand genus Cyathodes (Epacridaceæ) has been -already noticed in the case of Hawaii (page 282). The two Tahitian -species occur on the elevated mountain-ridges forming the summits of -Tahiti, one of them, C. tameiameiæ, occurring also in Hawaii, and the -other, C. pomaræ, being restricted to the group. I have shown that the -fruits are dispersed by frugivorous birds, and I can only include the -genus as another example of the representation of the New Zealand flora -in Tahiti.... There remain of these so-called Tahitian mountain-genera -the Antarctic Nertera and the north-temperate Luzula, each represented -by a solitary widely ranging species, N. depressa and L. campestris, -which I have fully discussed under Hawaii (Chapter XXIII), in which -group they also occur. - -When we look at the evidence of origin supplied by the four Tahitian -mountain-genera possessing species that are found outside the group, -namely Coriaria, Cyathodes, Nertera, and Luzula, we find that the first -three hail from high southern latitudes, and more especially from New -Zealand; and when with this clue in our hands we take up the four genera -Weinmannia, Coprosma, Vaccinium, and Astelia, possessing only species -restricted to the Tahitian region, we find that all but the third-named -genus hail also from the south. It would thus appear that the element of -the Antarctic flora is much more evident in the Tahitian mountain-genera -than with those of Hawaii. In the Hawaiian mountain-flora, excluding, of -course, the endemic genera, it includes about a fourth of the -mountain-genera, which number about thirty-eight or forty in all; whilst -in the Tahitian mountain-flora it comprises six out of the eight genera. -It may, indeed, be said that the resemblance between the mountain-genera -of Hawaii and Tahiti is mainly restricted to genera that are found in -high southern latitudes, namely, Nertera, Coprosma, Cyathodes, and -Astelia, the only other genera linking the mountain-floras of both -groups together being Vaccinium and Luzula, which probably hail from -high northern latitudes. The agency of the frugivorous bird is plainly -marked in the case of five out of the six genera that connect the -cloud-capped peaks of Tahiti and Hawaii. In two of these genera, -Cyathodes and Nertera, the same species occurs in both archipelagoes. - -_The Mountain-flora of Rarotonga._—A word may here be said on the -representation of these mountain-genera in Rarotonga, a small island -2,250 feet in height and about eight miles in length, which is, however, -the most important island of the Cook group. The recent important -explorations of Mr. Cheeseman show that its flora is essentially -Tahitian in character. As in Tahiti, the early age of the Compositæ and -Lobeliaceæ is well represented in the high levels by peculiar species of -Fitchia and Sclerotheca which are discussed in Chapters XXI and XXII. On -account, however, of its relatively low altitude and its small size, we -could not expect any extensive representation of the eight non-endemic -mountain-genera of Tahiti. Yet three of these occur, a Tahitian species -of Vaccinium (page 281) growing on its summits, whilst peculiar species -of Weinmannia (page 290) and Coprosma (page 295) are found in its -interior. The prevailing condition of many of the genera growing in the -higher levels is one of isolation, since other genera, like Pittosporum -and Elæocarpus, only possess peculiar species; but seeing that in -several cases the species are closely allied to others found in the -Western Pacific, as in Samoa, Fiji, and the Kermadec group, it is -apparent that the period of isolation has not long commenced. - - - THE MOUNTAIN-FLORA OF THE FIJIAN REGION. - - Weinmannia, Saxifragaceæ, Fiji and Samoa. } Derived from - Lagenophora, Compositæ, Fiji. } New Zealand or - Coprosma, Rubiaceæ, Fiji. } from the - Astelia, Liliaceæ, Fiji and Samoa. } Antarctic flora. - - Vaccinium, Vacciniaceæ, Samoa, from the northern hemisphere. - Nertera depressa, Rubiaceæ, Samoa, from the Antarctic flora. - - Dammara, Coniferæ, Fiji. }Not as a rule belonging - Podocarpus, Coniferæ, Fiji and Tonga. }to the present age - Dacrydium, Coniferæ, Fiji. }of dispersal - -But little can be said of the mountain-flora of Fiji, since on account -of the relatively low elevation of the islands there are but few special -mountain-genera; and as a rule we find only here and there a solitary -species on some isolated peak that recalls the upland flora of the -Hawaiian mountains. “None of the mountains of Fiji,” remarks Horne (page -60), “are high enough for an alpine flora to exist. Many of the plants -found on the tops of the mountains are also found near the level of the -sea. On the other hand sea-level plants may also be found on the tops of -the hills.” - -Fiji lacks the endemic genera of Compositæ and of Lobeliaceæ that often -give a character to the mountain-floras of the Hawaiian and Tahitian -regions, though, as remarked in Chapters XXI and XXII., their absence -involves something more than a question of station. We find, however, -four genera of the Antarctic or New Zealand flora, Weinmannia, -Lagenophora, Coprosma, and Astelia. The first-named genus possesses four -or five species ranging up to 2,000 feet, some of which are endemic, and -it has been already discussed in this chapter. The United States -Exploring Expedition found a single species of Lagenophora (L. -pickeringii) on the mountains of the Mathuata coast of Vanua Levu, and -no other species seems to have since been found. The subject is dealt -with in Chapter XXIII in the case of Hawaii, but it may be here observed -that there is an Hawaiian mountain species, and that the route followed -by the ancestor of the Fijian species from the New Zealand home of the -genus is indicated by a species in the intermediate Kermadec group. The -genus Astelia has been discussed on page 291. It is represented in -Hawaii and in most of the oceanic groups of elevated islands. The -solitary species, A. montana, discovered by Seemann on the summit of -Kandavu in Fiji, has since been found in Samoa, and probably Mr. Horne’s -collections contain another species. - -The Rubiaceous genus Coprosma needs a few special remarks, since a -particular genus of birds seems to have been concerned in dispersing it -in the South Pacific. About fifty species are enumerated in the _Index -Kewensis_, and if we include a few other species from the collections of -Hillebrand, Horne, Cheeseman, &c., the total would be about sixty. Of -these, about half are restricted to New Zealand, which may be justly -regarded as the home of the genus, the rest being confined to Australia -and the islands of the Pacific, excepting a Chilian and three or four -Malayan species. Hawaii with its nine species, Tahiti with two, -Rarotonga with one, and Fiji with two or three species represent -approximately the distribution of the genus in the oceanic archipelagoes -of the tropical Pacific. (It most probably exists on the high peaks of -Samoa, though it has not yet been recorded from the group.) In all, or -in almost all cases, the species are restricted to their particular -groups, so that we may regard the dispersal of the genus over the -Pacific as suspended, though, as will be observed below, the period of -suspension in the South Pacific has not been of sufficient duration to -obliterate the affinities of species in distant groups and to prevent us -from tracing out the route followed by the genus. - -This genus of temperate latitudes, which in its New Zealand home ranges -from near the sea-level to the region of the alpine floras, finds its -usual station in the tropics on the summits of mountains. Thus, on Mount -Kinabalu, in Borneo, it is found at altitudes of 10,500 to 13,000 feet -(Stapf), and on the mountains of East Java at elevations exceeding 9,000 -feet (Schimper). In Hawaii its species grow at elevations ranging from -3,000 to 9,000 feet, and in Tahiti at altitudes of 2,600 to 3,300 feet; -whilst in Rarotonga it grows in the hilly parts of the island, its -elevation in Fiji not being recorded. - -When we come to consider the route by which the genus (Coprosma) entered -the tropical Pacific, we must remember that unless we establish some -special connection with its New Zealand home it will always be open for -any one to suggest that the genus might have been derived, like -Vaccinium, from other regions than the south, as from the summits of the -Malayan mountains. However, a curious connection has been discovered by -Mr. Cheeseman in his examination of the Kermadec and Rarotongan floras, -and it would indeed appear that he has traced the Rarotongan peculiar -species to its New Zealand home. Thus, he says that Coprosma lævigata, -his new Rarotongan species, is very closely allied to the Kermadec -endemic plant, C. acutifolia, Hook., which itself comes near C. lucida, -Forst., a New Zealand species. The connection between Rarotonga and New -Zealand by way of the Kermadec group is rendered yet more probable by -the occurrence of two New Zealand species of Coprosma in the Kermadec -flora (_Journ. Linn. Soc._ i. 1857; _Trans. Linn. Soc. Bot._ vi. 1903; -_Trans. N.Z. Instit._ xx. 1887). - -When speaking of the genus in Hawaii (page 275), mention was made of the -inter-island dispersal of the fruits of one of the species by the native -mountain-goose, Bernicla sandwicensis. We learn from Sir W. Buller’s -_History of the Birds of New Zealand_ that when the Coprosma is in fruit -the Swamp-Hens (Porphyrio melanotus) come out to feed on it. These -birds, he says, are capable of prolonged flight; and I chance to have -beside me a cutting from the _Field_ of July 9, 1904, in which “Hy. S.” -refers to a Black-backed Porphyrio that was captured in 1876 four -hundred miles off the coast of New Zealand. This genus, which is widely -dispersed in the tropics, the birds being commonly known as Sultanas, -Blue Gallinules, Purple Water-Hens, &c., has probably been a very -important factor in the dispersal of plants, especially in connection -with insular floras. The birds live on a variety of food. The Messrs. -Layard observed that Porphyrio vitiensis, which abounds in the swamps of -New Caledonia, fed on maize, yams, &c. (_Ibis_, 1882); whilst in the -stomach of a bird of the same genus shot in the Rewa swamps in Fiji I -found a number of the stony fruits of Scleria, a genus of the Cyperaceæ. -According to Mr. Wiglesworth, each region in the South Pacific has its -own species of Porphyrio. There is one in the Tahitian Islands, and -another common to Fiji, Tonga, and Samoa; whilst New Caledonia and the -New Hebrides have their species (“Aves Polynesiæ”). However, it is -evident that the power of dispersing seeds from group to group is not -quite suspended, since, as we learn from Sir W. Buller, the New Zealand -species, above named as partial to Coprosma drupes, is distributed over -Tasmania and Australia, and reaches also Niue and New Caledonia; whilst -the Messrs. Layard evidently regarded one species as common to Fiji and -New Caledonia. - -It is doubtless to birds of this description that we owe some of the -specific connections of Coprosma between groups of the Western Pacific. -That the dispersal of the species over distant regions was recently in -active operation is shown by the close affinity, according to Dr. Stapf, -of two species growing on the summit of Kinabalu, the Bornean mountain, -with certain species from New Zealand and South-east Australia. Other -Rubiaceous species, like Nertera depressa, possessing Coprosma-like -fruits and fitted for the same mode of dispersal, link the heights of -Kinabalu with the flora of high southern latitudes. - -Being included in the Fijian area, the scanty mountain-flora of Samoa -may be here referred to. As in Fiji, the endemic genera of Compositæ and -Lobeliaceæ are not to be found, but we find in the central elevated -district of Savaii, which rises to over 5,000 feet above the sea, a -peculiar species of Vaccinium (4,900 feet), the Antarctic Nertera -depressa (4,000 feet), and two species of Weinmannia, a genus hailing -probably from high southern latitudes. - - - THE FIJIAN CONIFERÆ. - -It has been found most convenient to discuss here these interesting -plants, which belong in a general sense to the mountain-flora of this -archipelago. That which the Fijian flora loses in interest in the eyes -of the student of plant-dispersal in not possessing the mysterious -Composite and Lobeliaceous genera of Hawaii and Tahiti, it regains in -the possession of its genera of Coniferæ. If he felt loth to apply his -empirical principles to the above-named Hawaiian and Tahitian endemic -genera, he feels more than uneasy when he comes to deal with the three -Coniferous genera of Fiji, Dammara (Agathis), Podocarpus, and Dacrydium. - -These three genera represent an order that has not found a home either -in Tahiti or in East Polynesia generally, or in the more distant Hawaii; -and they present at first sight in their existence in Fiji a powerful -argument in favour of the previous continental condition of the islands -of the Western Pacific. But in advocating this view we should remember -that it involves the original continuity of the Fijian land-area, not -only with the neighbouring islands of the New Hebrides and of New -Caledonia where these genera alike occur, but also with New Zealand, -Tasmania, and Australia, where they sometimes attain a great -development. - -In Fiji these trees often chiefly form the forests of the larger -islands, extending in the moister regions from near the sea to the -mountain-tops, and being often abundant on the great mountain-ridges of -the interior. It may be at once remarked that, viewed merely from the -standpoint of dispersal, there is no great difficulty in regarding it as -probable that the seeds of Podocarpus and Dacrydium have been dispersed -by frugivorous birds over tracts of ocean 500 or 600 miles across. -Dammara, however, so far as my Fijian observations show, possesses none -of the means of dispersal across oceans that we are at present -acquainted with. The two first-named genera occur in South America as -well as in the Australo-Polynesian region, some of the species in these -two regions, though the Pacific Ocean divides them, being closely -related. Dammara is, on the other hand, confined to a much more limited -area, extending from New Zealand to Borneo. It is from the distribution -of this genus that the continental theory derives its chief support. - -Yet it may be remarked that something more than questions relating to -the capacity for dispersal are involved here. This is at once indicated -by the circumstance that although Podocarpus is known to be dispersed by -frugivorous birds, it is not found in Polynesia east of Tonga, and the -same may be said of Dacrydium, which does not occur east of Fiji. In -this connection it is necessary to notice the intrusion of Araucaria -into the tropical Pacific from Eastern Australia to New Caledonia and -the New Hebrides. The fact of this genus not having been recorded from -Fiji or any of the groups east of the New Hebrides is very remarkable, -and scarcely in accordance with the continental hypothesis. There is a -persistence in type of these genera of the Coniferæ during geological -time that prevents us from dealing with them on the lines that are -required by the mass of the flowering-plants. Other factors intervene, -and we apply with hesitation the same canons of dispersal that we employ -for the general bulk of the plants of the Pacific islands. If, as often -happens, a specific distinction alone separates the Conifers of the same -genus on either side of the Pacific Ocean, it must possess in point of -time a very different value from that which we would usually attach to -specific distinctions in the floras of the Pacific islands. - -DAMMARA (AGATHIS).—The Dammara region includes Eastern Australia, New -Zealand, New Caledonia, with the New Hebrides, Fijian, and Santa Cruz -groups, and extends north-west to Java and Borneo. Only ten species are -named in the _Index Kewensis_, and of these four are assigned to New -Caledonia and two to Fiji, the focus of geographical distribution being, -therefore, as Seemann long since pointed out, in the islands of the -Western Pacific. The absence of the genus from the neighbouring Samoan -and Tongan groups is very significant; and it is evident that the -ordinary agencies of dispersal, whether birds, winds, or currents, have -here failed to extend the genus over a few hundred miles of sea. - -When by means of observation and experiment we turn to the fruits and -look for a reply, we find in the first place that they are never to be -noticed either whole or in part in the floating drift of sea or river, -or amongst the stranded materials of the beaches. This is at once -explained when we ascertain that the fresh cones sink in the -river-water, and thus could never reach the coast in their entire -condition. Nor could they do so in fragments, since the detached cone -falls to pieces on the ground and the separate scales and seeds sink at -once or float only for a few hours. In order to test the buoyancy of a -cone after drying, it is necessary to bind it round with string to keep -it from breaking down. One such fruit, after being kept for ten days, -was placed in sea-water, where it floated heavily for eleven days and -then sank. This is, of course, a most unnatural experiment, but it was -well to have carried it out. That the entire fruit could never be -transported by water is indirectly implied by Kirk respecting the fruit -of Dammara australis, the Kauri Pine of New Zealand. In this case, when -the fruit reaches maturity the scales, he remarks, fall away from the -woody axis of the cone and the seeds are freed. - -The fleshy, unprotected seeds, which, as above noted, possess little or -no floating power, could scarcely withstand the injurious effect of -sea-water; and they are absolutely unfitted for any known mode of -dispersal by birds. It is observed by Kirk that the seeds of the New -Zealand tree are widely spread by winds. But this could only avail them -for local dispersion, and they appear ill-suited for being transported -for more than a few paces. The seeds are winged, and are in form a -little like the samara of the Maple (Acer); but they have not the same -protective coverings, the wing being, however, only a little more than -half the length of the entire seed. Those of both Dammara australis and -D. vitiensis are about two-thirds of an inch in length, and are -heavy-looking; and the agency of the wind could never be invoked except -for local dispersion. - -Looking at these results, the cones of Dammara may be regarded as most -unsuited for any of the ordinary means of dispersal over an ocean except -through the agency of man. There is, however, no necessity to introduce -man’s aid here, unless the gum or resin which the Fijian burns in his -torches and employs as a glaze for his pottery gave his ancestors an -object in carrying the cones with them in their migrations. But in that -case the same argument would have to be applied to all partially useful -plants, and much of the Fijian flora would lose its indigenous -reputation. The endemic character of the Fijian species also militates -against such a view, and we should have to apply the same explanation to -the New Zealand species, concerning which no one, so far as I know, has -ever ventured to suggest that it was introduced by the Maoris. - -The native names of the trees seem to have been sometimes connected with -general words for gums or resins; whilst at other times the tree and the -resin have separate designations. Thus the Fijians call the tree -“ndakua” and the resin “makandre,” which last Hazlewood in his -dictionary seemingly connects with “ndrenga,” the word for “gum.” In my -work on the Solomon Islands, page 190, I have endeavoured to show that -the Maori name of “kauri” may be connected with “gatah,” the general -Malayan word for gums and resins, transitional stages being presented in -the names of resin-yielding trees in the intermediate regions, as, for -instance, by “gutur,” a species of Canarium, on the Maclay coast of New -Guinea, and by “katari,” a species of Calophyllum, in Bougainville -Straits, Solomon group. It may be pointed out that these facts of -plant-nomenclature do not promise us any aid in determining the mode of -dispersion of Dammara in the Western Pacific. There is a suspicious -resemblance between the Fijian name of “ndakua” and “dundathu,” the -Queensland aboriginal name for Dammara robusta; but even if the -comparison is legitimate, its explanation may lie far back in the ages -in some root-word as ancient as the Malayan “gatah.” - -If there is a real difficulty in applying our canons of -plant-dispersal to the distribution of Dammara, it is merely the same -difficulty that has so often perplexed the botanist with other -Coniferous genera in continental regions, such as, for instance, the -occurrence of Pinus excelsa on the far-removed mountains of Europe and -of the Himalayas, and the existence of the cedar in its isolated homes -on the Atlas, the Lebanon mountains, and the Himalayas. Such -difficulties largely disappear if we regard the present distribution -of the Coniferæ as the remnant of what it was in an ancient geological -period. In the case of Dammara it seems almost as idle to puzzle over -its means of dispersal as to consider the mode of dispersal of the -Marsupials. The questions, indeed, that affect the Dammaras of Fiji -and the Western Pacific far ante-date any questions concerning a -previous continental condition of those regions. The attitude of the -palæobotanist to such questions would probably be one of indifference; -yet to the student of plant-distribution they are of prime importance; -and _nolens volens_ we must admit that Dammara may well be cited in -support of any continental hypothesis affecting the Western Pacific. - -PODOCARPUS.—In this connection I will mainly depend on Pilger’s recent -monograph on the Taxaceæ (heft 18, Engler’s _Das Pflanzenreich_, 1903). -More than sixty species are here enumerated, which are distributed in -Africa, Asia, Australasia, and South America. With a range that extends -north to Japan and south to Southern Chile in latitude 48°, this genus -attains its greatest development in respect of species in Malaya, in the -region comprised by Australia, New Zealand, and New Caledonia, in South -America, and in Africa. Eastward of New Caledonia it is found in Fiji -and in Tonga, but not in Samoa, and it is altogether absent from the -Tahitian region as well as from Hawaii. Of the four species accredited -by Seemann to Fiji, two are enumerated by Pilger, namely, P. affinis and -P. vitiensis. The first-named, according to Stapf, is allied to P. -bracteata, which occurs on the upper slopes of Kinabalu, in Borneo, and -is distributed not only over Malaya, but occurs in Japan and in the -Himalayas. The Tongan species, P. elatus, is, according to Hemsley, -found in East Australia. - -This Tongan tree is suggestive of bird-agency in the dispersal of the -genus, and the same may be said of the occurrence of another species, P. -ferrugineus, found in both New Caledonia and New Zealand. Since the -seeds of the genus possess an outer fleshy and an inner bony covering, -they would appear to be well fitted both to attract and to be dispersed -by birds. In fact, we learn from Sir W. Buller that the New Zealand -fruit-pigeon feeds on the seeds of the “matai” tree (Podocarpus spicata) -and of the “kahikatea” (P. dacrydioides), and no doubt to the agency of -frugivorous birds we can attribute the presence of the genus in Fiji and -Tonga. Yet it is strange that bird-agency should have failed both with -Tahiti and Hawaii. In point of size the seeds, which range from -one-quarter to an inch across, present no great difficulty, and one -would have thought that the birds that carried the “stones” of -Elæocarpus to Hawaii could have also carried the seeds of Podocarpus. - -It is, however, necessary to remember, in dealing with a genus that has -a wide distribution both in time and space, that specific affinities may -have a very different significance with the Gymnosperms than with most -other flowering plants. When Hemsley remarks (_Introd. Chall. Bot._ p. -56) that the New Zealand Podocarpus spicata is closely allied to the -South American P. andina, he does not imply that the two regions are in -touch with each other though some 5,000 to 6,000 miles of ocean -intervene. One is prepared to credit these seeds with a capacity of -dispersal by birds over tracts of sea such as the extent of ocean -separating New Caledonia and New Zealand, which are some 900 miles -apart; but one hesitates to admit that frugivorous birds could carry -them across the Southern Ocean. If we assign a home in the high -latitudes of the northern hemisphere to a genus that was well -represented in Europe in the Tertiary period, a movement of migration -southward would explain most of the difficulties in its present -distribution. The great vertical range of some of the species leads us -to attribute a corresponding power of adaptation to the genus in respect -of widely different climates. Thus, according to Stapf, the vertical -range of P. bracteata in the Malay Archipelago extends, including -varieties, from the coast to an altitude of 12,000 feet. With such a -capacity for adaptation, migrations of the genus would be rendered easy -over the globe. - -DACRYDIUM.—It may happen that some additional light on the mystery of -the Fijian Coniferæ may be afforded by Dacrydium elatum, a tree that -occurs not only in Fiji, but in Further India and in Malaya. Pilger -confirms Seemann’s view in his identification of the Fijian tree, and -this opinion is, in the main, shared by Stapf. This species, so to -speak, affords us a _point d’appui_ in the history of the distribution -of the genus in the Western Pacific. This distribution somewhat -resembles that of Dammara in extending from New Zealand (its principal -centre) to Malaya and Further India; but, unlike Dammara, Dacrydium is -represented in America by a solitary species in South Chile. Of the -sixteen species enumerated by Pilger, seven belong to New Zealand, four -to New Caledonia, three to Malaya, one to Tasmania, and one to Chile. -The seeds are, as a rule, smaller than those of Podocarpus, and on -account of their somewhat similar structure would serve as bird-food, -and might be distributed in this fashion. Yet the genus has been only -recorded from Fiji, and is not only unrepresented in Hawaii and Tahiti, -but is also not known from the Tongan and Samoan groups that belong to -the Fijian floral region of the Pacific. Capacities for dispersal appear -meaningless here, especially when we have regard to the solitary -American species, Dacrydium fonkii, that as a shrub finds a refuge in -the bleak region of Southern Chile. - -The three Fijian genera of the Coniferæ, Dammara, Podocarpus, and -Dacrydium, appear at first sight to be beyond the reach of our canons of -plant-dispersal, by which we connect specific affinity with a continuity -of range, and by which we co-ordinate means of dispersal and area of -distribution. We begin to realise that there may have been an age of -Coniferæ in the Pacific islands that is even less amenable to our -methods than the later era of the Compositæ and Lobeliaceæ in Hawaii and -Tahiti. Such an age would be concerned only with that region in the -Western Pacific which is now held by the genera Dammara, Podocarpus, and -Dacrydium, a region that did not participate in the era of the Compositæ -and Lobeliaceæ. We thus have evidence of an ancient era of the Coniferæ -that was confined to the Western Pacific, and of a later era indicated -by the peculiar genera of Compositæ and Lobeliaceæ that was restricted -to Hawaii and to Eastern Polynesia (Tahiti, Rarotonga, &c.). The key to -the situation here presented seems to lie in the following -considerations. - -It is assumed that there was an age of Coniferæ in the Pacific, or -rather that this region shared in an era of dispersion of existing -genera of the order. In this age only the islands of the Western Pacific -participated, neither the Hawaiian nor the Tahitian islands taking a -part in it. Such a result is to be attributed either to the inability of -these genera of Conifers to reach Hawaii and the islands of East -Polynesia, or to the non-existence of the Hawaiian and Tahitian -archipelagoes at that epoch. The first explanation seems scarcely -acceptable, since, although the powers of dispersal of the genus Dammara -are very limited, there seems no reason why the genera Podocarpus and -Dacrydium could not have reached those distant regions of the Pacific. -The second explanation is most probable, and it is the one suggested by -Hillebrand (p. xxx) in the case of Hawaii, namely, that “the absence of -Gymnosperms militates for the view that the islands were formed -subsequent to the age in which these were universally distributed.” - -If this conclusion is legitimate we have here a datum-mark in the -history of the islands of this ocean. Before the appearance of the -Hawaiian and Tahitian islands (using the term Tahitian to cover the East -Polynesian region) there existed a land-area in the Western Pacific held -by the Coniferæ, probably in the late Secondary period. After the -formation of the Hawaiian and Tahitian islands, perhaps in the early -Tertiary epoch, came the age characterised by the ancestors of the -present endemic genera of the Compositæ and Lobeliaceæ, and of a few -other orders in Hawaii and Tahiti. In this age the islands of the -Western Pacific do not seem to have participated, and it is to be -inferred that this was an age of extensive but probably not of complete -submergence in that part of the ocean, since at least the genus Dammara -was able in places to hold its ground. Then ensued the great Tertiary -emergence of the land-areas of the Western Pacific, when small islands -that dotted the sea-surface in this region became the nuclei for the -formation of the large islands of the present Fijian, New Hebrides, and -Solomon groups. This prepared the way for the migration of Malayan -plants which now predominate over the islands of the tropical Pacific; -and in a later age man, following the same track from Indo-Malaya, -occupied these islands. - -In my volume on the geology of Vanua Levu it was shown that the Tertiary -period was an age of submergence in the Western Pacific, and a disbelief -in any previous continental condition was expressed. My later view is -more in accordance with that of Wichmann, who, on geological grounds, -contended that the islands of the Western Pacific were in a continental -condition during the Palæozoic and Mesozoic periods, and that their -submergence and subsequent emergence took place in Tertiary times. The -distribution of the genus Dammara has thus led me to modify the views -expressed in the final chapter of my first volume on the geology of -Vanua Levu. Though still holding that there is no geological evidence -that the various islands of the Fijian group were ever amalgamated, or -that they were joined as such to the westward groups, it is quite -possible that their position was indicated by a few small islands a few -miles across and a few hundred feet in height in early Tertiary times. -On these small islands, which probably represented the remains of a -submerged Mesozoic land-area, such as is in part implied in Dr. Forbes’ -_Antipodea_, or in Mr. Hedley’s _Melanesian Plateau_, the genus Dammara -survived. Such islands merely indicated the situation of some of the -present groups of the Western Pacific, which have been since largely -built up by submarine eruptions, and the greater number of the islands -were no doubt completely submerged. Between the groups as we know them -now there never was any land connection, since they are the product of -later eruptions, mainly submarine; and they have acquired their present -composite character during the emergence that followed the period of -volcanic activity. Except, perhaps, in New Caledonia, which does not -seem to have shared in the Tertiary submergence, the islands of the -Western Pacific have a configuration acquired in comparatively recent -times, and one that gives no idea of the character of the Mesozoic -continent. - -Such, as I understand them, are the indications of the Fijian Coniferæ -and particularly of Dammara. In the distribution of this genus we have -outlined an ancient, more or less continuous land area which, with the -exception of a few isolated points, disappeared beneath the sea in -Tertiary times to re-appear near the close of that period in the form of -a number of archipelagoes that were largely built up by submarine -eruptions, and probably altogether mask the form of the original -land-area. It may be remarked that New Zealand, which largely shared in -the Tertiary submergence, especially in the Miocene age, is included in -the range of the genus Dammara, as well as in those of the genera -Podocarpus and Dacrydium. - - - _Summary._ - -(1) The evidences of a mountain-flora in Tahiti, as indicated by the -non-endemic genera, though, as we would expect, of a scanty nature when -contrasted with Hawaii, are nevertheless of considerable interest. There -is much kinship with the Hawaiian mountain-flora, but it is mainly -confined to genera from high southern latitudes, such as Nertera, -Coprosma, Cyathodes, and Astelia, which are all dispersed by frugivorous -birds. Amongst other plants linking the Tahitian mountains with the -region of the Antarctic flora, and with New Zealand in particular, may -be mentioned Coriaria ruscifolia and the genus Weinmannia. - -(2) On account of their relatively low altitude the Fijian islands do -not present the conditions for an alpine flora. Traces, however, of the -Antarctic flora, or of the New Zealand flora, occur on occasional -mountain-tops, as is indicated by the occurrence of species of -Lagenophora, Coprosma, and Astelia. In Samoa the mountain-flora is also -scantily developed, as we might have expected; but here occurs the genus -Vaccinium as well as a widely-ranging species of the Antarctic flora, -Nertera depressa. - -(3) The route by which some of the representatives of the flora of high -southern latitudes reached the mountains of the islands of the tropical -Pacific is directly indicated by the genus Coprosma to have been from -New Zealand by way of the Kermadec Islands. - -(4) In the distribution of plants possessing drupes or berries that -connect the tropical islands of the South Pacific with New Zealand, it -is highly probable that birds of the genus Porphyrio (Swamp-Hens or -Purple Water-Hens) have taken a prominent part. - -(5) In the possession of species of the three genera of Coniferæ, -Dammara, Podocarpus, and Dacrydium, which often largely form the forests -of the mountain-slopes, Fiji is distinguished from all the other groups -of the open Pacific with the exception of Tonga, which owns a species of -Podocarpus probably introduced by birds. From the circumstance that -Dammara has no known means of crossing a tract of ocean, whilst -Podocarpus and Dacrydium could be dispersed by frugivorous birds, all -three genera having, however, much the same limited distribution in the -Western Pacific, it is apparent that something more than a question of -means of dispersal is here involved. It is assumed that they mark the -site of a Mesozoic continental area in this region, and that at this -period the Tahitian and Hawaiian groups which possess no Conifers did -not exist. This area was submerged during the Tertiary period with the -exception of a few peaks that formed small islands on which the Conifers -held their ground. During the Tertiary submergence of the Western -Pacific region, the Hawaiian and Tahitian islands were built up by -subaërial volcanoes and received the ancestors of the Compositæ and -Lobeliaceæ that now exist as endemic genera in those groups. Then -followed the emergence of the islands of the Western Pacific and their -occupation mainly by Indo-Malayan plants that extended eastward over the -Pacific. Thus in the Pacific there has been first an age of Conifers in -which the islands of the Hawaiian and Tahitian regions could not -participate, since they did not exist. Then ensued an era of American -forms of Compositæ and Lobeliaceæ in which only Hawaii and Tahiti -participated, since the Western Pacific region was submerged. Lastly -came the invasion of Indo-Malayan plants, which have largely occupied -every group in the tropical Pacific. - - - - - CHAPTER XXV - - THE ERA OF THE NON-ENDEMIC GENERA OF FLOWERING PLANTS (_continued_) - - THE AGE OF THE MALAYAN PLANTS AS REPRESENTED IN THE LOW-LEVEL FLORA OF -HAWAII AND IN THE BULK OF THE FLORAS OF THE FIJIAN AND TAHITIAN REGIONS - - - _The Age of Wide Dispersal over the Tropical Pacific._ - -The widely dispersed genera which possess only peculiar species in - Hawaii.—Pittosporum.—Reynoldsia.—Gardenia.—Psychotria.—Cyrtandra—Freycinetia.—Sapindus.—Phyllanthus.—Pritchardia.—Summary. - -WE pass now from the consideration of the mountain-flora of Hawaii and -its scanty representation in the Fijian and Tahitian regions to a -discussion of the low-level Hawaiian flora, belonging to stations under -4,000 or 5,000 feet, and of the corresponding floras of the other two -regions. It has been previously pointed out that in mass the plants of -Fiji and Tahiti correspond to the low-level flora of Hawaii. - -There are numerous ways of comparing this era of the non-endemic genera -of these three regions of the Pacific. The necessities of space, -however, compel me to treat the subject only in an illustrative fashion, -and in adopting the plan which seems easiest and simplest I have also -been obliged to keep my limitations mainly in view. - - - THE WIDELY-DISPERSED GENERA WHICH POSSESS ONLY PECULIAR SPECIES IN - HAWAII. - -Amongst the oldest denizens of the Pacific islands in this era of the -non-endemic genera may be taken those genera of flowering plants which -are found in all three regions, Hawaii, Fiji, and Tahiti, but possess in -the first group only endemic species, whilst in the other two regions -they may include species both confined to and occurring outside the -respective groups. They represent an age of wide dispersal over the -Pacific, an age which for Hawaii has long since passed away, since all -the genera have been disconnected from the outer world, whilst in the -groups of the South Pacific they as a rule in each case remain in touch -through some of the species with the groups around. - -The problem of plant-distribution in the Pacific thus assumes a -different aspect in an age which we term Malayan or Indo-Malayan, since -the bulk of the plants are thence derived. The earliest age of the -Coniferæ was, as we have seen in the previous chapter, restricted to the -region of the Western Pacific. The following age of the Compositæ and -the Lobeliaceæ was concerned with the regions of Tahiti and Hawaii. Now, -however, in the Malayan era, the whole of the tropical Pacific is -concerned. Yet, although we shall still regard, for purposes of -convenience, the groups of Fiji, Tahiti, and Hawaii as the three foci of -plant-distribution, it will soon become apparent that in future there -will be in reality only two regions to deal with, the Hawaiian in the -North Pacific, and the whole region of the South Pacific extending from -Fiji to Tahiti and as far east as the islands stretch. It will be also -seen that in making our comparison we shall sometimes have to regard -each of the principal Hawaiian islands as the equivalent as a -plant-centre of an entire archipelago of the South Pacific. - -The genera that are here selected to represent this epoch of wide -dispersion are very characteristic of the floras of the Pacific islands. -Genera like Pittosporum, Gardenia, Psychotria, Cyrtandra, Freycinetia, -and others one meets with everywhere in the larger islands, and it -should be observed that they are predominantly Old World, and more -especially Malayan, in their origin, not a single purely American genus, -unless we except the decadent genus of fan-palms, Pritchardia, occurring -among them. Here we notice [what we shall see is especially typical of -the era of the non-endemic genera, excepting those of the lofty uplands -of Hawaii] that the frugivorous bird has been the principal agent in -dispersing the plants, quite two-thirds of the total genera possessing -drupes or berries that would attract such birds. The transport of seeds -or seedvessels in birds’ plumage, which was a conspicuous feature in the -case of the mountain-flora of Hawaii, is not a feature of this age of -wide dispersal of tropical plants over the Pacific. - -The genera selected to represent this age are given in the following -table. Those on which my observations directly bear, or in which I was -particularly interested when in the Pacific, will be discussed in detail -from the standpoint of dispersal; whilst only a brief reference will be -made to a few of the others, not, however, from lack of materials at my -disposal, but merely to keep this volume within moderate bounds. - - -_Genera selected to represent the Age of Wide-dispersal of Indo-Malayan -or Malayan Plants over the Pacific, and possessing in Hawaii only -Endemic Species._—Most of the genera of this age are exclusively from -the tropics of the Old World, whilst those found on both sides of the -Pacific can be shown in most cases to have been derived from the same -source, and only very few, like Pritchardia, can be traced to America. - - Pittosporum (Pittosporeæ). - Sapindus (Sapindaceæ). - Reynoldsia or Trevesia (Araliaceæ) - Gardenia (Rubiaceæ). - Psychotria (Rubiaceæ). - Cyrtandra (Gesneraceæ). - Phyllanthus (Euphorbiaceæ). - Pritchardia (Palmaceæ). - Freycinetia (Pandanaceæ). - - - PITTOSPORUM (Pittosporeæ). - -This genus, which contains nearly a hundred species, usually of small -trees, is widely spread in the warmer regions of Africa, Asia, -Australia, and New Zealand. It is also especially a genus of oceanic -islands, occurring not only in those of the Pacific but also in Madeira -and Teneriffe in the Atlantic. - -Though found in most of the larger Pacific groups, it has apparently -never been recorded from Samoa. From Hawaii ten species are known, all -peculiar to that group. About half a dozen have been described from -Fiji, of which three at least have been observed outside the group in -the neighbouring Tongan Islands. Rarotonga possesses a peculiar species -which, however, is so near to two other Fijian and Tongan species that, -according to Cheeseman’s memoir, they may have to be subsequently -united. Tahiti is credited by Drake del Castillo with a solitary species -widely distributed in the Old World, whilst in the _Index Kewensis_ a -peculiar species is assigned to it. They form small trees of the wooded -mountain-slopes of Fiji; whilst in Hawaii, beside occurring in the lower -forests, they may extend to altitudes of between 5,000 and 7,000 feet. -In the connection that more or less exists between the species of the -South Pacific archipelagoes, and in the endemic character of all the -Hawaiian species, we see the principle exemplified that there are two -regions of distribution in the islands of the tropical Pacific—the -Hawaiian region and the South Pacific region. - -Before their dehiscence, the wrinkled, woody capsules would seem very -unlikely to attract birds; but the observer on handling an opening -fruit, with its orange or brightly coloured lining and displaying black -or dark-purple seeds immersed in a semi-liquid pulp, would form a -different idea of the plant’s capacity for this mode of dispersal. The -mature dehiscing fruits are very conspicuous on the tree; and the seeds -covered with the “sticky” material of the pulp might possibly adhere to -birds pecking at the fruit. But this would only aid in local dispersion, -since the weight and size of the seeds, 5 to 8 millimetres (2/10 to 3/10 -inch), would unfit them for this mode of transport across an ocean. They -are, however, sufficiently protected by their hard tests to be able to -pass unharmed through a bird’s intestinal canal. - -Yet the distribution of the species of Pittosporum in the Pacific would -show that their dispersal is more a matter of the past than of the -present. Out of the ten peculiar Hawaiian species, Hillebrand designates -none as generally distributed over the group. But it is evident that, -though it is on the point of breaking off, some sort of connection still -exists in the South Pacific between the Tongan and Fijian species, and -until recently between the species of those two groups and of Rarotonga. - - - REYNOLDSIA (Araliaceæ). - -The Polynesian genus of Reynoldsia, originally established by Gray, is -merged by Hooker and Bentham into the Malayan genus Trevesia, a step -that brings the Pacific plants into line with many other of the plants -hailing originally from the Old World. The significant fact in the -distribution of this genus of small trees in the Pacific is that its -dispersal over the ocean has ceased long ago, since the three species -here occurring are restricted each to a particular group, namely, to -Hawaii, Tahiti, and Samoa. Yet the inter-island dispersal still -continues in the Hawaiian Group, the species characteristic of that -archipelago being found in all the islands. - -Reynoldsia sandwicensis came frequently under my notice in Hawaii, and -the fairly fleshy drupes, about one-third of an inch, or 8 millimetres, -in size, with their crustaceous pyrenes appeared to me well fitted for -assisting the dispersal of the plant by frugivorous birds. Yet here the -same question arises that presents itself with so many other Hawaiian -plants, and that is, How has it happened that the birds have continued -to disperse the species over the scattered islands of this group long -after they ceased to transport fresh seeds from the outside world? The -answer is an obvious one. The birds that originally brought the seeds of -the parent species from some distant region came at last to remain -permanently in the Hawaiian Group, and not only the plant but probably -also the bird has since undergone specific differentiation. This link -between bird and plant in the floral history of a group of Pacific -islands is the common theme of the story of most of the endemic species -of plants in this region of the globe. - - - GARDENIA (Rubiaceæ). - -This genus, comprising about a hundred known species, is spread over -tropical Africa, Asia, and America, and over all the groups of the -tropical Pacific. On account of their handsome, white, scented flowers -these shrubs are much appreciated by the Pacific islanders, who employ -the flowers for personal decoration. Some ten species have been -described from the groups of the open Pacific, all of which, with the -exception of Gardenia tahitensis, which ranges the South Pacific from -Fiji to the Marquesas and Tahiti, are seemingly peculiar to the -different archipelagoes. Thus there are some six species endemic to -Fiji, one to Samoa, and two to Hawaii. - -The Hawaiian Islands are, however, quite isolated in this respect, since -the group possesses only peculiar species; whilst a solitary species -keeps up the connection between the groups on the south side of the -equator. The Gardenias thus tell the same story of complete isolation in -Hawaii, and of partial isolation in the archipelagoes of the South -Pacific that is repeated by many other Pacific genera. Yet in Hawaii -there has subsequently been some inter-island dispersal, since the -species are not restricted each to a single island, but are found on two -or three islands. The significance of the relation of the Hawaiian -Gardenias to those of the combined Fijian and Tahitian areas consists in -regarding the two regions, the Hawaiian and the South Pacific, as of -equivalent value, and each large Hawaiian island as equivalent to one of -the southern archipelagoes. - -_The Station of the Pacific Gardenias._—Although they may occur in the -forests, the Gardenias of the Pacific are most characteristic of dry, -thinly vegetated localities, and they have an inclination for the -vicinity of the coast. In the Tahitian Group, as we learn from the -writings of Nadeaud and Drake del Castillo, Gardenia tahitensis thrives -much better on coral islands than on volcanic soils, and, in fact, -rarely quits the “_région madréporique_.” It is sometimes planted in -Polynesia near the houses, and both Nadeaud in Tahiti and Cheeseman in -Rarotonga consider that it was probably introduced into those islands -before the arrival of Europeans. The aborigines may have assisted in the -dispersal of the genus to a small extent, but from the presence of -peculiar species in Hawaii, Samoa, and Fiji it is apparent that the -genus is truly indigenous in the Pacific islands, and long antedated -their occupation by man. This is also evident from the station of the -species in Hawaii, Samoa, and Fiji. In Hawaii they may be found on the -dry forehills in the vicinity of the sea-border. In Samoa, as Reinecke -informs us, Gardenia tahitensis is very widely spread in the -mountain-forests, whilst the endemic species is found thriving in -inundated coast districts. In Fiji I found the Gardenias to be -especially characteristic (as is also pointed out by Horne) of the dry -districts on the leeward side of the larger islands. On the rolling -“talasinga” or “sun-burnt” plains of the north side of Vanua Levu they -thrive in numbers; and here their leaf-buds and the extremities of the -young shoots are often tipped or covered over with an amber-like -gum-resin which the natives chew. - -_The Mode of Dispersal of the Pacific Gardenias._—The fruits of this -genus are usually described as indehiscent. If this were true of Pacific -plants it would be very difficult to explain the dispersal of hard, dry -fruits an inch in size over this region. In the case of two or three -Fijian species, I paid especial attention to this point by examining the -plants in fruit. As exhibited in Fiji the fruits are globose, hard, and -almost stony, with persistent adherent calyx, the seeds lying -horizontally in a pulp at first firm and subsequently softening as the -fruit matures. The fruits are not as a rule to be observed opening on -the plant; but they are to be seen dehiscing septicidally on the ground -beneath, the detached woody valves being scattered around. If one of the -fruits gathered from the plant is kept soaking in water for some time it -will begin to dehisce; and this is probably what occurs with fallen -fruits in wet weather. Dr. Hillebrand regards the fruits of the Hawaiian -species as indehiscent. I did not myself examine them, but it is not -improbable that, like those in Fiji, they dehisce whilst lying soaking -on the ground. - -Judged merely from the dispersal standpoint, the fruits of the Fijian -Gardenias come near to those of Pittosporum, and both can be in a sense -described as baccate capsules. The flat, crustaceous seeds of Gardenia, -which are usually two or three millimetres in size, are also well fitted -for passing without injury through the digestive canal of a bird. It is -likely that the two genera have been dispersed in the Pacific by the -same kind of birds; and it should be remarked that their distribution is -somewhat similar, both belonging to the warm regions of the Old World. - -It might at first appear from some experiments of mine made in Fiji that -the dried fruits of Gardenia could be dispersed over oceans by the -currents. This receives some support by the preference for a littoral -station sometimes shown by G. tahitensis in Tahiti, and by the -occurrence of G. zanguebarica in the East African strand-flora -(Schimper’s _Ind. Mal. Strand-flora_, p. 131). It will, however, be -pointed out that currents could only have aided the dispersal of the -genus to a limited extent. The fresh fruits of Fijian species, with or -without the adherent calyx, have little or no buoyancy, and the seeds -sink even after drying for months. But it was ascertained that fruits -which had been kept for three months floated after four or five weeks’ -immersion in sea-water. On examination, however, it was found that the -valves gaped a little, being only held in apposition by the adherent -calyx, and that water had penetrated into the interior, the pulp being -in a state of decay. The fruits were, in fact, kept afloat in the latter -part of the experiment partly by the investing calyx and partly by gas -generated in the decomposing pulp. Ultimately they broke down altogether -and the seeds sank. In the “rough-and-tumble” of ocean-transport this -could scarcely be deemed an effective means of dispersal; and in the -open sea a fortnight would probably represent the limit of the floating -power. It is to the agency that has distributed the genus Pittosporum -over the Pacific that we must look for the explanation of the dispersal -of Gardenia over the same ocean, namely, to birds. - - - PSYCHOTRIA (Rubiaceæ). - -We find in this large genus of the Old and New Worlds a typical example -of the plants with fleshy drupes containing hard pyrenes that represent, -from the standpoint of dispersal, a common Rubiaceous type of plant in -the tropical Pacific. Such plants, of which those of Coprosma and -Nertera may be cited as other instances, are in a generic sense always -widely distributed in these islands. They are eminently suited for -dispersal by frugivorous birds; and it is a matter for surprise, -therefore, that in a genus like Nertera the solitary Pacific species has -such a wide range, whilst with Psychotria and Coprosma the numerous -species are usually restricted to particular groups. Genera doubtless -have their periods of development and decadence in the Pacific, and -probably Nertera is to be regarded as a decadent genus. These Rubiaceous -genera, however, appear to be well fitted for the investigation of the -centres of dispersal of particular genera and of their relative age. - -The Psychotrias in these islands are typically shrubs of the shady -woods, and they may be seen thriving best where the forest-growth is -rank and the humidity greatest. Their bright red ovoid drupes, which -range from eight to twenty-five millimetres in length (1/3 to 1 inch), -would readily attract birds, and their crustaceous pyrenes, that vary -between five and eight millimetres (1/5 to 1/3 inch) in length, would -pass unharmed through a bird’s digestive canal. That fruit pigeons can -distribute their seeds over the Pacific has been long established, and -Mr. Hemsley includes Psychotria amongst those genera which, from the -collections of fruits and seeds found in the crops of fruit-pigeons, -made by Professor Moseley, myself, and others, in the groups of the -Western Pacific, are “known to be dispersed by birds in Polynesia” -(_Introd. Bot. Chall. Exped._, p. 45). It is thus hardly necessary to -point out that neither the entire fruits nor the separate pyrenes could -be transported by the currents, my observations showing that in both -cases they sink at once or in a day or two. - -Psychotria, however, is an enormous genus including, according to the -_Index Kewensis_, some 600 or 700 described species, distributed in the -tropics all over the world, and also extending into subtropical regions, -the greatest concentration being in America. It is described in the -_Genera Plantarum_ as a polymorphous genus distinguished by no certain -characters from some other genera of the tribe of the Rubiaceæ to which -it has given its name. We have here a genus that has overrun the -tropical regions of the world, probably originating in America; and we -may contrast it with the relatively small Rubiaceous genus of Coprosma -(with its three score of species, and quite comparable with it from the -standpoint of capacity for dispersal), that, having its birthplace in -New Zealand, is only beginning to reach the mainlands of the New and the -Old World. - -One is a genus of the tropics and the other is a genus of south -temperate latitudes; and both have occupied the Pacific islands; but -Coprosma naturally finds its most appropriate station on the cool -uplands of Hawaii and Tahiti. We may ask, indeed, whether the great -contrast in the fecundity of the two genera, dispersed as they are in -the same fashion by the agency of frugivorous birds, is to be connected -with questions of relative antiquity or with geographical position. It -would certainly have been a more difficult task in the past, other -things being similar, for a New Zealand genus to stock the temperate -regions with its species than for a tropical American genus to overrun -the warmer regions of the globe. However that may be, the age of -dispersal of both genera is largely over now. - -A vast genus like Psychotria, that is not sharply defined from other -genera, presents difficulties to the systematic botanist which are -reflected in a complex synonymy; but there are certain broad facts which -the student of dispersal can gather for himself without much difficulty. -When we look at its distribution in the islands of the open Pacific, we -find that the genus attains its greatest development in the Western -Pacific, there being from thirty to forty species known from Fiji and -quite a dozen from Samoa, and that it shades away as we proceed eastward -and northward, some six species being recorded from Tahiti and the -Marquesas, two from Hawaii, and one from Juan Fernandez near the South -American mainland. The arrangement of the species shows fairly -conclusively that the genus Psychotria, as it is found in the Pacific, -has, like most of the other plants of this era of non-endemic genera, -been derived from the Asiatic side of the ocean. (The absence of species -of this genus from Mr. Cheeseman’s Rarotongan collections seems strange. -It is represented by some species in Tonga, and it is extremely probable -that it will be subsequently found also in the Rarotongan group.) - -That the age of dispersal of the genus Psychotria over the Pacific -islands has almost passed away is evident from the circumstance that of -the half-hundred species known from these groups, all but some four or -five are confined to particular groups. There is one species, P. -insularum, that ranges over the South Pacific from Fiji to the Tahitian -region; and there are two or three others that keep up a connection -between the adjacent groups of Fiji, Samoa, and Tonga, the last having -no peculiar species; but, apart from these indications, isolating -influences generally prevail. The two Hawaiian species are both endemic -and are only recorded from the island of Kauai, so that in that -archipelago there has not even been inter-island dispersal of the genus. -For Fiji it would seem from the _Index Kewensis_ and other authorities -that at least two-thirds of the species are confined to the group. Of -the dozen Samoan species only two or three are known outside the -islands. Four out of the five Tahitian species are peculiar, and the -only Marquesan species named by Drake del Castello is endemic. Even the -solitary species of Juan Fernandez is endemic, there attaining the -dimensions of a fair-sized tree. It forms the subject of an illustration -in Schimper’s _Plant-Geography_, page 491. - -Speaking generally, birds may be said to have almost ceased dispersing -this genus over the Pacific. This is not because birds have ceased to be -partial to the fruits, but because the frugivorous birds that used to -range over the Pacific archipelagoes now restrict their wanderings to -the limits of a single group. If we find occasionally in other parts of -the world, as in the occurrence of a Florida species of Psychotria in -the Bermudas, some evidence of a dispersal still in operation, this is -nothing more than we observe in the case of a few of the Polynesian -species now. The connection between birds and plants in the Pacific is -discussed in Chapter XXXIII. In this ocean the dispersal of the genus is -now practically dead, and Psychotria presents no exception to that -general tendency towards isolation and differentiation exhibited by most -genera of the tropical Pacific as the result of failure of the means of -dispersal. - - - CYRTANDRA (Gesneraceæ). - -This remarkable genus of shrubs, which forms the subject of an important -memoir by Mr. C. B. Clarke (_De Cand. Mon. Phan._ v. 1883-87), offers, -as Mr. Hemsley remarks, an example of a Malayan genus extending to -Polynesia and there developing numerous species. Of some 180 known -species, about 80 or nearly half are confined to Polynesia, the rest -being mainly Malayan. Of the Polynesian species about thirty are -Hawaiian, twenty Fijian, fifteen Samoan, and twelve Tahitian; whilst -solitary species are restricted to Tonga and Rarotonga respectively. - -The most significant feature in the distribution of this genus in -Polynesia is not only, as is pointed out by Mr. Clarke, that every group -has its peculiar species, but that very few species are found in more -than one group, and that even in the same archipelago each island has -its own species. Thus, of the thirty Hawaiian species, all of which are -peculiar to the group, only two or three, according to Hillebrand, are -at all generally distributed over the islands, whilst four-fifths have -not yet been found to be common to more than one island. So again, all -the species found in the Tahitian Group proper are peculiar, with the -exception of one extending to the neighbouring Paumotu Islands; and even -Rarotonga has its own species. In the region comprising Fiji, Tonga, and -Samoa the same rule prevails, only two or three species connecting the -three groups together. There thus seems to be not only a complete -suspension of the dispersal agencies between the various archipelagoes, -but also often between the several islands of a group. This is -particularly to be remarked with the relatively contiguous groups of -Fiji, Samoa, and Tonga, since with most other genera a number of species -are common to all three archipelagoes. “The polymorphism of the Hawaiian -Cyrtandras,” says Hillebrand, “is extraordinary: no single form extends -over the whole group, and not many are common to more than one island. -The variations affect nearly every part of the plant, and branch out and -intercross each other to such an extent that it is next to impossible to -define exact limits of species.” Genera, however, run riot in other -groups of the Pacific besides Hawaii, and Reinecke uses much the same -language with reference to Elatostema, an Urticaceous genus in Samoa, -attributing the wealth of forms to the sensitiveness of the plants to -the varying conditions of station (see Chapter XXVII). - -The behaviour of Cyrtandra in the Pacific is rather startling to the -student of plant-dispersal when he reflects on the suitability of the -berries for dispersing the plant through the agency of birds. That the -vegetation of oceanic islands should be of an endemic character is a -fact, remarks Mr. Clarke, that is illustrated by many other orders -besides the Gesneraceæ. But the point we have to remember is that not -only does the genus Cyrtandra display the same prolific character in the -large continental islands of Malaya, such as Java, Sumatra, and Borneo, -each of which possesses at least a couple of dozen species, but that -this seems to be a feature of the tribe Cyrtandreæ and of the whole -order. The genera, as observed by Mr. Clarke, are very continuous in -their areas of distribution, and in the tribe Cyrtandreæ there are very -few species that extend to more than one region, whether on the mainland -or in an oceanic archipelago. In the Himalayas, he says, closely allied -species of Didymocarpus are confined to single districts, although there -appears no reason either in soil or climate why they should not spread -to the adjacent valleys. - -There is therefore, we may infer, nothing peculiarly characteristic of -insular floras in this prolific display of the genus Cyrtandra in the -Pacific, except that it is rather more pronounced in an oceanic group -than in a continent. The same general cause is working alike in an -island in mid-ocean, in a large continental island bordering the -mainland, and on the mainland itself. With the Pacific Cyrtandras as -with the British species of Rubus the variability may be so great that -the ordinary agencies of dispersal fail to keep it in check; and when, -as in the Pacific islands, the suspension of the activity of these -agencies is complete, the formative energy of the species knows no -bounds other than the determining limits of station. Our lesson from the -Pacific Cyrtandras is therefore this. The isolation of the oceanic -archipelagoes may not explain the endemic character of the flora, but -only the extreme degree to which the endemism is carried. When a genus -is in its prime, it can defy all the limiting conditions imposed by -similarity of station and by free and unchecked means of dispersal, the -essential marks of a species or a genus having probably in their -development little or no connection with environment. - -The Cyrtandras of the Pacific Islands are most frequent where vegetation -is rank, as in moist woods, in humid valleys, and in shady ravines and -gorges; but they may also occur in more exposed and drier stations. They -often grow gregariously, and Schimper says the same of them in the Java -forests (_Plant-Geography_, pp. 291, 297). - -The fruit of the genus is described by Clarke as a fleshy or a -coriaceous berry. Almost everywhere in the Pacific groups the berry is -white and fleshy; but it is noteworthy that out of the nine Tahitian -species where the fruit is particularised by Drake del Castello, in two -cases it is designated a capsule and in seven a berry. It is in this -connection worth remarking that in Malaya other genera of the tribe -often have capsular or dry and coriaceous berries. The conspicuous white -berries of the Pacific species would readily attract birds, and their -minute roughened seeds scattered through the pulp might readily adhere -to their plumage or even be ejected unharmed in their droppings. As -respecting the capacity for dispersal, the Pacific Cyrtandras come near -the Hawaiian endemic genera of Lobeliaceæ with baccate fruits and minute -seeds. Speaking of Malayan genera of the tribe Cyrtandreæ, Mr. Ridley -says that their dry, dull-coloured, and inconspicuous corky fruits are -often devoured by animals. The seeds, on account of their roughened -surface, adhere to rocks and other surfaces and readily germinate. - - - FREYCINETIA (Pandanaceæ). - -If there is any genus of tropical plants to which the student of -distribution can look for guidance in the region of the Pacific, it is -to Freycinetia as dealt with by Dr. Warburg in his monograph on the -order (Engler’s _Pflanzenreich_, iv. 9, 1900). Its characters and its -distribution are well defined; and here, if anywhere, we might be able -to work out the history of a genus. In the words of the German botanist, -it stands quite apart from Pandanus and Sararanga, the two other genera -of the order. When Hillebrand was preparing his work on the Hawaiian -flora, more than a quarter of a century ago, only about thirty species -were known. Warburg’s list, excluding doubtful forms, comprises sixty -species, and even this number the author surmises will be doubled in -future years. The later investigators, however, have not materially -extended the range of the genus; and the statement of the botanists of a -generation ago, that it extends from Ceylon through Malaya and Australia -to New Zealand, and is found on almost every elevated island of the -Pacific, can only be supplemented by extending its area to the Asiatic -mainland in Burma where a wide-ranging Malayan species exists. - -It is, however, remarkable that no endemic species can be with certainty -accredited to the mainland of Asia either in Burma or in the Malay -peninsula where the genus also occurs. The Malayan region from Java to -the Philippines possesses quite three-fifths of the species, and it is -singular how few wide-ranging species there are. The Philippine Islands, -Borneo, Celebes, Sumatra, Java, New Guinea, &c., have all their own -species, the only wide-ranging plant being Freycinetia angustifolia, -which occupies the region from Burma to Java and Borneo. So also in the -Pacific, there is no widely distributed species, every group possessing -its own plant or plants, and there does not appear to be any Freycinetia -that is common to two groups. Thus, Hawaii and Tahiti each have their -own species. Rarotonga, according to Cheeseman, owns a peculiar but not -yet fully described form. Samoa has two and Tonga has one species. -Westward from Tonga and Samoa the numbers of species increase, Fiji -possessing five and New Caledonia four. Australia and New Zealand each -claim two species as their own. - -Dr. Warburg, who has studied the genus in its home, remarks on page 43 -that none of the species possess any means of dispersal enabling them to -cross an ocean; and he connects with this the fact that the genus is -only found (to use his own words) on islands like those of Samoa, -Tahiti, and Hawaii, that possess a “palæobiotic” nucleus -(_paläobiotischen Kern_) and not on islands like the Bonin Islands of -new formation (_auf Neubildungen_). This attitude towards the problem of -plant-distribution in the Pacific is backed by a great experience; but -it is one, of course, that is directly opposed to the line of argument -followed in these pages; and it is needless to say that it is not -encouraging to the student of plant-dispersal. Yet one could hardly look -upon the islands of the Tongan Group with their representative of the -endemic Freycinetias as of more ancient origin than the Bonin Islands -that have none; and plants that find their homes on the peaks and in the -forests of mountainous islands would rarely find a suitable station on -the low coral islands of the Pacific. It is, however, noteworthy that -Professor Schimper is inclined to include a species of Freycinetia as -amongst the strand-flora of the coral islands of the Java Sea (_Ind. -Mal. Strand-flora_, p. 134). With regard to the question of the means of -dispersal of Freycinetias, it will at once be shown that these plants -possess many opportunities for dispersal by birds. - -Though in our own time dispersal by birds between the various Pacific -archipelagoes is often largely suspended, the inter-island dispersal in -each group is usually active through the agency of birds, now like the -plants they distribute confined to each group. Thus with Freycinetia we -find that, notwithstanding that each Pacific group is, as regards this -genus, isolated from the others, the separate islands, as in the case of -those of Hawaii, may possess a common species dispersed over the area. -The ripe fruit, which consists of a number of berries in a head or -spike, is juicy and pulpy, and contains in each berry a large number of -minute oblong or fusiform seeds, usually one or two millimetres long and -possessing thick toughish tests. Birds, indeed, are fond of pecking at -the ripe fruit-heads in Hawaii. Thus we learn from the _Aves -Hawaiienses_ of Wilson and Evans that a Grosbeak (Psittacirostra) and -the Hawaiian Crow (Corvus tropicus) feed principally on ripe Freycinetia -fruits, the seeds having been often found by Mr. Wilson in the stomach -of the former bird. No doubt these birds distribute the seeds over the -islands of the group. Mr. Perkins tells me that the Grosbeak is found -unmodified all over the group, and that it no doubt frequently gets -carried _nolens volens_ from one island to another. In his memoir on the -birds in the _Fauna Hawaiiensis_, he remarks that the essential food of -the “Ou,” the native name of this bird, is the fruiting inflorescence of -Freycinetias. The “Oo” (Acrulocercus) and the Hawaiian Crow above -mentioned, as he also observes, feed on these ripe red fruits. Like Mr. -Wilson, he sometimes found the Crow absolutely filled with this food to -the exclusion of all others (see Chapter XXXIII). Facts of a similar -kind came under my notice whilst in these islands. Thus on one occasion -I observed, on a leaf below a fruit-head that had been partly eaten by a -bird, a pellet half an inch long composed entirely of Freycinetia seeds -well soaked with the gastric juices and apparently only recently -disgorged. Sir W. Buller refers to different New Zealand birds, as the -Banded Rail (Rallus philippensis), the Kaka Parrot (Nestor -meridionalis), and the “Tui” (Prosthemadera), that live on the “sugary -flowering spadices” of Freycinetia Banksii. One can legitimately suppose -that they also attack the juicy berries. It is singular that as we learn -from Dr. Warburg (p. 17), Flying-Foxes (Pteropidæ) feed on the flowers -and top-leaves of many species of Freycinetia, and he considers that -they would aid in fertilisation by carrying about the pollen in the hair -of the head. Here again it would seem to us highly probable that whilst -brushing past a ripe fruit-head these bats might readily carry away in -their fur some of the minute seeds, which in the fresh berry are -“sticky” or adhesive. - -Just as it was possible in the case of Coprosma in the South Pacific -(see page 296) to connect its distribution with the range of the Purple -Water-Hens (Porphyrio), so it may perhaps be legitimate to associate the -range of Freycinetia over Polynesia with the distribution of the -Honey-Eaters (Meliphagidæ) in the Pacific, a family sometimes possessing -peculiar genera as in New Zealand and Hawaii, and one in which the -species have usually a very confined range, being sometimes limited to a -single island (Newton in _Encycl. Brit._ xii. 139). To this family -belongs the New Zealand “Tui” above mentioned; and it may be remarked -that these birds as a rule feed on soft fruits, such as figs, and -bananas. It is to Acrulocercus, one of the Hawaiian genera of the -Meliphagidæ, that Mr. Perkins refers me, on my asking him to name some -of the fruit-eaters in that group. - -These climbing shrubs, as Dr. Warburg observes, mostly frequent the -tropical forests up to 4,000 feet and over. Though their most familiar -habit is as tree-climbers in the forests, in localities where there are -no trees they adopt a trailing habit and cover mountain peaks and ridges -with a dense growth to the exclusion of almost all other plants. Many a -peak in the Pacific islands would be inaccessible if it were not for the -dense growth of these plants on their precipitous sides. It was owing to -the friendly aid of a tangled mass of Freycinetia stems that Lieutenant -Heming and myself were able to clamber to the summit of Fauro Island -(1,900 feet) in the Solomon Group, where I discovered a tree that under -the name of Sararanga forms the type of the third genus of the -Pandanaceæ. - -Whilst describing their station, it will be of interest to also record -the altitudes at which these plants have been observed in the tropical -Pacific. Since they can be independent of trees and are as much at home -on treeless rocky peaks and mountain crests, the upper limit would -usually be determined by climatic conditions, abundance of rain and -great humidity being the chief requisites; but, as will be seen below, -this limit does not seem to be reached in the tropical islands of the -South Pacific except perhaps in Tahiti. In the Fijis the Freycinetias -ascend to the highest mountain peaks. Thus, three of the species -discovered here by Seemann were found at elevations of about 4,000 feet -on Voma Peak in Viti Levu and in the highlands of Taviuni. In Vanua -Levu, as I found, they cover the highest peaks 3,500 feet above the sea. -They are especially abundant on the lofty mountain ridges, and clothe -the higher slopes of the Mbatini Ridge which terminates in the highest -peak of the island. In no locality did I find them growing in such -densely tangled masses as on the long ridge-like crest that forms the -upper part of Mount Freeland, 2,740 feet above the sea. For more than an -hour in order to reach the summit I had to clamber along the crest of a -ridge covered with a dense growth several feet deep of these trailing -plants, without touching the ground beneath. - -In Samoa, as we learn from Reinecke, Freycinetias are common on the -mountain ridges, climbing the trees and forming also a dense undergrowth -covering the ground and concealing the rocks. They occur at all levels -from 1,000 feet above the sea up to the highest region of Savaii, rather -over 5,000 feet in elevation. In Rarotonga, according to Mr. Cheeseman, -the Freycinetias are very abundant on the mountains, which reach a -height of 2,200 feet, the plants scrambling up the trunks of trees or -over rocks and frequently rendering the forest almost impenetrable. In -Tahiti, Nadeaud tells us, the Freycinetias often cover in an -inextricable network the sides of the valleys at elevations of 2,000 to -3,300 feet, extending in their vertical range from the lower levels of -the island to the highest inaccessible peaks which attain a maximum -height of about 7,300 feet. - -These plants in the Hawaiian group are common in the lower woods as -Hillebrand informs us, that is to say, at elevations of 2,000 or 3,000 -feet. During my descent from Mauna Kea through the Hamakua forests on -the north-east side I observed that the Freycinetias commenced at an -altitude of 3,900 feet, and that they attained their greatest -development between 3,200 and 2,000 feet. These plants ascended quite a -thousand feet higher on these mountain slopes than the Bird’s Nest Fern -(Asplenium Nidus), which reached an altitude of 2,800 feet. In the -forests on the west side of Mauna Loa they were abundant at altitudes of -3,500 to 4,000 feet and were not noticed above 4,500 feet. On the slopes -of Mount Eeka in West Maui they abounded between 3,500 and 4,400 feet. -In those localities where the forest descends to the sea, Freycinetias -occur at the coast, and on Oahu they are often found at elevations under -a thousand feet. - -I have but few data showing the altitude obtained by Freycinetias in -other regions, as, for instance, in their most southerly habitat in New -Zealand, where they give a tropical luxuriance to the forests, or in -their chief home in Malaya. From Schimper’s observations -(_Plant-Geography_, p. 293) it would seem that they thrive in the Gedeh -forest of Java at elevations of about 5,000 feet. Except for the lower -levels, Warburg makes but few references to this subject in dealing with -the species. It appears to me that some very interesting results might -be obtained by comparing the vertical range of this genus in different -regions, as, for instance, in New Zealand and in Borneo or in Java. We -might get indications that since the age of Freycinetia began the -climate in tropical latitudes has been getting warmer, and that the -erstwhile plants of the lower levels are now as a result climbing the -mountain slopes. The student of distribution may find here a genus that -has been “cornered” not only in space and time, but as regards its -conditions of existence. Since it is obvious that during a gradual -increase of temperature it would ascend the mountains and during a -lowering of temperature it would descend to the plains, it follows that -in the mountains of an oceanic island it might be driven into the sea or -await extinction on a mountain-top. In the tropics also there would be -no escape during a gradual increase of temperature. Here again it would -make its last stand on the strand, and, forced to choose between Death -and Adaptation, the genus might select the latter alternative and -present us with a startling new form. In this sense Freycinetia seems to -offer itself as “fair game” for the speculative botanist, and at all -events he will be able to interrogate it as to the connection between -its existing range of altitude and the climatic conditions of the -earlier phases of its history. - -The Freycinetias bear the same name over Polynesia, “ie-ie” in Hawaii, -“ie” and “ie-ie” in Tahiti and Samoa, which appear in their full form in -the Rarotongan and Maori “kie-kie.” The secret of the wide distribution -of the name lies in the circumstance that this is a mat-word over much -of Polynesia, as in Fiji, Tonga, Samoa, the Gilbert group, Tahiti, &c., -Freycinetia leaves being often employed for making mats, as in Samoa and -New Zealand. The same word is applied in some groups to small species of -Pandanus that were also used in mat-making. Thus in Fiji “kie-kie” was -not only the name for a mat-dress, but also of Pandanus caricosus that -supplied the material. In the home of the Polynesians in Malaya and its -vicinity the same word for mat and Pandanus occur. Thus, “gerekere” in -the Motu dialect of New Guinea and “keker” or “kekel” in Amboyna are the -names of small species of Pandanus employed in mat-manufacture; whilst -“kihu” and “kiel” in Celebes are the words for the mats themselves. -Therefore in one form or another the word, originally applied to the -mats, but now often restricted to the plants from which the materials -were derived, ranges over the great region extending from Malaya to New -Zealand, Tahiti, and Hawaii, and, as I have shown in the table given in -my paper on Polynesian Plant-Names (_Journ. Victor. Inst._, London, -1896), it may be traced even to Further India, as in Annam, and to -North-East Australia. It thus covers the area to which the migrations of -the Polynesians of the Pacific have been confined, and it covers also -the area of the genus Freycinetia. There is something far more than mere -analogy between man and plants in their occupation of the Pacific -islands. The plants are Malayan and the Polynesians are from Malaya -also, whilst in both man and plants we experience the same difficulty in -explaining their dispersal over the ocean. Divesting his mind of all -previous conceptions, the ethnologist might profitably study _de novo_ -the dispersion of man in the Pacific from the standpoint of -plant-dispersal (see Chapter XXVIII). - - - SAPINDUS AND PHYLLANTHUS. - -Brief reference can alone be made to these two genera. Foremost comes -Sapindus, which is represented by two endemic species, one in Hawaii and -one in Fiji, and by another species, found in Tahiti, the Marquesas, and -Easter Island, which is identified by some botanists with the well-known -American “soap-tree,” S. saponaria. There are several difficulties -connected with the presence of this genus of the Old and New World in -the Pacific. Not the least of them is connected with the transport of -the large seeds of this genus, an inch in size, to the isolated Hawaiian -Group, where it is represented by a solitary endemic species in the -island of Oahu. The fleshy mesocarp of the fruits might attract birds; -but it is not easy to perceive how birds could carry such large seeds -over some 1,500 or 2,000 miles of ocean. Yet the same difficulty exists -with a few other genera, such as Osmanthus and Sideroxylon, that are -only represented in Hawaii by endemic species, genera which require the -agency of birds to explain their occurrence unless we wish to postulate -a continental connection for this group. (See under those genera in -Chapter XXVII.) - -The large Euphorbiaceous genus Phyllanthus, spread universally over the -tropics and containing some 500 known species, clearly indicates by its -distribution in the Pacific islands that genera with dry fruits, such as -are typical of the order, are as widely distributed and just as much at -home in these islands as the genera with fleshy fruits, such as -Psychotria and Cyrtandra. The small trees and shrubs of Phyllanthus are -common in dry, open, partially wooded districts near the sea-border. The -genus attains its greatest development in this ocean in New Caledonia -and Fiji; and since the number of species diminishes the further we -penetrate the Pacific, it can be scarcely doubted that the genus has -entered this ocean from the west. In Fiji there are at least 20 species, -of which probably half are not recorded from elsewhere. In Samoa there -are seemingly but few peculiar species. In Hawaii there is only one -indigenous species, and that is endemic. The genus, however, has -developed a lesser centre of distribution in East Polynesia, there being -about a dozen species known from Tahiti and the Marquesas, of which half -are peculiar to one or other of those groups. From experiments made by -me in Fiji on the fruits and seeds of two species it was evident that -they possessed little or no capacity for dispersal by the currents. We -look, therefore, to the birds, and in this connection it is of interest -to note that this genus is included amongst those known to be dispersed -by birds in the Pacific, some of the fruits having been found in the -crops of fruit-pigeons shot by Prof. Moseley in the Admiralty Islands -(_Bot. Chall. Exped._, Introd. 46; iv. 308). - - - PRITCHARDIA (Palmaceæ). - -This genus of Fan Palms supplies an instructive lesson for the student -of plant-distribution, more especially with reference to the loss of the -endemic reputation of a genus. Regarded by the earlier botanists who -visited the Pacific as identical with the familiar Asiatic Talipot Palm -(Corypha umbraculifera), the Fan Palms of this region, as represented in -Fiji and Hawaii, were subsequently placed by Seemann and Wendland in a -new genus restricted to Polynesia and named after a former British -Consul in Fiji. Since that time it has lost its reputation as a -peculiarly Pacific genus, since a species (Pritchardia filifera) has -been found lingering in a few valleys in Arizona, where it enjoys the -distinction of being the most northerly in station of all the world’s -palms (Linden in _Illustr. Hort._ vol. 24, 1876-77). It would thus -appear that the Pacific islands have derived this genus of palms from -the western part of North America, but the whole question is beset with -many difficulties, and not the least is that connected with the -confusion that seems to reign in several cases as regards the allocation -and identity of the species. - -Six species are named in the _Index Kewensis_, viz.: Pritchardia -macrocarpa, restricted to Hawaii; P. martii and P. gaudichaudii, of the -Pacific islands; P. pacifica, assigned to Fiji; P. vuylstekeana, from -the Paumotus; and P. filifera, from the west side of North America. -Though it is sometimes difficult to reconcile this account of the -distribution of the genus in the Pacific with views held by other -botanists, it offers the safest basis for the future investigation of -the subject. It would be, however, necessary to remember that -Pritchardia gaudichaudii and P. martii are regarded by Hillebrand as -peculiar to the Hawaiian Islands, and that the exact locality of the -Paumotu species is not very definitely settled, if it depends on the -remarks made on this species in the _Gardeners’ Chronicle_ for 1883. No -mention is indeed made by Drake del Castillo of any Tahitian or -Paumotuan species. - -Whilst in Hawaii and Fiji I was much interested in these palms, and the -following remarks are merely intended to be a contribution to the -subject. According to Seemann, Hemsley, Drake del Castillo, and Burkill, -Pritchardia pacifica, which often attains a height of thirty to -thirty-five feet, occurs in Fiji, Tonga, Samoa, and the Marquesas, but -it does not exist in Tahiti, and Cheeseman does not include it in the -Rarotongan flora. Except in the Tonga Group, where, according to Lister -as quoted by Hemsley, the palms form conspicuous objects along the -weather shore of the island of Eua, this species is rarely found in the -wild state in the South Pacific. This especially applies to Fiji, as Mr. -Horne also observes; and at most one is accustomed to see (to employ the -words of Dr. Seemann) one or two trees outside a village which are -reserved, as in many parts of Polynesia, for the use of the chiefs who -employ the leaves for fans and for other purposes. But even this reason -for preserving the palms scarcely now exists in Fiji, and at the time of -my sojourn in Vanua Levu (1897-99) the trees were rare enough to be -regarded as curiosities. In the Marquesas, according to Bennett (quoted -by Seemann), they grow in groves in the valleys of the interior. Dr. -Reinecke does not even include the species in the Samoan flora, but -mentions it with the Date-Palm (Phœnix dactylifera) as if it were -recently introduced. It was, however, found in that group by the United -States Exploring Expedition about 1840, and this is evidently the palm -referred to by Captain Cook as existing at his time in the Tongan Group. - -The Hawaiian species of the palm appear to be three in number, -Pritchardia gaudichaudii and P. martii, both regarded by Hillebrand as -confined to the group, and P. macrocarpa of Linden, also endemic -(_Illustr. Hort._ vol. 26). The two first-named species are evidently on -the road to extinction in the wild state, and often find their last -refuge on rocky, almost inaccessible, inland cliffs. Pritchardia -gaudichaudii, about twenty feet in height, is found in the wild state, -as we learn from Hillebrand, on the islands of Molokai and Hawaii. It -was at one time frequently met with near native dwellings; but during my -sojourn in 1896-97 on the last-named island it was not at all frequent, -and as a rule only came under my notice occasionally in clumps of three -or four trees on the Kona and Puna coasts, as near Kiholo, Milolii, and -Kalapana. However, it was more frequent in the Waimanu district of -Kohala in the same island. Here I noticed it growing in clumps in -precipitous rocky situations at elevations ranging from 1,200 to 2,000 -feet. The other palm mentioned by Hillebrand, P. martii, is only five or -six feet high, and is confined mostly to Oahu and Molokai. - -The agency of man in introducing these interesting Fan-Palms into the -Hawaiian Islands seems out of the question, since they are home -productions in a specific sense and are doubtless ancient components of -the flora; and, of course, grave objections exist on ethnological -grounds, if this genus had originally its home in America. With -reference, however, to Pritchardia pacifica of the South Pacific, it is -not unlikely that man has aided in the distribution of a palm mainly -preserved by planting in and about the villages and set apart from time -immemorial for the use of the chiefs. - -In this connection the aboriginal names are of some importance and may -be very briefly here referred to. The Fijian “Viu,” the “Piu” of Samoa, -Tonga, and Futuna, and the Tongan “Biu” are forms of the same name -applied to this palm all over West Polynesia; and I have shown in my -paper on Polynesian Plant-Names that in the form of “Firo” in the -Solomon Islands (Bougainville Straits) and of “Wiru” in Sundanese, one -of the Malayan languages, the same name is given to another genus of Fan -Palms, namely, Licuala. But since these West Polynesian names do not -always conform with the laws of consonantal interchange in this region, -they cannot all be considered as indigenous in the languages concerned. -If, for instance, “Viu” is an indigenous Fijian name, as no doubt it is, -since it follows the phonetic laws affecting the Malayan and Fijian -languages, “Piu” must be a foreign word in Samoa and Tonga, and “Biu” -must be another introduced Tongan name.... The Fijians have in “Sakiki” -(contracted into “Saii” in the Somosomo dialect) another name for this -palm. This is probably derived from “Kiekie,” a mat-word in different -forms in various Polynesian groups, and applied in many islands to the -plants that supply the materials for mat-making, such as Pandanus and -Freycinetia. - -The Hawaiian generic name of “Loulu” for these palms appears to be quite -local; but it may possibly have a common origin with “Roro,” one of the -Fijian names of Cycas circinalis. It is pointed out by Hillebrand that -the Hawaiian name of the edible kernels of these palms, “Hawane” or -“Wahane,” occurs in the Marquesas as “Vahana” applied to the palm, a -comparison that is on linguistic grounds quite legitimate. “Vaake” is -another Marquesan name, which recalls “Vakoa,” the Malagasy word for -Pandanus. - -When we compare the variety of the names of the Pritchardia fan-palms in -the Pacific Islands with the prevailing uniformity of the names of -cultivated plants transported by the aborigines in their migrations from -Malaya, such as the taro, the yam, the sugar-cane, the coco-nut, and the -Malay-apple, we perceive that the testimony of the names points to the -same conclusion as the botanical evidence, namely, that the ancestors of -the Hawaiians found these palms in the group at the time of its -occupation. In the South Pacific much uncertainty prevails. The -ancestors of the West Polynesian peoples evidently brought the word for -a fan-palm from their Malayan home; but it is doubtful if they found -Pritchardia already established in all the islands; and the apparent -home of the genus in America prevents us from attributing to a palm, -that is by some botanists regarded as confined to the Western Pacific, a -home in the neighbouring regions to the west. There is thus a lack of -agreement between the botanical and ethnological indications as regards -the original American origin of Pritchardia in the South Pacific. - -There remain then the agencies of the currents and of birds. A singular -feature in the distribution of the Hawaiian species, Pritchardia -gaudichaudii, at once affords a clue as concerning the dispersal in the -North Pacific. Dr. Hillebrand remarks that this palm covers part of Bird -Island, a small volcanic rock forming an outlier of the Hawaiian group -about 400 miles north-east of Kauai. Here the agency of birds is -suggested, since it is scarcely likely, though, as shown below, not -impossible, that stranded fruits of the palm could have established -themselves in this fashion. Mr. Perkins has an interesting note on the -food of Ciridops anna, an Hawaiian bird, now nearly extinct, that feeds -principally on the blossoms and unripe fruits of the Loulu palms, -probably of this species. The drupes when fresh have a somewhat fleshy -mesocarp and are about 9/10 of an inch (22 mm.) across, and their -crustaceous inner shell would undoubtedly fit the seeds for dispersal by -frugivorous birds like pigeons. The fruits of the other two Hawaiian -species are considerably larger, that of P. macrocarpa being, according -to Linden, of the size of a nut of Juglans regia, that is, about 1-1/8 -inch or 29 mm., whilst that of P. martii, as we learn from Hillebrand, -is from 1-1/2 to 2 inches or 37 to 50 mm. Allowing for the variation in -size of the fruits within the limits of the genus, there need be no more -difficulty in assuming that the original species had fruits that could -have been brought by birds, than in holding that the fruits of -Elæocarpus have been carried to Hawaii in the same fashion. The drupes -of Pritchardia pacifica are barely half an inch in diameter. They are -fitted by reason of their hard crustaceous endocarp for dispersal by -fruit-pigeons; and I may here add that these birds are known to -distribute the fruits of other palms, such as Kentia and Areca, in the -islands of the South Pacific (_Bot. Chall. Exped._ iv. 308, 312). - -Both in Hawaii and in Fiji I experimented on the capacity of Pritchardia -drupes for dispersal by the currents. Those of the Hawaiian species, P. -gaudichaudii, have when well dried a light buoyant rather fibrous -mesocarp which enables them to float in the case of a good proportion of -the fruits for at least five weeks. I had no opportunity of testing the -buoyancy of the fruits of P. martii, another Hawaiian species; but, -judging from the existence in the coats of a fibrous layer as described -by Hillebrand, they ought to display some floating power. The fruits of -P. pacifica, the South Pacific species, lack the light buoyant covering -of the Hawaiian species above referred to, and display little or no -floating power even after drying for weeks. Looking at the results of -these experiments, it would seem that it is not impossible that Hawaii -received the genus through the agency of the currents; but it seems -scarcely probable, since it could only have been derived from America, -and the American species grows in the interior of the continent and not -near the sea-border. The possibility of course exists; but I am inclined -to attribute the presence of Pritchardia in Hawaii to bird-agency. - -My position from the standpoint of dispersal with regard to Pritchardia -in the Pacific is this. The Hawaiian species I would consider as -American in origin. The Marquesan species, unless recently described, -still awaits detailed investigation. The West Polynesian species of Fiji -and Tonga, according to the principles of distribution prevailing in the -South Pacific, ought to hail from the west. - - - _Summary._ - -(1) Whilst the earliest age characterised by the Coniferæ was restricted -to the Western Pacific, and whilst the following age of the Compositæ -and Lobeliaceæ, mainly American in their affinities, was concerned with -the regions of Hawaii and Tahiti, we have now to discuss the Malayan era -during which the bulk of the plants were derived from the nearest -tropical regions of the Old World. Here we have to deal with the -low-level flora of Hawaii, that is to say, with the plants of the levels -below 4,000 or 5,000 feet, and with almost the entire floras of the -areas of Fiji-Samoa and of East Polynesia. The whole of the tropical -Pacific is here concerned, and not a portion of it, as in the two -preceding eras; and in our comparison we shall see that there are two, -and not as heretofore three, regions to be regarded—the Hawaiian in the -North Pacific, and the whole Polynesian area of the South Pacific -extending from Fiji to Tahiti. - -(2) Here the frugivorous bird has been the principal agent in dispersing -the plants, quite two-thirds of the genera possessing drupes or berries -that would attract such birds. - -(3) The genera representative of the first part of this era are those -which have only peculiar species in Hawaii, and are composed in the -South Pacific either entirely of peculiar species or sometimes of a -mixture of endemic and non-endemic species. It is an era of complete -isolation in Hawaii and often of a partial connection between the groups -of the southern region. Except to some extent in the South Pacific, the -dispersing agencies are now no longer active between the groups. - -(4) Amongst the genera typical of this period are Pittosporum, Gardenia, -Psychotria, Cyrtandra, and Freycinetia. - -(5) The two genera of the Rubiaceæ, Psychotria and Coprosma (the last -belonging to the mountain-flora), appear to be well suited for the -investigation of the effect on distribution of the geographical position -of the home of the genus, the first with 600 to 700 species distributed -over the tropics of the Old and New Worlds, the second with some sixty -species having its home in New Zealand. - -(6) From the Pacific Cyrtandras we derive the lessons that the display -of great formative power in a genus may not be a peculiarity of an -insular flora; that the isolation of an oceanic archipelago does not -necessarily induce “endemism,” but merely intensifies it; and that the -production of new species within the limits of a genus like Cyrtandra -may be nearly as active on the mainland as in an island in mid-ocean. - -(7) From the Freycinetias we learn that it may be possible to connect -the distribution of a genus of plants with that of a genus or a family -of birds. Just as in Chapter XXIV we endeavoured to connect Coprosma and -Porphyrio (the Purple Water-Hens), so we here suggest a connection, in -their range over the Pacific, between the Freycinetias and the -Meliphagidæ (the Honey-eaters), a connection that in the last case at -least belongs to the past. - -(8) From the genus Phyllanthus we learn that genera with dry fruits may -be as widely distributed and may display the same formative power in the -Pacific as those with fleshy fruits that would seem much more likely to -be dispersed by birds. Here again we obtain an indirect indication that -species-making in these islands is not altogether dependent on -isolation. - -(9) In the case of the genus Sapindus we are apparently compelled to -infer that its large seeds (in the present species an inch in size) have -been transported by birds to Hawaii. Yet in point of size the -difficulties here raised are no greater than those arising from the -existence of such genera as Sideroxylon and Elæocarpus in Hawaii, the -fruits of which are known to attract frugivorous birds. - - - - - CHAPTER XXVI - - THE MALAYAN ERA OF THE NON-ENDEMIC GENERA OF FLOWERING PLANTS - (_continued_) - - THE AGE OF WIDE DISPERSAL OVER THE TROPICAL PACIFIC (_continued_) - -The widely dispersed genera that are as a rule not entirely - represented by endemic species in any - archipelago.—Elæocarpus.—Dodonæa.—Metrosideros.—Alyxia.—Alphitonia.—Pisonia.—Wikstrœmia.—Peperomia.—Eugenia.—Gossypium.—The - last stage in the general dispersal of plants of the Malayan - era as illustrated by the widely-dispersed genera having - as a rule no peculiar - species.—Rhus.—Osteomeles.—Plectronia.—Boerhaavia.—Polygonum.—Pipturus.—Dianella.—Summary. - - -A LATER period in the era of the general dispersal of Malayan plants -over the Pacific is indicated by those genera that as a rule are never -entirely represented by endemic species in any archipelago. Hawaii now -comes into touch with the world outside, and all the groups possess some -connecting link. But the beginning of the effect of the isolating -influence is shown in the association in each principal archipelago of -peculiar species with those that occur in other groups. - -We see here illustrated in all but the final stage that process by which -a solitary widely-ranging species, alone representing its genus, becomes -ultimately in each group the parent of a number of peculiar species. The -polymorphous, or extremely variable, species plays in this period the -all-important part. The earliest stage is exhibited by such genera as -Alphitonia, Dodonæa, Metrosideros, Pisonia, and Wikstrœmia, that possess -in the tropical Pacific a solitary widely-ranging species, varying -independently in every group and giving rise to forms that, in their -degree of differentiation, sometimes approach a specific value. Later -stages are shown when the polymorphous species, having done its work of -distributing the genus, settles down and “differentiates” in every -group; and this we see now illustrated in the genera Elæocarpus, Alyxia, -Peperomia, and others. - -The bulk of the genera of this period, of which only a few can be -mentioned here, hail from the tropics of the Old World through Malaya. -Thus Alyxia, Elæocarpus, Morinda, and Wikstrœmia are Malayan; whilst -genera like Eugenia, Peperomia, and Pisonia, that occur in the Old and -New Worlds, can similarly be traced to the Asiatic side of the ocean by -the distribution of their species. Others again have their home in New -Zealand like Metrosideros, or in Australia, as with Dodonæa and Scævola. -None are exclusively American. Some of the genera, as Morinda and -Scævola, have littoral as well as inland species; but, as shown in -Chapter XIV, there is rarely anything to suggest a derivation of the -inland from the coast species, both being, from the standpoint of -dispersal, of independent origin. - -About half of the plants have fleshy or sappy fruits (drupes and -berries) that would attract frugivorous birds, such as we find in -Xylosma, Elæocarpus, Eugenia, Scævola, Wikstrœmia, &c., whilst the -others have often dry capsular fruits, with minute seeds as in -Metrosideros, or with larger seeds as in Dodonæa. Some of them, like -Pisonia, have fruits that excrete a viscid material that causes them to -adhere firmly to plumage. Birds both granivorous and frugivorous have -been actively at work; and there are few difficulties relating to -dispersal connected with the genera, except with such as Gossypium and -Elæocarpus. - -I will adopt the method employed in the preceding chapter of discussing -in detail from the standpoint of dispersal some of the genera that came -most frequently under my notice, or in which I am greatly interested, -and of dealing briefly with some of the rest. Those dealt with in other -connections will not be treated. - - - ELÆOCARPUS (Tiliaceæ). - -This is a genus of trees containing, according to the _Index Kewensis_, -about 130 species, most of which are confined to tropical Asia, -including Malaya; but a fair number occur in the Pacific region, in -Australia, New Zealand, and the islands of the tropical Pacific, and the -genus is also found in Japan. It will thus be seen that Elæocarpus is -not only a continental but also a typical insular genus. It has reached -not only some of the most isolated island-groups of the Pacific, but it -is to be found also in the smaller islands of the Indian Ocean, there -being an endemic species in Mauritius. Amongst the Pacific Islands, a -region with which we are more immediately concerned, it has been -recorded from the Solomon Islands, New Caledonia, Fiji, Tonga, Samoa, -Rarotonga, and Hawaii. It is strange that the genus is not accredited to -Tahiti, but since it is represented in Rarotonga we may regard it as not -altogether absent from East Polynesia. Reinecke does not include it -amongst the Samoan plants, but Horne, in a short list of plants -collected in Upolu about 1878, mentions Elæocarpus græffei, a Fijian -species (_Year in Fiji_, p. 285). - -New Caledonia represents the principal centre of the genus in the -tropical Pacific, thirteen species being accredited to it in the _Index -Kewensis_. Seemann found six species in Fiji, a number that does not -seem to have been added to by Horne. Of these one is found in Tonga and -Samoa, and of the rest perhaps most are peculiar; but one of them is -closely allied to a second peculiar Tongan species. Tonga possesses the -two species just alluded to, whilst Rarotonga and Hawaii have each a -peculiar species. - -From an interesting comparison made by Mr. Burkill of some of the -Polynesian species, it would seem that Elæocarpus, if not actually -possessing a widely-spread polymorphous species in the tropical Pacific, -presents us with the next stage in the differentiation of the species. -Thus, he says in his paper on the flora of Vavau that an endemic Tongan -species, E. tonganus, is allied to three different species—E. græffei -from Fiji, E. floridanus from the Solomon Group, and E. glandulifer from -Ceylon—three species, he remarks, which are “so closely allied that it -is possible to regard them as insular subspecies.” It would thus appear -that some of the species of the Western Pacific are almost in touch with -Asiatic species. It would be of importance to determine whether some -affinity can be detected between the species of this part of the Pacific -and some of the widely-ranging species of Indo-Malaya, such as E. -ganitrus and E. oblongus. Mr. Burkill goes on to say that the solitary -Hawaiian and Rarotongan species are closely allied, an inference which -is of interest as indicating the route by which Hawaii received its -species. The genus, we may fairly infer, once possessed a widely-ranging -polymorphous or very variable Asiatic species in the tropical Pacific; -and we see it now in the next stage of specific differentiation in -various far-removed regions. In this connection Seemann significantly -remarks that all the Fijian species are evidently very local in the -group. - -It will be appropriate here to refer briefly to the station and mode of -occurrence of the species. They occur most typically as forest-trees, -often of considerable height. In New Zealand, according to Hochstetter, -they form a feature in the temperate rain-forest; and, as we learn from -Kurz, they are similarly conspicuous in the tropical rain-forests of -Pegu. To this seeming indifference to the varying thermal conditions of -different latitudes we shall have subsequently to refer again. The tree -of the Hawaiian Group, as Hillebrand tells us, is common in the forests -of Oahu and Kauai, but is scarce in Maui and Hawaii, a singular -distribution that may be due to the inflorescence being “often -monstrously deformed by oviposition of some dipterous insect.” The -Rarotongan species, according to Cheeseman, is common throughout the -island from the sea-level to the tops of the hills. In Vanua Levu I -found that these trees preferred the crests of wooded mountain-ridges or -the partially vegetated mountain peaks. They came under my notice in the -forests of the island of Fauro, in the Solomon Group, associated with -other large trees of the genera Canarium and Calophyllum. - -Much interest is attached to the mode of dispersal of this genus, since -in some species the size of the drupes and of the included “stone” is so -great that, judged by those species only, it might be deemed impossible -to attribute the existence of the genus in isolated oceanic groups to -the agency of frugivorous birds. We are, however, compelled to appeal to -the bird, since, as my experiments in Fiji indicate, the genus has -little or no capacity for dispersal by currents, the “stone” when -containing a seed always sinking, whilst the entire fruit either sinks -at once or floats heavily for a few days. - -The degree of fleshiness of the drupes of Elæocarpus varies in different -species, being sometimes slight and at other times pronounced, but, -speaking generally, they would be expected to attract frugivorous birds. -The colour of the fruits of some species is dark and purplish, whilst in -others it is a bright blue. In the last case the fruits are very -conspicuous and sappy. A Solomon Island species collected by me and a -Malayan species observed by Ridley had bright blue fruits, and Cheeseman -refers to the Rarotongan species as possessing fruits of this hue. Their -colour, therefore, would often aid in attracting birds, and we are not -surprised to learn that they form a favourite food with fruit-pigeons, -parrots, and other frugivorous birds in different regions. Amongst the -fruits found by Professor Moseley in the crops of fruit-pigeons in the -Admiralty Islands were those of Elæocarpus; whilst in the Solomon -Islands I noticed that the blue fruits of the “Toa,” a species of the -genus, were a favourite food of the same birds (_Bot. Chall. Exped_., -iv. 307, 308; Guppy’s _Solomon Islands_, 293, 295). We learn also from -Hochstetter and from Sir W. Buller that the drupes of the “Hinau” -(Elæocarpus) form a favourite food of the parrots and fruit-pigeons of -New Zealand (Hochstetter’s _New Zealand_; Buller’s _Birds of New -Zealand_). - -The question of size acquires considerable importance when we come to -consider the transport of the seeds of the genus to a group of islands -lying, like Hawaii, in the middle of the Pacific Ocean. The protection -of the seed is also another important matter. There can, however, be no -doubt that the hard woody or often osseous “stone” sufficiently protects -the seed. With regard to size, if we were to judge from the dimensions -of the fruits of some of the Fijian species, where, as I found, the -“stone” measures from 3 to 5 centimetres (1-1/4 to 2 inches) in length, -we might be led to form a very erroneous opinion of the capacity of the -genus for conveyance through the agency of frugivorous birds to Hawaii. -But when we turn to the Hawaiian species we find the difficulty much -diminished, though still serious, the fruits being smaller and -possessing a “stone” 2-1/2 centimetres or about an inch long. In other -regions, however, the genus may possess fruits yet smaller in size. The -Tongan endemic species, as described by Burkill, has fruits 1·7 cm. or -7/10 of an inch in length; and closely similar dimensions are given by -Kirk for a New Zealand species. In both these cases the “stone” would -not be more than half an inch or 1·2 cm. in length, and this would also -apply to the Solomon Island species above mentioned. In another New -Zealand species, where the drupe is only half an inch, the “stone” would -be still smaller. It is thus evident that the fruits of different -species vary greatly in size in different regions, and that there is no -difficulty in assuming that a small-fruited species could be dispersed -over the Pacific by frugivorous birds, and carried either to Hawaii or -New Zealand. - -It might be an interesting point to determine to what extent a species -in an oceanic island could effect its own isolation by developing a -“stone” too large and too heavy to be transported across an ocean by -birds, such as seems to have happened with some Fijian species. But a -similar curious question is raised by the deterioration of a drupe in -its capacity for dispersal by frugivorous birds, when, as in the case of -the Hawaiian species of Elæocarpus, the drupes become dry and almost -sapless. As remarked in Note 68, this same feature is to be noticed in -the fruits of some of the Hawaiian endemic genera. This, of course, -would be quite in accord with what we should expect from the standpoint -of dispersal. - -I will conclude these remarks on Elæocarpus with a reference to the -similarity of its distribution with that of Freycinetia. Both genera are -at home in the temperate rain-forests of New Zealand and in the tropical -rain-forests of the Pacific islands and of Malaya. Their capacities for -dispersal are so different and so unequal, the dispersal of Freycinetia -being seemingly so much more readily effected, that we can only suppose -that time has long since discounted any special advantage one genus -possesses over the other as regards distribution. - - - DODONÆA (Sapindaceæ). - -This genus of small trees and shrubs includes between fifty and sixty -known species, of which about forty are confined to Australia; but a few -species are found over the tropical and subtropical regions of the -world, extending sometimes into temperate latitudes. There are, it -seems, only three species known from the oceanic groups of the tropical -Pacific: one, the cosmopolitan Dodonæa viscosa, that occurs in every -island of volcanic formation; and two others associated with it in the -Hawaiian Group, to which they are restricted. We have thus repeated in -this genus what is true of several other genera in Hawaii, such as -Metrosideros and Wikstrœmia, namely, the occurrence in that group of a -widely-ranging species accompanied by other species peculiar to those -islands. In the case of Dodonæa in Hawaii we should not expect to find -it very difficult to connect the endemic species with the widely-ranging -D. viscosa, which is a very variable species. The extreme forms in -different parts of the world are so different in character that Bentham -viewed this species as probably including the whole of the -extra-Australian species, excepting perhaps the Hawaiian endemic species -and one or two South African and Mexican plants (_Bot. Chall. Exped._, -iii. 136). - -Of the two Hawaiian peculiar species, one, Dodonæa eriocarpa, is a -mountain shrub found in most of the large islands and occurring -sometimes at elevations of 6,000 to 8,000 feet. The other species, D. -stenoptera, is, according to Hillebrand, a very distinct species found -only on Molokai. Bentham was only acquainted with the first-named, and -his hesitation to include it as one of the innumerable forms of the -widely-ranging D. viscosa is very suggestive. However, whether or not -one or both of these peculiar forms are connected in their origin with -this species, it is certain that the genus has been established for ages -in Hawaii; and from D. viscosa we can learn how a species of the genus -can cross an ocean, and also how from a widely-ranging species -exhibiting extreme variability species peculiar to a group of islands -could have been derived. - -The great variability of Dodonæa viscosa is associated with great -adaptability to different stations. Thus, as Mr. Hemsley tersely puts -it, it is one of those plants that thrive on the sea-coast as well as -inland, and in almost any soil or situation—provided, it may be added, -that the station is well exposed to the sun. Although Mr. Ridley -characterises it as a regular sea-shore plant in the Malay peninsula, -and although Prof. Schimper places it in the Indo-Malayan strand-flora, -it is as an inland plant that it is most characteristic of the Pacific -islands; and the key to its powers of adaptation to different stations -is to be found in its xerophilous habit. It is essentially a plant of -sunny places, and is equally at home on the parched inland plain, in the -open wood, on the sandy beach, on an old lava-field, or on rocky -declivities. It is not a plant of the rain-forest, preferring dryness to -humidity and sunshine to shade. - -The following remarks on the mode of dispersal of the wide-ranging -Dodonæa viscosa will serve to roughly indicate the capacity of the genus -for distribution. It is a subject, however, that requires further -detailed investigation. The light, inflated, winged capsules of this -species, about an inch across, could be blown for long distances along -the ground and carried for short distances in the air by strong winds, -but, as is also remarked by Prof. Schimper (_Ind. Mal. Strand-flora_, p. -157), they are much too large to be transported by winds across a broad -tract of sea. The currents, however, may have aided in the dispersal of -the species in the case of island-groups 500 or 600 miles apart. -Although the membranous capsules before dehiscing would be unable to -withstand the “rough-and-tumble” of ocean-transport for more than a few -days, the seeds possess some floating powers of a purely accidental -nature due to the imperfect filling up of the seed-cavity in some of the -seeds. In an experiment made in Hawaii I found that only half the seeds -floated in sea-water. Prof. Schimper, in an experiment conducted in -Germany with seeds that must have been well dried by keeping, found that -they floated for from ten to sixty days. This limited capacity for -flotation might possibly allow the species to reach Tahiti by easy -stages from Fiji; but it is not sufficient to explain its occurrence in -the more isolated Hawaiian Group. The fruits and seeds of this plant -never, however, came under my notice in the floating or stranded -seed-drift of Fiji; and I am not inclined, for this and the reasons -above mentioned, to consider that the currents have been very effective -agents in dispersing this plant over the Pacific islands. - -Hillebrand endeavoured to account for the wide distribution of Dodonæa -viscosa by “the glutinous capsules which would easily adhere to the -plumage of birds.” It may be here remarked that in the dried state -specimens of the plant have a varnished appearance as respecting the -leaves, branchlets, and capsules. In the living condition this is -represented by a glutinous or viscid condition of the surface of these -portions of the plant, rendering them adhesive to the touch. I found, -however, that only the immature capsules are markedly “sticky,” and that -in any case the adhesive power was quite insufficient to allow of -adherence for any length of time of fruits of this size to a bird’s -feathers. Mr. Ridley, who allows much latitude to birds in matters of -dispersal, remarks that the stickiness only appears when the specimen is -dry (_Trans. Linn. Soc. Bot._, 1888-94, p. 289). It is, nevertheless, -likely that the crustaceous seeds, which do not exceed 1/5 of an inch (5 -mm.) in size, when swallowed by a bird granivorous in its diet, might be -voided unharmed, and the dispersal of the species assured. It is in this -fashion, I imagine, that the plant reached distant groups like Tahiti -and Hawaii. - -There is, of course, the possibility that man has in past times aided in -the distribution of Dodonæa viscosa over the warmer regions of the -globe. But such an agency seems largely discounted in the case of an -isolated archipelago like Hawaii by the occurrence of endemic species. -Nor does the usual station in the Pacific islands support the view that -it was introduced by the aborigines. According to Hillebrand, it -possesses a variety (var. spathulata) in Hawaii which seems also to -occur in Tahiti and New Zealand. Nadeaud observes that in Tahiti it -grows as a bush on dry crests, and as a small tree, ten feet in height, -in the mountains. - -Nor do the aboriginal names of Dodonæa viscosa point in the direction of -man’s agency. It possesses a different name in every group, and is -evidently not a plant with which the ancestors of the Polynesians were -familiar in the home of the race. Thus it is named “aalii” in Hawaii, -“apiri” in Tahiti, “ake” in Rarotonga, “lala vao” in Samoa, and I may -add “usi” or, as Seemann writes it, “wase” in Fiji. - -Looking at these various facts, I am not inclined to exclude altogether -any one of the three agencies above discussed; but I should imagine -that, placed in their order of effectiveness, we should have first -birds, then the currents, and lastly man. - - - METROSIDEROS (Myrtaceæ) - -Whilst this genus of trees and shrubs has its home in New Zealand and -Australia, there is an extremely variable Polynesian species, -Metrosideros polymorpha, ranging over all the volcanic groups of the -tropical Pacific, from Fiji to Pitcairn Island and from Hawaii to the -Kermadec group, but seemingly only in the Hawaiian group associated with -endemic species. According to the _Index Kewensis_ the genus comprises -about forty known species, of which two-thirds are confined to New -Zealand and Australia in equal proportions; whilst, among the rest, six -species belong to New Caledonia, two to Hawaii, and three to Malaya, and -there are solitary species in Chile, Madagascar, and South Africa. - -I will attack the problem connected with the distribution of the genus -through the widely-ranging Polynesian species, Metrosideros polymorpha. -“This genus,” wrote Dr. Seemann, “is in a fair way of becoming in -Polynesia what Rubus is in Europe. It is very much given to variation, -and it is very difficult to find out the limits of the different -species.” In making these remarks he had this species in view, and his -adoption of Gaudichaud’s specific name of “polymorpha” to cover almost -all the Polynesian forms has been generally followed. Although so widely -distributed over the Pacific, it is in the Hawaiian Islands that this -tree attains its greatest development, growing gregariously and often -forming almost exclusively entire forests; and it is here that it -displays the greatest variation. But it was remarked by Seemann, and -this was confirmed by Hillebrand, that almost all the Hawaiian forms -occur in the Society or Tahitian Islands. - -In connection with the great variability of Metrosideros polymorpha must -be considered its variety of stations and its great range in altitude. -Hillebrand describes seven Hawaiian forms of this species, and their -various stations and characters are well illustrated in his -descriptions. Thus, whilst the trees may attain a height of forty feet -in the forests, in elevated exposed situations they may be small and -gnarled or low and shrubby; whilst in the bogs and swamps of the high -levels of Maui and Kauai the plant grows as a prostrate shrub. It is not -at all unlikely that the two peculiar Hawaiian species of the genus had -a common origin from a widely-ranging species, which, if not the present -M. polymorpha, was its immediate ancestor. One of them was, indeed, -included by Dr. Seemann within the wide limits of this species, and the -other was accepted with a doubt. - -To illustrate the great vertical range in the Hawaiian Group of -Metrosideros polymorpha, I will take it as I found it in the island of -Hawaii. Here it ranges from the coast up to about 8,000 feet above the -sea. But it is in the middle forest-zone at elevations of 2,000 to 4,000 -feet, where it is often associated with the Koa and Olapa Trees (Acacia -koa and Cheirodendron Gaudichaudii), that it is most at home and attains -its greatest size. Higher up at heights of 5,000 to 7,000 feet in the -more open forests it is still in the company of the trees just named -together with Sophora chrysophylla and Myoporum sandwicense. At 8,000 -feet it becomes very stunted and is accompanied usually by bushes of -Cyathodes and other plants of similar bushy growth. In the lower parts -of its range, from 2,000 down to 1,000 feet, it forms forests with the -Kukui Tree (Aleurites moluccana), mingled also with smaller trees such -as the Hawaiian Olive (Osmanthus), and the Kopiko (Straussia). Below -1,000 feet, and wherever bold promontories reach the coast and the -inland forest descends to the sea, we find it associated with such trees -and shrubs as the Lama (Maba sandwicensis) and different Akeas -(Wikstrœmia). On the partially vegetated surfaces of old lava-flows near -the coast it grows beside bushes of the Ulei (Osteomeles -anthyllidifolia) and of Cyathodes. - -Compared with its behaviour in Hawaii, Metrosideros polymorpha takes a -relatively unimportant part in the vegetation of Fiji. As Horne -observes, the trees are most common in the dry parts of the two largest -islands and grow in the poorest soil. I found them in Vanua Levu usually -in open exposed situations, generally in the dry “talasinga” plains on -the north side of the island, where they were associated with Acacia -Richii, Dodonæa viscosa, and Casuarinas; and sometimes they occurred in -a shrubby form on the rocky peaks of the highest mountains. In Rarotonga -also, as we learn from Cheeseman, it is on the tops of the rocky peaks -and along the crests of the ridges that this species, which is abundant -in the island, is frequently found. - -I may here allude to the curious fact observed by me on the upper open -wooded slopes of Mauna Kea at elevations of 6,000 to 7,000 feet, and -therefore on the outskirts of the true forest-zone. Here the Ohia Tree, -as the Hawaiians name Metrosideros polymorpha, often grows in close -association with the Olapa Tree (Cheirodendron Gaudichaudii). In one -locality, for instance, a large Olapa was growing in the fork of an Ohia -at about eight feet from the ground, and sending down roots on either -side. Sometimes the trunks of the Olapa and the Ohia were to be seen -growing in such close contact as to look like one tree. In one such case -a young tree, four feet high, of Myoporum sandwicense was growing in a -fork of the Ohia, whilst in a fork of the Olapa a plant of Vaccinium -penduliflorum, three or four feet in height, had established itself. -This remarkable instance of epiphytic growth also proved to be quite a -revelation with regard to the dispersal of seeds in this island. Amongst -these four associated plants, which include three trees and one shrub, -all except the Ohia, which was probably the original tree, have fruits -that would attract frugivorous birds; and in succession these birds had -first dropped a pyrene of the Olapa in the fork of the Ohia, and -afterwards the seeds of Myoporum again on the Ohia, whilst finally the -Vaccinium seeds were dropped into the fork of the Olapa after it had -developed into a tree. - -The mode of dispersal of the seeds of Metrosideros polymorpha now -invites our attention. Since the fruits are dry, dehiscent capsules -possessing minute fusiform seeds, we are not able to appeal directly to -the agency of frugivorous birds to explain the wide dispersal of this -species. The seeds are light in weight and remind one a little of those -of the succulent fruits of Freycinetia. For purposes of dispersal, -however, they must be placed in the same category with other plants with -dry, dehiscent fruits and small seeds, such as the Vota (Geissois -ternata) of Fiji, a tree that in those islands grows in similar -stations. On a later page I have suggested that the seeds of the Vota -are dispersed by large bats that visit the trees for the sake of the -honey in the red flowers. With Metrosideros polymorpha birds act -probably in the same way. We are, in fact, informed by Mr. Perkins that -the nectar-feeding birds of the Hawaiian Drepanids now obtain their main -supply of this food from the blossoms of this tree. If bats or birds -visit the large red flowers of Metrosideros polymorpha for the same -purpose, it is not difficult to imagine that they might carry away in -their fur or in their plumage some of the small seeds shaken out of old -dehiscent capsules. In this connection we may note that the Kaka Parrot -(Nestor meridionalis) of New Zealand is said to feed largely on the -scarlet blossoms and nectar of Metrosideros robusta (Evans’ _Birds_, p. -374). - -The seeds of Metrosideros polymorpha might no doubt be carried by winds -from one mountain-top to another and across narrow straits, but only -whilst adherent to a bat or a bird could they be carried across a wide -tract of ocean. Speaking of the genera Metrosideros and Lobelia in -connection with their occurrence in the Kermadec Islands, Sir J. Hooker -long ago referred to their minute seeds as not adapted for transport -across oceans unless their minuteness and number fitted them for it -(_Journ. Linn. Soc._, i. 127). The point that is raised here for these -genera in the Kermadec Group can be raised for the same two genera in -Hawaii and for a multitude of other small-seeded genera in those -islands. - - - ALYXIA (Apocynaceæ). - -This genus of climbing or straggling shrubs tells its own story of the -widely dispersed Indo-Malayan genera in the Pacific islands. Containing -about forty known species, it is distributed over the tropical regions -from Madagascar and the Mascarene Islands eastward to the Paumotu Group -and Pitcairn Island in mid-Pacific, and has its focus in the area -comprised by Malaya, Australia, and New Caledonia. In the _Index -Kewensis_ about eight species are assigned to New Caledonia, seven to -Australia, and seven to Malaya. One species, Alyxia stellata, ranges -over nearly the whole of the area of the genus from tropical Asia, -through Malaya, across the South Pacific to Tahiti. It will be for the -future investigator to determine how far the present distribution of the -genus can be connected with one or two widely-ranging polymorphous -species. The data at my disposal seem to show that in the open Pacific, -at all events, the history of the genus has gone a step beyond this -stage. - -Of the seven or eight species recorded from the Pacific islands east of -New Caledonia, only two or three seem to be now recognised as restricted -to particular groups, namely, one in Hawaii (Schumann), one in Fiji, and -one in Rarotonga. The other species indirectly connect together all the -groups, although no single species occurs over the whole region. Thus -the Hawaiian species, Alyxia olivæformis (Gaud.) has in recent years -been found in Upolu, in the Samoan Group, by Dr. Reinecke, an -exceedingly interesting though unusual specific link between these two -archipelagoes. Two species, A. stellata and A. scandens, range over the -South Pacific from Fiji to Tahiti, the last-named also occurring in the -Paumotu or Low Archipelago; whilst Rarotonga possesses a form closely -allied to the first-named, and to it Cheeseman has given specific rank. -Another species, A. bracteolosa, links together the contiguous Fijian, -Tongan, and Samoan groups. This distribution is what we should have -expected if one or two polymorphous species had originally ranged over -the Pacific and were advancing towards that stage of differentiation -when each group possesses its own peculiar species. (It may be here -remarked that an undetermined species of Alyxia is accredited by Maiden -to Pitcairn Island, which indicates that the genus has extended east in -the Pacific almost as far as the extreme limit of the Polynesian -region.—_Australas. Assoc. Reports_, Melb., 1901, viii.) - -All visitors to these islands that are interested in their floras will -be familiar with the Alyxias; and there are few of their plants that the -natives take more pleasure in pointing out to white men. They are -readily recognised on account of their black moniliform drupes and their -milky sap. All over Polynesia, whether in Hawaii, Tahiti, Samoa, or -Fiji, the aborigines value the plants on account of the delicate -fragrance of their foliage and bark. These materials they use for -personal decoration and in making wreaths, stripping off the bark of the -young branches with their teeth in the same fashion in Fiji and Hawaii -and probably in all the Pacific islands. Throughout Polynesia, excluding -Fiji, they bear the same name, which takes the form of “maile” in Hawaii -and Samoa, and of “maire” in Tahiti and Rarotonga—a name which the -Maoris, remembering the Alyxias of their tropical home in the South -Pacific, have applied to New Zealand species of Olea and Eugenia. The -Fijian generic name for Alyxia is “vono.” - -A word may be said about the station of these plants in the Pacific -islands. In Hawaii they occur in the middle and lower forests, and -usually between 2,000 and 4,000 feet in elevation. In Tahiti they -frequent the crests and precipitous rocky slopes of the mountains at -elevations of from 3,000 to over 6,000 feet. The Rarotongan species -often forms extensive thickets in rocky localities on the hills. In -Samoa they are found usually in the mountain forests. In Fiji they grow -on the outskirts of the virgin forests and on rocky sparingly vegetated -mountain peaks. I found them often in Vanua Levu growing amongst the -open vegetation on the summits of isolated mountains at elevations of -2,000 to 2,500 feet, where they were associated with other plants like -Elæocarpus, Pleiosmilax, and Scævola, possessing similar fleshy fruits -likely to be dispersed by frugivorous birds. - -The Alyxias indeed seem well suited for dispersal by birds. The black -fleshy drupes would readily attract them; and the solitary seed -protected by a very tough horny albumen might be ejected unharmed in -their droppings. - - -------------- - -It would be possible to enter into similar detail with several other -genera of this period; but here I can only direct attention to their -principal indications, permitting myself a little more license when -discussing the means of dispersal. - -ALPHITONIA (Rhamnaceæ).—Amongst other genera with polymorphous species -closely following the lines taken by Metrosideros in the Pacific is -Alphitonia, a small Malayan and Polynesian genus of tall trees, -containing at most three or four species, one of which (A. excelsa) has -almost the range of the genus and is found in most of the Pacific -archipelagoes. So variable is this widely-ranging tree that Bentham -suggested that there was only one species in the genus (_Bot. Chall. -Exped._, iii. 133), a suggestion especially interesting in connection -with the _rôle_ taken by polymorphous species in the Pacific. As bearing -on the mode of dispersal of this species, it may be observed that my -Fijian experiments show that the fruits are not fit for transport by -currents. With the mature drupe the outer coverings become pulverulent, -and the fruit breaks down, freeing the pyrenes which do not float; nor -have the seeds any buoyancy. Although the dry drupes would seem -unattractive to birds, it is to birds we must look for the dispersal of -the genus. - -PISONIA (Nyctagineæ).—Like Dodonæa, Metrosideros, and Alphitonia, the -cosmopolitan genus Pisonia possesses a polymorphous species that -displays its variation in every Pacific group and occupies a -considerable number of stations. The earlier botanists in the Pacific -differed much as to the species of this region, and this led Mr. Hemsley -to observe in his paper on the Tongan flora that it is difficult to -understand the various Polynesian and Australian species except on the -assumption that there is one very variable species. Recognising this -difficulty, Drake del Castillo deals somewhat summarily with nearly all -these forms, uniting them under one comprehensive species, P. -umbellifera (Seem.), thus constituting “une espèce très-polymorphe” that -ranges (generally in maritime districts) over tropical Asia and the -islands of the Indian and Pacific Oceans, extending to North-East -Australia and to New Zealand. On account of the unusual capacity for -dispersal possessed by this species—a subject to be immediately -discussed—the tendency to specific differentiation has been kept in -check, though the process has gone farther in some groups than in -others, as in the case of Hawaii, where Hillebrand’s endemic species -has, however, been included by Drake del Castillo in his polymorphous -species, P. umbellifera. - -The fruits of this genus possess no capacity for dispersal by currents. -They never came under my notice either in floating or stranded -seed-drift, and have little or no buoyancy. Prof. Schimper, -experimenting on the well-dried fruits of Pisonia aculeata, a seaside -shrub common in America and in the Old World, and destined probably to -be brought by the systematist into touch with the polymorphous P. -umbellifera, found that they sank in a day or two (_Ind. Mal. -Strand-flora_, p. 156). Dismissing the agency of the current, he looked -to that of the bird for the explanation of the dispersal. The -probability of the effectiveness of this last-named agency has long been -surmised. It attracted the notice of Darwin and especially invited the -attention of another student of plant-dispersal, Dr. H. O. Forbes. The -long, narrow, often fusiform fruits are invested by a somewhat -coriaceous perigone and range from less than an inch to three inches in -length (2-7·5 cm.). They excrete a very viscid fluid often in quantity, -and sometimes also possess glandular spines. The Hawaiians, according to -Hillebrand, used this material as bird-lime for catching birds, and the -fruits, he says, will stick fast to the paper in the herbarium for -years. In that group I often found the fruit adhering firmly to my -clothes. Writing of these trees on Keeling Atoll, Forbes observes that -their sticky fruits are often such a pest to birds roosting in their -branches that they have proved fatal to herons and boobies by collecting -in their plumage. “It is easy to perceive,” he remarks, “how widely this -tree might be disseminated by the birds that roost on it” (_The Eastern -Archipelago_, p. 30). In New Zealand, as we learn from Kirk, the viscid -fruits of Pisonia brunoniana attract small birds which become firmly -caught and die miserably. A cat has been known to wait under a tree -watching its opportunity of preying on the entangled birds. Sir W. -Buller states that the New Zealand fruit-pigeon feeds at times on the -green fruits of P. umbellifera; and we can infer that it occasionally -carries off some of the riper fruits in its feathers. - -WIKSTRŒMIA (Thymelæaceæ).—This is a small genus of shrubs and small -trees, with red or yellowish drupes fitted for dispersal by frugivorous -birds, that is confined mainly to tropical Asia, Australia, and -Polynesia. Following Seemann and Drake del Castillo, we may say, that -like several other genera of this period, this genus possesses in the -tropical Pacific a widely-ranging species, W. indica, that occurs in -Hawaii, the Marquesas, Tahiti, Samoa, and Fiji, growing amongst the -vegetation immediately behind the beaches and in the plains and open -wooded districts inland. In Hawaii it is associated with half a dozen -peculiar species, and in Tonga there is also an endemic species. The -widely-ranging species has its home in the Indian Archipelago and in the -Asiatic mainland, and occurs also in Australia. According to Gray, the -American botanist, it is represented by a different variety in almost -every group in the tropical Pacific, and it presents us therefore with -another example of a polymorphous species which links Polynesia directly -with Malaya. As bearing on the dispersal of the genus by birds, it may -be added that Mr. Perkins in the _Fauna Hawaiiensis_ speaks of some of -the Drepanids and of a species of Phaeornis as feeding at times on the -fruits of these plants. - -PEPEROMIA (Piperaceæ).—All observers of tropical plant-life will be -familiar with this genus of low herbs growing on tree-trunks, on the -soil, on rocks, and on stonewalls, and comprising about 500 known -species distributed over the warmer regions of the globe and sometimes -extending into cooler latitudes. In Polynesia it attains its greatest -development in Hawaii, where Hillebrand enumerates about twenty species, -of which, after excluding doubtful forms, at least a third must be -endemic. Tahiti, Samoa, and Fiji are each known to possess three or four -species, of which one is usually restricted to the group. Two species, -P. reflexa and P. leptostachya, link together nearly all the groups of -the tropical Pacific, including Hawaii, the first cosmopolitan, and the -second hailing from North-East Australia and indicating that the genus -has entered Polynesia from the west.... These plants possess spikes of -small berries containing a single seed, and are evidently, like other -Piperaceæ, dispersed by frugivorous birds. It is to be noted that the -presence of a West Indian and Mexican species in the Bermudian caves is -attributed by Mr. Hemsley to frugivorous birds (_Bot. Chall. Exped._, -Introd. 49, i. 62). In Vanua Levu they occur on the bare rocky peaks of -some of the mountains under such conditions that the seeds could only -have been brought by birds. Thus, on the bare surface of a large block -of tuff forming the highest peak of Koro-Mbasanga, 2,500 feet above the -sea, I found only two plants, Oxalis corniculata and a species of -Peperomia. - -EUGENIA (Myrtaceæ).—This is a very extensive genus split up into -different subgenera, and comprising some 600 or 700 known species -scattered over the warm regions of the globe. Their fleshy, usually red, -berries contain as a rule one or two large seeds, and attract birds and -animals of all descriptions. The feature most interesting to us is the -dispersal of the genus over the Pacific islands eastward to the Low -Archipelago and northward to Hawaii. The track by which it has entered -the Pacific from the west is indicated in the distribution of the -species. The genus is only well represented in the Western Pacific, -whilst eastward and northward of Samoa and Tonga the distribution is -fitful and irregular, it being evident that the extension beyond these -two groups has been accomplished with difficulty. - -There are at least twenty-five species in Fiji, of which perhaps half -would be peculiar; in Tonga eight species, of which two may be endemic; -in Samoa thirteen species, of which four are peculiar; in Rarotonga -none; in Tahiti a single non-endemic species; and in Hawaii two species, -of which one is peculiar. Only truly indigenous species are here -recorded, and Eugenia malaccensis, which has accompanied the aborigines -in their migrations, is not included. A solitary species, E. rariflora, -connects together all the principal archipelagoes from Fiji to Tahiti -and the Gambier Islands, and northward to Hawaii. Nine species are known -to be common to the region in which lie the three groups of Fiji, Tonga, -and Samoa; and since some of these species occur in the groups further -west they may be regarded as keeping up the connection with the original -home of their ancestors in the Malayan region. - -Looking at these facts of distribution of the genus Eugenia in the open -Pacific, it is evident that whatever dispersal of the genus is now in -progress in this ocean is mainly confined to an interchange between the -groups of Fiji, Tonga, and Samoa in the Western Pacific, and doubtless -between the islands further west of these groups. The smaller islands -lying between and around these three groups participate in the -distribution of the species common to all. Thus Wallis Island, according -to Drake del Castillo, possesses two of these species. Over the rest of -the ocean the dispersal of the genus seems to be no longer effective, -since Eugenia rariflora, which links together Fiji, Tahiti, and Hawaii, -shows signs of differentiation in nearly every group. In Hawaii, where -it is very rare and is only recorded from two of the islands, it has -developed a small-leaved variety. In Tahiti it displays the same -variation; and Seemann observes that there are differences between the -Tahitian and Fijian species which may be almost specific in value. It -would also appear that both in Hawaii and Tahiti the fruits have become -less attractive to birds, being described as “dryish” and “dry,” which -is, as Dr. Seemann remarks, certainly not true of the Fijian plant. - -In Fiji the Eugenias, as small trees and shrubs, find their home usually -on the banks of streams and rivers, on the outskirts of forests, and -occasionally at the coast. One of them, E. richii (Gray), is a -characteristic littoral tree in the group. A tree near it in character -was found by me of common occurrence in the interior of coral islets in -the Solomon Group (_Solomon Islands_, p. 297). E. rariflora occurs also -in the interior of coral islets in Fiji and amongst the vegetation at -the back of the mangrove-swamps. - -Coming to the mode of dispersal of the genus in the Pacific, I may -remark that all the species, with the doubtful exception of the Fijian -and Samoan Eugenia neurocalyx (the Lemba of Fiji), are wild trees and -shrubs useless to man, but much appreciated by pigeons, pigs, &c., on -account of their fleshy fruits. Since exact observations on the -possibility of their dispersal by currents seemed to be wanting, I made -some experiments in Fiji. Out of six species, which included E. -corynocarpa, rariflora, richii, and rivularis, the mature fruits of most -species sank in sea-water in from seven to ten days. However, those of -the beach tree, E. richii, floated for a fortnight. The cause of sinking -in all cases lay in the decay of the outer fleshy covering. As I have -observed in river and sea drift, fish bite at the floating fruits, and -in this manner the seeds would soon be liberated and sink. The seeds of -all the plants sank at once in my experiments except with one species, -where the seed loosely filled its test and thus a floating-power of a -few days was acquired. Currents, it is apparent, could never account for -the dispersal of the genus over a broad extent of ocean, though in a few -cases, as in that of the littoral tree above noted, it is quite possible -that the fruits could be successfully transported across a tract of sea -200 or 300 miles in width. - -It has long been known that fruit-pigeons are fond of the fruits of wild -species of Eugenia, and I found the Solomon Islanders and the Fijians -well acquainted with the fact. The fruits of a tall Eugenia tree, near -E. richii, common in the interior of the coral islets of Bougainville -Straits in the Solomon Group, were found by me in quantities in the -crops of fruit-pigeons shot by Lieut. Heming and Lieut. Leeper on the -islets (_Solomon Islands_, pp. 293, 297; _Bot. Chall. Exped._, Introd. -46, iv. 312). Dr. Seemann remarks that in Fiji the red fruits of E. -brackenridgei are eaten by pigeons. The somewhat thin coverings of the -seeds of this genus would seem to offer but a slight protection in a -bird’s stomach, though in one species the test was almost crustaceous. - -Most species possessed only one or two large seeds in each fruit, though -this number may vary in the same individual. Thus, out of ten fruits of -Eugenia rariflora in Fiji, six had one seed, three had two seeds, and -one had three seeds. In the fruit of E. neurocalyx, however, the seeds -range from three to five. - -It is the question of size that is of importance in considering the -possibility of birds transporting the seeds over a broad tract of ocean. -Eugenia rariflora, the species found all over the Pacific, has seeds -that measure in the Fijian plant one-fourth to one-third of an inch (6 -to 8 mm.) across; and in Hawaii, according to Hillebrand, they would -perhaps be rather smaller. In point of size there is less difficulty -with regard to the transport by birds across the ocean to Hawaii of the -seeds of Eugenia rariflora than with the “stones” and seeds of some -other genera, like Elæocarpus, Osmanthus, and Sideroxylon, that must -have been conveyed there by the same agency. The fruits of several of -the Fijian species are of the size of a large cherry; but it is -noteworthy that in those species like E. corynocarpa and E. neurocalyx, -where the fruits are large and the seeds about an inch in size, the -plants are confined to the Western Pacific only, namely, to the -Fiji-Samoa region. - -There is therefore no difficulty, from the standpoint of size, in -accounting for the distribution by birds of the widely-ranging Eugenia -rariflora over Polynesia; but at first sight there seems to be a real -difficulty with regard to the protective coverings of the seed. Yet -Nature speaks with no hesitating voice in the matter. The West Indian -and Florida species, E. monticola, regarded as indigenous in the -Bermudas, must have reached that group through the agency of birds that -carried its seeds over quite 800 or 900 miles of sea; and it may here be -noted that South Trinidad, lying some 600 miles off the coast of Brazil, -and Rodriguez, distant about 330 miles from Mauritius, each possess -species (_Bot. Chall. Exped._, Introd., 12, i. 32, ii. 128). If -fruit-pigeons can transport Eugenia seeds across 600 or 800 miles of -ocean, there would be no difficulty in accounting for the stocking of -the Fijian, Tongan, and Samoan Islands with the genus from regions to -the west. But the occurrence of the genus in Hawaii seems to compel us -to assume that the seeds have been carried in a bird’s stomach over -1,500 to 2,000 miles of ocean. This difficulty, however, does not really -exist. Eugenia rariflora, the Polynesian species found in Hawaii, -frequents, as before observed, coast districts and coral islets in Fiji, -and if we suppose that the low islands of the Fanning and Phœnix Groups, -lying between Hawaii and Samoa, have served as stepping-stones, a -capacity of crossing 1,000 miles of ocean would be alone required. This -is not much in excess of the distance that must have been traversed by -the bird that first brought the seeds of Eugenia monticola to the -Bermudas. - -Other genera like Morinda and Scævola, possessing fleshy fruits -dispersed by frugivorous birds, have been mentioned in different -connections in other parts of this work, and will not be further dealt -with here. But before concluding this chapter I will refer briefly to -one of the disquieting mysteries in the flora of the Pacific which is -presented to us in the genus Gossypium. Three species are, or were, -truly indigenous in this region. One is Gossypium drynarioides, a small -endemic tree found by Nelson, the companion of Captain Cook, in Hawaii, -which was very rare in Hillebrand’s time, and is perhaps now extinct. -The second is G. tomentosum (Nuttall), which is also peculiar to Hawaii, -where it is found on the beaches. I am following here the _Index -Kewensis_; but it should be remarked that this species occurs also in -Fiji, though Seemann regards it as introduced. The third is G. -religiosum (L.), found by Captain Cook’s botanists growing wild in -Tahiti, and hailing from the tropics of the Old World. The seeds of the -first species are covered with a short brownish tomentum, and could -never have been of any value. The tawny wool of the seeds of the second -species has a staple too short for cultivation; whilst the Tahitians do -not seem to have made any use of the third species. It is difficult to -draw any conclusion concerning the presence of these plants in the -Pacific islands at the time of their discovery; nor can Dr. Seemann, who -was especially well informed in these matters, aid us much in our -endeavours to solve the mystery. From the aboriginal names we get no -clue. The Hawaiian name of “huluhulu” seemingly refers to the hairy -covering of the seed; whilst the Tahitian “vavai” and “ovari” simulate -the Fijian “vauvau,” which is merely the reduplicated form of “vau” (the -word in many shapes for Hibiscus tiliaceus in Malaya and Polynesia), and -is applied by the Fijians to Hibiscus esculentus and to the introduced -species of Gossypium. - -When in Hawaii I ascertained that neither the seeds of the littoral -plant, Gossypium tomentosum, nor those of two cultivated species -possessed any fitness for dispersal by the currents, the scraped seeds -sinking at once, whilst when covered with the wool they floated only for -a few days. Further references to G. tomentosum in Hawaii are given in -the index of this volume. - - -_The Last Stage of the General Dispersal of Plants of the Malayan Era._ - -We arrive now at the close of the era of the general dispersal of -tropical plants, mainly Malayan, over the Pacific, and this brings us -down to our own age. The few genera that are still dispersed have no -peculiar species in particular groups. The species which often range -over all the groups, and retain as a rule their characters in most of -them, do not therefore display, except in a few cases, that extreme -variation which would give them a place in the ranks of the polymorphous -species. The dispersing agencies, in fact, are sufficiently active to -check marked variations, and the process of isolation has scarcely -begun. - -We perceive the reason of this when we look at the nine genera which are -taken as samples of this period, viz., Rhus, Osteomeles, Viscum, -Plectronia, Boerhaavia, Polygonum, Pipturus, Boehmeria, and Dianella, -most of them being known to be dispersed by birds at the present day. -Six of the genera possess fruits likely to attract frugivorous birds; -whilst one of them, Boerhaavia, has sticky fruits that would be apt to -adhere to plumage. Actual observations in the cases of Rhus, Viscum, and -Plectronia establish the fact of their dispersal by fruit-eating birds; -and there is no difficulty in postulating the same agency for -Osteomeles, Pipturus, and Dianella. A method by which Boerhaavia fruits -would be transported in the plumage of birds has been observed by Mr. -Lister; whilst the nutlets of Polygonum are known to afford food to a -variety of birds and to be thus distributed. - -In this period the plants all hail from the Asiatic side of the Pacific. -Three of the genera, Plectronia, Pipturus, and Dianella, belong almost -exclusively to the Old World. Five occur in both the Old and New Worlds, -but, as with Rhus, Viscum, Boerhaavia, and Boehmeria, are represented by -Old World species in the Pacific, or, as with Polygonum, possess a -cosmopolitan species (P. glabrum) ranging over the warm regions of the -globe. Even Osteomeles presents no exception to the rule, since the -Pacific plant is the only one of its species that is not American. - -We have in Polygonum glabrum the only aquatic or semi-aquatic plant -widely distributed over the Pacific islands that can lay claim in all -groups to be indigenous. It is associated in Hawaii with species of -Potamogeton and Naias, aquatic genera that have, however, a limited -distribution in Polynesia. - -I will now make a few remarks on each genus such as bear on their -distribution and on their mode of dispersal in the Pacific. - -RHUS (Anacardiaceæ).—The representation of this genus by indigenous -species in oceanic islands not only in the Pacific but also in the -Atlantic, as in the Bermudas, is of especial interest in connection with -dispersal by frugivorous birds, since the drupes are typically dryish -and might appear to be not very attractive to birds. There are two Old -World species known from the Pacific islands: one being R. simarubæfolia -(Gray), distributed over the South Pacific groups from Fiji to Tahiti -and hailing from Malaya; the other, R. semialata (Murray), alone -recorded from the Hawaiian Group and derived probably from China or -Japan. This indication that the groups of the North and South Pacific -have derived their species, the first from Temperate Asia and the second -from Tropical Asia, is of some interest. In Samoa, according to -Reinecke, the fruits of R. simarubæfolia, which are of the size of a -pea, form the favourite food of the fruit-pigeons. That birds disperse -the seeds of the various Sumachs is familiarly known. In the United -States, as we learn from Barrows, Beal, and Weed, crows, woodpeckers, -and other birds feed extensively in winter on the fruits of different -species of Rhus, including the Poison Ivy (R. toxicodendron). The crows -discharge the seeds in pellets after retaining them for about thirty -minutes. Some seeds we must infer would pass into the intestines, where -they might be retained for ten to twelve hours (see Chapter XXXIII.), -which would be long enough, according to Gätke’s views of bird-velocity, -to enable them to be transported over a thousand miles of ocean. - -OSTEOMELES (Rosaceæ).—One of the most interesting cases of dispersal in -recent times over the Pacific islands is that of O. anthyllidifolia. Of -the ten known species of the genus, nine are confined to South America; -whilst the Pacific species, which is not recorded from America, has been -found in Upper Burma, Japan, the Liukiu and Bonin Groups, Hawaii, -Pitcairn Island, Mangaia, and Rarotonga. The remarkable distribution of -the Pacific plant at once attracts attention. I was very familiar with -it in Hawaii, where it forms one of the commonest bushes in open-wooded -and thinly vegetated districts at elevations usually ranging from the -coast to 3,000 feet. Its small, white, somewhat fleshy fruits would -attract birds, and the hard pyrenes would be able to pass unharmed -through a bird’s digestive canal. It seems probable that, like Rhus -semialata, this plant entered the Pacific Ocean from the north-west, -taking the route by Japan and the Bonin Islands, and following the trend -of the archipelagoes over Polynesia (see _Bot. Chall. Exped._, Introd. -p. 18; _Journ. Linn. Soc. Bot._, vol. 28, 1891, &c.). - -VISCUM (Loranthaceæ).—A single species, V. articulatum, which has its -home in Southern Asia, is found in most of the Pacific groups, such as -Hawaii, Marquesas, Tahiti, Rarotonga, Fiji, &c. The dispersal of the -genus by frugivorous birds is well known. - -PLECTRONIA (Rubiaceæ).—I have found it more convenient to place this -genus here, although there are probably one or two species peculiar to -Fiji. This genus of shrubs, which is spread over the warm regions of the -Old World, is represented by two widely distributed species in -Polynesia, Plectronia odorata (B. and H.) and P. barbata (B. and H.), -the first alone extending to Hawaii. I was very familiar with P. odorata -in Hawaii and was much interested in its mode of dispersal, since the -species has also been found in Fiji, Tahiti, the Marquesas, and Pitcairn -Island (Maiden). In one locality, where an old lava-field was partially -covered by its bushes then in fruit, the doves were feeding greedily on -the drupes, the “stones” of which, as well as the partially digested -fruits, were to be seen in quantity in their excrement near a -water-hole. The stones are very hard and about a third of an inch (8 -mm.) in length, and are exceedingly well suited for transport by -frugivorous birds. It was very probably to one of these species of -Plectronia that Peale alluded when he wrote of the berries of a species -of Canthium forming the principal food, on one of the Paumotu Islands, -of Numenius tahitensis, a curlew that has its home in Alaska, migrating -south in autumn to Hawaii, Tahiti, and the Paumotu Group (Wilson’s AVES -HAWAIIENSES). - -BOERHAAVIA (Nyctagineæ).—Two or three Asiatic species of this genus, B. -diffusa, B. tetranda, &c., are spread all over the Pacific islands from -the Fijis to the Paumotus and northward to Hawaii. Similar or allied -species occur on the coral islands of the Indian Ocean, as on Diego -Garcia and on Keeling Atoll. Though these plants have often been -accidentally spread by man with his cultivated plants, it is probable -that sea-birds have regularly aided in their dispersal. The fruits, on -account of their small size and their glutinous sticky surfaces, are -well suited for transport in a bird’s feathers. Mr. Lister, as quoted by -Hedley (from _Proc. Zoolog. Soc._, 1891), made an interesting note in -this connection on one of the islands of the Phœnix Group, where he -found a fruit of Boerhaavia tetrandra entangled in some of the down that -had been preened by a booby (Sula piscatrix) out of its feathers whilst -roosting in a clump of Tournefortia trees. - -POLYGONUM (Polygonaceæ).—This genus is represented by the cosmopolitan -Polygonum glabrum, the only aquatic or semi-aquatic plant that is -generally distributed in the Pacific islands. It occurs in fresh-water -swamps and beside streams and ponds in Tahiti, Tonga, Fiji, Hawaii, &c., -and was gathered by Banks and Solander when Captain Cook first visited -Tahiti. That this plant has been distributed by geese, ducks, and -waterfowl over the tropics of the globe can scarcely be doubted. In -England I have found the nutlets of Polygonum convolvulus, P. -persicaria, and P. aviculare in the stomachs of a wild duck and a -curlew; and they came frequently under my notice in the crops and -intestines of different kinds of partridges and of wood-pigeons. Though -most of the fruits were generally injured, a few of them were not -uncommonly obtained in a sound condition. - -PIPTURUS (Urticaceæ).—This is a genus of small trees and shrubs found in -the Mascarene Islands, Malaya, Australia, New Zealand, and throughout -Polynesia. Besides P. albidus, which is confined to Hawaii and Tahiti, -there are two Malayan species, P. argenteus and P. velutinus, which are -widely distributed over the islands of the South Pacific, extending to -Tahiti and the Marquesas. The fleshy receptacle and small achenes of the -compound fruit of Pipturus give it the appearance of a white immature -strawberry, and as such it would be likely to attract frugivorous birds. -Plants of this genus are included amongst the numerous plants from the -bast of which the natives used to prepare their native cloth or from -which they obtained the fibres for their fishing-lines. - -BŒHMERIA (Urticaceæ).—There is an Asiatic species widely spread in the -South Pacific and another closely-allied species in Hawaii; but I -possess no data relating to the dispersal of the genus. The fruits are -dry and consist of an achene in a persistent perianth. - -DIANELLA (Liliaceæ).—This is a genus of herbs, possessing often pretty -blue berries, that extends over tropical Africa, tropical Asia, the -Mascarene Islands, Malaya, Australia, and New Zealand, and is found in -all the larger Pacific archipelagoes. Of the twelve species named in the -_Index Kewensis_ only two belong to America, occurring respectively in -Cuba and Venezuela. There are two species in the islands of the tropical -Pacific: (_a_) Dianella ensifolia, found in Hawaii and ranging over the -Mascarene Islands, India, China, Malaya, and tropical Australia; and -(_b_) D. intermedia, recorded from most of the groups of the South -Pacific (Fiji, Tonga, Rarotonga, Tahiti), and occurring also in Norfolk -Island and New Zealand. These two plants occur in similar stations all -over Polynesia, sometimes growing in the grassy plains on the dry side -of an island, at other times extending up the thinly wooded mountain -slopes and reaching the hill-crests some 2,000 or 3,000 feet above the -sea. Their berries would readily attract birds; and their seeds, about -one-fifth of an inch (5 mm.) in size in the case of D. ensifolia, could -be carried uninjured in the stomach and intestines of a bird. - - - _Summary._ - -(1) A later period in the era of the general dispersal of Malayan plants -over the Pacific is indicated by the genera that contain species found -outside each group as well as species restricted to it. - -(2) In this period the extremely variable or polymorphous species plays -a conspicuous part, as represented in such genera as Alphitonia, -Dodonæa, Metrosideros, Pisonia, and Wikstrœmia. - -(3) The first stage is displayed by a solitary widely-ranging species -found over most of the Polynesian archipelagoes, and varying -independently in every group. - -(4) The next stage is shown where the polymorphous species, having done -its work of distributing the genus, ceases to wander and settles down -and “differentiates” in all the groups; and the genus thus includes both -peculiar and widely-ranging species in each group. Most of the genera -possessing polymorphous species are in this stage. - -(5) The following stage is displayed by those genera like Elæocarpus, -Eugenia, and Peperomia, where peculiar species are especially developed -in particular groups, and we get subcentres of distribution for the -genus, that is to say, small gatherings of peculiar species. A few -species, however, still keep up a connection with neighbouring -island-groups. Should this be severed we get the type of genus belonging -to the earlier period of the Malayan era as described in the preceding -chapter, a genus possessing only peculiar species and destined, after -ages of further isolation through the failure of the dispersing -agencies, to give rise to a new generic type or types. - -(6) Frugivorous birds were chiefly active in dispersing these genera -over the Pacific. Some of the genera possess seeds or “stones” of such a -size that at first sight their transport by frugivorous birds to Hawaii -seems improbable; but, as in the case of Elæocarpus, it is shown that -this difficulty does not apply to all species of a genus, some of them -having much smaller seeds or stones. - -(7) The close of the era of the general dispersal of Malayan plants over -the Polynesian Islands is indicated by those genera that are represented -more or less entirely by widely ranging species. Though such species may -vary among the different groups, they rarely take the rank of -polymorphous species, the agencies of dispersal being sufficiently -active to check marked variations. - -(8) Several of the genera of this concluding stage, like Rhus, Viscum, -and Plectronia, are known to be dispersed by frugivorous birds, whilst -others, like Osteomeles and Dianella, are equally well suited for this -mode of dispersal. - -(9) Distinct indications are afforded by the genera Rhus, Osteomeles, -and Dianella that the Hawaiian Group has been often supplied with its -plants directly from the Old World by the Asiatic mainland, whilst the -groups of the South Pacific have received different species of the same -genus by Malaya and tropical Australia. - - - - - CHAPTER XXVII - - THE MALAYAN ERA OF THE NON-ENDEMIC GENERA OF FLOWERING PLANTS - (_continued_) - - THE AGE OF LOCAL DISPERSAL - - - _Synopsis of the Chapter._ - -HAWAII.—(1) _The Hawaiian residual genera, being those not found in - either the Fijian or the Tahitian regions._ The genera especially - discussed are Osmanthus, Sicyos, Jacquemontia, Cuscuta, Rumex, - Dracæna, Naias, Potamogeton; and amongst others mentioned are - Perrottetia and Embelia. - - (2) _The Hawaiian genera found in Tahiti and not in Fiji._ Very few, - and illustrated by Byronia, Reynoldsia or Trevesia, Phyllostegia, and - Pseudomorus, though it is likely that most of these will be - subsequently discovered in Fiji. - - (3) _The Hawaiian genera found in Fiji and not in Tahiti._ Illustrated - by Eurya, Gouania, Maba, Sideroxylon, Antidesma, Pleiosmilax, Ruppia. - - (4) _The absentees from Hawaii._ Illustrated amongst the orders by the - Sterculiaceæ (see text), the Meliaceæ, the Rhizophoreæ, the - Melastomaceæ, and the Coniferæ, and amongst the genera by - Trichospermum Loranthus, Stylocoryne, Ophiorrhiza, Alstonia, Hoya, - Ficus; and a great many others might be cited. - -TAHITI.—(1) _The Tahitian residual genera._ Only six in number—Cratæva, - Buettneria, Berrya, Coriaria, Bidens, Lepinia. - - (2) _The Tahitian genera found in Hawaii and not in Fiji._ See above - under (2). - - (3) _The Tahitian genera found in Fiji and not in Hawaii._ (a) Those - possessing only species confined to the Tahitian region or to East - Polynesia, of which Meryta, Ophiorrhiza, Alstonia, and Loranthus are - examples. - - (b) Those possessing widely-ranging species besides, often, species - confined to the Tahitian region, such as Grewia, Nelitris, Melastoma, - Randia Geniostoma, Tabernæmontana, Fagræa, Bischoffia, Macaranga, and - Ficus. The widely-ranging species is in many genera polymorphous. - - (4) _The absentees from Tahiti._ Amongst the orders are the Meliaceæ, - the Rhizophoreæ, and the Coniferæ. Amongst the genera, usually those - with “stones” or large seeds an inch in size, such as Canarium, - Dracontomelon, Myristica, Sterculia, Veitchia, &c. Numerous other - absent genera might be named. - -FIJI.—_The Fijian genera not found either in Tahiti or Hawaii._ These - genera compose about half the Fijian flora, being at least 160 in - number. Those especially discussed here are the following:—Hibbertia, - Cananga, Sterculia, Trichospermum, Micromelum, Canarium, - Dracontomelon, Begonia, Geissois, Dolicholobium, Lindenia, Myrmecodia, - Hydnophytum, Couthovia, Limnanthemum, Myristica, Elatostema, - Ceratophyllum, Gnetum, Veitchia, Rhaphidophora, Lemna, Wolffia, - Scirpodendron. The Coniferæ are dealt with in Chapter XXIV. - - -------------- - -Note appended on Marsilea - - -HAVING completed our discussion of the general dispersal of tropical -genera, chiefly Indo-Malayan, over the Pacific islands, we pass on now -to consider the more restricted distribution of non-endemic genera over -this region. Here as before we take Hawaii, Tahiti, and Fiji as the -three centres of distribution; and here also we deal with the flowering -plants after excluding the orchids, the sedges, the grasses, the -mountain-plants, and all plants introduced either by the aborigines or -by white men. - - - HAWAII. - -After excluding the endemic genera as well as those that are confined to -the mountains, we find that this group possesses very few genera that do -not occur in the Fijian and Tahitian regions, and fewer still that it -owns in common with Tahiti to the exclusion of Fiji. On the other hand, -we observe that Fiji possesses a great number of genera, mostly Asiatic -in origin, that have not reached Hawaii, and in several cases are not -known, from the Tahitian region. These contrasts might have been -expected, since the Pacific islands have in later ages been mainly -stocked from the Asiatic side of the Pacific, the principal route lying -through the Fijian region. - -As far as the flora of the lower levels (below 4,000 feet) is concerned, -Hawaii only possesses a portion of that which Fiji has derived from the -Old World, chiefly through Malaya. Although, as will be shown below, -there is a noticeable contribution from America, it is very far from -counterbalancing the loss which the Hawaiian flora has sustained in -comparison with Fiji through the isolated position of the group. The -want of variety, however, in the flora of the Hawaiian lower levels, -which up to 4,000 or 5,000 feet represent the islands of the less -elevated Fijian region, is in a small degree compensated for by the -development of new genera and new species and by the great number of -individuals. Trees like Metrosideros polymorpha and Aleurites moluccana, -that in the southern groups form only one of many contributors to the -forests, rise suddenly into prominence in the northern archipelago and -form entire forests. Pandanus odoratissimus largely composes extensive -forests in the province of Puna in the large island of Hawaii, extending -several miles inland and nearly 2,000 feet up the mountain slopes. - -The remarkable contrast between the Fijian flora, which is almost -entirely tropical, and the Hawaiian flora, which on account of the great -elevation of the islands is temperate as well as tropical, is brought -into yet greater prominence when we look at it more closely and treat it -numerically. The Hawaiian Group, it must be first observed, though -possessing the same area as Fiji and presenting a far greater variety of -climatic conditions, has only two-thirds the number of genera of -flowering plants (see Chapter XXI., Table B). Whilst at least 200 of the -Fijian genera of indigenous plants (excluding the orchids and the -grasses) are not found in Hawaii, only about 100 of the Hawaiian genera -are absent from Fiji, and the two groups possess about 100 genera in -common. When we look more closely at the hundred Hawaiian genera not -found in Fiji, we find that about sixty represent endemic genera -(thirty-seven) and non-endemic mountain-genera (twenty-two), which -naturally are not to be found in Fiji, so that there remain but a small -number of genera distinguishing the tropical flora of Hawaii from the -Fijian flora. When we take from them a few that occur in the Tahitian -region, there is left a very small residuum characteristic of Hawaii -alone to the exclusion of the Fijian and Tahitian regions of the South -Pacific. - - - THE HAWAIIAN RESIDUAL GENERA. - -It is my purpose now to deal in an illustrative fashion with this -Hawaiian residual flora which is composed, as above explained, of the -non-endemic tropical genera that are not represented in the Fijian and -Tahitian regions. Up to the present we have been dealing with the -characters that the floras of Fiji, Tahiti, and Hawaii possess in common -as far as tropical genera are concerned. We will now proceed to discuss -their differences in this respect, and will begin with the residual -Hawaiian flora. - -After eliminating two or three genera that will probably be found in -Fiji, but including one or two others that are best treated under the -endemic genera, about twenty-seven present themselves for our purpose. -Nearly all of them possess only endemic species, and belong therefore to -an age of dispersal that has passed away. These residual genera plainly -indicate that although Hawaii largely received its flora during the age -of general dispersal of Old World genera over the Pacific, it was at the -same time independently stocked with plants from other sources. They -include among others—Cocculus (4), Cleome (1), _Perrottetia_ (1), -Mezoneuron (1), _Lythrum_, _Sicyos_ (8), Peucedanum (2), Campylotheca -(12), Senecio (2), Lobelia (5), Embelia (1), _Chrysophyllum_ (1), -Rauwolfia (1), _Nama_ (1), Osmanthus (1), _Jacquemontia_ (1), Breweria -(1), Cuscuta (1), Lycium (1), _Sphacele_ (1), _Phytolacca_, Rumex (2), -_Urera_ (2), Pilea, Dracæna (1), Naias, Potamogeton. Those printed in -italics are regarded as derived from America; whilst the figures in -brackets indicate the number of endemic species, nearly all of the -genera except the five above indicated possessing only peculiar species, -and these five (Lythrum, Phytolacca, Pilea, Naias, Potamogeton) are only -represented by species found outside the group. - -American genera form a more conspicuous element than they do amongst the -genera that have been generally dispersed over the Pacific, those -exclusively American being fairly represented, making a third of the -whole. We find, for instance, in the Hawaiian “Olomea,” Perrottetia -sandwicensis, a small tree that represents in the woods of all the -islands the Perrottetias of Mexico and the Andes; whilst with some of -those genera that, like Sicyos and Urera, are at home in both the Old -and New Worlds, we obtain indications of America being the source of the -Hawaiian plants. A few genera again, like Lythrum and Phytolacca, are -represented in Hawaii by American species. - -Plants with drupes, berries, or other fleshy fruits likely to attract -frugivorous birds compose about a third of the total number of these -residual genera, whilst fruits or seeds, that were in all probability -originally brought entangled in a bird’s feathers, are represented by -Sicyos. Some of the genera with stone fruits, such as Osmanthus, to -which belongs the Hawaiian Olive, present special difficulties on -account of the size of the stone, in this case two-thirds of an inch in -length. There are also a number of genera with large dry fruits and -sometimes large seeds, of which the method of dispersal is not easy to -discover. Thus, Mezoneuron, a Leguminous genus with seeds an inch across -(2·5 cm.), and Peucedanum, of the Umbelliferæ, with mericarps half to -three-quarters of an inch (1·2 to 1·8 cm.) in length, offer serious -difficulties to the student of plant-dispersal. In discussing the -difficulty connected with Mezoneuron (see Chapter XV.) he will keep in -view the possibility that the original species may have been a littoral -plant possessing seeds dispersed by the currents, seeds that lost their -buoyancy when the plant established itself inland, just as is now taking -place with Afzelia bijuga, a Leguminous littoral tree of Fiji (see -Chapter XVII.). - -He will also find much to puzzle him in the mode of dispersal of the -Hawaiian residual genera of the Convolvulaceæ (Breweria, Jacquemontia, -and Cuscuta) that possess only endemic species, and he will speculate as -to the manner in which seeds that would seem to possess but little -attraction for birds and have no capacity for transportation by the -currents could ever have reached these islands, and he will ask himself -why it is that the agencies of dispersal, whatever they are, have now -ceased to be active. He will perhaps see a way out of his difficulties -when he perceives that if isolation has led to the development of -peculiar species in Hawaii, it has strangely enough in the case of the -Myrsinaceous genus Embelia produced the same effect over the whole range -of the genus, and that Hawaii has in this respect derived no advantage -from being an oceanic group. According to Carl Mez, nearly all the -ninety species of this Old World genus are restricted in their areas, -whether continental or insular (“Myrsinaceæ,” _Das Pflanzenreich_, -1902); and indeed we do not seem justified in assuming that the -isolating influences in the case of this genus have been more effective -in Hawaii in the mid-Pacific, or in Mauritius in the Indian Ocean, than -they have been in continental regions like the Deccan and Nyassa Land, -in all of which localities endemic species occur. - -The remarkable development of the Cucurbitaceous genus Sicyos, in Hawaii -alone of all the tropical Pacific groups, will attract his attention, -and he will find here another instance of that predominant principle in -the distribution of Pacific plants, where in a widely-ranging genus we -find one of its species covering most of its area, whilst the other -species are more or less localised. He will wonder at the limitation to -Hawaii of a genus like Dracæna, that is so well adapted for dispersal -over the Pacific by frugivorous birds; and in endeavouring to explain -the presence in the Hawaiian forests of the gigantic Rumex, R. -giganteus, he will remember that the small group of Tristan da Cunha, -equally isolated in the South Atlantic, possesses an endemic species of -the same genus. He will discover in the recognised dispersing agencies -of wild ducks and other waterfowl an explanation of the occurrence in -Hawaii of the aquatic genera Naias and Potamogeton; but he will be -puzzled at their restriction to this group alone of the three tropical -Pacific archipelagoes here especially discussed. - -Amidst these various perplexities he will probably look with relief on -the appearance of Phytolacca brachystachys, an endemic species of the -American “pokeweeds”; and he will feel grateful to the American -botanists like Professor Weed when they tell him that in the United -States crows, blackbirds, and other birds successfully disperse these -plants, the seeds of which are sometimes able to pass through the -alimentary canal undigested. - -But by far the most significant lesson that the student of distribution -will carry away from his study of the Hawaiian residual genera will be -that which he learns from the genera Embelia and Naias. He perceives -here that not only with a typical land-genus has specific -differentiation occurred to much the same extent in the continental and -insular localities of its range, but that even with a typical genus of -submerged aquatic plants, where the conditions of existence are as -uniform as they are varied in the case of land plants, the process of -differentiation has proceeded on the same broad lines in the interior of -a continent and in an island in mid-ocean. - -The following notes on some of the residual genera refer more -particularly to matters connected with distribution and dispersal. - -_Osmanthus_ (Oleaceæ).—This genus, according to the _Index Kewensis_, -contains six species localised in their several habitats of North -America, Hawaii, Japan (two), Hongkong, and the Himalayas. Its -representative in this group is the Hawaiian Olive, the Olea -sandwicensis of Gray, a prevailing tree in the lower and middle woods -(1,000 to 4,000 feet) of all the islands, which, like other Hawaiian -plants, such as those of the genera Eurya and Antidesma, indicates that -the group has been sometimes independently stocked from the regions of -the northern hemisphere. The drupe of this tree contains a stone -two-thirds of an inch (17 mm.) in length, and suitable for dispersal by -frugivorous birds; and birds have evidently distributed the tree all -over the group. In fact Mr. Perkins in mentioning the favourite food of -birds of the Hawaiian genus, Phaeornis, refers to the fruits of this -tree as well as of the Opiko (Straussia) and of the Olapa -(Cheirodendron). When, however, we come to consider the feasibility of -the stones of the genus having been thus originally carried to Hawaii -either from Japan or from North America, we meet with the difficulty -presented to us by other Hawaiian genera with stone-fruits, such as -Elæocarpus, or with berries containing large seeds, such as Sideroxylon. - -_Sicyos_ (Cucurbitaceæ).—This genus comprises about thirty-five known -species, of which three-fourths are confined to the New World, being -mainly South American, whilst the remainder are restricted to Hawaii, -with the exception of two species in the Galapagos Group and Norfolk -Island, and a widely-ranging species, S. angulatus. The plant just -named, the small fruits of which possess hooked spines, adapting them -for dispersal in a bird’s plumage, occurs in Africa, Australia, New -Zealand, and America, but has only been recorded in the Pacific islands -from the Kermadec Group. - -North America was probably the home of the original Hawaiian species. -Hillebrand describes eight species, of which five are not found in more -than one island, whilst one species is spread over most of the islands. -The fruits vary much in size, and only in a couple of species do they -now possess any fitness for attaching themselves to plumage, some of -them being pubescent or even glabrate, so that deterioration in the -capacity for dispersal has here taken place. Their size is usually a -quarter to half an inch (6-12 mm.); but it is noteworthy that the -species with the largest fruit (Sicyos cucumerinus, one to two inches, -or 25 to 50 mm.) is the species most widely dispersed over the group. -This appears to indicate that there is some other means of inter-island -dispersal in this archipelago than by attachment to birds’ plumage. The -isolation of the genus in Hawaii from the rest of the world is, however, -complete, since all the species are endemic; and when, therefore, we -come to ask how Sicyos angulatus, that has been dispersed in the recent -era over America, Australia, and New Zealand, is not found in these -islands, we are brought face to face with the ever-recurring difficulty, -the suspension in later times of the agency of dispersal in the tropical -North Pacific. - -_Jacquemontia_ (Convolvulaceæ).—This genus, which is chiefly American, -is represented in Hawaii by a peculiar species, J. sandwicensis. This -species grows occasionally on the sandy beaches associated with -Heliotropium anomalum and Tribulus cistoides; but it is most at home on -rocky ground and on old lava-flows near the sea-border, making its abode -often in the pockets of black sand produced by the disintegration of the -lava. Its small seeds sink in sea-water even after prolonged drying; and -it can perhaps be supposed that the original seeds were brought from -North America in the crevices of a drifting log. According to Ridley, -Fernando Noronha possesses a peculiar species also growing near the sea; -and it may be that the drifting log has here been the agent also: but in -neither case would this explanation account for the endemic character of -the species. - -_Cuscuta_ (Convolvulaceæ).—It would seem that with the exception of -Hawaii, where an endemic species, C. sandwichiana, occurs, no other -oceanic group in the globe possesses a peculiar species of the Dodders. -With the exception of an endemic species in New Zealand, and an -introduced species in Fiji which is found usually near the gardens of -the white residents on Viti Levu, the genus takes but little part in the -Pacific floras. The Hawaiian species is a characteristic beach-plant -growing on Ipomœa pes capræ, Scævola Kœnigii, Tribulus cistoides, and on -other plants that find a permanent or a temporary abode on the beaches. -We learn from Ridley and Moseley that Cuscuta americana in Fernando -Noronha finds its host also in Ipomœa pes capræ. Since the seeds of the -Hawaiian plant and of the introduced Fijian species possess no buoyancy, -even after drying for years, we cannot look to the agency of the current -unless we call the drifting log to our assistance, and in that case the -endemic character of the Hawaiian species would present the difficulty -already alluded to in the case of Jacquemontia. The seeds of the -Hawaiian plant are about one-twelfth of an inch (2 mm.) in diameter, and -as far as size is concerned they might have been transported in a bird’s -stomach; but, on account of the rapidity with which the seeds of the -genus absorb moisture and swell up, it is most unlikely that they would -escape injury. This is one of the several difficulties in -plant-dispersal which New Zealand and Hawaii share in common. Further -remarks on the germination of the Hawaiian species are made in Note 69. - -_Rumex_ (Polygonaceæ).—Hawaii possesses two peculiar species of Rumex, a -genus not recorded from any other of the Polynesian groups. One of these -species, R. giganteus, is a very remarkable plant, growing to a height -of thirty or forty feet when supported by trees. It is noteworthy that -the small group of Tristan da Cunha in the South Atlantic possesses a -species, R. frutescens, confined to those islands (_Bot. Chall. Exped._, -ii. 154). Both Hawaii and Tristan da Cunha lie in mid-ocean, cut off -from the nearest continent by some 1,800 or 2,000 miles of sea; and we -may have to choose between the bird and the current in selecting the -agency concerned with the transportation of the original seeds; or -perhaps they have co-operated. Birds could disperse the nutlets of Rumex -as readily as they do those of Polygonum, and I have found these fruits -at times in the stomachs of partridges. On the other hand, Rumex fruits -occur amongst the drift stranded on beaches in England and in -Scandinavia; and, as indicated by the observations of Sernander and -myself in these two localities, they float through the winter in ponds -and rivers, germinating afloat in the spring. The nutlets sink, but they -owe their buoyancy to the persistent perianth. In my sea-water -experiments the fruits of Rumex hydrolapathum and R. conglomeratum were -still afloat after from six to twelve months’ immersion, and their seeds -subsequently germinated. It is quite possible, therefore, that currents -can carry these fruits unharmed to oceanic island-groups like Hawaii and -Tristan da Cunha. - -_Dracæna_ (Liliaceæ).—This Old World genus, which on account of its -berries is eminently suited for dispersal by frugivorous birds, is -represented in Polynesia by a solitary species (D. aurea) peculiar to -the Hawaiian Group. Attaining a height of twenty to twenty-five feet, it -often forms a striking feature in the vegetation of the open wooded -regions up to altitudes of 3,000 feet. I found it growing in abundance -in the large island of Hawaii between Waimanu and Waipio, and on the -northern slopes of Hualalai. It grows in a variety of stations, and I -came upon it once in the broken-down caverns of an old lava-flow that -were frequented by pigeons which no doubt brought the seeds. Its -conspicuous yellow berries have hard rounded seeds a quarter of an inch -(6 mm.) across and weighing two to three grains when dry, which would -probably withstand injury in a bird’s stomach, the minute embryo being -protected by a very tough albumen. Neither the entire berry nor the seed -could be transported by currents, the last sinking even after drying for -six years. - -_Naias_ (Naiadaceæ).—If we except New Caledonia, where two or three -species have been found, Hawaii is the only island-group in the tropical -Pacific from which this interesting world-ranging genus of submerged -aquatic plants has been recorded. Chamisso, the celebrated naturalist of -Kotzebue’s expedition, collected Naias marina in Oahu in the early part -of last century; but apparently it did not come under Hillebrand’s -observation in the group. However, in 1897 I found it in another -locality, namely, just within the mouth of the Waipio, a river on the -north-west side of the island of Hawaii. The mature fruits of this genus -have never been experimented on by me; but there is nothing in the -structure of the fruits to indicate that they have any buoyancy, or to -show that they differ in this respect from the fruits of other -completely submerged aquatic plants like Ceratophyllum, Ruppia, and some -of the Potamogetons. It is to ducks and other waterfowl that we must -attribute the dispersal of this and the other genera just mentioned over -wide tracts of ocean, a subject dealt with in discussing those plants. - -The Hawaiian Group probably represents the most isolated locality -occupied by this genus, since none of the other islands from which -species have been recorded, such as New Caledonia, Mauritius, and -Bourbon, are so far removed from continental regions. The source of the -Hawaiian form of Naias marina lies evidently on the Asiatic side of the -Pacific, since it is referred by Mr. Rendle to the variety -“angustifolia,” an Asiatic plant found also in the island of Bourbon and -in West Australia, but not recorded from the New World. The important -little monograph of the genus by Mr. Rendle (“Naiadaceæ,” in Engler’s -_Das Pflanzenreich_, 1901) is full of suggestiveness for the student of -plant-distribution. His interest is excited when he discovers that one -of the most typical genera of aquatic plants displays the same principle -of differentiation at work that is so well illustrated by many of the -land genera of the Pacific islands. I refer to the principle implied in -the existence of a widely-ranging genus comprising “a polymorphic -species occurring over almost the whole area of the genus,” as well as a -number of less widely distributed species, most of which have -“restricted areas and fall for the most part into small geographical -groups.” I have just been quoting Mr. Rendle’s description of the -distribution of Naias, the “polymorphic” species concerned being N. -marina; but it need scarcely be remarked that it would apply just as -well to several of the land genera dealt with in the previous chapter -(XXVI.), such as Alphitonia, Metrosideros, Pisonia, &c. - -Although there is such a contrast in the degree of uniformity of their -life-conditions between land and water plants, a strictly aquatic plant -being but slightly affected by changes in the physical conditions that -are accompanied by a complete transformation in the character of the -terrestrial vegetation, yet—and this is the important point—we find the -same principle of differentiation at work with both land and water -plants. If one wished to produce proof of the contention that the -production of new species is largely independent of external conditions, -one could not do better than take the cases of Elæocarpus, Metrosideros, -and Naias. In all cases we see a widely-ranging polymorphous species -settling down and “differentiating” in particular localities or regions, -and forming subcentres for the distribution of the genus. - -_Potamogeton_ (Potameæ).—Though well suited for dispersal by waterfowl, -the Potamogetons have been recorded from the Hawaiian and Marianne -Islands alone among the tropical groups of the open Pacific. The genus, -though not so well represented in insular floras as we might have -expected, is still not infrequently to be found. Widely-ranging species -have been observed in the Azores, Madeira, and the Canaries in the -Atlantic, as well as in Hawaii in the Pacific; whilst species have been -recorded that are peculiar to Martinique, the Mascarene Islands, and to -the Marianne Group. Hillebrand gives for Hawaii, Potamogeton fluitans, a -plant of the Old and New Worlds, and P. pauciflorus, a North American -species; whilst in the _Index Kewensis_ a peculiar species, P. -owaihiensis of Chamisso (which is, however, regarded by Hillebrand as a -form of P. fluitans), is also accredited to the group. Owing, however, -to the paucity of streams and rivers this genus takes no prominent part -in the Hawaiian flora, and the species seem to have been recorded alone -from Oahu. As they were discovered by Chamisso in the early part of last -century they are in all probability truly indigenous in Hawaii, even if -none are peculiar to the group. - -That ducks and similar birds are the agents in carrying the seeds of -Potamogeton to oceanic islands cannot be doubted. About twelve years ago -I examined the stomachs and intestines of thirteen wild ducks obtained -in the London market. Three of them contained in all forty-one -Potamogeton seeds, or rather “stones,” most of which subsequently -germinated in water. In one of my experiments, carried out in the month -of December, I fed a domestic duck with the fruits of Potamogeton -natans. They appeared in quantity in the droppings, for the most part -divested of their soft coverings, but otherwise uninjured. Sixty per -cent. germinated in the following spring; whilst of those left in the -vessel, from which the duck had been fed, only one per cent. germinated -in the next spring, and another year elapsed before any number did so. -These results were published in _Science Gossip_ for September, 1894. - -One often reads in books of travel interesting remarks bearing -indirectly on the dispersal of the Potamogetons. Thus, when Sir Joseph -Hooker (then Dr. Hooker) noted in his _Himalayan Journals_ the -occurrence of P. natans in the Neongong Lake in the Himalayas, and the -presence of coots, he most probably mentioned the bird that brought the -plants, coots being active distributors of the seeds of water plants. It -is of importance to remember that (as shown in my experiment on the -duck) seeds of water-plants are voided in a condition peculiarly -favourable to early germination. Ducks, coots, and other water birds -might often be characterised as “travelling germinators.” My experiment -showed that seven to eight hours at least were occupied by Potamogeton -nutlets in passing through the digestive canal of a duck, and that -probably nine or ten hours would be required after an average full meal. -But this does not represent the possible maximum period, since the bared -“stone” may remain in the gizzard for a long time with ordinary gravel. -Most of the Potamogeton fruits found by me in wild ducks were obtained -from the gizzard, where they were mixed with gravel and other hard seeds -or seedvessels, as described in Chapter XXXIII. Such fruits afterwards -germinated. With regard to the chances, therefore, of the fruits of -Potamogeton being carried by a bird without injury across an ocean, we -may infer that, whether they are retained in its body for only ten hours -or for as long as three or four days, they will preserve in some cases -their germinating power. - - - HAWAIIAN GENERA FOUND IN TAHITI TO THE EXCLUSION OF FIJI. - -Taking only the genera that are strictly indigenous, and excluding -therefore all those introduced by the aborigines, the number available -for establishing an independent connection between the Hawaiian and -Tahitian regions is exceedingly few. Amongst the Hawaiian shore-plants -not found in Fiji proper but occurring in the Tahitian region are -Heliotropium anomalum and Sesuvium portulacastrum. The last-named, -however, has been recorded from Tonga, which lies within the Fijian -area; whilst the first will probably be found in the same region. -Amongst the Hawaiian and Tahitian mountain genera not recorded from Fiji -proper are Nertera, Vaccinium, Cyathodes, and Luzula. As is pointed out -in Chapter XXIII., the absence of these genera from Fiji is connected -with the relatively low elevation of those islands, though it is quite -possible that one or more of them may yet be found on the highest -summits of Fiji; and indeed Nertera depressa and Vaccinium have been -discovered in the more elevated uplands of Savaii in Samoa. - -After removing the littoral plants and the mountain genera, there are -probably not more than half a dozen inland genera that connect the -Hawaiian lowlands with the Tahitian region to the exclusion of the -Fijian Group; and Byronia (Ilicineæ), Reynoldsia or Trevesia -(Araliaceæ), Phyllostegia (Labiatæ), and Pseudomorus (Urticaceæ) may be -taken as examples. Of these, Pseudomorus, which has a small drupaceous -fruit suitable for dispersal by frugivorous birds, has been recorded -from New Caledonia, and not improbably it exists in the Fijian area; and -the same may be postulated of Reynoldsia, which is discussed in a later -page, since it has been found in Samoa. We may almost form the same -opinion of Byronia, since it exists in Australia. This genus of small -trees contains only three known species, one in Australia, one in -Tahiti, and one in Hawaii. Its fleshy drupes, about a third of an inch -(8 mm.) in size, would attract birds, and their numerous cartilaginous -pyrenes would probably pass unharmed through a bird’s alimentary canal. -Phyllostegia, a Labiate genus with fleshy nucules that might attract -birds, is, with the exception of a solitary Tahitian species, entirely -confined to Hawaii (see Chapter XXII.). - -From these data it may be inferred that the interchange of plants -between the regions of Hawaii and Tahiti to the exclusion of Fiji has -been very slight. The facts of distribution are just such as we might -look for in the case of a general dispersal over the oceanic groups of -the tropical Pacific, with the altitudes of the islands playing a -determining part. In this general dispersal Hawaii has shared; and -except in the case of Phyllostegia it is evident that this group has -kept nearly all it received and has distributed but little. - - - HAWAIIAN GENERA FOUND IN FIJI TO THE EXCLUSION OF TAHITI. - -We shall be able to throw further light on the floral history of Hawaii -by discussing the few tropical genera, not a score in all, that it -possesses in common with Fiji to the exclusion of the Tahitian region. -The following genera offer themselves for treatment:—Eurya -(Ternstrœmiaceæ), Gouania (Rhamnaceæ), Maba (Ebenaceæ), Sideroxylon -(Sapotaceæ), Antidesma (Euphorbiaceæ), Pleiosmilax (Smilaceæ), and -Ruppia (Potameæ). - -These seven genera, which with the exception of Ruppia, an aquatic -genus, are only represented in Hawaii by peculiar species, possess in -all cases, except Gouania and the last-named genus, drupaceous or -baccate fruits likely to attract frugivorous birds. Two of them, Eurya -and Antidesma, have their home in Malaya and in the Asiatic continent; -three of them, Gouania, Maba, and Sideroxylon, are found on both the -Asiatic and the American sides of the Pacific Ocean; whilst Pleiosmilax -should, strictly speaking, be regarded as a Polynesian subgenus of -Smilax, a world-ranging genus; and Ruppia is a cosmopolitan brackish- -and salt-water genus. - -It is highly probable that Fiji received almost all these genera from -the Old World through Malaya; and in some cases the resemblance between -the Malayan and the Fijian species is so close that, as in Gouania, Dr. -Seemann questioned if they were not forms of the same species. In other -instances, as with Maba, we find a widely-ranging Asiatic and Malayan -species, like Maba buxifolia, extending into Western Polynesia, where it -is accompanied by other species peculiar to that region. But if the -genera were able subsequently to extend their range thence to Hawaii, it -is difficult to understand why they have not reached the Tahitian -region. It is therefore likely that Hawaii received most of these genera -by a northern route and not through the South Pacific; and it is -legitimate to suppose that when Old World genera like Eurya and -Antidesma occur in north-eastern Asia, as in Japan and in the -neighbouring mainland, Hawaii received the genus by that route. In the -case of Eurya it is noteworthy that Fijian and Samoan forms, regarded by -Seemann and Gray as distinct species, are viewed by Reinecke as forms of -E. japonica, an extremely variable species found in Japan. With genera -like Gouania and Maba, that exist on both sides of the Pacific, it is -possible that they may have originally reached Hawaii from America. - -A noticeable feature in the instance of genera like Maba and Sideroxylon -is that hard seeds or pyrenes 3/4 to 1 inch (18 to 25 mm.) in length -have seemingly been transported by frugivorous birds across the ocean to -Hawaii. This at first sight seems improbable; but it is known that -fruit-pigeons can swallow very large drupes, as in the case of those of -Canarium, Dracontomelon, and Elæocarpus, afterwards disgorging the -“stones.” They have carried such stones to Fiji, across some 500 or 600 -miles of ocean; and unless we impute a continental origin to Hawaii we -must assume that in some cases, as with Elæocarpus, Maba, and -Sideroxylon, they have been able to transport these large stones or -pyrenes to that group. The extent of ocean to be crossed is no doubt -much greater, but this area of the Pacific is not without some small -half-way groups that would serve as resting-places. - -That fruits of the order Sapotaceæ are much appreciated by fruit-pigeons -is already known. We learn from Kirk that the fruits of Sideroxylon -costatum (Sapota costata) are a favourite food of the New Zealand -fruit-pigeon, the fruits, about an inch long, containing three hard -crescentic bony seeds nearly as long as the fruit. The natives of Vanua -Levu informed me that a Fijian species of Sideroxylon with hard seeds -about an inch long was much appreciated on account of its fruit by the -pigeons. I found the hard, sound seeds of a species of Sapota, -two-thirds of an inch (or 16 mm.) in size, in the crop of a Fijian -fruit-pigeon. The similarly large seeds of a species of Achras were -identified by Mr. Charles Moore, of Sydney, amongst a collection of -seeds, &c., found by me in the crops of fruit-pigeons shot in the -Solomon Islands (Guppy’s _Solomon Islands_, p. 293). It may be added -that the difficulty concerned with Sideroxylon in Hawaii is the -difficulty concerned with other large-seeded Sapotaceous trees in Fiji -and New Zealand, and the same explanation must be applied to all. Some -further remarks on the Sapotaceæ in the Pacific are given below. - -The mode of dispersal of some of these genera is illustrated in other -regions. The berries of Pleiosmilax, a subgenus of Smilax, are well -suited for aiding the dispersal of the genus by frugivorous birds; and -we learn from Prof. Barrows (Weed, p. 42) that in the United States -crows feed on the fruits of Smilax rotundifolia and disperse the seeds. -On the other hand, it is not at first sight easy to understand how a -genus like Gouania has been distributed over the tropics of the globe, -since it possesses dry capsular fruits about half an inch across, -separating into three woody cocci that appear most unlikely to attract -birds. The same difficulty exists, however, with other dry-fruited -widely-ranging genera like Alphitonia and with many of the Euphorbiaceæ. - -Amongst these genera found in Hawaii and Fiji to the exclusion of Tahiti -we can at times detect indications of the operations of a polymorphous -species as described in Chapter XXVI., when a widely-ranging highly -variable species is associated in some groups with peculiar species. We -see some evidence of this in the genera Gouania, Maba, and Eurya, -alluded to on a previous page. (See also _Bot. Chall. Exped._, iii. 134, -under “Gouania.”) - -One of the mysteries of the Pacific is concerned with the distribution -of the Sapotaceæ, the dispersal of which by frugivorous birds has been -dealt with above. It is strange that whilst the order seems to have -found a _rendezvous_ in Tonga, no one except Horne appears to have -recorded any of the genera from Samoa. They are fairly well represented -in Fiji; but it is in Tonga that we especially note the gathering -together of several Sapotaceous trees with large heavy seeds, of the -genera Bassia, Mimusops, and Sideroxylon. Besides owning one or two -species of Sideroxylon in common with Fiji (Burkill), this small group -possesses Bassia amicorum and Mimusops kauki, both of which were found -there by Forster at the time of Cook’s visit. In a list of a small -collection of plants made by him in Upolu in the Samoan Group about -1879, Horne includes two species of Sideroxylon (_Year in Fiji_, p. -286); and according to Seemann there is a Sapotaceous tree in Wallis -Island. A species of Bassia exists in Rarotonga, the seeds of which, -from Mr. Cheeseman’s description of the fruit, must be almost an inch -long. Drake del Castillo refers to an endemic Tahitian tree near -Mimusops; but its fruit was not known to him. - -As already indicated, the difficulties connected with the Sapotaceæ -affect the whole Pacific from New Zealand north to Hawaii and from Fiji -east to Tahiti. We are driven to appeal to the agency of frugivorous -birds, at least in the case of Sideroxylon, since some fruits -experimented on by me in Fiji sank at once or in a day or two, the seeds -having no buoyancy. That birds actually disperse the seeds of this and -other genera of the order has been already pointed out, yet it is -possible that currents have at times aided in the dispersal of some of -the genera. This is indicated by the circumstance that, as we learn from -Schimper, some Sapotaceous trees are to be included in the Malayan -strand-flora, namely, Sideroxylon ferrugineum, Mimusops kauki, and M. -littoralis, all occurring as well on the Asiatic mainland, the first -growing also in the Liukiu Islands, and the last in the Andaman and -Nicobar Groups. - -_Ruppia maritima_ (Potameæ).—This cosmopolitan aquatic plant has only -been recorded in Polynesia from Hawaii, Samoa, and Fiji. It had not been -collected in Fiji before my discovery of it in 1897. Amongst other -oceanic islands where it occurs may be mentioned the Bermudas, where, -according to Hemsley, it exists as an indigenous plant in the lagoons. -Chamisso first noticed it in Hawaii, and Hillebrand remarks that it -grows in shallow waters along the coasts. Amongst other localities where -I noticed it in this group may be mentioned the north-west coast of the -large island of Hawaii between Kailua and Keahole Point. Here in 1896 it -was thriving in brackish-water ponds, with Sesuvium portulacastrum -growing at the edges. Reinecke observes that it occurs in similar ponds -in Samoa. In 1897 I found it in abundance in the Rewa estuary (Fiji), -both in the creeks and in the main channel. In the following year it was -not to be found in this locality, a circumstance noticed both by the -natives and by resident whites. The fruits of this plant possess no -floating power, sinking, even after prolonged drying, in a few hours. It -is to ducks and to birds of similar habit that its dispersal must be -attributed. - - - THE ABSENTEES FROM HAWAII. - -It has been before remarked that of the 330 or 340 genera of -flowering-plants recorded from Fiji some 200 are not known in Hawaii. It -will only be possible to deal with the absent genera in a cursory -manner; but enough will be done to show that we are face to face here -with a multitude of the seeming inconsistencies that so often beset the -study of plant-distribution. - -A host of plants are unrepresented in Hawaii, of which it may be said -that their seeds or fruits are not less suited for being carried across -the Pacific than those of many that are now in that group. On the other -hand, a number of genera exist there which we should never expect to -have been endowed with the capacity, and to have received the -opportunity, of crossing nearly 2,000 miles of ocean. Yet perhaps when -Nature acts in a wholesale fashion and excludes entire orders we may be -able to perceive the dim outlines of a principle of exclusion at work. -But even here much caution and some clearing of the ground are needed. - -For example, having regard to the several modes of dispersal possessed -by the great variety of fruits and seeds of the Sterculiaceæ, it would -be almost meaningless to remark that the order so well represented in -Fiji is practically non-existent in Hawaii as far as truly indigenous -plants are concerned. It is true that two species of Waltheria are here -present, but one of them W. americana, is a weed probably introduced by -the aborigines whilst the other, W. pyrolæfolia, recorded from a -solitary locality by the Wilkes Expedition, has seemingly never been -found since. From the standpoint of dispersal the genera Sterculia, -Heritiera, Kleinhovia, Melochia, and Commersonia, that are represented -in Fiji but not in Hawaii, cannot be discussed together. With Sterculia -is concerned the dispersal by birds of large seeds, an inch in length, -not particularly well protected, the genus being confined to Fiji alone -of all the oceanic Pacific groups. Heritiera is only represented by a -littoral species, the large fruits of which are carried great distances -by the currents; and no other agency of dispersal is here possible. The -last three genera are distributed over the South Pacific, their -relatively small seeds being probably in the main dispersed by -granivorous birds; whilst the setose fruits of Commersonia may have been -at times transported in birds’ plumage. - -It is more legitimate, perhaps, to speak collectively of the orders -Meliaceæ and Melastomaceæ as absent from Hawaii; but even here the issue -raised is one concerned rather with opportunities than with capacities -for dispersal. Several years ago, M. Casimir de Candolle remarked that -“it is hardly credible that the Meliaceæ should be entirely absent from -the Sandwich archipelago” (_Trans. Linn. Soc. Bot._, vol. i. 1880). Yet -it can scarcely be said that this is a matter connected with means of -dispersal. Amongst the Meliaceous genera represented in Fiji, Vavæa and -Aglaia have a berry, Melia has a drupe, and Dysoxylum has a capsule. So -again with the Melastomaceæ; it possesses at least six genera in Fiji, -two in Tahiti, and none in Hawaii. Whilst the genera Melastoma and -Medinilla have baccate fruits with minute seeds, Astronia has a capsule -with similar seeds, and Memecylon has a single-seeded berry. Since, -however, minute seeds are most typical of the order, those of Melastoma -denticulatum being about one-fiftieth of an inch or ·5 mm. in size, it -would seem that this character has not aided its dispersal in the -Pacific so far as Hawaii is concerned. From the circumstance that -berries, drupes, and capsules are represented in these two Fijian orders -we may form the opinion that their non-occurrence in Hawaii is due not -so much to lack of capacities for dispersal as to failure of -opportunities. - -This opinion is much strengthened when we come to deal with the -individual genera, where the predominant cause of the absence of so many -Fijian genera from Hawaii is concerned with the failure of the agencies -of dispersal. It is not a question of a difference in size between the -groups, since, although the surface-area is approximately the same in -both groups, Hawaii possesses only two-thirds of the number of genera -occurring in Fiji. It is not a question of capacity for dispersal across -an ocean, since birds have transported across the Pacific to Hawaii the -“stones” and large seeds of genera like Elæocarpus and Sideroxylon, a -feat that would have been deemed impossible by many botanists. It is no -lack of capacity for dispersal that has excluded Loranthus from Hawaii -and has admitted Viscum. - -Few genera, indeed, would seem to be better fitted for dispersal by -frugivorous birds in the Pacific than that of Ficus. Its fruits are -known to be eaten by birds all over the area of the genus; and we find -the species distributed over the South Pacific from Fiji to Tahiti, but -they are quite absent from Hawaii. This is the more remarkable on -account of the occurrence of a species of Ficus resembling a banyan in -Fanning Island about 900 miles south of the group (_Bot. Chall. Exped._, -iii. 116, 194), and because the Hawaiian Islands possess the Meliphagidæ -or Honey-eaters, which are widely distributed in Polynesia and are known -to feed on these fruits—a matter further discussed in my treatment of -Ficus later on in this chapter. - -Of several Rubiaceous genera with fleshy fruits that are represented -both in Fiji and Tahiti, such as Stylocoryne and others, and of those -Rubiaceous genera with minute seeds that, like Ophiorrhiza, are -distributed over the South Pacific, none occur in Hawaii. Here we find -represented other genera of the order, like Gardenia, Plectronia, and -Coprosma, that do not appear to be better fitted for dispersal by -frugivorous birds than many of the genera not existing there. If birds -have carried to Hawaii in their plumage the fruits of Pisonia and -Sicyos, it cannot be merely a question of capacity for dispersal that is -concerned with the restriction to the South Pacific of genera with hairy -seeds, such as Trichospermum, Alstonia, and Hoya. - -It is unnecessary to dwell longer here on the subject of the Hawaiian -absentee-genera, since many of the absent plants will be discussed when -dealing with the peculiarities of the Fijian flora. The data there given -all go to show that mere lack of capacity for dispersal over the Pacific -often counts for little in supplying us with an explanation of the -absence of so many likely genera from the Hawaiian flora. Hawaii has -only been stocked with those genera common to Fiji and Tahiti that could -have reached it during each age of general dispersal over the Pacific. -In later eras the dispersing agencies have been mainly active in the -tropical South Pacific; and thus it is that, as will be pointed out in a -later page, the bulk of the plants of the Malayan era are confined to -the region between Fiji and Tahiti. In a still later period the -dispersing agencies have confined their operations mainly to Western -Polynesia and the last immigrant genera have not reached beyond the -Fijian region. - -The whole story of plant-life in the tropical Pacific is bound up with -these successive stages of decreasing activity of the dispersing -agencies. The story of plant-distribution in this region is well -illustrated in its earlier phases of general dispersion in the floral -history of Hawaii, in its later phase by those Asiatic genera that have -only crossed the South Pacific to Tahiti, and in its last phase by those -genera that have never extended beyond the groups of the Fijian area. -The area of active dispersion, that first comprised the whole of the -tropical Pacific, was afterwards restricted to the South Pacific, and -finally to the western portion of that area. It can scarcely be doubted -that these successive stages in the contraction of the area of active -dispersion of plants in the Pacific were accompanied by a corresponding -diminution in the general distribution of birds in the same ocean, to -which it stood in the relation of an effect to a cause. - - - TAHITI. - -The peculiarities of the Tahitian flora as compared with Hawaii and Fiji -may be discussed by treating first those genera that are alone -represented in Tahiti, the “residual” genera; then those that it -possesses in common first with Hawaii and then with Fiji; and lastly by -pointing out the more noticeable gaps in the flora. By Tahiti is -typically signified the whole Tahitian region, which includes the -Austral and Cook Groups, the Society Islands, the Paumotus, and the -Marquesas. - - - THE TAHITIAN RESIDUAL GENERA. - -The non-endemic genera occurring alone in the Tahitian region and not -found either in Hawaii or in one or other of the three groups of the -Fijian region (Fiji, Tonga, Samoa) are not more than half a dozen. These -six genera are exceedingly interesting; but since each tells a different -story and gives its own independent indication they cannot be treated in -a collective sense. Nor are they all to be regarded as anomalies in -plant-distribution, since with a single exception there is scarcely one -concerning which it is not in some way possible to give an explanation -of its isolation without coming into conflict with the principles of -plant-dispersal. The exception is Lepinia tahitensis, which, without -presenting any very evident capacity for dispersal, has not been -recorded from any other localities in the Pacific than the far-separated -Solomon and Tahitian Groups. There is a suspicion that, as in the case -of the residual genera of Hawaii, America may have contributed some of -the original plants, since three of the genera, Buttneria, Coriaria, and -Bidens, occur in that continent, and in the case of Coriaria Tahiti -possesses a species found in South America as well as in New Zealand. - -One of the trees in question is Cratæva religiosa, an Asiatic species, -which may be placed among a group of trees, including Cananga odorata -and Fagræa Berteriana, which, whilst they are much esteemed by the -inhabitants of the South Pacific for their fruits or their flowers, and -are often planted in and around their villages, possess fruits that -attract birds, and in the case of Cananga are known to be dispersed by -fruit-pigeons. Probably the aborigines and the birds have worked -together in the distribution of these trees. - -The genera Buttneria of the Sterculiaceæ and Berrya of the Tiliaceæ are -represented in this region by species that must owe their dispersal to -birds, though I have no data relating to the matter of their dispersal, -their fruits being capsular, in the first case prickly. Coriaria is a -mountain genus in Tahiti and will be found discussed in Chapter XXIV. in -connection with the Tahitian mountain-flora. Its absence from the West -Polynesian groups is no doubt to be connected with their insufficient -altitude. In addition to the introduced Bidens pilosa, a common tropical -weed, Tahiti possesses two other truly indigenous species of Bidens, of -which one at least is peculiar to the region. The achenes of this genus -are well known to be adapted for dispersal in a bird’s feathers; and -since the genus has its principal home in America, no other indigenous -species having been recorded from South Polynesia, it is not unlikely -that the parent species was American. - -One of the numerous enigmas of the Pacific floras is concerned with the -presence in the islands of Tahiti and Moorea (Eimeo), in the Society -Group, of the Apocynaceous tree, Lepinia tahitensis. The genus contains -this solitary species, which has been collected only in one other -locality, namely, in the Solomon Group, where it was obtained by the -Rev. R. B. Comins. Such an instance of disconnected distribution is rare -in the Pacific Islands, and undoubtedly it represents one of the -difficulties of the Tahitian flora. The fruits, which are indehiscent -and five or six inches in length, possess a fibrous pericarp and a -single seed. No data are to hand relating to the capacities for -dispersal possessed by this plant, but it is certain that it has had -some means of crossing the sea between the adjacent islands of Tahiti -and Moorea. (See Hemsley, _Journ. Linn. Soc. Bot._, xxx. 165.) - - - TAHITIAN GENERA FOUND IN HAWAII TO THE EXCLUSION OF FIJI. - -This subject has been already discussed in this chapter in dealing with -the genera restricted to Hawaii and Tahiti. - - - TAHITIAN GENERA FOUND IN FIJI TO THE EXCLUSION OF HAWAII. - -Excluding the orchids, sedges, and grasses, as well as the few endemic -genera, between sixty and seventy genera, or rather less than half of -the genera of the flowering-plants of Tahiti, are found in Fiji to the -exclusion of Hawaii. Of these, rather over a half are Old World genera; -about a third occur in both the Old and the New World; four are confined -to Polynesia, and not one is exclusively American. One-third are genera -now possessing in the Tahitian region endemic species either entirely or -in part, and in such cases we may consider that the agencies of -dispersal are now inactive or partially suspended; the others belong -entirely to the present era of dispersal. About half have more or less -fleshy fruits fitted for dispersal by frugivorous birds. About a fourth -have capsular or other dry fruits that must have been also dispersed by -birds preferring a drier diet. Three only possess hairy seeds or fruits -suitable for being carried in a bird’s plumage, namely, Commersonia, -Weinmannia, and Alstonia. There remain about a fourth of the total that -are shore-plants dispersed by the currents, being in two cases (Ximenia -and Kleinhovia) assisted by birds; whilst Triumfetta, another littoral -genus, is probably distributed by birds alone. - -There are no cases of special difficulty from the standpoint of -dispersal in these sixty and odd non-endemic genera that Tahiti -possesses in common with Fiji to the exclusion of Hawaii. The lack of -difficulties connected with the dispersal of all these Tahitian genera -is worthy of note, because there are very few difficult genera amongst -the rest of the Tahitian flora. Excluding Lepinia tahitensis, which has -been already referred to, there are scarcely any “impossible” plants in -the Tahitian region; and even in this case, when the modes of dispersal -of Lepinia come to be investigated, it is likely that much of the -difficulty will disappear. Hawaii, as we have before seen, abounds with -perplexing questions of this nature. When dealing with the absentee -Tahitian genera, later on in this chapter, it will be shown that “size” -has played a prominent determining part in the exclusion of genera from -Tahiti, genera with seeds or “stones” exceeding half an inch or twelve -millimetres in dimension being, as a rule, unrepresented amongst the -truly indigenous plants. - -My further remarks on these Tahitian genera found in Fiji but not in -Hawaii will be limited to some general observations from the standpoint -of dispersal. I will first discuss some of those genera that possess -only peculiar species. They belong to an era of dispersal that, as far -as Tahiti is concerned, is passing or has passed away. Here we have the -species of each genus more or less localised in the various South -Pacific archipelagoes; but, as with Meryta, Alstonia, and Loranthus, it -is often apparent that, although the Tahitian region is mainly outside -the zone of present dispersal, the different groups of the Western -Pacific are kept in touch by the possession of species in common. This -testifies to the activity of dispersal in that region after it had -become suspended in Eastern Polynesia. The connection between the -isolated endemic species of Eastern Polynesia and a species ranging over -the Western Pacific can sometimes be shown, as in the case of Loranthus, -where a species confined to the Society Islands and to the Marquesas is -very closely related to L. insularum, a widely-ranging West Polynesian -species that reaches eastward as far as Rarotonga. - -We next have those genera like Grewia, Nelitris, Melastoma, Randia, -Geniostoma, Tabernæmontana, Fagræa, Bischoffia, Macaranga, and Ficus, -that possess in Polynesia one or more widely-ranging species. The agency -of the polymorphous species, which I have described in the preceding -chapter in connection with the general dispersal of Malayan plants over -the whole of Polynesia, is evidently also active when the work of -dispersal is restricted to the South Pacific. Its operation is to be -distinctly traced in all the genera above named except in Fagræa and -Ficus. Thus, in the genera Grewia, Melastoma, Randia, Geniostoma, and -Macaranga we find a single variable species ranging over the South -Pacific from Fiji to Tahiti, keeping all the groups in touch, but -associated in each, as a rule, with one or more peculiar species. A yet -earlier stage in the process is to be seen in those genera which, like -Nelitris, Tabernæmontana, and Bischoffia, possess only a solitary -species ranging over the South Pacific, varying in each group, but not -usually associated with endemic species. As with Melastoma, Macaranga, -and others, we can often trace the widely-ranging species of Polynesia -back to its home in Malaya, and with these and other genera the -connection between a species confined to a group and a variable species -ranging through all the archipelagoes of the South Pacific can sometimes -be detected in the affinity of their characters. - -It is thus seen that one of the principal determining causes of the -differentiation of species in Polynesia lies in the failure of the -dispersing agencies, a widely-ranging species becoming in consequence -gradually isolated in the various groups. With some genera, as with -Ophiorrhiza, it is possible to show that the resulting endemic species -pass into each other by intermediate forms. - -My further remarks on the Tahitian genera occurring in Fiji but not in -Hawaii will be devoted mainly to those with which I was most familiar -from the standpoint of dispersal. - -The Tiliaceous genus GREWIA offers a good example of those Polynesian -genera which possess in the South Pacific a single widely-ranging -species associated often with endemic species in the individual groups. -It is likely that a polymorphous form, including most of the Polynesian -species, could be here constituted. The fruits are dryish drupes, -becoming black and moist when over-ripe, and containing three or four -pyrenes suitable for distribution by birds and five or six millimetres -in size. - -The berries of NELITRIS, a genus of the Myrtaceæ, contain a few hard -seeds that are well fitted for dispersal by frugivorous birds. I am -inclined to follow Drake del Castillo, who considers that there is only -one varying species, N. vitiensis (Gray), which is distributed over the -whole of the South Pacific from the Solomon Islands to Tahiti. The -tendency of this widely-ranging species to vary in different groups is -indicated in the fact that some botanists have distinguished other -species within these limits. It is noteworthy that N. paniculata in -Indo-Malaya and N. vitiensis in the Pacific cover the whole range of the -genus. It would be interesting to establish a connection between them. - -MELASTOMA, an Old World genus of forty and more species, has one very -variable species, M. denticulatum, which, as defined by Bentham, has the -range of the genus from tropical Asia across the Pacific to Tahiti. This -plant is associated in some groups, as in Fiji, Tonga, and Samoa, with -other more or less localised species, and it affords a good example of -the principle of polymorphism in species-making. The berry-like fruits -contain an abundance of minute seeds, half a millimetre in size, which, -when rendered adhesive by adherent pulp, might readily stick to -feathers, or they might pass unharmed through the alimentary canal of a -bird. It is noteworthy that amongst the plants regarded by Prof. Penzig -as introduced by frugivorous birds into Krakatoa since the eruption is a -species of Melastoma. - -Few genera in these islands would better repay a careful study of their -species with regard both to the influence of station on specific -characters and to the question of “mutations” than OPHIORRHIZA. I found -the three Fijian species of this Rubiaceous genus so often in close -association, that I cannot doubt there is some connection between them. -Seemann and Gray, indeed, characterise two of them as confluent species. -The island of Tahiti alone possesses five peculiar species, and it is -evident that this island has been a centre of development for species of -Ophiorrhiza, just as Samoa has become the birthplace of many species of -the Urticaceous genus Elatostema. The minute angular seeds of -Ophiorrhiza would probably be transported in a bird’s feathers or in -adherent soil. As my experiments showed, they do not become adhesive -when wet. - -The genus LORANTHUS as distributed in the South Pacific has already been -briefly noticed. There is a species confined to the Tahitian region, and -there is another peculiar to Samoa, whilst one widely-ranging species, -L. insularum, that connects these regions together, reaching east to -Rarotonga, is closely related with the Tahitian species. There was no -doubt originally a single polymorphous plant that ranged over the -tropical South Pacific. With regard to the mode of dispersal of the -seeds of this genus of parasites, I should at once refer to the -systematic and careful observations made by Mr. F. W. Keeble in Ceylon -(_Trans. Linn. Soc._, v. 1895-1901). He formed the opinion that the -seeds of Loranthus usually reach their host without passing through the -alimentary canal of a bird, being merely wiped off its bill. This method -would never carry the seeds to Tahiti or even to Fiji; and since this -observer remarks that, although most of the seeds in the droppings were -completely rotten, some of them “possibly pass through the gut -uninjured,” we may accept this possibility as sufficient for the purpose -of dispersal in the Pacific Ocean. Mr. Keeble notes the observation in -Teil 3 of Engler’s _Die Natürlichen Pflanzenfamilien_ that the seeds may -germinate after passing through a bird’s intestine; and we may therefore -infer that whilst the method he describes is typical of local dispersal, -the other method is required in the instance of oceanic dispersal. - -ALSTONIA, an Apocynaceous genus of tropical Asia and Australia, yields -the caoutchouc of Fiji. Besides possessing in Fiji and Samoa peculiar -species, the islands of Western Polynesia have in A. plumosa a species -common to Fiji, Samoa, and New Caledonia. Another species, A. costata, -is restricted to Eastern Polynesia, occurring in the different islands -of the Tahitian Group as well as in Rarotonga. It is possible that the -Pacific species may be connected with A. scholaris, a species possessing -the range of the genus with the exception of Polynesia. The long -ciliated or hairy seeds, six to nine millimetres in length, are fitted -for transport by the winds and in birds’ plumage. The follicles dehisce -on the tree, and, according to Horne, the light seeds are distributed -locally by the wind. It is probable that the thick white juice oozing -from a broken branch would at times aid the adhesion of the seeds to a -bird’s feathers. - -GENIOSTOMA, a genus of the Loganiaceæ, is found in Malaya, Australia, -and New Zealand. It possesses in G. rupestre a species that ranges -across the South Pacific from New Caledonia to Tahiti, being associated -with one or more endemic species in most of the groups. The fruit is a -dehiscent capsule containing numerous small seeds imbedded in a -yellowish pulp; and from the standpoint of dispersal it may be placed in -the same category with Pittosporum and Gardenia (see pages 310, 313). - -The same principle involved in the occurrence of a species ranging the -South Pacific from New Caledonia to Tahiti, and associated with one or -more endemic species in most of the principal groups, is illustrated in -the Euphorbiaceous genus MACARANGA. It is specially noteworthy that M. -tanarius, which ranges from India to East Australia and the New -Hebrides, comes in touch in the group just named with M. harveyana, the -widely-ranging plant of the South Pacific above alluded to, and itself -an Asiatic species (see Burkill; _Bot. Chall. Exped._, iii. 191; _Index -Kewensis_). The connection between M. harveyana, the widely-ranging -species of the South Pacific, and the endemic species in the various -groups is indicated by its affinity with M. reineckei, a Samoan species. -The Macarangas in Fiji grow in a variety of situations, on the borders -of estuaries, in the mountain forests, and on the isolated mountain -peaks. It is to birds that we must look for the dispersal of the genus. -In the case of a species, apparently M. seemanni, common in the Rewa -delta, the seeds, which soon fall out of the cocci, are not infrequently -found in the drift of the estuary, but they sink in a week or two. Other -species examined showed no capacity for dispersal by currents. The fruit -of M. harveyana is provided with a few prickles, but since it breaks up -into the cocci, from which the seeds soon fall out, these appendages -could scarcely aid its dispersal. - -Like many other genera, TABERNÆMONTANA, an Apocynaceous genus -distributed through the tropics, is represented in Polynesia by a -widely-ranging species, T. orientalis, which extends from Malaya and -Eastern Australia through all the large groups of the South Pacific from -the New Hebrides to Tahiti, and is associated in Fiji with one or two -peculiar species, one of which, according to Mr. Burkill, is nearly -related to it. This genus therefore seems to illustrate the earliest -stage in the Pacific of that process by which a widely-ranging species -takes on a polymorphous habit and through its variations gives rise to -different species in various groups. Prof. Schimper ranks T. orientalis -amongst the Malayan strand-flora; but in Fiji the Tabernæmontanas are -only littoral where the soil is rich as in alluvial regions; and they -have no capacity for dispersal by currents that is worth speaking of, -the seeds in the case of T. orientalis and another species sinking after -drying for years, whilst the follicles soon open in water and go to the -bottom in a few days. The observations of Gaudichaud and Moseley -indicate that some Malayan species are dispersed locally by the currents -(_Bot. Chall. Exped._, iii, 279, 293); but the fruits of the genus are -evidently quite unfit for oceanic dispersal by this agency. We find in -the bird the agent that has carried the genus to the distant -island-groups of the Pacific; and from the standpoint of dispersal the -fruits may be placed with those of Pittosporum and Gardenia, being -follicular, and in the Fijian plants possessing seeds, 5 to 10 -millimetres in size, embedded in a pulp. - -FAGRÆA, an Asiatic and Malayan genus of the Loganiaceæ, is represented -in the Pacific by F. berteriana ranging through all the groups and -islands of the South Pacific from the Solomon Islands and New Caledonia -to Tahiti and the Marquesas, and by one or two other species in Fiji. It -is with Fagræa berteriana that we are entirely concerned. The tree is -often planted by the Pacific islanders near their villages; and since -they value its timber and use its large fragrant flowers for personal -decoration and for other purposes, it is probable that they have aided -in its dispersal. But, as shown below, it behaves in most localities as -an indigenous plant; and its berries are well fitted for promoting its -dispersal by frugivorous birds. - -I was familiar with Fagræa berteriana both in the Solomon Islands and in -Fiji; and in the last-named locality I especially studied it from the -standpoint of dispersal. All over the South Pacific, whether in the -Solomon Islands, in Fiji, in Rarotonga, or in Tahiti, this tree, though -thriving also in the lower levels, especially frequents rocky scantily -vegetated or open-wooded hill-tops and crests up to 2,000 or 2,500 feet -above the sea. In the rich alluvial soil of the Rewa delta in Fiji it -attains a height of 25 or 30 feet or more, whilst in the poor, dry soil -of the “talasinga” plains in this group it is much dwarfed, and often -does not exceed 10 feet, and may be only 6 feet high. It is in these -“talasinga,” or “sun-burnt,” plains of Fiji, especially in the Mbua -province of Vanua Levu, that the tree, although dwarfed, seems most at -home. Here it flowers and fruits abundantly whilst associated with -Acacia, Casuarina, and Pandanus trees, and it is in such dry localities -that this tree reflects in its choice of station the behaviour of -different species of the genus in the Malay Peninsula, where they grow -in open heath-country and sometimes on sandy heaths (Ridley in _Trans. -Linn. Soc. Bot._, iii, 1888-94). The fruits and seeds of F. berteriana -have little or no capacity for dispersal by currents. On the Fijian -plains the berries partially wither and rot on the tree. In the western -part of its area this tree almost comes in touch with the Asiatic -species, F. obovata, that ranges from India and Ceylon to the Malayan -region, a species that must be indebted to frugivorous birds for its -wide distribution. - -The Euphorbiaceous genus BISCHOFFIA seems to offer another example of -polymorphism in a wide-ranging species. Following Drake del Castillo, I -take the genus as including only a single species, B. Javanica, a tree -distributed over tropical Asia, Malaya, and Polynesia as far east as -Tahiti. The variable character of the species is indicated by the -different views held by the several botanists who have discussed the -South Pacific species. Whilst it is a common forest-tree in Indo-Malaya, -it affects in the Pacific islands the open-wooded districts of the lower -levels, and it is not uncommon on the dry “talasinga” plains of Fiji. -The fruits and seeds displayed in my experiments little or no capacity -for dispersal by currents; nor do these dryish berries, with seeds four -or five millimetres long, seem to be especially attractive for -fruit-eating birds; and it is likely that the same birds that distribute -Macaranga seeds also disperse those of this genus. The tree bears the -same name over the South Pacific, “koka” in Fiji and Rarotonga, and “oa” -in Samoa. Like many other Polynesian trees, it has its uses, but there -is no reason to believe that the natives have aided materially in its -dispersal. - -FICUS, a large genus comprising several hundred species, attains its -greatest development in tropical Asia and in Malaya. It is well -represented in the Western Pacific from the Solomon Islands to Fiji and -Samoa; but in Eastern Polynesia the species are very few, and the genus -is altogether absent from Hawaii, although a species has been found in -the North Pacific in Fanning Island, about 900 miles south of the -Hawaiian group (see page 377). - -The Polynesian species are for the most part restricted to the Pacific -islands, but there are only two species that range over the South -Pacific as far east as Tahiti, namely, Ficus prolixa, the Tahitian -banyan, and F. tinctoria. Some species are confined to Western -Polynesia, such as F. obliqua, the Fijian banyan, F. scabra, and F. -aspera, the last occurring in East Australia. Among the individual -groups Fiji possesses probably fourteen or fifteen species, of which, -perhaps, a third would be peculiar. According to Dr. Warburg, as cited -in Dr. Reinecke’s paper, Samoa owns eight species, of which six may be -endemic. In Rarotonga and Tahiti we find only F. prolixa and F. -tinctoria. The species in the groups where they are best represented -belong to three or four sections of the genus. - -The banyans of the South Pacific are represented by three or four -species, namely, Ficus prolixa, the Tahitian banyan, found all over the -tropical groups of the South Pacific from the New Hebrides and New -Caledonia to Tahiti, the Marquesas and Pitcairn Island (Maiden); F. -obliqua, the Fijian banyan, confined to the islands of the Western -Pacific from the New Hebrides to Tonga; and two new banyans in Samoa, as -described by Dr. Warburg in Dr. Reinecke’s paper. In my paper on -Polynesian plant-names it is shown that the banyans possess two names in -the Pacific, one being “aoa,” the Polynesian name, found in all the -groups from Samoa eastward, and connected linguistically with the -Malayan and Malagasy banyan-words; the other, the Melanesian name -typified in the Fijian “mbaka,” and represented in a variety of forms in -the New Hebrides and neighbouring groups. - -It is probable that the Pacific islanders have assisted in the dispersal -of one or two of the species of Ficus, such as F. tinctoria, which they -employ for different purposes, but, generally speaking, birds are active -agents in distributing the genus. I need scarcely say that the agency of -the currents is quite insufficient to explain the distribution of Ficus. -When in Fiji I experimented on three or four different species of Ficus -belonging to the sections of the genus there represented. The fruits may -float at first, but within a week or ten days they break down, and the -seeds escape and sink. Beneath a tree of F. scabra growing on the banks -of the Wai Tonga in Viti Levu, I noticed a number of its fruits floating -in a sodden condition among the reeds at the river-side. - -It is with the banyans that the dispersal of the seeds by frugivorous -birds becomes most evident. This is at once indicated by the frequent -occurrence of these trees in the interior of coral islets in the Western -Pacific, as in Fiji and in the Solomon Islands. Fruit-pigeons roost in -their branches, and birds shot on these islets often contain the fruits -in their crops (_Bot. Chall. Exped._, iv, 310). The process may also be -seen in operation in Krakatoa. Professor Penzig found in 1897 that three -species of Ficus had established themselves there since the eruption of -1883 through the agency of frugivorous birds. Besides pigeons, we find -that parrots, hornbills, honey-eaters, &c., feed on these fruits, and I -possess a large number of references to this subject. The Messrs. Layard -in New Caledonia, Dr. Meyer in Celebes, Mr. Everett in Borneo, Dr. -Forbes in Sumatra, and several other contributors to _Ibis_ might be -here mentioned. Dr. Beccari, in his _Wanderings in the Great Forests of -Borneo_, speaks of “the facile dissemination of the various species of -Ficus through the agency of birds,” and he arrives at certain important -conclusions which are discussed in Chapter XXXIII. - -I have before alluded to the absence of Ficus from Hawaii. This group -possesses the Honey-Eaters (Meliphagidæ), birds well suited for -dispersing species of Ficus over Polynesia; but this family of birds is -only represented by peculiar genera in Hawaii, and therein lies the -explanation. At the time when the Honey-Eaters roamed over Polynesia, -the genus Ficus had not arrived from Malaya. The connection between the -bird and the plant is well shown on Fernando Noronha, which possesses a -peculiar species of Ficus and a peculiar species of dove, the only -fruit-eating bird in the island (Ridley). - - - THE ABSENTEES FROM TAHITI - -Generally speaking, all the “difficult” genera which puzzle the student -of plant-dispersal in Fiji and Hawaii are absent from the Tahitian -region. Those with stone-fruits and with large seeds, where the stone or -seed is an inch in size and over, are absent from Tahiti. Thus the -genera Canarium, Dracontomelon, Myristica, Sterculia, and others, of -which the three first-named are known to be dispersed by fruit-pigeons, -have not advanced into the Pacific eastward of the Fijian region. We -miss in the Tahitian islands the large-fruited palms of Fiji, such as -the Veitchias with fruits two to two and a half inches (5 to 6 cm.) -long, and we find in their place a Ptychosperma, evidently very rare, -and the widely spread Pritchardia pacifica, that may have been -introduced by man, both with drupes not far exceeding half an inch (1·2 -cm.) in size. The islands of the Tahitian region also lack the Coniferæ; -and genera like Dammara, Dacrydium, and Podocarpus that give such a -character to the Fijian forests are not to be found. In this region we -do not find many of the large-seeded Leguminous genera, such as -Cynometra, Storckiella, and Afzelia, that occur in Fiji, the only -large-seeded genera that it possesses being such as are brought by the -currents, namely, Mucuna, Strongylodon, Cæsalpinia. The difficulties -presented by the occurrence of the inland species of Canavalia and -Mezoneuron in Hawaii do not offer themselves in Tahiti (see Chapter XV). -Tahiti also lacks, as often before observed, the mangroves and most of -the plants of the mangrove-formation. - -As above remarked, the Fijian trees with large “stones” and heavy seeds -an inch in size are not to be reckoned amongst the indigenous Tahitian -plants, “size” being an important determining factor in the exclusion. -The occurrence of Elæocarpus in Rarotonga presents no real difficulty, -as I have explained in Chapter XXVI. An apparent exception is presented -by the existence in Tahiti of Calophyllum spectabile, where the stones -are about an inch across; but since its fruits can float in sea-water -for nearly a month, and on account of the value placed on its timber by -the Polynesians, we cannot altogether exclude the agencies of man and -the currents. One seeming exception is also offered by the presence of -Serianthes myriadenia, a tree which in Fiji grows both in the forests -and on the banks of the tidal estuaries. Its seeds, which are six to -seven-tenths of an inch (15 to 18 mm.) in length, have no buoyancy, and -the pods float only two or three weeks. The case of Lepinia tahitensis -is alluded to elsewhere, but it may be added that these and other -difficulties await further investigation. - -A great many Fijian plants are not found in the Tahitian region, such as -Micromelum, those of the order Meliaceæ, the Melastomaceous genus -Medinilla, Myrmecodia, Ophiorrhiza, &c., which are often quite as well -fitted for over-sea transport as are several of the plants already -established there. But it should be remembered that crowding out would -often come into play in such a contracted region. The area, however, has -been very generously dealt with as regards plant genera. Though the -total land-surface cannot be more than one-fourth or one-third that of -Fiji or Hawaii, it possesses more than half the number of genera found -in Fiji, and four-fifths of the number found in Hawaii. - - - FIJI - -_The Fijian Genera not found in either the Tahitian or Hawaiian Regions_ - -We have already in some degree dealt with Fiji in so far as the partial -dispersal of genera over the Pacific islands is concerned. We have seen -that it possesses very few genera (not a score in all) in common with -Hawaii that are not found in the Tahitian region, and it is assumed that -in most cases such genera reached Hawaii independently and not through -the South Pacific. On the other hand, excluding the grasses, sedges, and -vascular cryptogams, Fiji owns in common with Tahiti between sixty and -seventy genera that do not occur in Hawaii. This shows unmistakably the -trend of plant migration in the Pacific islands. Several interesting -features in plant-distribution have been already brought out, and -notably the fact that Indo-Malayan genera with large seeds or “stones” -an inch in size have been arrested in the Fijian region in their passage -into the South Pacific. Thus Canarium, Dracontomelon, Myristica, and -Sterculia have not extended eastward of the Fijian area. - -Yet a very large proportion of the Fijian genera, quite half of the -total number, are not represented either in the Tahitian or in the -Hawaiian region; and of many of them it is obvious that they are as well -fitted to be carried over the Pacific as are those that have actually -reached Tahiti and Hawaii. Take, for instance, Begonia, which has not -extended east of Fiji, though Hillebrandia, a genus of the order, is -peculiar to Hawaii. Nor can we explain why with three genera like -Geissois, Dolicholobium, and Alstonia, possessing seeds dispersed by the -winds, only the last-named has passed beyond Fiji. However, as before -remarked, it is probable that lack of opportunity rather than capacity -for dispersal has determined the matter, and we must, therefore, assume -that many of the genera have halted in the Fijian region because they -entered the Pacific after the age of active general dispersal over that -ocean. - -Occasionally we notice in this region that which we have observed in the -case of Cyrtandra in different Pacific groups, namely, a sudden -development of what Hillebrand terms “formative energy” in a genus, such -as we find in the case of Elatostema in Samoa, and in that of Psychotria -in Fiji and Samoa. The principle of polymorphism in the development of -species is also illustrated by Micromelum and by Limnanthemum. In the -last case we possess a typical polymorphous species in Limnanthemum -indicum that has played in this respect the _rôle_ of Naias marina in -the warm waters of the globe. - -With several genera that like Gnetum, Myristica, and Sterculia occur -both in the Old and the New World, it is evident that in explaining -their distribution we are dealing with something more than questions of -means of dispersal. With these genera, and with others like Lindenia, it -seems almost futile to talk of means of dispersal, when to all -appearance their existing distribution is but the remnant of an age of -general dispersion over the greater part of the warm regions of the -world. These genera, with others, might be cited in favour of the -continental hypothesis relating to the islands of the Western Pacific. -Trees with stone-fruits, such as Canarium, Couthovia, Dracontomelon, and -Veitchia, where the stones are an inch and more in length, might be also -adduced by some in evidence of this theory. But in these cases the -lesson of Elæocarpus (Chapter XXVI) should always be remembered, since -the “stones” of drupes may vary greatly in size amongst the different -species of a genus, and species seemingly “impossible” from the -standpoint of dispersal in one group may be represented in other groups -by species where the size of the “stone” presents no difficulty in -attributing the dispersal of the genus to frugivorous birds. - - - _Sterculia_ - -The problem connected with the presence of this genus in Fiji is but a -part of the still more difficult problem connected with the dispersal of -the genus over the tropics. The riddle presented by the Fijian species -seems, indeed, difficult enough; but it merely presents in miniature the -great mystery surrounding the whole genus. According to the _Index -Kewensis_ no other species have been found in oceanic islands except -those occurring in the Western Pacific, as in Fiji, the New Hebrides, -and New Caledonia, and most of these seem to be confined to those -islands. We have here a genus that repeats the Dammara difficulty of the -Western Pacific. - -The trees are common in places in the Vanua Levu forests, where the -large, woody, open follicles may be seen lying in numbers on the ground, -empty and in all stages of decay. The seeds of one species, near -Sterculia vitiensis, were nearly an inch long and sank like stones. The -unopened follicles will float for weeks; but it is evident that Nature -does not disperse the genus in this fashion, since the fruits before -dehiscence remain on the tree. It is also noteworthy that Gaudichaud, -when describing the floating drift of the Molucca seas, refers to the -open follicles of two or three species of Sterculia (_Bot. Chall. -Exped._, iii, 279). The fruits never came under my notice in the drift -of Fiji. The seeds of a Fijian species examined by me were four-fifths -of an inch (2 cm.) long. They had a thin, brittle, outer skin and -crustaceous inner test, and, being edible, might attract birds; but such -birds would be ground feeders, like the Megapod, and the Goura pigeon of -New Guinea, and the Nicobar pigeon, birds of this habit being rare in -Fiji. I should doubt whether the seeds are sufficiently protected to be -preserved from injury in a bird’s stomach during a long sea-passage; and -they may thus be placed in the same category with the seeds of -Myristica, a genus that has also failed to reach Tahiti and Hawaii. - -But the distribution of Sterculia raises other more important questions -than that connected with its occurrence in Fiji, which involves an -over-sea passage of only 500 or 600 miles. As in Podocarpus amongst the -Coniferæ, which has a similar distribution in the Western Pacific, we -have to explain the existence of the genus in the three great -continental masses of Africa, Asia, and America, now separated by oceans -several thousands of miles across. Here also we must look far back into -the ages for a common centre of diffusion in the extreme north, such as -is in a sense suggested by the occurrence of the order in the Eocene -beds of Europe. - -As showing unmistakably that Fiji received its species from the Old -World, it may be observed that one of its trees, Sterculia vitiensis, is -very closely allied to S. fœtida, widely spread in tropical Asia, in -Malaya, and Australia, as well as in Africa. - - - _Trichospermum_ (Sterculiaceæ) - -There are only two species of this tree recorded in the _Index -Kewensis_, one in Java, and one in Fiji as well as in Samoa. The fruit -is a capsule with small, flat seeds, margined by long hairs, that might -possibly attach themselves to a bird’s feathers. - - - _Micromelum_ (Rutaceæ) - -This small genus of tropical Asia, Malaya, tropical Australia and the -islands of the Western Pacific, has one species, Micromelum pubescens, -possessing the range of the genus with other species that are restricted -to different localities. We thus have apparently another illustration of -the part played by a wide-ranging polymorphous plant in providing new -species. The red berries would easily attract frugivorous birds; but the -seed-tests seem too delicate to allow the seeds to remain more than a -few hours in a bird’s stomach without injury. - - - _Cananga odorata_ (Anonaceæ) - -This tree, which is cultivated in many places in tropical Asia and -Malaya, but is certainly indigenous, according to the authors of the -_Flora Indica_, in Ava and Tenasserim, has apparently extended into the -Pacific by cultivation. But though much valued by the natives on account -of its fragrant flowers, and in consequence often planted by them near -their villages, it grows in some localities in Fiji and Samoa as an -indigenous plant. The berries are especially suited for dispersal by -frugivorous birds, their flat seeds, 8 mm. in length, possessing hard -crustaceous tests that would enable them to pass unharmed in a bird’s -droppings. According to Reinecke the fruits are sought after by pigeons, -and particularly by Didunculus strigirostris, the Samoan Tooth-Billed -Pigeon. The tree has not travelled eastward of Tonga and Samoa, with the -exception of its occurrence in Rarotonga; and according to Mr. Cheeseman -the Rarotongans received it from Samoa several years ago. - - - _Geissois_ (Saxifragaceæ) - -This genus of seven or eight known species is found in Australia, New -Caledonia, the New Hebrides, and Fiji. Since New Caledonia possesses -four species, it may be considered the home of the genus. To the Fijian -endemic species, G. ternata, I paid special attention. The capsules -dehisce on the tree and allow the small seeds to escape. These seeds, -which are very light, 150 to 200 going to a grain, are 3 to 4 mm. long -and are winged at one end. They could no doubt be carried some distance -by strong winds; but they possess no buoyancy. Large bats probably aid -in their dispersal. The Fijians assert that these animals are in the -habit of visiting the trees for the sake of the honey furnished by the -conspicuous red flowers. When they see a bat flying towards these trees, -they are wont to remark that it is going to drink the “se ni vota,” that -is, to suck the flowers of the Vota tree. It is very likely that seeds -would sometimes be carried in their fur for considerable distances. - - - _Begonia_ - -Before the discovery of Hillebrandia, a new genus of the Begoniaceæ, in -Hawaii, the order was not known from Polynesia. However, in 1878 Mr. -Horne collected a species of Begonia in Fiji, and it was probably this -species that frequently came under my notice in the rain-forests of the -Vanua Levu mountains. In 1883 I collected a Begonia in the Solomon -Islands, which I gave to Baron F. von Mueller, who informed me that it -was the first record of the genus east of New Guinea, the description of -Mr. Horne’s Fijian plant apparently not having been published (see -Guppy’s _Solomon Islands_, p. 288). It is not easy to explain why a -genus with such minute seeds, which are apparently as well fitted for -dispersal as those of the orchids, should have such a limited -distribution in the Pacific. - - - _Dolicholobium_ (Rubiaceæ) - -In the _Index Kewensis_ this genus, containing five species, is -restricted to Fiji. It must, however, be more generally distributed in -the Western Pacific, since the genus was identified at Kew among my -Solomon Island collections, and it is recorded in the list given in my -book on that group (pages 283, 288, 297). - -The showy, large, white, fragrant flowers of these small trees recall -those of Lindenia, with which Dolicholobium is often associated in Fiji -by the sides of streams and rivers. As Horne observes, the Fijian -Dolicholobiums range from the sea-shores and the heads of the estuaries -to the tops of the highest mountains. As noticed by me in the Solomon -Islands they affected the same station, being especially common on the -banks of streams. The genus has a long, narrow capsule six inches or -more in length. The linear seeds, though very light, are an inch or more -long, the coats being drawn out into a long tail at either end, and thus -differing greatly from those of Lindenia, the other Rubiaceous genus, -with which these plants are so frequently associated at the river-side. -I can only suppose that the seeds are transported by the winds. The -history of the genus is suggested in my remarks on Lindenia. - - - _Lindenia_ (Rubiaceæ) - -Respecting its distribution in the Pacific, this genus of showy -river-side shrubs takes the same place amongst the plants that Galaxias -takes among the fishes. It is full of mystery. Of the four species -known, two grow on the river-banks of Central America and two in similar -stations in the islands of the Western Pacific. Of the last-named both -occur in New Caledonia, one of them being endemic, whilst the other, -Lindenia vitiensis, is found also in Fiji and Samoa. Reinecke seemingly -records no Samoan species, but in the list of additions at the end of -his _Flora Vitiensis_, Seemann refers to the Fijian species as having -been found in Samoa by Dr. Graeffe. - -Lindenia vitiensis, as Horne aptly remarks, adorns the rocky banks of -many Fijian streams with its cream-coloured flowers, which impregnate -the air with their sweet odour. I found it in Vanua Levu, both at the -heads of the estuaries and beside the stream and the torrent in the -heart of the mountains. It was often associated with a species of -Dolicholobium, which it resembled strangely in its large, showy, scented -flowers and in the form of the leaf. Seemann says it is also accompanied -at the river-side in Viti Levu by Ficus bambusæfolia and Acalypha -rivularis. It is noteworthy that all the four plants here mentioned as -being associated river-side plants in Fiji possess the long, narrow -leaves of the willow type, a subject that is discussed in note 79. - -The capsules of Lindenia vitiensis contain numbers of small, angular -seeds about 1·5 mm. across, some 400 of them when well dried going to a -grain. The seeds float buoyantly by reason of their outer covering of -crisp, air-bearing, cellular tissue. When this outer covering is -stripped off, the minute nucleus, or seed proper, which is barely a -millimetre across and is but slightly protected, sinks at once. As the -seeds float on the surface of a stream they might readily get on the -plumage of an aquatic bird; but they have no special means of -attachment; though, if they dried on the feathers they might adhere to -some extent. That they could be carried in mud adhering to a bird across -an ocean’s breadth I think most unlikely; and it should be remembered in -this connection that only the dead or sickly seeds would be found at the -bottom of a stream. - -The most reasonable explanation of the extraordinary distribution of -Lindenia is that it was in a past age found over the tropical regions of -both America and the Old World, and that it has died out over the -greater part of its original area. To study the means of dispersal of -plants with such a distribution seems almost futile. I am inclined to -think that the limited range of Dolicholobium, so frequently its -station-companion in Fiji, may be similarly explained. - - - _Limnanthemum_ (Gentianaceæ) - -This interesting genus of aquatic plants is dispersed over the tropical -and temperate regions of the globe, but with the exception of Fiji and -the New Hebrides it is not found in oceanic groups, though it occurs in -large continental islands like New Caledonia and Cuba. About twenty -species are enumerated in the _Index Kewensis_, but it is stated in the -_Genera Plantarum_ that they can probably be reduced to ten, the -reduction being chiefly applicable to the tropical species, nearly all -of which are reducible to varieties of L. indicum, the temperate species -being often very distinct. It would thus appear that although dispersal -is still active in the tropics, it is in part suspended in the temperate -zone, and we seem to possess in L. indicum a typical polymorphous -species that has played the _rôle_ of Naias marina in the warm, fresh -waters of the globe (see page 368). - -Although some of the temperate species, like Limnanthemum nymphæoides in -Europe and Northern Asia, have a wide range, it is probable that this is -connected not so much with means of dispersal, as with its relation to -present and past drainage-areas. Rivers in the lapse of ages change -their courses and carry their aquatic floras with them, leaving, -however, a few of their plants around the springs and in the lakes which -serve still as centres of dispersal. Rivers may even exchange their -plants in flood-time in extensive level districts. Nor is the occurrence -of the genus in the Old and New Worlds in the northern hemisphere to be -connected with questions of dispersal across an ocean. Except in the -case of small-seeded plants, like Nasturtium and Lythrum, where the -dispersal could be carried on by water-fowl, the plant-species being -often identical on both sides of the Atlantic, it is probable that most -of the large-seeded river-side genera common to Europe and North -America, such as Iris and Acorus, had in past ages their home in the -extreme north, whence the plants spread as from a focus into the -continents of America and Eurasia. It is also to be doubted whether even -in the tropics there has been much over-sea dispersal of Limnanthemum -without the aid of man, and reasons will be given for the belief that -probably in Fiji, in the New Hebrides, and in New Caledonia the seeds of -the first plants were unintentionally introduced by the aborigines. - -Following Bentham we may regard the species of the Western Pacific -Islands as a form of the wide-ranging Limnanthemum indicum. These plants -in Fiji do not play the part in river-vegetation that they do in the -temperate regions, as for instance in the Upper Thames. They are not -common except in places, and seem to be chiefly confined to Viti Levu, -particularly to ponds in the Rewa delta, where their _rôle_ is that of -an Indian tank plant. In the Rewa delta they may be sometimes seen -thriving in brackish water having a density of 1·005. - -Looking at the mode of dispersal to which the Limnanthemums owe their -existence in the Western Pacific, we cannot disregard, especially in -Fiji, the possibility of the seeds having been unintentionally -transported by the natives when they carried in their migrations their -edible tubers, such as Colocasia antiquorum, Alocasia indica, and -Cyrtosperma edulis, that are cultivated in wet places. It is in the -ponds around which these plants grow that the Limnanthemums thrive. The -Chinese, with their peculiar methods of cultivation, are now carrying -with them strange water-plants over the warmer regions of the globe; and -it would be surprising if the Pacific islanders in their migrations did -not do the same. If such an introduction, however, took place, it must -have happened before the time of Captain Cook, when the plant was found -in New Caledonia. (It may be remarked in this connection that the seeds -of the genus will germinate after being kept dry for years. Seeds of the -British species which I had kept dry for two and a half years germinated -healthily when placed in water.) - -Some years ago I ascertained that the seeds of the British plants were -enabled, by means of their fringe of hairs, to attach themselves firmly -to the downy plumage of a bird’s breast. This could not happen with the -Fijian plant as the seeds are naked, and the same may be said of some -species described by Gray and Chapman as widely spread over the United -States. The seeds of the genus appear quite unsuited for safe transport -inside the body of a bird. The Fijians give the plants a variety of -names, nearly all of which are associated with the word for a duck, and -none of them bear an ancient impress. Thus we find such names as -“Ndambe-ndambe-ni-nga” and “Vothe-vothe-ni-nga,” meaning respectively -“the duck’s seat” and “the duck’s paddle.” - - - _Ceratophyllum demersum_ - -This wonderful aquatic has been dispersed over most of the globe; but I -will only mention its occurrence in oceanic islands, such as Fiji, -Samoa, the Bermudas, and the Azores, to indicate the necessity of -attributing its distribution in islands to birds. Several years ago I -made a careful study in England of the habits and mode of germination of -this plant, the results of which are given in _Science Gossip_ for -November, 1894; but reference can only be made here to such points as -bear on the occurrence of the plant in the Pacific islands. - -It is well known that in our English ponds and rivers the plant -propagates itself, as a rule, by budding; and that it is only in -unusually hot and dry summers, such as that of 1893, when many ponds -became very low and were excessively heated, that the fruits mature in -any quantity. My observations clearly showed that a higher temperature -is required for the completion of maturation than for the early stage of -the fruiting process and for the flowering. After a comparison of my -river and pond temperatures, I formed the conclusion that whilst in -water 12 to 18 inches deep this plant requires for a week or more an -average daily maximum water temperature of 70° F. to produce its -flowers, a warmth of 80° and over is necessary to mature its fruit, a -condition to be found in England only in shallow ponds, where the plants -may fruit abundantly, but not in rivers, where they flower and rarely -mature the fruit (see also for the thermometric conditions my paper in -_Proc. Roy. Phys. Soc. Edin._, xii, 296). Since a yet lower temperature -(an average maximum water temperature of 66° for a week or more) is -sufficient for germination, it follows that the thermal conditions of -our English climate will allow Ceratophyllum to germinate and to flower, -though but rarely to mature the fruit. - -Even in Fiji we can notice the distinction between the cooler river and -the superheated ponds and swamps of the Rewa delta as regards the -maturation of the fruit. In 1897 I found Ceratophyllum thriving in the -main channel of the Lower Rewa where the water was quite fresh; whilst -lower down where the water was often brackish its place was taken by -Ruppia maritima. In the main river, where the water unmixed with -sea-water rarely acquires a temperature of 80° F., the reading being -usually 78° to 79°, I never found the plants in fruit, and it is only in -the superheated shallow waters of the swamps and back-waters that they -mature their fruits. - -Since Ceratophyllum even in tropical climates would probably only mature -its fruits in the superheated waters of shallow ponds, tanks, and -ditches, it follows that its dispersal by birds is confined to warm -regions. In the cold waters of the Siberian lakes and rivers it would -never mature its seeds, and could only be propagated by budding. If it -existed in the head-springs of the sources of a river in these -latitudes, it would be distributed by means of its floating shoots and -fragments along the length of the river basin, and in the times of flood -it might pass in the lower plains from one river system to another. When -rivers changed their courses it would be left behind in the lakes and -ponds and springs, and would also be carried away to the new region. In -this manner it would in the course of ages be distributed over a -continent without the aid of seed, propagating itself in a vegetative -fashion. - -In the case of oceanic islands, however, we have to appeal to the seed. -Since the fruits sink in sea-water even after prolonged drying, and -since a few days’ immersion in sea-water, as I found, kills the floating -plant, we are driven to the agency of birds. The fruits, which without -appendages are a quarter of an inch (6 mm.) in length, are too large and -heavy to be carried in dry mud adhering to birds. The chances of their -becoming entangled in a bird’s feathers by means of their basal spines -and terminal style seem small, since they would be lying usually on the -mud under the water. They are quite fitted for safe transport in the -stomach and intestines of birds, such as is established in Chapter -XXXIII for Potamogeton and Sparganium in the case of ducks. As my -experiments show, drying for a period of three months does not injure -the germinating capacity of the seeds. - - - _Dracontomelon_ (Anacardiaceæ) - -This is a genus accredited in the _Index Kewensis_ with eight species, -of which three belong to Borneo, one to Sumatra, one to Java, one to the -Philippines, and two to Fiji, all the species being restricted in their -range. My observations were confined to D. vitiense, Engler (D. -sylvestre in Seemann’s work), the Tarawau of the Fijians, who regard it -as a tree that is planted by the dead in Naithombothombo, the place of -departed spirits, according to the legend given by Hazlewood in his -Fijian Dictionary. Its method of dissemination in the Fijian forests is, -however, far more prosaic. Pigs and fruit-pigeons assist in the -dispersal of the seeds in these islands. Pigs are often found in the -vicinity of a Tarawau tree; and evidently they much appreciate the -fallen fleshy fruits, which are about 1-1/3 inch (3·3 cm.) across and -inclose a large stone 7/8 inch (2·2 cm.) in diameter. The entire fruit -and the detached stone sink in sea-water, the last floating only a few -hours, even after drying for four years. Mr. Hemsley regards the genus -as probably dispersed by the currents, since a stone was found amongst -the floating drift collected by the _Challenger_ Expedition off the -coast of New Guinea. The stone, however, is described as seedless, which -may explain its buoyancy. It is, however, to the fruit-pigeon that we -must look for the dispersal of this genus. In the crop of one of these -birds shot in Fiji I found the entire fruit of a Tarawau tree. - - - _Canarium_ (Burseraceæ) - -This genus of trees, to which nearly a hundred species are referred in -the _Index Kewensis_, belongs mainly to tropical Asia and Malaya, a few -species occurring in tropical Africa, Madagascar, the Mascarene Islands, -and Polynesia. Its great home is in Malaya, to which two-thirds of the -species are confined; but its distribution in the oceanic islands of the -Indian and Pacific Oceans is especially interesting, Mauritius, Bourbon, -Fiji, Tonga, and Samoa (Horne) each possessing a species. - -The large drupes of the genus, as I found in Fiji, have no capacity for -dispersal by currents; and we are, therefore, compelled to appeal to the -agency of the frugivorous bird. Yet to a person unaccustomed to the ways -of fruit-pigeons the transportation across a broad tract of ocean of -large heavy “stones,” an inch and more in size, would seem impossible; -and even to a student of dispersal improbable. Unless, however, we -prefer to accept the Lemurian theory for the Indian Ocean and the theory -of a Melanesian continent for the Pacific we are compelled to appeal to -these birds; and it can scarcely be said that our appeal is without some -justification. Both in the Solomon Islands and in the Fijis I was -familiar with the dispersal of the stones of these trees by -fruit-pigeons; and Wallace, amongst other writers, observed the same -long ago in the Malayan Islands (_Malay Archipelago_). Stones obtained -from the crops of Fijian pigeons measured 1-2/10 × 1 inch (3 × 2·5 cm.). -In the Solomon Islands these birds stock the interior of the coral -islets with trees of the genus, and the ground below the trees is often -strewn with the disgorged stones (_Bot. Chall. Exped._, iv, 310; Guppy’s -_Solomon Islands_, p. 85). - -Although the difficulty concerned with the transport of the seeds across -a broad tract of ocean seems very great, it is quite possible that -further investigation will enable us to overcome this objection, just as -we have done in Chapter XXVI when explaining how the genus Elæocarpus -may have reached Hawaii. It is, indeed, not unlikely that, as with -Elæocarpus, the stones of the drupes may in some species be much smaller -and far more fitted for being carried in a bird’s body over several -hundred miles of ocean. - - - _Couthovia_ (Loganiaceæ) - -Reference is here made to this genus because its mode of dispersal is -known, and because I was familiar with it in Fiji. Seemann gives two -species for Fiji, C. corynocarpa and C. seemanni, and the few other -species known seem to be confined to the Western Pacific. Solereder -gives a third species, C. densiflora, for Kaiser-Wilhelmsland in New -Guinea (Engler’s _Pflanz. Fam._ teil 4, abth. 2); and a Solomon Island -species, nearly allied to, if not a variety of, the Fijian species, C. -seemanni, is referred to in the list of plants from that group given in -my book on those islands. I found C. corynocarpa not infrequently -growing on the banks of small rivers in the heart of Vanua Levu. Its -drupes, which float for a few days in sea-water, are, according to -Seemann, eaten by fruit-pigeons. The “stone” varies from 2 to 4 -centimetres (3/4 - 1-1/2 inch) in length; and from the standpoint of -dispersal the genus ranks with Canarium and Dracontomelon. Seemann -describes and figures this species, which was constituted by Gray, in -his _Flora Vitiensis_; but, apparently through an error, it is in the -_Index Kewensis_ accredited to Hawaii. Hillebrand makes no reference to -the genus in his book on the Hawaiian flora. - - - _Veitchia_ (Palmaceæ) - -This genus of palms is closely allied to Ptychosperma, a Malayan genus -also represented in Fiji. The _Index Kewensis_ names four species, one -New Hebridean, and three Fijian. The fruits of two of the last-named -species tested by me had no floating power. The seed is about an inch -long, and the genus would be likely to be spread by fruit-pigeons. From -the standpoint of dispersal the genus would be placed with Canarium and -Couthovia; but possibly its presence in the Pacific may be indicative of -an ancient Western Pacific continent. - - - _Hibbertia_ (Dilleniaceæ) - -This genus of some eighty known species is almost entirely Australian, -with the exception of a few species found in New Caledonia, Tasmania, -and apparently also in the Mascarene Islands. Horne was the first to -record a species from Fiji, where it grows commonly in the “talasinga” -plains on the lee sides of the islands, and also on the scantily -vegetated mountain summits. In Vanua Levu I often found these plants -growing on the rocky peaks of the highest mountains of the island, as on -Mbatini, 3,500 feet, and on Mariko, 2,900 feet. Their presence on these -isolated peaks can only be attributed to birds. The carpels contain one -or two seeds, which have a membranous aril; but in the plains the seeds -are usually destroyed by grubs. - - - _Myrmecodia_ and _Hydnophytum_ (Rubiaceæ) - -These two genera of epiphytes, distributed over Malaya and extending to -the islands of the Western Pacific, possess tuber-like stems, which are -extensively chambered by ants that find a home in the interior. They -were familiar to me in the Solomon Islands, where they frequently grow -on the mangroves and on other littoral trees. They do not form such a -feature in the shore vegetation of Fiji, and judging from the -observations of Dr. Seemann and myself they occur most often on the -wooded mountain-peaks. The berries of these plants would attract -frugivorous birds; and their pyrenes, which in a Fijian Myrmecodia I -found to be 4 millimetres long, appear quite suitable for dispersal -through this agency. It would seem that germination may occur in the -berry on the plant. A specimen of Myrmecodia in fruit, that had been -lying overlooked for a fortnight between newspapers during one of my -mountain journeys, displayed on examination the pyrenes in a germinating -condition, the process being subsequently completed. The reader will -find these interesting plants described and illustrated in the English -edition of Schimper’s work on _Plant-Geography_, pp. 149, 150. - - - _Myristica_ - -The Nutmeg trees, though principally at home in Indo-Malaya, are found -also in the warm regions of Africa and America, as well as in the -islands of the Western Pacific from the Solomon group eastward to Fiji, -Tonga, and Samoa. The Tongan and Samoan groups possess two species in -common, whilst Fiji seems to possess its own species, four or five in -number. - -The seeds of this genus have long been known to be dispersed by -fruit-pigeons. Mr. Moseley, in his _Notes of a Naturalist_, and in the -_Journal of the Linnean Society_ (vol. xv), tells us how at one time -these birds in their dissemination of the seeds in the Banda Islands -were active opponents of the policy of the Dutch Government in -preserving their monopoly of the cultivation of the nutmeg of commerce. -He found numbers of wild nutmegs in the crops of these birds in the -Admiralty Islands, some of which were partially digested and others -seemingly sound; and Mr. Hemsley includes the genus as amongst those -dispersed in the Western Pacific by birds (_Bot. Chall. Exped._, Introd. -46; iv, 229, 308). In my book on the Solomon Islands I refer to the -occurrence of these seeds in the crops of fruit-pigeons; and I found -that the seeds were similarly dispersed by these birds in Fiji. It is -likely that the absence of the genus from Eastern Polynesia is to be -partially connected with the insufficient protection of the seeds -against injury during such a long ocean passage in a bird’s body. - -Gaudichaud, as quoted by Hemsley, refers to the occurrence of the fruits -of three or four species of Myristica in the drift floating in the -Molucca Sea. When in the Solomon Islands I noticed that the unopened -fruits of a species floated in sea-water. In later years in Fiji I -tested this point, and found that whilst the fruits just before -dehiscing will float between three and seven days in sea-water, the -seeds sink. As I have pointed out in the chapter on Drift, rivers carry -down to the sea an abundance of seeds and fruits that can float a few -days but do not imply dispersal by currents. - -Although, as I have above remarked, the localised range of the genus in -Polynesia may be in part connected with the insufficient protection of -the seed, it is apparent that in the case of a genus found in Asia, -Africa, and America we are brought into contact with questions other -than those of means of dispersal. No one would pretend that Myristica -seeds could be carried by birds uninjured across the Pacific Ocean; and -to explain the present distribution of the genus we must recall cases of -a similar kind, such as Podocarpus, where the genus in past ages had a -home in the north, from which, as from a focus of dispersion, it -extended into the continents of the Old and the New World (see p. 302). - - - _Rhaphidophora_ (Araceæ) - -This genus of climbing aroids, which gives a character to the forests of -Indo-Malaya as well as to those of the Western Pacific, is represented -in the New Hebrides, Fiji, Tonga, and Rarotonga by a variety of the -widely spread R. pertusa that ranges over Indo-Malaya and Eastern -Australia. The ripe berries would readily attract birds; and the seeds, -4·5 millimetres long in the case of a Fijian plant, appear hard enough -to pass unharmed through a bird’s digestive canal. We seem here to have -evidence of a somewhat recent connection between Indo-Malaya and -Polynesia through the agency of frugivorous birds. That the genus has -been long established in Polynesia is, however, indicated by the -occurrence there of a species seemingly peculiar to Fiji. We are -disappointed that in Engler’s recent contribution to the _Pflanzenreich_ -(in his volume on the Araceæ-Pothoideæ) he has not been able to include -this genus in the field of his studies. - - - _Gnetum_ (Gnetaceæ) - -This Gymnospermous genus, which is found in the warm regions both of the -Old and the New World, is represented in Fiji by a Malayan species, -Gnetum gnemon, which exists also in the Solomon group with other species -of the genus (Guppy’s _Solomon Islands_, pp. 288, 301). I was familiar -with this species in both Fiji and the Solomon group; but in the -first-named locality it is seemingly restricted to the borders of -Wainunu Bay on the south side of Vanua Levu, where Dr. Harvey first -found it. It grows there abundantly in young wood. - -It seems almost idle to discuss the mode of dispersal of a genus that is -placed in a class apart with the African Welwitschia and the European -Ephedra, possessing with them a history of which we know nothing. Yet it -is ranked by Mr. Hemsley amongst those genera that are dispersed in -Polynesia by birds, and he produces better evidence in support of this -view than we possess for many other plants. Thus a fruit of a species of -Gnetum, perhaps G. gnemon, has been found in a New Guinea fruit-pigeon; -and the fruits of two species of the genus were found in the crops of -fruit-pigeons shot by Mr. Moseley in the Admiralty Islands (_Bot. Chall. -Exped._, Introd. 46; iv, 308). The red drupes of Gnetum gnemon of Fiji -would readily attract birds, and their nut-like stones, about 8 -millimetres long, are well suited for this mode of dispersal. My -experiments in Fiji show that neither the drupe nor the stone of this -species floats in sea-water; and it is probable that the fruits of this -genus referred to by Mr. Hemsley as having been picked up on the beach -in the Aru Islands possessed only a temporary buoyancy. - -This genus presents us with the same puzzling question put to us by -several Fijian genera, such as Myristica and Podocarpus, that occur in -both Asia and America; and until we answer that query it seems almost -futile to study modes of dispersal. - - - _Elatostema_ (Urticaceæ) - -This genus of annual and perennial herbs belongs to the tropical regions -of the Old World. It is represented in Samoa by fifteen known species -and by at least four or five in Fiji, whilst with the exception of a -solitary Tahitian species it is not recorded from East Polynesia. -Reference is here made to it particularly on account of its great -development in Samoa. We have here a genus that, like Psychotria in Fiji -and Cyrtandra in Fiji, Samoa, and Hawaii, runs riot in respect to the -production of species (see p. 317). Dr. Reinecke describes fifteen -Samoan species, of which, with the exception of two found in Malaya, all -seem to be described for the first time. So sensitive, he remarks, is -the genus to external conditions that station-forms abound; and he -points out that if we were to follow the dividing lines usually -recognised between species, we should account every station-form a new -species. It is, of course, obvious that the polymorphism of the Samoan -Elatostemas depends primarily not on the varying influence of station -but on their sensitiveness to external conditions. One might put the -question to the Samoan Elatostemas that Hillebrand put to the Hawaiian -Cyrtandras, and ask why nature in this particular genus in this -particular locality thus luxuriates in formative energy. Almost every -Pacific group in respect of some of its plants presents the problem so -well stated by Dr. Reinecke for this genus in Samoa. It is noteworthy -that Schimper, in his work on _Plant-Geography_ (English edition, pp. -291, 297, 299), especially singles out Elatostema and Cyrtandra as -growing socially in the tropical rain-forests of Java and of the Asiatic -mainland. - - - _Scirpodendron costatum_ (Cyperaceæ) - -As far as I can gather, this giant-sedge has not been previously -recorded from Fiji; but it is included in the Samoan flora, and has also -been found at Penang and Singapore, as well as in Borneo, Java, and -Queensland. In Samoa, as we learn from Reinecke, it grows both in the -coast swamps and on dry ground. In Fiji it is very common in the -mangrove-swamps at the mouths of rivers, especially in the Lower Rewa; -but in Vanua Levu it is also frequent in the marshy localities of inland -plateaux, 700 to 800 feet above the sea, as well as by the side of -streams in swampy districts on the lower hill slopes. This double -station in the salt-water swamp of the coast and in the fresh-water -marsh of the interior seems to be repeated in Java, where the plant was -first discovered by Zippelius on the banks of torrents in mountainous -regions and in swampy places. - -The genus comprises, according to the _Index Kewensis_, only this -species, though variations are to be observed in plants from different -localities. The species was described by Kurz in the _Journal of the -Asiatic Society of Bengal_ (vol. 38, 1869) and by Bentham in his _Flora -Australiensis_; and an illustration is given by Miquel in his -_Illustrations de la Flore de l’Archipel Indien_ (1871). The plant is so -common in Fiji that one can only suppose that its resemblance to a -stemless Pandanus, from which, as Kurz observes, it is with difficulty -distinguished except when in flower or fruit, led to its being -overlooked by both Seemann and Horne. Its leaves, from 9 to 12 feet in -length, are commonly used for making mats and for thatching, both in -Fiji and Samoa. The plant usually attains a height of 3 to 5 feet. - -The fruits occur abundantly in the floating and stranded river and sea -drift in Fiji, a circumstance that led to my discovery of the parent -plant in the swamps. The fruit, which is about half an inch (12 mm.) -long, consists of a hard, stony nut invested by a thick ribbed cork-like -covering, to which it owes its buoyancy, since the nut sinks. The -detached fruit is perforated at the base through both coverings, and -only a little soft tissue closes the aperture in the inner shell, the -protection against the entry of sea-water in the case of floating fruits -being quite inadequate. This explains also why the stranded fruits were -so frequently found by me germinating on the beach, where, as my -observations showed, they never established the plant. This early -germination would prove to be an advantage in the case of fruits -stranded in a suitable locality. - -But though the perforation in the fruit favours its early germination, -it lessens its ability to withstand a long sea-passage without injury to -the embryo. I found in different experiments on fruits of plants growing -in the mangrove swamps, that when placed in sea-water 40 per cent. sank -during the first fortnight, whilst 15 per cent. floated after five or -six weeks, but all were at the bottom in two months. On the other hand, -fruits from plants of the swamps of the inland plateaux displayed much -feebler floating power, in some cases sinking at once, in others -floating for a few days, and in others again floating for a week or two. -In this case the outer cork-like covering proved to have lost most of -its floating power. - -From the number of empty seed-vessels found, both in the floating and -stranded drift, it appeared evident that the seed had often rotted away -during the flotation. It is apparent from these observations and -experiments that Scirpodendron costatum is not suited for dispersal by -currents over wide tracts of ocean. The fruits might be able to float -unharmed for a few weeks, but they would be unable to accomplish much -more than the 500 or 600 miles intervening between Fiji and the nearest -groups to the west. - - - _Lemnaceæ_ - -This order, judging from the writings of Hegelmaier, Schenck, and -Hemsley, is represented by one or other of the common species, Lemna -minor, L. gibba, L. polyrrhiza, in various Atlantic islands, as in the -Bermudas, the Azores, Madeira, the Canary Islands, and St. Helena; but -doubts frequently arise as to their being truly indigenous. Lemna -trisulca is regarded by Hemsley as indigenous in the Bermudas. Lemna -minor has been introduced in recent years into Hawaii, where I observed -it flowering and sometimes fruiting abundantly in the heated waters of -the ponds. Two species found in other regions were recorded by Seemann -from Fiji, and I have come upon few other records of the occurrence of -the order in the tropical islands of the open Pacific. I am inclined to -the opinion, based not only on the facts of distribution, but also on -the results of numerous experiments on the means of dispersal, that this -order has in most cases reached oceanic islands with man’s assistance. - -Some years ago I made a systematic study of the habits of the British -Lemnæ, most of the results being published in the _Linnean Society’s -Journal_ (vols. xxix and xxx), as far as concerned Lemna minor, L. -gibba, and L. polyrrhiza. During this inquiry I ascertained that with -these species, as well as with L. trisulca, the chances of a bird’s -carrying their fronds uninjured in its plumage over a wide extent of -ocean were small. None of them survived twenty-four hours’ drying in -fine weather, whether in the sun or in the shade; but in rainy weather -they withstand an exposure of one or two days. It is, therefore, -unlikely, even if the fronds were entangled by their rootlets in a -bird’s feathers, that they would be able under ordinary conditions to -reproduce the plants after a day’s flight of some five hundred miles -across the sea. It must also be remembered that the drying capacity of -the air when a bird is in full flight in ordinary weather would be that -displayed during a gale of wind with a velocity of at least thirty to -forty miles an hour. For this reason I do not think with Kerner that -under usual conditions drops of water would be a factor of importance in -causing the adherence of minute seeds of any kind to birds’ plumage. -Where the seeds are not available, it is most probable that birds -disperse the duckweeds by their fronds over short distances, but not -across broad seas. This would certainly apply to temperate latitudes, -where these plants rarely seed. Thus with Lemna, as with Ceratophyllum, -it would seem that the dispersal of the seeds by birds takes place -normally only in warm latitudes. Those of the duckweeds could be -transported in adherent mud over land-areas. - -According to Hegelmaier, the two species of Lemna found in Fiji are L. -paucicostata, an Asiatic species, and a variety of an Australian -species, L. oligorrhiza, possessing dark root-sheaths. These plants -mostly came under my notice in the Rewa delta. They were rarely seen in -Vanua Levu, where in one locality I found the typical Lemna minor. The -first species is also Samoan. - -In 1897 and in 1899, in a pool near Notho in the Rewa delta, in Viti -Levu, Fiji, I found a great abundance of a species of Wolffia, specimens -of which were sent to Prof. Schimper with my mangrove collections, but -his death intervened, and I have not been able to follow up the matter. -On comparing the specimens with Hegelmaier’s descriptions and plates, it -would seem that the species is near W. arrhiza and W. brasiliensis, but -differs from both in the greater length of the fronds. As concerning the -means of dispersal of the genus, I may add that the fronds were killed -after being allowed to dry for eighteen hours. - - - _Marsilea_ (Marsileaceæ) - -A species of this genus, apparently near Marsilea villosa, was common in -the ditches and ponds around Notho, in the Rewa delta, Fiji, in 1897-99. -The genus is included by Horne in his list of Fijian plants; but is not -given by Seemann. The villous sporocarps, when dry, are very light and -readily catch in cloth and in feathers. Hillebrand includes in the -Hawaiian flora M. villosa and M. crenulata. The first-named, which was -collected by Chamisso and Gaudichaud, finds (he says on the authority of -Braun) its nearest relative in a species from Oregon and California. The -other has been collected in the Liukiu Islands, the Philippines, -Mauritius, and Bourbon. It is very probable that the occurrence of the -genus in oceanic islands is due to the agency of birds. - - - _Summary of the Chapter_ - -(1) We are here concerned with the more restricted distribution of -non-endemic tropical genera over the Pacific. The general trend eastward -of these genera is well brought out in the fact that whilst Fiji -possesses some sixty or seventy genera in common with Tahiti to the -exclusion of Hawaii, it does not possess a score in common with Hawaii -to the exclusion of Tahiti. The grasses and sedges and the mountain -genera are not here included; and we are comparing the flora of the -Hawaiian lowlands below 4,000 feet with the floras in mass of Fiji and -Tahiti. - -(2) Hawaii possesses very few genera (less than thirty) that are not -found either in Fiji or in Tahiti, or in both; and of these quite a -third are to be traced to America. - -(3) From two of these genera, Embelia, a land genus, and Naias, an -aquatic genus, we obtain two important indications, namely, that -specific differentiation has taken place to much the same extent in a -water plant as in a land plant, whether in a continent or in an island. -In other words, new species have been developed or are developing -independently of the immediate environment and of isolating influences. - -(4) The interchange of plants between the regions of Hawaii and Tahiti -to the exclusion of Fiji has been very slight. Probably not half a dozen -genera belong to this category. - -(5) Excluding plants brought by man and by the currents, Tahiti -possesses very few that present any difficulty from the standpoint of -dispersal, plants with seeds or “stones” an inch in size being, as a -rule, absent. - -(6) With the genera (60-70) common to Fiji and Tahiti, and distributed, -therefore, over the South Pacific, the wide-ranging highly variable -plant is an important factor in the development of peculiar species in -the different groups, just as it has been shown to be in the previous -chapter in the case of genera dispersed over the whole Pacific. The -_rôle_ of the polymorphous species has always been an important one in -this region. - -(7) In the case of several Fijian genera it seems almost futile to talk -of existing means of dispersal, since the present distribution of genera -like Sterculia and Gnetum, that occur on both sides of the Pacific, in -America and in Asia, is not to be thus explained. - -(8) On account of the large size of their seeds and “stones” it might be -argued that certain of the Fijian plants afford evidence of a previous -continental condition of the islands of the Western Pacific, since it is -not easy to understand how such large seeds and “stones” could have been -transported over broad seas by birds. It is, however, pointed out that -in these respects the species of a genus may vary greatly, and that the -seeds and stones may be large in some species and small in others. - -(9) The greater number of the genera that have entered the Pacific from -the Old World have not advanced eastward of the Fijian region, half of -the Fijian genera not occurring in the Hawaiian and Tahitian regions; -and the explanation of this is to be found not in any lack of capacities -for dispersal, but in a want of opportunities. The story of -plant-distribution in the Pacific is bound up with the successive stages -of decreasing activity in the dispersing agencies. The area of active -dispersion that at first comprised the whole of the tropical Pacific was -afterwards restricted to the South Pacific, and finally to the Western -Pacific only. The birds that in an early age carried seeds all over this -ocean became more and more restricted in their ranges, probably on -account of increasing diversity of climatic conditions. The plants of -necessity responded to the ever narrowing conditions of bird-life in -this ocean, and the differentiation of the plant and of the bird have -taken place together. - - - - - CHAPTER XXVIII - - THE POLYNESIAN AND HIS PLANTS - -Identity of the problems presented by the indigenous plants and the - peoples of the Pacific islands.—The food-plants of the Polynesians and - the pre-Polynesians.—Their weeds.—The aboriginal weeds.—The white - man’s weeds.—Weeds follow the cultivator but are distributed by - birds.—The general dispersion of weeds antedates the appearance of the - Polynesian in the Pacific.—Weeds of little value to the - ethnologist.—Aleurites moluccana.—Inocarpus edulis, Gyrocarpus - Jacquini, Serianthes myriadenia, Leucæna Forsteri, Mussænda frondosa, - Luffa insularum.—Summary. - - - MAN AND THE SEED - -MAN in his distribution in the islands of the Pacific reproduces in a -minor degree nearly all the difficulties presented there by plants, -birds, and other forms of animal life. Like the plant he entered the -ocean from the west; and as with the plants, so with the aborigines, -there was an era of general dispersion over this ocean, followed by an -age in which Polynesian man, ceasing to migrate, tended to settle down -in the several groups, there undergoing differentiation in various -respects, as in physical characters, in language, and in manners. Just -as we can now recognise the type of a plant, of a bird, or of an insect, -that belongs to a particular group of islands, so we can distinguish -between the Hawaiian, the Tahitian, and the Maori, whether in physical -characters, in his speech, or in his customs. Fiji possesses in the -Papuan element of its Melanesian population the earliest type of man in -the Pacific, just as it also possesses in the Coniferæ the most ancient -types of trees in this region. Divesting his mind of all previous -conceptions, the ethnologist, as I have remarked in my discussion of the -distribution of Freycinetia in Chapter XXV, might profitably study _de -novo_ the dispersion of man in the Pacific from the standpoint of -plant-dispersal. - -Man and the seed have battled their way over the Pacific apparently in -defiance of the prevailing winds and currents, and both have failed to -reach the New World. Man in the Pacific is almost as enigmatical as the -plant. As a denizen of this region he is by no means a recent -introduction; and though his food-plants are mainly Asiatic, they belong -to distinct ages in the history of man’s occupation of these islands. - -I venture to think that a great deal lies behind the Indo-Malayan mask -of the Polynesian, and that there is a story concerned with his origin -that has yet to be told. We have by no means solved the riddle when by -following the evidence we assign to him a home in Asia. It is only then -that the real difficulties begin. It required many centuries of European -civilisation for the discovery of America; but the voyages of Columbus -sink into insignificance when we reflect on what had been dared and -accomplished by uncivilised man when he first landed on the shores of -Hawaii and Tahiti. - -The problem of man in the Pacific bristles with difficulties differing -in degree but not in kind from those relating to the flora. Whenever a -particular theory seems on the point of being well established, some -disturbing question arises, and as with the plant, we are never able to -push our facts quite home. Since I first visited the Solomon Islands, -now twenty-four years ago, the Pacific islander and his flora have -deeply interested me. The history of man and of the plant cannot be -separated in the Pacific; and the same determining principles of -distribution have affected both. - - - THE FOOD-PLANTS OF THE POLYNESIANS AND PRE-POLYNESIANS - -One can imperfectly distinguish two sets of food-plants in this region; -the first comprising such plants as Pachyrrhizus trilobus, Tacca -pinnatifida, Amorphophallus campanulatus, the Mountain Bananas, the Wild -Yams, and several others that grow wild, and, as a rule, only serve as -food in times of scarcity; the second including the plants that are -extensively cultivated by the present islanders, such as the Breadfruit, -the Banana (Musa paradisiaca), the Taro (Colocasia antiquorum) and the -two Yams (Dioscorea alata and D. sativa), &c. Those of the first set -probably formed the food of the earliest inhabitants of the Pacific -Islands, pre-Polynesian peoples that practised only a rude sort of -cultivation, as with the present “bush-men” of the islands of the -Western Pacific. Those of the second set belong to the later occupants -of these islands, the Polynesians. - -(a) _The Pre-Polynesian food-plants._—In addition to those above named -one may mention Cycas circinalis, Cyrtosperma edulis, Lablab vulgaris, -Pandanus odoratissimus, Saccharum officinarum, Sagus vitiensis, &c. -Inocarpus edulis is probably to be here included, and amongst the Wild -Yams should be named Dioscorea nummularia and D. pentaphylla. Some of -them are now occasionally cultivated; but most of them only occur in the -wild condition, either as weeds or as larger plants growing -spontaneously in uncultivated localities. Even the knowledge of them as -food-plants has sometimes been altogether lost, the present inhabitants -of the Fijis, for instance, knowing nothing of Lablab vulgaris and Sagus -vitiensis as sources of food. The question of the antiquity of the -Coco-nut Palm in Polynesia was discussed at length by Seemann; but for -various reasons we cannot be absolutely certain whether or not it is an -older denizen of the Pacific islands than the Polynesian. It is, -however, to be inferred that it came originally from the home of the -genus in America, perhaps as a gift brought by the Equatorial Current -from the New World to Asia. Several chapters might be devoted to the -discussion of the earlier food-plants of these islanders; but here only -a brief reference can be made to a few of them. - -Perhaps the oldest of the earliest aboriginal food-plants are those -that, like Cyrtosperma edulis and Sagus vitiensis, are apparently -confined to Fiji. Here we seem to possess indications of the development -of new species since that group was first occupied by man. Others, like -Pachyrrhizus trilobus and Cycas circinalis, that are restricted to the -groups of the Western Pacific may come next in relative antiquity. - -Although most of the early food-plants hail from the Old World, the home -of Pachyrrhizus is in America. One may indeed wonder how a plant with -such a history ever reached the Western Pacific. It seems to be -generally distributed in this part of the ocean, having been recorded -from New Caledonia, the New Hebrides, Fiji, Tonga, and Samoa. Although -its edible roots are only used in times of scarcity, the plant grows -wild all over Fiji, being especially frequent in the “talasinga” plains. -Though I searched diligently, it never presented me with its seed. In -Tonga, according to Graeffe (as quoted by Reinecke) the plant is much -employed in preparing the land for yam-cultivation, since it restrains -the growth of weeds and keeps the soil moist. - -Amongst the food-plants of this early period that are distributed over -the South Pacific as far east as Tahiti may be mentioned the Wild Yams -(D. nummularia and D. pentaphylla), the Mountain Bananas, Tacca -pinnatifida, Amorphophallus campanulatus, and others. Of these Tacca -pinnatifida and Dioscorea pentaphylla are alone found in Hawaii. I will -only now refer to the Mountain Bananas. - -The Mountain Bananas of the tropical South Pacific, distinguished by -their erect fruit bunches and their seeded fruits, present us with one -of the mysteries connected with aboriginal man in this ocean. Whether in -New Caledonia, Fiji, Samoa, Rarotonga, or in Tahiti, they grow wild in -the interior, and form often a conspicuous feature of the vegetation in -the mountains and at the heads of the valleys. They are occasionally -cultivated. Their Fijian and Samoan names of “Soanga” and “Soa’a” -reproduce the names of the banana, “Saguing” and “Saing” in the Tagalog -language of the Philippine Islands. The Tahitian appellation is “Fehi” -or “Fei,” and this reappears in Samoa in the form of “Fa’i,” the word -for the common cultivated banana, Musa paradisiaca. The Rarotongan name -of “Uatu,” as given by Cheeseman, is suggestive of the Micronesian form -(Ut, Uut, &c., in the Carolines) of a widely spread banana word in -Malaya, Melanesia, and West Polynesia (Fudi, Vundi, Undi, &c., &c.). It -is not unlikely that all these South Pacific mountain bananas with erect -inflorescences and seeded fruits belong to one species, variously -designated by botanists as Musa fehi, M. uranoscopus, M. troglodytarum, -&c., and confined to this region. Under the name of Musa fehi Schumann -includes the New Caledonian and Tahitian plants, and he views the Samoan -plant as probably identical with them. This botanist, in his monograph -on the Musaceæ (Engler’s _Das Pflanzenreich_, 1900), establishes the -home of the bananas in tropical Asia, and considers that their -occurrence in America before the time of Columbus has not been proved. -Birds have no doubt often assisted in the dispersal of the wild, seeded -plants; but it is likely that man is responsible for the occurrence of -the mountain forms in the Pacific, and probably their fruits formed when -cooked one of the principal articles of diet of the earliest immigrants. -(There evidently exists in Vanua Levu a plant very like the African Musa -Ensete. Its presence was only indicated by the occurrence of its empty -seeds in the stranded beach-drift, and reference is made to it in that -connection in Chapter XXIX.) - -(b) _The Polynesian food-plants._—The cultivation of the yams, the -taros, the breadfruits, and the bananas in later ages all over the -Pacific islands cannot here be dealt with. My readers will already know -that a very ancient cultivation is in each case indicated by the -occurrence of a great number of varieties. Much has been written upon -this matter, and amongst the recent contributions to the subject may be -reckoned Mr. Cheeseman’s interesting paper published in the -_Transactions and Proceedings of the New Zealand Institute_ (vol. -xxxiii). - -I may here mention in connection with the Sweet Potato (Batatas edulis), -a plant that may have an American origin, though much mystery surrounds -its home, that it rarely seeds in Fiji except when it is grown in poor, -sandy soil, and in dry, rocky situations. The Fijians were quite -incredulous as to its maturing seed; but after much searching I found a -solitary plant in seed and removed their doubts. - - - THE POLYNESIAN WEEDS - -Some curious questions are raised in connection with the weeds of this -region. Polynesia, says Dr. Seemann, presents a most interesting problem -with regard to its weeds. It is, however, necessary to point out that -these plants arrange themselves into two groups, the aboriginal weeds -comprising those existing in the islands at the time of Captain Cook’s -expeditions in the latter half of the eighteenth century, and the white -man’s weeds that have been since introduced. - -As concerning the Fijian Islands, Dr. Seemann remarked that although the -majority of the non-endemic plants of the flora is Asiatic, “the bulk of -the weeds is of American origin, or, at all events, is now found in -America.” His principal point was to show that American weeds displayed -a greater disposition than Asiatic weeds to spread in Fiji, because Fiji -was to American plants altogether virgin ground. This is a purely -botanical matter, and we are not in a position to oppose a conclusion -formed by such a careful observer of plant life. But to the ethnologist -it is a very different matter whether most of the Fijian weeds are of -American origin or merely now exist in America. His interest lies -entirely in the aboriginal weeds. To the student of plant-dispersal this -distinction is also a very important one; and his interest again is all -on the side of the aboriginal weeds. - -Dr. Seemann enumerates 64 Fijian weeds, of which at least 37 were in the -Pacific islands when Captain Cook’s botanists made their collections -(see Note 82). Of these 22 occur in continental regions on both sides of -the Pacific, 13 are natives of the Old World alone, and two only are -seemingly American exclusively, namely, Waltheria americana and Teucrium -inflatum. The first is claimed to be American because most species of -the genus are American, but it is now widely distributed in the Old -World as well as in America. The second, though widely distributed in -tropical America, has strangely enough only been found in the islands of -the Western Pacific. - -The important point is thus brought out that although in Captain Cook’s -time the food-plants cultivated by the Polynesians, such as the banana, -the breadfruit, the taro, and the yams, were almost exclusively Asiatic -in origin and bore Malayan names, a large proportion of the weeds were -not exclusively Asiatic, but occurred in America as well as in the Old -World. The inference to be drawn, however, is not, as Dr. Seemann -implies, that the Polynesians derived several of their weeds from -America (since with few exceptions all the aboriginal weeds named in -Note 82 occur in the Old World, and in more than a third of the plants -in the Old World only), but that many so-called cosmopolitan weeds were -distributed very much as they are now, when the Polynesians brought -their food plants from Indo-Malaya into the Pacific. - -Weeds follow the cultivator in all climates; and it is very natural -that, as Mr. Hemsley points out, plants which seem to owe their wide -dispersal to cultivation are not found in Australia (_Bot. Chall. -Exped._, iii, 142). The Australian native as a rule cultivates nothing. -Yet I fancy that man’s share in weed dispersal is as often as not merely -restricted to producing the conditions favourable to the growth of -weeds, and that the seeds are often brought by birds and other agencies. -Many weeds of the genera Atriplex, Polygonum, and Ranunculus are -dispersed by partridges in England, and I have often found the uninjured -fruits of the plants in the stomachs of these birds. Many weeds, like -Prunella vulgaris, Plantago major, Capsella bursa-pastoris, Luzula -campestris, and several others named in Note 43, possess seeds or fruits -that become “sticky” when wet, and would readily adhere to a bird’s -plumage. - -We can also say of tropical weeds that many of them are distributed by -birds. In the crop of a dove in Hawaii I found a number of the small dry -fruits of Waltheria americana, the widely spread tropical weed before -mentioned, and of another weed of the order Compositæ. On the bare rocky -peak of one of the Vanua Levu mountains the only plants found growing -were Oxalis corniculata and a species of Peperomia, both of them -evidently growing from seed dropped by birds. The fruits of Urena lobata -and of species of Sida, as well as those of Bidens pilosa and Ageratum -conyzoides, could be readily dispersed, entangled by their appendages in -the plumage of birds, whilst the sticky achenes of Adenostemma viscosum -would easily adhere to feathers. Weeds with drupes or berries like -Geophila reniformis and Solanum oleraceum would attract frugivorous -birds, and I have often seen berries of the last-named pecked by birds. -Man has doubtless often been the agent in dispersing the seeds of -Leguminous weeds like Lablab vulgaris. On the other hand, we know from -the observations of Focke (see page 150) that birds can distribute the -seeds of a plant like Vicia faba; and in the Pacific islands it is -evident from the frequent occurrence of Tephrosia piscatoria on bare -rocky hill tops that its seeds are dispersed through the same agency. -Birds also probably carry about the seeds of Cardiospermum halicacabum. - -If we based our conclusion solely on the distribution of weeds without a -previous study of their means of dispersal we might, as students of the -distribution of man, acquire some startling and very erroneous notions -on the history of the races of man, especially in the New World. Lacking -such an acquaintance with existing modes of dispersal it would not be -prudent to attach too much importance to the occurrence of Asiatic weeds -in America and of American weeds in Asia. Mr. Hemsley, in his work on -the botany of Central America (_Biologia Centrali-Americana_), gives a -list of ten British plants of world-wide range, which we will designate -plants of waste places rather than weeds. They are plants often found -not only in the Old and New Worlds, but also in the southern hemisphere, -and I will here name them: Radiola millegrana, Alchemilla vulgaris, -Cotyledon umbilicus, Lythrum salicaria, Convolvulus sepium, Sibthorpia -europæa, Prunella vulgaris, Lycopus europæus, Aira cæspitosa, Luzula -campestris. According to this authority these plants “are most unlikely -to have been aided, intentionally or unintentionally, by man” and -“possess no special means of dispersion by animals or birds or the -elements” in the way, as is implied, of appendages like hooks, hairs, a -pappus, &c. - -Five of these plants are referred to in various connections already in -this work. In all I have tested the means of dispersal of six or seven -of them; and although my results are not always conclusive, I venture -here to indicate some of them. The nutlets of Prunella vulgaris and the -seeds of Luzula campestris emit mucus when wetted and adhere firmly to -feathers on drying, whilst the nutlets of Lycopus europæus are sticky in -the dry state and adhere to the fingers on handling. This last-named -plant is occasionally to be noticed on rubbish heaps growing with other -waste-plants. No such adhesive qualities, whether in the wet or dry -condition, came under my notice with Alchemilla arvensis or with Lythrum -salicaria. With Alchemilla the seed-like fruits fall from the plant, -inclosed in the dried-up calyx. The seeds of Cotyledon umbilicus are so -minute (1/3 mm. or 1/75 inch) that they can be compared with Juncus -seeds from the standpoint of dispersal. They are naturally a little -sticky and tend to adhere to feathers, but more probably they are -transported in adherent soil. The case of Convolvulus sepium is a very -remarkable one, and I have referred to it on page 29 and in the notes -there indicated. The species of Radiola, Sibthorpia, and Aira have not -been tested by me. Dispersion, however, would be favoured by the small -size of the seeds in the first two species and by the awned glumes in -the case of Aira. - -The distribution of aboriginal weeds might be expected by some to supply -data of profound interest to the student of the races of mankind; and I -think the botanist rarely realises how often he tantalises the -ethnologist by the remark that certain weeds have been spread by -cultivation all round the tropics. De Candolle many years ago, in his -_Géographie Botanique_, gave a list of nearly 100 plants, made up of Old -World species naturalised in America and of American species naturalised -in the Old World, and quite half of them were classed as plants -distributed in one way or another through man’s agency. Now this is -either a subject of supreme importance or it is of no interest to the -student of man’s history. If it should prove that birds have done most -of this dispersal, then the story of the aboriginal weed would be of -little interest in connection with the races of man in the New World. - -I will now refer briefly from the standpoint of dispersal to a few -interesting Polynesian plants in which man has been in most cases more -or less concerned in their distribution. - - - ALEURITES MOLUCCANA (THE CANDLE-NUT TREE) - -Much interest is attached to this tree, which is found in India, Malaya, -and North-east Australia, and occurs all over the Pacific, extending -north to Hawaii, south to the Kermadec Islands, and east to Tahiti and -Pitcairn Island (Maiden). In the Hawaiian Islands it is often so -frequent as to form whole forests, or at all events to give a character -to the forest zone up to 2,000 feet above the sea. Its prevalence in -Hawaii might be regarded as evidence of its indigenous character; but -its predominance there is due to the circumstance that it is one of the -few forest-trees that the cattle and other animals avoid, most other -trees falling victim to their depredations by the loss of the bark. In -Fiji, though frequent in places, it does not form such a conspicuous -feature in the vegetation as in Hawaii. In Samoa it is abundant in the -coast-bush. In Rarotonga it forms with Hibiscus tiliaceus, as we learn -from Cheeseman, the major portion of the lower forests, a circumstance -which seems to indicate, since both these trees were probably introduced -by the natives, that this island like Hawaii has lost or is losing many -of its original forest-trees. In Tahiti, according to Nadeaud, it is -common from the sea-level up to 3,000 feet above the sea. - -As a Polynesian tree, Aleurites moluccana presents itself to me as an -intruder which has often taken the place of trees of the primeval -forests of these islands. That the natives usually employ the oily seeds -for illuminating purposes is well known; and its prevailing name of -Tuitui (Kukui in Hawaii) is derived from the Polynesian custom of -threading the seeds before using them for lighting purposes. One of the -Fijian names, “Sikethi,” is suggestive of “Saketa,” a name for the tree -in the Ternate dialect of the Indian Archipelago. To the modes of -dispersal of this tree, I have devoted much attention. - -The more or less empty seeds of this tree are to be commonly found -floating in rivers and stranded on beaches. I have found them in numbers -on the beaches of Fiji and Hawaii in the Pacific, and of the south coast -of Java and of Keeling Atoll in the Indian Ocean. In all I have examined -many hundreds of these seeds, whether stranded on the beaches in the -localities above named, or floating in the Fijian rivers and at sea -amongst the islands of that archipelago. The seeds were always either -empty or contained a kernel in an advanced stage of decay. A sound seed -has no floating power under any condition; and sound seeds are only to -be found in beach drift near the mouths of estuaries, where they have -been freed by the decay of the fruits brought down by the rivers. During -some dredging operations in the harbour at Honolulu several years ago, -quantities of old Aleurites fruits and seeds were brought up. It is only -by means of the floating fruit that the sound seed can be carried any -distance by the currents; but even in this case the opportunities of -wide dispersal are very limited. If one places in sea-water a number of -well-dried fruits, most of them will sink within a week, and all will be -at the bottom in a fortnight. - -The seeds stranded on a beach are often found cracked. This I think -arises from long exposure to the scorching rays of the sun. On account -of the hardness of the shell it is very difficult to obtain the kernel -entire; but the Mangaians get over this difficulty, as we learn from the -Rev. Wyatt Gill, by slightly baking the seeds; whilst the Fijians, -according to Mr. Horne, effect the same object by throwing the heated -seeds into cold water. On one occasion I placed an empty seed on a tin -plate kept at a temperature 115° to 120° F., a temperature near that -which the seed would acquire when lying exposed to the sun on a tropical -beach. After five days I found it had reproduced the cracks noticed in -another empty seed from the Keeling beach. - -Facts are not wanting with regard to the dispersal of the seeds by -birds; but since the kernel alone is sought for by birds, and as there -is no means of cracking the shell in their stomachs, such an agency is -only available for local distribution. The Messrs. Layard inform us that -in New Caledonia a small crow (Physocorax moneduloides) and a parrot -(Nymphicus cornutus) are very partial to these seeds (_Ibis_, 1882). -They were told that the crow cracked them by carrying them to a -considerable height and letting them fall on a stone. We are not told -how the parrot cracks the seed, which has a shell so hard that the -Malays, I may remark, term the seed “bua kras,” or “the hard seed,” -whilst a hammer is required to break it. However, since Indian parrots, -according to Mr. J. Scott, are able to split open with their beaks the -hard beans of Adenanthera pavonina (_More Letters of Charles Darwin_, -ii, 349), they evidently possess ingenuity in seed-cracking. - -My general conclusion with reference to this tree in Polynesia is that -it could not have been distributed, except locally, by birds and -currents; and that it owes its dispersion there principally to man. A -contrary indication seems to be offered by the occurrence of the tree in -the uninhabited Kermadec group; but since Cordyline terminalis also -exists there, a cultivated plant widely dispersed by the Polynesians, it -would appear that these islanders have formerly visited the group. It is -also contended by Canon Walsh that the Cordyline of the Maoris was -introduced into New Zealand by that race. (See Cheeseman in vols. xx and -xxxiii, _Trans. N.Z. Inst._, for papers on the Kermadec flora and on the -food-plants of the Polynesians.) - - - INOCARPUS EDULIS (THE TAHITIAN CHESTNUT) - -Like Aleurites moluccana this tree presents a _primâ facie_ case for -dispersal by currents. As the result of inquiries in this direction I -have formed the opinion, however, that it has been mainly distributed by -man. Though occurring in all the South Pacific groups, as far east as -Tahiti and the Marquesas, it does not occur in Hawaii. With its home in -Malaya it possesses a range closely resembling that of the breadfruit -tree; and yet, although its fruits are often a common article of food in -Polynesia, it requires no cultivation, and reproduces itself so -abundantly in favourable situations that, as Dr. Seemann observes, only -the dense shade of the parents checks the occupation by the seedlings of -all the adjacent ground. It possesses in the Pacific two sets of names, -neither of which I have been able to identify with any Malayan names, -and both occur over much of the region. Thus the Fijian “Ivi” and the -Samoan and Tongan “Ifi” are represented by “Ii” in Rarotonga, “Ihi” in -Tahiti and the Marquesas, “Hi” in Ualan in the Carolines, “Ifi” in -Futuna in the New Hebrides, and “If” in a New Guinea dialect. Then we -have the Tahitian “Mape,” the “Marap” of Ponape in the Carolines, and -the “Mamape” of Fate in the New Hebrides, besides other forms found in -Melanesia. - -In the South Pacific islands, as in Fiji, Samoa, Rarotonga, and in the -Tahitian group, it flourishes in low, moist localities at and near the -coast, by the side of streams and estuaries, and in the rich soil of the -lower valleys. In the Rewa delta in Fiji it is especially abundant, -often bordering the creeks in the mangrove swamps, and occupying -stations that are under water when the river is in flood. It may extend -inland in the various groups, but it is in the low-lying, moist, coast -regions that it mostly thrives; and in Fiji it presented itself to me as -essentially a tree of the estuaries, a station strongly suggestive of -dispersal by currents. Schimper, it may be remarked, includes it amongst -the shore vegetation of the Indian Archipelago. - -When in Fiji I paid especial attention to the dispersal by currents of -these large fruits, the agency of birds being, of course, negatived by -their size. They are to be commonly observed floating in the rivers when -in flood, as well as at sea between the islands, and stranded on the -beaches. Of those found afloat in the Rewa River not more than a fourth -had a sound seed. Of those stranded on the beaches two-fifths were -empty, two-fifths displayed a rotten seed, and one-fifth had sound -seeds. Of those picked up at sea all were empty. These fruits, unlike -many others in the drift of the Fijian rivers, do not germinate afloat. -They soon lose in the water their outer, fleshy, non-buoyant coat; -whilst the inner fibrous coat, to which the floating power of the fruit -is due, the seed having no buoyancy, is not water-tight, and moisture -soon enters and leads to the decay of the seed. In order to test their -floating power, I placed in sea-water ten mature fruits. Five of them -floated after forty-five days, having then lost most of the outer, -fleshy coat. Two were afloat after sixty days, but their seeds were -rotting. One fruit that sank after five weeks had a sound seed. Most of -them were sown out afterwards in a place where the trees were thriving, -but none germinated, and of two or three examined all had a decaying -seed. The empty fruits may float a long time after the decay of the -seed. Forty days would probably be the extreme limit for the flotation -in sea-water of a fruit with a seemingly sound seed, though a very small -proportion would reach this limit, and I much doubt whether such a fruit -would germinate afterwards. - -I, therefore, inferred that currents are only available for the local -dispersion of the fruits of Inocarpus edulis. It is to man that the tree -owes its existence in Tahiti and other groups of the open Pacific; and -it is to be concluded that the occurrence of this tree on Christmas -Island in the Indian Ocean marks an early Malayan occupation of the -island. - - - GYROCARPUS JACQUINI - -The cosmopolitan distribution of this seemingly useless tree, growing, -as Hemsley remarks, in maritime districts throughout the tropics, in -America, Australia, Asia, and Africa, presents one of the puzzles of -plant-distribution. It is by no means universally spread in the Pacific -islands, and I find reference to it only in Fiji and Tahiti. Seemann -says that in Fiji it is common on the beaches of Taviuni and other -islands. I found it to be a rare coast tree on Vanua Levu. It does not -seem to have been recorded by the botanists of the 18th century in the -Pacific. It, however, has evidently been long established there. Nadeaud -does not speak of its littoral station in Tahiti, and says that it grows -best in the regions of the interior up to elevations of 2,000 feet, -where it attains a great size; and its abundance is implied by his -remark that he had to fell many trees to collect the fruits. - -The singular fruit, which has two long wings and looks like a -shuttlecock, dries up on the tree; and in course of time it is detached -and falls to the ground. The falling fruit in its descent twists round -like a screw, and hence the Fijians call the tree the Wiri-Wiri tree, -the same name in the form of Wili-Wili being given for a similar reason -to Erythrina monosperma in Hawaii. Schimper (p. 157) truly remarks that -the fruits are too heavy to be carried by the wind across a wide extent -of sea; and I ascertained by experiment that in an ordinary trade-breeze -they would only be carried a few paces. Birds are quite out of the -question as agents of their transport to oceanic islands. We are driven -then either to the agency of man or to that of the current. The trees -grow rapidly and the timber is soft and perishable. The fruits are not -edible, and as far as I could ascertain the tree is of little or no -value to the Pacific Islander, there being at all events no reason to -believe that he has distributed it. - -We appeal lastly to the currents, the agency which Mr. Bentham selected -on _a priori_ grounds (Presidential address, _Linnean Society_, 1869). -My experiments in Fiji showed that the fruits, when dried on the tree -and afterwards detached, are able to float over long distances in -sea-water. After two months they were still afloat, the seeds inside -being dry and unharmed. The fruit’s buoyancy was tested in different -conditions, either without the wings, or with both wings, or with but -one wing, and it was found that the wings, which float for only a day or -two by themselves, lessen the buoyancy of the fruit. Of fruits with both -wings attached forty per cent. floated after two months, whilst of those -deprived of the wings all floated after two months. In the ordinary -course of flotation the wings in most cases break off during the first -few weeks, and in the rough-and-tumble of current-transport this would -occur sooner, so that the floating power of most of the fruits would not -be much affected. The cause of the buoyancy in a structural sense -belongs to the Convolvulaceous type. The kernel has no buoyancy, but it -incompletely fills the cavity of the seed-vessel, the coats of which are -quite waterproof, but have no independent floating power. - -It is thus evident that like many other shore-trees Gyrocarpus Jacquini -is distributed by the currents. It is not unlikely that its present -sporadic occurrence in the Pacific islands may be due to the gradual -extinction of the tree in this region, either on account of some insect -pest introduced since Cook’s time or from the use of the timber for -fire-wood by the aborigines. - - - SERIANTHES MYRIADENIA - -This is a striking looking Acacia-like tree that might have been fitly -discussed in the chapter on the enigmas of the Leguminosæ. Only four or -five species are named in the _Index Kewensis_, of which one occurs in -Malacca and in the Philippines, a second in New Caledonia, a third in -Fiji, and the fourth, S. myriadenia, over the South Pacific groups of -Fiji, Tonga, and Tahiti. Reinecke does not include the genus in the -Samoan flora; and it is merely assigned to that group by Seemann on the -authority of Mr. Pritchard, the British Consul in Fiji. Though common in -the forests of the larger islands of Fiji, S. myriadenia is most at home -on the banks of the estuaries, usually behind the mangrove belt, but not -beyond tidal influence. The peculiar species, S. vitiensis, I found on -the banks of the estuary of the Mbua River in Vanua Levu, the locality -from which Gray described it. According to the French botanists, S. -myriadenia, in Tahiti, ranges from near the sea to an elevation of 800 -metres. The Fijian name of the trees is “Vaivai,” the name also of -Leucæna Forsteri, and of some other introduced trees of the Acacia -habit. The Tahitians apply the same name in the form of “Faifai” to S. -myriadenia. - -The Fijians value the trees on account of the wood; but unless the -Polynesians were in the habit of transporting the seeds of their -numerous timber trees, which is most unlikely, it seems at first sight -useless to look to man’s agency for an explanation of the wide dispersal -of a tree like S. myriadenia in the South Pacific. The tough, woody, -indehiscent pods, from 3-1/2 to 4 inches long, floated in my sea-water -experiments in the case of both S. myriadenia and S. vitiensis between -seven and twenty-five days, after drying for some months. The seeds, -about two-thirds of an inch (17 mm.) in length, are only freed by the -decay of the fallen pod, and have no buoyancy. The agency of birds is -evidently excluded; and it is, therefore, to the currents that we must -make our final appeal; but their powers of dispersing the species appear -quite insufficient to explain the occurrence of these trees in Tahiti. -Perhaps, as in the case of Calophyllum spectabile, another Polynesian -timber-tree found in Tahiti (see p. 136), man and the currents have -worked together. - - - LEUCÆNA FORSTERI - -This bush of the Mimoseæ frequents maritime sands in the South Pacific, -and is confined to this region. It has been found in New Caledonia, -Fiji, Tonga, Rarotonga, and Tahiti. The seeds sink and the pods dehisce -on the plant, so that the agency of currents, unless we invoke the -intervention of the drifting log, bearing the seeds in its crevices, -seems to be excluded. Sea-birds might carry the seeds unharmed in their -stomachs, but there is no evidence bearing on birds as agents in the -dispersal of the species. Since the plant has not been recorded from -localities outside the Pacific islands, and since it was collected by -Cook’s botanists in Tonga and Tahiti, it cannot be placed amongst plants -of recent introduction. Although growing on maritime sands in Fiji, -Rarotonga, and Tahiti, it may grow inland, and according to Cheeseman is -particularly abundant in Rarotonga. In Fiji it is apt to occupy -newly-formed alluvial land at the mouth of the rivers, as in the case of -the Rewa; but the “how and why” caused me much fruitless speculation, -and I abandoned the plant in despair. The Fijians sometimes give it the -native name of Serianthes myriadenia, which they then term “Vaivai ni -Viti,” or the Fijian Vaivai. In Tahiti it is named “Toroire,” and in -Tonga “Toromiro.” - - - MUSSÆNDA FRONDOSA - -Mussænda frondosa is the only one of the sixty species of this tropical -Asiatic and African genus that extends into Polynesia. This beautiful -shrub, which is easily recognised by its conspicuous white, leaf-like -calyx lobe, is common everywhere in Fiji, decorating, as Horne fitly -remarks, in the contrast presented by its golden flowers, its large -white calyx leaf, and its green foliage, many an acre of waste, grassy -land, where the orange-coloured doves and the red and the green parrots -flit to and fro. With its home in India, China, and Malaya, it ranges -all over the South Pacific, from the Solomon Islands to Tahiti. Its -berries contain an abundance of small, minutely-pitted seeds, 0·7 mm. or -1/35 of an inch in size, and weighing when well dried about 600 to the -grain. The seeds retain after years of drying the property of clinging -to passing objects by means of a few microscopic, thread-like fibres, -that are attached to their surfaces. In this manner they will fasten -themselves to the point of a knife, and the observer is astonished to -see them dangling in the air from a pin’s point. I suppose that this is -connected with some hygroscopic quality. At all events, it would enable -these light seeds to be carried about not only by birds and bats but -also by insects. It is possible that man has aided in the dispersal of -this interesting plant; but birds, bats, and insects have, I think, -mainly done the work. - - - LUFFA INSULARUM - -This is regarded as a maritime form of Luffa cylindrica, a plant -commonly cultivated throughout the tropics. The South Pacific plant, -which occurs also in Australia and Malaya, has been found in New -Caledonia, Fiji, Tonga, Rarotonga, and Tahiti. In Fiji it grows chiefly -on the “talasinga” plains and in places once under cultivation. I -noticed it in one locality climbing over the branches of an Inocarpus -tree on the banks of the Rewa. In Rarotonga it is common in the lower -regions. It is, according to Nadeaud, fairly frequent on the shore and -in the lower valleys of Tahiti, where it was collected by Banks and -Solander, the companions of Cook. The Pacific islanders, as far as can -be gathered, make little or no use of the plant; and unless it was -introduced accidentally with their cultivated plants, they could -scarcely have been concerned in its dispersal. - -In Fiji I made a special point of investigating the mode of dispersal of -this plant. The fruits, which ultimately become dry and fibrous, are to -be seen hanging vertically from the plant as it climbs among the -branches of a tree. The apical disk usually falls off, and many of the -seeds drop out through the hole thus produced; but a few remain -entangled in the fibrous material occupying the interior of the fruit. I -have noticed such fruits floating down the stream of the Rewa River; but -my experiments showed that they do not float more than a week, whether -in fresh or salt water. The seeds, however, possess a hard, impervious -shell, and are well adapted to withstand unharmed prolonged immersion in -the sea. They will evidently float for months. Out of one hundred -selected seeds placed in sea-water, sixty were found afloat and sound -after two months. The cause of the seed-buoyancy is purely mechanical. -Neither the shell nor the kernel has any floating power, the buoyancy -arising, as with Convolvulaceous seeds, from the unfilled space in the -seed-cavity. When in Fiji, I tested the seeds of the ordinary cultivated -tropical form of the plant which had been introduced into a garden from -Australia. They all sank in a few days, and on being cut across the seed -displayed but little unoccupied space in its cavity. I have no doubt -that the Pacific form of this plant has been at times dispersed by the -currents, not, however, through the fruits, but through the seeds. It is -also quite possible that it may have been introduced by a pre-Polynesian -people into the Pacific. - - - _Summary of the Chapter_ - -(1) Man in his distribution over the Pacific islands reproduces, but in -a less degree, nearly all the difficulties presented by the plant in its -dispersal. In both we have the age of general dispersion followed by a -suspension more or less complete of the migrating movements; and in both -we have differentiation associated with the isolation. - -(2) The Pacific islanders possess two sets of food-plants. In addition -to those commonly cultivated in our own time, such as the yam, the taro, -the banana, &c., there are a number of food-plants now growing wild, but -rarely cultivated, and only used when the others fail. These plants, -which include the wild yams, the mountain bananas, Tacca pinnatifida, -Pandanus odoratissimus, and several others, are regarded as older than -the Polynesians in the Pacific, and as having probably formed the food -of a pre-Polynesian race that practised only a rude sort of cultivation. - -(3) The weeds of Polynesia also fall into two groups. In the first place -there are the aboriginal weeds, of which those found in this region by -Captain Cook’s botanists in the latter part of the 18th century are -taken as examples. These include species of Urena and Sida, besides -Waltheria americana, Oxalis corniculata, Bidens pilosa, and many other -weeds. In the second place, there are the numerous weeds that are known -to have been introduced by the white man since the voyages of the -English and French navigators of Captain Cook’s time. - -(4) There is reason to believe that many weeds now cosmopolitan in the -tropics had obtained their present distribution in America and in the -Old World before the Polynesians entered the Pacific. It is thus that we -can explain how there existed in these islands at the time of their -discovery by Cook, Bougainville, and other navigators of that period, a -number of weeds that have their homes in America. - -(5) It is not considered that the distribution of aboriginal weeds can -materially aid the ethnologist in his study of the early history of man, -since birds are regarded as the chief distributors of their seeds and -fruits. Whilst man has prepared the conditions for the growth of weeds, -the bird has usually brought the seeds. - -(6) Amongst interesting plants concerned with man in the Pacific are -Aleurites moluccana and Inocarpus edulis, which are regarded as in the -main distributed through man’s agency. Gyrocarpus Jacquini is viewed as -a tree originally widely dispersed by the currents in the Pacific, but -now becoming extinct. - - - - - CHAPTER XXIX - - BEACH AND RIVER DRIFT - -In the south of England.—On the coast of Scandinavia.—In the - Mediterranean.—Southern Chile.—Very little effective dispersal by - currents in temperate latitudes.—Cakile maritima.—In tropical - regions.—River drift.—River and beach drift of Fiji.—Musa Ensete.—The - coco-nut.—River and beach drift of Hawaii.—Comparison of the beach - drift of the Old and New Worlds.—Summary. - - - THE BEACH DRIFT OF TEMPERATE LATITUDES - -DISPERSAL by currents seems to be mainly restricted to warm latitudes. -Whilst in the tropics seed-drift is abundant on the beaches, in the -cooler regions of the globe it is usually very scanty and often masked -by other vegetable _débris_. - -Let us take, for instance, a beach in the south of England. We can find -by careful searching amongst the stranded drift the seeds and -seed-vessels of various littoral plants of the buoyant group, such as -Arenaria (Honckeneya) peploides, Cakile maritima, Crithmum maritimum, -Convolvulus soldanella, Euphorbia paralias, &c., and such sundries as -bits of stems of Salsola kali bearing fruits; but their amount is -scanty; and they are often difficult to find on account of the great -amount of rubbish with which they are associated, such as empty stones -of cherries, plums, and peaches; empty seeds of grapes; hazel-nuts, -beech-nuts, chestnuts, acorns, all either empty or with decaying seed; -the spiral pods of Medicago; besides quantities of leaves, sticks, and -bark. Although the occasional shell of a Spirula, or the horny skeleton -of a Velella, or a genuine pumice pebble (see Note 76), may tell us of -long wanderings in mid-ocean, we find little that is not English or -derived from neighbouring coasts on a beach in the south of England. I -have examined numerous beaches on the coasts of Devon and Cornwall, and -have never come upon any indubitable tropical seed-drift. - -On one occasion I examined many of the beaches between Ilfracombe and -Padstow with the object of finding tropical seeds, but to no purpose. -Portions of bark, generally 2 to 4 inches across and much water-worn, -together with a quantity of steamer-slag or cinders, often largely -composed the stranded drift. No doubt this bark is stripped off by the -waves from floating trees, which are generally stranded in a bare -condition after a long ocean voyage. This is the case with the timber -brought in the Oregon drift to Hawaii; and Sernander (p. 117) remarks -that bark seldom occurs on the trees washed ashore with the Atlantic -drift on the coasts of Scandinavia. Modern marine deposits ought to -contain much bark _débris_. - -On the beaches in the vicinity of estuaries we find a certain amount of -river drift, and amongst it fruits or seeds of Sparganium ramosum, Iris -pseudacorus, Alnus glutinosa, Rumex, and many other river-side plants, -such as I have mentioned in my paper on the Thames drift (_Journ. Linn. -Soc. Bot._, xxix). Most of them are capable of reproducing the plant, -but not on the sandy beach where the waves have stranded them; and we -thus see here one of the limits of the efficacy of currents as -seed-dispersers. - -From the labours of Lindman and Norman, the results of which are summed -up by Sernander (p. 116) we can learn what are the components of the -“Atlantic Drift” on the Scandinavian coast; and a strange assortment we -here find, in which it is difficult to detect much indication of -effective seed dispersal. Besides the seeds of Cæsalpinia Bonducella, -Entada scandens, and Mucuna urens, familiar to us as occurring in the -drift of tropical beaches, there is a quantity of vegetable drift -hailing sometimes from North America, sometimes from the Canary Islands, -and sometimes from the West Indies, mingled with much local drift in -which the larch and steamer-slag or cinders predominate. The seed-drift -derived from the proper beach-plants of the coast plays a subordinate -part, though it is stated by Norman and others that seeds and -seed-vessels, as the case may be, of Arenaria peploides, Cakile -maritima, Convolvulus soldanella, and Lathyrus maritimus, with those of -other plants, are also to be found. - -The Mediterranean beach drift, as illustrated by the results of my -examination of numerous beaches in Sicily as well as in the islands of -Stromboli and Lipari, and of the beach at Cumæ, is of a scanty nature. -If we eliminate the various evidences of cultivation which seem to occur -over much of the temperate regions of the globe, very little remains of -an interesting character. As in the south of England and in other -regions, the empty stones of the cherry, plum, and peach, the empty nuts -of the oak, hazel, &c., together with the spiral pods of Medicago figure -largely in the drift; and here and there we come upon the seeds of -littoral plants, such as Convolvulus soldanella and Euphorbia paralias. - -I have found Medicago fruits in all these localities on the beaches. -They often contain seeds, which, it may be added, have no buoyancy, the -seeded pods themselves floating from two to five days. The pods of -several kinds of Medicago form the great feature of Sicilian drift and -are often indications in other places of the vicinity of cultivated -districts. A small hairy species thrives on Letojanni beach near -Taormina, and I observed its seeds together with those of Euphorbia -paralias germinating in the drift stranded on the same beach. Arcangeli, -in his _Flora Italiana_, enumerates as many as thirty-three species of -Medicago. Many of the species grow in maritime districts, and their -fruits must often get into the beach drift independently of cultivation. -I noticed the pods amongst the drift brought down by the Alcantara, a -river near Taormina, a fact which goes to explain their presence in -beach drift.... On the beach of Trogilus Bay, near Syracuse, I gathered -several fruits of a Vitex, apparently V. agnus castus. After being kept -afloat for six weeks in sea-water some were placed in soil, when they -soon germinated and reproduced the plant. - -The beach drift of temperate Chile is described in Chapter XXXII. There, -as in other beaches of cool latitudes, it is not easy to find seeds -amongst the rubbish; but amongst the scanty seed-drift may be recognised -much of what we are familiar with in the Old World, such as the seeds of -Convolvulus soldanella, bits of the fruiting stems of Salsola kali, as -well as the rubbish indicating the white man’s presence, such as empty -stones of cherry, plum, and peach, Medicago pods, &c. In addition, we -find the seed-vessels of plants like Franseria and Nolana that are -peculiar to American beaches; and now and then, the seeds of Sophora -tetraptera, a tree of the immediately adjacent hill-slopes, come under -our notice. - -Before quitting this subject of the beach seed-drift of temperate -latitudes, it may be observed that when at San Francisco I visited the -beach running south from the Golden Gate. With the exception of the -fruits of Cakile maritima, a plant growing on the beach, few other seeds -or fruits were observed in the drift. - -The inference that there is very little effective dispersal by currents -in temperate regions is of some importance, and Sernander arrived at a -similar conclusion when discussing the origin of the Scandinavian flora. -The few plants with buoyant seeds and fruits, such as Arenaria -peploides, Cakile maritima, Crithmum maritimum, Convolvulus soldanella, -Euphorbia paralias, and Lathyrus maritimus, are no doubt thus dispersed, -and Norman is quite right in attaching some value to the distribution by -currents of certain plants within the region of the Arctic flora; but -after all it amounts to little, and geographical and climatic conditions -have often had a predominant influence in determining the distribution -in the temperate latitudes of littoral plants possessing buoyant seeds -or fruits. - -Nowhere is this shown more plainly than with the littoral plants with -buoyant seeds or seed-vessels that are found on our English beaches. -Some have evidently acquired their present distribution before ice and -snow reigned supreme in the extreme north. Though it may be possible, it -seems highly improbable, that either Arenaria peploides or Lathyrus -maritimus, both of which occur on beaches in high northern latitudes in -the Atlantic and Pacific Oceans (as in Arctic Norway, Spitzbergen, and -Behring’s Straits), could possess in our own day any means of -communication between their areas of distribution on the borders of -these two ocean-basins. - -So again with Cakile maritima, the occurrence of this or of two closely -allied species on both sides of North America cannot be attributed to -any present working of the currents for two reasons. In the first place, -as is remarked in Note 18, the results of two independent experiments -made by me show that the fruits will not float more than a week or ten -days in the sea, a capacity that will not admit of their transportation -by the currents over tracts of ocean more than one or two hundred miles -across. In the second place, this species is not an Arctic plant like -Arenaria peploides and Lathyrus maritimus; and the possibility of -inter-communication between the Atlantic and Pacific Oceans having any -effective value from the standpoint of dispersal, shadowy as it is with -the two Arctic species, is still more so in the case of Cakile maritima. -Norman’s observations on the coast of Norway, as quoted by Sernander -(page 123), indirectly indicate how hopeless it would be for this plant -to attempt to traverse the Arctic region. Just as I have noticed on the -north coast of Devonshire, the fruits occur plentifully in the beach -drift and germinate freely in the upcast wrack as far north as Senjen in -latitude 69°. Further north the plant has been recorded from only eight -localities, and since it is there sterile and but a summer annual, the -seed-vessels, it is argued, must have been brought by the currents from -the south. - -The reference to Cakile maritima as a summer annual on the north coast -of Norway is of interest; but I may point out that it displays a similar -behaviour in England on the north coast of Devonshire. Here, during the -latter half of July, 1903, I found the fruits common in the stranded -drift, and often in a germinating condition, whilst numerous seedlings -one to two inches high with the fruit-shell still attached were growing -out of the sand. From this arises the curious reflection that an annual -which germinates in the end of July could scarcely be expected to mature -its fruit before the winter. It would seem that this beach plant hampers -its own dispersal by its misdirected efforts; and the idea suggests -itself that we have here the explanation of its sterility in the north -of Norway. Had it been a perennial like Arenaria peploides and Lathyrus -maritimus it might have had a similar distribution within the Arctic -Circle. - -Quite other considerations seem to be suggested by the perennials -Crithmum maritimum and Euphorbia paralias. In these cases, although the -seeds or fruits, as the case may be, will float for months in sea-water -without apparently sustaining any injury, the species are confined to -the warmer parts of the European region. - -From Convolvulus soldanella we obtain another story. Its occurrence in -the temperate regions of both the northern and southern hemispheres, -great as the floating powers of the seeds may be, is concerned with -something more than with questions relating to modes of dispersal. The -circumstance that in its distribution in the temperate regions it is -practically coterminous with Ipomœa pes capræ in the tropics is very -significant (see Note 49). - -Each one of the English beach plants with buoyant seeds and fruits has -its own story of the past to tell. Time has indeed gathered on our -beaches current-dispersed plants, which, if they could speak, would tell -us strange stories of many latitudes, stories of change within the -Arctic Circle, and stories of great events within the temperate regions, -and, as in the case of Convolvulus soldanella, stories of a past within -the tropical zone. It cannot be said that investigators lack clues -leading to lines of inquiry into the age that immediately precedes our -own. - -Yet valuable as our British plants would be for this purpose, they do -not afford any indication that currents have played an important part in -plant distribution in temperate and arctic latitudes. Ekstam strikes the -true note for these regions when discounting the agency of currents in -the instance of the Spitzbergen flora, he regards the wind as the -greatest factor in seed-dispersal and after that the bird. The several -interesting points raised by this botanist are discussed in Chapter -XXXIII. - - - THE BEACH-DRIFT OF TROPICAL LATITUDES. - -Tropical beaches, as a rule, present a much greater abundance and -variety of stranded seeds and fruits than we find on beaches in -temperate latitudes. Observers in different parts of the tropics have -alluded to the enormous amount of vegetable drift floating in the sea -off the coasts, particularly in the vicinity of estuaries. Though much -of it is brought down by rivers, a good proportion is also derived from -the luxuriant vegetation that lines the beaches. Gaudichaud speaks of -the immeasurable quantity of drift (trees, branches, leaves, flowers, -fruits, and seeds) floating amongst the islands of the Molucca Sea; and -Hemsley, who quotes this author, gives other facts illustrating the same -point. Moseley tells us that seventy miles off the coast of New Guinea, -H.M.S. _Challenger_ found the sea in places blocked with drift (_Bot. -Chall. Exped._ iv. 279, 284). When the author of this book was in the -Solomon Group, long lines of vegetable drift were frequently observed -floating among the islands. The Rewa River in Fiji carries down a great -amount of drift to the sea; and as described in Chapter XXXII, the -Guayaquil River in Ecuador bears seaward an enormous quantity of these -materials. - -When we come upon this floating drift out at sea off an estuary, we -find, as Mr. Moseley pointed out, that the leaves have gone to the -bottom, whilst the floating islets, composed of the matted vegetation -lifted up from the shallows of a river channel, which form such a -feature in the Guayaquil River, have been dispersed or sent to the -bottom. However, a very large proportion of the seed-drift brought down -by a river from the interior has no effective value for the purposes of -dispersal. Many of the fruits and seeds brought down from inland owe -their presence in river-drift entirely to the buoyancy acquired by the -decay of the seeds. It is in its lower course when it traverses the -mangrove belt that a river picks up most of the material that is of -service in distributing the species; and this is mingled out at sea with -the numerous buoyant seeds and fruits of littoral plants that are swept -off the beaches by the currents. - -A description is given in Chapter XXXII of the enormous amount of -vegetable drift brought down by the Guayaquil River to the coast of -Ecuador. Besides the huge tree-trunks and the floating Pistias, we -observe large islets formed mainly of Pontederias and Polygonum, -together with a host of seeds and seed-vessels, both large and small, -including those of Anona paludosa, Entada scandens, Erythrina, Hibiscus -tiliaceus, Ipomœa, Mucuna, Vigna, &c., accompanied by the empty seeds of -Phytelephas macrocarpa and of many other strange plants from the slopes -of the Chimborazo mountains. In addition, we notice the seedlings of -Avicennia and of Rhizophora mangle together with the seeded joints of -Salicornia peruviana and the germinating fruits of Laguncularia. - -When in Fiji I made an especial study of the drift of the Rewa Estuary -within tidal influence, the results of which are incorporated in various -parts of this work. In the rainy season, when the drift is most -abundant, the following would be its most characteristic components: - - Seedlings of Bruguiera and Rhizophora. - Fruits of Barringtonia racemosa and B. speciosa, the first-named most - abundant and often germinating. - Seeds of Carapa obovata, most of them far advanced in germination. - Fruits of Lumnitzera coccinea. - Fruits of Cerbera odollam, abundant. - Fruits of Inocarpus edulis, with the seed generally rotten. - Fruits of Heritiera littoralis, Parinarium laurinum, and Pandanus. - Empty seeds of Aleurites moluccana. - Fruits of Scirpodendron costatum, abundant. - Fruits of Clerodendron inerme and Smythea pacifica, both of them in - some cases germinating. - Pyrenes of Morinda citrifolia. - Small fruits of Vitex trifolia and Premna taitensis, both sometimes - abundant. - Seeds of Entada scandens, Mucuna, and Vigna lutea. - Pods of Dalbergia monosperma and Derris uliginosa, the last sometimes - in a germinating condition. - Seeds of Hibiscus tiliaceus and of different species of Ipomœa, such as - I. peltata and I. pes capræ. - -Amongst other seeds and fruits brought down by the Fijian rivers and -stranded with a large amount of miscellaneous vegetable _débris_ on the -beaches in the vicinity of the estuaries are the seeds of Dioclea, -Strongylodon lucidum, and Afzelia bijuga; the empty seeds of Musa Ensete -(as identified with a query at Kew); the empty stones of the Sea tree, -apparently a species of Spondias; the seeds of Colubrina asiatica; the -fruits of an inedible indigenous Orange (Citrus vulgaris?) referred to -in Chapter XIII; the cocci of Excæcaria Agallocha and Macaranga; and -Coco-nuts. - -The occurrence in Fijian beach-drift of the seeds of Musa Ensete, or of -a wild banana much like it, is very remarkable. This species is found in -the mountains of Abyssinia and on the slopes of Kilima-njaro in -Equatorial Africa; but according to the monograph by Schumann on the -Musaceæ (Engler’s _Pflanzenreich_, 1900) the species is confined to -Africa, whilst all the other species of the subgenus are mostly -restricted to the same continent with the exception of one or two in -Further India. The empty seeds are frequent on the beach at Duniua at -the mouth of the Ndreke-ni-wai in Savu-savu Bay, Vanua Levu, and are -doubtless brought down by that river. Strangely enough the natives could -give me no information about the parent plant which I never discovered. -The seeds did not come under my notice in any other locality in Fiji. -They answer to the description and to the figure given by Schumann for -Musa Ensete; and their presence in the drift is one of the mysteries of -the Pacific floras. - -To enumerate the seeds and fruits found stranded on beaches in Fiji -would be to give a list of all the littoral plants with buoyant seeds or -fruits that are included in the list given in Note 2. I may here allude -to the fact that the Coco-nut, whether brought down by a river or -transported by a current, is able to germinate and establish itself when -washed up on the Fijian beaches. I have found these fruits germinating -amongst the drift stranded on the beaches near the mouths of rivers, -some just beginning to germinate and others already striking into the -sand and showing the first leaves. White residents living for years in -one locality were quite convinced that this frequently happens. One of -them pointed out to me some newly formed land at a river’s mouth, not -over two years old, on which were growing young plants three or four -feet high of Barringtonia speciosa, Calophyllum Inophyllum, and several -other plants including young Coco-nut palms, all growing from fruits -washed up by the waves and therefore self-sown. - -Like the littoral flora the beach-drift proper to the Hawaiian Islands -is very scanty. This is due to the scarcity of rivers, to the absence of -the mangrove-formation from which much of the drift is derived in other -tropical regions, and to the paucity of shore-plants with buoyant seeds -or fruits. As is observed in Note 30, where the composition of the beach -drift is described, the presence of a large amount of timber and of -other materials brought by the currents from the north-west coast of -America masks much of the local drift. - -Remarks on the beach-drift of the Panama Isthmus, and of the Ecuadorian, -Peruvian, and Chilian coasts of South America will be found in Chapter -XXXII. I have examined beach drift in other tropical regions, as in the -Solomon group, on Keeling Atoll, and on the south coast of West Java; -whilst there are at my disposal the data supplied by Schimper and Penzig -for the Malayan region including Krakatoa, and by Hemsley for tropical -regions generally. It will, I think, be best, if instead of describing -in detail the composition of the drift for each locality, I refer -briefly to the features that distinguish the tropical beach-drift of the -Old World from that of the New World. - -The beach-drift reflects the characters of the coast flora; and since -tropical littoral floras belong to two great regions, the Asiatic -including Polynesia and the African East Coast, and the American -including the African West Coast, the seeds and fruits stranded on the -beaches may be similarly referred to the same two regions. - -All over tropical Asia, as well as in the tropical islands of the Indian -and Pacific Oceans, the drift stranded on the beach presents the same -general character, and as a rule possesses seeds and fruits of the same -species that range over the whole or the greater part of this region. -Almost everywhere we find seeds or fruits of the same plants of the -beach formation, such as Barringtonia speciosa, Cæsalpinia Bonducella, -Calophyllum Inophyllum, Canavalia obtusifolia, Cerbera Odollam, Cordia -subcordata, Entada scandens, Guettarda speciosa, Hernandia peltata, -Hibiscus tiliaceus, Ipomœa pes capræ, Mucuna, Scævola Kœnigii, Sophora -tomentosa, Terminalia Katappa, and Tournefortia argentea. In those -localities where mangrove-swamps occur we find generally diffused in the -stranded drift of this region the seedlings of Bruguiera and Rhizophora, -the seeds of Carapa moluccensis, the fruits of Heritiera littoralis and -Lumnitzera coccinea, and the pods of Derris uliginosa. Amongst sundries -found over much of this region may be mentioned, the drupes of Pandanus, -the seeds of Erythrina, Vigna lutea, and Hibiscus tiliaceus, and the -“nuts” of Aleurites moluccana. With the exception of the last-named all -the fruits and seeds here enumerated are effectively dispersed by -currents over great areas. The sound nuts of Aleurites have no buoyancy; -and the nuts only acquire their floating power through the decay of the -kernel (see p. 419). - -The beach drift of the American region, a region which comprises both -the Pacific and Atlantic coasts of tropical America as well as the -African West Coast, has some features in common with the Asiatic -beach-drift and other features peculiar to itself. The plants, however, -that are represented in the drift of both regions are comparatively few, -and none of the large fruits of the Asiatic region are here to be -noticed. We observe, however, that the drift of the two regions possess -in common the seeds of Cæsalpinia Bonducella, Canavalia obtusifolia, -Entada scandens, Erythrina, Mucuna, Sophora tomentosa, and Vigna lutea, -all belonging to the Leguminosæ; and to these we must add the seeds of -Hibiscus tiliaceus and of Ipomœa pes capræ, and the seedlings of -Rhizophora and Avicennia. (Avicennia occurs in tropical Asia, but not in -Polynesia.) The distinctive characters of the beach-drift of both coasts -of America and of the west coast of Africa would be shown in the -presence of seeds of Anona paludosa, the fruits of Laguncularia -racemosa, Conocarpus erectus, Spondias lutea, and other plants. But the -beach-drift of the American region is much more scanty. Of the shore -plants generally dispersed in this region there could not be more than a -couple of dozen that are indebted for their wide dispersal to the -currents, and these alone figure in the effective beach drift. In the -Asiatic region these plants would number at least seventy or eighty. - - - _Summary._ - -(1) Effective dispersal by currents is mainly restricted to warm -latitudes, as is indicated by the scanty character of the seed-drift -stranded on the beaches of the south of England, Scandinavia, the -Mediterranean, and Southern Chile. - -(2) The present distribution in temperate latitudes of littoral plants -possessing buoyant seeds or seed-vessels is to be attributed more to the -influence of geographical and climatic conditions than to the agency of -currents. With some of them, such as those that occur on both sides of -North America, it is evident that their distribution antedates the -present climatic conditions within the Arctic Circle. - -(3) Time has gathered on an English beach current-dispersed plants that -could tell us strange stories of many latitudes. - -(4) The seed-drift that is often found in such abundance in tropical -seas is partly brought down by rivers and partly swept off the coast. -Very little of the seed-drift brought down by the rivers from the -interior is of any service for plant-dispersal, nearly all the floating -seed-drift found at sea which has any effective value being derived from -the plants of the beach and of the mangrove belt. - -(5) The tropical beach drift of the Old and New Worlds reflects the -characters of the littoral floras of those regions, more especially with -regard to the plants provided with buoyant seeds or seed-vessels. The -plants represented in the beach drift common to both these regions -belong mostly to the Leguminosæ. The large fruits so characteristic of -Old World beach-drift are not found in the New World. The number of -shore plants with buoyant seeds or seed-vessels that are widely -dispersed in the American region are only one-quarter or one-third of -those in the Old World region; and this difference is reflected in the -scanty character of tropical American beach-drift. - - - - - CHAPTER XXX - - THE VIVIPAROUS MANGROVES OF FIJI - - RHIZOPHORA AND BRUGUIERA - -_Rhizophora._—Represented by Rhizophora mucronata, Rhizophora - mangle, and the Selala, a seedless intermediate form.—Their - mode of association and characters.—The relation of the - Selala.—Polyembryony.—The history of the plant between the - fertilisation of the ovule and the detachment of the seedling.—Absence - of a rest period.—Mode of detachment of the seedling. Capacity - for dispersal by the currents.—_Bruguiera._—The mode of - dispersal.—Peculiar method of fertilisation.—Length of period between - fertilisation and the detachment of the seedling.—Mode of detachment - of the seedling.—Summary. - - -BETWEEN 1897 and 1899 I made numerous observations on the Fijian species -of Rhizophora and Bruguiera (mostly around the coasts of Vanua Levu and -in the Rewa delta); and these were supplemented in the early part of -1904 by observations on the first-named genus in Ecuador. I did not make -any collections in Fiji until Prof. Schimper asked me to obtain -specimens; and a fair-sized collection containing specimens dried, and -preserved in spirit, was sent to him. His illness and death shortly -followed, and I lost the advantage of his great experience in these -matters. In a letter written to me in 1898 he expressed the hope that I -would publish my notes on the mangroves of Fiji. Years have since passed -by, and as I read again his words of encouragement I take up once more -the interrupted task. - - - RHIZOPHORA - -Of the three species of this genus, two of them, Rhizophora mucronata -and R. conjugata, are Asiatic and are unknown in America; whilst the -third, R. mangle, was until recently regarded as peculiar to the -American and West African regions. - -When Mr. Hemsley wrote the Report on the Botany of the Challenger -Expedition he remarked (iii, 149) that the American Rhizophora (R. -mangle) appeared to be restricted to that region, and he questioned its -existence in the Pacific Islands as indicated by Jouan for New -Caledonia. The same view was taken by Prof. Schimper in his work on the -Indo-Malayan strand-flora published in 1891. There was, in fact, much to -support this view, since Dr. Seemann, one of the most accomplished -botanists who have explored the Pacific, describes only the Asiatic -Rhizophora (R. mucronata) in Fiji, and nothing is said of any other -species collected by the United States Exploring Expedition under Wilkes -in Fiji and Samoa. - -However, in a paper on the flora of Tonga, read before the Linnean -Society in 1893, Mr. Hemsley includes the American mangrove, Rhizophora -mangle, amongst the collections made there by Mr. Lister; and he refers -to its occurrence also in Stewart Island (I suppose in the Solomon -Group), but he suggests that it was accidentally introduced with ballast -in both these localities. In 1897 I found a species of Rhizophora, to -all appearances identical with the American species, in great abundance -in the Rewa delta in Fiji. Subsequently the same mangrove came under my -notice as the prevailing species in Vanua Levu in the same group; and on -sending photographs of a branchlet in flower and fruit and of the -germinating fruit to Prof. Schimper he expressed the opinion that they -belonged to the typical Rhizophora mangle. - -There are four typical mangroves in Fiji, namely (1) Bruguiera rheedii, -the “Dongo” proper of the natives; (2) Rhizophora mangle, usually known -as “Tiri-wai,” that is to say, the Tiri of the river, or rather of the -estuary; (3) Rhizophora mucronata, the “Tiri-tambua” of the Fijian, -signifying the Whale’s Tooth Tiri in allusion to the form of its fruit; -and (4) a seedless form intermediate between the two species of -Rhizophora, which the Fijians designate “Selala,” a name signifying “the -tree with empty flowers.” - -Bruguiera rheedii and Rhizophora mucronata were alone recorded by Dr. -Seemann and his predecessors; but he significantly refers to the natives -speaking of four mangroves. Mr. Horne, who spent twelve months in the -group some years later, also overlooked the American Rhizophora; but it -is apparent that both these botanists were naturally more interested in -the vegetation of the inland regions than of the coast swamps, and we -have before observed that they failed to record Scirpodendron costatum, -a giant-sedge very common and conspicuous in the swamps. It is not easy -to understand Dr. Seemann’s remark that “mangroves are restricted to but -few parts of the larger islands.” Horne, who was in the islands eighteen -years afterwards, makes frequent allusion to them. The natives whom I -questioned closely on this subject scouted the idea that any of the four -mangroves above named were recent arrivals. The coasts, as they said, -had always been extensively fringed by mangroves; and the reader has -only to refer to my remarks in the second chapter of my volume on the -geology of Vanua Levu to convince himself that mangrove swamps of -considerable extent existed in the time of Commodore Wilkes (1840). - - - _The Relative Abundance and Mode of Association of the three Fijian - forms of Rhizophora._ - -Stated in their order of frequency, we have first Rhizophora mangle, the -American species, then Rhizophora mucronata, the Asiatic species, and -lastly the Selala. The first is equally at home at the sea-border and on -the banks of brackish estuaries. The second is, as a rule, more -exclusively at home on the sea-coasts; and the same may be said for the -Selala. Usually all three kinds occur in the lower part of an estuary; -but as we ascend the river and the water freshens, the Asiatic -Rhizophora and the Selala disappear, and the American plant is alone -found in the higher reaches, where the density of the water ranges -according to the state of the tide between 1·000 and 1·010. I examined -the distribution of these three forms of Rhizophora in numerous -estuaries of Vanua Levu, as well as in the Rewa estuary in Viti Levu; -and it was ascertained that in all cases they followed the rule above -indicated. When the estuary receives but few streams and the water is -mostly salt, the three Rhizophoras may extend miles inland; but when it -contains a large body of fresh-water, Rhizophora mangle may be the only -form observed from the mouth of the river to the head of the estuary, -and it may monopolise the adjacent coasts. On the other hand, Rhizophora -mucronata may occupy almost exclusively a long extent of coast; or the -Selala may prevail in certain localities, as on parts of the Mathuata -coast of Vanua Levu. - -The manner of association of these three Rhizophoras is of interest in -connection with the origin of the seedless Selala. They very rarely -occur mingled together, but grow gregariously in contiguous colonies; -and not uncommonly all three may occur on the same line of coast within -a distance of a few hundred yards. The colonies pass into each other -without a break, and there is no fixed rule of association. Whilst on -the south side of Vanua Levu the Selala is generally associated with the -American Rhizophora, on the north side it is usually in touch with the -Asiatic species. In other localities all three occur in contiguous -colonies. The Selala colony may be exposed on the line of a river-bank -or along the sea-coast, or it may lie in the heart of an extensive -mangrove tract. The most extensive mangrove region in Fiji, that of the -Rewa delta, is in great part occupied by Rhizophora mangle; but all -three forms grow together in the eastern part of the delta; and here, -strangely enough, as at Daku, the Selala may grow sporadically, and all -three may grow mixed together with their branches intercrossing. - - -_The Characters of the Selala or Seedless Rhizophora compared with those - of the American Mangrove (R. mangle) and the Asiatic Mangrove (R. - mucronata)._ - -The three kinds of Rhizophora, when seen at the same time along a tract -of coast, may be readily distinguished by the different shades of green -of their foliage, that of Selala being dark green, that of Rhizophora -mucronata light green, and that of Rhizophora mangle intermediate in -shade. The Selala is usually the tallest of the three, and attains a -height of from 20 to 30 feet or even 40 feet and over, the aerial roots -dropped from the higher branches giving it a characteristic aspect. -Rhizophora mangle is generally the shortest, and at the coast is from 10 -to 12 feet high; but where the mangrove vegetation is most luxuriant, as -in the great swamps in the interior of the Rewa delta, it forms tall -trees as much as 40 feet in height, displaying the aerial roots hanging -from the higher branches. Rhizophora mucronata is, as a rule, -intermediate in height, and is distinguished by its stout, reddish trunk -and reddish aërial roots. - -The trunks of Selala are often in an inclined position and supported -entirely by the trestle-like aërial roots, the lower end raised some 5 -or 6 feet above the ground with the rest of the trunk inclined upwards. -They then look like gigantic walking-stick insects. The same habit may -be sometimes observed with the larger trees of Rhizophora mucronata, and -in fact all three may present at times the same habit of growth. The -taller trees of Rhizophora mangle may resemble the Selala in habit, and -the smaller trees of the Selala may approach the habit of Rhizophora -mangle. - -The distinctive characters of the Selala are given in the table -opposite. It will be there seen that this form is intermediate between -the other two species as regards the form and size of the petioles and -peduncles; the size of the bracts and bracteoles; the colour, form, and -size of the flowers; and in the length of the style. Its leaves are -smaller than in the case of the other two species, but pointed and -semi-aristate like those of Rhizophora mucronata. There are, however, -two varieties of the Selala, both with larger foliage than that -belonging to the prevailing type of the tree, and from 10 to 15 feet in -height. In one the flowers are more numerous, each flowering stem -branching four or five times and bearing at least twenty-four flowers, -the first branch being trichotomous and the rest dichotomous. In the -other, which is the prevailing form on the Mathuata coast, there is a -nearer approach to Rhizophora mucronata in the rounding of the peduncles -and in the length of the style. Then, again, there are divergent -varieties of Rhizophora mangle which in the larger bracts and bracteoles -and in the greater size, form, and paler hue of the flowers come nearer -to the Selala. Taking all the characters together, the Selala, though -intermediate between the Asiatic and the American species, comes in the -most critical diagnostic points, as in the inflorescence, in the -individual flowers, and in the form of the apex of the leaf, nearest to -Rhizophora mucronata, the Asiatic species. - -The seedless character of the Selala is well known to most Fijians of -the coast districts, the native name signifying empty (lala) flowers -(se). Now and then they aver that it produces fruit, but the numerous -offers of rewards in money never resulted in their bringing me the -fruits. During my residence of two years in the group I examined the -Selala trees in a great number of localities and never succeeded in -finding them in fruit. - -With all three kinds the anthers burst in the bud before it begins to -open, and we may ask why the process of self-fertilisation, which is -effectual with the other two kinds, produces no result with the Selala. -In all three cases the flower-buds and expanded flowers hang downwards, -and the expanded flowers retain their parts for the first twenty-four -hours, the pollen being caught in quantity on the hairy edges of the -petals. During the next day the withering stamens fall out, and on the -following day the petals fall too. With the Selala, the style soon -begins to blacken and wither, and in a few days the flower becomes -detached and drops off. With Rhizophora mucronata and Rhizophora mangle, -the style preserves its healthy condition, and shortly evidences of -fertilisation appear in the altered shape of the ovary. It is apparent, -therefore, that in the case of the Selala fertilisation has not -occurred, although the mechanical processes connected with it have been -carried out. The cause of this is not far to seek. - - +-----------------+--------------------------------------------------------+ - | | FIJI | - | CHARACTERS +------------------+------------------+------------------+ - | | RHIZOPHORA | SELALA (a | RHIZOPHORA | - | | MUCRONATA. | seedless form). | MANGLE. | - +-----------------+------------------+------------------+------------------+ - |Height of tree | | | | - | in feet | 12-20. | 20-40. | 9-12. | - | | | | | - |Colour of foliage|Pale green. |Dark green. |Intermediate | - | | | |shade. | - | | | | | - |Base of leaf |Tapering. |Sub-rounded. |Tapering. | - | | | | | - |Apex of leaf |Acute, and |Acute, and |Very obtuse, | - | |terminating |terminating in |with no | - | |in a twisted |a twisted point |twisted point. | - | |point a line |less than a line | | - | |(2·5 mm.) long. |(2·5 mm.) long. | | - | | | | | - |Leaf-stalk |Rounded, 1 - |Rather flattened |Length as in | - | (petiole) |1-2/10 inch, |horizontally, |Selala, but | - | |(25-30 mm.) long, |5-8/10th inch |flattening | - | |about as long as |(12-20mm.) long, |very marked. | - | |the peduncle. |shorter than the | | - | | |peduncle. | | - | | | | | - |Inflorescence |Branching |Branching usually |Usually branching | - | |(dichotomous) two |twice, but |only once | - | |or three times |sometimes three |(trichotomous) and| - | |with four to eight|times; first |bearing only three| - | |flowers. |branching |flowers; but | - | | |trichotomous, rest|sometimes | - | | |dichotomous; six |branching again | - | | |to twelve flowers;|(dichotomous) and | - | | |in one variety, |bearing then six | - | | |flowers as many as|flowers. | - | | |twenty-four. | | - | | | | | - |Peduncle (lowest}|Rounded. |Flattened above. |{Flattening more | - | flower-stalk) }| | |{marked than with | - | | | |{Selala. | - | | | | | - |Pedicels |As stout as the |More slender than |As in Selala. | - | |peduncle, and |the peduncle, and | | - | |rounded. |angular. | | - | | | | | - |Bracts and |Large, 1-1/2 line |Small, 2/3 line |Very small or | - | bracteoles |(4 mm.). |(2 mm.). |absent. | - | | | | | - |Calyx |Very pale yellow, |As in R. |Pale or bright | - | |or dirty white, |mucronata. |green, angular at | - | |rounded at base | |base in the bud, | - | |in the bud, lobes | |lobes 3-1/2 lines | - | |4-1/2 - 5 lines | |(8 mm.). | - | |(11-12 mm.). | | | - | | | | | - |Length of style |1-1/2 lines |1 line (2·5 mm.). |1/2 line | - | |(4 mm.). | |(1·5 mm.). | - | | | | | - |Fruit |Ovoid and usually |No fruits |Conical, somewhat | - | |symmetrical, with |produced. |curved, and thus | - | |large persistent | |not symmetrical; | - | |bracteoles at | |bracteoles at base| - | |base. | |very small or | - | |(Hypocotyl 16 | |absent. | - | |inches.) | |(Hypocotyl 9 or | - | | | |10 inches.) | - | | | | | - |Colour of trunk, | | | | - | rootstock and | | | | - | roots |Reddish. | — | — | - +-----------------+------------------+------------------+------------------+ - - +-----------------+-------------------------------------+ - | | ECUADOR. | - | CHARACTERS. +------------------+------------------+ - | |RHIZOPHORA MANGLE |RHIZOPHORA MANGLE | - | | (Mangle chico). | (Mangle grande). | - +-----------------+------------------+------------------+ - |Height of tree in| 10-15. | 50-80 and more. | - | feet | | | - | | | | - |Colour of foliage|Pale green. |Dark green. | - | | | | - |Base of leaf |Tapering or |Tapering. | - | |sub-rounded. | | - | | | | - |Apex of leaf |Very obtuse, with |Very obtuse, with | - | |no twisted point. |no twisted point. | - | | | | - |Leaf-stalk |Flattened above |Flat above with a | - | (petiole) |and below, with no|median groove, | - | |median groove, |1 inch (25 mm.) | - | |1/2 inch (12 mm.) |long, two-thirds | - | |long; not half as |the length of the | - | |long as the |peduncle. | - | |peduncle. | | - | | | | - |Inflorescence |As described under|Branching at least| - | |R. mangle of Fiji.|three times, | - | | |sometimes four or | - | | |five times, | - | | |trichotomous or | - | | |dichotomous, | - | | |twelve to | - | | |forty-eight | - | | |flowers. | - | | | | - |Peduncle (lowest}| | | - | flower-stalk) }|}Sub-angular. |Rounded. | - | | | | - |Pedicels |More slender than |More slender than | - | |the peduncle, and |the peduncle, and | - | |rounded. |angular. | - | | | | - |Bracts and |Scarcely |Well developed, | - | bracteoles |developed, 1/2 |1 line (2·5 mm.). | - | |line (1 mm.). | | - | | | | - |Calyx |As with R. mangle |As with R. | - | |of Fiji. |mucronata and | - | | |Selala of Fiji, | - | | |but lobes 4 lines | - | | |(10 mm.). | - | | | | - |Length of style |Less than a line |1-1/2 lines | - | |(2·5 mm.). |(4 mm.). | - | | | | - |Fruit |As in R. mangle of|Conical, not | - | |Fiji. |symmetrical, and | - | |(Hypocotyl 9 or |somewhat curved; | - | |10 inches.) |large persistent | - | | |bracteoles at base| - | | |as in R. | - | | |mucronata. | - | | |(Hypocotyl 12 to | - | | |15 inches.) | - | | | | - |Colour of trunk, | | | - | rootstock and | | | - | roots | — | — | - +-----------------+------------------+------------------+ - -Although the ovaries of the Selala contain four ovules, which in size -and appearance do not differ from those of Rhizophora mangle and R. -mucronata, its pollen when compared with that of the other two forms -presents a remarkable difference. The pollen of these three mangroves -was examined in five localities far apart from each other, and in all -the same results were obtained. The pollen-grains of the Selala are much -smaller than those of the other two, and differ much from them in form. -They are irregularly oval in outline, and have a shrunken look beside -the spherical symmetrical grains of the two species with which they are -compared. They are from one-fourth to one-third the size of those of -Rhizophora mucronata, and from one-third to one-half the size of those -of Rhizophora mangle. - -There is much to support the view that the Selala is a cross between the -other two species, its intermediate characters and its seedless -condition being especially indicative of such a derivation; but there -are several difficulties in accepting this explanation. - -(1) The circumstance of the anthers bursting in the flower-bud would -considerably lessen the chances of cross-fertilisation; but this -objection is not insurmountable, since numerous insects, such as flies, -ants, and small coleoptera, visit the newly opened flowers, and they -might sometimes produce a result. When I made this suggestion to Prof. -Schimper he replied that insect-pollination was quite possible after the -expansion of the flowers. - -(2) If, as seems highly probable, the pollen of Selala is impotent -and the ovules fertilisable, then its seedless condition implies not -only an incapacity for self-fertilisation, but also for -cross-fertilisation; and if Selala with its impotent pollen does not -admit of cross-fertilisation, this would still less be expected of -Rhizophora mucronata and R. mangle where the pollen is potent and -where fertilisation takes place in the bud. I endeavoured to -fertilise the Selala flowers with the pollen of the two other -species; but there were no results, the flowers falling off in a few -days. It may here be remarked that on one Selala tree I found a -solitary flower with an enlarged ovary, as if through fertilisation. - -(3) It is not easy to explain the gregarious growth of the Selala if it -is a seedless hybrid. The colonies could not be renovated by mere -intercrossing, especially in places where, as on the north coast of -Vanua Levu, the dense belt of mangrove is for many miles composed in -mass of Selala trees, with a few trees of the Asiatic and American -Rhizophoras growing on the outskirts. - -It is obvious that in order to clear the way for considering this -problem the means of renovating the Selala colonies should be inquired -into. In the first place, whilst seedlings occur in numbers under the -trees of the other two Rhizophoras they are never to be found under the -Selala trees. The mode of reproduction of the Selala is evidently -vegetative, and the question arises as to what mode of vegetative -reproduction occurs. The Selala trunks, as already observed, are often -inclined, the trunks being supported on trestle-like aërial roots. These -trunks send out branches which in their turn drop aërial roots; and when -the decay of the parent trunk takes place, the branches are able to live -independently. The primary branches in due time send out secondary -branches which also let fall aërial roots; and thus the process is -repeated indefinitely, the result being a maze of semi-prone trunks, -branches, and aërial roots. The first stage of the process ends with the -death of the parent trunk, and the primary branch, supported by its own -aërial roots, is often all that the observer can distinguish in the -centre of a colony. This is evidently the mode by which the Selala -colonies are renovated in their interior. One sometimes observes in the -midst of one of these colonies extensive bare mud-flats 100 to 500 yards -across from which apparently the trees have died off _en masse_. The -natives assert that when part of a Selala tract is cleared the trees -never grow again. - -But _pari passu_ with this process of vegetative reproduction of the -Selala, by which the mass of the colony is preserved and renovated, -there is evidently some other process of reproduction in operation -amongst the trees of Rhizophora mangle and R. mucronata at the edge of -the colony, as a result of which Selala seedlings are produced. Whilst -no seedlings are to be observed striking into the mud under the Selala -trees, numbers occur, as before observed, under the trees of the other -two species. Those under the trees of R. mangle possess in nearly all -the cases the distinctive leaf-characters of that mangrove, and would be -recognised at once as belonging to that species. On the other hand, -those beneath the trees of R. mucronata are of two kinds, some of them -being readily recognised by their foliage as of the Selala type others, -again, being typical seedlings of R. mucronata. Only those seedlings, or -“keimlings” as we might term them, were noted that had dropped plumb -from the branches above. - -Such were the results of my investigations on Vanua Levu. My field of -inquiry was then shifted to the Rewa delta, where, with the assistance -of the Daku natives, who, like most Fijians, display a keen interest in -matters relating to their plants, I spent a few days in investigating -the origin of the Selala trees that grow sporadically in that locality. -On pulling up some of the young trees we found that the original -radicular or hypocotyledonary portion of the keimling could be still -distinguished. My zealous native friends also pointed out to me that -though the leaves in form and colour were those of the Selala, the -rootstock was reddish like that of R. mucronata, and not white as with -R. mangle. The natives averred that the Selala trees are produced in the -first place from fruits of R. mucronata. When young, they said, they are -Tiri-tambuas (R. mucronata), but when old, Selalas. Yet although R. -mucronata may be now regarded as the source of the Selala trees, and my -Vanua Levu observations pointed unmistakably in this direction, it could -not be definitely settled whether this was the result of a cross with -the male element of R. mangle or whether the Tiri-tambua (R. mucronata), -in producing two types of seedlings, one fertile with the parent -characters and the other seedless of the Selala type, brought about the -same end. On the whole I am inclined to the view that the Asiatic -Rhizophora presents us in the dimorphism of its seedlings the true -explanation. - -This inference is supported by the behaviour of Rhizophora mangle on the -coast of Ecuador, a subject which is discussed in Chapter XXXII, and I -have given the results of my observations on the Ecuadorian Rhizophoras -side by side with those on the Fijian trees in the table before given. -There are two very distinct forms of the American Rhizophora (R. mangle) -in the swamps of Ecuador. There is the low coast tree, the “Mangle -chico” of the Ecuadorians, ten to fifteen feet in average height, which -grows on the sea-front of the swamps and has all the general appearance -and the more conspicuous characters of the American Rhizophora in Fiji. -There is also a tall tree, 60, 80, or even 100 feet high, that forms the -great mass of the mangrove swamps. In its inflorescence, in the dark -green colour of its foliage, and in other characters, it comes near the -Fijian Selala; but it differs in fruiting abundantly. This is locally -termed the “Mangle grande,” and its true relation to the Fijian Selala -appears to be as follows. Whilst both as regards the flowers approach -the Asiatic Rhizophora (R. mucronata), the Fijian Selala resembles the -Asiatic tree also in its foliage, whilst the “Mangle grande” or the -Ecuadorian Selala more resembles the typical American tree (R. mangle) -in its leaves and also in its seedlings. Here in the Ecuadorian swamps -there can be no question of crossing, since both, according to Baron von -Eggers, belong to one species. Therefore I am inclined to the opinion -that whilst the Asiatic Rhizophora displays dimorphism in Fiji, the -American Rhizophora displays dimorphism in Ecuador. The reversion on the -part of the “Mangle grande” of Ecuador to some of the characters of the -Asiatic plant is remarkable, and points to the greater antiquity of the -Asiatic R. mucronata as compared with the American R. mangle. - -This accords with the opinion expressed by Schimper in his work on the -Indo-Malayan strand flora that the American Rhizophora is either a -degenerated descendant of the Asiatic R. mucronata or a sister form -derived from a common ancestor. America, as we have seen, possesses only -one of the three species of Rhizophora, and this is the only -representative that it owns of the four Asiatic genera (Rhizophora, -Kandelia, Ceriops, Bruguiera) that constitute the tribe Rhizophoreæ. The -rule prevailing with current-dispersed plants that America is a -distributor and not a recipient evidently does not apply to the -Rhizophoreæ; and to explain their distribution we must go back to some -epoch very remote from the present. That Fiji derived its -representatives of Rhizophora mangle from America by the agency of the -currents I do not for a moment admit. The restriction of the species and -indeed also of the genus to the Western Pacific is very significant. It -is far more likely that, as I have pointed out in the case of Lindenia -(see page 396), the American Rhizophora was once widely distributed over -the tropics of the Old and New Worlds, and that it is now on the “down -grade” towards extinction. Its survival in the Western Pacific could -thus be explained without our being obliged to suppose that the -seedlings or keimlings have been carried uninjured across the Pacific -Ocean, an ocean voyage for which, as shown in a later page, they are not -well fitted. - - - _The Occasional Occurrence of more than one Seed in the Fruits of - Rhizophora mucronata and Rhizophora mangle (Polyembryony)._ - -The bilocular ovary contains four ovules, one of which only as a rule -becomes a seed. But it is incorrect to say that the fruits are always -one-seeded, since two or even three seeds are occasionally produced, and -they may all germinate. In November, 1897, I noted eight hundred fruits -of Rhizophora mangle germinating on the trees in one of the creeks of -the Rewa delta. Out of this number eight fruits had two germinating -seeds and one had three, the protruding radicles being in all stages of -growth. Just two years afterwards I counted eight hundred more fruits in -the same locality, and then observed seven with two germinating seeds -and none with three, the radicles protruding in all cases. On another -occasion at Wailevu in Savu-Savu Bay I counted four hundred, and none -had more than a single radicle protruding. The results appear to vary -with the locality, but in the Rewa creek the proportion of fruits in -which more than one seed germinated was fairly constant at dates two -years apart, namely, about one per cent. Occasionally, however, in -particular localities a greater proportion may be noticed. Thus near -Daku in the Rewa delta I found that the proportion was between two and -three per cent. for the same species (R. mangle), those with three -germinating seeds being about half per cent. - -The case of more than one seed germinating in the fruits of Rhizophora -mucronata never came under my observation; but in one locality, where I -examined a considerable number of fruits near the stage of germination, -between ten and fifteen per cent. showed two seeds approaching maturity. - -Warming thoroughly investigated the polyembryony of Rhizophora more than -twenty years ago, seemingly from materials brought to him from the West -Indies (Engler’s _Botanische Jahrbücher_, band iv., 1883). With the -usual German thoroughness he deals with the work of earlier observers, -and goes back to Piso in the middle of the 17th century. Of the four -ovules, he remarks, three usually abort, and only in rare cases are two -seeds developed. He quotes Baron von Eggers to the effect that only in -three per thousand cases was more than one seedling observed protruding -from a germinating fruit. These remarks evidently all apply to the -American species. I do not find any reference in my notes to -polyembryony in Ecuador, and evidently its occurrence is not so frequent -there as in Fiji. - -It is frequently apparent in the cases where more than one seed -germinates in a fruit that on account of the difference in the length of -the protruding seedlings germination does not always begin at the same -time. Thus in Fiji the difference in the length varied between one and -three inches, an amount representing at least from ten to twenty days’ -growth, as will be subsequently pointed out. Warming in one of his -figures gives a fruit where an interval of some months seems to be -indicated, since one of the seedlings has fallen out and the other is -protruding less than an inch. By cutting across a fruit containing two -seeds one may sometimes observe one seed quiescent and the other -beginning to germinate. The significance of this occasional interval -between the germination of seeds in the same fruit will be referred to -in a later page. - - - _The Seasons of Flowering and Fruiting of the Species of Rhizophora in - Fiji._ - -The Selala flowers all the year. With the two American and Asiatic -species there are considerable variations between different localities. -Generally speaking, they flower and fruit all the year through; but the -flowers are usually less abundant in the warm season from December to -February, and the germinating fruits which are to be observed on the -trees every month of the year are more numerous in that season. - - - _The History of the Reproductive Process in Rhizophora from the - Fertilisation of the Ovule to the Falling of the Plantlet or Seedling - from the Tree._ - -I devoted great attention to this subject in the instance of Rhizophora -mangle, being desirous of determining two points, in the first place as -to whether there was any period of rest between the maturation and -germination of the seed, and in the second place as to the period that -elapsed between the commencement of germination and the fall of the -seedling. - -The principal change in the ovary for the first three or four weeks -after fertilisation is shown in its increased breadth. The increase in -height is but slight during this period; and in fact after thirty days -the ovary only added 2 millimetres to its original height of 3 -millimetres. After this the growth of the fruit proceeds until the tip -of the radicle pierces its summit, the fruit being then about eleven -lines (2·8 cm.) long. _From the date of fertilisation to the time the -radicle pierces the top of the fruit a period of about fifteen weeks -elapses._ (The fruit, it should be here remarked, continues to grow in -length and breadth after the radicle has protruded, attaining a length -of thirteen or fourteen lines [3·5 cm.] when the seedling or “keimling” -is ready to fall.) - -By referring to the table below it will be observed that there is no -period of rest in the growth of the fruit up to the date of the -protrusion of the radicle. It will now be shown that there is normally -no pause between the epoch of the maturation of the seed and the -beginning of germination, or, in other words, that from the time of the -fertilisation of the ovule to the onset of germination there is no -cessation in the process of growth of the embryo. That period of dormant -vitality which almost all seeds pass through forms no normal feature in -the life-history of this species of Rhizophora. - - - RHIZOPHORA MANGLE AND R. MUCRONATA. - - +-------------------------------------------+ - | RHIZOPHORA MANGLE. | - +-------------------------------------------+ - | _Growth of fruit in height._ | - +---------------------+---------------------+ - | Lines or tenths of | Number of days | - |an inch (millimetres | since | - | in brackets). | fertilisation. | - +---------------------+---------------------+ - | | | - | 2 (5) | 30 | - | 3 (7·5) | 42 | - | 4 (10) | 50 | - | 5 (12·5) | 61 | - | 6 (15) | 67 | - | 7 (17·5) | 74 | - | 8 (20) | 83 | - | 9 (22·5) | 92 | - | 10 (25) | 100 | - | 11 (28) | 105 | - | | { Protrusion| - | | 107{ of the | - | | { hypocotyl | - | | - | _Growth of the protruding hypocotyl._ | - | | - | 10 (25) | 127 | - | 20 (50·5) | 141 | - | 30 (76) | 151 | - | 40 (101·5) | 160 | - | 50 (127) | 167 | - | 60 (152) | 175 | - | 70 (177·5) | 185 | - | 80 (203) | 202 | - | 90 (228) | 222 | - | | {Fall of | - | | 229{the | - | | {seedling | - +---------------------+---------------------+ - - _Explanation of the Table._ - -We have here shown the period between fertilisation and the fall of the -seedling from the tree. - -This period divides itself into two parts, the first being concerned -with the continuous growth of the fruit and of the inclosed embryo until -the tip of the hypocotyl appears through the apex of the fruit, the -second being indicated by the growth of the protruding hypocotyl until -the fall of the seedling. - -The height of the fruit is measured from the base of the calyx-lobes, -and the length of the hypocotyl at first from the apex of the fruit and -afterwards from the edge of the protruding neck of the cotyledonary -body. The height of the ovary at the time of fertilisation is about 3 -millimetres; and from that time onward it is to be regarded as a fruit. - - [_To face page 452._ - - FIGURES ILLUSTRATING THE DEVELOPMENT OF - THE SEED AND THE GERMINATING PROCESS - OF RHIZOPHORA AND BRUGUIERA - - (Natural size. Drawn for convenience of description in the erect - position.) - - 1. Rhizophora mucronata Fruit 3-1/3 lines (8-9 mm.) high, six to - seven weeks after fertilisation. The - micropyle is but slightly dilated, and is - occupied by a small plug of endosperm. - - 2. Rhizophora mucronata Seed of fruit represented in Fig. 1. - - 3. Rhizophora mucronata Fruit 5 lines (12 mm.) high, eight to nine - weeks after fertilisation. Germination is - about to begin. A large plug of endosperm now - protrudes through the dilated micropyle, but - still covers the lengthening hypocotyl. - - {Seed of fruit represented in Fig. 3. In - 4.} Rhizophora mucronata {Fig. 4 the plug of endosperm is shown on the - 5.} {upper end of the seed; whilst in Fig. 5 it - {has been removed, exposing the tip of the - {hypocotyl. - - 6. Rhizophora mucronata Fruit 7-1/2 lines (18 mm.) high, eleven to - twelve weeks after fertilisation. - - 7. Rhizophora mucronata Fruit, seventeen or eighteen weeks after - fertilisation. - - 8. Rhizophora mucronata Full-grown fruit with upper portion of - seedling just before detachment from the - tree. The long tapering plumule is here - inclosed in the cotyledonary body. The * - indicates the point of detachment of the - seedling. - - 9. Rhizophora mucronata The cotyledonary body of Fig. 8. - - 9A.} {Illustrating different stages in the - 9B.}Rhizophora mucronata {development of the plumule and of the neck of - 9C.} {the cotyledonary body resulting finally in - 9D.} {the expulsion of the plumular end of the - {seedling from the fruit cavity as in Fig. 8. - {(See page 458.) - - 10. Rhizophora mucronata Fruit with two seeds. - - FIGURES ILLUSTRATING THE DEVELOPMENT OF - THE SEED AND THE GERMINATING PROCESS - OF RHIZOPHORA AND BRUGUIERA—(_continued_) - - - (Natural size. Drawn for convenience of description in the erect - position.) - -11. Rhizophora mangle Fruit, six weeks after fertilisation. - -12. Rhizophora mangle Seed with plug of endosperm, as shown in Fig. 11. - -13. Rhizophora mangle Fruit, eight weeks after fertilisation. The tip of - the hypocotyl is now piercing the plug. - -14. Rhizophora mangle Embryo (enlarged) shown in Fig. 13. - -15. Rhizophora mangle Fruit, ten weeks after fertilisation. The growing - hypocotyl has now pierced the plug. - -16. Rhizophora mangle Embryo shown in Fig. 15. - -17. Rhizophora mangle Fruit, nearly sixteen weeks after fertilisation. - -18. Rhizophora mangle Full-grown fruit, just before the detachment of the - seedling from the tree. The long tapering plumule - is inclosed in the tube of the cotyledonary body. - The point of detachment of the seedling is - indicated by *. - -19. Rhizophora mangle The cotyledonary body of Fig. 18. - -20. Rhizophora mangle Fruit with two seedlings in different stages of - growth (given in the first plate). - -21. Bruguiera Rheedii Fruit, about four weeks after fertilisation. (The - shaded portion is the calyx-tube or cup, in the - midst of which rises the style.) - -22. Bruguiera Rheedii Germinating seed. - -23. Bruguiera Rheedii Germinating fruit, about eight weeks after - fertilisation. - -24. Bruguiera Rheedii Germinating fruit, about ten weeks after - fertilisation. Here the growing hypocotyl, - carrying the style with it, has pushed upwards the - lining membrane of the floor of the calyx-tube, - which has ruptured and forms a cap on its extremity. - -25. Bruguiera Rheedii Germinating fruit, thirteen or fourteen weeks after - fertilisation. - -26. Bruguiera Rheedii Fruit with full-grown seedling just before its - detachment from the tree. - - [_To face page_ 453. - - - RHIZOPHORA MUCRONATA. - - _Growth of the first seven inches of the hypocotyl after it protrudes - from the fruit._ - - 10 lines (25 mm.) after 26 days - 20 lines (50·5 mm.) after 41 days - 30 lines (76 mm.) after 51 days - 40 lines (101·5 mm.) after 61 days - 50 lines (127 mm.) after 70 days - 60 lines (152 mm.) after 78 days - 70 lines (177·5 mm.) after 86 days - -In my description of the germinating process of Rhizophora mangle from -this particular standpoint I adopt the general views of Prof. Schimper, -the observations being my own, the phraseology employed being his. It -would be out of place here to deal with the biological significance of a -process to which observers like Warming, Goebel, Karsten, Schimper and -Haberlandt have applied their greater talents as well as their greater -experience. I investigated the subject carefully from my own standpoint -of inquiry, and whilst the reader will find in my rough sketches of the -various stages of the process a little aid in following the argument, he -is referred for detailed treatment of the subject to the memoirs of the -above-named botanists as well as to those of yet more recent -investigators. - -After fertilisation, according to Prof. Schimper (_Ind. Mal. -Strandflora_), the embryo-sac is filled with endosperm, which -subsequently protrudes and forms a plug completely closing the micropyle -(see my figures). As my observations showed, the seed during the first -eight weeks after fertilisation increases continuously in size, and the -plug of endosperm, at first inconspicuous, becomes of considerable size, -the seed attaining a length of seven millimetres. The embryo meanwhile -grows rapidly, and at the end of this period of eight weeks the -radicular tip or the point of the hypocotyl begins to protrude from the -micopyle, still covered by the plug of endosperm, the fruit being -between four and five lines (10-12 mm.) in length (figures 11-14). In -another week, when the fruit has grown another line in length, the tip -of the radicle is on the eve of piercing the plug, and this may be -termed _the commencement of germination, nine weeks after the act of -fertilisation_. The next stage, after an interval of one and a half -weeks, is illustrated in figure 15; and _after a period of about fifteen -weeks from the date of fertilisation the tip of the radicle pierces the -top of the fruit_. As shown in the figures, the fruit grows in length -throughout the process. - -The question as to whether the matured seed passes through a stage of -quiescence before it germinates finds its answer in the statement that -only nine weeks elapse between fertilisation and germination. It may, -however, be urged that the maturation of the seed could be accomplished -in a few weeks, and that after this a period of dormant vitality might -follow. This objection can be at once disposed of and the whole matter -placed beyond reasonable doubt by making, as I did, a large number of -vertical sections of the fruit in all its stages. It will then be -perceived that there is a fairly constant relation in all stages of -growth between the seed and the fruit, whether maturating or -germinating. Since the growth of the fruit is continuous (see Table) up -to the time of the protrusion of the tip of the hypocotyl through its -coats, it follows that there can be no appreciable pause between the -completion of maturation and the commencement of germination of the -seed. In other words, both fruit and seed preserve the same relation -during the process, and the absence of any period of rest is to be -inferred from the uninterrupted growth of the fruit. - -We will take, to illustrate this point, a fruit between four and five -lines long in the stage that immediately precedes germination (see -figure 11). The fruit proceeds with its growth, and the seed, we will -suppose, remains quiescent for a month. At the end of that time (see -Table) the fruit would be eight lines long, and the seed, of course, -would be unchanged. This condition of things never presented itself to -me. Fruits eight lines long were always far advanced in germination (see -figure 15). If the seed passes through an interval of rest before -germination, it must be of a very short duration and practically _nil_. - -This absence of any period of rest between the final maturation of the -seed and its fertilisation had already been assumed by Prof. Schimper. -Writing to me on July 14, 1898, when my observations were in progress, -he says:—“I am ready to assume, according to my own experience, that -there is continuous development until the falling off of the embryo. -More accurate observations on the subject would be interesting, and -would not present any great difficulties.” At the end of the same month -he wrote the preface to his great work on Plant-Geography; and he -expresses himself decidedly on this point. Speaking of Rhizophora -mucronata (English edition, p. 396), he says that “the fruit ... soon -after the completion of its growth is pierced at its summit by the green -hypocotyl, as the embryo does not undergo any period of rest, but -continues to develop without interruption.” - -Though the rest-period is normally non-existent with the seeds of -Rhizophora, it has already been observed that it is indicated in rare -cases and under exceptional conditions. Thus I have already remarked -that in Fiji about one per cent. of the germinating fruits of the -American species exhibit more than one seed. These seeds usually begin -to germinate about the same time, but in a few cases, say, one in ten, a -marked difference in the length of the protruding hypocotyls points to -the fact that one of the seeds began to germinate some weeks after the -other. We at times also meet with fruits which when cut across display -two seeds, of which only one is beginning to germinate. Such cases -indicative of a pause between the maturation of the seed and the -beginning of germination would be very rare. With Rhizophora mangle, -probably one in a thousand fruits would be a generous estimate. - -In passing it may be remarked that the same stages occur with Rhizophora -mucronata in the development of the seed and in the subsequent -germinating process. When the fruit is three lines long the micropyle is -but slightly dilated (see figures 1 and 2). When it is four lines long -the endosperm begins to escape from the gaping micropyle and forms a -projecting plug. The growth of the embryo now becomes rapid, the -endosperm escapes in greater quantity, and by the time the fruit is five -lines long the tip of the radicle is on a level with the micropyle, -although still covered by the plug (see figures 4, 5). After this, -germination begins; and when the fruit is six lines in length the -radicle is in the act of penetrating the plug. Ultimately the tip of the -radicle pierces the top of the fruit when this last is nine or ten lines -long. As shown in the figures there is continuous growth of the fruit -during the maturation and germination of the seed, until, in fact, the -plantlet drops into the water. With reference to the stage when -germination begins, it should be remarked that the formation of the -large plug of endosperm outside the micropyle does not necessarily -indicate the beginning of germination. Germination is in progress only -when the hypocotyl or radicle begins to lengthen and is on the point of -piercing the plug of endosperm that fills up the gaping micropyle. This -is well shown in this species in the case of fruits with two seeds. Both -seeds may have large plugs of endosperm, and yet only one may show -indications of germination in the lengthening hypocotyl. - -We must now return to the subject of the growth of the hanging seedling -of Rhizophora mangle. We have already remarked that, as shown in the -Table, about fifteen weeks (107 days) is the average time elapsing -between the fertilisation of the ovule and the protrusion of the tip of -the radicle through the top of the fruit. A further period of seventeen -and a half weeks (122 days) is occupied by the growth of the seedling on -the tree, at the end of which period it drops into the water or mud -according to the state of the tide. This gives a total period of nearly -thirty-three weeks (229 days) as the duration of the time between -fertilisation and the fall of the seedling. This may be divided, as has -been already implied, in the following manner:— - - (1) Period between fertilisation and germination. 9 weeks. - - (2) Period between the commencement of germination 6-1/2 - and the protrusion of the tip of the radicle weeks. - through the top of the fruit. - - (3) Period occupied by the growth of the hypocotyl 17-1/2 - outside the fruit, and terminating in the fall weeks. - of the seedling from the tree. - - —— - - Total 33 weeks. - -This represents the average of numerous observations, the deviations -being from two to three weeks on either side. In the latter part of its -growth, the lower end of the hypocotyl becomes thickened or club-like, -and during the last week or ten days the increase in length is arrested -altogether. - -My observations on the growth of the seedling on the tree of Rhizophora -mucronata were comparatively few; but, as shown in the Table on page 453 -they give nearly the same rate of growth. Taking the average length -attained by the hypocotyl on the tree at sixteen inches, and employing -as well the data supplied by Rhizophora mangle, a period of 26-1/2 weeks -would elapse from the time the hypocotyl pierces the top of the fruit -until the plantlet falls from the tree. If we then add, as in the case -of the other species, 15-1/2 weeks for the preceding period between -fertilisation and the protrusion of the hypocotyl, we get a total of 42 -weeks for the whole period from fertilisation to the fall of the -seedling. In the extreme cases where a length of almost two feet is -attained on the tree, the period would somewhat exceed twelve months; -and in those rare instances in other regions, when, according to -Schimper, the seedling is a metre in length, probably eighteen months -would be required. The period for Rhizophora mucronata is thus -considerably longer than for R. mangle, which is sufficiently indicated -by the difference in the average length of their hypocotyls on the tree -in Fiji, that for R. mucronata being sixteen inches, and that for R. -mangle nine or ten inches. - -The only other observations that have come under my notice relating to -this subject are those made by Jacquin on Rhizophora mangle in the West -Indies in the middle of the eighteenth century. The results are -literally quoted by Warming; but I have referred to the original account -in the work of Jacquin, entitled _Selectarum Stirpium Americanarum -Historia_, Vindobonæ, 1763. According to this observer the seedling -falls from the tree in the twelfth month from the fecundation of the -flower. This happened in my observations on the same species in Fiji in -the eighth or ninth month. Jacquin states that the tip of the radicle -protrudes from the fruit in the third month, whilst my results give it -as taking place in the fourth month. The difference in the length of the -total period, it may be remarked, would be to a great extent determined -by the varying length acquired by the seedling before it drops from the -tree. In ordinary conditions it averages about ten or eleven inches, and -the hypocotyl itself attains a length of nine or ten inches on the tree, -both in Fiji and Ecuador; but in sheltered localities it may attain a -length half as long again. I have already pointed out in the case of the -fruits of Rhizophora mucronata that a year and more would be sometimes -required, and the same remark would apply to unusually long fruits of R. -mangle. Local conditions would often produce varying results, both in -the rate of growth of the hanging seedling and in the duration of the -period of its attachment to the tree; but it is probable that nine or -ten months would represent for the genus the average length of the -period between fertilisation of the ovule and the detachment of the -seedling from the parent tree. - - - _The mode of separation of the seedlings of Rhizophora mangle and - Rhizophora mucronata_ - -This is a process of expulsion almost akin to parturition, and is -brought about by the outward growth of the neck of the cotyledonary -body. There is much that is of great interest in this subject; and I may -add that Haberlandt, in a memoir published in the _Annales du Jardin -Botanique de Buitenzorg_ for 1894, gives the results of an elaborate -study of the viviparous process in this and other genera of mangroves. -The same analogy seems also to have presented itself to him, but only in -connection with the means employed in some of the genera, as with -Bruguiera, for conveying nourishment to the growing embryo. He remarks -that he was involuntarily reminded by these structures of the -chorion-tufts and lobes in the placenta of mammals, and that such -structures in the mammal are functionally nothing more than true -_haustoria_ as found in the viviparous mangroves. - -When studying the germination of the American and Asiatic Rhizophoras in -Fiji, I observed that the neck of the cotyledonary body did not begin to -form, nor the inclosed plumular bud to show signs of differentiation, -until the hypocotyl had protruded about 4-1/2 inches with R. mangle, and -between 6 and 7 inches with R. mucronata. The neck of the cotyledonary -body then proceeds to grow in length, pushing before it the plumular end -of the embryo-seedling, which it surrounds as a sheath. This operation -continues until the hypocotyl has acquired a length of about seven -inches with R. mangle, and about nine inches with R. mucronata, when the -neck begins to protrude outside the fruit. The cotyledonary neck -proceeds with its growth, and by the time the seedling is ready to fall -from the tree it protrudes about an inch from the fruit-shell, having -carried the growing plumular bud with it. The plumular end of the -seedling has been now more or less expelled from the fruit-cavity, and -the connection between the suspended seedling and the fruit now alone -depends on a slight bond between the base of the plumule and the inner -margin of the cotyledonary neck, as indicated by a cross in the figures -given in the plate. The union is soon broken and the seedling falls. - -Whether there is anything more than an analogy between the expulsion of -a Rhizophora seedling and the birth of a mammal seems most unlikely; but -the process is at all events a very remarkable one. - - - _The means of dispersal of the genus Rhizophora_ - -My experiments and observations were for the most part made on the -Asiatic and American species in Fiji; but I enjoyed the opportunity of -confirming some important points on the coast of Ecuador. We can only -look to the currents for the explanation of the capacity of the genus to -cross tracts of ocean; but, given this capacity, there is much that is -difficult to understand in the distribution of the genus and of a -species like Rhizophora mangle; and it is probable that we shall have to -look behind the means of dispersal to a distant age in the distribution -of shore-plants of the mangrove type. - -When Schimper published his work on the Indo-Malayan strand flora in -1891, but little was known of the duration of the floating capacity of -Rhizophora seedlings (p. 166). In giving the results of my -investigations I am merely describing the agencies of dispersal at -present in operation. Such agencies have their limitations, and we may, -perhaps, be thus able to explain why Rhizophora is restricted in the -Pacific islands to the archipelagoes of the Western Pacific; but many -serious objections would at once present themselves if we regarded the -occurrence of the genus in America, as well as in Asia and Africa, as a -matter depending on capacities and means of dispersal. - -The fruits of Rhizophora, as they display themselves before the -protrusion of the germinating seed, have no buoyancy, and the -germinating fruits until the hypocotyl has protruded for some inches (6 -inches in the case of R. mangle) also sink in sea-water. With a further -increase in the length of the hypocotyl, the germinating fruit acquires -buoyancy; and when the seedling, usually 10 or 11 inches in length, -becomes detached from the fruit on the tree and falls into the sea, it -floats readily in 95 per cent. of the cases. Such seedlings occur very -commonly in the floating drift of the estuaries and out at sea both in -Fiji and in Ecuador. - -Out of five seedlings of the Asiatic species, Rhizophora mucronata, that -had fallen naturally from the tree, three were afloat and healthy after -eighty-seven days’ immersion in sea-water. Out of twenty seedlings of -the American mangrove, Rhizophora mangle, sixteen floated after ninety -days and four were afloat and healthy after one hundred and twenty days, -the greater number sinking during the fourth month. These results -indicate considerable powers of buoyancy, and go to show that extensive -tracts of ocean could be traversed by the floating seedling. - -It should, however, be observed that not all the full-sized seedlings -float. With Rhizophora mangle about 5 per cent. sink in sea-water and -from 20 to 50 per cent. sink in fresh-water; whilst with R. mucronata -the proportion of non-buoyant seedlings is rather greater. There would -thus appear to be a rather nice adjustment of the specific weight of the -seedlings to the density of sea-water. Generally speaking, they may be -seen floating vertically or steeply inclined in the fresh-water of -estuaries and horizontally in the sea. With the buoyant seedlings of -Rhizophora mucronata, as a rule, about 90 per cent. float horizontally -in sea-water, and about 70 per cent. float vertically or steeply -inclined in fresh-water. The same general rule applies to R. mangle, -whether in the rivers and seas of Fiji or in those of Ecuador. In those -cases where the seedling drops prematurely on account either of storms -and floods or of the depredations of a grub that frequently attacks the -fruit, this rule would not apply. One may frequently notice in Fiji -after heavy weather that seedlings detached prematurely, and often -carrying the fruit, are floating in numbers horizontally in the rivers. -In a few days, as a rule, the fruit-case becomes detached and sinks. - -It may be remarked that the horizontal position is much better adapted -for the safety of the seedling in transport than the vertical position. -In the last case the plumule, which protrudes above the water, would be -unable, as indicated in my experiments, to withstand the scorching rays -of the sun in a smooth sea; whereas in the horizontal position, which -the seedlings assume in sea-water, the plumule is more or less -completely submerged, and the risk of withering in the sun is very much -less. The Rhizophora seedlings would certainly have little chance of -crossing in safety a large tract of sea, if they floated, as they do in -river-water, with the plumule exposed above the surface. It is not -unlikely that the comparatively restricted area occupied by Rhizophora -conjugata may be due to the attitude its seedlings assume when floating -in sea-water. - -The stranded seedlings of Rhizophora readily establish themselves for a -while in very different situations; and it is by no means necessary that -they should be washed ashore on a muddy coast. When half-buried amongst -the heap of vegetable drift piled up on a sandy beach they are -frequently to be found striking into the sand and showing their first -leaves. Here they ultimately perish in the great majority of cases; but -when protected long enough to reach the moist sand four or five inches -below, they may give rise to a little mangrove colony. When caught in a -fissure in the bare reef-flats these plantlets are sometimes able to -establish themselves. Rhizophora seedlings would, however, require a -coast prepared by them by the work of ages before they could form -extensive swamps. It is, therefore, not surprising that Prof. Penzig -found no evidence of mangrove-settlements on the shores of Krakatoa -fourteen years after the eruption. - -Yet suited as Rhizophora seedlings are for crossing tracts of sea, I -regard them as quite unfitted for being transported by the currents -unharmed across an ocean. The plumular bud is insufficiently protected -for such a long voyage of many months, and perhaps of years. Though the -horizontal position of the seedling would secure the plumule against -being scorched in the sun, it increases considerably the risk of injury -from direct impact. - -As bearing on their capacity for dispersal in other fashions, it may be -remarked that Rhizophora seedlings can withstand long drying. Five which -had been kept dry for nine weeks, after having been found stranded on a -beach, were planted in the mud of a mangrove-swamp. In a fortnight two -of them were developing the first leaves and throwing out roots. As long -as they are protected by a covering of vegetable _débris_ and sand, the -stranded seedlings might retain their vitality for months. - - - BRUGUIERA RHEEDII (Blume) - -This species is reduced in Hooker’s _Flora of British India_ to -Bruguiera gymnorhiza (Lam.), and thus viewed it has a very wide range in -the Old World, corresponding very much to that of Rhizophora mucronata, -namely, tropical East Africa, tropical East Asia to the Liukiu Islands, -the Indian Archipelago, New Guinea, tropical Australia, and Western -Polynesia, as in New Caledonia, Fiji, Tonga, and Samoa. There are four -or five species of the genus, but all are confined to the Eastern -Hemisphere, none occurring in America. - -As with the species of Rhizophora, this plant is indebted for its -present dispersal to the floating seedling, which, however, often falls -from the tree whilst still attached to the fruit, but is generally freed -in a day or two. The seedlings float for a long time in sea-water. I -kept one of them afloat for 117 days, when it was quite sound and -healthy. They appear to be better fitted than the species of Rhizophora -for the “rough-and-tumble” of ocean transport, since the plumule is much -less prominent, projecting only one line (2·5 mm.) or less, whilst with -the two Fijian species of Rhizophora the plumule measures from seven to -twelve lines (18 to 30 mm.). In the latter part of the year they are to -be found in abundance in the floating drift of rivers, and there they -readily develop the first leaves and roots. They are also frequent in -the sea off the coasts, and they are stranded in large numbers on the -beaches, where they readily strike into the sand when partially buried -amongst the vegetable drift. - -The empty flowers and the germinated fruits containing the cotyledons -are very common in floating drift. They look much alike, but the flowers -are much smaller and possess the long style, whilst the fruits contain -the cotyledons at the bottom of the seed-cavity. - -As with Rhizophora, there is a rather curious adjustment of the buoyancy -of the seedling to the density of sea-water. About 75 per cent. of those -afloat in the fresh-water of rivers assume the vertical position, the -plumular end protruding between two and five lines (5 to 12 mm.) above -the surface, while the remainder float horizontally or nearly so. In -sea-water about 50 per cent. float either vertically or steeply -inclined, and the other half float horizontally. - -With regard to the times of flowering and fruiting, it may be remarked -that the trees are mostly in flower during the hot months from November -to February, and that the fruiting is in active operation in the latter -half of March. The floating seedlings occur in abundance in the -river-drift at the end of the year, a circumstance which corresponds -with the fact that a period of six months passes between the -fertilisation of the ovule and the fall of the seedling into the water. - -Fertilisation, or, more correctly speaking, the discharge of the pollen, -takes place after the opening of the flower, and not before, as in the -case of the species of Rhizophora. The flower-bud is at first erect, but -subsequently it begins to bend downwards, and ultimately it hangs more -or less vertically. The provision to secure fertilisation under these -circumstances is rather curious. Without some such contrivance as is -below described, the pollen would merely fall out of the flower. Each -petal has its sides rolled or folded inwards so as to completely inclose -two stamens. In the bud the folded petals are white and flexible, but as -the flower expands they redden and become dry and elastic, and are only -prevented from flying open with a spring by the interlocking of the -hairy tips of their lobes. Whilst the folded petals are becoming stiff -and elastic during the opening of the flower, the inclosed stamens are -at the same time preparing themselves for their function. The anthers -are dehiscing and the filaments are acquiring elasticity. All is now -ready, and a slight shake or a touch puts the mechanism into action. The -petals unfold themselves with a spring, and the stamens thus suddenly -exposed and released fly forward, and a little shower of pollen is -thrown towards the centre of the flower. This process is accomplished in -ordinary fine weather during the first twenty-four or thirty-six hours -after the expansion of the flower. When the opening occurs in the early -morning, half of the stamens will be found released in the evening and -the rest on the following day. During the next day or two the petals and -the stamens fall out of the flower. In wet weather, the petals never -acquire elasticity, and in consequence do not unfold. In this case -pollenisation is never effected, and the folded petals soon fall to the -ground, carrying the stamens within them. Cross-fertilisation would be -much more likely to occur with species of Bruguiera (if, as is probable, -the same process of pollenisation is usually followed) than with species -of Rhizophora, since the stamens are securely inclosed in the petals for -some hours after the expansion of the flower. - -Nearly eight weeks pass between the date of fertilisation and the -commencement of germination. This is somewhat similar to the period -given for Rhizophora mangle, namely, nine weeks, and it obviously leaves -little or no time for any stage of quiescence or dormant vitality in the -case of the seed. The changes which the fruit undergoes in this interval -are a considerable increase in girth and a thickening of the calycine -walls, together with a contraction of the mouth of the tube. However, I -found no method sufficiently accurate for recording the rate of increase -of the fruit. - -It is known that germination is in progress when the end of the -hypocotyl begins to lift up the lining membrane at the bottom of the -calycine tube (see Figs. 21 to 26). The floor of the tube begins to -bulge up, but since this cannot be well seen at first, a better index is -afforded in the elevation of the style which accompanies it. The top of -the style preserves previous to this time a constant level with regard -to the tips of the calycine teeth. But this does not indicate the actual -beginning of germination. As shown in Fig. 21, the seed lies about two -and a half lines (6 mm.) below the floor of the calycine tube, and the -tip of the hypocotyl has to penetrate the intervening tissues before it -can push up the lining membrane and raise the style. Judging from the -subsequent rate of growth, seven or eight days at least, and perhaps as -much as two weeks, are requisite for this purpose. It is not necessary -to give further details here, and it may be at once stated that the -average of numerous observations on the length of the interval between -fertilisation and the elevation of the style was sixty-four days, the -range being fifty-nine to sixty-nine. After deducting ten days for the -time occupied for the radicle in reaching the floor of the calycine tube -(see Figs. 22 and 23), we obtain, as already remarked, nearly eight -weeks as the time elapsing between fertilisation and germination. - -The radicle or hypocotyl, therefore, in the first stage of germination -pierces the tissues above it and reaches the floor of the calycine tube. -It does not, however, pierce the lining membrane of the tube but pushes -it upward until it ruptures about 4 millimetres below the base of the -style which is carried up with it. Thus a kind of cap is formed, as -shown in Fig. 24, which does not fall off from the end of the hypocotyl -until it has protruded rather more than an inch. The hypocotyl attains a -length varying between 5 and 11 inches, the average being about 8 -inches. - -The whole period may be thus divided up:— - - (1) Period between fertilisation and germination 7-1/2 - weeks. - - (2) Period between the beginning of germination and the 1-1/2 - protrusion of the point of the hypocotyl at the floor of the weeks. - calycine tube - - (3) Period occupied in the growth of the hypocotyl 8 inches 18 weeks. - outside the fruit and terminating in the fall of the - seedling - - —— - - Total 27 weeks. - -The total period of twenty-seven weeks between fertilisation and the -fall of the seedling is thus six weeks shorter than that estimated for -Rhizophora mangle. On comparing the two tables it will be seen that the -difference mainly lies in the length of the second period, namely, that -between the commencement of germination and the protrusion of the -hypocotyl from the fruit. With Rhizophora mangle the fruit grows -considerably in length during this period of the germinating process. On -the other hand with Bruguiera rheedii there is, during this period, -practically no increase in the length of the fruit, and the radicle has -only to penetrate the tissues, 2-1/2 lines in thickness, between the -seed and the floor of the calycine tube. - -In the mode of separation of the seedling there are very marked -differences between this species of Bruguiera and the species of -Rhizophora. With Bruguiera rheedii the four small cotyledons, which are -united at the base, are, however, left behind at the bottom of the -seed-cavity, when the seedling is detached. But there is no expulsion of -the seedling, the connection being ultimately severed at the contracted -base of the cotyledons inside the fruit. When the seedling is full-sized -the nutritive supply begins to fail, and in consequence the pressure of -the sides of the fruit on the inclosed plumular end of the seedling -becomes slacker, the union with the cotyledons becomes weaker, and the -connection of the fruit with its peduncle at the basal joint becomes -slighter. Usually the fruit falls before the seedling is ready to drop -out, and the connection is severed after a few days’ flotation in the -water; but sometimes the union between the seedling and fruit is weaker -than that between the fruit and its peduncle, and in that case the -seedling falls and leaves the fruit containing the cotyledons on the -tree. The whole process of separation is much simpler than with species -of Rhizophora. Here it is mainly a matter of the failure of the -nutritive supply, whilst with Rhizophora it is almost a process of -parturition. - -Haberlandt, in the memoir before quoted, describes quite a different -mode of detachment in the case of Bruguiera eriopetala. Here the -seedling falls normally whilst still attached to the fruit, and the -separation is subsequently effected by the expansion of the mouth of the -calyx-tube due to the swelling of the “endosperm-neck” from the entrance -of water. - - - _Summary_ - -(1) There are four typical mangroves of the Rhizophoraceæ in Fiji, -Bruguiera rheedii, Rhizophora mucronata (the Asiatic species), -Rhizophora mangle (the American species), and the Selala, a seedless -form intermediate between the two species of Rhizophora just named, but -nearest to the Asiatic species. - -(2) It is shown that the sterility of the Selala is connected with the -impotent character of the pollen; and since the ovules appear capable of -fertilisation this is held to indicate that cross fertilisation has not -been in operation in producing the barren form. - -(3) Good reasons are given for the belief that the Asiatic species of -Rhizophora is the parent of the Selala, not as the result of a cross -between the Asiatic and American species, but as connected with -dimorphism, the Asiatic species producing two kinds of offspring, one of -them with impotent pollen. - -(4) In support of this view it is pointed out that there are two forms -of Rhizophora mangle in Ecuador, one of which comes near the Fijian -Selala, though producing seed. There could thus be no question of -crossing, since but one species occurs there. - -(5) The Selala reproduces itself in a vegetative fashion when growing, -as it often does, in an inclined position. The parent trunk dies and the -primary branches supported by the aërial roots, remain alive and in -their turn give rise to secondary branches similarly supported. - -(6) Although, as a rule, only one of the four ovules of Rhizophora -becomes a seed, occasionally a fruit contains more than one seed. With -R. mangle in Fiji about one per cent. of the germinating fruits -displayed more than one hypocotyl. - -(7) As a result of a protracted series of observations in Fiji, it was -established that in the case of a seedling of average length of -Rhizophora mangle a period of thirty-three weeks elapsed between the -date of fertilisation of the ovule and the detachment of the seedling -from the tree. In the instance of R. mucronata it was placed at -forty-two weeks. A period of thirty-eight weeks, or nine to ten months, -is regarded as typical for the genus. - -(8) It is established that normally there is no rest-period for the seed -in the case of Rhizophora, the seed at once beginning to germinate on -reaching maturity. In those exceptional instances, however, where there -is more than one seed, it is shown that in some cases the seeds do not -begin to germinate together, and that a rest-period of at least some -weeks can be at times postulated for one of the seeds. - -(9) An analogy exists between the process of expulsion ending in the -detachment of the seedling of Rhizophora from the fruit and the process -of parturition. - -(10) Experiments show that Rhizophora seedlings can float unharmed in -sea-water for a period of at least three or four months. Though -nine-tenths or more float in sea-water, as much as a fourth or a half -sink in fresh-water. As a rule they float vertically in fresh-water and -horizontally in sea-water, the horizontal position safe-guarding the -plumule against the risk of being withered up by the sun in a calm sea. - -(11) It is shown that in the case of Bruguiera rheedii the seedlings -when detached from the tree can float unharmed in sea-water for months. -In their specific weight they display a similar fine adjustment to the -density of sea-water, as is above described in the case of Rhizophora. - -(12) With this species of Bruguiera, fertilisation takes place not in -the unopened flower, as in Rhizophora, but after the flower’s expansion; -and a very singular mechanism is here described which secures the -completion of the process. - -(13) A period of twenty-seven weeks elapses between the fecundation of -the ovule and the detachment of the seedling from the tree in the case -of Bruguiera rheedii; and it is shown that there is normally little or -no room for any rest-period, and that, as with Rhizophora, the seed on -reaching maturity begins to germinate. - -(14) Though the seedlings of Rhizophora and Bruguiera could be -transported in safety a few hundred miles across the sea, it is held -that they could never cross the Pacific and reproduce the plant. That -the American species of Rhizophora has reached the Western Pacific from -the New World is not accepted. Rather is its present distribution -regarded as representing its original wide range over much of the -tropical zone. - - - - - CHAPTER XXXI - - A CHAPTER ON VIVIPARY - -The significance of vivipary.—The scale of germinative capacity.—A lost - habit with many inland plants.—The views of Goebel.—The shrinking in - the course of ages of tropical swamp areas.—The variation in the - structures concerned with vivipary.—Abnormal vivipary.—Summary. - - -IT was remarked in Chapter IX that the study of the germination of the -floating seed carried us to the borderland of vivipary; and we may now -observe that our study of the mangroves, Rhizophora and Bruguiera, in -the previous chapter, has brought us into contact with vivipary in its -most complete development in the tropical swamps of our age. There is a -great gap between the two extremes, represented by the occasional -germination of a seed in a capsule or in a berry on the plant, and by -the elaborate process of vivipary exemplified by Rhizophora; but most of -the intermediate stages can be illustrated by known examples of -vivipary. There is, however, no pretension to deal with this subject -here in anything but a cursory fashion; but it will, I venture to think, -add completeness to a work in which germination on and off the plant has -been such a frequent theme if I endeavour to connect together some of -the various sets of facts known to us concerning germination from the -standpoint of vivipary. - -The principal argument here followed has been already outlined in -Chapter IX, where I have remarked that it is possible to construct a -scale of the germinative capacities of plants, presenting a continuous -series beginning with the mangroves, where germination takes place on -the tree, and ending with those numerous inland plants where seeds are -liberated in an immature condition. It is suggested that vivipary was -the rule under the uniform climatic conditions of early geological -periods, and that with the differentiation of climates that has marked -the emergence of the continents the viviparous habit has been lost over -much of the globe, the mangrove-swamps alone illustrating the climatic -conditions once prevailing. The rest-period of the seed is regarded as -an adaptation to climatic differentiation and to seasonal variation; and -even the seed-stage may be broadly regarded as the price paid for -adaptation on the part of the evolutionary or determining power that -lies behind plant-development. When discussing the germination of -Cæsalpinia in Chapter XVII, I have shown that the contraction and -induration of the seed-tests appear merely as an adaptation to climatic -differentiation and to seasonal variation, and that it would be quite -possible by exposing the maturing seed to very warm and moist conditions -to induce germination without any rest-period, as actually occurs with -Rhizophora. One would then dispense altogether with the final processes -of the contraction and induration of the seed-coats, as illustrated in -the Leguminosæ; and the rest-stage would appear as an adaptation to -secular differentiation of climate in the later epochs of the world’s -history. - -The significance of occasional vivipary was long ago pointed out by -Goebel in his _Pflanzenbiologische Schilderungen_ (teil I., 117-134, -Marburg, 1889), when he observed that vivipary, as displayed in the -mangroves, and particularly in the Rhizophoreæ, represented the fullest -expression of a habit that is only occasionally exhibited by other -plants under exceptionally moist conditions. His view was that the seeds -of plants living in wet places are suited in a varying degree for rapid -germination, and that vivipary presents itself as the most complete -development of this capacity. If I regard the views of Goebel and of -Kerner aright, vivipary as normally developed in the mangrove is to be -traced in a descending scale to small beginnings, the principal -determining condition lying in the great difference that exists amongst -plants in the readiness of the seed to germinate. In the ascending scale -we would have first the detachment of the immature seed, where the -embryo is often in a rudimentary state, the ripening of the seed taking -place in the soil. Then would come those plants where the seeds on being -detached are quite mature and are ready to germinate as soon as they -fall to the ground. Then would follow the stage represented by those -plants where the seeds merely begin to germinate on the plant, such as -occurs more or less normally with some mangroves like Laguncularia, and -abnormally with a number of plants living in drier stations. After this -come those mangroves, where, as in Avicennia, germination is completed -on the tree or shrub, but the seedling at once liberates itself from the -parent. Last of all there is the stage of the typical mangroves, -Rhizophora and Bruguiera, where the seedling remains for months growing -on the tree and hangs from the branches. - -Vivipary, as above stated, presents itself as a matter of small -beginnings. My own view, however, is that it is a matter of small -“endings”; and that if we were to commence the scale not with the -immature seed lying on the soil, but with the seedling suspended from -the branches of a Rhizophora tree, we should record the various epochs -in the history of vivipary throughout the plant-world. From this -standpoint the occasional cases of incomplete vivipary displayed outside -the mangrove-swamp represent a lost habit belonging to a primeval period -when the climatic conditions were uniform over most of the earth, an age -almost of eternal gloom, when the air was ever saturated with aqueous -vapour, and when the sun’s rays were screened off by a dense -cloud-covering that enveloped the globe, an age of which the existing -mangrove swamps alone afford an imperfect indication. Yet even now we -can say with Schimper that “dense and frequently repeated cloudiness -apparently represents the most essential climatic condition for the -occurrence of mangrove in the tropics” (_Plant Geography_, p. 409). - -But, to return to the subject immediately under consideration, if my -view is correct we ought to find indications of the lost habit in the -anomalous structure of the seeds of some inland plants; and, indeed, it -is shown in Note 50 that this view can be taken of the singular -structure of the seeds of the Myrtaceous genera, Barringtonia and -Careya, and of the genera of some other orders, and can be extended by -implication to several other plants possessing similar seed-structures. - -With regard to the subject generally, it may be remarked that although -normal vivipary is mainly restricted to the plants of a mangrove swamp, -by no means all mangrove plants are typically viviparous. This habit in -its most complex form is exhibited as a rule by plants with firm, -somewhat fleshy, usually one-seeded, indehiscent fruits, such as we find -with Rhizophora and Bruguiera; but plants with follicular fruits, such -as occur with Ægiceras, may also display it in a fashion nearly as -complex. Generally speaking, however, plants with hard, dry fruits, such -as are owned by Excæcaria, Heritiera, and Lumnitzera, are -non-viviparous, though to all appearances quite at home in a -mangrove-swamp. Others again, like Carapa, Laguncularia, and Nipa, -whilst displaying vivipary in a varying degree, in some cases as a -general rule, in others only occasionally, exhibit no special structures -connected with it. This point is well brought out by Schimper in his -work on the Indo-Malayan strand-flora (p. 43), and no further mention -need be made of it here. - -The structures connected with vivipary vary greatly in their degree of -specialisation. At the one end of the scale we have highly complex -structures, such as are described in the preceding chapter. At the other -end we have those cases of occasional germination on the parent plant -where there is seemingly no special structure of any sort. That the -complex arrangements concerned with the vivipary of Rhizophora, -Bruguiera, Ægiceras, and Avicennia are adaptations is argued by -Haberlandt and Schimper, both of whom devoted much attention to the -study of these plants. This is seemingly indicated by the circumstance -that complex structures concerned with vivipary are found in plants so -divergent in their characters (the four genera above-named representing -three orders, Rhizophoreæ, Myrsinaceæ, and Verbenaceæ) that they only -possess their stations in common. It does not, however, follow that all -mangroves that exhibit a complex form of vivipary are of the same -antiquity. I should be inclined to regard those of the Rhizophoreæ as -the more primitive types, whilst it is possible that plants of other -orders, though ancient denizens of a mangrove-swamp, may be more recent -intruders into the mangrove-formation after the differentiation of a -dry-land flora. - -Of particular interest in this connection are the cases of abnormal -vivipary, or of “precocious germination,” that have been recorded from -time to time respecting a number of plants not denizens of a mangrove -swamp, none of which would appear, according to Schimper’s views, to -present anything of the nature of an adaptation. Goebel mentions a -number of instances, such as that of wheat-grains germinating on the -stalk in a wet summer, and that of Dryobalanops camphora, the Borneo -camphor-tree, when during a prolonged wet season in Java the seed -germinates in the fruit on the parent tree. Amongst other examples he -cites the Cacti, Epilobium, Agrostemma, and Juncus, the last case coming -also under my observation in a wet season in England. One may here -notice the instance of Dracæna, of which Mr. Hemsley, in April, 1902, -exhibited at a meeting of the Linnean Society of London a specimen -showing the seeds germinating in the berries on the plant. - -Several cases of this kind came under my notice in Fiji. Pulpy fruits -rather favour the precocious germination of seeds. Thus I sometimes -found the seeds germinating in the Mandarin orange and in the Papaw -fruit (Papaya) shortly after they had been gathered. But more -interesting examples were displayed in those instances where the seed -was found germinating on the plant. When the Convolvulaceæ grew in wet -situations, as on the borders of a mangrove swamp, the seeds were -sometimes observed germinating in the capsule. This came under my notice -with Ipomœa glaberrima (Boj.) and with I. peltata, more particularly in -wet weather. With some other plants, like Hibiscus diversifolius, that -grow in wet places, this at times occurs. A species of Croton, employed -as a support for the Vanilla plants in a plantation near Suva, displayed -seeds germinating on the plant. I was informed that the seeds of the -common cultivated Luffa (L. cylindrica) growing in a garden on Vanua -Levu sometimes germinated in the fruit still attached to the parent. It -is possible that the seeds of the parasitical genus, Myrmecodia, may -occasionally germinate on the plant, since I found them germinating -inside some of the small berries that had been lying forgotten within a -newspaper for a fortnight. - -Perhaps the most curious case of abnormal vivipary observed by me in -Fiji was that concerned with the Coco-nut palm. Though not known to many -residents in the island, this habit was described to me by Mr. Matthew -Simpson, a planter on Vanua Levu, who told me that he had noticed nuts -germinating on the tree in unusually dry seasons. Coco-nut palms -displaying the nuts germinating on the tree came under my observation -near Bale-bale, Savu-Savu Bay. In these cases the mature fruit, instead -of falling, remains attached and dries on the stalk. In one case the -seedling was about eighteen inches high. This seems to be what takes -place normally according to Blume with Nipa fruticans, the swamp palm of -Indo-Malaya. Goebel quotes this author to the effect that the fruits are -not separated from the head before germination is so far advanced that -sea-water can no longer injure the seedling. The fruits, we are told, -may remain for years attached in a state of incomplete germination. - - - _Summary_ - -The scale of germinative capacity, that begins with the seedling hanging -from the branches of a mangrove like Rhizophora and ends with the -detached immature seeds of many inland plants that only germinate after -lying for some time in the soil, is regarded as supplying a record of -the various epochs in the history of vivipary throughout the -plant-world. In the occasional cases of incomplete vivipary occurring -among inland plants and in the singular structure presented by the seeds -of certain genera of the Myrtaceæ and other orders we perceive -indications of a lost viviparous habit belonging to a primeval period -when vivipary was the exception and not the rule, an age when the same -climatic conditions prevailed over much of the globe. At such a period -the sun’s rays were screened off by a dense cloud-covering that -enveloped the earth, and the atmosphere was ever charged with moisture. -With the differentiation of climate that has marked the emergence of the -continents during the secular drying of the earth, the viviparous habit -has been alone retained within the confines of the mangrove-swamp, where -the conditions once almost universal now survive; and as an adaptation -to the differentiation of climate and to the resulting seasonal -variation the rest-period of the seed has been developed. - - - - - CHAPTER XXXII - - THE WEST COAST OF SOUTH AMERICA - -The littoral floras of the West Coast of South America.—The Convolvulus - soldanella zone of Southern Chile.—The plantless or desert zone of - Northern Chile.—The Sesuvium zone of Peru.—The Mangrove zone of - Ecuador and Colombia.—The two varieties of Rhizophora mangle, the - “mangle chico” and the “mangle grande.”—The floating vegetable drift - of the Guayaquil River.—The Humboldt current and the climate of the - West Coast of South America.—The advance northward of the arid - climatic conditions of the Peruvian sea-border.—The retreat of the - mangroves.—Evidence of ancient coral reefs on the coast of Peru.—The - shore plants and stranded seed-drift of the Panama Isthmus.—Summary. - - -MY acquaintance with the strand-flora of the west coast of South America -began at Corral, the port of Valdivia, in Southern Chile in lat. 40° S., -and terminated at the mouth of the Guayaquil River, in Ecuador, about 2° -south of the equator. During the period December 23, 1903, to March 17, -1904, I examined the coast plants at sixteen localities in this region, -which covers 38 degrees of latitude and thus measures about 2,300 miles. -Travelling in a steamer to Callao that was trading on the coast I had -opportunities of staying for periods ranging from half a day to a couple -of days at a considerable number of places; and a week spent at -Valparaiso gave me a good opportunity of examining the beaches north and -south of it. At Lima I spent some weeks, and from that centre examined -the shore-plants at Callao, Ancon, and Chancay to the northward. North -of this I had not the same opportunities, until we passed the Peruvian -and Ecuadorian boundary; but from a visit to the shore at Paita, from -the general look of the country in places as we coasted along, and from -information derived from other sources, I was able to obtain a fair -general idea of the prevailing character of the beach plants. After my -previous experience to the southward, one could fairly gauge the -character of the beach-flora from the appearance of the land behind. In -the Gulf of Guayaquil and in the vicinity of the city of that name I -spent about three weeks in the investigation of the coast flora. - -If it were not for the interposition of the great rainless deserts of -Northern Chile and for the scantily vegetated, scantily watered and -semi-sterile condition of almost the whole coast of Peru, the botanist -would be presented with a splendid opportunity of studying the -distribution of shore-plants along a meridian stretching through some -fifty degrees of latitude from Patagonia to Ecuador. As it is, drought -and sterility in one form and another reign over about half of this -great stretch of continental coast. This is reflected in the -beach-flora; and though the observer will often have his interest -attracted by the wonderful climatic anomalies arising from the presence -on the coast of the cold Humboldt current, to which the sea-border of -North Chile owes its desolation and the coast of Peru its -semi-sterility, yet for a long time he will feel as if Nature had hardly -dealt fairly with him. - -[Illustration: - - THE WEST COAST - OF - SOUTH AMERICA - - John Bartholomew & Co., Edin^r. -] - -Along the sea-border corresponding to the deserts of North Chile there -would seem to be practically no plants growing on the beaches, except -here and there where some stray plant from the saline districts inland -intrudes on the coast. Along the whole sea-border of Peru from Arica -north to Tumbez on the borders of Ecuador, the coast-districts, though -more or less rainless, receive the benefit of the drizzly garuas and -sea-fogs, and the sterility of the land immediately backing the beaches -is much less pronounced than with the sea-border corresponding to the -deserts of Northern Chile. This difference shows itself in a peculiar -type of littoral vegetation, a strand-flora that is very scanty but one -where on the beaches Sesuvium prevails. North of Tumbez the -mangrove-formation predominates along the sea-borders of Ecuador and -Colombia to Panama, excepting on a stretch of sterile coast extending -north from the Gulf of Guayaquil to the equator. - -Though in one sense the botanical observer will be disappointed with the -littoral floras of the west coast of South America, in another sense -when he remarks the manner in which the coast-vegetation reflects the -abrupt changes in the prevailing climatic conditions he will be -fascinated by the interesting problems presented to him. We are -accustomed to connect a tropical coast with mangroves, coral-reefs, and -beaches of calcareous sand supporting a luxuriant littoral flora. -Climatic conditions banish all these from the tropical west coast of -South America until within four degrees of the equator, and then with -startling suddenness the dominion of the mangrove begins, the -neighbouring hills commence to be clothed with tropical jungle, and the -climate is completely changed. Mr. John Ball, who sailed along this -coast about twenty years ago, referring to this remarkable phenomenon on -the borders of Peru and Ecuador, remarks that no such abrupt and -complete change both in climate and vegetation is known elsewhere in the -world, and he adds that few parts of the American coast better deserve -careful examination (_Naturalist in South America_). This subject has -since been discussed at length by Dr. Wolff in his “Geografia y -Geologica del Ecuador,” and by Baron von Eggers in a paper to be -subsequently quoted, two very competent observers, but the latter -considers that the subject still requires a systematic investigation, -and suggests that an observing station should be established on this -coast by the combined meteorological societies of Europe. A sojourn of -more than a week in the swamps at Puerto Bolivar, a few miles from -Tumbez, enables me to appreciate the nature of the problem, and to throw -a little light on the line of investigation required. - -But to return to the general subject of the littoral floras of the west -coast of South America, I may say that beginning with the island of -Chiloe in lat. 42° S., this coast may be divided into four zones. - -(1) The Convolvulus soldanella zone of Southern Chile, which extends as -far north as Coquimbo about 30° S. lat. - -(2) The Plantless or Desert zone stretching north to the vicinity of -Arica in lat. 18° 30ʹ, and corresponding to the coast of Northern Chile. - -(3) The Sesuvium zone, extending north from Arica to the 4th parallel of -south latitude in the vicinity of Tumbez, a sea-border of semi-sterility -that comprises the entire coast of Peru. - -(4) The Mangrove zone, stretching from Tumbez, on the frontiers of -Ecuador, to the equator and on to Central America, but interrupted at -first by a strip of sterility on the coast extending from the Gulf of -Guayaquil to the borders of Colombia, or, strictly speaking, to the -equator. - - - THE CONVOLVULUS SOLDANELLA ZONE (SOUTHERN CHILE). - -This zone, which answers to the coast of Southern Chile, from Chiloe as -far north as Coquimbo, corresponds to watered and vegetated inland -regions, in which, however, the amount of rain and the degree of -fertility decreases from south to north, that is to say, as we approach -the desert regions. Here we find none of the dry beaches that prevail -for twenty-five degrees of latitude north of Coquimbo. When we scoop -with our hands to a depth of three or four inches in the sand we find it -relatively cool and more or less moist, as in an English beach. In a hot -summer’s day on a Valparaiso beach we should find that the temperature -of the sand at the surface (half-inch deep) was about 112° F., and at a -depth of four inches about 80°. This would be above the average for the -zone, which would be probably near the typical summer-temperature of an -English beach, namely, 102° at the surface and 77° four inches down. -This subject of beach temperature is discussed in Note 70. - -Plants typical of the beaches of this zone, and evidently occurring over -the length of it, are Convolvulus soldanella, Nolana (paradoxa?); -Polygonum maritimum, Salsola Kali, and Selliera radicans. Nolana is a -Chilian and Peruvian genus. This beach plant, which is especially -abundant on the beaches near Coronel and at Bahia San Vincente, has the -creeping habit of its associate, the Convolvulus. However, it possesses -seeds, or rather seedvessels, of more limited buoyancy; and it is shown -in Note 71 that prolonged drying is needed for effective dispersal by -currents over great distances. This beach species of Nolana has narrowly -escaped being a widely-spread littoral plant; whereas it is now -restricted to the Chilian beach flora. Selliera radicans, a little -creeping Lobeliaceous plant, growing under the shade of tall clumps of -Juncus at the edge of the beach or in wet places where springs ooze out -in the sand, is a very interesting species that occurs also on the other -side of the Pacific in Australasia. Of the mode of dispersal of its -small seeds I know nothing, as the fruits were not ripe at the time of -my visit; but I would suggest that some resident botanist should -investigate this important point. I found it at Corral and at Coquimbo; -and Gay speaks of it as growing on wet coast places from Chiloe to -Coquimbo, a range of 12-1/2° of latitude. - -It is probable that all the shore-plants of this zone extend south to -Chiloe in latitude 42° S.; and it is likely that some of them reach -towards the Straits of Magellan. I did not find any of them within the -Straits on the beaches in the vicinity of Punta Arenas, where, however, -I noticed the three plants recorded by Ball, namely, Armeria maritima, -var. andina; Senecio candidans, also found in the Falkland Islands; and -Plantago maritima; besides a Chenopodiaceous plant not in fruit. The -Plantago has no capacity for dispersal by currents, and probably none of -the other plants are thus dispersed. I formed the opinion when in the -Straits that the beach plants on the Pacific and Atlantic coasts of -Patagonia could have but little communication by the currents, and that -they are in this respect quite cut off from each other. A botanist who -investigates the strand-flora of Patagonia and Tierra del Fuego in -connection with the littoral plants of the opposite coasts ought, if he -has not already done so, to obtain some very interesting results from -the standpoint of plant-dispersal. - -The northern limit of the plants of this zone near Coquimbo, in lat. 30° -S., is not determined by the change in climatic conditions that goes -normally with decrease in latitude, but by the vicinity of the great -deserts of Northern Chile, the aridity extending to the beaches. - -Amongst the other plants occurring generally in the Convolvulus -soldanella zone of Southern Chile, species of Salicornia and Samolus are -to be observed in wet places. On the beaches near Valparaiso and in the -vicinity of Talcahuano there thrives a species of Franseria, a Composite -plant possessing prickly fruits well suited for conveyance in bird’s -plumage, but not adapted, as shown in Note 71, for dispersal by -currents. Mesembryanthemum is a typical beach-plant at Coquimbo, and an -intruder from the adjoining hill-slopes at Valparaiso. Raphanus, -seemingly R. maritimus, occurs in places, but apparently only as an -intruder from the cultivated districts behind the beaches. One or two -species of Euphorbia are not uncommon. A few small trees or bushes of -Acacia farnesiana grow typically on the beach at Coronel and in -neighbouring sandy tracts at Talcahuano, though the plant, as Gay -observes, has been introduced. Sophora tetraptera, found also in New -Zealand, and one of the most interesting plants of the Antarctic flora, -thrives as a small tree on the hill slopes overlooking the harbour of -Corral, becoming bushy where in places it intrudes on the beaches, and -fruiting there as freely as on the slopes above. It was by testing the -buoyancy of the seeds of this plant that I was led to the discovery of -its mode of dispersal by the currents (I am indebted to Mr. Holland for -the specific determination of the fruits sent by me to the Kew Museum). -Other shore-plants, of course, occur in this zone; but I have gone far -enough to illustrate the subject. Of the numerous occasional intruders -from the neighbouring inland districts, frequently Compositæ, I say -nothing. The results of my observations on the floating power of the -seeds and seed-vessels of some of the shore-plants of this zone are -given in Note 71. - -Stranded seeds and fruits that belong to the proper beach-drift are not -easily found on the beaches of Southern Chile, as they are often buried -in rubbish. Those most characteristic are seeds of Convolvulus -soldanella and drupes of Nolana (paradoxa?), both typical beach-plants -of the zone. Portions of Salsola Kali bearing mature fruits, as -described in Note 17, are also frequent. Seeds of Sophora tetraptera -were found on the beach of Bahia San Vincente, whither they must have -been brought by the Humboldt Current from the south, as I did not -observe the tree in the vicinity. On this beach, as well as at -Valparaiso, the prickly fruits of Franseria were abundant in the drift, -doubtless derived from the plants growing on the same beaches. In -addition we get as frequent components of the beach-drift materials that -mark the white man’s presence over much of the globe. Corks are widely -distributed over the beaches of the world; but on no coast have I found -them more numerous than on the Valparaiso beaches. Here we find Medicago -fruits, the empty stones of the cherry, the plum, and the peach, empty -filberts and other materials, all of which I have gathered on the shores -of the Straits of Messina and on English beaches. Amongst this medley we -find also Casuarina cones and fruits of Eucalyptus. Then we find special -indications of the New World in the pea-nut (Arachis hypogæa) and in the -abundant seeds of a huge pumpkin (Cucurbita), which is a favourite food -with the Chilian indigenes. These seeds are cited as an example of -futile buoyancy in Chapter XIII. - - - THE PLANTLESS OR DESERT ZONE (NORTHERN CHILE). - -This zone of the coast, which stretches north for some 700 miles from -Coquimbo to near Arica (30°-18°30ʹ S. lat.), corresponds to the great -desert region of North Chile. On the beaches of Antofagasta, Tocopilla, -and Iquique, which are situated in the midst of this zone, I found no -plants. This rainless sea border of barren mountains, presenting to the -eye of the traveller from the deck of a passing steamer nothing but rock -and sand, must be one of the most desolate coasts on our globe. It is -therefore not a matter for surprise that the beaches are of dry loose -sand in which the hand fails to find on scooping below the surface that -refreshing coolness which is the character of beaches in all latitudes -where the land is vegetated and a subsoil drainage seaward exists. Under -ordinary conditions the sensation of moisture in the sand a few inches -down is not produced by the mere proximity of the sea. On the -Antofagasta and Iquique beaches the temperature in the heat of the day -of the surface half inch ranged from 120° to 130° F., whilst four inches -down it was 95° to 100°, and no moisture was found by scooping five or -six inches down. On the Taltal beach, which lies towards the southern -end of the desert region, I noticed, besides a few plants of Suæda -fruticosa, two other species of the orders Santalaceæ and Nolanaceæ, -evidently intruders from the inland regions. Where the zone of extreme -aridity terminates at the north between Pisagua and Arica a few bushes -are to be seen on the hill-slopes behind the beaches. - -Very little seed-drift came under my notice on the beaches of the desert -zone. Here and there I found a few Medicago pods and some seeds of the -large pumpkin above noticed, but that was all. This is due as a rule to -the seed-drift being masked by an enormous amount of rubbish, mostly -brought from the south by the Humboldt Current. My walk for five miles -along the beaches immediately north of Antofagasta gave me an experience -in the way of stranded drift such as I have never met with on the -beaches of any other region. All the dead bodies of the Chilian coast to -the southward seem to have been stranded in the bend of Moreno Bay, on -the shore of which Antofagasta lies; and the air was tainted with -decaying flesh, the past being mixed up with the present in a most -unrefreshing fashion. Besides carcases of sea-lions, six feet in length, -sharks, dog-fish, and fish of many sorts, some of them dried up, others -in a state of putrefaction, there were dead penguins, dead pelicans, -dead sea-birds of other kinds, the bodies of horses, cattle, dogs, &c., -all preyed upon by the numerous vultures and skuas, and in some -localities by hungry-looking dogs of large size that took no notice of -me as they slunk along. The past was represented by great quantities of -bones that lay bleaching on the sand, with here and there a vertebra of -a whale, making in all quite a varied osteological collection. But this -was not all. Carcases of all sorts were drifting towards the beach. Here -a vulture, there a skua, there again a dog stood just beyond the -tide-wash looking keenly seaward; and by following the direction of -their gaze one could see that each had marked down a carcase slowly -drifting in. Now and then they would make a dash, scarcely waiting for -the new arrival to be washed up by the waves. But there was no -competition, since there was enough for all. - -Under such conditions my investigation into the seed-drift was out of -the question; but I saw what would be considered by some as more -interesting, namely, the dead of many latitudes piled up on the beach by -the Humboldt Current. - - - THE SESUVIUM ZONE (THE PERUVIAN COAST). - -This zone, which comprises the whole Peruvian sea-border from Arica in -18°30ʹ S. to the vicinity of Tumbez in about 3°30ʹ S., usually possesses -in its scanty littoral flora one or two species of Sesuvium, and in some -places Sesuvium alone occurs on the beach. The beaches here do not line -a region of almost complete aridity, as in the coast corresponding to -the great desert region of North Chile. Though here also scarcely any -rain falls, the sea-border receives the benefit of the “garuas” or -drizzling sea-fogs; whilst the region immediately behind the coast may -either be desert or semi-sterile during much of the year, or may be -scantily vegetated, or, as along the river-valleys, may display a -vegetation more fitting to the latitude. The general aspect, however, of -the coast of Peru is one of aridity; but there are probably few beaches -where a certain amount of subsoil drainage from the land sea-ward does -not exist. This is well exhibited at Ancon, north of Callao, where in -the most unlikely situations water is reached by digging wells; but in -spite of this the Sesuvium alone grows on the beach. The beaches -examined by me in the heat of the day in February, as at Mollendo and -Ancon, had much the same surface-temperature noticed in the preceding -month on the beaches of North Chile, namely, 120° to 130° F., and in one -place 135°; whilst at a depth of four inches the sand was rather cooler, -and instead of being between 95° and 100°, as on the Antofagasta and -Iquique beaches, it was here usually only about 90°. But it was only -occasionally that the sand felt at all moist at a depth of five or six -inches; and in this zone, therefore, only a few shore plants of a -peculiar type could be expected to find a station on the beaches, -excepting, of course, those localities where low marshy districts or -lagoons lie behind the beach. - -The beach plants of the coast of Peru as observed by me though usually -scanty, presented two types according to the character of the district -bordering the beach. I make no mention here of those local plants, often -belonging to the Compositæ, that as at Callao and Arica descend the -valleys to the beaches, or to those numerous introduced plants that -accompany cultivation, such as we find at Arica. In those coast -localities, as at Arica, Callao, and Chancay, where salt-water pools or -brackish lagoons lie behind the beach, or where a stream or a river -empties into the sea, Sesuvium portulacastrum, Heliotropium -curassavicum, and a Salicornia are to be generally noticed, and, as at -Callao, Batis maritima may also abound. On the Chancay coast, about 30 -miles north of Callao, there lies inside the shingle-beach a large -shallow lagoon of brackish water (spec. gr. 1·012) with extensive muddy -marginal flats, the temperature of the water at the edge being at -mid-day on Feb. 3rd, 90° F. In the water flourished Ruppia maritima, -which was also exposed in dead, dry, matted masses on the bordering -mud-flats. On these mud-flats grew Sesuvium portulacastrum, which near -the water’s edge was associated with a small species of Salicornia, -whilst further away from the water it was accompanied by Heliotropium -curassavicum. - -But the most typical beach-flora of the Peruvian coast is such as we -find on the dry beaches skirting the base of sand-covered or barren -hill-slopes such as occur at Mollendo, Ancon, and Paita. As at Ancon, -sand-covered hills and plains may extend miles inland, displaying here -and there lines of shifting sand-mounds or “medanos.” On such beaches we -may often find only a solitary plant, a species of Sesuvium which seems -to differ only in its larger flowers, its much larger leaves (2 inches -long), and its stout stems, of the thickness of the little finger, from -the ordinary Sesuvium portulacastrum. This seems to be the only plant -that can make its home on such beaches. At Mollendo, where there are -signs of desiccated pools behind the beach which are occasionally filled -with sea-water, the vegetation was of an intermediate character and more -abundant; and here grew Sesuvium portulacastrum, a tall Salicornia, and -Suæda fruticosa; whilst the commonest plant was a prostrate Nolanaceous -species with a handsome purplish flower. - -Excepting with the fruits of Batis maritima, and perhaps the buoyant -joints of Salicornia, scarcely any of the prevailing shore-plants of the -coast of Peru possess a capacity for dispersal by currents. In this zone -I rarely found any seed-drift on the beaches. Much rubbish, such as -roots of bamboos, however, may be brought down by the rivers; and where -the Humboldt Current strikes a bend in the coast we get a repetition, on -a smaller scale, of the scenes on the Antofagasta beaches. Ancon Bay, -for instance, receives much of the floating offal of the south. - - - THE MANGROVE ZONE (THE COASTS OF ECUADOR AND COLOMBIA) - -We come now to the mangrove zone which comprises, with the remarkable -exception of a long stretch of arid sea-border to the north of the Gulf -of Guayaquil, the whole remaining western sea-border of South America, -namely, the Ecuadorian and Colombian coasts. My own acquaintance with -this region is limited to the estuary of the Guayas or the Guayaquil -River and to the southern shore of the Gulf of Guayaquil; but I am able -to avail myself of the researches of Baron von Eggers, which cover the -entire Ecuadorian coast; and with Ecuador, therefore, I will bring this -brief sketch of the littoral flora of one side of a large continent to a -conclusion. - -The Ecuadorian coast, lying, as Baron von Eggers observes, between the -rainless and desert coasts of Peru and the “ewig grüne” coasts of -Colombia, may be regarded as a transition-area presenting very varied -and complicated conditions. With the cause of the remarkable contrasts -exhibited by the strand-flora, not only on the coast of Ecuador, but -along the whole west coast of South America through some forty-five -degrees of latitude from Patagonia to Colombia, I will presently deal. -Here it may be remarked in passing that the Humboldt Current has played -the determining part in producing the abnormal climatic conditions to -which these remarkable contrasts in the strand-flora of this coast of -the continent are mainly due. - -The mangrove zone, marking a more or less abrupt transition from a -region of drought and semi-sterility to one of humidity and rank -tropical vegetation, begins about lat. 3° 30ʹ S., that is, in the -vicinity of Tumbez, or perhaps nearer the boundary-line between Ecuador -and Peru in lat. 3° 20ʹ (see Note 72). Occupying the southern shore of -the Gulf of Guayaquil it extends up the Guayas estuary to Guayaquil and -rather beyond. But when we follow the coast of Ecuador northward from -the island of Puna towards Santa Elena Point, we come upon one of the -most remarkable phenomena presented on the west coast of South America. -The dry region begins again and the mangroves disappear; and these -conditions continue through about 2-1/2 degrees of latitude until we -reach the equator, when the mangrove zone soon recommences, and, as I -infer, continues northward without a break to the coast of Central -America. - -Dealing first with the mangrove districts of the south side of the Gulf -of Guayaquil and of the Guayas or Guayaquil estuary, we may observe that -probably in few localities of the globe have the forces of nature worked -more in unison to produce the conditions favouring the growth of the -mangrove. The reason why this particular locality has been thus favoured -will be discussed later on in this chapter. I may here observe that -Baron von Eggers was so struck with the exceptional features of the -mangrove-growth in this region that he was inclined to look for the -American centre of the genus Rhizophora, the prevailing mangrove, in the -estuary of the Guayas River. - -I will not enter into a detailed description of the mangrove-formation -of this coast, which has indeed been given by the German botanist; but I -will merely refer to the leading features such as they presented -themselves to me. In the first place, reference will be made to the -sea-border of the province of Eloro, where I spent nine or ten days, -making Puerto Bolivar, the port of Machala, my headquarters—a locality -about thirty miles east of Tumbez. Except in the Guayas estuary I have -never seen such a magnificent growth of mangrove. - -By following the line of light railway that runs about six kilometres -inland from Puerto Bolivar to Machala, the capital of the province, we -obtain a good section of the mangrove-belt, which may be thus described. -The mangrove-swamp proper extends about three kilometres inland. Whilst -the small variety of Rhizophora mangle (mangle chico) immediately fronts -the sea, Laguncularia grows on the islets close to the seaward margin of -the swamp. When we enter one of the numerous broad creeks that intersect -the border of the mangrove-belt we soon find ourselves in the true -mangrove forest, where prevail tall trees of Rhizophora mangle (mangle -grande) that rise to a height of 70 or 80 feet or more. Gloomy as the -depths of the swamp are, they acquire quite a funereal aspect, the -branches of the trees being draped with pendent Tillandsias. These long, -hair-like, tangled growths hang vertically from the branches of the -trees and may be 20 or 30 feet in length. In the rear of the zone of -tall mangroves we come upon a more open district of the swamp. The -forest proper gives place to a tract occupied by small trees of -Rhizophora, Laguncularia, and Avicennia, with here and there whole acres -occupied only by the shrubby Salicornia peruviana which attains the -height of a man. - - [_To face page 484._ - -[Illustration: - - Emery Walker sc. - Rough Plan of the Gulf of Guayaquil. -] - -(The main stream of the Humboldt current, as indicated by the arrows, - turns off to the north-west at Cape Blanco; whilst a small branch - crosses the mouth of the Gulf of Guayaquil and flows along the Ecuador - coast north of Santa Elena Point.) - -Here terminates the mangrove-swamp proper, and about three kilometres -from the sea it passes gradually into a region of extensive bare -mud-flats which are penetrated by salt-water creeks, two or three yards -across and a foot or two in depth, that are bordered by low and shrubby -Avicennias, the Salicornia bushes above noted, and dwarfed trees of -Rhizophora mangle only four or five feet high. These flats, which are -evidently only overflowed by the sea at the higher spring tides, were at -the time of my visits much sun-cracked and in some parts incrusted with -salt; but the mud was rather soft, and in places Sesuvium portulacastrum -and Batis maritima flourished in quantity on it. These mud-flats, about -two kilometres across, pass by degrees into the low-lying level district -known as the Machala plains, on which the capital of the province is -built. Here the soil is dryish, and, notwithstanding that it displays on -its surface when exposed to the sun a white saline efflorescence, a dry -jungle type of vegetation of the xerophilous character here thrives. I -noticed casually the Algaroba (Prosopis), a yellow-flowered Cordia, -cacti of the Opuntia and Cereus kinds, besides several small trees and -shrubs often thorny. - -These Machala plains, on account of the fine saline incrustation above -mentioned, are of much interest, since at a distance of six kilometres -from the coast they thus display on their surface the effect of -sea-water infiltration, their level above the sea being only a few feet. -We have seen the three stages of this infiltration landward of -sea-water: first, the mangrove-swamps daily overflowed by the tide; -second, the mud-flats behind them which are only overflowed by the -fortnightly spring-tides; third, the vegetated plains behind all, which -are sufficiently raised to be above the reach of the tides, but which -are nevertheless soaked with sea-water that displays its presence in the -salt left by evaporation on the surface of the soil. - -But another interesting point is here raised. At the back of the -mangrove-belt, in most parts of the world, we usually find a -particularly rank and luxuriant vegetation where the Scitamineæ often -take a leading part; whereas on the sea-border of this part of Ecuador -the mangrove-swamps pass gradually into arid saliniferous plains. With -this singular fact is to be associated the circumstance that we see here -in operation, some four or five miles from the coast, a process by which -great quantities of sea-salts are accumulating below the surface. This -may possibly be concerned with the origin of the great saline deposits -of Northern Chile. However this may be, there is some reason for -believing—and I understand that this is the opinion of Dr. Wolff, the -historio-geographer of Ecuador—that in the course of ages the tendency -will be towards an extension of the dry, sterile regions of Northern -Peru into Ecuador. This subject is referred to again in a later page of -this chapter. - -Whilst in this neighbourhood I made the ascent for some fifteen miles of -the Santa Rosa River, which opens into the sea near Puerto Bolivar. It -is a tidal estuary that has no proportion in size to the small river -that enters it. In its lower third we passed at first between long -mangrove-islands formed almost entirely, as viewed from the boat, of the -tall Rhizophora trees draped with Tillandsias, and presenting really a -magnificent spectacle. In the middle third we were penetrating into the -rear of the mangrove-belt. The giant swamp-fern (Chrysodium aureum) -abounded, and here and there we passed by a patch entirely held by the -large shrubs of Salicornia peruviana. The tall Rhizophora trees were -replaced by the short variety, the “mangle chico,” which ceased -altogether about ten miles from the mouth of the estuary, but probably -only about five miles from the nearest part of the coast. The water at -the place where the Rhizophora trees ceased was evidently quite fresh -during nine out of the twelve hours, being only salt in the latter part -of the rising tide. Above the mangroves, in the upper third of the -ascent, Hibiscus tiliaceus, with Chrysodium aureum, flourished on the -banks. The shallows at the margins were occupied by a considerable -variety of semi-aquatic and other plants, such as Pontederia (two -species); one of the Alismaceæ, with the flower and fruit of Sagittaria -and the leaves of Alisma; Typha, Polygonum, and an Amaryllid like -Crinum. Plants of Pistia and Pontederia floated in the stream. - -I have said enough to give a general idea of the composition of the -mangrove-belt of the Ecuador littoral, and will refer but briefly to the -mangroves and other river-side plants in the neighbourhood of the city -of Guayaquil, some forty miles up the Guayas estuary. As I have remarked -in Note 38, the water of the river off the city is usually quite fresh -except at high water; but the sea has much freer access to the channels -at the back of Guayaquil, where at high water the density was 1·014. In -these channels are displayed the typical mangrove formation, trees of -Rhizophora mangle bordering the water, whilst behind they are mingled -with Avicennia tomentosa and Laguncularia. On the banks of the main -river, where they are overflowed at high water, Anona paludosa was the -most frequent tree, being associated with the Rhizophora, Hibiscus -tiliaceus, and other trees. Above the city, Polygonum glabrum was -growing in dense masses at the river’s edge, whilst Pontederia and -Pistia flourished on the low muddy banks and floated in quantities in -the river. - -Before quitting the subject of the mangrove-formation of Ecuador, I will -refer shortly to the two varieties of Rhizophora mangle that here occur. -Baron von Eggers received the impression that the common type of this -species, a low tree bordering the coast, did not exist in Ecuador, such -a type as he says is characteristic of the West Indies and of Central -America, and, I may add, also of Fiji. The species he regards as -acquiring a new facies in Ecuador, where it exists as tall forest-trees, -branchless for half their height, and exhibiting other divergent -characters. However, I found that the common type of the species occurs -normally on the coast in the vicinity of Puerto Bolivar, thirty miles -east of Tumbez, a district above described. - -There are two distinct forms of Rhizophora mangle exhibited in the -mangrove-belt of the coasts around Puerto Bolivar. One of them, which -the indigenes name “mangle chico,” is a small tree, 10 to 15 feet high, -with useless timber, that immediately borders the sea, and, in fact, -largely forms the margin of the swamp, not only on its seaward side, but -also on the land side, where it passes into drier ground. The other, the -“mangle grande,” a tall tree reaching to 60 or 80 and sometimes perhaps -to 100 feet in height, composes the interior, and indeed the bulk, of -the mangrove-belt, and possesses a hard and durable timber much employed -in the district. - -Distinct as these two types are, it is not difficult to find -intermediate forms, and, in truth, in some localities they prevail. But -the interesting point is that this peculiar Ecuadorian type of the -species, a type that attracted the attention of the eminent German -botanist, comes near the “Selala,” the mysterious seedless Rhizophora of -the Fijian swamps—a subject fully discussed in Chapter XXX., where I -have compared the Fijian and Ecuadorian Rhizophoras. Both the “Selala” -of Fiji and the “mangle grande” of Ecuador are intermediate between the -American Rhizophora mangle and the Asiatic R. mucronata, resembling the -last in their inflorescence, but in other points approaching the -American species. The “Selala,” however, comes nearer to the Asiatic -tree, whilst the “mangle grande” comes nearer to the American tree. -Unlike the Fijian tree, that of Ecuador is not sterile, but matures its -fruit; and it displays no evidence of the vegetative reproduction so -characteristic of the “Selala.” - -Sandy beaches are not common on the mangrove-fronted shores of the south -side of the Gulf of Guayaquil. However, on the seaward side of the long -low mangrove island of Jambeli, on which the lighthouse is placed off -Puerto Bolivar, there is a long stretch of beach of whitish, mainly -non-calcareous, sand. The Coco palms behind the beach give the coast -quite the aspect of a Pacific island strand. Ipomœa pes capræ flourishes -on the sand nearest to the sea; and immediately behind, the beach is -more or less occupied by a Cyperus 2 to 3 feet high, and by Canavalia -obtusifolia. Further back grows a small Acacia tree, and behind it the -yellow-flowered Cordia tree of the district; and in the rear of all lie -extensive mud-flats, partly occupied by stunted bushes of Avicennia -tomentosa and by Sesuvium portulacastrum, which in their turn pass into -the mangrove-swamps. - -On account of the enormous amount of drift of all descriptions that is -carried to the sea by the Guayas or Guayaquil River, floating vegetable -materials are abundant in the Gulf of Guayaquil, and are thrown up in -quantity on the coasts of Ecuador. One of the most interesting -spectacles at Guayaquil is presented by the floating river-drift. Huge -tree-trunks and floating islets, the last-named ranging from 3 or 4 to -30 or 40 feet or more across, were, at the time of my visit in February, -being carried to and fro unceasingly in front of the city by the tide, -gradually making their way down the river, and ultimately reaching the -open waters of the gulf. Floating plants of Pistia were in abundance; -and their fate when they reached the sea must have been tragical. The -islets were exceedingly interesting; they were evidently formed of -materials lifted up bodily from the shallows at the margin of the river, -and then carried off in the stream. They were mainly composed of two -species of Pontederia and of Polygonum glabrum in the position of -growth; the first often in flower. Pistia and a variety of smaller -plants nestled among them, such as Salvinia, portions of Azolla, Lemna, -&c.; and in one islet I noticed, oddly enough, the growing rhizome of a -sensitive plant (Mimosa pudica). A great quantity of floating seeds -collect amongst the roots and stems of the plants composing the islets, -and here I obtained much of the smaller seed-drift. - -Most frequent in the floating drift of the river at Guayaquil were the -seeds of Anona paludosa, often in a germinating condition. The seeds are -liberated by the decay of the floating fruit, which was also common in -the drift. Amongst the larger materials were the seeds of Entada -scandens and of Mucuna; the empty seeds of the vegetable-ivory palm -(Phytelephas macrocarpa), the sound seed possessing no floating power; -the “stones” of Spondias lutea, L., as identified by Mr. Holland, of the -Kew Museum; the empty small nuts of several palms, including, -apparently, Oreodoxa, &c. Amongst miscellaneous materials were small -gourds, which are referred to in Note 47, and an occasional empty cacao -fruit. Smaller seeds were also abundant, and included those of Hibiscus -tiliaceus, Erythrina, Vigna, Ipomœa, and others. Carried into the river -from the neighbouring mangrove-creeks, where they abound, there were -floating seedlings of Rhizophora and Avicennia, fruits of Laguncularia -often germinating, and the seeded joints of Salicornia peruviana. - -There was of course, in addition, much that was strange in the floating -drift of the Guayas River, which received its contributions not only -from the river-side vegetation and the neighbouring mangrove-swamps, but -also from the interior mountain ranges culminating in Chimborazo, the -slopes of which are drained by its tributaries. I had several -opportunities of meeting the drift of the Guayas River in the open -waters of the Gulf of Guayaquil. Much of it is carried along the south -side of the gulf; and I picked up at sea, ten to twenty miles from the -mouth of the estuary, many of the things above enumerated, such as -Erythrina and Mucuna seeds, seeds of Hibiscus tiliaceus, the empty -vegetable-ivory seeds, the seedlings of Rhizophora and Avicennia, and -the germinating fruits of Laguncularia and Salicornia peruviana. Much of -these materials mingled with local drift is stranded on the long beach -of Jambeli Island, thirty miles from the mouth of the estuary. Here, -besides the seeds of Canavalia obtusifolia and Ipomœa pes capræ derived -from the locality, I found the seedlings of Rhizophora and Avicennia, -and the fruits of Laguncularia and Salicornia peruviana, that might have -been in part derived from the adjacent swamps, as well as much of the -drift of the Guayas River, such as the seeds of Anona paludosa, Entada -scandens, Erythrina, and Mucuna, the small gourds, the same small -palm-nuts, the empty seeds of Phytelephas, the “stones” of Spondias -lutea, and much other material previously familiar to me, but nowhere a -sign of the floating Pistias and of the flowering Pontederia islets of -that estuary. - -_The Stretch of Dry Coast from the Vicinity of Puna Island to the -Equator._— This remarkable piece of sea-border, covering nearly three -degrees of latitude, and in its aridity and general character recalling -the sterile sea-coast of Peru, is placed between the humid -mangrove-fronted coast of the Guayas estuary and the similarly humid and -mangrove-fronted coasts of Northern Ecuador and Colombia. The mangrove -seems to be almost absent from this stretch of dry coast. Mr. F. P. -Walker, of the Santa Elena Cable Station, tells me that some time ago a -little mangrove-growth existed near the Point, but that it has -disappeared; and Baron von Eggers implies the absence of mangroves from -the whole coast. The first-named speaks of the dry character of the -coast district from Santa Elena Point to within half a degree of the -equator; and the last-named, in his description of the coast, mentions -cacti and thorny plants as typical of the vegetation. Since this region -represents a typical locality where the direct influence of the Humboldt -current on the climate of almost the whole west coast of South America -can be put to the proof, I will refer to its peculiar climatic -conditions below in my discussion of the general question, and will here -content myself with saying that on this dry portion of the coast of -Ecuador we have reproduced, but in a less pronounced degree, the -climatic conditions of the coast of Peru. - -_The Humboldt or Peruvian Current and the Climate of the West Coast of -South America._—The question we will now briefly consider is one that is -concerned with the determining causes of the singular distribution of -coast-plants on the west coast of South America. The reader will have -already seen that the matter is an affair of climate; but it is an -affair of climate in which (although it affects forty or more degrees of -latitude), latitude, in a general sense, scarcely counts. All the -naturalists, from Humboldt onward, who have sojourned in this region of -the globe have displayed a deep interest in this subject; and I suppose -there can be no region of the globe where there are so many climatic -anomalies as interesting to the meteorologist. Here, for instance, might -be obtained materials for solving the irritating mystery of a London -fog; and if the suggestion of Baron von Eggers, before alluded to, is -carried out, and a station is established by the Meteorological -Societies of Europe and America at some suitable locality like Santa -Elena on the coast of Ecuador, we might obtain, among other results, -another line of investigating the causes of the fogs of our metropolis, -a subject about which Captain Carpenter has recently made an important -preliminary inquiry. - -I will assume that my readers are already acquainted with the nature of -the problem to be discussed relating to the climate of the west coast of -South America, and that they are familiar with the view generally held -that the aridity of this extensive coast region, stretching from the -thirtieth parallel of south latitude to the equator, arises from the -loss by the trade-winds of all their moisture in the interior of the -continent before reaching the western countries of Chile and Peru. Mr. -Ball, in his book on South America, opposed this view, though from -reasons only partially valid, since he instanced the Ecuador coast as -being, contrary to the theory, a wet coast, whereas we know that a large -stretch of it is arid and not unlike Peru. The parting of the ways in -this discussion lies in the answer to the query, Why should the -south-east trade carry so much moisture to the east side of South -America, whilst the south-west winds, that are equally prevalent on the -west coast of the continent, are drying winds which convert the -sea-border into a desert, as in Northern Chile, or into a region of -semi-sterility, as in the instance of Peru? Other things being equal, we -should expect both sea-borders of the continent in these latitudes to be -well watered. In the answer to the question why the south-east trade -should be a wet wind and the south-west wind a dry one lies a fatal -objection to the prevailing view. - -When Professor Davis, in his article on North America in the -_Encyclopædia Britannica_ (vol. 25), observes in connection with the -arid coast regions on the west side of the continent that the southerly -flow of the winds along the Pacific coast gives them a drying quality, -thus causing the extension to the coast in South California and in North -Mexico of the arid regions of the interior, he seems to imply that these -winds acquire their drying capacity in flowing from cooler to warmer -latitudes. On this view all trade-winds should be drying winds, whereas -the reverse would appear to be the case. - -There is some condition, present on one coast of the South American -continent and absent on the other, which determines why a southerly -wind, blowing landward, is in the one case moist and in the other dry. -According to my own view the winds of the arid coast regions of western -North America cross the cool waters of the Californian current, and thus -acquire their drying quality on striking a sea-border more highly heated -than the winds themselves. On the tropical west coast of South America -the winds also become drying winds by passing over the cold waters of -the Peruvian or Humboldt current, where mists are in consequence of -frequent occurrence; and on striking the more highly heated land-surface -at the sea-border the moving air does not part with any more moisture -until an altitude of some thousands of feet above the sea is reached, -when the cloud-belt forms. On the mountains bordering the coast of the -Antofagasta province, in January, the clouds gathered at an elevation of -4,000 to 5,000 feet. Perhaps the best way to contrast the east and west -coasts of tropical South America in this respect would be to say that -whilst the wind blows landward in both regions, the land is the -condenser on the east side, and the sea, owing to the interposition of -the cold Humboldt current, is the condenser on the west side. - -During a fortnight spent in February at Ancon, about twenty miles north -of Callao, I noticed that with the prevailing cool south-westerly wind -the coast was clear, but it was misty at sea. On the few days when there -were warm westerly and north-westerly breezes, the weather was thick at -sea; and if this condition was pronounced, the whole coast was enveloped -in mist; but more usually the coast-line was fairly clear except at the -promontories, along the sides of which clouds blown in from the sea -rolled in lines inland, not generally attaining an elevation over 300 or -400 feet, but sometimes reaching 900 or 1,000 feet, and gradually -disappearing a mile or two inside the coast-line. These sea-born clouds -thus vanished as they traversed the more highly heated land-surface; and -the air-current continuing its inland course mounted the slopes of the -adjacent mountain ranges of the Andes, some three or four miles from the -coast, until at an altitude of some 5,000 or 6,000 feet condensation -again occurred and the cloud-belt was formed at those cooler levels. - -From the summit of a range rising to a height of about 2,500 feet to the -north-west of Lima I had presented to me a splendid spectacle, on -February 12th, in the formation of the coast-belt of clouds. The -forenoon was clear, but about 2 p.m. the sea-born clouds began to roll -inland, concealing the lower two-thirds of the island of San Lorenzo, -which has an elevation of almost 1,400 feet, and completely covering up -Callao and the low country bordering the sea, but extending only a mile -or two from the coast-line. The dense cloud that covered Callao -appeared, as I looked down upon it from my mountain-peak, like a billowy -field of snow sparkling in the sun, with the summit of San Lorenzo -standing out like some bare alpine summit from amidst the snows. Yet -beneath that dazzling covering Callao lay all in gloom; whilst only six -miles up the broad valley of the Rimac the city of Lima stood in a blaze -of sunlight, its domes and towers reflecting back the light as I looked -at the strange contrast it presented with the buried city of the coast. -The mystery of a London fog seemed to lie unfolded at my feet, ready for -the man who can read the signs aright. - -That the mere presence of a cold current on a coast with the winds -blowing off the land (as in the case of the Labrador current, which -extends down the Atlantic coast of North America to Cape Hatteras and -beyond) produces no sterilising effect on the vegetation of the -sea-border of a continent is well brought out in the beautifully -executed maps in Prof. Russell’s recent work on North America. The -essential condition for producing sterility on the sea-border of a -continent is not only that the waters of a cold current should wash its -coasts, but that the regular winds should blow landward across its cool -surface. These are what we find on the west coast of South America. - -Not with the hope of adding anything new to our knowledge of the -climatology of this region, but with the purpose of becoming personally -acquainted with the problem involved, I paid considerable attention to -this subject during the three months passed on the west coast of South -America between Port Valdivia and Guayaquil. It was not until I had -dropped my thermometer into the cool water of the Humboldt current and -had watched the formation of the fogs on the sterile coast of Peru that -the real nature of the problem presented itself. From the pages of a -work like Tschudi’s _Travels in Peru_ one acquires an excellent idea of -the extraordinary climatic conditions of this region, and the same may -be said of the narratives of Darwin and other travellers; but it is -necessary to be brought into personal contact with these conditions -before one can appreciate their significance. - -As is well known, says Baron von Eggers, the Humboldt current explains -the anomalous climate of the coast of Peru, and one may add of North -Chile and Ecuador. The current, which represents the extension -northwards of the west wind-drift of the Roaring Forties (see Dickson in -_Encycl. Brit._, xxxi. 404; and Admiralty Current Charts of the -Pacific), begins on the coast between the 33rd and 40th parallels of -south latitude, according to the season. North of Valdivia, as we -approach Valparaiso, in lat. 33°S., the effect of its presence is at -once seen in the increasing dryness of the climate and in the alteration -in the character of the vegetation. It has, however, been shown that the -current needs the co-operation of the prevailing southerly and westerly -winds as they blow landward over its cool waters. On the coast of Peru -these moist winds often generate fog and mist as they cross the current. -They reach the coast as drying winds, having a temperature much cooler -than the lower coast regions; and the air-currents do not precipitate -any moisture on the land until an elevation of 4,000 to 6,000 feet is -attained where the cloud-belt is formed. - -In order to establish this theory it is, however, necessary to show that -when the Humboldt current leaves the coast normal conditions of humidity -occur, to which the vegetation responds, and that when the current -strikes the coast again the conditions of aridity reappear. In its -course northward the current divides off Cape Blanco, the principal mass -of its waters making towards the Galapagos Group, whilst the remainder, -after crossing the Gulf of Guayaquil, flow along the coast of Ecuador -between Santa Elena and the equator. Now, it is along this stretch of -the Ecuador coast that the conditions of aridity reappear and that the -climate of the Peruvian sea-border is in a modified form reproduced. In -the interior of the Gulf of Guayaquil, on the other hand, where the -sea-border is no longer subjected to the influence of the cold waters of -the Humboldt current, the genius of the tropics, repressed through so -many degrees of latitude, bursts its bonds, and presents us with a -spectacle of littoral vegetation that, so far as mangrove-growth is -concerned, is probably unrivalled on our globe. - -This contrast is well shown in the mean annual temperatures on the -opposite sides of the Gulf of Guayaquil. Baron von Eggers, quoting Dr. -Wolff, states that whilst the mean for the year at Puna is about 75° F., -and at Santa Elena about 73°, on the south side of the gulf at Balao it -is several degrees warmer and is evidently not under 80°. The mean -temperature for the second week of March during my sojourn at Puerto -Bolivar, which is near the beginning of the mangrove region on the south -side of the gulf, was 79°, the mean daily range being 74° to 83·5°. This -stretch of dry coast reaching north from Puna to the equator is -evidently regarded by Baron von Eggers and others who have studied the -climatology of Ecuador as the critical area required to confirm the -theory connecting the aridity of the west coast of South America with -the Humboldt current. Here the sea for the greater part of the year has -a temperature (according to the British Admiralty chart of -surface-temperatures) of 70° to 75°; the mean temperature of the air is -73° to 75°; the rainy season, instead of covering a period of six months -and over, as in the humid regions north and south of this coast, has a -duration of only two or three months; the prevailing wind is south-west; -whilst the direct influence of the cool waters of the current is shown -in the general cloudiness that prevails during the last half of the year -and in the drizzling mists that are frequent from June to October. -Reference has already been made to the manner in which the vegetation on -this dry coast of Ecuador responds to the arid conditions, as, for -instance, in the absence of mangroves and in the prevailing character of -the plants of the sea-border, cacti, thorny plants, and such like. For -my information on this exceedingly interesting tract of coast, which is -the test-ground of the Humboldt current theory, I am indebted to the -papers of Baron von Eggers (see end of this volume) and to Mr. F. P. -Walker, of the Central and South American Telegraph Company’s Station at -Santa Elena, who very kindly communicated with me by letter. Some -additional remarks are given in Note 73, and my own observations on the -temperature of the Humboldt current from Antofagasta northward are -summarised in Note 74. - -Before quitting the Ecuador coast a word may be said relating to the -prediction of Villavicencio that the climate of this sea-border will -assimilate itself to that of the rainless coasts of Peru. This is, I -believe, also the opinion expressed by Dr. Wolff in his _Geografia y -Geologia del Ecuador_ (Leipzig, 1892); and it is referred to by Mr. -Webster in his article on Ecuador in the seventh volume of the -_Encyclopædia Britannica_. There is a prevailing impression amongst the -more observing residents that I met in the Ecuadorian province of Eloro, -on the Peruvian border, that the country is drying up. A few pages back -I have described how in the Machala district of this province the -mangrove-belt passes landward into an arid region suggestive of the -sea-border of Peru. This transition is startling to one who expects to -find behind the mangrove-belt, as he would find in most parts of the -world, a humid region where Nature revels in the rank luxuriance of -plant-growth. This is, however, not always the case, since on the lee or -dry sides of the large islands of Fiji the mangrove-belt is backed by -extensive arid plains, for an explanation of which, as I have shown in -Note 22, we have to appeal rather to the hygrometer than to the -rain-gauge. This is true also of Ecuador; but whilst the reason is -intelligible enough in Fiji, it only carries us a step farther back in -the case of the Machala plains in Ecuador. These plains are continuous -with similar districts across the Peruvian border where they reach the -coast; and if the reader will refer again to my description of the -section of the mangrove-belt and the plains in its rear from Puerto -Bolivar to Machala, he will incline to the view that the desiccation of -the sea-border of Ecuador is now in progress. - -Evidence of a more direct nature could doubtless be supplied by those -who have long resided on the coast of Ecuador, and in illustration I -will give an extract from one of Mr. Walker’s letters dated May, 1904, -from Santa Elena.—“The rainfall here might for the last ten years be put -down at two showers per year. It is said that the last good rainy season -was in 1891. The inhabitants say that formerly it always rained enough -to make the grass grow every year, but during the eleven years I have -been here there appears to be a marked falling off of the rainfall.” - -It has been only possible to touch the fringe of this interesting -question here; but from the standpoint of the study of the littoral -flora of the west coast of South America it is of some importance. -Immediately behind the epoch of the present marine molluscan fauna of -this coast there lies an age when, as we learn from Philippi, the shells -of Chile were more akin to those of the Atlantic and Mediterranean -faunas than to those now found on the Chilian coasts. The transition is -a sudden one; and amongst other explanations of this strange -transformation Suess suggests the sealing up of a communication through -the Panama isthmus by volcanic eruptions and the appearance of the -Humboldt current (_Das Antlitz der Erde_, French edit. by Margerie, ii. -825). May it not be, my readers may ask, that the west coast of South -America is still in the age of progressive sterility; and that before -this age began Peru possessed a normal tropical strand-flora? It has -been remarked in Chapter VIII. that the same species of mangroves occur -on both the Atlantic and Pacific coasts of America, and that at all -events their present distribution belongs to an age when the Gulf of -Mexico was in communication with the Pacific Ocean. May we not, again, -suppose that in that age the mangroves extended far south on the coast -of Peru, just as they do now on the coast of Brazil? - -Coral reefs are stated not to exist in tropical latitudes on the west -coast of South America in our own day; but we might almost expect that -at the close of the Tertiary period, and perhaps before the appearance -of the Humboldt current, they existed with the mangroves on the coast of -Peru. As bearing on the subject of a change of climate on that coast in -times geologically not remote, I may allude to the circumstance, which -is discussed more in detail in Note 75, that I found, sometimes in fair -quantity, blocks of massive coral, long since dead, much pierced by -boring shells, and in places undergoing a chemical change, at Arica -(lat. 18° 25ʹ S.), at Callao (12° 3ʹ S.), and at Ancon (11° 45ʹ S.) on -the coast of Peru. - -These masses, which varied from a few inches to two or three feet in -size, gave me the impression of having been torn off the bottom, in some -cases in recent times, in others perhaps centuries ago, by the huge -sea-waves that from time to time overwhelm this coast. At Ancon, where -they were sufficiently abundant to be used for bordering the flower beds -in the hotel garden, they were most numerous in the vicinity of a rocky -spur of andesite that protruded from the beach between the tide levels -and was more or less covered at high water. A few paces inland from the -beach some of these coral masses, evidently stranded long ago, were -undergoing that queer process of disintegration which everything -calcareous seems to undergo on the beaches and plains of this almost -rainless coast. Like the bones of the Incas lying bleaching on the -neighbouring plains, like the sea-shells and bones of bird and beast -cast up long ago on the beach, they were falling to powder where they -lay, and the coral fragment lay often in the midst of its own _débris_. -The blocks on the beach proper were for the most part still hard and -compact, and the same may be said of those observed on the beaches of -Callao and Arica. - -The corals were quite different from those with which I was familiar in -the reefs of the Pacific islands, and, bearing in mind the known -distribution of coral reefs, I was a little dubious about them. -Accordingly I sent some specimens to the British Museum, and Mr. Jeffrey -Bell has kindly informed me that they seem to be decayed and much -injured perforated examples of Porites. When powdered they effervesce in -an acid, but the bulk of the material remains undissolved. - -No more eloquent testimony could be afforded of the rainless climate -than these corals crumbling on the Ancon plains when washed a few paces -inland from the beach. They could be noticed in all stages of -disintegration from the block surrounded by a little line of -disintegrated material, representing the initial products of its own -decay, to the crumbling mass, almost friable in the fingers, that was -lying in the midst of its own dust and loose polyp-tubes, and finally to -the little mound of _débris_ that alone remained. Mr. Darwin, in his -_Journal of Researches_ (chap. xvi.), refers to a similar process of -decay in the elevated shell-beds of San Lorenzo, off the coast of -Callao. On the higher terraces a layer of saline powder, consisting of -sulphates and muriates of lime and soda but with very little carbonate -of lime, was the sole indication of the shell-beds. Dry climatic -conditions at the sea-border evidently favour, as he observes, the early -decay of exposed calcareous remains. - - - _The Shore-plants and Stranded Seed-drift of the Panama Isthmus._ - -I spent two days at Panama and two days at Colon in examining the -neighbouring beaches and estuaries of the Pacific and Atlantic coasts of -the isthmus. On the Panama side the mangrove-belt was formed on the -seaward border of “mangle chico” (the small prevailing type of -Rhizophora mangle), Laguncularia, and Avicennia; whilst behind it passed -into extensive swampy tracts occupied by the Swamp Fern (Chrysodium -aureum), Hibiscus tiliaceus, and other plants. On the Colon or Atlantic -side the mangrove-belt had precisely the same composition and presented -the same species, Rhizophora and Avicennia usually forming the outposts -on the reef-flat, whilst Laguncularia was abundant in the rear. In the -estuary of the Rio Chagres, Rhizophora and Laguncularia were abundant -near the mouth, and Chrysodium aureum and Hibiscus tiliaceus by the -waterside higher up. Dr. Seemann, in his volume on the botany of the -voyage of H.M.S. _Herald_, observes that the species of Laguncularia -common on both the Atlantic and Pacific coasts of the Panama isthmus is -L. racemosa. This species differs in the form of its fruit from the -Ecuador tree. Laguncularia racemosa, Rhizophora mangle, and I may add -Anona paludosa and Conocarpus erecta, are all plants of the -mangrove-formation that occur not only on the Pacific and Atlantic -coasts of America but also on the west coast of Africa. It is likely, I -may add, that the “mangle grande,” the Ecuadorian type of Rhizophora -mangle, exists in the Panama isthmus, since in the higher part of the -estuary of the Chagres I found trees approaching it in characters. - -Amongst the plants growing on the Panama beaches I noticed Canavalia -obtusifolia, Hibiscus tiliaceus, and Ipomœa pes capræ, all of which -occur also on the Atlantic side of the isthmus. The Manchineel -(Hippomane mancinella), found also on the Atlantic side of the -continent, grows on the Panama beaches. Its fruits, which look like -crab-apples, lose their outer fleshy covering when drying on the sand. -Not being familiar with this poisonous tree, I allowed some of the milky -sap of the fruits to touch the skin, and suffered great pain for five or -six hours. The fruit possesses an inner coat of air-bearing cork-like -tissue; and the stone, if I may so term it, thus acquires great floating -power. I kept some afloat in sea-water for five weeks, and no doubt they -will float for months. - -The seed-drift to be observed stranded on the beaches and floating in -the estuaries on both sides of the isthmus is, generally speaking, the -same—a circumstance of great importance in plant-distribution, since we -can here see rivers bringing down the same seeds from the same “divide” -to the shores of the Pacific and Atlantic oceans. In the case of a plant -like Entada scandens, which grows in the interior, this is a matter of -much interest, as it thus possesses here a centre of dispersal from -which its seeds can be carried by the currents eastward to the West -African coast and westward across the Pacific to Malaya and (given time) -around the shores of the Indian Ocean to the East African coast. In -describing the possible routes of dispersion from this centre I have -described the distribution of the species. - -I am indebted to Mr. Holland, of the Kew Museum, for the identification -of some of the drift-seeds and fruits collected by me on the isthmus, -those identified by him being followed by the letter H. On the beaches -and floating in the estuaries on both sides of the isthmus I found -Rhizophora seedlings; seeds of Entada scandens and Mucuna urens -(medic.), H.; seedvessels of Spondias lutea (Linn.), H.; Prioria -copaifera (Griseb.), H., with decayed seed; and the empty nuts, 1-1/2 to -2 inches in size, of more than one species of Astrocaryum, H. Although -in the case of the two last-named genera the seedvessels were useless -for dispersal, being evidently brought down from the interior by the -rivers, they serve to illustrate the important principle that the rivers -bring down the same seed-drift on both the Atlantic and Pacific coasts -of Central America. Mr. Hemsley includes amongst the seed-drift stranded -on the coast of Jamaica the seedvessels of Spondias (probably S. lutea) -and of Astrocaryum (_Bot. Chall. Exped._, iv. 299, 304). - -Those of Spondias lutea were found by me floating in the Guayaquil River -and stranded on the beaches of Ecuador and of the Pacific and Atlantic -coasts of the Panama isthmus. This is the Hog-plum, which in tropical -America and the West Indies is both wild and cultivated. Its buoyant -“stone” has a covering of cork-like air-bearing tissue. This is a -remarkable case of non-adaptation in the matter of buoyancy. The -seedvessels cut across contained sound seeds; and they are provided with -the essential qualities of “long floaters.” - - - _Summary._ - -(1) The strand-district of the west coast of South America is divided -into four zones:— - -(_a_) The Convolvulus soldanella zone of Southern Chile. - -(_b_) The Desert or Plantless zone of Northern Chile. - -(_c_) The Sesuvium zone of Peru. - -(_d_) The Mangrove zone of Ecuador and Colombia. - -(2) The mangroves do not extend south of Ecuador or, more strictly, -south of Tumbez (3° 30ʹ S.). - -(3) The absence of mangroves on the tropical coasts of Chile and Peru is -attributed to the Humboldt current, which has so influenced the climate -that it has converted the sea-border of North Chile into a desert and -that of Peru into a region of semi-sterility. - -(4) It is considered that this has been effected through the prevailing -winds acquiring drying qualities on crossing the cold waters of the -current in tropical latitudes. - -(5) To establish this it is shown that when the Humboldt current leaves -the coast at Cape Blanco mangroves thrive in the Gulf of Guayaquil, and -that when it strikes the coast again near Santa Elena Point and courses -along that seaboard to the equator we find the Peruvian conditions of -semi-sterility reproduced. - -(6) The probability that the arid climate of Peru is in our own time -extending northward into Ecuador is pointed out; and from the presence -of old coral blocks on the Peruvian beaches it is considered likely that -when these corals throve the mangroves extended far down the coast of -Peru. - -(7) It is shown from the presence of the same species of mangroves on -the Pacific and Atlantic coasts of America that there must have been, -not long ago, a communication between these two oceans across Central -America; but it is at the same time observed that this could not be -inferred from shore-plants with buoyant seeds that, like Entada -scandens, occur inland, since, although they occur on both sides of the -continent, their distribution can be explained by the transport of their -seeds by rivers to the Atlantic and Pacific coasts, such as we see in -operation on the Panama isthmus in our own day. - -(8) Stress is laid on the great development of mangroves in the Gulf of -Guayaquil and in the Guayas estuary; and it is pointed out that there -are in this locality two varieties of Rhizophora mangle, a large and a -small variety, the first approaching in some of its characters the -Asiatic species, R. mucronata, and being akin also to the seedless -intermediate form of Fiji. - -(9) Amongst other matters dealt with in this chapter are the floating -seed-drift of the Guayas or Guayaquil River and the shore plants and -stranded seed-drift of the Panama isthmus. From the locality last named -we learn that rivers bring down from the interior to the Atlantic and -Pacific coasts much the same seed-drift, and that from this centre -littoral plants with buoyant seeds can be distributed over the whole -tropical zone. - - - - - CHAPTER XXXIII - - SEED-DISPERSAL AND GEOLOGICAL TIME - -The shifting of the source of the Polynesian plants from the New to the - Old World.—The floral history of Polynesia stated in terms of - geological time.—The suspension of the agencies of dispersal in later - periods.—Parallel differentiation in the course of ages of climate, - bird, and plant.—New Zealand.—Insects and bats as agents in - plant-dispersal.—The effective agency of sea-birds in other - regions.—The observations of Ekstam.—The Spitzbergen controversy.—The - efficacy of ducks as distributors of aquatic plants.—Summary. - - -IN the matter of the dispersal of seeds by birds in the tropical -Pacific, there are at least two questions which my readers must have -frequently put to themselves. The one would be concerned with the -shifting of the source of the Polynesian plants from America to the Old -World, which occurred probably near the close of the Tertiary period. -The other would be connected with the suspension of the work of -dispersal over a large portion of Polynesia, which has become more and -more pronounced as we approach our own day. - -_Suggested Cause of the Shifting of the Source of the Polynesian Plants -from the New to the Old World._—In previous chapters we have discussed -the various epochs in the plant-stocking of these islands. There was -first the age of Coniferæ, in which the islands of the Western Pacific -were only concerned, an age prior to the appearance of the volcanic -groups of the Tahitian and Hawaiian regions, and placed in the Secondary -period. Then followed, in the Tertiary period, after the birth of Hawaii -and Tahiti, and when the island-groups of the Western Pacific were -mainly submerged, the general dispersion from America of the Compositæ, -Lobeliaceæ, and other orders, now represented only by genera peculiar to -the Hawaiian and Tahitian islands. Last of all, towards the close of the -Tertiary period, when the island-groups of the Western Pacific had -re-emerged, a general dispersion of Old World plants, mainly Malayan, -took place over all the present archipelagoes of the tropical Pacific. - -Since the climate of Hawaii must have, to a great extent, shared in the -vicissitudes of the continental climates of the northern hemisphere -before, during, and after the Glacial epoch, it is assumed that in the -Ice Age no tropical plants reached the group, and that only the plants -now represented in its mountain-flora could have then reached there. The -area of active dispersion of tropical plants was pushed far south. -During the Ice period, Indo-Malayan plants doubtless crowded into the -equatorial region of the Western Pacific; but, cramped and confined -within this limited area, they were cut off by a climatic barrier from -the cool latitudes of Hawaii. As the cold ages passed away, migratory -birds, confined during that period to the southern hemisphere, would -extend their ranges north, sometimes reaching Hawaii, and transporting -to it the seeds of New Zealand and Antarctic genera, now represented by -endemic species on its mountain-slopes. The Indo-Malayan plants, with -the increasing warmth in the climate of the northern hemisphere, would -overrun the Pacific, set free from their prison in the south-west -portion of that ocean. Dispersal, we might imagine, would be at first -very active over the whole ocean. - -My point is, then, that whilst the Malayan era of the plant-stocking -began after the Ice Age in the northern hemisphere, the dispersion of -the New Zealand and Antarctic genera over the Pacific took place during -that period; whilst, as before noticed, the dispersion of the Compositæ, -Lobeliaceæ, and other orders, represented now in Hawaii by endemic -genera, would be pre-Glacial and well back in the Tertiary epoch. I -would, therefore, suggest the following scheme, in illustration of the -floral history of the tropical Pacific. - -(1) The Age of Conifers of the Western Pacific during the Mesozoic -period, and before the appearance of the Hawaiian and Tahitian -archipelagoes. - -(2) The Age of Compositæ and Lobeliaceæ, and of other genera. This is an -era of American plants, and it is referred to the Tertiary period. In it -only the newly-formed Hawaiian and Tahitian groups shared, the islands -of the Western Pacific being largely submerged. - -(3) The Age of Malayan plants, regarded as mainly post-Glacial, and -subsequent, therefore, to the re-emergence of the Western Pacific -islands at the close of the Tertiary period. - -Dispersion then was general over the Pacific. The distribution of the -New Zealand and Antarctic genera, plants that take a subordinate part in -the floras of the Pacific islands, is regarded as having occurred during -the glaciation of the northern hemisphere. - -_On the Suspension of the Agencies of Dispersal in the Tropical -Pacific._—If the remark of Drake del Castillo that genera possessing -only non-endemic species in the Pacific islands owe their presence in -this region to existing agencies of dispersal looks something like a -truism, we must remember that, assuming Nature to be uniform in her -methods, it involves not merely the original co-operation of the same -agencies with genera that own only peculiar species, but also the -subsequent suspension of the work of these agencies. - -The nature of the connection between freedom of dispersal and specific -differentiation is well brought out by Beccari in contrasting the -species of Ficus and the palms of Borneo; whilst out of fifty-five -species of Ficus collected by him in that island, 30 per cent. were -apparently peculiar, 85 per cent. of the 130 Borneo palms had not been -found elsewhere. In the English edition of his _Nelle Foreste di Borneo_ -he says that “the explanation lies in the fact of the facile -dissemination of the various species of Ficus through the agency of -birds, an explanation which applies to all trees which produce edible -fruits specially relished by animals.” He shows, also, that the same -principle applies within the limits of the genus Ficus, since of those -Bornean species known to him as belonging to the section Urostigma, -which possesses fruits most preferred by birds (pigeons, hornbills, -&c.), nearly all (fourteen out of sixteen) are found elsewhere; whilst -of ten species belonging to the section Covellia, where the fruits are -more or less hidden and inconspicuous, and with difficulty discovered by -birds that would effectively distribute the species, four, at the most, -are found elsewhere. “Such facts,” he goes on to say, “show that in -tropical countries the various kinds of Ficus are, to a large extent, -biologically connected with birds, which, perhaps, on their part, also -owe some of their peculiarities in the shape of the bill or in the -plumage to the nature and coloration of the fruits which form their -food.” - -Whilst Dr. Beccari as a botanist lays especial stress on the biological -connection in Malaya between the plant, as illustrated by the genus -Ficus, and the bird, Mr. Perkins, as a zoologist, is similarly emphatic -on the biological connection in Hawaii between the bird, as illustrated -by the peculiar family of the Drepanididæ, and the plant. The plants -here are the arborescent Lobeliaceæ and the Freycinetias. To the flowers -of the arborescent Lobeliaceæ the nectar-feeding Drepanids are -particularly partial; and the development of the extreme forms of these -birds, as Mr. Perkins observes in the _Fauna Hawaiiensis_, “is not -comprehensible without a knowledge of the island flora.” Not only does -he point to the modifications in the form of the bill of the bird in -connection with the tubular form of the flowers; but in at least one -species of these arborescent Lobeliaceæ he shows that it is dependent on -the Drepanid for its fertilisation, and he inclines to the view that -changes such as that of lengthening of the bill may have taken place -side by side with the increasing length of the tubular flowers. In -connection with the Freycinetias of Hawaii, Mr. Perkins regards the bill -of the Ou, a finch-like Drepanid of the genus Psittacirostra, as -“entirely formed and adapted for the purpose of picking out the -component parts” of the fruiting inflorescence. - -That in an isolated island-group birds and plants often “differentiate” -together is a fact well known in distribution. In Hawaii, for instance, -as I learn from Mr. Perkins, quite 45 per cent. of the birds are -peculiar; whilst according to Dr. Hillebrand 80 per cent. of the -flowering plants are confined to the group. Then, again, in the -Galapagos Islands, half of the plants and two-thirds of the birds are -confined to that archipelago. At the other end of the series we have the -Azores, with about a tenth of its plants peculiar, and about 4 per cent. -of its birds peculiar to the islands, and Iceland with no endemic plants -and, as far as I can gather, few peculiar birds. - -Accepting Mr. Charles Dixon’s view (_The Migration of Birds_, 1897) that -specific differentiation does not occur along lines of migration, we -must assume that the differentiation of the avifauna of an isolated -group like Hawaii began with the breaking off of its regular -communication through birds with the outside world. I do not consider -that in the past these Pacific archipelagoes received their birds in any -haphazard fashion, as, for instance, in the guise of stragglers that had -lost their way. From the circumstance pointed out to me by Mr. Perkins -that 25 of the 67 genera of Hawaiian birds are peculiar, we must -postulate a high antiquity for the bird fauna dating far back into the -Tertiary period. Mr. Perkins, who kindly supplied me with his general -views of the nature of the Hawaiian fauna, tells me that it is -“positively oceanic-insular and could be continental only on the -supposition that everything continental had been at some time destroyed -and that the group had been subsequently re-stocked as would any oceanic -island.” - -The view naturally presents itself that in past ages birds in the -Pacific were much more uniform in their characters, and the agencies of -dispersal far more active in their operations and far more general in -their range than in more recent times, “It may be accepted (says Mr. -Dixon) as an axiom of geographical distribution that all existing -species are surviving relics of more ancient forms or ancestral types, -whose dispersal in a remoter past was more continuous, and whose -affinities and characteristics were therefore more homogeneous.” I -assume that in past ages the differentiation of birds has largely been -favoured by differentiation of climate acting through the limitation of -their ranges. To these changes, plants, so often biologically connected -with birds, have largely responded. - -There is, of course, no difficulty in imputing to birds the capacity of -reaching Hawaii in the mid-Pacific, and there are many regular migrants -now (sea-birds, waders, ducks, &c.). The only difficulty is in the -estimation of the time occupied in the trans-oceanic journey. According -to Gätke the journey, which is 1,500 to 2,000 miles, ought to be -accomplished within the limit of fifteen hours, which he regards as “the -longest spell during which a bird is able to remain on the wing without -taking sustenance of any kind.” As he considers that a bird might cover -the 1,600 miles between Newfoundland and Ireland in nine hours -(_Heligoland as an Ornithological Observatory_, p. 140), the Hawaiian -traverse would offer to him no difficulties. It has frequently occurred -to me in this connection that in ancient times, when the volcanoes of -the mid-Pacific were in full activity, their light at night-time would -have often given a direction to the migrating bird, and that they might -have sometimes determined the line of migration across the Pacific. - -It has not been possible to discuss here the capacity of pigeons to -cross an ocean, a subject bearing directly on the floras of all the -Pacific groups (excepting Hawaii, which possesses no indigenous Columbæ) -and as concerning these islands generally presenting no difficulty. Dr. -H. de Varigny, who amongst his other studies has long displayed an -active interest in plant-dispersal, has directed my attention to two -very important papers on the flight of pigeons in the _Revue -Scientifique_, one by M. A. Thauziès (April 30, 1898) and the other by -M. M. Dusolier (Nov. 28, 1903). That land birds, as well as swimmers and -waders, cross the Atlantic is well known, and in this connection the -reader might profitably consult Prof. Heilprin’s _Geographical -Distribution of Animals_ (vol. 58, _Internat. Sci. Ser._ 1887). - -Much might be said of these matters, but it would be out of place here; -and I will content myself with stating the view above indicated that the -suspension of the agencies of seed-dispersal over the Pacific is -probably connected with a general principle affecting the whole -plant-world. The tendency in the course of ages has been towards the -differentiation of climate, bird, and plant, the range of the bird being -largely controlled by the climate, and the range of the plant being -mainly dependent on the range of the bird. - -It is evident that in some cases the plants themselves may make the -endemism of a flora more pronounced by creating their own difficulties -and by standing in the way of their own dispersal to outside regions. It -has been shown that some of the endemic Hawaiian genera (see Note 68) -have deteriorated in their capacity for dispersal by birds; and similar -remarks are made with reference to the genera Sicyos (page 365) and -Eugenia (page 350). Genera with stone fruits like Elæocarpus possess in -the different species stones of various sizes, some of them suitable in -point of size for conveyance in a bird’s body over an ocean, others so -large that one could only predicate for them a limited capacity for -distribution by birds over a few hundred miles of sea. One, for -instance, could safely assume that species of Elæocarpus, with stones an -inch and over in size, that occur in Fiji and Hawaii, are not suited for -distribution over an ocean now; whilst other species found in New -Zealand and Rarotonga have stones less than half this size, which are -quite fitted for distribution by birds over broad tracts of ocean (page -337). - -This brings us to discuss the relative difficulties presented from the -dispersal-standpoint by the forest floras of Hawaii, Fiji, and New -Zealand. It is with the forest floras that nearly all the difficulties -of distribution lie; and I hope I shall not be considered presumptuous, -or at all events too heterodox, in expressing the opinion that judging -from the details given in Kirk’s _Forest Flora of New Zealand_ those -islands present no greater difficulties for the student of -plant-distribution, if we exclude the Coniferæ, than either Fiji or -Hawaii. Indeed, even with the Conifers included, New Zealand presents -fewer problems than Fiji, and Hawaii has its own special difficulties -connected with the inland species of the Leguminosæ. There is, on the -other hand, no special New Zealand difficulty. It possesses the Conifers -in common with Fiji; and it shares with Fiji and Hawaii genera like -Elæocarpus and Sideroxylon, that take a foremost place amongst the trees -of the Pacific forest flora presenting puzzling points to the student of -distribution. The existence of Elæocarpus in New Zealand admits of a -simpler explanation than the occurrence of the same genus in Hawaii. -Pandanus in Fiji is a more difficult genus from the standpoint of -dispersal than Corynocarpus in New Zealand, and in fact, than any of the -non-coniferous genera of forest trees in that region. - -Whilst it is likely that birds of the genus Porphyrio have, up to almost -recent times, been active in distributing the seeds of New Zealand -plants outside the region (see p. 296), it would seem that the -fruit-pigeons, as represented by a solitary peculiar species of -Carpophaga, have long since ceased to be active in this direction. It is -true that Sir W. Buller gives a long list of trees, including -Corynocarpus, Elæocarpus, Litsea, Olea, Podocarpus, and many others, the -fruits of which are appreciated by this pigeon; but since the bird is -confined to this region its efforts in plant-dispersal possess only a -local interest. Mr. G. M. Thomson, indeed, has expressed the opinion -(_Trans. and Proc. N.Z. Instit._ vol. 33) that in recent times not a -single plant has been added through the agency of birds to the New -Zealand flora. Besides the regular migratory birds two cuckoos only -reach the region, the one from Australia and the other from Polynesia; -whilst Australian birds which had managed to survive the long flight -across the ocean have been met with only at times on the west coast of -the North Island. From the standpoint adopted in this work we should -have expected that, with the exception of current-dispersed plants, New -Zealand would be out of touch with the world outside. Varied only by -occasional inrushes of plants, its history, dating back to the Mesozoic -age, has been one of insular isolation. - -When, however, we apply the principles of plant-dispersal in the -Pacific, deducible from the study of the Hawaiian flora, we learn that -the stocking of New Zealand with its plants could have been carried out -(with the exception of the Coniferæ and a few other genera like Fagus -that are in a geological sense ancient denizens in this region) by the -agencies that stocked Hawaii with its flora. New Zealand genera like -Elæocarpus, Sideroxylon, Sophora, etc., that are represented in the -forests of Hawaii, could not be taken to illustrate any former -continental connection. If we, so-to-speak, put the New Zealand forest -flora in the Hawaiian sieve, all will pass through with the exception of -Fagus, the genera of the Coniferæ, and plants of similar history in high -southern latitudes. This residuum belongs more to the palæobotanist than -to the student of means of dispersal. - -I should be inclined to think that the tropical genera of the New -Zealand flora, more especially of the forest flora, reached that region -during the glaciation of the northern hemisphere, when the Indo-Malayan -plants were, so-to-speak, cornered in the Western Pacific. Yet it must -be noted that these are, as a rule, genera that either display an -indifference to the varying thermal conditions of different latitudes or -are known to at times extend their range beyond the tropics. Thus -Elæocarpus and Freycinetia are equally at home in the temperate -rain-forests of New Zealand and in the tropical rain-forests of -Polynesia and Malaya; whilst widely-spread tropical genera like -Pittosporum and Peperomia, that occur in New Zealand, exhibit their -power of adaptation to varying climates by extending outside the tropics -in other regions and by their vertical range in the Hawaiian mountains, -where they are found alike at low elevations a few hundred feet above -the sea and at altitudes of 6,000 or 7,000 feet. - -All these plants, however, are in a relative sense recent intruders. -When the student of dispersal looks at the long list of the conifers of -the New Zealand forest flora and reflects that he knows but little of -their means of dispersal, and that if his acquaintance were far greater -it would not avail him much, he has no choice but to take his place -behind the earlier investigators of the flora, and to see in these trees -evidence in favour of a remote continental period, probably referable to -the mesozoic age. - -_A Discussion of some Means of Dispersal._—Not many authors seem to have -discussed the possibility of insects as agents of seed-dispersal in the -Pacific. They appear to me quite suited for transporting the spores of -ferns and lycopods as well as the minute seeds of plants like the -orchids and the begonias. Darwin, who allowed few possible means of -dispersal to escape his notice, procured the germination, as my readers -will remember, of grass seeds found in the dung of Natal locusts. When -on the barren summit of Mauna Loa, I noticed that the recently dead -bodies of some butterflies, that had been carried up the slopes from the -forests below by the “southerly updraught,” were already attacked by -bugs, parasites that must have been transported from the lower regions -by some of the numerous larger insects that are blown up the slopes. - -In Note 61 the occurrence of the wind-blown insects on the summit of -Mauna Loa is described. That insects can be transported into the upper -regions of the atmosphere by ascending air-currents was long ago -remarked by Humboldt, and the subject has been discussed with his usual -acumen by Whymper (_Travels amongst the Great Andes of the Equator_). -Carried along in the higher air-currents these insects might finally be -deposited at places far distant from their home. One reads occasionally -extraordinary accounts of a rain of insects. A very circumstantial -account was given to me when I was on Keeling Atoll of a shower of -dragon-flies that fell on the islands, their remains being found in -quantities in the lagoon. Dragon-flies, it is known, are often found at -sea far from land, and one species has been observed nearly all over the -world, including the Pacific islands. In this connection it is -interesting to recall Mr. M’Lachlan’s remark in his article on the -dragon-fly in the seventh volume of the _Encyclopædia Britannica_ that -some of the earliest fossil forms seem to have been washed ashore after -having been drowned at sea. - -Another creature that has been often ignored as a possible agent in -seed-dispersal is the bat. Bats are found all over the world, including -the oceanic groups, and one can scarcely doubt that they must have often -transported seeds, at all events in their hair. They are found at times -high up in mountainous regions, and Sir H. Johnston, in his recent work -on the Uganda Protectorate, refers to the occurrence of bats at an -altitude of 13,000 feet. The large frugivorous bats (Pteropidæ) are -known to be very destructive feeders; but I doubt whether they swallow -the seeds. Dr. Warburg, as is remarked in Chapter XXV, says that they -feed on the flowers of Freycinetias, and I have already observed that -they visit the flowers of Geissois ternata in Fiji (p. 394). In this -fashion Dr. Warburg regards them as agents in pollenisation; and it -seems to me that if, as appears likely, they are attracted by trees with -large, brightly-coloured flowers, they would often aid in the dispersal -of the minute seeds of trees like Metrosideros. - -Until recently sea-birds, and some particular birds of passage, have -been generally considered as only fitted for dispersing seeds in their -plumage. That they can also transport seeds inside their bodies is shown -below. Dr. R. Brown in his book entitled _Our Earth and its Story_, -1888, gives a general account of plant-dispersal with numerous -references to the Literature on the subject. On the direct route between -Scotland and Cape Farewell in Greenland snow-buntings (Plectrophanes -nivalis) and other birds of passage frequently alight, as we are told, -on ships when hundreds of miles from land. Dr. Brown says that when -taking this voyage he examined dozens of these birds. Only in one case, -however, did he find any seeds, namely, in the case of a snow-bunting -which carried, attached to its plumage, an achene of, perhaps, a -Ranunculus, and in its gizzard a seed like that of a Suæda. My discovery -of a small, hard seed in the gizzard of a Cape-pigeon (Daption capensis) -550 miles east of Tristan da Cunha has been referred to by Mr. Hemsley -in his introduction to the _Botany of the “Challenger” Expedition_ (p. -45). On p. 188 I have mentioned the probable dispersal of the seeds of -Cæsalpinia by frigate-birds and boobies; and in Note 59 reference is -made to some large seeds found in the crop of the Fulmar petrel. - -Gulls, when they nest at the coast, where the sea-thrift (Armeria -vulgaris) and the sea-campion (Silene maritima) thrive, or inland -amongst the heather-covered slopes, must often carry the seeds of these -plants from place to place in their plumage (see Notes 15 and 16); but, -as shown below, they can also disperse plants with fleshy fruits which -at times form their food. Gulls, geese, and arctic grouse take an -important part in the dispersal of seeds in the cold latitudes of the -northern hemisphere; and few things are more suggestive in this way to -the student of distribution than the data supplied by Ekstam, Hesselman, -Sernander, and others for the region including Spitzbergen, Nova Zembla, -and Arctic Norway. The history of the discussion relating to the flora -and fauna of Spitzbergen reproduces in its main features the various -stages in the controversy that has been waged in connection with the -Pacific islands. - -When Ekstam published, in 1895, the results of his observations on the -plants of Nova Zembla, he observed that he possessed no data to show -whether swimming and wading birds fed on berries; and he attached all -importance to dispersal by winds. On subsequently visiting Spitzbergen -he must have been at first inclined, therefore, to the opinion of -Nathorst, who, having found only a solitary species of bird (a -snow-sparrow) in that region, naturally concluded that birds had been of -no importance as agents in the plant-stocking. However, Ekstam’s -opportunities were greater, and he tells us that in the craws of six -specimens of Lagopus hyperboreus shot in Spitzbergen in August he found -represented almost 25 per cent. of the usual phanerogamic flora of that -region, in the form of fruits, seeds, bulbils, flower-buds, leaf-buds, -&c. This observer now also realised the importance of gulls and geese in -dispersing certain types of plants in those latitudes. Species of Larus, -he says, consume greedily all kinds of berries, and especially those of -Empetrum nigrum, the stones of which were found uninjured in their -droppings by Professor Lagerheim in Arctic Norway. Geese, as we are also -informed, are hearty plant-eaters in Spitzbergen; and Ekstam found in -their droppings the fruits of Oxyria digyna as well as an abundance of -uninjured bulbils of Polygonum viviparum, some of which proved to be -capable of growth (See Ekstam in _Tromso Museums Aarshefter_, vols. 18 -and 20, 1895-7). - -The result of Ekstam’s observations in Spitzbergen was to lead him to -attach a very considerable importance in plant-dispersal to the agency -of birds; and when in explanation of the Scandinavian elements in the -Spitzbergen flora he had to choose between a former land connection and -the agency of birds, he preferred the bird. - -I have gone into some detail in this matter because the Spitzbergen -controversy in some respects might have equally centred around New -Zealand or some of the large continental islands of the tropical -Pacific. There is at first the endeavour in the absence of precise -knowledge to disregard the bird and to look for a land connection. With -the increase in our acquaintance with the efficacy of bird-agency in -seed distribution there is the abandonment of such a view. In both -localities, however, there are the same counter-indications of the -insect faunas, and the same considerations are raised by the absence or -presence of larger animals in the regions concerned. The principal -difference lies in the frozen sea, and yet, strangely enough, it does -not seem to affect the problem much. It would indeed appear that the -questions raised by the floras and faunas of the Pacific islands are not -peculiarly Pacific in their character; and it is probable that the -difficulties here presented are repeated in one form or other in the -case of large islands over all the globe. - -_On the efficacy of Ducks and other Waterfowl in the Distribution of -Aquatic Plants._—It is highly probable that aquatic plants, like the -beach plants distributed by the currents and the ferns and lycopods -distributed mainly by the winds, have changed much less in the course of -ages than the plants of the inland forest. This in all three cases is -chiefly due to the uninterrupted freedom of communication by means of -the dispersing agency. - -Wild ducks and their kind are active agents in the distribution of the -seeds of aquatic plants; but it is curious that the early experiments of -Caspary went far to discredit them in this respect. As quoted by Dr. -Schenck in his _Die Biologie der Wassergewächse_, 1886, he fed tame -ducks with the seeds of water-lilies and found that in a short time they -thoroughly digested the seeds. Those familiar with the seeds of our -British species of Nuphar and Nymphæa will not be surprised at such a -result; but, unfortunately, the inference drawn from this experiment has -been by some extended to aquatic plants in general. Since the seeds or -seed-vessels of some aquatic or semi-aquatic plants of the genera -Potamogeton, Sparganium, &c., appeared to me to be quite fitted for -conveyance without injury in a duck’s body, I made several years ago a -number of observations on this subject, the results of which were -published in _Science Gossip_ for September, 1894. - -Out of 13 wild ducks obtained in the London market and stated to have -been sent from Norfolk and Holland, eleven contained in their stomach -and intestines 828 seeds, which I thus classed:— - - 295 seeds of Sparganium in 8 birds - 41 seeds of Potamogeton in 3 birds - 270 seeds of Cyperaceæ in 5 birds - 222 not identified - -In the case of four birds the germinating capacity of the seeds was -tested, and in three cases very successfully. The seeds of Potamogeton, -Sparganium, and of the Cyperaceæ germinated readily in water, but few of -them failing, the process beginning in a few days or a few weeks. At -that time I was conducting an extended series of observations on the -postponement of germination of the seeds of aquatic plants, the results -of which were published in the _Proceedings of the Royal Physical -Society of Edinburgh_ for 1897. It was there shown that the seeds of -these plants often postpone their germination to the second and even to -the third spring. It thus happened that, whilst seeds obtained from the -stomach and intestines of the wild duck germinated in a few days or -weeks, I had to wait often a year and more for such a result with seeds -in their ordinary condition. This was well brought out in another -experiment made on a domestic duck, which I have described on page 369. -That wild ducks are to be regarded in the light of “flying germinators” -is thus very evident. - - - _Summary._ - -(1) In explanation of the shifting of the source of the Polynesian -plants from the New to the Old World, it is suggested that during the -glaciation of the northern hemisphere the Indo-Malayan plants entering -this region were “cornered” in the tropical Western Pacific, and were -only set free after the cold period had passed away, when they overran -Polynesia. - -(2) Whilst the age of the Conifers is placed in the Mesozoic period, -that of the Compositæ is accredited to the Tertiary period, and the era -of Malayan immigration followed the glacial epoch. - -(3) The suspension to a great extent of the agencies of plant dispersal -in the Pacific in later times is connected with a general principle -affecting the whole plant-world. With the secular drying up of the globe -the differentiation of climate, bird, and plant have gone on together, -the range of the bird being mainly controlled by the climate and the -range of the plant being largely dependent on the bird. - -(4) Accepting Hawaii as entirely insular in its history, it is pointed -out that the principles deducible from the study of its flora can be -applied to the forest-flora of New Zealand, with the exception of the -Conifers and some genera that are ancient denizens of Antarctic -latitudes, and indicate a remote continental age dating back to the -Mesozoic period. It is suggested that the Indo-Malayan element in its -flora arrived there during the glaciation of the northern hemisphere. - -(5) Insects and bats have probably been effective agents in -seed-dispersal in the Pacific, and it is shown that sea-birds carry -seeds in their stomach and intestines as well as in their feathers. - -(6) It is shown that birds of the grouse family, gulls, and geese are -active seed-dispersers in cold northern latitudes, and that the -discussion of their influence in stocking Spitzbergen with its plants -reproduces many of the points of the controversy concerning the floras -of the continental islands of the South Pacific. - -(7) The results of experiments and observations are cited to establish -the efficacy of ducks in distributing the seeds of aquatic plants, the -seeds ejected in their droppings germinating in a few days or weeks, -whilst those remaining in the pond or river often do not germinate for a -year or more. - - - - - CHAPTER XXXIV - - GENERAL ARGUMENT AND CONCLUSION - - -THE problems concerned in the study of the floras of the Pacific islands -from the standpoint of dispersal are here approached through the buoyant -quality of the seed and fruit; and it is shown when dividing the plants -into two groups, those with buoyant and those with non-buoyant seeds or -fruits, that there has been at work through the ages a great sorting -process, by which the plants belonging to the group first named have -been mostly gathered at the coast. Its operation may be also observed -within the limits of a genus, where the species possessing seeds or -fruits that float is stationed at the coast, whilst the species with -seeds or fruits that sink makes its home inland. - -When the principle here involved is applied to the British flora, it -presents itself as part of a much wider principle, by which plants -endowed with buoyant seeds and fruits have been stationed at the -water-side, whether on a river-bank, or beside a lake or pond, or on a -sea-beach. The broader principle proves in its turn to belong to a far -larger scheme, in which the fitness or unfitness of a plant to live in a -physiologically dry station appears as the primary determining quality, -the xerophyte (the plant of the dry station), provided with buoyant seed -or fruit, finding its way to the coast, and the hygrophyte (the plant -growing under more moist conditions), that is similarly endowed, -establishing itself by the side of the river, or the lake, or the pond. - -When dealing with the general character and composition of the -strand-plants of the tropical Pacific, it is shown that in Fiji the -beach-plants often assert their primary xerophilous habit or fitness for -occupying any dry station by extending into the inland plains on the dry -sides of the islands. The Fijian shore-plants are divided into three -formations, those of the beach, those of the mangrove-swamp, and those -of intermediate stations on the borders of the swamps. The great -majority of the Fijian shore-plants are dispersed by the currents. The -Tahitian Islands, which are representative of Eastern Polynesia, lack -the mangroves and most of the plants that grow at the margin of a -mangrove-swamp; and their strand-flora is mainly composed of plants of -the beach, such as are dispersed by the currents far and wide in -tropical regions. The Hawaiian strand-flora is very meagre in its -character, lacking not only the plants of the mangrove and intermediate -formations, but almost all the large-fruited beach-trees of the South -Pacific. Since Hawaii possesses but few current-dispersed shore-plants -that are not found in the New World, reasons are given for the inference -that such shore-plants were originally brought by the currents from -America, and not from the South Pacific. - -We are led on various grounds to the conclusion that tropical -shore-plants distributed by currents belong to two great regions, the -American including the west coast of Africa, and the Asiatic, or Old -World Region, which includes the African east coast. It is held that -America is so placed with regard to the currents, that it is a -distributor, and not a recipient of tropical shore-plants dispersed by -that agency. From this it follows that all cosmopolitan tropical -beach-plants that are dispersed by the currents have their homes in -America. - -The results of observation and experiment are given to show that there -is no direct relation between the specific weight of seeds and fruits -and the density of sea-water. Yet, although the floating or sinking of a -seed or fruit is but an accidental attribute, it has had indirectly a -far-reaching influence not only on plant-distribution, but on -plant-development. In accordance with this want of relation between the -specific weight of seeds and fruits and the density of sea-water, the -great variety of structures concerned with buoyancy are regarded in the -main, after a detailed examination of their character, as not arising -from adaptation. Rather, it is urged, is buoyancy connected with -structures that now serve a purpose for which they were not originally -developed. Nature, it is held, has never concerned herself directly with -providing means of dispersal of any sort. - -In the discussion of the relation between the littoral and inland -Pacific floras, it is shown, as a result of the examination of those -genera possessing both shore and inland species, that they have been on -the whole developed on independent lines. Two special difficulties in -explaining the modes of dispersal of plants of the Pacific islands here -come into prominence. There is the Hawaiian difficulty, where with -genera containing both shore and inland species only the last are found -in Hawaii; and although the shore-plants are known to be dispersed by -the current, the inland plants display little or no capacity for this or -any other mode of dispersal. Here belong the Leguminous genera -Canavalia, Erythrina, Mezoneuron, and Sophora, and the Apocynaceous -genus Ochrosia; and it is assumed that the inland Hawaiian species are -derived from a current-dispersed shore-plant that has since disappeared -from the group. The Fijian difficulty is displayed in those genera where -both coast and inland species occur in the islands, but no known -existing means of dispersal across an ocean can be postulated for the -inland plants, though the shore species are distributed by the currents. -Of such genera Pandanus is the best example, and it is pointed out that -this genus presents the same difficulty in the Mascarene Islands, in -which case the agency of the extinct Columbæ is invoked. - -As illustrating the methods of observation and experiment employed by -the author, the Leguminous shore-plants Afzelia bijuga, Cæsalpinia -bonducella, and Entada scandens are discussed at length; and in the -chapters on the enigmas of the Leguminosæ in the Pacific it is pointed -out that the behaviour of the plants of this order is a source of much -perplexity, and that they conform to no single rule of dispersal. - -Coming to the inland plants of this region, the Fijian, Tahitian, and -Hawaiian groups are taken as the chief centres of distribution in the -Pacific. After discussing the relative sizes, the altitudes, and the -climates of these three archipelagoes, it is shown that Hawaii, on -account of the far greater altitude of the islands, is characterised by -a special mountain flora, and that it is comparable with Fiji, and to a -great extent also with Tahiti, only as regarding the plants of the -levels below 4,000 or 5,000 feet. - -The first era of the plant-stocking is designated the Age of Ferns, and -it is observed that, whilst in Hawaii nearly half of the ferns and -lycopods are peculiar to that group, very few new species have been -developed in the Fijian and Tahitian regions. - -The next era in the floral history of these islands is represented in -the first era of the flowering plants. This is indicated by the endemic -genera, which are particularly numerous in Hawaii, relatively scanty in -Fiji, and very few in Tahiti. On account of their preponderance, the era -is designated the Age of Compositæ and Lobeliaceæ. The genera of these -two orders, though mainly characteristic of Hawaii, are also to be found -in the Tahitian region, but they are absent from the Fijian area. -Chiefly American in their affinities, their dispersion over the Pacific -took place during the Tertiary submergence of the archipelagoes of the -Western Pacific, in which are included the groups of the Fijian area -(Fiji, Samoa, Tonga). These early forms of Compositæ and Lobeliaceæ are -often arborescent in habit; and it is observed that Tree-Lobelias also -occur high up the slopes of lofty mountains in tropical regions, as in -Equatorial Africa, under conditions similar to those prevailing on the -slopes of the Hawaiian mountains, where the Tree-Lobelias, termed by Dr. -Hillebrand “the pride of our flora,” abound. - -The other Hawaiian endemic genera, marking the first chapter in the -history of the flowering plants, arrange themselves in two groups, one -chiefly American in general affinities, and containing highly -differentiated Caryophyllaceæ, Labiatæ, &c.; the other largely Malayan, -and indicating the close of the first era of the flowering plants, when -the main source of the plants was shifted from America to the Old World. -The Fijian endemic genera, which are few in number, miscellaneous in -appearance, and disconnected in character, are regarded as having -probably acquired their endemic reputation through their failure at -their sources in the regions to the west. - -The second era of the flowering plants is indicated by the non-endemic -genera. Here we are concerned on the one hand with a mountainous flora -mainly Hawaiian, in which genera from the New Zealand and Antarctic -floras take a conspicuous part, and on the other with a low-level flora -chiefly derived from Indo-Malaya, and including the plants of the lower -slopes of Hawaii below 4,000 and 5,000 feet, and the floras in mass of -Fiji and Tahiti. - -On account of their lower altitude, the extensive mountain flora of -Hawaii is but scantily developed in Tahiti, and is represented by a mere -remnant in Fiji and Samoa. Two-thirds of the Hawaiian non-endemic -mountain genera contain only species restricted to the group, and, -although amongst these disconnected genera, Acæna, Gunnera, Coprosma, -Lagenophora, &c., of the New Zealand and Antarctic floras take a -prominent part, a large proportion of the genera like Ranunculus, Rubus, -Artemisia, Vaccinium, and Plantago represent generally the flora of the -north temperate zone on the summits of tropical mountains. The Tahitian -mountain flora, scanty as it is when judged by the non-endemic genera, -displays much kinship with the Hawaiian mountain flora; but this kinship -is mainly confined to genera from high southern latitudes, such as -Coprosma, Cyathodes, Astelia, &c. In the possession on its mountain -slopes of the three genera of the Coniferæ, Dammara, Podocarpus, and -Dacrydium, the Fijian region is distinguished from that of Tahiti and -Hawaii; and it is assumed that they mark the site of a continental area -in the Mesozoic period, when the Tahitian and Hawaiian groups did not -exist. - -The era of the non-endemic genera, in so far as it is concerned with the -low-level flora of Hawaii and the floras in mass of the areas of Fiji, -Samoa and East Polynesia, is termed Malayan, because many of the genera -are thence derived. Here we are dealing with all the oceanic groups of -the tropical Pacific, and not with a portion of them, as in the case of -the Age of Coniferæ, in the Secondary period, that was limited to the -Western Pacific, or in the case of the Age of Compositæ and Lobeliaceæ -that was restricted during the Tertiary epoch to the Hawaiian and -Tahitian regions. The first part of this era, as is indicated by the -endemic species, is an age of complete isolation in Hawaii, and of -partial isolation in the groups of the southern region. Amongst the -genera typical of this period are Pittosporum, Gardenia, Psychotria, -Cyrtandra, and Freycinetia. A later period in this era of the general -dispersal of Malayan plants over the Pacific is one where the extremely -variable or polymorphous species plays a conspicuous part, as -represented in such genera as Alphitonia, Dodonæa, Metrosideros, -Pisonia, and Wikstrœmia, the general principle being that each genus is -at first represented by a widely ranging very variable species, which -ultimately ceases to wander and settles down, and becomes the parent of -different sets of species in the several groups. - -The facts of distribution in this age of general dispersion are just -such as we might look for in the case of a general dispersal over the -oceanic groups of the Pacific, with the altitudes of the islands playing -a determining part. But it should be remarked that the greater number of -the genera that have entered the Pacific from the Old World have not -advanced eastward of the Fijian region, half of the Fijian genera not -occurring in the Hawaiian and Tahitian regions. The explanation of this -is to be found, not in any lack of capacities for dispersal, but in a -want of opportunities. The story of plant-distribution in the Pacific is -bound up with the successive stages of decreasing activity in the -dispersing agencies. The area of active dispersion, as illustrated by -the non-endemic genera, at first comprised the whole of the tropical -Pacific. It was afterwards restricted to the South Pacific, and finally -to the Western Pacific only. The birds that carried seeds all over this -ocean became more and more restricted in their ranges, probably on -account of increasing diversity of climatic conditions. The plants of -necessity responded to the ever narrowing conditions of bird-life in -this ocean, and the differentiation of the plant and the bird have taken -place together. - -During the stages of decreasing activity in the dispersing agencies, the -widely-ranging highly variable species continued to be an important -factor in the development of new species in the different groups. The -_rôle_ of the polymorphous species has always been a conspicuous one in -the Pacific. - -Yet, as in the case of the Cyrtandras, it is shown that the display of -great formative power within a genus is not a peculiarity of an insular -flora; that the isolation of an oceanic archipelago does not exclusively -induce “endemism,” but only intensifies it; that the development of new -species may be nearly as active on a mountain in a continent as on an -island in mid-ocean; and that this is equally true of a land genus, like -Embelia, exposed to an infinite variety of conditions, and of an aquatic -genus, like Naias, where the conditions of existence are relatively -uniform all the world over. - -In framing a scheme by which the eras of the floral history of the -Pacific are brought into correlation with those of geological time, the -age of the Coniferæ is placed in the Secondary period, that of the -Compositæ and Lobeliaceæ in the Tertiary period, whilst the era of -Malayan immigration is regarded as mainly post-glacial. The age of the -Coniferæ is concerned only with the Western Pacific, since the Hawaiian -and Tahitian islands had not then been formed. The age of the Compositæ -and Lobeliaceæ is concerned only with Hawaii and Tahiti, since the -islands of the Western Pacific were then more or less submerged. That of -the Malayan plants affects the whole Pacific as at present displayed to -us. - -In the chapter on the viviparous mangroves of Fiji it is shown that both -the Asiatic and the American species of Rhizophora (R. mucronata and R. -mangle) exist in that group, and that there is in addition a seedless -form, the Selala, which, although intermediate in character between the -two other species, comes nearest to the Asiatic plant. Reasons are given -for the belief that the Selala is derived from the Asiatic species (R. -mucronata), not as the result of a cross but as connected with its -dimorphism; and in support of this it is pointed out that on the Ecuador -coast of South America, where only the American species exists, a -dimorphism is also displayed, one of the forms approaching in several of -its characters the Fijian Selala, though fruiting abundantly and bearing -the impress of a closer connection with the typical American species -than with the Asiatic plant. The view that Rhizophora mangle reached the -Western Pacific from America is rejected, and it is considered that this -species was originally as widely diffused in the Old World as in -America, and that it now survives only in a few places in the tropics of -the Old World. The results of detailed observations on the modes of -dispersal and on the germinating process both with Rhizophora and -Bruguiera are given; and the absence, as a general rule, of any period -of rest between the fecundation of the ovule and the germination of the -seed is established. - -A special chapter is devoted to the significance of vivipary, and it is -considered that a record of the history of vivipary on the globe is -afforded in the scale of germinative capacity that begins with the -seedling hanging from a mangrove and ends with the seed that is detached -in an immature condition from an inland plant. It is suggested that with -the drying up of the planet in the course of ages the viviparous habit, -which was once nearly universal, has been for the most part lost except -in the mangrove swamp, which to some extent represents an age when the -earth was enveloped in cloud and mist and the atmosphere was saturated -with aqueous vapour. The lost habit is at times revived in the abnormal -vivipary of some inland plants, and traces of it are seen in the -abnormal structure of the seeds of some genera of the Myrtaceæ, like -Barringtonia, and in the seeds of genera of other orders. With the -desiccation of the planet and the emergence of the continents there has -been continual differentiation of climate resulting in seasonal -variation and in the development of the rest-period of the seed. - -With the secular drying of the globe and the consequent differentiation -of climate is to be connected the suspension to a great extent of the -agency of birds as plant-dispersers in later ages, not only in the -Pacific Islands but over all the tropics. The changes of climate, bird, -and plant have gone on together, the range of the bird being controlled -by the climate, and the distribution of the plant being largely -dependent on the bird. - -The history of climate, the history of the continents and of the oceans, -the history of life itself, but only in the sense below defined, all -belong to that of a desiccating world, or rather of a planet once -sunless and enveloped in mist and cloud, that through the ages has been -drying up. Life’s types were few and the sea prevailed, and one climate -reigned over the globe. With the diminution of the aqueous envelopes the -continents began to emerge, climates began to individualise, and -organisms commenced to differentiate, and thus the process has run on -through the past, ever from the general to the special both in the -organic and in the inorganic world. - -The same story of a world drying up is told by the marine remains left -stranded far up some mountain slope, or by the bird akin to no other of -its kind that Time has stranded on some island in mid-Pacific. The bird -generalised in type that once ranged the globe is now represented over -its original range by a hundred different groups of descendants, -confined each to its own locality. Climate, once so uniform, now so -diversified, has by restricting the range of the bird favoured the -process of differentiation, and the plant dependent on the bird for its -distribution has in its turn responded to these changes. - -The _rôle_ of the polymorphous species belongs alike to the plant and to -the bird. A species that covers the range of a genus varies at first in -every region and ultimately gives birth to new species in some parts of -its range. Then the wide-ranging species disappears and the original -area is divided up into a number of smaller areas each with its own -group of species. Each smaller area breaks up again, and forms, yet more -specialised, are produced; and thus the process of subdivision of range -and of differentiation of form goes along until each island in an -archipelago owns its bird and each hill and valley has its separate -plants. This is not the path that Evolution takes, since beyond lies -extinction whether of plant or of bird. Such is the upshot of the -process of differentiation exhibited in the development of species and -genera in the Pacific Islands, or, indeed, in any oceanic groups. It can -never do more than produce a Dodo or a Kiwi, or amongst the plants a -Tree-Lobelia. - -Evolution here and elsewhere is a thing apart from species and genera, -which are but eddies on the surface of its stream. It is a scheme of -life introduced into a much conditioned world, and adaptation in endless -forms is the price it has had to pay. The whole story of life on this -earth is a story of a sacrifice, of an end to be won, but of a price to -be paid. Immortality is in the scheme, but death is the price of -adaptation. The same theme runs through our conceptions of the spiritual -life. There is the same duality, evolution adapting its scheme to the -exigencies of the physical world, the good principle ever in conflict -with the evil, and at times compelled to adapt itself to attain its -ends. There is the tale of adaptation in the one case and of sacrifice -in the other, and success is reached in both. - - - - - APPENDIX - - LIST OF NOTES - - -Note 1. On the number of known species of Fijian flowering plants. - -Note 2. The littoral plants of Fiji. - -Note 3. Results of long flotation experiments on the seeds or - seedvessels of tropical littoral plants. - -Note 4. Table illustrating the degree of buoyancy of the seeds and - fruits of inland Fijian plants. - -Note 5. The inland Fijian plants possessing buoyant seeds or fruits. - -Note 6. Table showing the degree of buoyancy of the seeds and fruits of - some inland Hawaiian plants. - -Note 7. Some inland Hawaiian plants possessing buoyant seeds or fruits. - -Note 8. The pyrenes of Morinda. - -Note 9. The buoyancy of the fruits of Calophyllum. - -Note 10. The buoyancy experiments on British plants. - -Note 11. The effect of sea-water immersion on the germinating capacity - of seeds and seed-vessels. - -Note 12. The buoyancy of the fruits of Galium aparine. - -Note 13. The buoyancy of the seeds of Convolvulus sepium. - -Note 14. Other long flotation experiments. - -Note 15. The occurrence inland of Silene maritima. - -Note 16. The buoyancy of the seeds or fruits of the British beach-plants - that also occur inland. - -Note 17. The buoyancy of the seeds or fruits of the British littoral - plants that frequent salt-marshes and muddy shores. - -Note 18. The buoyancy of the seeds or fruits of the British littoral - plants that are confined to the beach. - -Note 19. On germination in sea-water. - -Note 20. On the maximum heights reached by some shore plants in their - extension inland in Vanua Levu, Fiji. - -Note 21. On the dwarfing of shore plants when extending inland in the - “talasinga” plains in Vanua Levu. - -Note 22. The “talasinga” plains of Vanua Levu. - -Note 23. Schimper’s grouping of the Indo-Malayan strand flora. - -Note 24. Grouping of some of the characteristic plants of the strand - flora of Fiji. - -Note 25. The strand flora of the Tahitian region. - -Note 26. The Fijian shore plants not found in Tahiti. - -Note 27. The intruders into the beach flora from the inland plants of - Tahiti. - -Note 28. The littoral plants of the Hawaiian islands. - -Note 29. Botanical notes on the coast plants of the Hawaiian islands. - -Note 30. The beach drift of the Hawaiian islands. - -Note 31. The inland extension of the shore plants of the Hawaiian - islands. - -Note 32. The Fijian species of Premna. - -Note 33. De Candolle’s list of plants dispersed exclusively by currents. - -Note 34. The littoral plants of the eastern-most Polynesian islands. - -Note 35. Distribution of the littoral plants with buoyant seeds or - fruits that occur in the Fijian, Tongan, Samoan, Tahitian, and - Hawaiian Groups. - -Note 36. Hawaiian plants with buoyant seeds or fruits known to be - dispersed by the currents either exclusively or with the assistance of - frugivorous birds. - -Note 37. On vivipary in the fruits of Barringtonia racemosa and Carapa - obovata. - -Note 38. On the temperature and density of the surface water of the - estuaries of the Rewa River in Fiji and of the Guayaquil River in - Ecuador. - -Note 39. On the Pacific species of Strongylodon. - -Note 40. Precautions in testing seed-buoyancy. - -Note 41. The buoyancy of the seeds of Convolvulus soldanella in - fresh-water and sea-water compared. - -Note 42. On secular changes in sea-density. - -Note 43. On the mucosity of small seeds and seed-like fruits when wet. - -Note 44. Upon the effects of inland extension on the buoyancy of the - seeds or fruits of littoral plants. - -Note 45. Tabulated results of the classification, according to - Schimper’s application of the Natural Selection Theory, of the buoyant - seeds and fruits of tropical littoral plants. - -Note 46. On the modes of dispersal of the genus Brackenridgea. - -Note 47. On the transport of gourds by currents. - -Note 48. On the useless dispersal by currents of the fruits of the Oak - and of other species of Quercus, as well as of the Hazel (Corylus). - -Note 49. On the distribution of Ipomœa pes capræ, Convolvulus - soldanella, and Convolvulus sepium. - -Note 50. On the structure of the seeds and fruits of Barringtonia. - -Note 51. On a common inland species of Scævola in Vanua Levu, Fiji. - -Note 52. On the capacity for dispersal by currents of Colubrina - oppositifolia. - -Note 53. On the genus Erythrina. - -Note 54. On the genus Canavalia. - -Note 55. The inland extension of Scævola kœnigii. - -Note 56. On the capacity for dispersal by currents of Sophora tomentosa, - S. chrysophylla, and S. tetraptera. - -Note 57. On the species of Ochrosia. - -Note 58. On Pandanus. - -Note 59. Seeds in petrels. - -Note 60. Schimper on the halophilous character of littoral Leguminosæ - and of shore plants generally. - -Note 61. Meteorological observations on the summit of Mauna Loa. - -Note 62. On the relative proportion of vascular cryptogams in Fiji. - -Note 63. On the table of vascular cryptogams of Tahiti, Hawaii, and - Fiji. - -Note 64. On the distribution of the Tahitian ferns and lycopods. - -Note 65. Distribution of some of the mountain ferns of Hawaii that are - not found either in Fiji or in Tahiti. - -Note 66. Endemic genera of ferns in Hawaii. - -Note 67. On the dispersal of Compositæ by birds. - -Note 68. On some of the Hawaiian endemic genera excluding those of the - Compositæ and Lobeliaceæ. - -Note 69. On the germination of Cuscuta. - -Note 70. On beach-temperature. - -Note 71. On the buoyancy of the seeds or seed-vessels of some Chilian - shore plants. - -Note 72. On the southern limit of the mangrove formation in Ecuador. - -Note 73. Additional note on the temperature of the dry coast of Ecuador, - between the island of Puna and the equator. - -Note 74. Observations on the temperature of the Humboldt current from - Antofagasta northward between January and March, 1904. - -Note 75. On the stranded massive corals of the genus Porites (?) found - on the coast of North Chile and Peru at Arica, Callao, and Ancon. - -Note 76. Stranded pumice on English and Scandinavian beaches. - -Note 77. On the mode of dispersal of Kleinhovia hospita. - -Note 78. On the “Sea”, an unidentified wild fruit tree in Fiji. - -Note 79. On willow-leaved river-side plants. - -Note 80. Mr. Perkins on the Hawaiian Lobeliaceæ. - -Note 81. On the vertical range of some of the most typical and most - conspicuous of the plants in the forests on the Hamakua slopes of - Mauna Kea, Hawaii. - -Note 82. Aboriginal weeds. - -Note 90. On the buoyancy of the seeds of Euphorbia amygdaloides and E. - segetalis. - -Note 91. Mr. E. Kay Robinson on Aster tripolium. - - - NOTE 1 (page 13) - - ON THE NUMBER OF KNOWN SPECIES OF FIJIAN FLOWERING PLANTS - -Rather over 600 species of flowering plants are included in Seemann’s -_Flora Vitiensis_, excluding the weeds and the plants introduced by man. -Horne’s collections would probably add another 300 species; and many -more remain to be discovered. - - - NOTE 2 (page 13) - - THE LITTORAL PLANTS OF FIJI - -In the following table are incorporated the results of an extensive -series of observations and experiments on the buoyancy of the seeds and -fruits of the shore plants made by the author during his sojourn of two -years in Fiji, and based not only on prolonged buoyancy-tests, but also -on systematic examination of the stranded and floating seed-drift, both -of sea and river. The details would occupy many chapters: and it is only -possible here to give the bare results. Since Professor Schimper went -over much the same ground in the Malayan region, one enjoys in many -cases the great advantage of his authority; but a fair proportion of the -results are new; and, besides, there are a number of plants included, -the buoyancy of whose seeds or fruits has long been well established. In -all cases the seed or fruit is taken as it presents itself for dispersal -by the currents. Many of the plants are discussed with some detail in -various parts of this book, as indicated in the reference column of the -table. - -Since the Gramineæ and the Cyperaceæ contain very few species suited for -direct transport by the currents over wide areas of sea, this list may -be regarded as containing nearly all the littoral flowering plants -possessing seeds or seed-vessels with any buoyancy of importance. - -Nearly all the Tahitian strictly littoral plants are represented in -Fiji, and the few that have not been found there yet, such as Sesbania -grandiflora, Heliotropium anomalum, &c., may exist, as in the -first-named species, in the neighbouring Tongan group, and may probably -even exist in Fiji. Two other Tahitian littoral plants, that are widely -spread in the Pacific, namely, Suriana maritima and Sesuvium -Portulacastrum, are found in Tonga, and are included in my list of -Fijian shore plants, though not yet recorded from that group, where, -however, they will, without a doubt, be found by some future observer. - - TABLE SHOWING THE BUOYANCY OF THE SEEDS OR FRUITS OF THE LITTORAL PLANTS - OF FIJI, EXCLUDING THE GRASSES AND, WITH ONE EXCEPTION, THE SEDGES - -The letters placed before the plant name indicate that the species is -also found in Hawaii (H), in Tahiti (T), and in the Marquesas (M). The -Marquesan locality is only given where the plant is not in Tahiti. - -The abbreviations in the reference column are as follows: - -S=Schimper; G=Guppy; P=Earlier authorities and particularly the list -given by Hemsley in the Introduction to the _Botany of the Challenger -Expedition_. - - Column headings: - - A. Buoyancy of seeds or fruits. - B. Float for months. - C. Sink at once or in a week or two. - D. Authorities. - E. Pages of further reference. See also Index. - - +--------------------------------------+--------------+-------+-------+--------+ - | | | A | | | - | Species. | Family. +---+---+ D | E | - | | | B | C | | | - +--------------------------------------+--------------+---+---+-------+--------+ - |HT Calophyllum inophyllum |Guttiferæ. | + |...|S.G.P. | 18 | - | | | | | | | - |HT Hibiscus tiliaceus |Malvaceæ. | + |...|S.G.P. | 21 | - | | | | | | | - | Hibiscus diversifolius (Jacq.) |Malvaceæ. | + |...|G. | 21 | - | | | | | | | - |HT Thespesia populnea |Malvaceæ. | + |...|S.G.P. | Note 3 | - | | | | | | | - | H Gossypium tomentosum (Nutt.) |Malvaceæ. |...| + |G. | | - | | | | | | | - | Heritiera littoralis |Sterculiaceæ. | + |...|S.G.P. | 45, 48 | - | | | | | | | - | T Kleinhovia hospita |Sterculiaceæ. | + |...|G. | 21 | - | | | | | | | - | T Triumfetta rhomboidea |Tiliaceæ. |...| +?| | | - | | | | | | | - | T Triumfetta procumbens |Tiliaceæ. |...| + |G. | 45 | - | | | | | | | - | T Suriana maritima |Simarubeæ. | + |...|S.G. | | - | | | | | | | - | Carapa moluccensis |Meliaceæ. | + |...|S.G.P. | 45 | - | | | | | | | - | Carapa obovata |Meliaceæ. | + |...|S.G.P. | 45 | - | | | | | | | - | T Ximenia americana |Olacineæ. | + |...|S.G. | 113 | - | | | | | | | - | Smythea pacifica (Seem.) |Rhamneæ. | + |...|G.P. | 106 | - | | | | | | | - |HT Colubrina asiatica |Rhamneæ. | + |...|G. | 137 | - | | | | | | | - |HT Dodonæa viscosa |Sapindaceæ. | + |...|S.G. | | - | | | | | | | - |HT Tephrosia piscatoria |Papilionaceæ. |...| + |G. | 45 | - | | | | | | | - | M Desmodium umbellatum |Papilionaceæ. |...| + |G. | | - | | | | | | | - |HT Dioclea violacea |Papilionaceæ. | + |...|G.P. | 82 | - | | | | | | | - | T Canavalia obtusifolia |Papilionaceæ. | + |...|S.G.P. |Note 54 | - | | | | | | | - | T Canavalia sericea |Papilionaceæ. | + |...|G. |Note 54 | - | | | | | | | - | T Canavalia ensiformis, var. turgida.|Papilionaceæ. | + |...|S.G.P.?|Note 54 | - | | | | | | | - |HT Mucuna gigantea |Papilionaceæ. | + |...|S.G.P. | 81 | - | | | | | | | - | T Erythrina indica |Papilionaceæ. | + |...|S.G.P. | | - | | | | | | | - |HT Strongylodon lucidum |Papilionaceæ. | + |...|G. | 82 | - | | | | | | | - |HT Vigna lutea |Papilionaceæ. | + |...|S.G. | 139 | - | | | | | | | - | Dalbergia monosperma |Papilionaceæ. | + |...|S.G. | 106 | - | | | | | | | - | Derris uliginosa |Papilionaceæ. | + |...|S.G.P. | 111 | - | | | | | | | - | Pongamia glabra |Papilionaceæ. | + |...|S.G.P. | | - | | | | | | | - | T Sophora tomentosa |Papilionaceæ. | + |...|S.G. |Note 56 | - | | | | | | | - | T Inocarpus edulis |Papilionaceæ. | +?|...|G.P. | | - | | | | | | | - |HT Cæsalpinia Bonducella |Cæsalpinieæ. | + |...|S.G.P. | 193 | - | | | | | | | - | T Cæsalpinia Bonduc |Cæsalpinieæ. | + |...|G.P. | 193 | - | | | | | | | - | Afzelia bijuga |Cæsalpinieæ. | + |...|G. | 173 | - | | | | | | | - | Cynometra sp. |Cæsalpinieæ. | +?|...|S.G. | | - | | | | | | | - | Entada scandens |Mimoseæ. | + |...|G.P. | 181 | - | | | | | | | - | Acacia laurifolia |Mimoseæ. |...| + |G. | 164 | - | | | | | | | - | T Leucæna Forsteri |Mimoseæ. |...| + |G. | | - | | | | | | | - | T Serianthes myriadenia |Mimoseæ. |...| + |G. | 424 | - | | | | | | | - | Parinarium laurinum |Rosaceæ. | + |...|G.P. | | - | | | | | | | - | Eugenia Richii |Myrtaceæ. |...| + |G. | | - | | | | | | | - | T Barringtonia speciosa |Myrtaceæ. | + |...|S.G.P. | | - | | | | | | | - | Barringtonia racemosa |Myrtaceæ. | + |...|G. | | - | | | | | | | - | Rhizophora mucronata |Rhizophoreæ. | + |...|S.G.P. | | - | | | | | | | - | Rhizophora mangle |Rhizophoreæ. | + |...|S.G.P. | | - | | | | | | | - | Bruguiera Rheedii |Rhizophoreæ. | + |...|G.P. | | - | | | | | | | - |HT Terminalia Katappa |Combretaceæ. | + |...|S.G.P. | | - | | | | | | | - | M Terminalia littoralis |Combretaceæ. | + |...|S.G.P. | | - | | | | | | | - | Lumnitzera coccinea |Combretaceæ. | + |...|S.G.P. | | - | | | | | | | - | T Gyrocarpus Jacquini |Combretaceæ. | + |...|G. | 423 | - | | | | | | | - | T Pemphis acidula |Lythraceæ. | + |...|S.G. | | - | | | | | | | - | T Luffa insularum (Gray) |Cucurbitaceæ. | + |...|G. | 426 | - | | | | | | | - |HT Sesuvium Portulacastrum |Ficoideæ. |...| + |G. | | - | | | | | | | - |HT Morinda citrifolia |Rubiaceæ. | + |...|S.G.P. | | - | | | | | | | - | T Guettarda speciosa |Rubiaceæ. | + |...|S.G.P. | | - | | | | | | | - | T Wedelia biflora |Compositæ. | + |...|G. | | - | | | | | | | - |HT Scævola Koenigii |Goodeniaceæ. | + |...|S.G.P. | | - +--------------------------------------+--------------+---+---+-------+--------+ - - TABLE SHOWING THE BUOYANCY OF THE SEEDS OR FRUITS OF THE - LITTORAL PLANTS OF FIJI, EXCLUDING THE GRASSES, AND WITH - ONE EXCEPTION, THE SEDGES (_continued_) - - Column headings: - - A. Buoyancy of seeds or fruits. - B. Float for months. - C. Sink at once or in a week or two. - D. Authorities. - E. Pages of further reference. See also Index. - - +--------------------------------------+--------------+-------+-------+--------+ - | | | A | | | - | Species. | Family +---+---+ D | E | - | | | B | C | | | - +--------------------------------------+--------------+---+---+-------+--------+ - | T Cerbera Odollam |Apocynaceæ. | + |...|S.G.P. | | - | | | | | | | - | T Ochrosia parviflora |Apocynaceæ. | + |...|G.P. | | - | | | | | | | - |HT Cordia subcordata |Boraginaceæ. | + |...|S.G.P. | | - | | | | | | | - | T Tournefortia argentea |Boraginaceæ. | + |...|S.G.P. | | - | | | | | | | - |HT Ipomœa pes capræ |Convolvulaceæ.| + |...|S.G.P. | | - | | | | | | | - | H Ipomœa glaberrima (Boj.) |Convolvulaceæ.| + |...|G. | | - | | | | | | | - | Aniseia uniflora |Convolvulaceæ.| + |...|G. | | - | | | | | | | - | T Premna tahitensis |Verbenaceæ. | + |...|G. |Note 32 | - | | | | | | | - | Clerodendron inerme |Verbenaceæ. | + |...|S.G. | | - | | | | | | | - |HM Vitex trifolia |Verbenaceæ. | + |...|G. | | - | | | | | | | - |HT Cassytha filiformis |Lauraceæ. | + |...|G. | | - | | | | | | | - | T Hernandia peltata |Lauraceæ. | + |...|S.G. | | - | | | | | | | - |HT Wikstrœmia fœtida |Thymelæaceæ. |...| + |G. | | - | | | | | | | - | Drymispermum Burnettianum |Thymelæaceæ. |...| + |G. | | - | | | | | | | - | T Euphorbia Atoto |Euphorbiaceæ. |...| + |S.P.G. | | - | | | | | | | - | Excæcaria Agallocha |Euphorbiaceæ. | + |...|S.G. | | - | | | | | | | - | T Casuarina equisetifolia |Casuarineæ. |...| + |G. | | - | | | | | | | - |HT Tacca pinnatifida |Taccaceæ. | + |...|G. | 19 | - | | | | | | | - |HT Cocos nucifera |Palmeæ. | + |...|P. | | - | | | | | | | - |HT Pandanus odoratissimus |Pandaneæ. | + |...|S.G.P. | | - | | | | | | | - | Crinum asiaticum |Amaryllideæ. |...| + |G.P. | | - | | | | | | | - | Scirpodendron costatum |Cyperaceæ. | + |...|G. | 407 | - | | | | | | | - | Cycas circinalis |Cycadeæ. | + |...|G.P. | | - +--------------------------------------+--------------+---+---+-------+--------+ - - - NOTE 3 (page 13) - - RESULTS OF LONG FLOTATION EXPERIMENTS ON THE SEEDS OR SEED-VESSELS OF - TROPICAL LITTORAL PLANTS - -At various times during the past twenty years I have made lengthened -experiments in England on the buoyancy in sea-water of the seeds or -seed-vessels of beach plants collected by me in the Solomon Islands, the -Fijis, Hawaii, Keeling Atoll, &c. In all the species enumerated below, -the floating powers were retained after twelve months’ immersion, the -seed-contents being to all appearance unharmed. In six species I -succeeded in getting the seeds to germinate after the experiment; and -there can be no doubt that the number of successful results would have -been largely increased, if I had not been obliged to resort to very -primitive methods in conducting the experiments. Some of the results are -referred to in a note to my paper on the flora of Keeling Atoll, dated -about 1889; and if I remember aright, Mr. Hemsley mentioned those -relating to Thespesia populnea and Ipomœa grandiflora in the _Annals of -Botany_, not long after. The others have not been previously published. -In one instance (Cæsalpinia bonducella) the flotation experiment was -prolonged to two and a half years, the seeds floating buoyantly and -being apparently quite sound at the end of the experiment. - -As demonstrating that tropical seeds can be transported unharmed by -currents through cold latitudes, it should be noted that all these -experiments were conducted in England. In the cases of the Keeling Atoll -seeds the experiment was carried on through a very severe winter, the -vessel of sea-water being exposed to a degree of cold that kept -fresh-water frozen for three weeks on the same table. This did not -prevent the subsequent germination of the seeds of Thespesia populnea -and Ipomœa grandiflora. The same thing was established in a more natural -way by Lindman, who planted seeds of Entada scandens and Mucuna urens, -that had been stranded on the Norwegian coast, and found that they -retained their germinating capacity (see Sernander, p. 7). - -The following are the seeds or seed-vessels that remained afloat after a -year’s flotation in sea-water, those that subsequently germinated being -preceded by G. In the other cases the germinating capacity was not -tested; but they were always sound in appearance when cut across at the -close of the experiment. - - G Thespesia populnea (Malvaceæ) - Dioclea (violacea?) (Papilionaceæ) - G Mucuna gigantea, D C (Papilionaceæ) - G Mucuna urens, D C (Papilionaceæ) - Mucuna, sp. (Papilionaceæ) - Mucuna, sp. (Papilionaceæ) - G Strongylodon lucidum, Seem. (Papilionaceæ) - Sophora tomentosa, (Papilionaceæ) - G Cæsalpinia bonducella (Cæsalpinieæ) - Entada scandens (Mimoseæ) - Morinda citrifolia (Rubiaceæ) - Scævola Koenigii (Goodeniaceæ) - Cordia subcordata (Boragineæ) - Tournefortia argentea (Boragineæ) - G Ipomœa grandiflora, Lam. (Convolvulaceæ) - Tacca pinnatifida (Taccaceæ) - - - NOTE 4 (page 13) - - TABLE ILLUSTRATING THE DEGREE OF BUOYANCY OF THE SEEDS AND FRUITS OF - INLAND FIJIAN PLANTS - - (Unless otherwise indicated, the seeds or fruits sink at once or in a - day or two) - - Abrus precatorius. - Acacia Richii. - Ageratum conyzoides. - Alphitonia excelsa. - Alpinia sp. - Alyxia (scandens?). - Artocarpus incisa. - Artocarpus integrifolia. - Barringtonia edulis (1 month) - Barringtonia sp. - Bauhinia sp. - Bischoffia javanica. - Cæsalpinia sp. - Calophyllum spectabile (2-4 weeks). - Calophyllum Burmanni (4-10 days). - Cananga odorata. - Canarium sp. - Canarium sp. - Canna indica. - Citrus aurantium (3-4 weeks). - Citrus decumana (1 month). - Citrus limonum (5 weeks). - Citrus vulgaris, R. (6-7 weeks). - Coix lachryma (2-7 days). - Commersonia platyphylla. - Cordyline sepiaria. - Couthovia corynocarpa (a few days). - Cucumis acidus (a few days). - Cucurbita sp. (several months). - Cupania sp. - Dammara vitiensis (7-10 days). - Dioscorea sativa (a few days). - Dioscorea sp. - Dracontomelon sylvestre. - Dracontomelon sp. - Elæocarpus sp. - Elæocarpus sp. (a few days). - Eranthemum sp. - Eugenia malaccensis (2-4 weeks). - Eugenia effusa? (4-7 days). - Eugenia confertiflora? (10-12 days). - Eugenia rariflora (a few days). - Eugenia corynocarpa (a few days). - Eugenia rivularis (a week). - Fagræa Berteriana (a few days). - Ficus Harveyi (7-10 days). - Ficus scabra (7-10 days). - Ficus sp. (7-10 days). - Gardenia vitiensis (4-5 weeks). - Geissois ternata. - Geophila reniformis. - Gnetum gnemon. - Grewia sp. - Guettarda sp. (a few weeks). - Hibiscus Abelmoschus (months). - Hibiscus seculentus. - Hydrocotyle asiatica (months). - Ipomœa batatas. - Ipomœa insularis (_nil_ or months). - Ipomœa peltata (weeks or months). - Ipomœa turpethum (_nil_ or weeks or months). - Ipomœa sp. (7-10 days). - Lindenia vitiensis (weeks or months). - Maba sp. (7-10 days). - Macaranga sp. (1-2 weeks). - Melastoma denticulatum. - Micromelum minutum. - Momordica Charantia (a few days). - Morinda Forsteri. - Mussænda frondosa. - Myristica sp. (3-7 days) - Myristica sp. (3-7 days) - Myrmecodia sp. - Nelitris vitiensis (a few days). - Nephelium pinnatum (a few days). - Ophiorrhiza leptantha. - Phyllanthus sp. - Phyllanthus sp. - Piper Macgillivrayi. - Pittosporum sp. - Pleiosmilax vitiensis. - Portulaca (lutea?). - Portulaca quadrifida. - Premna serratifolia. - Pritchardia pacifica. - Psychotria sp. - Psychotria sp. - Psychotria sp. - Psychotria sp. - Psychotria sp. - Ptychosperma sp. - Rhaphidophora vitiensis. - Sapota sp. (a few days) - Sapota sp. (a few days) - Scævola floribunda. - Spondias dulcis (a month). - Sterculia sp. (seeds _nil_, fruits months). - Stylocoryne sambucina (2 or 3 days). - Tabernæmontana (orientalis?) (a few days). - Tacca maculata (_nil_ or a few days). - Trichospermum Richii (a few days). - Urena lobata. - Veitchia Joannis. - Veitchia sp. - - - NOTE 5 (page 14). - - THE INLAND FIJIAN PLANTS POSSESSING BUOYANT SEEDS OR FRUITS - -They come under the following heads: - -(a) Plants of the stream-border or the pond-side or of the inland swamp, -_e.g._, Lindenia vitiensis and Hydrocotyle asiatica. The extension of -the principle by which plants with buoyant seeds or fruits are located, -not only at the sea-side but at the water-side generally, is here -involved, as explained in Chapter III. - -(b) Plants following the rule deduced by Schimper for Terminalia, that -when a genus comprises several species possessing buoyant fruits, only -those having fruits with the greatest floating power are found at the -coast, the least buoyant plants occurring inland; examples, Calophyllum -and Guettarda. - -(c) Plants that like Ipomœa behave irregularly in respect to -seed-buoyancy, a difference in behaviour often associated with varying -stations both at the coast and inland. - -(d) Plants with dehiscent buoyant capsular fruits, like Sterculia, where -dehiscence takes place on the tree and the seeds have no buoyancy. -Although the unopened fruit may float a long time, it does not in that -condition come under the influence of the currents. - -(e) Plants like Citrus Decumana, Gardenia, sp., &c., that, although -apparently exceptions to the principle, do not offer much opposition to -it, since the first is most at home at the river-side and the second -often displays a decided inclination for a station at the coast. - -(f) Genuine exceptions to the principle, such as Hibiscus Abelmoschus -(see page 21). - - - NOTE 6 (page 15) - - TABLE SHOWING THE DEGREE OF BUOYANCY OF THE SEEDS AND FRUITS OF SOME - INLAND HAWAIIAN PLANTS - - (Unless otherwise stated, the seeds or fruits sink at once or in a day - or two) - - Acacia Koa. - Aleurites moluccana (1-2 weeks). - Alyxia olivæformis. - Argemone mexicana. - Argyreia tiliæfolia (_nil_ or months). - Bidens pilosa. - Campylotheca sp. - Canavalia galeata. - Capparis sandwicensis. - Cassia Gaudichaudii. - Cassia occidentalis. - Cheirodendron Gaudichaudii. - Colubrina oppositifolia (weeks). - Commelina nudiflora. - Coprosma ernodeoides. - Coprosma sp. - Coprosma sp. - Cyathodes Tameiameiæ (a few days). - Cyrtandra sp. (a few days). - Cyrtandra sp. (a few days). - Cyrtandra sp. (a few days). - Dianella odorata (a few days). - Dracæna aurea. - Eclipta alba (months). - Erythrina monosperma. - Gossypium tomentosum (a week). - Gossypium barbadense (a few days). - Gossypium sp. cultiv. (a few days). - Hibiscus Youngianus (weeks). - Hydrocotyle verticillata (weeks). - Ipomœa bona nox (_nil_ or months). - Ipomœa insularis. - Ipomœa pentaphylla. - Ipomœa reptans. - Ipomœa tuberculata. - Jacquemontia sandwicensis. - Jussiæa villosa (a few days). - Lobeliaceæ (Clermontia). - Maba sandwicensis. - Metrosideros polymorpha. - Mezoneuron kauaiense (pod, a week). - Mucuna urens (months). - Myoporum sandwicense. - Olea sandwicensis, see page 364. - Phyllostegia grandiflora. - Phyllostegia mollis. - Plectronia odorata. - Pritchardia Gaudichaudii (5 or 6 weeks). - Ricinus communis (7-10 days). - Rubus Macraei. - Scævola Chamissoniana. - Scævola Gaudichaudii. - Sida fallax. - Sisyrinchium acre. - Solanum aculeatissimum. - Sophora chrysophylla (pod, 1-2 weeks). - Viola Chamissoniana. - Waltheria americana. - - - NOTE 7 (page 15) - - SOME INLAND HAWAIIAN PLANTS POSSESSING BUOYANT SEEDS OR FRUITS - -Three of these, Eclipta alba, Hibiscus Youngianus, and Hydrocotyle -verticillata, frequent wet places, and come under the principle that -water-side plants generally have buoyant seeds or fruits. The buoyancy -of the seeds of Argyreia tiliæfolia and of Ipomœa bona nox varies with -station and may be explained as under Ipomœa in Note 5. The floating -power of the fruits of Colubrina oppositifolia may be akin to that of -inland species of Terminalia as indicated in Note 5, since another -species of the genus C. asiatica, which is a coast plant, has very -buoyant seeds. Mucuna urens was no doubt originally, as it now is in -tropical America, a littoral plant. The buoyant fruits of Pritchardia -Gaudichaudii offer a genuine exception to the principle (see page 330). - - - NOTE 8 (pages 18, 112) - - THE PYRENES OF MORINDA - -The pyrenes of the two Malayan inland species of Morinda (M. umbellata -and M. longiflora) examined by Professor Schimper do not possess the -bladder-like cavity to which those of M. citrifolia owe their floating -power, and it is to be inferred from his remarks (p. 183) that they have -little or no buoyancy. The pyrenes of a Fijian inland species, near M. -Grayi, had no floating power as tested by me, and they lacked the -bladder-like cavity. - - - NOTE 9 (page 18) - - THE BUOYANCY OF THE FRUITS OF CALOPHYLLUM - -Professor Schimper found that whilst the fruits of Calophyllum -inophyllum, the shore tree, remained afloat after 126 days, those of C. -amœnum, an inland species, sank in from three to fourteen days, both -possessing similar buoyant structures, but to a less degree in the case -of the inland species. This genus presents a parallel case to Terminalia -referred to on page 17; but the general discussion of the subject will -be found in Chapter XIII. According to the above authority C. Calaba, a -West Indian coast tree, has buoyant fruits. The same is also true of the -fruits of a large inland tree in the Solomon Islands experimented on by -me (_Solomon Islands_, p. 305). It would thus appear that the fruits of -the genus are as a rule buoyant, and that, as in Terminalia, the least -buoyant fruits belong to the inland species. Professor Schimper also -shows (p. 182) that the diminished floating power of the fruits of the -inland species is associated with diminution in thickness of the buoyant -seed-shell which is most developed in the buoyant fruits of the strand -species. - - - NOTE 10 (page 24) - - THE BUOYANCY EXPERIMENTS ON BRITISH PLANTS - -The experiments in all cases were made to test the floating power of the -seed or fruit in the condition in which it is detached from the plant. -It usually makes very little difference whether sea-water or fresh water -is employed, since in my numerous experiments there were but few -exceptions to the general rule that seeds or seed-vessels that sink in -fresh water sink also in sea-water. This subject is discussed in Chapter -X. However, it may be here observed that the chief effect of the -increased density of sea-water is merely to increase the proportion of -buoyant seeds or fruits in any particular species. - -It is necessary in such experiments to imitate Nature as much as -possible. The seed or fruit, as the case may be, must be experimented -upon in the condition in which it falls from the plant, or in the -condition in which it would be ultimately found in river and pond drift. -The seed or fruit should be thoroughly wetted, and air-bubbles removed. - -Prolonged drying has but a slight effect on the great majority of seeds -and seed-vessels experimented on, and this is just as true of tropical -plants. Those that sink at once in the mature and fresh condition rarely -float more than a day or two even after drying for a year. The usual -effect is to increase the floating capacity of seeds and fruits already -buoyant, and not to develop the capacity. - -The results given in the table refer only to sound seeds. In fresh-water -experiments, in nearly all cases, the seeds ultimately germinate in the -water, and this is the usual cause of the close of the experiment. In an -ordinary collection of floating seed drift from a pond or river, -germination will go on for years at each successive spring, the -postponement of germination being a very striking feature with a fair -proportion of seeds in river and pond-drift. This subject is dealt with -in detail in my paper published in the _Proceedings_ for 1897 of the -Royal Physical Society of Edinburgh. - - THE TABLE OF RESULTS OF OBSERVATIONS AND EXPERIMENTS ON THE - BUOYANCY OF THE SEEDS OR SEED-VESSELS OF MORE THAN 300 - BRITISH FLOWERING PLANTS - -EXPLANATION OF TABLE.—The capacity of floating for months is thus -indicated, ++; of floating for 1 to 4 weeks, +; and where sinking occurs -at once or within a week there is no entry. When buoyancy continued in -my experiments after 6 and 12 months, it is indicated by Roman numerals -(VI and XII). A=an aquatic plant; M=a beach plant; R=a river-side or -pond-side plant; var.=variable in floating power. - - +--------------------------------------+----+------+ - |R Thalictrum flavum | | | - |A Ranunculus aquatilis | | | - |R Ranunculus hederaceus | | | - |R Ranunculus flammula | | | - | Ranunculus ficaria | | | - |R Ranunculus sceleratus VI | ++ | Var. | - | Ranunculus acris | | | - | Ranunculus repens XII | ++ | Var. | - |R Caltha palustris | + | | - | Berberis vulgaris | | | - |A Nymphæa alba | | | - |A Nuphar luteum | | | - | Papaver rhœas | | | - | Papaver dubium | | | - | Chelidonium majus | | | - | Rœmeria hybrida | | | - |M Glaucium luteum | | | - | Barbarea vulgaris | | | - |R Nasturtium officinale | | | - |R Nasturtium sylvestre | | | - |R Nasturtium amphibium | | | - | Arabis hirsuta | | | - | Arabis thaliana | | | - |R Cardamine pratensis | | | - | Cardamine hirsuta | | | - | Alliaria officinalis | | | - | Brassica campestris | | | - | Brassica alba | | | - |M Cochlearia officinalis | | | - |M Alyssum maritimum | | | - | Draba verna | | | - | Thlaspi arvense | | | - | Capsella bursa pastoris | | | - |M Cakile maritima | + | | - |M Crambe maritima | + | | - |M Raphanus maritimus | + | | - | Reseda luteola | | | - | Helianthemum vulgare | | | - |R Viola palustris | | | - | Viola canina | | | - | Viola tricolor | | | - | Polygala vulgaris | | | - | Silene cucubalus | | | - |M Silene maritima | | | - | Lychnis diurna | | | - | Sagina procumbens | | | - |M Arenaria peploides (Honckeneya) XII | ++ | | - | Mœnchia erecta | | | - | Cerastium vulgatum | | | - |R Stellaria aquatica | | | - | Stellaria media | | | - | Stellaria graminea | | | - | Stellaria holostea | | | - | Spergularia rubra | | | - |M Spergularia marina | | | - | Spergula arvensis | | | - |R Montia fontana | | | - |R Elatine hydropiper | | | - | Hypericum perforatum | | | - | Hypericum quadrangulum | | | - |R Hypericum elodes | | | - | Linum usitatissimum | | | - | Linum angustifolium | | | - | Malva rotundifolia | | | - | Malva sylvestris | | | - | Oxalis acetosella | | | - | Oxalis corniculata | | | - | Impatiens parviflora | | | - |R Impatiens fulva VI | ++ | | - | Acer campestre | | | - | Ilex aquifolium | | | - | Euonymus europæus | + | | - | Ulex europæus | | | - | Cytisus scoparius | | | - | Ononis arvensis | | | - | Medicago lupulina | | | - | Medicago denticulata | | | - | Trifolium incarnatum | | | - | Lotus corniculatus | | | - | Anthyllis vulneraria | | | - | Vicia sativa | | | - | Lathyrus pratensis | | | - |M Lathyrus maritimus | ++ | | - |R Spiræa ulmaria | + | | - | Fragaria vesca | | | - | Potentilla tormentilla | | | - | Potentilla sp. | | | - |R Potentilla comarum XII | ++ | | - | Alchemilla arvensis | | | - | Rosa arvensis | | | - | Cratægus oxyacantha | + | | - |R Epilobium hirsutum | | | - | Epilobium parviflorum | | | - |R Lythrum salicaria | | | - |R Peplis portula | | | - | Bryonia dioica | | | - | Cotyledon umbilicus | | | - | Saxifraga granulata | | | - | Saxifraga tridactylites | | | - |R Chrysosplenium alternifolium | | | - |R Chrysosplenium oppositifolium | | | - | Drosera rotundifolia | + | | - |A Myriophyllum spicatum | | | - |A Myriophyllum alternifolium | | | - |R Hydrocotyle vulgaris XII | ++ | | - |M Eryngium maritimum | | | - |R Cicuta virosa | ++ | | - |R Apium graveolens | | | - |R Apium nodiflorum | | | - |R Apium inundatum | | | - |R Sium latifolium | ++ | | - |R Sium angustifolium | ++ | | - |R Œnanthe crocata VI | ++ | | - |R Œnanthe phellandrium | + | | - | Æthusa cynapium | | | - |M Crithmum maritimum | ++ | | - |R Angelica sylvestris XII | ++ | | - |R Peucedanum palustre | ++ | | - | Pastinaca sativa | | | - | Chærophyllum sylvestre | | | - | Smyrnium olusatrum | | | - | Hedera helix | | | - | Viscum album | | | - | Sambucus nigra | | | - |R Galium palustre VI | ++ | | - | Galium mollugo | | | - | Galium aparine | | | - | Centranthus ruber | | | - | Valerianella olitoria | | | - | Eupatorium cannabinum | | | - |M Aster tripolium | | | - |R Bidens cernua VI | ++ | | - |R Bidens tripartita VI | ++ | | - | Chrysanthemum segetum | | | - | Chrysanthemum leucanthemum | | | - | Matricaria inodora | | | - |M Matricaria inodora, var. maritima | ++ | | - | Matricaria chamomilla | | | - | Achillea millefolium | | | - | Artemisia vulgaris | | | - | Artemisia absinthium | | | - | Tussilago farfara | | | - | Tussilago petasites | | | - | Senecio vulgaris | | | - |R Senecio aquaticus | | | - |R Senecio palustris | | | - | Carduus nutans | | | - | Carduus lanceolatus | | | - |R Carduus palustris | | | - | Carduus arvensis | | | - | Tragopogon pratensis | | | - | Tragopogon porrifolius | | | - | Tragopogon echioides | | | - | Leontodon autumnalis | | | - | Sonchus oleraceus | | | - | Taraxacum dens leonis | | | - | Crepis virens | | | - | Crepis fœtida | | | - | Lapsana communis | | | - |A Lobelia Dortmanna | | | - |R Hottonia palustris | | | - |R Lysimachia vulgaris | + | | - |R Lysimachia thyrsiflora | + | | - |M Glaux maritima | | | - | Anagallis arvensis | | | - |M Samolus valerandi | | | - | Pinguicula lusitanica | | | - | Ligustrum vulgare | | | - |R Menyanthes trifoliata | ++ | | - |R Limnanthemum nymphæoides | + | | - | Convolvulus arvensis | | | - | Convolvulus sepium XII | ++ | | - |M Convolvulus soldanella XII | ++ | | - | Cuscuta europæa | | | - | Lithospermum officinale | | | - |R Myosotis palustris | | | - | Myosotis arvensis | | | - | Myosotis versicolor | | | - | Lycopsis arvensis | | | - | Symphytum officinale | | | - | Borago officinalis | | | - | Datura stramonium | | | - | Solanum dulcamara | | | - | Solanum nigrum | | | - | Linaria vulgaris | | | - | Linaria cymbalaria | | | - | Scrophularia nodosa | | | - |R Scrophularia aquatica | | | - |R Veronica anagallis | | | - |R Veronica beccabunga | | | - | Veronica agrestis | | | - | Veronica arvensis | | | - | Bartsia odontites | | | - | Rhinanthus crista galli VI | ++ | Var. | - |R Pedicularis palustris | ++ | | - | Salvia verbenaca | | | - |R Lycopus europæus XII | ++ | | - |R Mentha aquatica VI | ++ | | - | Thymus sp. | | | - | Calamintha officinalis | | | - | Nepeta glechoma | | | - | Nepeta cataria | | | - | Prunella vulgaris | | | - |R Scutellaria galericulata XII | ++ | | - | Stachys betonica | | | - | Stachys sylvatica | | | - |R Stachys palustris | ++ | | - | Stachys arvensis | | | - | Galeopsis tetrahit | | | - | Ballota nigra | | | - | Lamium purpureum | | | - | Lamium album | | | - | Lamium galeobdolon | | | - | Teucrium scorodonia | | | - | Ajuga reptans | | | - | Verbena officinalis | | | - |M Armeria vulgaris | | | - | Plantago major | | | - | Plantago media | | | - | Plantago lanceolata | | | - |M Plantago maritima | | | - |M Salicornia herbacea | | | - |M Salsola kali | + | | - |M Suæda fruticosa | | | - |M Suæda maritima | | | - | Chenopodium album | | | - |M Beta maritima | | | - | Atriplex patula VI | ++ | | - |R Rumex aquaticus | + | | - | Rumex crispus | ++ | | - | Rumex obtusifolius | | | - |R Rumex hydrolapathum XII | ++ | | - |R Rumex conglomeratus XII | ++ | | - | Polygonum aviculare | | | - |M Polygonum maritimum | | | - | Polygonum convolvulus | | | - |R Polygonum amphibium | | | - |R Polygonum persicaria | | | - | Polygonum lapathifolium | | | - |R Polygonum hydropiper | | | - | Euphorbia helioscopia | | | - | Euphorbia segetalis | | | - |M Euphorbia paralias | ++ | | - | Euphorbia peplus (?) | | | - | Euphorbia amygdaloides | | | - |A Ceratophyllum demersum | | | - |A Callitriche aquatica | | | - | Urtica dioica | | | - | Ulmus campestris | | | - |R Alnus glutinosa XII | ++ | | - | Betula alba | | | - | Corylus avellana | + | | - | Quercus robur | + | | - | Pinus sylvestris | + | | - | Taxus baccata | | | - |R Typha latifolia | | | - |R Typha angustifolia | | | - |R Sparganium ramosum XII | ++ | | - |R Sparganium simplex VI | ++ | Var. | - |R Sparganium minimum | ++ | | - | Arum maculatum | | | - |R Calla palustris XII | ++ | | - |A Lemna minor | ++ | | - |A Lemna gibba | | | - |A Naias marina | | | - |A Zannichellia palustris | | | - |A Ruppia maritima | | | - |A Potamogeton natans XII | ++ | | - |A Potamogeton oblongus VI | ++ | | - |A Potamogeton lucens VI | ++ | | - |A Potamogeton perfoliatus | ++ | | - |A Potamogeton crispus | | | - |A Potamogeton densus | | | - |A Potamogeton obtusifolius | | | - |A Potamogeton pusillus | | | - |R Butomus umbellatus | | | - |R Sagittaria sagittifolia VI | ++ | Var. | - |R Alisma plantago VI | ++ | Var. | - |R Alisma ranunculoides | | | - |A Alisma natans | | | - |R Damasonium stellatum | | | - |R Scheuchzeria palustris | ++ | | - |R Triglochin palustre | | | - |R Triglochin maritimum | | | - |A Hydrocharis morsus ranæ | | | - |R Iris pseudacorus XII | ++ | | - | Iris fœtidissima | | | - | Tamus communis | | | - | Fritillaria meleagris | ++ | | - | Scilla nutans | | | - | Narthecium ossifragum | | | - |R Juncus communis | | | - | Juncus glaucus | | | - |R Juncus articulatus | | | - | Juncus squarrosus | | | - |R Juncus bufonius | | | - |M Juncus maritimus | | | - | Luzula campestris | | | - |R Cladium mariscus | ++ | | - |R Blysmus rufus | ++ | | - |R Scirpus palustris | | | - |A Scirpus fluitans | | | - | Scirpus setaceus | | | - | Scirpus holoschœnus | | | - |R Scirpus lacustris | | | - |M Scirpus maritimus | + | | - | Scirpus sylvaticus | | | - | Eriophorum alpinum | | | - | Eriophorum vaginatum | | | - | Eriophorum polystachion | | | - |R Carex leporina | ++ | | - | Carex stellulata XII | ++ | | - |R Carex canescens | ++ | | - | Carex remota XII | ++ | | - |R Carex paniculata XII | ++ | | - |R Carex vulpina XII | ++ | | - |R Carex acuta VI | ++ | | - | Carex hirta | | | - |R Carex flava VI | ++ | Var. | - | Carex distans | | | - | Carex panicea | | | - |R Carex pseudocyperus VI | ++ | | - |R Carex ampullacea | + | | - |R Carex paludosa XII | ++ | | - | Carex sp. | | | - | Carex sp. | | | - |R Leersia oryzoides | | | - |R Alopecurus geniculatus | | | - | Agrostis sp. | | | - | Agrostis sp. | | | - |R Poa aquatica | | | - |R Poa fluitans | | | - | Melica nutans | | | - |R Arundo phragmites | +? | | - +--------------------------------------+----+------+ - -Total of the original list: 320 species belonging to 192 genera and 65 -families. Of these, about 260 were tested by the author, the data for -the remaining species being mainly derived from the writings of Thuret, -Kolpin-Ravn, and Sernander, with a few from those of Darwin and Martins. - -NOTE.—Whilst this work has been going through the press, the author has -added thirteen species, seven genera, and two families to the list above -given; but the general inferences are not affected by the additions. The -corrected total would, therefore, be 333 species, 199 genera, and 67 -families. - - - _On the effect of drying on the buoyancy of seeds and seed-vessels_ - -It has been already observed that this is as a rule but slight, and that -in the great majority of cases the effect of prolonged drying for many -months, or even for years, is at the most to give a seed or fruit -originally non-buoyant a floating power of a few days’ duration. This is -a subject to which I have paid especial attention in my experiments, -since, of course, much depends on it in the way of dispersal by -currents. It is obvious that a seed or fruit possessing impermeable -coverings at the time of its separation from the parent can scarcely be -compared with one where the coverings only attain their water-proof -capacity by drying. Most gardeners know that seeds which dry easily take -up moisture easily, and the principle applies in a varying degree to the -great majority of seeds and fruits. - -Darwin was inclined to attach importance to adventitious buoyancy -acquired by drying; and in the _Origin of Species_ he refers to -instances offered by the fruits of the Hazel (Corylus), the Asparagus, -and Heliosciadium. In Note 48 I have referred to the cases of the Oak -and the Hazel; and, indeed, we have only to examine the beach-drift in -various parts of the world, and to look at their respective stations, to -learn that this is not an effective mode of dispersal. Buoyancy of seed -or fruit is only one of many other qualities that is concerned with -distribution by currents. Nature does not act in this way in -seed-distribution, and there can be little doubt that the author of the -_Origin of Species_ would have been the first to abandon this view, if -his researches had been continued. It should be especially noted that -plants of the sea-beach, where the floating power happens to be _nil_, -or limited only to a week or two, would have derived great advantage -from the drying of their seeds or fruits if it was really effective in -aiding dispersal by currents. However, with plants like Cakile maritima, -Eryngium maritimum, Glaucium luteum, &c., the effect of drying is very -small. - - - NOTE 11 (page 25) - - THE EFFECT OF SEA-WATER IMMERSION ON THE GERMINATING CAPACITY OF SEEDS - AND SEED-VESSELS - -Berkeley, Darwin, Martins, and others, long ago established the capacity -of seeds to germinate after prolonged immersion in sea-water. The reader -will find a _resumé_ of their results in the appendix to Mr Hemsley’s -volume on the _Botany of the Challenger Expedition_. The subject is well -illustrated in the original papers of those authors, and in my later -papers on the flora of Keeling Atoll, and on the seed-drift of the -Thames. - -I may here remark that the earlier observers often pay more attention to -the retention of the germinating capacity after sea-water immersion than -to the degree of buoyancy. For this reason I have not been able to make -great use of the buoyancy results of Martins, since he frequently does -not distinguish between temporary and long-sustained buoyancy, an -objection also pointed out by Thuret and Hemsley. - - - NOTE 12 (page 27) - - THE BUOYANCY OF THE FRUITS OF GALIUM APARINE - -Norman and Sernander (see p. 172) attribute considerable buoyancy to -these fruits on account of the hollow cavity in each. I used to find -them in England in floating river-drift in autumn; and Norman observed -them on the Scandinavian beaches. They do not, however, float long, as -the cavity is open; and in two sets of my experiments they sank within a -few days. - - - NOTE 13 (page 29) - - THE BUOYANCY OF THE SEEDS OF CONVOLVULUS SEPIUM - -This plant seeded freely in 1893 in the Lower Thames Valley, as at -Molesey. I kept some of the seeds afloat for thirty-three months, of -which the first nine months were spent in sea-water and the rest in -fresh-water. One seed, at the end of the period, germinated healthily in -the fresh-water. - - - NOTE 14 (page 26) - - OTHER LONG FLOTATION EXPERIMENTS - -Whilst keeping my collections of Thames seed-drift in water from year to -year, I obtained a number of records of long “flotations.” Thus in -several cases, as with Bidens cernua and different species of Carex, -germination of the floating fruit took place in the water after a period -of two years. The same is also true of the seeds of Iris pseudacorus and -of the drupes of Sparganium ramosum. The last-named remained afloat in -the vessels, with the seed still sound, after four years; and the fruits -of Carex paludosa germinated afloat after three years in water. Many -drift fruits and seeds did not germinate freely in the vessels until the -second spring, that is, after a lapse of eighteen months; and in those -cases where the experiments were still further prolonged, a few -germinated in the vessels in the third and sometimes even in the fourth -year. - - - NOTE 15 (pages 33, 280) - - THE OCCURRENCE INLAND OF SILENE MARITIMA - -Prof. Schimper appeared to be in doubt as to the inclusion of this -littoral plant amongst those found in elevated mountain districts. -However, an interesting note on the occurrence of this plant on the -summit of one of the inland Norwegian mountains is given by Sernander -(p. 405), and is referred to by me on page 280 of this work. - - - NOTE 16 (page 34) - - THE BUOYANCY OF THE SEEDS OR FRUITS OF THE BRITISH BEACH-PLANTS THAT - ALSO OCCUR INLAND - -My experiments in the case of Armeria vulgaris, Artemisia, Cochlearia -officinalis, Plantago, the maritime forms of Spergularia rubra with and -without winged seeds, and Silene maritima disclose little or no floating -capacity even after prolonged drying. Thuret obtained similar results -for the Spergularia. It is unlikely that other plants of the group -possess any floating power worth speaking of. As indicated in Note 71, -the fruits of Raphanus maritimus float only for 7 to 10 days. - -Nature disperses the fruits of Armeria vulgaris inclosed in the -persistent calyx; but in this condition they float only for 2 to 4 days -in sea-water, and the buoyancy of the capsule and seed is still more -limited. They are sufficiently light to be blown some distance by strong -winds, and the stiff hairs would cause them to adhere to a bird’s -plumage in the case of gulls nesting where the plants grow. - -Reference to Matricaria inodora is made under Note 18. - - - NOTE 17 (page 35) - - THE BUOYANCY OF THE SEEDS OR FRUITS OF THE GROUP OF BRITISH LITTORAL - PLANTS THAT FREQUENT SALT MARSHES AND MUDDY SHORES - -_Aster tripolium._ The achenes, with or without the pappus, sink in -fresh and salt water in a day or two even after a year’s drying. - - _Glaux maritima_ } - _Plantago maritima_ } The small seeds, or the seed-like nucules as - _Samolus valerandi_ } in Suæda, have but little floating power - _Suæda fruticosa_ } even after prolonged drying. - _Suæda maritima_ } - -_Salicornia herbacea._ Would be dispersed probably by floating portions -of the plant, which, however, soon break down and the liberated seeds -sink. The floating seedling thrives in sea-water and could be carried -great distances (see Note 19). - -_Salsola kali._ I experimented on this plant, both on the coast of -Devonshire and in Chile, with the same results in both localities -whether in the fresh state or after drying for weeks. The fruit sinks, -but when the plant dries the fruit is often detached inclosed in the -perianth, and floats in that condition in sea-water for a few days. -Portions of the plant of various sizes bearing mature fruits all sank -within ten days. It would seem at first sight, from the observations of -Prof. Martins, that the fruits float for several weeks; but his -experiments were mainly directed to testing the powers of germination -after sea-water _immersion_; and it is often not at all clear whether -_flotation_ is implied or even to be correctly inferred. There is a -slight suspicion of germination on the plant. Sea-birds doubtless aid in -the dispersion of this plant; the dry crisp portions of the plant -carrying fruits catch readily in one’s clothes on account of the -prickly-pointed leaves. - -_Scirpus maritimus._ The fresh fruits float a few weeks in sea-water in -most cases, but 10 per cent. remain afloat after two months. After -drying for some months 30 per cent. remain floating after two months’ -immersion. - - } The fruits float a few days or a week. Drying - _Triglochin maritimum_ } somewhat increases the buoyancy. Sir W. - _Triglochin palustre_ } Buller in New Zealand found in the gullet - } of Anas superciliosa, the Grey Duck, numbers - } of the fruits of Triglochin triandrum. - - - NOTE 18 (page 35) - -THE BUOYANCY OF THE SEEDS OR FRUITS OF THE BRITISH LITTORAL PLANTS THAT - ARE CONFINED TO THE BEACH - -_Arenaria (Honckeneya) peploides._ The seeds float for many months in -sea-water unharmed, 75 per cent. floating after a year. They never -germinate in sea-water; but on being transferred to fresh water after -many months in sea-water they germinate healthily in a few days. These -seeds only float a few days in fresh water, all sinking within 10 days, -and even after a year’s drying they sink in a week or two. Precisely the -same results were produced in my experiments in 1892 on Cornish seeds, -and in 1904 on Devonshire seeds. In the great contrast between their -floating capacity in sea-water and in fresh water the seeds of this -plant defy the general rule that seeds that float a long time in -sea-water float also a long time in fresh-water. According also to -Sernander the seeds float a long time in the sea. He says that the -capsules float, but since they ultimately dehisce this could scarcely be -efficacious in dispersal. Floating portions of the plant also aid in its -dispersal, according to the same authority (p. 174). The plant forms -great extended masses on the pebbly shores of Spitzbergen (Ekstam, p. -28). - -_Beta maritima._ Thuret found that the dried fruits of this plant -floated only two or three days in sea-water; whilst in my sea-water -experiments the freshly gathered fruits floated only one or two days. -Sernander speaks of them as fitted for dispersal from shore to shore; -but this could only be to a limited extent. Martins and Thuret -established by experiment the capacity of the germination of seeds of -other species of Beta after long immersion in sea-water; and the first -seems to imply that those of Beta vulgaris float for many weeks; but I -am inclined to think an error lies here. - -_Cakile maritima._ The fruits, even after long drying, float, as a rule, -only a week and sink within ten days, the same results being afforded in -my sea-water experiments in 1893 on fruits from Cornwall, and in 1904 on -fruits from Devonshire. The fruits are common in the stranded drift on -the north coast of Devonshire and may often be seen germinating there. -They are also frequent in the beach drift of the Scandinavian coasts -(Sernander, p. 156). - -_Crambe maritima._ The fruits were kept floating by Sernander more than -13 days (p. 165). Martins implies that they floated for 45 days. Darwin -says that they germinated after 37 days’ immersion in sea-water, but -does not specify that they floated all the time. - -_Crithmum maritimum._ The ripe fruits readily separate into the two -carpels, which are very buoyant and float in sea-water for months. In my -experiments, 95 per cent. remained afloat after 10 months. It is -remarkable that whilst in sea-water the spongy covering of the carpels -retains its vitality, in fresh-water it becomes sickly and decays and -the carpels lose their floating power, so that they float weeks instead -of months as in the sea-water. The carpels are extremely light, being -washed up in the spray and blown up by the wind amongst the lightest of -the stranded drift of the Devonshire beaches. In a moderate gale they -are often blown off the beach and up the cliff-faces. - -_Convolvulus soldanella._ From 40 to 50 per cent. of the seeds float -after six months in sea-water, and about 30 per cent. float after -eighteen months, retaining up to the end their germinating capacity. -Sernander implies that the plant is not found on the Scandinavian coast -to the north of Nissum Fjord in Denmark. It is known, however, to occur -in the south of Scotland. (I am indebted to Mr. Millett for his -extremely kind assistance in experimenting on this plant about ten years -since.) - -_Eryngium maritimum._ The fruits float in sea-water, as a rule, only 3 -or 4 days and all sink within a week. After drying for three months, the -floating period is only increased by a day or two. Though not at all -suited for transport for any distance by the currents, the carpels, on -account of their long prickly calyx teeth, would readily become -entangled in a bird’s plumage, and doubtless they are dispersed usually -in that fashion. - -_Euphorbia paralias._ The seeds float a long time unharmed in the sea. -In my experiments at least 90 per cent. remained afloat after six weeks -in sea-water. On account of their small size they are liable to be -overlooked in beach drift; but they are to be found stranded on the -sands of our southern coasts, and they came under my notice in abundance -in the seed-drift of the Sicilian beaches. - -_Glaucium luteum._—The seeds have no proper buoyancy even after -prolonged drying. On account of their oiliness they will float at first -on still water; but they can be made to sink at once or in a day by -dropping water upon them. The mode of dispersal is problematical. - -_Lathyrus maritimus._—The seeds are evidently able to float a long time. -They were, according to Sernander (p. 178), found in quantities by J. -Schmidt cast up on some sand-islets near Falster in Denmark; and the -plant is regarded by Norman as distributed over the coasts of Arctic -Norway through the agency of the currents. They have, as observed by -Schmidt, considerable floating powers. Some small leguminous seeds, -seemingly of this species, which I found in the beach drift of -Woollacombe Sands, Devonshire, floated uninjured for many weeks in -sea-water. - -_Matricaria maritima,_ maritime variety of M. inodora. The fruits -floated in my experiments unharmed after eight months in sea-water. In -an experiment made some years since on the fruits of the inland form I -noted that they had little or no buoyancy; but it is necessary to repeat -the observation. Sernander (p. 181) supports Norman’s view that these -plants are spread by the currents in Arctic Norway. The fruits occur in -the Baltic sea-drift and also in fresh-water drift. M. inodora is found -on sandy beaches in Nova Zembla. I am inclined to regard the maritime -form from the dispersal standpoint as a distinct species. - -_Polygonum maritimum._—I have made observations on this plant in -Devonshire, the Lipari Islands, and the coast of Chile. As in the case -of several other species of Polygonum tested by me the fruits have -little or no buoyancy, but inclosed in the perianth they float three or -four days. The entire plant floats; but portions placed in sea-water -sank within five or six days. Shore-birds can alone explain the wide -distribution of this species. - -The structural characters of some of these fruits or seeds are in their -relation to buoyancy discussed on page 115. It may be here observed that -the valuable results obtained by Prof. Martins in testing the -germinating capacity of the fruits and seeds of several of the -shore-plants above mentioned, after long immersion in sea-water, are at -times not to be depended on for the flotation indications, the -persistence of the seed’s vitality being the special purpose of his -research. His negative results as regards germination are not, however, -always conclusive, since the period employed from April to June was -quite insufficient. In many of my experiments seeds after long flotation -in sea-water did not germinate for a year or more afterwards. If his -investigation had been extended, the opinion that the Ranunculaceæ, the -Malvaceæ, and the Convolvulaceæ are apparently least able to resist the -action of sea-water would never have been formed. A very large amount of -evidence now shows that most seeds or fruits that are at all well -protected will germinate after long immersion in sea-water. But all -experiments must be well safeguarded and extended over a year or two. -The necessity of this was long since shown by Thuret. By employing -double sets of seeds he ascertained that in a third of the species -germination failed not only in the case of the seeds immersed in -sea-water, but also in those that had not been placed in sea-water at -all. Future investigators may, however, regard the buoyant qualities of -seeds or fruits with their associated structural characters as offering -now the true line of research. Observers beginning with Berkeley and -Darwin down to the present time have quite established the fact that -seeds as a rule germinate freely after long sea-water immersion. - - - NOTE 19 (page 35) - - ON GERMINATION IN SEA-WATER - -During my experiments on the buoyancy of about 270 British plants, about -a fourth of them (including most of those with buoyant seeds or fruits) -were subjected to prolonged immersion in sea-water from periods varying -from six to thirty-three months. If we except plants like Aster -tripolium, Salicornia herbacea, Triglochin maritimum, &c., that live -normally in salt marshes, or on the muddy banks of estuaries, only one -of the whole number, namely, Ranunculus sceleratus, displayed the -capacity of germination in sea-water. Amongst the plants that failed may -be mentioned the following that are confined to the sea-beach—Arenaria -peploides, Cakile maritima, Convolvulus soldanella, Eryngium maritimum, -Euphorbia paralias, Glaucium luteum, and we may here include Crithmum -maritimum of the rocky coasts. Of the beach-plants that also grow -inland, Silene maritima and Spergularia rubra (excepting the form found -on muddy coast flats) likewise failed. Amongst the plants of -miscellaneous inland stations that failed were Atriplex patula, Bidens -cernua, B. tripartita, Calla palustris, several species of Carex both -from dry and wet situations, Convolvulus arvensis, C. sepium, -Hydrocotyle vulgaris, Iris pseudacorus, several species of Juncus, -Lycopus europæus, Mentha aquatica, Ranunculus repens, Rhinanthus crista -galli, several species of Rumex, Scutellaria galericulata, Sparganium -ramosum, &c. - -In nearly all the plants that failed to germinate in sea-water the -capacity of readily germinating in fresh water was displayed. The -restraining power of immersion in sea-water was illustrated over and -over again in my experiments. During the course of an experiment seeds -removed from the sea-water vessel and placed directly in a vessel of -fresh water kept beside the other germinated in a few days, whilst those -left in the sea-water never germinated, though often kept there for -months after. It was also noticeable that a previous sea-water immersion -favoured early germination in fresh water. It may be added that most of -the experiments were on floating seeds and seedvessels, though -germination also occurred in the sunken state. - -It was ascertained in the exceptional case of Ranunculus sceleratus, -that although germination took place in sea-water, it was only after a -prolonged soaking of months had prepared the way. Of a number of its -seed-like fruits placed in fresh water and in sea-water in April and -kept under the same conditions, those in fresh water germinated freely -in a week or two, whilst those in sea-water did not begin to germinate -until the following October. Whilst the floating seedlings produced by -germination in fresh water grew vigorously and developed roots, those -resulting from germination in sea-water and left in the vessel only -attained a length of four millimetres in two months, developed no roots, -and showed only the first leaf. The sea-water seedlings were pale green, -and in their stout fleshy appearance contrasted greatly with the slender -fresh-water seedlings. - -With regard to the germination in sea-water of the plants of the salt -marsh and of the mud-flats of estuaries, the following observations may -be made. With Aster tripolium the seeds germinate readily in sea-water -even when its density is raised by evaporation to 1·040; and I think -that by a carefully graduated series of experiments they could be -induced to germinate in brine. The seeds of Salicornia herbacea -germinate in sea-water more readily than in fresh water; and the -sea-water seedling is much the more vigorous and healthy of the two. I -kept the floating seedlings in sea-water for about ten weeks from the -date of germination, when they had developed the second joint and were -throwing out rootlets. After that, unless placed in salt-mud, they -became sickly and died. The floating seedling can evidently disperse the -species. I found with Spergularia marina, the maritime form of S. rubra, -that seeds of the plants growing on a sandy beach did not germinate in -sea-water, only those from plants growing on muddy coast-flats doing so. -But the sea-water seedlings, unlike those of Salicornia herbacea, but -like those of Ranunculus sceleratus, when left in sea-water did not -thrive. The seeds of Triglochin maritimum, as well as those of T. -palustre, behave very similarly in sea-water, germinating readily, the -liberated seedlings thriving afloat and producing the plumule. The -ultimate test of the capacity for germinating in sea-water seems to lie -in the behaviour of the seedling when left in the sea-water. Unless it -belongs to a characteristic plant of the salt marsh or of the estuary, -like Salicornia, it makes but little attempt at growth whilst afloat in -sea-water, showing no rootlets, though at times developing the plumule. - -The germination of seeds in sea-water also attracted the notice of -Darwin; but his results in some respects are scarcely those I should -have looked for (_Gardener’s Chronicle_, May, 1855, and _Journ. Linn. -Soc._, vol. i., p. 130, 1857). Out of the seeds of 87 plants placed in -sea-water to test their capacity of germination when afterwards planted, -in three cases, those of Tussilago farfara, Convolvulus tricolor, and -the garden Orache (Atriplex), the seeds germinated under the water, the -freed seedlings, as with the two first named plants, living in the -sea-water for some time after. Darwin was evidently himself surprised at -these results, and I am quite unable to understand them. In England and -in the tropics I have carried on prolonged sea-water experiments on the -seeds of at least fifteen species of Convolvulus and Ipomœa (including -the beach plants C. soldanella and I. pes capræ) and have never obtained -such a result. The seeds will nearly always germinate well in fresh -water; but in sea-water the process begins, as indicated by the swollen -seed, and then aborts, the embryo dying (see page 83). The seeds of -Atriplex patula, though a long time in sea-water in my experiments, made -no attempt to germinate there. Neither Prof. Martins, who experimented -upon the effects of sea-water immersion on the seeds of nearly 100 -plants, including many coast species, nor M. Thuret, who experimented in -sea-water on the seeds of 251 plants, the experiments being in some -cases prolonged for more than a year, make any reference, as far as I -could gather from their writings, to any cases of germination in -sea-water. Darwin’s results, however, are always significant in matters -of dispersal; and perhaps one of my readers will be able to experiment -again on his three plants. - -When in Hawaii, I made some observations on the germination of Batis -maritima in sea-water, a plan with which I was also familiar in its home -in the salt-water pools of the coast of Peru. The mature fruits, on -being freed from the parent plant in sea-water, float away, and in from -one to two weeks they break down from decay, setting free the seeds. The -seeds float in sea-water indefinitely, their buoyancy only terminating -with their germination, the first seeds germinating afloat about six -weeks after the breaking down of the fruit, whilst the rest continue to -float in the sea-water during the next three months, some of them -germinating at intervals, and all of them doing so eventually. Strange -to say, although the seedlings remained healthy whilst afloat in the -sea-water, they made no effort either to separate the cotyledons or to -produce a plumule. - - - NOTE 20 (page 42). - - ON THE MAXIMUM HEIGHTS REACHED BY SOME SHORE PLANTS IN THEIR EXTENSION - INLAND IN VANUA LEVU, FIJI - -Since they occupy the “talasinga” districts described in the following -note, these shore plants would be expected to extend as high as those -districts extend, namely, to about 1,500 feet above the sea. This indeed -represents their limit excepting in one instance; but many fall -considerably short of this elevation. - - _Canavalia obtusifolia_, variety, 700 feet, rare. - _Cassytha filiformis_, 950 feet. - _Cerbera Odollam_, 1,200 feet: 2,600 feet in one exceptional case on - the slopes of Mbatini. - _Colubrina asiatica_, 400 feet. - _Cycas circinalis_, 1,100 feet. - _Derris uliginosa_, 1,000 feet, rare. - _Ipomœa pes capræ_, 1,300 feet. - _Morinda citrifolia_, 700 feet. - _Scævola Kœnigii_, not common inland, and rarely over 100 feet above - the sea; but it may occur miles from the beach, as near Vatu Levoni, - where a few stunted plants were growing five miles from the coast. - _Vitex trifolia_, 1,300 feet, usually more or less unifoliolate and - procumbent. - -Unless otherwise stated all the plants above named are common inland, as -also are _Premna tahitensis_, _Tacca pinnatifida_, _Tephrosia -piscatoria_, _Hibiscus tiliaceus_, &c.; but I have made no note of -_Thespesia populnea_ occurring far off the beach. - - - NOTE 21 (pages 42, 43) - - ON THE DWARFING OF SHORE PLANTS WHEN EXTENDING INLAND INTO THE - “TALASINGA” PLAINS IN VANUA LEVU. - -_Premna tahitensis_, 9 or 10 feet high at the coast, may here be only 3 -feet high. Other trees like _Morinda citrifolia_ become also stunted. -_Cerbera Odollam_, a moderate-sized tree at the coast, may in the -“talasinga” plains be only 4 to 6 feet high, but it here displays -distinct varietal characters. Whilst the shore trees of Cerbera Odollam -have broad leaves (length 3 times the breadth) with obtuse points, and -short, stout flower-peduncles (1-1/2 - 2 inches), the inland or -“talasinga” species has long lanceolate leaves (length 7 or 8 times the -breadth), and long, slender flower peduncles (3 inches). However, -intermediate forms are common, the broad-leaved coast tree approaching -the inland plant and _vice versâ_. - - - NOTE 22 (page 43) - - THE “TALASINGA” PLAINS OF VANUA LEVU, FIJI - -Amongst the most conspicuous features of the north and north-west or lee -sides of the large islands of Vanua Levu and Viti Levu are the extensive -rolling plains that extend from the sea-border for some miles inland to -the foot of the mountains. It is to those of the first-named island that -the following remarks strictly apply; but no doubt they will serve -equally well for those of the other island. In the first volume on the -geology of Vanua Levu, reference is frequently made to this subject, and -the reader may profitably look at the remarks there made. - -Here the mountain-forests more or less abruptly cease, and we have an -undulating region of grass, reeds, and ferns dotted over with -Casuarinas, Pandanus trees, Cycads, Acacias, and shrubby growths. Though -the list of plants characteristic of these plains is not small, they are -not, as a rule, numerous in any one locality, and the general appearance -is one of aridity. A dry, crumbling soil, often deeply stained by -iron-oxide, is plentifully exposed; and blocks of basic volcanic rocks -in all stages of disintegration are strewn over the surface in many -localities. Rivers, fed by the heavy rainfall of the forested slopes of -the mountains, traverse these regions, but, as a rule, receive no -tributaries; and the districts have, in fact, well earned the name given -to them by the natives of the “talasinga,” or sun-burnt, lands. - -The vegetation, though sparse and scanty in comparison with that of the -forests, is sufficiently varied when it comes to be more closely -examined. In one locality we may have extensive tracts covered with -Gleichenia, Pteris, and other ferns of the bracken habit. In another, -tall reeds (Eulalia) and grasses cover large areas. Here, more than one -species of Tacca (T. pinnatifida and T. maculata) thrive. There, the -Turmeric (Curcuma longa) abounds. Trailing over the soil in one place we -notice Ipomœa pes capræ, in another the Yaka (Pachyrrhizus trilobus), -and in another the procumbent unifoliolate form of Vitex trifolia. -Amongst the shrubs and small trees we observe in different localities -the Sama (Commersonia echinata), the Mbulei (Alstonia plumosa—one of the -rubber plants), Mussænda frondosa, Melastoma denticulatum, and Nelitris -vitiensis, the Nunga-nunga. Dodonæa viscosa, found in similar regions in -Australia and New Zealand, abounds in places; and here and there may be -seen species of Hibbertia, another Australian genus. Fagræa Berteriana, -the Mbua tree, grows abundantly in certain districts, as in the Mbua -plains, and Gardenias are at times abundant. One or two characteristic -beach-plants have been already mentioned, and amongst others -particularly frequent in these plains are Cassytha filiformis, Cerbera -Odollam, Morinda citrifolia, and Premna tahitensis. - -When these talasinga districts approach the forests, patches of wood -occur at intervals, and we observe here the Candle-nut Tree (Aleurites -moluccana), the Vunga (Metrosideros polymorpha), and the Thau-kuro -(Casuarina nodiflora). Such are some of the botanical features of these -districts; but the reader will acquire a sufficiently correct general -notion of the floral physiognomy of these regions if he bears in mind -their most conspicuous characters, those of an undulating region more or -less covered with ferns, tall reeds, and grass, and dotted over, either -separately or in clumps, with Casuarinas (C. equisetifolia), Screw-pines -(Pandanus odoratissimus), Cycads (C. circinalis), and Acacias (A. -Richii, &c.). - -However, the peculiar vegetation of these plains often ascends the lower -slopes of the mountains, reaching to various elevations. In Vanua Levu -it often ceases at 900 or 1,000 feet, but it may only reach to 400 or -500 feet, and, on the other hand, not uncommonly it ascends to as much -as 1,500 feet, the greatest elevation recorded by me being 1,600-1,700 -feet in the Sealevu district. It extends miles inland, and where -conditions are suitable it may reach the heart of the island. - -Different explanations have been offered of the origin of the peculiar -vegetation of the leeward slopes of these islands. It is, however, a -phenomenon that is presented over much of the globe by islands lying in -the track of regular winds, the weather, or wet, side being densely -wooded, whilst the lee, or dry, side is covered with grass, ferns, and -similar vegetation. The predisposing cause must be climatic; and -although Mr. Horne’s explanation attributing it to the effect of fires -and to a faulty system of native cultivation (pp. 80, 132) may be -doubtless true in certain localities, the influences at work here must -be the same as are at work in other islands and on continental coasts in -other parts of the world. - -But for all that it is not easy to give a definite explanation even from -a meteorological standpoint. Those who are interested in this subject -will recall the desert districts of Australia and the dreary sandy -wastes of the coast of Northern Chile and Peru; and they will be -cautious in venturing on a definite explanation even with such -relatively unimportant examples of the same principle as are exhibited -by the islands of Fiji. Dr. Seemann, writing of these “talasinga” plains -(p. xii), remarks that “their very aspect is proof that rain falls in -only limited quantity,” the mountainous backbone of the islands -intercepting, as he holds, much of the rainfall. But the subsequent -observations of Mr. Holmes, at Delanasau, in the “talasinga” district on -the north-west side of Vanua Levu, have shown that there is by no means -a small rainfall in this locality, the average rainfall, for instance, -for the seven years ending December, 1877, being 113 inches, which must -be quite two-thirds or three-fourths of the fall on the weather side of -the island (see p. 215); whilst the average number of days on which rain -fell was 156. The true cause would seem to lie in the excessive dryness -of the air on the lee side of the islands between the rains, and the -whole matter may, perhaps, be one rather for the hygrometer than for the -rain-gauge. I have no comparative data bearing on this point; but Mr. -Holmes, whose observations as here quoted are from Horne’s _Year in -Fiji_, found that the mean relative humidity for 1875 at 1 P.M. was 63, -which is certainly very low for the tropics. I may remark that, as far -as personal experience goes, the climate on the lee side of Vanua Levu -is much more enervating, much less healthy, and the air is far more -“drying” than on the side exposed to the trade-wind. - -Geological characters, as I found, explained nothing in this connection, -the “talasinga” vegetation sometimes occurring on basaltic areas, at -other times on the “soapstone” or calcareous mud-stone, and again on -coarser tufaceous rocks. In my volume on the geology of Vanua Levu (p. -57), it is pointed out that the extensive disintegration of the basaltic -rocks, that are exposed on these plains in places, affords evidence of -the great antiquity of these “talasinga” districts in their present -unforested condition. The extent to which these rocks have weathered -downward is remarkable. In some places they are decomposed to a depth of -ten feet and more. The same inference is to be drawn from the occurrence -of fragments of limonite, or bog-iron ore, over these plains, marking as -they do original swampy tracts that, with a few exceptions, have long -since disappeared. Such deposits indicate that these plains have been -for ages in the same condition. ... It may be added that, according to -Mr. Lister and Mr. Crosby, the features of the “talasinga” plains occur -in the Tongan Group on the leeward sides of the islands of Eua and -Vavau. - - - NOTE 23 (page 43) - - SCHIMPER’S GROUPING OF THE INDO-MALAYAN STRAND-FLORA - -It is divided into four formations—the Mangrove, the Nipa, the -Barringtonia, and the Pes-capræ. The two last make up my -Beach-formation, the Barringtonia formation comprising the trees, -shrubs, &c., immediately lining the beach, and the Pes capræ including -the creepers and bushes of the beach itself. In the Pacific islands it -is not always easy to preserve this distinction. The Nipa formation -corresponds in some respects with my Intermediate or Transition -formation, lying as it does between the mangrove-belts and the woods of -the interior; but the swamp-palm (Nipa fruticans) that forms it in the -mass is not found in Fiji or, indeed, in the Pacific islands, excepting -the Solomon and Caroline Groups. - - - NOTE 24 (page 44) - - GROUPING OF SOME OF THE CHARACTERISTIC PLANTS OF THE STRAND-FLORA OF - FIJI - -(a) _Beach-formation._—Calophyllum inophyllum, Thespesia populnea, -Triumfetta procumbens, Carapa moluccensis, Canavalia obtusifolia, Vigna -lutea, Pongamia glabra, Sophora tomentosa, Cæsalpinia Bonducella, Acacia -laurifolia, Barringtonia speciosa, Terminalia Katappa, Gyrocarpus -Jacquini, Pemphis acidula, Morinda citrifolia, Guettarda speciosa, -Wedelia biflora, Scævola Kœnigii, Cordia subcordata, Tournefortia -argentea, Ipomœa pes capræ, Cassytha filiformis, Hernandia peltata, -Pandanus odoratissimus, &c. - -(b) _Mangrove-formation._—Carapa obovata, Rhizophora mucronata, -Rhizophora mangle, Bruguiera Rheedii, Lumnitzera coccinea, Scirpodendron -costatum, &c. (See below.) - -(c) _Intermediate or Transition-formation._—Hibiscus tiliaceus, -Heritiera littoralis, Smythea pacifica, Derris uliginosa, Entada -scandens, Barringtonia racemosa, Cerbera Odollam, Clerodendron inerme, -Vitex trifolia, Excæcaria Agallocha, &c. - -N.B.—It is not possible to draw a definite line between the plants of -the mangrove swamp and those of the tracts around. Several of the plants -placed in the intermediate formation, such as Heritiera littoralis, -Entada scandens, Excæcaria Agallocha, &c., are just as much at home -amongst the mangroves. In the same way it is often difficult to -distinguish between the Beach and the Intermediate formations, and -plants like Cerbera Odollam, Hibiscus tiliaceus, and Vitex trifolia -belong equally to both. - - - NOTE 25 (page 47) - - THE STRAND-FLORA OF THE TAHITIAN REGION - -Drake del Castillo’s _Flore de la Polynésie française_ deals mainly with -the Society or Tahitian Islands, but also with the Marquesas, Paumotus, -Gambier Islands, and Wallis Island. The last-named, however, lies in -Western Polynesia, and is not dealt with in this connection. There is no -reason to believe, judging from the general character of the islands and -from Cheeseman’s memoir on the Rarotongan flora, that the strand-plants -of the islands of the Cook and Austral Groups, which also belong to this -region, differ materially from those of the Tahitian islands proper. -Rarotonga, however, possesses Entada scandens, not recorded as a growing -plant from any other part of East Polynesia, excepting perhaps Mangaia -in the same group. - - - NOTE 26 (page 48) - - THE FIJIAN SHORE-PLANTS NOT FOUND IN TAHITI - -Although most of these plants, such as Barringtonia racemosa, -Clerodendron inerme, Entada scandens, Excæcaria Agallocha, Heritiera -littoralis, Smythea pacifica, &c., have fruits that float for months, -and could have reached Tahiti as readily as some of the beach-plants -that have successfully established themselves, there are a few like -Dalbergia monosperma, Derris uliginosa, and Scirpodendron costatum, the -fruits of which only float for weeks, and it is possible that they may -have been unable to reach there. - - - NOTE 27 (page 49) - - THE INTRUDERS INTO THE BEACH-FLORA FROM THE INLAND PLANTS OF TAHITI - -Drake del Castillo mentions several, such as species of Boerhaavia, that -could only be occasional intruders; but it is noteworthy that Gardenia -tahitensis appears to be a genuine recruit from inland. The xerophilous -habit of the Pacific Gardenias and their station, usually near the -coast, however, would render this possible. - - - NOTE 28 (page 52) - - THE LITTORAL PLANTS OF THE HAWAIIAN ISLANDS - - Column headings: - - A: Indigenous. - B: By aborigines in ancient times. - C: By Europeans soon after discovery. - D: Old World. - E: New Worlds. - F: Both Worlds. - G: Large. - H: Small. - I: Float for months. - J: Sink at once or in a few days. - K: Attract frugivorous birds. - - +---------------------------+---+---+---+---+---+---+---------------------+ - | | Origin. | Distri- | Characters of fruit | - | | | bution. | or seed. | - | +---+---+---+---+---+---+-------+---------+---+ - | Species.[5] | |Intro- | | | | Size. |Buoyancy.| | - | | |duced. | | | | [6] | | | - | | +---+---+ | | +---+---+-----+---+---+ - | | A | B | C | D | E | F | G | H | I | J | K | - +---------------------------+---+---+---+---+---+---+---+---+-----+---+---+ - | Acacia Farnesiana |...|...| + |...|...| + |...| + |weeks|...| ? | - | Cæsalpinia Bonducella | + |...|...|...|...| + |...| + | ... | + |...| - | Calophyllum inophyllum |...| + |...| + |...|...| + |...| + |...|...| - | Cassytha filiformis | + |...|...|...|...| + |...| + | + |...| + | - | Colubrina asiatica | + |...|...| + |...|...|...| + | + |...|...| - | Cordia subcordata |...| + |...| + |...|...|...| + | + |...|...| - |E Cuscuta sandwichiana | + |...|...|...|...|...|...| + | ... | + |...| - | Cocos nucifera |...| + |...|...|...| + | + |...| + |...|...| - |P Gossypium tomentosum | + |...|...|...|...|...|...| + | ... | + |...| - |P Heliotropium anomalum | + |...|...|...|...|...|...| + | ... | + |...| - | Heliotropium curassavicum| + |...|...|...|...| + |...| + | ... | + |...| - | Herpestis Monnieria | + |...|...|...|...| + |...| + | ... | + |...| - | Hibiscus tiliaceus |...| + |...|...|...| + |...| + | + |...|...| - | Ipomœa glaberrima | + |...|...| + |...|...|...| + | + |...|...| - | Ipomœa pes capræ | + |...|...|...|...| + |...| + | + |...|...| - |E Jacquemontia sandwicensis| + |...|...|...|...|...|...| + | ... | + |...| - |E Lipochæta integrifolia | + |...|...|...|...|...|...| + | ... | + |...| - | Morinda citrifolia |...| + |...| + |...|...|...| + | + |...| ? | - | Mucuna gigantea | + |...|...| + |...|...| + |...| + |...|...| - | Pandanus odoratissimus |...| + |...| + |...|...| + |...| + |...|...| - | Portulaca oleracea |...| + |...|...|...|...|...| + | ... | + |...| - | Scævola Kœnigii | + |...|...| + |...| + |...| + | + |...| + | - | Sesuvium Portulacastrum | + |...|...|...|...|...|...| + | ... | + |...| - | Tacca pinnatifida |...| + |...| + |...| + |...| + | + |...|...| - | Tephrosia piscatoria | + |...|...| + |...|...|...| + | ... | + |...| - | Terminalia Katappa |...|...| + | + |...|...| + |...| + |...|...| - | Thespesia populnea |...| + |...| + |...|...|...| + | + |...|...| - | Tribulus cistoides | + |...|...|...| + |...|...| + | ... | + |...| - | Vigna lutea | + |...|...|...|...| + |...| + | + |...|...| - | Vitex trifolia | + |...|...| + |...|...|...| + | + |...| + | - +---------------------------+---+---+---+---+---+---+---+---+-----+---+---+ -Footnote 5: - - There are three endemic species here included which are preceded by E. - Two species preceded by P are confined to Polynesia. Most of the - plants are at present typically littoral, though often also occurring - inland. - -Footnote 6: - - All fruits or seeds, an inch or over in size, that could not have been - transported to Hawaii by birds are regarded as large. - - - NOTE 29 (page 54) - - BOTANICAL NOTES ON THE COAST-PLANTS OF THE HAWAIIAN ISLANDS - -[The following remarks have been extracted from my journals and -represent some of the field-notes of journeys made in the more -interesting localities.] - -(1) _Walk along the Puna Coast, Hawaii, from Punaluu to Hilo_ (Dec. 26, -1896, to Jan. 6, 1897).—For the first two to three miles to Kamehame -Point, the following plants were noticed on the flows of smooth ropy -lava that formed the cliff-bound coast—Capparis sandwichiana, -Jacquemontia sandwicensis, Ipomœa insularis, Lipochæta lavarum, -Portulaca villosa, Tephrosia piscatoria, Tribulus cistoides, Waltheria -americana, &c. Beyond this point Scævola Kœnigii was abundant in places -on the old lava-flows near the sea, and further on patches of Myoporum -sandwicense growing, not as a tree 20 to 30 feet high, as in the -mountains, but as a prostrate shrub with fleshy leaves. Vegetation -similar to that above described occurred on the surface of the old -lava-flows that constituted the cliff-bound sea-border as far as -Kapapala Bay. On the sandy beach at Kapapala Bay grew Ipomœa pes capræ, -serving as host to Cuscuta sandwichiana. In the vicinity of the house at -Keauhou there were a few Coco palms and Pandanus trees, whilst Capparis -sandwichiana and Morinda citrifolia were growing on the adjacent -lava-fields. - -Morinda citrifolia and Tephrosia piscatoria grew on the lava flows -between Keauhou and Apua. On the beach at Apua, Ipomœa pes capræ and -Scævola Kœnigii were abundant, the last extending a few hundred yards -inland on the lava. Further east the inland bush, made up of Cyathodes -tameiameiæ, Metrosideros polymorpha, &c., descended to the coast to -within a few hundred yards of the sea. In crossing the lava coast plains -to Kapa-ahu I observed Morinda citrifolia growing frequently out of the -cracks in the bare lava-rock, and an occasional solitary tree of -Erythrina monosperma growing also from the fissures. - -Before reaching Kapa-ahu we passed the site of an old coast village, -named Laepuki, where there were growing from forty to fifty Coco-nut -palms, as well as another village, represented by a solitary house, and -named Kamomoa, where there were 27 Coco-nut palms and a few Pandanus -trees. Kapa-ahu, with its numerous Coco-nut palms, was more like a South -Sea coast village than any before seen; and the coast vegetation -suddenly acquired a South Pacific character. - -At Pulama, for instance, about a mile west of Kapa-ahu, where the -ancient lava-flows, fairly vegetated, terminate at the sea in cliffs 20 -or 25 feet high, there is a curious and quite unexpected development of -a littoral flora such as we should see in the South Pacific. Here, -growing on the broken lava surface at the brink of the cliffs and -overlooking the sea, thrive Cæsalpinia Bonducella, Cocos nucifera, -Ipomœa pes capræ, Ipomœa glaberrima, Morinda citrifolia, Pandanus -odoratissimus, Scævola Kœnigii, Sesuvium Portulacastrum, Thespesia -populnea, and Vigna lutea. This shore-belt of characteristic littoral -plants is backed by vegetation more inland in its character, amongst -which Aleurites moluccana, Dodonæa viscosa, Erythrina monosperma, Ipomœa -insularis, I. bona nox, Osteomeles anthyllidifolia, &c., are to be -observed. Such a shore-belt of typical littoral plants is rarely to be -found in the large island of Hawaii; and its usual position at the -margin of cliffs, and raised 20 or 25 feet above the sea, is rather -suggestive of an uplift in recent times of this part of the coast. - -Between Kapa-ahu and Kalapana is a low country occupied mostly by -Guavas, and often turfy. At Kalapana, which is a large village situated -on a grassy plain by the sea, Coco palms and Pandanus trees abound, and -Mucuna gigantea and Cæsalpinia Bonducella are frequent near the coast, -whilst Ipomœa pes capræ is common on the beach. Calophyllum inophyllum -is planted near the houses. Here Osteomeles anthyllidifolia in its -dwarfed form descends to the edge of the cliffs. About half a mile -beyond Kalapana is the hamlet of Kaimu, and here among the Coco palms -close to the beach I noticed four Loulu palms (Pritchardia -Gaudichaudii). Beyond Kaimu the trees and shrubs of the inland wood, -Metrosideros polymorpha, Cyathodes tameiameiæ, &c., descend on the spurs -of old lava-flows close to the coast; whilst Pandanus and Morinda -citrifolia with Mucuna gigantea are common near the sea as far as -Kehena, where there are plenty of Coco palms. I approached Opihikao -through as fine a Pandanus forest as I have ever seen, the large Bird’s -Nest Fern (Asplenium nidus) growing half-way up their trunks, adding -picturesqueness to the scene, whilst Mucuna gigantea was a common -climber. Beyond Opihikao the inland woods descend to the coast. Thence -on to Makuu the coasts are mostly occupied by Pandanus forests, and the -lower coast road from Makuu to Hilo traverses a region where these -Pandanus trees abound, extending far inland. Scævola Koenigii and Ipomœa -pes capræ are common on the coast near Coco-nut Island, Hilo Bay. - -It may be added that the agency of the wild goat explains the dispersal -of Myoporum sandwicense, Morinda citrifolia, Tephrosia piscatoria, -Waltheria americana, &c., over the almost bare surfaces of the lava -flows on the Puna coast. Goat droppings were frequent under the patches -of Myoporum and Waltheria. In some of them I found the entire seeds of -Portulaca oleracea and the small cocci of Euphorbia pilulifera, weeds -common in the district. - -(2) _Coasts of the Kalae Promontory and its Vicinity, Hawaii._— This is -the most southerly portion of the group, and it is on the eastern coasts -of this district that many of the North American drift logs are embayed -and stranded. At Kamilo, to the east of the promontory, there is a long -beach of calcareous sand where Heliotropium anomalum, Scævola Kœnigii, -and Tribulus cistoides grow in abundance, whilst Sesuvium Portulacastrum -thrives on the beach and in brackish pools. Portulaca lutea (Sol.), -Ipomœa glaberrima (Boj.), and Jacquemontia sandwicensis also occur. -Where the beach-sand has encroached on the adjacent lava surface, the -Scævola covers extensive tracts off the beach, and is stunted. I noticed -a solitary thicket of Thespesia populnea on the beach. - -The actual headland of Kalae is wind-swept and covered with grass, -amongst which Portulaca villosa and Sida fallax thrive. By the sea occur -Scævola Kœnigii and Ipomœa pes capræ, and there is some Sesbania -tomentosa near the point. Waiheiaukini beach is shut in between the -lofty arid slopes of the promontory on one side and a modern lava-flow -on the other side. Here Scævola Kœnigii grows in quantity, together with -Ipomœa pes capræ, Tribulus cistoides, Sida fallax, and Jacquemontia -sandwicensis, whilst Cuscuta sandwichiana is abundant, finding its hosts -in the first four plants just named. - -(3) _South Kona Coast, Hawaii._—The coast here, as exemplified by that -between Kapua and Hoopuloa, is mostly bare lava. Here and there, a -little coral sand collects amongst the lava blocks of the rubbly shore, -and it is in such places that Scævola Kœnigii and Ipomœa pes capræ find -a home and apparently thrive, whilst Hibiscus tiliaceus and Morinda -citrifolia grow behind. I observed Cordia subcordata and one or two -specimens of Pritchardia Gaudichaudii by the coast on the south side of -Milolii. Around a brackish pool at Kapua I observed Heliotropium -curassavicum, and Acacia Farnesiana was to be seen growing on the beach -at Okoe. On the lava coast between Hoopuloa and Papa, two miles to the -north, Tephrosia piscatoria was very abundant. - -(4) _North Kona Coast, Hawaii._—I examined the coast between Kailua and -Kiholo. White beaches are common south of Keahole Point, the coast -further north being usually lava-bound with sandy beaches here and -there. Heliotropium anomalum, Ipomœa pes capræ, and Sesuvium -Portulacastrum are the commonest beach plants on this coast. Scævola -Kœnigii is also abundant in places, whilst Tribulus cistoides and -Morinda citrifolia are also fairly common on the beaches. The Morinda -also grows on the adjacent lava flats; but on both sand and rock it is -evidently usually self-sown, since seedlings are to be seen near the -older plants. Heliotropium curassavicum is to be seen here and there on -the sand all along the coast, but nearly always associated with H. -anomalum. Jacquemontia sandwicensis occurs occasionally on the beach; -and Cuscuta sandwichiana is abundant in places, growing generally on -Ipomœa pes capræ, but sometimes on Scævola Kœnigii. Brackish water ponds -are common on the coast inside the beaches, Ruppia maritima flourishing -in the water, with Sesuvium Portulacastrum growing at the edges. -Sometimes Hala trees (Pandanus odoratissimus) fringe the borders of the -pools. I noticed Pritchardia Gaudichaudii on the coast at Kiholo, and I -learned that Cordia subcordata was once common here as on other parts of -the Kona coast; but it has died out as in most other localities. - -(5) _Kohala Coast, Hawaii._—Several littoral plants are scantily -represented on the beach of black sand at the mouth of the Waimanu -valley, especially Ipomœa pes capræ, Morinda citrifolia, Pandanus -odoratissimus, and Scævola Kœnigii. The Pandanus covers the adjacent -precipitous slopes up to a height of several hundred feet above the sea. -Ipomœa pes capræ is abundant on the sand dunes backing the beach at -Waipio. I observed Naias marina in the Waipio River just inside the -mouth. No one seems to have recorded the plant from the group since -Chamisso found it in Oahu. - -(6) _Hamakua Coast, Hawaii._—Not many opportunities presented themselves -on this cliff-bound coast of finding littoral plants. At the mouth of a -gulch between Ookala and Laupahoehoe I found growing at the coast Vitex -trifolia (var. unifoliolata) in quantity, together with Morinda -citrifolia, Scævola Kœnigii, and Pandanus odoratissimus, the last-named -clothing the hill-slopes overlooking the sea. - -(7) _The Coasts of Oahu._—The littoral vegetation of the south-east -portion of the island from Diamond Head round to Waimanalo is, as a -rule, scanty. Ipomœa pes capræ and Tribulus cistoides prevail to Koko -Head, and on the rubbly coast between that headland and Makapuu Point -occur Tephrosia piscatoria, different species of Lipochæta, &c. Between -Makapuu Point and Waimanalo, Scævola Kœnigii and Vitex trifolia (var. -unifoliolata) are fairly abundant, the former growing on the rocky slope -at the base of the cliffs, and raised perhaps some 20 feet above the -sea. Along the whole east coast of the island the littoral vegetation is -rarely well represented. However, Ipomœa pes capræ is common everywhere, -whilst Scævola Kœnigii occurs frequently, and here and there a few -plants of Morinda citrifolia are seen on the beach, while thickets of -Hibiscus tiliaceus mark in some localities the mouths of streams. - -On the north coast of Oahu, as on the Waialua and Waimea beaches, the -one-leaved variety of Vitex trifolia is common, together with Ipomœa pes -capræ and Euphorbia cordata; whilst Acacia Farnesiana is frequent on the -Waialua beach, its pods being much appreciated by the cattle. -Occasionally, as by the bridge at Waimea, Colubrina asiatica and -Thespesia populnea are to be noticed. - -Shore vegetation is a little better represented on the beaches at and -near Kaena Point, the north-west corner of the island. Here on the sand -we find often in abundance Heliotropium anomalum, the same variety of -Vitex trifolia, Scævola Kœnigii, and Ipomœa pes capræ; whilst on the -rocks bordering the beach occur Gossypium tomentosum, Jacquemontia -sandwicensis, Tribulus cistoides, Vigna lutea, and more than one species -of Lipochæta, the last being derivatives from the inland flora. - -On the west coast of the island true shore-plants play an inconspicuous -part. Ipomœa pes capræ is common on the beaches, and such plants as -Acacia Farnesiana, Jacquemontia sandwicensis, Gossypium tomentosum, and -Tribulus cistoides immediately border the beach. Ipomœa tuberculata is a -frequent intruder as well as the recently introduced Algaroba tree -(Prosopis dulcis). Acacia Farnesiana also extends inland, covering -entire large areas and forming in the Waianae valley extensive thickets -impenetrable for the cattle. It occupies great districts near the coast -in different parts of Oahu, and with Hibiscus tiliaceus is to be found -far inland. The cattle are active dispersers of its seeds. (See Note -30.) - -True beach plants are infrequent at the mouth of Pearl Harbour, although -the coast is well suited for them. Here I found Heliotropium anomalum, -H. curassavicum, Jacquemontia sandwicensis, Lipochæta integrifolia (a -true beach plant), Herpestis Monnieria, &c. Batis maritima occurs in one -or two localities around Oahu, but it is, according to Hillebrand, of -recent introduction. - - - NOTE 30 (page 58) - - THE BEACH-DRIFT OF THE HAWAIIAN ISLANDS - -It was pointed out by Dole long ago in one of the Hawaiian Club Papers -(1868) that the existing currents bring to this archipelago only huge -pine logs from Oregon, but no tropical fruits; and Hillebrand (p. xiv.) -refers to the driftwood of pine logs from the north-west coast of -America, stranded on the shores of these islands. This drift seems to -collect in quantity in particular localities, as on the south-east coast -of Hawaii between Honuapo and the Kalae promontory (especially on the -Kamilo beach near Kaluwalu) and on the east coast of Oahu; and probably -there are other favourable localities for catching the drift on the -northern shores of Maui and Molokai. - -It was on the south-east coast of Hawaii (on the beach at Kamilo and on -the eastern side of the Kalae promontory) that this drift came -particularly under my notice. Here the logs are stranded in abundance, -in sufficient quantity, in fact, to build a town, and they were employed -for building purposes by the manager of the neighbouring sugar-cane -plantation. Several of the logs are of huge size, as much as 4 feet in -diameter; and they are known locally as “white cedar” and “red cedar,” -and characterised as Oregon timber. Some of them are extensively -burrowed by the “teredo” and other boring mollusks. Others recently -stranded are covered with barnacles (Lepadidæ), whilst others that have -lain long on the beach are bare. I have seen these logs occasionally -washed up at Punaluu and at different places on the lava-bound Puna -coast. They apparently first strike the Puna coast, and are drifted -along until they become embayed near the Kalae promontory, and -ultimately stranded. Mingled with them on the beaches Pandanus trunks -occur in number; they evidently hail from those parts of the Puna coast -where Pandanus forests prevail, and thus they indicate the direction of -the drift on the coast of this island. In places there was a -considerable amount of small vegetable _débris_, sometimes partially -concealed by the sand, and containing seeds and fruits in fair quantity. - -The following seeds and fruits were collected:— - - _Pandanus_ drupes, common; most of them fresh-looking, but a few much - worn. - _Thespesia populnea_, a few seeds. - _Ipomœa pes capræ_, seeds, fair numbers. - _Ipomœa bona nox_, seeds, a few. - _Ipomœa glaberrima_, seeds, a few. - _Argyreia tiliæfolia_, fruits and seeds, a few. - _Strongylodon lucidum_, seeds, a few. - _Cæsalpinia Bonducella_, a single seed. - _Vigna lutea_, seeds, a few. - _Calophyllum inophyllum_, a few fruits. - _Ricinus communis_, a few, the seeds either free or in the cocci, and - often empty or decaying. - _Aleurites moluccana_, seeds, common, none sound, either empty or - containing a rotten kernel: also a single fruit. - One or two seeds not identified. - -There was seemingly a total absence of the fruits or seeds of any -littoral plant not found in these islands, such as I was familiar with -in the South Pacific. In the mass this seed-drift could have been -derived from the neighbouring coasts of the island. This is especially -indicated in the cases of the fruits and seeds of Aleurites moluccana, -Ricinus communis, and Argyreia tiliæfolia. The sound seeds of Aleurites -do not float, the buoyant seeds being always empty, or nearly so; and -the presence of the seeds in beach-drift, as explained on page 419, is -due partly to the buoyancy of the empty seed and partly to the decay of -the stranded fruit, the fruits being able to float for a week or two. -So, also, the seeds of Ricinus, whether free or inclosed in the coccus, -do not, when sound, float longer than a week or ten days. The capsules -of the Argyreia can float two or three weeks, whilst the seeds vary in -their behaviour, as observed on page 20. I noticed in places where the -vegetable _débris_ was heaped up and exposed to the sun’s heat, that -some of the Ipomœa seeds were germinating. It is to be remarked that -horse-dung and goat-dung are always common in the beach-drift of these -islands. Seeds are sometimes to be seen in the stranded material; and it -was evident that the droppings of these animals can float for some weeks -before breaking down.... I may add that large sponges, apparently of no -value, are thrown up in quantities on the east side of the Kalae -promontory. - -Excepting the pine logs, the only things coming under my notice in this -beach-drift that could be characterised without hesitation as -non-Hawaiian, were two well-worn pieces of acid pumice, less than an -inch in size. One of them was incrusted partially by the tubes of -annelids, and both of them had evidently been drifting about in the -Pacific for a long period, perhaps for years. They were such as occur in -abundance on the beaches of the South Pacific, and, in fact, on all the -shores of the Pacific Ocean, both temperate and tropical. Although I -carefully searched the stranded drift of many beaches in this group, no -other specimens of drift pumice were found. - -On different parts of Oahu the beach-drift was always made up of -materials derived from the vegetation of the coast adjacent. Of most -frequent occurrence were the seeds of Ipomœa pes capræ and Vigna lutea, -and the fruits of Scævola Kœnigii, Vitex trifolia, and Pandanus -odoratissimus. In addition, the empty seeds of Aleurites moluccana were -numerous, and there were occasional seeds of Thespesia populnea, -Colubrina asiatica, and Mucuna gigantea. On one beach there were a -number of fruits of Terminalia Katappa, showing but little signs of -ocean travel, and evidently derived from trees in the vicinity. This -tree was introduced by Europeans; but it is not unlikely that in a -generation or two it will become, without man’s aid, one of the -characteristic beach trees of Oahu. It may be remarked that the pods of -Acacia Farnesiana, a shrub now growing abundantly in Oahu near the sea, -are washed up in great quantities on the beaches of the west coast of -this island, and the seeds are to be seen germinating in numbers on the -beach, the seedlings striking into the sand. The pods float unharmed in -sea-water for four or five weeks, but the seeds, when freed, sink. - -Although the above evidence gives no indication of tropical drift of -non-Hawaiian origin on the beaches, it is probable, for reasons adduced -in Chapter VIII., that, in the winter, drift may be brought from -tropical America. - - - NOTE 31 (page 59) - - THE INLAND EXTENSION OF THE SHORE-PLANTS OF HAWAII - -_Cæsalpinia Bonducella._—According to Hillebrand, this plant, so -characteristic of the littoral floras of tropical regions, grows “in -gulches of the lower plains on all the islands,” no reference being made -to its occurrence on the beaches. It is very rarely to be seen on the -beaches of the large island of Hawaii; but it is to be found on the -lava-bound coasts, and from there it extends inland usually on old -lava-flows for five or six miles, and reaches sometimes considerable -elevations. In one locality I found it at 2,000 feet above the sea (see -page 188). - -_Cassytha filiformis._—Though a typical shore-plant in Fiji and other -tropical localities, it is rarely so in these islands. Hillebrand says -nothing of its station. It grows well in the lower open wooded regions, -and is frequently found amongst the blocks of old lava-flows near the -coast. - -_Cuscuta sandwichiana._—Unlike its fellow parasite Cassytha filiformis, -this species of Cuscuta, which is confined to this group, never came -under my notice away from the beach; and Hillebrand speaks of finding it -only at the coast (see page 366). - -_Ipomœa pes capræ_, as I observed it in the islands of Hawaii and Oahu, -is confined to the beach or to neighbouring sand-dunes. Hillebrand makes -no reference to its occurrence inland. This species in these islands -offers thus a great contrast to its behaviour in Fiji. - -_Scævola Kœnigii._—Whilst most at home on the sandy beaches, this plant -is also frequently met with in the island of Hawaii on scantily -vegetated lava-flows near the coast; but I never noticed it more than a -few hundred yards from the sea. - -_Tephrosia piscatoria._—Though it may occur on the beach, it is -generally found as described by Hillebrand on the rocky or rubbly ground -at the back of the beach, as well as further inland. It is common on the -old lava-fields of the island of Hawaii near the coast; and, according -to the natives, its seeds are disseminated by the wild goats that -frequent these localities. - -_Tribulus cistoides._—Hillebrand observes that this plant is found along -the sea-shore and on the lower plains. I found it most frequently on the -beaches and on the old lava-flows near the sea. - -_Vitex trifolia, var. unifoliolata._—It is confined, as Hillebrand -remarks, to the beaches. Neither in Oahu nor in Hawaii did I ever find -it straying inland, which is the more remarkable since this variety, or -one closely similar to it, is one of the most characteristic inland -plants of the Fijian strand-flora. - -_Vigna lutea._—This plant was found by me growing on the beaches and in -their vicinity. Hillebrand merely speaks of it as “growing at short -distances from the shore.” - -Some of the trees, usually littoral in their station in the tropical -Pacific, which are regarded as having been introduced in early times -into the Hawaiian group by the Aborigines (see Chapter VII.), behave, -nevertheless, quite like indigenous plants in the inland regions and in -the lower levels. This is true, for instance, of Hibiscus tiliaceus and -Pandanus odoratissimus, the last-named forming forests at the sea-board -extending in places far up the mountain slopes. The same, however, may -be said of other plants known to have been introduced since the -discovery of the islands, as in the cases of Cactus Tuna and of Ricinus -communis; and it also applies to Aleurites moluccana, the Candle-nut -Tree, which, although it could only have been introduced by the -Aborigines, now forms forests on the lower slopes of the mountains. - - - NOTE 32 (pages 19, 112, 165) - - THE FIJIAN SPECIES OF PREMNA - -I was much interested in the small trees and shrubs of this genus in -Fiji, more especially on account of the relation between the shore and -inland species. This is an Old World genus containing some eighty -species mainly characteristic of tropical Asia and Malaya, and -represented in the South Pacific archipelagoes by two species, one -Premna taitensis or tahitensis, spread over the region and very near P. -integrifolia, an Asiatic species; the other Premna serratifolia, an -Asiatic plant found in Fiji, the Marquesas, and other groups. Without -endeavouring to give a precise value to the Fijian plants, I will merely -describe the prevailing forms, which are, however, connected by -intermediate varieties. These trees, I may add, are known by the same -name in the various Pacific groups, “Avaro” or “Avalo” in Tahiti, -“Alo-alo” in Samoa, “Yaro” and “Yaro-yaro” in Fiji. - -The Fijian plants may be thus described.... (_a_) Premna serratifolia, -an inland tree, growing in open woods and on the outskirts of the -forest, 25 to 30 feet high, more or less hairy, leaves coarsely serrated -with long tapering points, putamen prominently tuberculated and -thick-walled. - -(_b_) Premna taitensis or P. integrifolia, a low straggling coast tree -or shrub of the beaches, the coral islets, the swampy borders of the -estuaries, and the inland talasinga plains, its usual height being eight -to ten feet, except in the inland plains, where it is dwarfed, and three -to five feet high. It is more or less glabrous, the leaves being -typically entire with obtuse or retuse and mucronate apices. The putamen -is thin-walled and relatively smooth. (_c_) Intermediate forms found -generally in the inland plains or talasinga regions. - -_On the Modes of Dispersal._—Speaking generally, the small drupes of -both species float at first, but the soft parts are soon removed by -decay, and the stone is freed. In the case of the coast species, P. -taitensis, the stones float indefinitely and are often found afloat in -rivers. In the case of the inland tree, P. serratifolia, most of the -stones sink at once, whilst the others sink in a few days. It is -probable that currents are one of the effective agencies in distributing -the coast species, but this could not apply to the inland tree. The -fruits of both the inland and the coast species would attract birds, and -the stones would resist injury in their crops. This is the agency -advocated by Prof. Schimper for the shore species, P. integrifolia, of -Indo-Malaya; and fruits referred with a query to this genus were found -in the collection of seeds and fruits obtained by me from the crops of -pigeons in the Solomon Islands (_Bot. Chall. Exped._, Introd. p. 46, -part IV. p. 312). - -_On the Cause of the Buoyancy of the Stone or Putamen of the Coast -Species._—This is primarily connected with the empty seed-cavities, the -four-celled stone usually developing only one seed, the other cavities -being empty. This inference was established by the dissection of a large -number of stones, but it will be seen from the table below that -one-seeded stones are also frequent in the case of the inland tree (P. -serratifolia), where they as a rule sink. With either species the -substance of the stone has no floating power, but with the shore -species, on account of the thin-walled stone, the empty seed-cavities -cause it to be specifically lighter than water whilst with the inland -species the walls of the stone are so thick that the empty spaces of the -unfilled seed-cavities do not effect the same result. It may be remarked -that when the coast species grows in the inland plains the buoyancy of -the stone is preserved. - - +-----------------------------+-------------+-------------+-------------+ - | | One-seeded | Two-seeded |Three-seeded | - | | stones. | stones. |stones. | - +-----------------------------+-------------+-------------+-------------+ - |Inland tree (P. serratifolia)| 73 per cent.| 23 per cent.| 4 per cent. | - |Coast tree (P. taitensis) | 92 per cent.| 8 per cent.| — | - +-----------------------------+-------------+-------------+-------------+ - - - NOTE 33 (page 63) - - DE CANDOLLE’S LIST OF PLANTS DISPERSED EXCLUSIVELY BY CURRENTS - -Drepanocarpus lunatus; Ecastaphyllum Brownei; Mucuna urens, D.C.; -Tephrosia piscatoria; Hibiscus tiliaceus; Rhizophora mangle; Guilandina -Bonduc, Linn.; Ipomœa pes capræ; Canavalia obtusifolia. - -I have experimented on the buoyancy of the fruits and seeds of all these -plants excepting the two first named. In five species the seeds float in -sea-water unharmed for several months. With Rhizophora it is the -floating seedling that disperses the plant. Neither the pods nor the -seeds of Tephrosia piscatoria are suited for dispersal by the currents. - - - NOTE 34 (page 64) - - THE LITTORAL PLANTS OF THE EASTERNMOST POLYNESIAN ISLANDS - -Except in the case of Hernandia peltata my authority here is the _Botany -of the “Challenger” Expedition_. Mr. J. H. Maiden gives some further -details of the flora of Pitcairn Island in a more recent paper -(_Austral. Assoc. Rep._, Melbourne, 1901, vol. 8), and Hernandia peltata -is included in his list. - - - NOTE 35 (page 68) - - DISTRIBUTION OF THE LITTORAL PLANTS WITH BUOYANT SEEDS OR FRUITS THAT - ARE FOUND IN THE FIJIAN, TONGAN, SAMOAN, TAHITIAN, AND HAWAIIAN GROUPS - -This list probably contains nearly all the species of the Polynesian -region, but it is not implied that these plants have been recorded from -all the groups (_vide infra_). - -(a) _Species found only in the Old World._—Calophyllum inophyllum, -Hibiscus diversifolius, Thespesia populnea, Heritiera littoralis, -Kleinhovia hospita, Carapa moluccensis, C. obovata, Smythea pacifica, -Colubrina asiatica, Mucuna gigantea, Erythrina indica, Strongylodon -lucidum, Dalbergia monosperma, Pongamia glabra, Inocarpus edulis, Derris -uliginosa, Afzelia bijuga, Barringtonia racemosa, B. speciosa, -Rhizophora mucronata, Bruguiera Rheedii, Terminalia Katappa, T. -littoralis, Lumnitzera coccinea, Pemphis acidula, Morinda citrifolia, -Guettarda speciosa, Wedelia biflora, Scævola Kœnigii, Cerbera Odollam, -Ochrosia parviflora, Cordia subcordata, Tournefortia argentea, Ipomœa -glaberrima, I. grandiflora, I. peltata, Aniseia uniflora, Clerodendron -inerme, Vitex trifolia, Hernandia peltata, Excæcaria Agallocha, Tacca -pinnatifida, Cycas circinalis, Pandanus odoratissimus, Scirpodendron -costatum. - -(b) _Species occurring in both the Old and New Worlds._—Hibiscus -tiliaceus, Suriana maritima, Ximenia americana, Dodonæa viscosa, -Canavalia obtusifolia, C. ensiformis, Vigna lutea, Sophora tomentosa, -Cæsalpinia Bonduc, C. Bonducella, Entada scandens, Gyrocarpus Jacquini, -Luffa insularum, Ipomœa pes capræ, Cassytha filiformis, Cocos nucifera. - -(c) _Species occurring in America to the exclusion of the Old -World._—Dioclea violacea, Mucuna urens, Rhizophora mangle. - -(d) _Species found only in Polynesia._—Canavalia sericea, Mucuna -platyphylla(?), Cynometra grandiflora, Serianthes myriadenia, Parinarium -laurinum(?), Premna tahitensis. - -_Remarks._—Of these seventy plants there is not one that has not come -within the scope of my observations and experiments. The West Coast of -Africa is included in the American region for reasons given in Chapter -VIII. For the other authorities on the buoyancy of these seeds and -fruits reference should be made to the list given under Note 2 and to -other parts of this work. About one or two of the plants, like Ipomœa -peltata, one scarcely knows whether they are most characteristic of the -coast-flora or of the inland-flora. - - - NOTE 36 (page 72) - - HAWAIIAN PLANTS WITH BUOYANT SEEDS AND FRUITS KNOWN TO BE DISPERSED BY - THE CURRENTS EITHER EXCLUSIVELY OR, AS IN A FEW SPECIES, WITH THE - ASSISTANCE OF FRUGIVOROUS BIRDS - -_Colubrina asiatica._—Usually regarded as confined to the Old World; but -since nearly all the species are American, that continent may be -considered as the probable home also of this species. Hillebrand gives -it a locality in the West Indies. - -_Dioclea violacea._—Tropical America. - -_Mucuna gigantea._—Old World. - -_Mucuna urens._—America, and extending to the African West Coast, which -is to be included in the American region of shore-plants. - -_Strongylodon lucidum._—Old World. - -_Vigna lutea._—Old and New Worlds. - -_Cæsalpinia Bonducella._—Old and New Worlds. - -_Scævola Kœnigii._—Usually regarded as confined to the Old World, but -according to the synonymy accepted by some authors it is also to be -ascribed to America. The genus is chiefly Australian, and it is possible -that the littoral species may have reached America through the agency of -birds, since all the species of the genus possess fruits that would -attract frugivorous birds. - -_Ipomœa glaberrima_ (Boj.).—Old World. - -_Ipomœa pes capræ._—Old and New Worlds. - -_Vitex trifolia._—Old World. The genus is also dispersed by pigeons. - -_Cassytha filiformis._—Old and New Worlds. Like Scævola the genus is -chiefly Australian, and here, also, the fruits of the littoral species -are not only dispersed by the currents, but are known to be also -disseminated by fruit-pigeons. - -It is possible that birds may have taken a predominant part in the -dispersal of the species of Scævola, Vitex and Cassytha. - -There thus remain nine species for consideration. Of these two are -exclusively American, three are found in both the Old and New Worlds and -four are usually regarded as exclusively Old World plants, but one of -them (Colubrina asiatica) has a fair claim to be regarded as of American -origin. Thus it is quite possible that six out of these nine plants were -brought to Hawaii from America through the agency of the currents. - - - NOTE 37 (page 78) - - ON VIVIPARY IN THE FRUITS OF BARRINGTONIA RACEMOSA AND CARAPA OBOVATA - -As observed by me in the Rewa delta, Fiji, there was no external -evidence of such a process in the case of the fruits on the trees; but I -did not pay very special attention to the matter, and it will be -gathered from Chapter XXX. that the initial stage of germination may -show no indication in the appearance of the fruit. More observation is -needed for both species. As indicated in Note 50, the structure of the -seed of Barringtonia racemosa is suggestive of a lost viviparous habit. -With regard to Carapa, Schimper (p. 43) remarks that he has never -observed vivipary; but Miquel, in his _Flora Indiæ Bataviæ_, -particularly speaks of the seeds germinating in the capsule. I think -this is very likely, and that perhaps even the rupture of the capsule -may be partly due to this cause. - - - NOTE 38 (page 78) - -ON THE TEMPERATURE AND DENSITY OF THE SURFACE-WATER OF THE ESTUARIES OF - THE REWA RIVER IN FIJI, AND OF THE GUAYAQUIL RIVER IN ECUADOR - -(a) _The Rewa Estuary._—My observations were made mostly in the warm, -wet seasons, from October to January, 1897-99, and generally in the -vicinity of the Roman Catholic Mission. The density varied usually -between 1·000 and 1·010, the water being quite fresh after heavy rains -inland. Though the density was usually greatest at high water, this was -by no means always the case. The temperature of the water in dry weather -varied from 79° to 84° F. With the river in flood after heavy rains it -fell to 75° and 76°. As a rule, the fresher the water the lower the -temperature, but this was not invariable. There was evidence of -super-heating in the estuary, the water there having sometimes a -temperature of 82° or 83°, when the water higher up the river as far as -Viria was two or three degrees cooler, the sea-temperature being 79° to -80°. The average temperature of the water of the estuary during the -season would be 80 to 81°. - -(b) _The Estuary of the Rio Guayas, also known, as the Guayaquil -River._—My observations were made in the last week of February and in -the first half of March, 1904. Whilst the sea-temperature a few miles -off the Ecuador coast varied from 76° to 80° F., the water of the -estuary from the mouth up to Guayaquil ranged from 79° to 86°, whilst -rather higher up the river the temperature was about 79° or 80°. The -super-heating of the estuary is thus directly indicated. It was well -marked in the lower part of the estuary during one of my ascents of the -river. - - _Surface-temperatures of estuary of the Guayaquil River, March 13, 1904, - 11 a.m. to 4 p.m.; tide running up._ - - Sea-temperature 5-10 miles off the mouth 79·7 - Estuary-temperature at the mouth, off Puna 82·7 - Estuary-temperature 3 miles above Puna 84·4 - Estuary-temperature 15 miles above Puna 86·5 - Estuary-temperature 25 miles above Puna 82·5 - Estuary-temperature off Guayaquil 81·8 - -The water of the estuary was, as a rule, cooler with the ebbing tide. - -The density of the estuary-water at the mouth opposite Puna during the -two days the ship was in quarantine ranged from 1·004 to 1·016, being -generally about 1·010, and salter with the up-going tide. Off Guayaquil -the water during the ebbing tide was quite fresh and, from an Ecuadorian -standpoint only, potable, whilst at high water it may be a little -brackish. The sea-water has much freer access to the channels in the -mangrove-district at the back of the city of Guayaquil, where at high -water I found the density to be 1·014. - -Off Puna, on Feb. 25, I noticed that the surface-current which was -running down the stream was from one to two fathoms deep, whilst below -it was a strong current running up the river which carried my -thermometer up against the surface-current. - - - NOTE 39 (page 82) - - ON THE PACIFIC SPECIES OF STRONGYLODON - -Hillebrand in his _Hawaiian Flora_, following Seemann, regards S. -lucidum, Seem., and S. ruber, Vogel, as one species found in Fiji, -Hawaii, and Tahiti, and by the former placed also in Ceylon. Hillebrand -and Seemann are followed by Drake del Castillo as regards the Tahitian -species. Taubert, in his monograph on the Leguminosæ (Engler’s _Pflanz. -Fam._, Teil 3, Abth. 3, 1894), takes the same view of the Polynesian -species and of its wide distribution. However, in the _Genera Plantarum_ -and in the _Index Kewensis_, the Asiatic and Polynesian species have -been always kept apart. The two species of the genus mentioned in the -first work are increased to five in the _Index Kewensis_, viz., one in -Fiji (S. lucidum), one in Hawaii (S. ruber), two in Madagascar, and one -in the Philippines. - - - NOTE 40 (page 88) - - PRECAUTIONS IN TESTING SEED-BUOYANCY - -Many seeds and fruits require a few hours’ soaking before they sink; and -when small they will rest a long time on the surface of still water, but -a touch with the finger or a drop of water will send them to the bottom. -A few will float a few days (3 or 4) before sinking; but such are -included in the non-buoyant group. Only in rare cases does prolonged -drying increase the period of flotation by more than a few days, -examples being given at the end of the Table of Buoyancy results under -Note 10. Adherent air-bubbles, a common cause of adventitious buoyancy, -must always be removed. - - - NOTE 41 (page 91) - - THE BUOYANCY OF THE SEEDS OF CONVOLVULUS SOLDANELLA IN FRESH WATER AND - SEA-WATER COMPARED - -The experiments were commenced at the close of September, 1894, and -covered six months. At the end of this period in Mr. Millett’s -experiment, 56 per cent. of the seeds were afloat in fresh water, and 62 -per cent. in sea-water; whilst in my own experiment 72 per cent. floated -in fresh water, and 65 per cent. in sea-water. I was indebted to Mr. -Millett’s courtesy for the seeds. - - - NOTE 42 (page 96) - - ON SECULAR CHANGES IN SEA-DENSITY - -Exact data bearing on this subject are not at my disposal; but it would -seem that geologists have formed conflicting conclusions from similar -premises. There is the view that the composition of the ocean water was -very different in early geological periods (_Encycl. Brit._, x., 221); -but I should imagine that the character of the crustacean fauna of those -seas would negative any great divergence from the present condition. -Suess implies that the ancient seas carried the same minerals in -solution that they do now, and it is to be inferred in a similar -proportion (E. de Margerie’s French edition of _Das Antlitz der Erde_, -ii., 343 and 345). - - - NOTE 43 (page 102) - - ON THE MUCOSITY OF SMALL SEEDS AND SEED-LIKE FRUITS WHEN WET - -I paid considerable attention to this subject from the standpoint of -dispersal some years ago, and published most of the results in _Science -Gossip_ for Sept., 1894. This peculiar quality of seeds had been noticed -by Dr. Kerner in his _Pflanzenleben_ (vol. i., 1887-91), and was -regarded as illustrating a mode of dispersal of seeds by adherence. As a -rule, such seeds when placed in water become coated with mucus in a few -minutes, or within an hour, and when allowed to dry on feathers they -adhere as firmly as if gummed. I found that this quality is not affected -by prolonged drying, as in the cases of Nepeta glechoma and Salvia -verbenaca, where it was exhibited to the same degree after the seed-like -fruits had been kept from one to three years. I especially tested about -110 British plants that were likely to display this quality, and found -that about a dozen exhibited it in a marked degree, and if to these we -add those plants with seeds that display it to a limited extent so that -they merely become adhesive when wetted, the total would be nearly -twenty. It will be noticed from the list subjoined that the plants -showing marked mucosity belong to twenty genera and to ten families, the -Labiatæ and Cruciferæ predominating. Although in some genera, like -Plantago, there is reason to suppose that the seeds of all the species -would behave in this fashion, it would be wrong to infer that this is -usually the case, six genera being indicated below to which such a rule -would not apply, and doubtless the number could be extended. These -plants in England mostly occur at the roadside, on waste ground, and in -dry meadows. It may be added that although in most cases the seeds -appear in water to emit mucus, “exuded mucilage” being the expression -used in the English edition of Kerner’s work, in some instances, as with -Helianthemum vulgare, there appears to be a dissolving process affecting -the outer seed-covering. - -I. _Plants with Seeds or Seed-like Fruits that emit Mucus to a Marked - Degree when placed in Water._ - - Arabis thaliana, G. Cruciferæ. - Camelina sativa, K. Cruciferæ. - Teesdalia, K. Cruciferæ. - Capsella bursa-pastoris, G. Cruciferæ. - Lepidium sativum, D. Cruciferæ. - Helianthemum vulgare, G. Cistaceæ. - * Viola tricolor (Field Pansy), G Violaceæ. - Linum usitatissimum, D. Linaceæ. - Linum, K. Linaceæ. - * Matricaria chamomilla, K. G. Compositæ. - * Senecio vulgaris, G. B. Compositæ. - Collomia, K. Polemoniaceæ. - Gilia, K. Polemoniaceæ. - * Veronica beccabunga, S. Scrophulariaceæ. - Ocimum basilicum, K. Labiatæ. - Salvia verbenaca, G., &c. Labiatæ. - Salvia, K. B. Labiatæ. - * Nepeta glechoma, G. Labiatæ. - * Dracocephalum, K. Labiatæ. - Prunella vulgaris, G. Labiatæ. - Plantago, K. Plantagineæ. - Plantago major, lanceolata, maritima, G. Plantagineæ. - Luzula campestris. G. Juncaceæ. - -_Explanation of Abbreviations._—The capital letter following the name -indicates my authority, which is not necessarily the oldest in each -case: B = Beal; D = Darwin; G = Guppy; K = Kerner; S = Scott Elliot. The -respective works quoted will be found at the end of this volume. The -papers of Darwin quoted will be found in _Journ. Linn. Soc._, “Botany,” -vol. i., 1857, and in the _Gardeners Chronicle_ for 1855. - -The asterisk is placed before those genera of which other species -examined by me exhibited no mucosity; these species are Arabis hirsuta, -Viola canina, V. palustris, Matricaria inodora, Senecio aquaticus, -Veronica agrestis, V. arvensis, Nepeta cataria, Dracocephalum -canariensis. - -II. _Plants with Seeds or Seed-like Fruits which in my Experiments only - exhibited Mucosity in a Slight Degree, becoming merely “Sticky” or - Adhesive when placed in Water._ - -Arabis albida, Chrysanthemum leucanthemum, Lamium purpureum -(occasionally), Thymus sp., Juncus bufonius, J. communis, J. glaucus, J. -squarrosus. - -III. _Plants with Seeds or Small Fruits that exhibit Adhesiveness in the - Dry State and are apt to stick to one’s fingers._ - -Adenostemma viscosum, Lycopus europæus, Piper Macgillivrayi, &c. One may -include here also Lagenophora (see page 276) as well as the familiar -instances of Pisonia (page 347) and Boerhaavia (page 356). - - - NOTE 44 (page 121) - - ON THE EFFECTS OF INLAND EXTENSION ON THE BUOYANCY OF THE SEEDS OR - FRUITS OF LITTORAL PLANTS - -When in Fiji I experimented on the buoyancy of the following -beach-plants that had extended far into the interior of Vanua Levu, as -will be found described in Note 22. Those tested were Cassytha -filiformis, Cerbera Odollam, Ipomœa pes capræ, Morinda citrifolia, -Premna tahitensis, Scævola Kœnigii, and Tacca pinnatifida. In all but -Cerbera Odollam, where I contented myself with establishing that the -fruits floated buoyantly in sea-water, the experiments were prolonged -for many weeks and often for several months; and in some cases, as with -Ipomœa pes capræ, three or four experiments were made on seeds from -different inland localities. The result was to establish in all cases -that the floating powers were as great with the inland as with the coast -plants of the same species; nor could any structural difference of -importance be noticed. It should be observed that there is every reason -to believe that the “talasinga” plains of Fiji have been occupied by the -intruding beach-plants for many ages. - - - NOTE 45 (page 122) - - TABULATED RESULTS OF THE CLASSIFICATION, ACCORDING TO SCHIMPER’S - APPLICATION OF THE NATURAL SELECTION THEORY, OF THE BUOYANT SEEDS AND - FRUITS OF THE TROPICAL LITTORAL PLANTS ON THE BASIS OF THE STRUCTURAL - CHARACTERS CONCERNED IN BUOYANCY - - Column headings: - - #: Number. - %: Per cent. - A: Percentage of non-adaptive species. - B: Percentage of adaptive species. - - +------------------+-----------------------------+---------------+--------+ - | | Classification of species. | | | - | +-------------------+---------+ Proportion of | Total | - | | Non-adaptive. |Adaptive.| non-adaptive | number | - | +---------+---------+---------+ and adaptive | of | - | Region. | First | Second | Third | species. | species| - | | group. | group. | group. | | dealt | - | +----+----+----+----+----+----+-------+-------+ with. | - | | # | % | # | % | # | % | A | B | | - +------------------+----+----+----+----+----+----+-------+-------+--------+ - |Pacific Islands | 27 | 40 | 10 | 15 | 30 | 45 | 55 | 45 | 67 | - | | | | | | | | | | | - |Pacific Islands, | | | | | | | | | | - |tropical America, | | | | | | | | | | - |and Indo-Malaya | 28 | 35 | 12 | 15 | 40 | 50 | 50 | 50 | 80 | - +------------------+----+----+----+----+----+----+-------+-------+--------+ - -NOTE.—If to the last we add the eight British shore plants, the buoyant -fruits of which are described in Chapter XII., three non-adaptive and -five adaptive, we get a proportion of adaptive species for temperate and -tropical regions of fifty-one per cent. This is probably fairly typical -of the world generally; but it must be remembered by the reader that the -author regards them all as non-adaptive. In that case, the table can be -used for the numerical results of the three groups which are based only -on structural characters without reference to any theory. - - - NOTE 46 (page 124) - - ON THE MODES OF DISPERSAL OF THE GENUS BRACKENRIDGEA. - -Seed-vessels of this genus found afloat in the New Guinea drift are -described by Mr. Hemsley as having two curved cavities crossing each -other one containing a seed, the other empty. “This empty cavity,” it is -stated “gives the fruit its buoyancy” (_Bot. Chall. Exped._, iii., 289; -plate 54) Dr. Beccari, in the English edition of his _Wanderings in -Borneo_, p. 187, speaks of the closed air-containing cavities in the -seed-vessels, or rather “stones,” of this genus as probably giving them -buoyancy and thus enabling them to be dispersed by currents. He points -out that the fleshy covering of these fruits would also aid their -dispersal by birds. The Italian botanist implies that the two Bornean -species grow in swamps. The Fijian species, as observed by me in flower -in Vanua Levu, grew in the dry talasinga districts bordering the -Mathuata coast, the locality where Seemann found the plant. One of the -most recent accounts of the genus is given by Van Tieghem in his memoir -on the Ochnaceæ in _Ann. des. Sci. Nat. Bot._, tome 16, 1902. According -to him there are nine species, all from Malaya and New Guinea, with the -exception of one in Fiji. Previous authors have also referred to -Queensland and Zanzibar species. However, all the species have a limited -distribution, a fact which plainly assigns to birds the principal share -in the dispersal of the genus. - - - NOTE 47 (page 125) - - ON THE TRANSPORT OF GOURDS BY CURRENTS - -Small calabashes or bottle-gourds are not uncommonly to be found -floating in the Fijian estuaries and stranded on the beaches; and I have -also found them in the sea off the coasts. They are usually more or less -globular, 3 or 4 inches across, and are evidently able to float for very -long periods and to carry the seeds unharmed. Most of those I examined -from the drift were dry inside and contained the seeds dried together -into a loose ball about an inch in size. The seeds are not those figured -in Gaertner’s _De Fructibus et Seminibus_, as belonging to Lagenaria -vulgaris, and more resemble those of Cucurbita, but are non-buoyant. One -of these gourds, picked up by me in the sea in Fiji, was placed in -sea-water, and two months later was still floating buoyantly. After -being then kept dry for seven months, it was broken open; and ten of the -seeds were put in soil, two of them germinating in a few days. - -In Ecuador gourds similar in size and shape were frequently observed by -me floating in the drift of the Guayaquil River and stranded on the -sea-beaches. The seeds are similarly caked together in a loose mass in -the cavity of the fruit. Their characters indicate that they belong to -another species of gourd; and they differ also from the Fijian seeds in -their buoyancy, some of them in my experiments floating two months and -afterwards germinating. - -It has been known since the days of Ström and Gunnerus, two Norwegian -naturalists of the 17th century, that gourds and calabashes are from -time to time stranded with other Gulf-stream drift on the coasts of -Norway. We learn from Sernander that those found are usually worked -calabashes; but he alludes to one that was unworked and contained -several seeds (see Sernander, p. 119). - -It is scarcely likely that a seed-carrying gourd stranded on a beach -would be able to establish the plant without the aid of man; but it -seems highly probable that gourds have often been introduced into new -countries by the currents and that man has afterwards cultivated them. -These plants may be contrasted with that remarkable Cucurbit, Luffa -insularum, a genuine littoral plant, the seeds of which, and not the -fruits, are dispersed in the Pacific by the currents (see page 426). - - - NOTE 48 (page 126) - - ON THE USELESS DISPERSAL BY CURRENTS OF THE FRUITS OF THE OAK (QUERCUS - ROBUR) AND OTHER SPECIES OF QUERCUS, AND ALSO OF THE HAZEL (CORYLUS - AVELLANA) - -The fruits of different species of Quercus are of not infrequent -occurrence in the seed-drift both of the temperate and tropical regions, -being brought down by the rivers to the sea and then stranded on the -neighbouring beaches. They were amongst the drift gathered by Mr. -Moseley in the open sea, 70 miles off the New Guinea coast (_Bot. Chall. -Exped._, iv., 294). I found them on the beaches of Keeling Atoll where -no oak exists, and on the beaches of the south coast of Java; whilst -Prof. Schimper noticed them among the stranded drift of the Java Sea, -and Prof. Penzig found them stranded on the shores of Krakatoa. They -also came under my notice on the Sicilian beaches and on the Italian -coast at Cumæ. Those of Quercus robur are to be found on the English -beaches and in the autumn drift of the Thames, but they soon sink and -disappear from river-drift. They are referred to by Dr. Sernander as -frozen with other floating seeds in the ice of the Scandinavian rivers; -but he evidently does not regard them as possessing much independent -floating power. - -Some years ago the author made a number of experiments on the buoyancy -of the acorns of Quercus robur, and he formed the conclusion that when -freshly collected not more than 4 to 8 per cent. of mature fruits will -float in fresh-water, and not more than about 10 to 12 per cent. in -sea-water, but that in either case they all sink in a day or two. -Immature acorns float much longer, and it is these that mostly figure in -the drift. However, unlike most fruits of little initial buoyancy the -mature fruits gain considerable floating power by drying. Of some that -had been kept for seven months 20 per cent. floated after four weeks in -sea-water and 15 per cent. after 10 weeks.... It may be added that, -according to Thuret, the fruits of Quercus ilex have little or no -floating power. - -The buoyancy of the fruits of Quercus is due entirely to the cavity left -by the shrinking of the kernel. I never remember to have found one with -a sound seed amongst the drift in England and Sicily; and I should doubt -much whether those in the tropical drift retain their germinating -powers. But, apart from this, the genus Quercus finds in its own -constitution or habit the greatest obstacle in most species to the -adoption of a littoral station. However, there are exceptional -tendencies displayed by the evergreen oaks; and this is very -significant, since in their xerophilous leaves they possess the -preliminary qualification for a station near the sea. Quercus ilex, it -is well known, shows a partiality for the sea-air, and Q. virens, the -“live oak,” flourishes near the sea in the southern states of America, a -maritime variety being distinguished by botanists. One of the -willow-oaks of America, Q. phellos, which grows in swampy land, also has -a beach variety. - -The Hazel-tree (Corylus avellana) must be placed in the same category -with Quercus. I found the empty nuts commonly amongst the stranded drift -of the Sicilian and English beaches. The fruits were also frequently -noticed by Dr. Sernander in the Scandinavian sea-drift; but he says -nothing of their empty condition. Mr. Darwin remarks, in the _Origin of -Species_, that he found that fresh hazel-nuts sank, but that after -drying a long time they floated for ninety days and subsequently -germinated. The floating-power is no doubt due to the cavity arising -from the shrinking of the kernel, and it is to this cause that Dr. -Sernander attributed the slight initial buoyancy observed by him. -However, the hazel, like the common oak, lacks the habit that would fit -it for a station by the sea, and, whatever capacity its fruits may -possess for dispersal by currents, it is quite useless for the spread of -the species. - - - NOTE 49 (page 131) - - ON THE DISTRIBUTION OF IPOMŒA PES CAPRÆ, CONVOLVULUS SOLDANELLA, AND - CONVOLVULUS SEPIUM - -Whilst Ipomœa pes capræ is cosmopolitan in the tropical zones, -Convolvulus soldanella is cosmopolitan in both the north and south -temperate zones; but, as might be expected, the two species at times -meet and their areas overlap. Thus, according to Mr. Cheeseman (_Trans. -New Zealand Inst._, xx., 1887), they meet in the Kermadec Islands, in -the South Pacific, in about latitude 30°. From my observations on the -coast of Chile it would seem that C. soldanella in its northward -extension fails somewhere between Valparaiso and Coquimbo, that is to -say, between 33° and 30° S. lat. Gay merely refers to the plant as -existing in North Chile, which in his time would include the coast -between 33° and 24° S. lat. It intrudes within the “thirties” on the -coast of California and is found in Madeira in about 33° N. lat. Ipomœa -pes capræ in its turn extends into subtropical regions, being recorded -from the Kermadecs, as above noted, and from the Bermudas in 32° N. lat. -Owing probably to special physical conditions of the coast, which are -referred to in Chapter XXXII., this plant is evidently limited to the -tropics on the west coast of South America. It did not come under my -notice on the beaches of North Chile, and it is apparently not mentioned -by Gay in his work on the Chilian flora. - -Convolvulus sepium, the frequent inland associate of the littoral C. -soldanella over the temperate regions of the globe, belongs to the same -section of the genus (Calystegia). Its extraordinary occurrence by -itself in the island of St. Paul, in the Southern Ocean, about fifty -yards from the shore (_Bot. Chall. Exped._, ii., 153, 264), almost -suggests that we have here a dimorphic species with a littoral and an -inland form; and its existence in the Azores is in this connection very -remarkable. It may be here noted that of three plants raised from seeds -found in the beach-drift near Palermo two had the foliage of C. sepium -and one of C. soldanella. Perhaps one of my readers, in imitation of De -Vries with Œnothera, might be able to settle this point by raising some -hundreds of seedlings from the seeds of the beach species. It is -possible that the relation between these two species of Convolvulus may -be in some respects akin to that between Cæsalpinia Bonducella and C. -Bonduc, two littoral plants that accompany each other over much of the -tropical zone. - -The student of dispersal will, however, find some curious gaps in the -distribution of Convolvulus soldanella even in the temperate regions; -and it will be curious to observe how they affect the distribution of C. -sepium. He will have to answer the query of De Candolle:... “Admitting, -if one wishes, that the currents have transported this marine species, -how comes it that it chances to be in the Pacific and in Europe, without -occurring on the east coasts of America and on the east and west coasts -of Africa?” (_Geographie Botanique_, ii., 1056). He will have to explain -why some botanists give C. soldanella a habitat in the tropics, as in -the Indian region. Schimper, who investigated this point, says that he -arrived at no certain result (p. 127). See Notes 13 and 41 and pages 29, -91, for further remarks on these two species of Convolvulus. - - - NOTE 50 (pages 79, 132) - - ON THE STRUCTURE OF THE SEEDS AND FRUITS OF BARRINGTONIA - -As regards the fruits and their coverings, the littoral and inland -species of Fiji evidently fall into different sections, the first named -(B. speciosa and B. racemosa) being distinguished by their outer fibrous -husk, to which the buoyancy is due, the last-named (B. edulis and an -undescribed species) possessing a hard stone surrounding the seed, and -here the fruits sink or float only for limited periods. - -The fruits of B. edulis have an outer almost fleshy covering, a little -fibrous at the outside, and the hard ligneous “stone,” containing an -edible seed, requires a hammer to break it. They float heavily for three -or four weeks, whereas those of the littoral species float for many -months. In the case of another inland species found by me growing as a -small tree 12 feet high on the slopes of Mount Seatura in Vanua Levu at -an elevation of 1,000 feet above the sea, the seed was similarly -protected by a hard “stone” that could only be broken with an axe, and -the fruit was non-buoyant, with thin and perishable outer coats. - -This mountain species of Fiji, which I may name Barringtonia seaturæ, -has the general habit of B. racemosa, with which the natives persisted -in linking it; whilst the fruit and foliage come nearer to those of B. -edulis. The leaves are entire, taper at the base, and have a petiole 1 -inch long. The fruits are oblong, at least 3 inches in length, and are -obscurely angled. - -It would appear from Schimper’s description (p. 173) that the fruits of -the Malayan Barringtonia excelsa possess both the hard stone-shell of -the inland Fijian species and the dry air-bearing fibrous husk of the -littoral species. This is of special interest, since the tree is both a -coast and an inland species. - -The following notes on the structure of the seeds of Barringtonia were -made whilst I was drifting about in my canoe in the creeks of the Rewa -delta in Fiji; and whatever may be their deficiencies they have the -merit of having been written in the home of the plants.... When we cut -across a seed like that of B. racemosa or B. speciosa, we observe that -the different parts of the embryo are indistinguishable, being united -into a homogeneous, firm, fleshy mass. But if we look closely we notice -a central fusiform portion marked out from the surrounding parts by a -faint line, along which a delicate membrane of vascular tissue has been -developed. When “germination” begins, though, as the reader will -subsequently perceive, this term is here hardly appropriate, the real -nature of this singular structure becomes more apparent, as is indicated -in the accompanying figure. The central fusiform portion proves to be -the young plant without cotyledons and growing at either end to form the -root and the stem. The delicate investing membrane becomes thicker and -more apparent as germination proceeds, extending upwards and downwards -with the growth of the stem and root and forming a cortical covering in -either case. The investing fleshy portion of the seed, which is now -separable with the fingers, remains attached to the lower part of the -seedling for some time, being evidently a source of nutriment, and gives -a bulbous appearance to the young plant. Young bulbous plants of B. -racemosa, 1 to 2 feet high, are very common on the edge of Fijian -mangrove swamps where the parent tree thrives. The seedlings of B. -speciosa have the same appearance, but the outer fleshy part of the bulb -is not so thick. - - [_To face page 574._ - -[Illustration: _B. racemosa._, _B. speciosa._] - -Diagrams illustrating the structure of the growing seeds of Barringtonia -(two-thirds the natural size). That of B. speciosa represents a seed -removed from a fruit displaying the young plant protruding two or three -inches. That of B. racemosa represents the lower end of the seedling -when the plant is eighteen inches high. - - _a_ = the exorhiza. - _b_ = the neorhiza invested by the medullary sheath. - -This structure of the seeds of Barringtonia speciosa and of B. racemosa -was for a long time meaningless to me, until one day, whilst seated on -the banks of the Lower Rewa, with a number of the sected seeds and -bulbous seedlings gathered around, I reflected that the fruits of the -latter species that floated past me in the river-drift were nearly -always germinating. This called up “vivipary” to my mind; and as I -looked at the Rhizophora seedlings dangling from the branches of the -mangrove-trees close by, it occurred to me that this seed-structure -might be the result of a lost viviparous habit. One apparently had to -deal here not with an ordinary seed containing an embryo in the midst of -albumen, but with a seed in an arrested stage of germination surrounded -by a body that might perhaps prove homologous with the “cotyledonary -body” of Rhizophora. The process of development that goes on without a -break in Rhizophora, from the fertilisation of the ovule to the -detachment of the seedling from the branch, was here, as I considered, -arrested after germination had begun, but before the protrusion of the -seedling from the fruit. With nearly all plants, as I reflected, there -is a rest-stage of varying length, which might be called the seed-stage. -With the mangrove-genera, Rhizophora and Bruguiera, I had convinced -myself by a long series of observations, the results of which are given -in Chapter XXX., that this rest-stage does not exist. It occurs, I -argued, in Barringtonia, but only after germination has begun, and, -therefore, displaced when compared with the typical seed-stage of most -plants. - -In this connection it may be noted that a difference in germinating -behaviour might be expected between the two shore species on account of -their difference in stations, Barringtonia speciosa growing on the sandy -beach, and B. racemosa in the wet ground around a mangrove-swamp. There -is a strong suspicion that the rest-stage in B. racemosa is very short, -though I never found germination in progress on a tree (see Note 37). -There is no doubt, on the other hand, that the rest-stage of B. speciosa -is often, as with most other plants, very long. This, then, was my -lesson from the Barringtonia fruits on the banks of the Rewa, and the -question arose whether this interpretation of these curious -seed-structures accorded with the opinion formed of their nature by -botanists. - -Curious seed-structures of this kind must have their significance in the -history of the plant; and on returning to England I looked a little -further into the matter. To follow up this kind of inquiry, however, -would carry me far beyond the limits prescribed for this note, and I -have only treated it here in a tentative fashion. Different botanists of -eminence have paid attention to this subject, amongst them Roxburgh, -Thomson, and Miers (see Dr. T. Thomson in _Journ. Linn. Soc. Bot._, vol. -ii., p. 47, 1858, and Mr. J. Miers in _Trans. Linn. Soc. Bot._, vol. i., -1880). It would appear that the seed-structure of Barringtonia is also -found in Careya, a genus of the same Myrtaceous tribe, and in Garcinia -and other genera of the Guttiferæ, as well as in other inland plants. - -Mr. Miers, after reviewing the opinions of his predecessors, gives the -results of his own investigations. The solid embryo found in -Barringtonia and many other genera consists, he observes, (_a_) of an -external portion, the “exorhiza,” which nourishes the germinating seed -and then dies away; (_b_) of an internal portion, the “neorhiza,” which, -growing at each end, forms the central portion of the stem and root; and -(_c_) the “medullary sheath” of Mirbel, that lies between the two, and -is composed of elementary vascular tissue, which ultimately gives origin -to the wood, bark, and leaves of the stem and yields woody fibre to the -root. The exorhizal portion in some cases, as in Barringtonia -acutangula, splits into four parts during germination. Mr. Miers -compares this seed-structure with that of Rhizophora, employing the same -terms, “neorhiza” for the internal portion which forms the seedling, and -“exorhiza” for the external portion which merely nourishes it. However, -I may add that the exorhizal portion in Rhizophora, as shown in Chapter -XXX., is now regarded as formed by the coalesced cotyledons, and is -termed the “cotyledonary body”; so that by implication the corresponding -part of a Barringtonia seed should be regarded from the same standpoint. - -It may be apposite to notice here that Barringtonia racemosa displays -one capacity which does not appear to belong to B. speciosa. The -branches stuck in wet soil throw out roots and establish themselves. -This capacity of vegetative reproduction is turned to account by the -Fijians, who make “live-fences” of this tree in wet localities. - - - NOTE 51 (page 135) - - ON A COMMON INLAND SPECIES OF SCÆVOLA IN VANUA LEVU, FIJI - -This is a tall shrub, or small tree, nine or ten feet high, which -corresponds with S. floribunda, Gray, as far as Seemann describes it. It -has small, black, juicy drupes, well suited for dispersal by birds, -having no “suberous” mesocarp as in the shore species (S. Kœnigii), and -no capacity for dispersal by currents. It grows, much like the Hawaiian -inland species, in exposed situations where there is plenty of light, as -on mountain-peaks, at the borders of forests, in open-wooded districts, -and in the plains, and is to be found at all elevations from near the -sea up to the highest mountain summit (3,500 feet) when the station is -suitable. I noticed it on the higher slopes and frequently on the tops -of nearly all the principal mountains that I climbed. It is evident that -birds carry the “stones” from one mountain-peak to another, and no doubt -they explain the presence of the species in Tonga. Dr. Seemann speaks of -it as a beach plant in Viti Levu. The plant familiar to me in Vanua Levu -is only on very rare occasions to be seen as an intruder in the -beach-flora. - - - NOTE 52 (page 137). - -ON THE CAPACITY FOR DISPERSAL BY CURRENTS OF COLUBRINA OPPOSITIFOLIA, AN - INLAND HAWAIIAN TREE - -The seeds in my experiments sank within ten days; but they are not -readily detached from the fruit, as in the case of the buoyant seeds of -the littoral species (C. asiatica). The fruits, which may float for a -week or two, break down, as Hillebrand observes, tardily and -imperfectly, and could give but little assistance to dispersal by water. - - - NOTE 53 (page 141) - - ON THE GENUS ERYTHRINA - -We have in E. indica a widely distributed littoral species, ranging from -India through Malaya to eastern Australia, and over nearly all the -groups of the Pacific, reaching to Tahiti and the Marquesas, but not -occurring in Hawaii. It is associated in Fiji and Tonga with another -shore-species, E. ovalifolia, Roxb., found also in India and Malaya. I -did not come on the second species in Fiji, and according to Seemann it -is rare. It is possible that there is a genetic connection between the -two; and it is noteworthy that in one case Seemann was uncertain (p. -426) whether the species was E. ovalifolia or only a variety of E. -indica. - -In different parts of their areas both these species may be found -inland. This no doubt is to be connected with their occasional -cultivation. The Polynesians who esteem E. indica for its handsome -scarlet flowers and its scarlet seeds often plant it near their houses; -but it is curious that if we look at the pages of Seemann, Horne, and -one or two other botanical authors who have written on the Pacific, we -find no reference to its littoral station, the first-named botanist -merely characterising it in Fiji as occurring “wild or planted.” - -However, in various localities in Fiji, as on the shores of Natewa Bay -in Vanua Levu, Erythrina indica thrives as a characteristic beach tree. -Dr. Reinecke speaks of it as widely spread on the Samoan coasts; and the -French botanists refer to it as a tree of the Tahitian beaches. Prof. -Schimper frequently mentions the two littoral species of Erythrina as -amongst the components of the Malayan strand-flora. Dr. Treub, when he -visited Krakatoa in 1886, three years after the eruption, noticed some -young plants of Erythrina growing on the shore; whilst Prof. Penzig in -1897 found that both E. indica and E. ovalifolia had established -themselves on the beach. Mr. Kurz again is quoted by Prof. Schimper (p. -170) as including E. indica amongst the “beach-jungle” of Pegu. - -There is abundant evidence in support of the dispersal of the genus by -currents. I have observed the seeds of Erythrina indica on the beaches -of Keeling Atoll. Schimper noted Erythrina seeds amongst the stranded -drift of the Java Sea. Treub remarked young plants of the genus growing -on the shore of Krakatoa three years after the great eruption, and -Penzig places Erythrina indica and E. ovalifolia amongst the -beach-plants brought to Krakatoa through the agency of the currents. The -seeds of E. indica not infrequently came under my observation stranded -on the Fijian beaches and floating in the Rewa estuary; and in an -experiment made in Fiji they still floated after five months in -sea-water. Mr. Hemsley years ago formed the opinion, from the drift -collections at Kew, that the genus was dispersed by the currents. I may -here add in further illustration of this point that Erythrina seeds were -found by me in South America floating in numbers in the Guayaquil -estuary and stranded on the beaches of Ecuador. - -It is noteworthy that, unlike some of the other shore-plants, Erythrina -indica has at least three sets of names in the South Pacific. Thus it is -known as Rara and Ndrala in Fiji, Ngatae in Samoa, Futuna, and -Rarotonga, Atae in Tahiti, and Kenae in the Marquesas. The Samoan and -Tahitian name recalls the Burmese name of Ka-thit, whilst the Marquesan -word is suggestive of the Makassar name Kăne or Kanur. The Hawaiian name -of E. monosperma is Wili-wili, which evidently has arisen from the -screw-like movement of the open pod when thrown into the air. The same -name in the form of Wiri-wiri is applied for a similar reason to -Gyrocarpus Jacquini in Fiji. It is possible that the Polynesians have -assisted the dispersal of the coast-species (E. indica); but the -currents could have performed the distribution unaided, and the variety -of aboriginal names is not in favour of human intervention. - -With reference to the possible extermination by insects of Erythrina in -Hawaii, it has been before remarked (p. 143) that this would not account -for the survival of an inland species, such as E. monosperma in Hawaii. -However, this species since the occupation of that group by the white -man is on the road to extinction. Dr. Hillebrand observes that the -species was much more common formerly than in his time (1851-1871), a -result evidently due to the ravages of the common tropical mealy bug, a -pest of relatively modern introduction (see Koebele in Stubb’s -_Agricultural Report on Hawaii_). It may be added here that Cordia -subcordata, a littoral tree, had been almost exterminated by the ravages -of a small moth even in Dr. Hillebrand’s time. During my examination of -the coasts of the large island of Hawaii, in 1896-7, I was shown several -places not long before occupied by this tree; and, as indicated in Note -29, it only came under my notice in a few localities. - - - NOTE 54 (page 145) - - ON THE GENUS CANAVALIA - -Of the three maritime species, C. obtusifolia, D.C., occurs on beaches -all round the tropical zone. I was familiar with it on North Keeling -Island in the Indian Ocean, in Fiji, and in Ecuador. C. ensiformis, -D.C., is just as widely spread; but it is both inland and maritime in -its station, and except when collecting it in the Solomon Islands I have -had but little acquaintance with it. C. sericea (Gray) is a -characteristic beach-plant in Fiji, but is infrequent. In Rarotonga, -according to Cheeseman, it is a common littoral plant. It was also found -in Tahiti by Banks and Solander, and is seemingly peculiar to the -Pacific islands. - -Besides C. ensiformis, the other two shore species may at times be found -inland. Thus it is singular that the French botanists do not, as a rule, -speak of C. sericea as a Tahitian beach plant; and Nadeaud only remarks, -concerning its station, that it frequents the wooded slopes of the -valleys of the interior. In North Keeling Island C. obtusifolia -presented itself to me not only as a beach-creeper, its usual habit, but -as a climber over the branches of the coast trees. In one locality in -Vanua Levu I found a variety of this species growing on a hill a mile -inland and about 700 feet above the sea. On one of the beaches it -approached C. sericea in some of its characters, as in the form of the -calyx and in the hairiness. - -Although the seeds of C. obtusifolia have long been known to be -dispersed by the currents, having been found in Moseley’s collections of -floating drift off the New Guinea coast (_Bot. Chall. Exp._, IV, 291), -they displayed remarkable fickleness when experimented on by me in Fiji. -As a rule, however, about 10 per cent. sank at once in sea-water, 50 per -cent. floated after three weeks, and 10 per cent. after twelve weeks. Of -seeds that had been kept three years, 50 per cent. floated after eleven -weeks. The seeds are to be found in numbers amongst the stranded drift -of the Fijian and Ecuador beaches, and I noticed them also afloat in the -Rewa estuary of Fiji. - -I tested the floating-power of the seeds of C. sericea in Fiji, and -found that half of them remained afloat after sixty days. On the seeds -of C. ensiformis I have not experimented; but their buoyancy is -indicated by the frequent occurrence of the plant on the Solomon Island -coral islets (Guppy’s _Solomon Islands_, pp. 290, 292, 296), and -probably the Canavalia seeds identified at Kew from my drift collections -on these islets belong to this species. Schimper (p. 166) refers to the -seeds of a Canavalia in Java that were still afloat after ten weeks. -These littoral plants are indebted for the floating capacity of the seed -to the buoyant kernel. - - - NOTE 55 (page 42 and Note 20) - - THE INLAND EXTENSION OF SCÆVOLA KŒNIGII - -Scævola sericea (Forst.), a hairy variety of this littoral plant, will -probably prove in some localities to be the inland form of the species. -Dr. Reinecke, who mentions only this variety for Samoa, says that it is -found in very moist ground in river-ravines, and no other station is -referred to. It would seem that both the glabrous and hairy forms occur -in Hawaii. Dr. Seemann speaks of the hairy variety as littoral in Fiji. - - - NOTE 56 (page 149) - - ON THE CAPACITY FOR DISPERSAL BY CURRENTS OF SOPHORA TOMENTOSA, S. - CHRYSOPHYLLA, AND S. TETRAPTERA - -(1) _Sophora tomentosa, Linn._—The moniliform pods will float for few -weeks, but it is to the seeds liberated by the breaking down of the pod -that the wide dispersal of this beach-plant by the currents is due. When -experimenting on the freshly obtained seeds in Fiji I found that -four-fifths of them floated after three months in sea-water. With seeds -that had been kept for three years, half floated after twelve months and -retained their sound condition. The seeds owe their floating power to -the buoyant kernel. - -(2) _Sophora chrysophylla, Seem._—The dry pods of this Hawaiian mountain -species float between one and two weeks in sea-water, but being brittle -they readily break down and the seeds escape. The seeds have no buoyancy -even after drying for four years. - -(3) _Sophora tetraptera, Ait._, from the coast of Chile.—After floating -from ten to fourteen days in sea-water, the dry pods become sodden and -begin to break up, the seeds escaping. Since, however, the pods tend to -decay and break open on the tree they would not be available for -dispersal by currents. Out of a number of freshly gathered seeds all -floated buoyantly after a month in sea-water, when the experiment ended; -and of seeds that had been kept over a year six out of ten floated after -four months in sea-water, two of them germinating afterwards in soil. -Like those of S. tomentosa the seeds possess buoyant kernels to which -the floating power is due. On account of the hardness of the tests the -seeds to ensure rapid germination require to be filed. - - - NOTE 57 (page 153) - - ON THE SPECIES OF OCHROSIA - -Schumann distinguishes the following species: - -(_a_) O. parviflora, Hensl., widely spread in the Pacific islands. - -(_b_) O. compta, Schumann, confined to Hawaii and corresponding to var. -B. of O. sandwicensis as given by Hillebrand. - -(_c_) O. borbonica, Spr., synonym O. oppositifolia, Lam., from Mauritius -and Madagascar to Java and Singapore. - -(_d_){O. sandwicensis, Gray, of Hawaii. } Both probably varieties of - {O. elliptica, Lab., of New Caledonia.} O. borbonica. - -(_e_) O. parviflora, Schumann, of New Guinea, probably identical with O. -mariannensis. - - - NOTE 58 (page 156) - - ON PANDANUS (from Warburg) - -(a) _The size (length) of the drupes of endemic species in oceanic -islands._—The drupes of P. reineckei of Samoa are 4-5 cm. (1-3/5 - 2 -inches). Those of P. joskei and P. thurstonii in Fiji measure -respectively 6 cm. (2-2/5 inch) and 2-1/2 cm. (1 inch). - -Out of about sixteen species in the Mascarene Islands (Mauritius, -Réunion, and Rodriquez) quite half have drupes 2 - 3-1/2 cm. -(4/5 - 1-2/5 inch) in size, whilst they run up to 8 or 10 cm. (3-4 -inches), and may be less than a centimetre (2/5 inch). - -(b) _The affinities of the Fijian and Samoan species._ - - P. odoratissimus | Wide-ranging | Section Keura. - P. joskei | Fiji | Section Lophostigma. - P. samoensis | Samoa | Section Lophostigma. - P. thurstonii | Fiji | Section Acrostigma. - P. reineckei | Samoa | Section Hombronia. - - - NOTE 59 (page 188) - - SEEDS IN PETRELS - -Darwin, in his correspondence (1859) with Sir Joseph Hooker, refers to -the occurrence of large West Indian seeds in the crops of some nestling -petrels observed by Sir William Milner at St. Kilda (_Life and Letters_, -II, 147, 148). Mr. Charles Dixon in _Ibis_ (1885) refers to Sir W. -Milner’s observation in the case of the Fulmar Petrel (Procellaria -glacialis) and speaks of them as Brazilian seeds brought by the Gulf -Stream, adding that he himself found a nut in the crop of one of these -birds in the same locality. He supposes that the birds pick them up from -the water. Mr. Hemsley very kindly wrote to Sir Joseph Hooker recently -on this point with the object of obtaining some idea of the nature of -the seeds; but after this lapse of time it has not been found possible -to satisfy my curiosity. I live in the hope of their proving to be -Cæsalpinia seeds. - - - NOTE 60 (page 202) - - SCHIMPER ON THE HALOPHILOUS CHARACTER OF LITTORAL LEGUMINOSÆ AND OF - SHORE PLANTS GENERALLY - -As a result of extensive microchemical investigations, this eminent -German botanist arrived at the conclusion that plants living on the -sea-shore, or in inland stations rich in chlorides, are able, as a rule, -to store up in their tissues a large quantity of these salts, a capacity -enabling them to live in localities where the subsoil is rich in these -materials. This inference, as shown in his experiments, is just as -applicable to the shore-plants of temperate regions, such as Aster -tripolium, Crambe maritima, and Eryngium maritimum, as it is to such -typical littoral plants of the tropics as Barringtonia speciosa, Ipomœa -pes capræ, Scævola Koenigii, Tournefortia argentea, &c. However, with -the Leguminosæ experimented upon, this capacity of storing up chlorides -was often exhibited but slightly or not at all; and characteristic -Pacific beach-plants, such as Canavalia turgida, Pongamia glabra, and -Sophora tomentosa are especially cited as examples (Schimper’s _Ind. -Mal. Strand-flora_, pp. 140-151; Wolff’s ash-analyses are here quoted). - - - NOTE 61 (page 215) - - METEOROLOGICAL OBSERVATIONS ON THE SUMMIT OF MAUNA LOA - -The summit is formed of bare rock and sand, the phanerogamic vegetation -ceasing a couple of thousand feet below. Some low plant-forms doubtless -occur under the moist, warm conditions near the steam-cracks, since -Wilkes mentions his finding a small moss; but with this exception the -surface may be described as sterile. - -_Dryness of the Air and Electrical Phenomena._—Wilkes refers to the -association of these conditions more than once in his narrative. -Whenever, as sometimes happened, the dew point could not be obtained -with Pouillet’s hygrometer, electricity was easily excited, and was -developed in large sparks. On taking off the clothes at night, sparks -would appear. As shown in the table subjoined, electrical phenomena were -noticed during the first few days of my sojourn on the summit when the -relative humidity was very low. My red blanket at night crackled in my -hands and emitted sparks, and a glowing line was produced by drawing the -finger along. Whilst the air was in this condition I observed that the -wings of dead butterflies lying on the ground stuck to my fingers -tenaciously like a needle to a magnet. The adhesiveness disappeared when -the excessive dryness gave place to humidity. The physiological effect -on me of the associated dryness and electrical state of the air was -displayed in a hot, dry, sweatless skin (cracking and chapping rapidly), -severe headache and sore-throat, general lassitude, and great -irritability. When the weather changed and the air became humid, these -unpleasant symptoms quickly disappeared. - -As a result of these dry conditions on the summit of Mauna Loa, -decomposition does not occur. I found in one place on the top, on the -site of an old camp, the remains of a quarter of beef, the meat fresh -but dried up. From a water-bottle left behind by one of the party and -subsequently restored to him, I learned that the visit had been made in -the previous summer. This non-decomposition seems a little strange, -since, as remarked below, flies and other insects were not infrequent on -the summit. However, as Hann remarks, when speaking of mountain -climates, everything dries much more quickly at great altitudes; animals -that have been shot, or killed by falling, become mummies without -undergoing decay (Schimper’s _Plant-Geography_, 697).... The scorching -power of the sun in a sky usually cloudless, or nearly so, was a trying -feature of my daily experiences; and I found that when I faced it with -unshaded eyes during my walks I suffered from severe pain in the -eyeballs at night. - -_Insects on the Summit._—It may seem a strange thing to relate, that in -a region apparently absolutely sterile, the flies and other winged -insects caused me much discomfort in my small tent when I was confined -to it through illness. When lying down one morning I noticed the -house-fly, the blue-bottle, and two or three other flies, small beetles -not over a fifth of an inch in size, a moth, and a wasp. They were no -doubt quite happy in the heat, as the temperature inside was over 80° -F., and the sun’s rays felt almost scorching through the thin duck -canvas. Butterflies (and occasionally large moths) were often observed -flying in a drowsy condition about the summit and were easily caught. -They were fond of fluttering around the steam-holes. In places, numbers -were to be seen dead and dried up on the ground, the detached wings -lying about. In the case of a recently dead butterfly I found its -carcase already attacked by numerous small bugs. The butterflies were -most frequent when there was a fresh southerly breeze, and were -doubtless blown up the slopes from the forests below. - -Whymper in his _Travels amongst the Great Andes of the Equator_ gives -many particulars of the occurrence of insects at great elevations. He -noticed beetles, diptera, butterflies, moths, and several other insects -at altitudes of 15,000 to 16,000 feet. At 16,500 feet he obtained a -small bug of the genus Emesa. He quotes Humboldt and Bonpland as showing -that insects are transported into the upper regions of the atmosphere -16,000 to 19,000 feet above the sea, and he remarks that the -transportation of insects by ascending currents of air has occasionally -been observed in operation. These facts bear directly on the dispersal -of insects. - -_The Winds._—My tent, which was pitched near the middle of the western -border of the crater, happened to be situated in the battle-ground of -the northerly and southerly winds, in a region of gusty winds, fitful -airs, and dead calms. The northerly winds were usually from N.-N.N.W. -and the southerly winds from S.W.-S.S.W., easting in either case being -rarely observed, the northerly winds rather prevailing at night. As a -result of this location miniature whirlwinds were frequent in the -vicinity of my tent, which carried sand into the air and more than once -threatened to lift up my tent bodily and carry it off into the crater -below. At the north end of the crater-border north-easterly winds -prevailed, and at the south end southerly winds occasionally showing -easting. When on one occasion I walked round the crater-margin, a fresh -south-easterly wind prevailed at most parts of the circumference except -in the vicinity of my camp, where there was a light S.S.W. wind both at -8 a.m. and 6 p.m. when I started and returned. The local character of -the winds was often displayed in my walks. On one occasion, having left -my camp, where a southerly wind was blowing, and walked half a mile to -the north, I found a bitterly cold N.N.E. gale in my face which so -impeded my progress that I returned to my camp where the same southerly -breeze continued. - -Commodore Wilkes was encamped on the east side of the crater, and there -(December and January) he experienced strong south-west winds, on at -least three days having the force of a gale. These are the prevailing -winds in this season over the group; whereas in August, the time of my -sojourn, south-westerly winds are quite out of season, this being in the -midst of the period of the N.E. trades. - -It will be gathered from the foregoing remarks that the mere record of -the winds is insufficient for the purpose of obtaining any definite -notion of the air-currents at this elevation (13,600 feet). It is to -close observation of the clouds that we must look for data of -importance. - -_The Clouds._—The clouds on the summit of Mauna Loa were an unending -source of interest to me, and I will give briefly the results of my -observations. The highest clouds were wispy cirri, often arranged as in -a mackerel sky, and evidently at a great altitude. They were only -observed on four or five days. (The lower clouds are indicated in the -accompanying diagram.) Below them and at no great height above the -mountain were to be not infrequently observed isolated woolly clouds -that were carried in a few minutes across the sky and had a brief -existence, often forming and melting away as one gazed at them. Next, -there was a heavy bank of cumulus, which formed on the south-west slope -near the top of the mountain, from which lines of cloud extended along -each flank. Lowest of all was a broad belt, or rather a sea, of cumulus -that was developed on both sides of the mountain about one-third way -down its slopes, and during the day-time isolated the peak from the -world below. It is with the last two cloud formations that we are most -concerned, and I will first describe the sea of cumulus. - -The sea of cumulus, as in the case of similar cloud-formations of most -other isolated mountains, when viewed from above, as from the -mountain-top, presents a cloud-field of dazzling whiteness, sparkling in -the sun. Seen from below, as from the coast, it has the dark lowering -appearance of the rain-cloud and indicates the rain-belt. Disappearing -during the night, this broad belt begins to form again between 8 and 9 -a.m., and by 10 or 11 a.m. the lower regions are completely hidden and -the mountain’s summit, cut off from the world, rises above the level of -the sea of clouds like an island in an Arctic ocean. As the day -progresses the clouds become more compact and dense. The usual altitude -of this broad belt of cloud is between 7,000 and 8,000 feet. This level -is indicated by the burying of the Kohala mountains, which rise to a -height of 5,500 feet in the distant north-west corner of the island, and -by the usual emergence of the highest summit of Hualalai, which rises, -still nearer, to an elevation of 8,275 feet. On some days, however, it -attains a height of nearly 9,000 feet. On such occasions the highest -peak of Hualalai kept reappearing and disappearing during the day, but -the distant summit of Haleakala in East Maui, 10,032 feet in elevation -and 80 miles away, was always visible. - -Words fail to describe the magnificent aspect of this sea of cloud which -shuts off the spectator from the world below. From the summit of the -mountain he gazes down on its surface lit up by a sun shining in a -typically cloudless sky. At one time it appears as an undulating Arctic -land covered with snow of dazzling whiteness. At another time it looks -like a hummocky frozen Polar sea sparkling in the sunshine. Through -occasional rifts, however, one can discern a dark dismal region of mist -and rain-cloud beneath. Miss Bird, who passed a night on the summit in -June, 1874, well describes this sea of cloud in her book on the Sandwich -Islands as “all radiance above and drizzling fog below.” - - [_To face page 585._ - -[Illustration: Diagram illustrating the prevailing cloud-formations of -Mauna Loa during August, 1897.] - -The heavy bank of cumulus, that forms at noon on the south-west slope at -an altitude of 10,000 to 13,000 feet above the sea, and sometimes rises -above the mountain, is one of the most conspicuous of the -cloud-phenomena on the summit of Mauna Loa. Apparently extending from -it, but in reality moving towards it, are two lines of small cumuli that -follow the same level along either flank above the sea of cumulus, as is -indicated in the accompanying diagram. It was observed by Wilkes in -mid-winter, 1840-41, but at a lower level. “Clouds would approach us (he -writes) from the south-west when we had a strong north-east trade wind -blowing, coming up with their cumulus front reaching the height of about -8,000 feet, spreading horizontally and then disappearing.” During my -sojourn this bank formed a very striking feature in the landscape during -the early afternoon. On two or three occasions when I visited the south -side of the summit and descended for about a thousand feet I passed -through this bank, being then exposed to a driving mist coming up the -slopes from the south-west. Though its upper surface viewed from a -distance is dazzling white, below it is dark and nimboid. - -It is to an updraught of warm moist air on the south or south-west -slopes of the mountain, and to the prevailing cool north-east trade that -strikes the north side of the summit, that we must look for the -explanation of the development and situation of this bank. Although the -trade-wind is markedly stronger than the south-west updraught, some of -the warm, moist, southerly air-currents find their way, as shown by the -observations at my camp, along the sides of the summit, and a line of -condensation is produced where they come into contact with the cool air -of the north-east trade as it sweeps past the flanks of the mountain. -Sometimes at my camp, when there was a light southerly breeze blowing, I -have noticed the line of small cumuli moving south along the mountain -side towards the bank of cumulus.... I may remark that on a few days a -small bank of cumulus formed under similar conditions on the north-west -side of the summit. - -From my study of the clouds I arrived at the conclusion that there were -three prevailing air-currents on the summit of Mauna Loa: - -(1) The updraught of warm moist air on the south and south-west slopes -of the mountain. - -(2) The north-east trade wind, the upper limit of this air-current being -probably not far above the summit. - -(3) An upper air-current from the south-east (E.S.E.-S.S.E.), which, -from the velocity of the clouds it carried, was often probably not over -a couple of thousand feet above the summit. It may be observed that on -the coast at the base of the southern slope of the mountain in the -middle of September, when the wind was N.E. and carried the lower clouds -with it, the upper clouds were, on several occasions, noticed travelling -in the opposite direction, namely, from the south. - -The volcano was quiescent during my visit and could have exercised but -little influence on the air-currents. - -_The Shadow of the Mountain._—Every morning and evening, in clear -weather, for about twenty minutes after sunrise and before sunset, the -shadow of the mountain was thrown back against the sky of the opposite -horizon. It seemed as if some Titanic brush, at work in the sky far -away, had painted in the profile of the mountain with a very uncanny -blue. At sunset the peak was the last to disappear. Commodore Wilkes, -who only records it once, namely, at sunset on the 1st of January, -describes it as “a beautiful appearance of the shadow of the mountain -projected on the eastern sky ... as distinct as possible, its vast dome -seemed to rest on the distant horizon.” This phenomenon is, of course, -well known in the case of other isolated mountains. According to -Murray’s _Handbook of Southern Italy_ (1892), the correct thing for a -visitor to Stromboli is to make an early ascent of the cone to observe -“the very curious triangular shadow of the mountain cast by the rising -sun upon the _sea_.” Unfortunately I neglected my opportunity when on -the island. The shadow of the mountain is also one of the sights of -Etna, a dark-violet, triangular shadow (Baedeker) being thrown at -sunrise over the surface of West Sicily, that is, on the _land_. I saw -the shadow but imperfectly outlined, as the weather was not favourable -at the time of my ascent. When at Nicolosi, on the south slope of Etna, -I noticed at sunset a faint shadow of the mountain thrown against the -eastern sky. I gathered from a short conversation with Prof. Ricco, the -director of the Catania Observatory, when I told him of the shadow of -the Hawaiian mountain, that the interest lay in its projection against -the sky. It is doubtless akin to the spectre of the Brocken and other -mountain spectres. - -_Some Previous Meteorological Observations on Mauna Loa._—.... Mr. -Douglas, the botanist, who was subsequently found dead in a cattle-pit -on Mauna Kea, spent a day on the summit of Mauna Loa in the middle of -January, 1834. He mentions that a little way below the top the -thermometer fell at night to 19° F. The wind on the top was N.W. The air -at 11.20 a.m. was 33°, the hygrometer registering 0·5. He remarks that -the great dryness of the air was evident without the assistance of the -hygrometer (_Hawaiian Spectator_, vols. I and II, 1838-9). - -Commodore Wilkes, in vol. IV of his _Narrative of the United States -Exploring Expedition_, gives the following observations on the -temperature and winds on the top of Mauna Loa between Dec. 23, 1840, and -Jan. 13, 1841. Those on the temperature are incomplete, but they give a -fair idea of the prevailing conditions. The degrees are in Fahrenheit’s -scale. - - Dec. 23, 1840: elevation, 13,190 feet; 3 p.m., 25° F.; strong S.W. - gale; night temperature, 15°. - Dec. 24, 1840: summit (13,600 feet); night minimum, 22°. - Dec. 26, 1840: summit (13,600 feet); violent S.W. gale; night min., - 17°. - Dec. 27, 1840: summit (13,600 feet); sunrise temp., 20°; night min., - 17°; wind, S.W. - Dec. 29, 1840: summit (13,600 feet); noon temp. in shade, 47°; night - min., 20°. - Dec. 30, 1840: summit (13,600 feet); noon temp., 55°; night min., 13°. - Dec. 31, 1840: summit (13,600 feet); night min., 17°. - Jan. 2, 1841: summit (13,600 feet); sunrise, 20°; wind, N.E. - Jan. 3, 1841: summit (13,600 feet); night min., 17°. - Jan. 4, 1841: summit (13,600 feet); daylight, 20°. - Jan. 8, 1841: summit (13,600 feet); S.W. gale. - Jan. 10, 1841: summit (13,600 feet); night temp., 16°. - Jan. 12, 1841: summit (13,600 feet); night temp., 17°. - Jan. 13, 1841: summit (13,600 feet); strong S.W. wind. - -The usual variation of temperature in the twenty-four hours is given as -17°-50°. The south-west was evidently regarded as the prevailing wind, -and the clouds are spoken of as sometimes moving from opposite -directions towards the same centre. - -When Miss Bird spent a night on the summit of Mauna Loa during the -eruption of June, 1874, the cold was described as intense, eleven -degrees of frost (21° F.). - -_Observations on the Summit of Mauna Kea._—.... When Prof. Alexander -with a party of scientists ascended this mountain (in the summer of -1892), the thermometer at night fell to 13° F., and the trade-wind was -found to be blowing as strongly on the summit as down below (Whitney’s -_Tourist Guide to Hawaii_). It is to be inferred that the party camped -by the small lake which is a few hundred feet below the actual summit -(13,800 feet). This lake, which I visited on May 20, 1897, is about 120 -yards across, and evidently shallow, probably not more than three or -four fathoms deep. A carpet of algæ covered the bottom. At noon, by the -lake, the air in the shade was 53° F., whilst the temperature of the -surface-water was 51°. The lower clouds were moving from S.S.E. This -lake is said to be permanently frozen over in the winter, and to have -been visited by skaters. - -_Permanent Water Supply on the Summit of Mauna Loa._—In this barren -rocky region water derived from the winter-snow is to be found all the -year through at the bottom of the deep cracks or fissures in the -lava-rock. Such fissures are from two to four feet wide, and in the case -of that near my tent the bucket had to be lowered to a depth of -seventeen or eighteen feet to reach the water, or rather the ice, since -it was often necessary to break the surface ice. In these deep, narrow -fissures, which the sun scarcely penetrates, the water would probably be -frozen over all through the seasons; but in those of less depth it would -remain liquid in summer. - -REGISTER OF OBSERVATIONS ON WIND, RELATIVE HUMIDITY, CLOUD, RAIN, AND - TEMPERATURE, MADE BY H. B. GUPPY ON THE SUMMIT OF MAUNA LOA AT AN - ELEVATION OF 13,500 FEET ABOVE THE SEA, AUGUST 9TH TO 31ST, 1897. - (CAMP ABOUT MIDDLE OF WEST SIDE OF CRATER MARGIN) - - +-----+-------+----------+----------+----------+----------+----------+----------+------------------+--------------+ - | | | | | | | | | Air in shade. | | - | | | | | | | | | | | - |Date.|Obser- | 12-4 | 4-8 | 8-12 | 12-4 | 4-8 | 8-12 |------------------+ Remarks. | - | |vation.| A.M. | A.M. | A.M. | P.M. | P.M. | P.M. | | | | | - | | | | | | | | | Min.| Max.|Range.| | - +-----+-------+----------+----------+----------+----------+----------+----------+-----+-----+------+--------------+ - |9 |Wind |S.S.W- |Variable | ... |W.S.W.- |N.N.W.- |N.N.W.- | F. | F. | F. |A beautifully | - | | |S.W. 2 | | |W.N.W. 3 |N. 1 | N. 3 | | | |coloured lunar| - | |Rel. | ... | ... | ... | ... | ... | ... | 27·5| 61·2| 33·7 |halo at 1 A.M.| - | | hum. | | | | | | | | | |Electrical | - | |Cloud | 0 | 0 | 0 | 0 | 0 | 0 | | | |condition of | - | |Rain | 0 | 0 | 0 | 0 | 0 | 0 | | | |the atmosphere| - | | | | | | | | | | | |(see text). | - | | | | | | | | | | | | | - |10 |Wind | ... |N.N.W. 2 |S.S.W.- |S.S.W.- |Calm | ... | | | |Electrical | - | | | | |W.S.W. 3 |W.S.W. 2 | | | | | |condition of | - | |Rel. | ... | ... | 34 | 42·5 | 46 | ... | 33·5| 59·7| 26·2 |the | - | | hum. | | | | | | | | | |atmosphere. | - | |Cloud | ... | 0 | 1 | 2 | 0 | ... | | | |Faint lunar | - | |Rain | 0 | 0 | 0 | 0 | 0 | 0 | | | |halo at 8 P.M.| - | | | | | | | | | | | | | - |11 |Wind |Calm |W.S.W. 1 | ... | ... |S.S.W.- | ... | | | |Electrical | - | | | | | | |S.W. 1 | | | | |condition of | - | |Rel. | ... | ... | {28·5 | ... | ... | ... | | | |the | - | | hum. | | | {21 | | | |}22·5| 61·2| 38·7 |atmosphere. | - | |Cloud | 0 | 0 | 0 | 1 | 0 | ... |} | | | | - | |Rain | 0 | 0 | 0 | 0 | 0 | 0 | | | | | - | | | | | | | | | | | | | - |12 |Wind |Calm |Calm |N.N.W.- |N.N.W. 1 |N.N.W.- |N. 2-4, | | | |Earth tremors.| - | | | | |N. 2-3 | |N. 2-4 |S.W.- | | | |Total rain, | - | | | | | | | |W.S.W. | | | |10/100. At | - | | | | | | | | 3-5 | | | |sunset, wind | - | |Rel. | ... | ... | 45 | 79 | 89 | ... | 23 | 54·7| 31·7 |N.W., wet | - | | hum. | | | | | | | | | |canvas of tent| - | |Cloud | 0 | 0 | 5 | 10 | 10 | 5 | | | |froze hard. At| - | |Rain | 0 | 0 | 0 |Rain |Rain |Rain | | | |10 P.M., | - | | | | | | | | | | | |strong | - | | | | | | | | | | | |southerly | - | | | | | | | | | | | |wind, canvas | - | | | | | | | | | | | |thawed, rain | - | | | | | | | | | | | |with strong | - | | | | | | | | | | | |gusty wind | - | | | | | | | | | | | |until 4 A.M., | - | | | | | | | | | | | |when wind | - | | | | | | | | | | | |less. | - | | | | | | | | | | | | | - |13 |Wind |S.W. |S.W. 3-5 | S.W.- |S.W. 3 |Calm; | ... | | | |Earth tremors.| - | | | 4-6 | | S.S.E. | |N.W. 1 | | | | |Total rain | - | | | | | 3-4 | | | | | | |10/100. | - | |Rel. | ... | ... | 86 | 86 | 78 | ... |} | | | | - | | hum. | | | | | | |}33·6|48·7 |15·1 | | - | |Cloud | ... | 10 | 7 | 10 | 9 | ... |} | | | | - | |Rain |Rain |Rain | 0 |Rain |Rain | 0 | | | |Butterflies | - | | | | | | | | | | | |flying about | - |14 |Wind |N.N.W.- |Northerly,|N.N.W. 3 |N.N.W. 2 |N.N.W. 1 | ... | | | |in a semi- | - | | |N. 1 |3 | | | | | | | |torpid state, | - | |Rel. | ... | 61·5 | 47 | 42 | 45·5 | ... |} | | |and easily | - | | hum. | | | | | | |}32·5|52·2 |19·7 |caught with | - | |Cloud | 3 | 0 | 0 | 0-2 | 0 | ... |} | | |the hand. | - | |Rain | 0 | 0 | 0 | 0 | 0 | 0 | | | | | - | | | | | | | | | | | | | - |15 |Wind |Northerly,|W.S.W. 1, |S.W.-W. |Calm; |Calm; |N.N.E.- | | | |Wind fitful | - | | |1 |N.N.W. 1 |2 |Southerly,|N.N.W. 1 | 1-2 | | | |during day; | - | | | | | |1 | | | | | |north-westerly| - | |Rel. | ... | ... | 38 | 44·5 |{62 | ... | | | |and south- | - | | hum. | | | | |{52·5 | |}28 |}54·7| 26·7 |westerly airs | - | |Cloud | 0 | 0 | 0 | 7 | 7-0 | 0 |} | | |with calms. | - | |Rain | 0 | 0 | 0 | 0 | 0 | 0 | | | | | - |16 |Wind | Calm |N.N.W. 2 |N. 3 | ... |Southerly,| ... | | | |Carefully | - | | | | | | |1 | | | | |observed the | - | |Rel. | ... | ... | {32} | ... | 61 | ... | | | |shadow of the | - | | hum. | | | {28} | | | |}26 | 53.2| 27.2 |mountain | - | |Cloud | ... | 0 | 0 | 7 | 0 | ... |} | | |which, at | - | |Rain | 0 | 0 | 0 | 0 | 0 | 0 | | | |sunrise and | - | | | | | | | | | | | |sunset, is | - | | | | | | | | | | | |projected | - | | | | | | | | | | | |against the | - | | | | | | | | | | | |opposite | - | | | | | | | | | | | |horizon. | - | | | | | | | | | | | | | - |17 |Wind | ... |N.N.W. 1 |N.N.W. 2, |N.- |N.N.W. 1 | ... | | | |Fitful | - | | | | |Southerly,|N.N.E 3, | | | | | |northerly and | - | | | | |2 |Southerly,| | | | | |southerly | - | | | | | |2 | | | | | |winds causing | - | |Rel. | ... | ... | 32 | 32·5 | ... | ... |} | | |miniature | - | | hum. | | | | | | |}20·5|58·7 |38·2 |whirlwinds | - | |Cloud | ... | 0 | 0 | 1 | 0 | ... |} | | |that carried | - | |Rain | 0 | 0 | 0 | 0 | 0 | 0 | | | |dust and paper| - | | | | | | | | | | | |up into the | - |18 |Wind |Calm |S.W. 1 |Variable | ... |S.W. 2 | ... | | | |air. | - | |Rel. | ... | ... | 26 | ... | 47 | ... |} | | | | - | | hum. | | | | | | |}23 | 58 | 35 | — | - | |Cloud | 0 | 0 | 0 | 1 | 0 | ... |} | | | | - | |Rain | 0 | 0 | 0 | 0 | 0 | 0 | | | | | - | | | | | | | | | | | | | - |19 |Wind | ... | ... |N.N.W. 1, |Northerly,|Northerly,|Southerly,| | | |Through the | - | | | | |W.S.W. 1 |3, |2, |3 | | | |day, fitful | - | | | | | |S.S.W. 3 |Southerly,| | | | |northerly and | - | | | | | | |2 | | | | |southerly | - | |Rel. | ... | ... | 20 | 23 |{24 } | ... | | | |breezes. | - | | hum. | | | | |{35·5} | |}22 | 58·7| 36·7 | | - | |Cloud | ... | ... | 0 | 0 | 0 | 0 |} | | | | - | |Rain | 0 | 0 | 0 | 0 | 0 | 0 | | | | | - | | | | | | | | | | | | | - |20 |Wind |Southerly,|Northerly,|Northerly,|S.S.W.- |Northerly,|Southerly,| | | |Fitful | - | | |2 |1, | 1, |S.W. 4 |2, |3 | | | |northerly and | - | | | |Southerly,|Southerly,| |Southerly,| | | | |southerly | - | | | |1 |1 | |1 | | | | |airs, often | - | |Rel. | ... | ... | 26·5 | 25 | 29·5 | ... |} | | |reversing | - | | hum. | | | | | | |}22 |57·2 |35·2 |several times | - | |Cloud | 0 | 0 | 0 | 0 | 0 | 0 |} | | |in a few | - | |Rain | 0 | 0 | 0 | 0 | 0 | 0 | | | |minutes. | - | | | | | | | | | | | | | - |21 |Wind |Southerly,|Southerly,|S.S.W. 3-4|S.S.W. 4-5|S.W. 4-5 |S.W. 4 | | | |At camp, | - | | |4 |3-4 | | | | | | | |strong | - | |Rel. | ... | ... | 35·5 | 31 | 47 | ... |} | | |southerly | - | | hum. | | | | | | |}26·5|53·7 | 27·2 |winds all | - | |Cloud | ... | 0 | 0 | 0 | 0 | ... |} | | |day. At 7 | - | |Rain | 0 | 0 | 0 | 0 | 0 | 0 | | | |A.M., walked | - | | | | | | | | | | | |half-mile | - | | | | | | | | | | | |north and | - | | | | | | | | | | | |found a | - | | | | | | | | | | | |bitterly cold | - | | | | | | | | | | | |N.N.E. gale | - | | | | | | | | | | | |blowing there,| - | | | | | | | | | | | |which forced | - | | | | | | | | | | | |me to return | - | | | | | | | | | | | |to camp where | - | | | | | | | | | | | |the south wind| - | | | | | | | | | | | |still blew | - | | | | | | | | | | | |freshly. | - | | | | | | | | | | | | | - |22 |Wind | ... |S.S.W. 1 | ... | ... |Calm, | ... | | | |Walked round | - | | | | | | |Southerly,| | | | |the crater | - | | | | | | |2 | | | | |from 8 A.M. to| - | |Rel. | ... | ... | 58 | ... | ... | ... |} | | |6 P.M. | - | | hum. | | | | | | |}20·5|46·7 |26·2 | | - | |Cloud | ... | 0 | 0 | 1 | 0 | ... |} | | | | - | |Rain | 0 | 0 | 0 | 0 | 0 | 0 | | | | | - | | | | | | | | | | | | | - |23 |Wind | ... |S.S.W. 2 | ... |N.N.W.- |Southerly,|Calm | | | |A few drops of| - | | | | | |N.E. 1-3, |1 | | | | |rain at 2 P.M.| - | | | | | |Southerly,| | | | | | | - | | | | | |3 | | | | | | | - | |Rel. | ... | ... | {60 | 52·5} | 70 | ... | | | | | - | | hum. | | | {54·5 | 64 } | | |}24 | 50·7| 26·7 | | - | |Cloud | ... | 0 | 3 | 3-8 | 3-6 | 0 |} | | | | - | |Rain | 0 | 0 | 0 |Rain | 0 | 0 | | | | | - |24 |Wind | ... |Northerly,|S.W. 1 |N.N.W.- |N.N.W.- |Northerly | | | | | - | | | |3 | |N. 2-3 |N. b. E. |2 | | | | | - | | | | | | |2-3 | | | | | | - | |Rel. | ... | ... | 54·5 | ... | 53 | ... |} | | | | - | | hum. | | | | | | |}20 | 52·7| 32·7 | | - | |Cloud | ... | 0 | 0-1 | 4 | 0 | ... |} | | | | - | |Rain | 0 | 0 | 0 | 0 | 0 | 0 | | | | | - | | | | | | | | | | | | | - |25 |Wind |Calm | ... |Calms |Calms |Northerly |Northerly | | | |A few drops of| - | | | | |with N.W. |with N.W. |and |and | | | |rain at 3 P.M.| - | | | | |and S.W. |and S.W. |Southerly |Southerly | | | | | - | | | | | airs | airs |airs |airs | | | | | - | |Rel. | ... | ... | {42·5 | 73·5} | 61 | ... |}17 | 52·2| 35·2 | | - | | hum. | | | {37·5 | 61·5} | | |} | | | | - | |Cloud | 0 | 0 | 0-1 | 2-9 | 0 | 0 | | | | | - | |Rain | 0 | 0 | 0 |Rain | 0 | 0 | | | | | - | | | | | | | | | | | | | - |26 |Wind |Northerly,| N.N.W. 1 | ... |N.N.W.- |N.N.W. 2 | ... | | | |Descended | - | | |2 | | |N. 3 | | | | | |through the | - | |Rel. | ... | ... | 49·5 | ... | 64 | ... |} | | |bank of | - | | hum. | | | | | | |}19·5| 53·7| 34·2 |cumulus on | - | |Cloud | 0 | ... | 3 | 3 | 0 | 0 |} | | |S.W. slope and| - | |Rain | 0 | 0 | 0 | 0 | 0 | 0 | | | |found driving | - | | | | | | | | | | | |mist coming up| - | | | | | | | | | | | |the slope from| - | | | | | | | | | | | |S.W. | - | | | | | | | | | | | | | - |27 |Wind |Calms |Calms |S.W.- | W.S.W.- |Southerly |Calm | | | |Rain not | - | | |with light|with light| W.S.W. 2 | W. 1 | | | | | |measurable. | - | | |airs |airs | | | | | | | |Rain-clouds | - | |Rel. | ... | ... | 35 | 73·5 | ... | ... |} | | |poured into | - | | hum. | | | | | | |}18·5| 50·7| 32·2 |and filled the| - | |Cloud | 0 | 0 | 4 | 10 | 4 | 0 |} | | |huge crater. | - | |Rain | 0 | 0 | 0 |Rain | 0 | 0 | | | | | - | | | | | | | | | | | | | - |28 |Wind |Calms |Calms |N.N.W.- |S.S.W. 3 |Northerly,|Calms | | | |9 A.M., high | - | | |with |with |N. 2-3 | |3 |with | | | |stationary | - | | |northerly |variable | | | |variable | | | |cirrus; at | - | | |airs |airs | | | |airs | | | |noon, solar | - | |Rel. | ... | ... | {46·5 | 64·5} | | | | | |halo; in | - | | hum. | | | {59·0 | 73·0} | ... | ... |}15 | 49·7| 34·7 |afternoon, | - | |Cloud | 0 | 0 | 5 | 7 | 0 | 0 |} | | |nimbus partly | - | |Rain | 0 | 0 | 0 |Rain | 0 | 0 | | | |filling crater| - | | | | | | | | | | | |causing a | - | | | | | | | | | | | |rainbow there;| - | | | | | | | | | | | |a few drops of| - | | | | | | | | | | | |rain at 4 P.M.| - | | | | | | | | | | | | | - |29 |Wind |N.N.W.- |N.N.W.- |N.- |N.N.W. 3, |Calms with|Northerly,| | | | | - | | |N. 3 |N. 2 |N. b. E. 3|S.S.W. 2 |variable |1 | | | | | - | | | | | | |airs | | | | | | - | |Rel. | ... | ... | 45 | 44 | 60·5 | ... |} | | | | - | |hum. | | | | | | |}21·5| 48·7| 27·2 | | - | |Cloud | 0 | 0 | 0 | 0 | 0 | 0 |} | | | | - | |Rain | 0 | 0 | 0 | 0 | 0 | 0 | | | | | - | | | | | | | | | | | | | - |30 |Wind |Calms with|Southerly,|N.N.W. |N.N.W. 3 |Calms with| ... | | | | | - | | |northerly |1 |2-3 | |variable | | | | | | - | | |airs | | | |airs | | | | | | - | |Rel. | ... | ... | 32·5 |{41·0} | ... | ... | | | | | - | | hum. | | | |{54·5} | | |}18 | 50·7| 32·7 | | - | |Cloud | 0 | 0 | 0 | 0 | 0 | 0 |} | | | | - | |Rain | 0 | 0 | 0 | 0 | 0 | 0 | | | | | - | | | | | | | | | | | | | - |31 | ... | ... | ... | ... | ... | ... | ... | 18·5| 50·0| 31·5 | | - +-----+-------+----------+----------+----------+----------+----------+----------+-----+-----+------+--------------+ - -_Method of Observation employed by the Author on the Summit of Mauna -Loa._—My camp was placed near the middle of the west margin of the -crater about 13,500 feet above the sea. The instruments employed were a -Sixe’s maximum and minimum thermometer made by Negretti and Zambra, -several unmounted thermometers, and a reference thermometer (with a Kew -certificate) by the above-named makers, which was used as a standard. -The freezing point was also tested for all the instruments on the summit -in melting powdered ice. The maximum air observations and those on the -relative humidity were taken in a small cave with a hole in the roof, -through which there was a steady flow of air. One day was occupied in -comparing the cave-observations with those obtained under a temporary -screen rigged up outside my tent, the only difference shown being as a -rule less than a degree. The minimum observations taken in my tent, -where there was no artificial heat, were usually only 1·5° higher than -those given by a thermometer outside the tent. - - _Results of the Observations on the Top of Mauna Loa, Aug. 9-31, 1897_ - - Mean minimum temperature of air in shade 23·2° F. - Mean maximum temperature of air in shade 53·8° - Mean daily range of temperature 30·6° - Lowest reading 15·0° - Highest reading 61·2° - Mean temperature for the period 38·5° - - } Many observations - Mean relative humidity, 8-9 a.m., 44·5 % } included which - Mean relative humidity, noon ... 43 % } are not given in - Mean relative humidity, 5-6 p.m., 56 % } the register. - -On Aug. 11th, at 10 a.m., wet bulb, 33·2°; dry bulb, 52°; difference, -18·8°. - -On Aug. 19th, at 11 a.m., wet bulb, 35·7°; dry bulb, 56°; difference, -20·3°. - -Owing to the varying winds at my camp, the relative humidity fluctuated -greatly in a short time. Thus, on Aug. 12 it was 46% at noon, and 79% at -2 p.m. - - _Average Cloudiness (10 indicating a Sky completely Overcast)_ - - 12-4 A.M. | 0 | Cloudless during 12 out of 13 days - 4-8 A.M. | 0 | Cloudless during 19 out of 20 days - 8-12 A.M. | 1·3 | Cloudless during 13 out of 22 days - 12-4 P.M. | 3·5 | Cloudless during 6 out of 22 days - 4-8 P.M. | 1·5 | Cloudless during 17 out of 22 days - 08-12 P.M. | 0 | Cloudless during 11 out of 12 days - -The winds at the camp were extremely variable and local from north and -south, usually light, with force 1-3: see under Winds and Clouds in the -text. - -Rain fell on six days, total 30/100 of an inch: but on four of the days -it was not measurable. - - - NOTE 62 (page 222) - - ON THE RELATIVE PROPORTION OF VASCULAR CRYPTOGAMS IN FIJI - -According to Seemann’s work, where about 617 indigenous flowering plants -and about 195 ferns and lycopods are enumerated, the vascular cryptogams -would form about 24 per cent. of the whole flora. (All weeds and -cultivated plants are here excluded.) The vascular cryptogams, however, -seem to figure too prominently in Seemann’s collections. From Horne’s -data, who says that he added 363 flowering plants to the flora, the -flowering plants would amount to about 980; and since Baker implies, in -_Trimen’s Journal of Botany_, 1879, that Horne added 42 species of ferns -and lycopods to the flora, this would increase the vascular cryptogams -to 237, which enables us to estimate the relative proportion of vascular -cryptogams in Fiji as about 20 per cent. of the whole flora of vascular -plants. This is probably near the truth. - - - NOTE 63 (page 222) - - ON THE TABLE OF VASCULAR CRYPTOGAMS OF TAHITI, HAWAII, AND FIJI - -In the case of Tahiti, I have gone carefully through the list given by -Drake del Castillo, comparing it with other Polynesian lists given by -Seemann, Horne, Hillebrand, Hemsley, &c., and have reduced his endemic -species from 19 to 13. The same thing has been done with Hillebrand’s -list for Hawaii, some of his species having been found in other parts of -Polynesia, thus reducing the endemic species from 75 to 70. The data -relating to Fiji are referred to in Note 62. - - - NOTE 64 (page 223) - - ON THE DISTRIBUTION OF THE TAHITIAN FERNS AND LYCOPODS - -I have arranged them as follows, according to the distributions given by -Drake del Castillo:—Cosmopolitan, 5; Tropics of Old and New Worlds, 33; -Tropics of Old World, mainly Indo-Malaya, 58; “Océanie,” including -Australia, 17; Polynesia, 26; South America, 2; peculiar to Tahiti, 13: -total, 154. - -Out of 141 non-endemic Tahitian species, 107 at least have been recorded -from the Fijian area comprising Samoa and Tonga, and 42 from Hawaii. Of -the last, all but four occur also in Fiji. There is thus a very small -element peculiar to Hawaii and Tahiti alone. Some of them will no doubt -be found in the Fijian area; whilst two of them, Acrostichum squamosum -and Lycopodium venustulum, are high-mountain forms in Hawaii and Tahiti, -which have evidently failed to find a suitable elevation in Fiji. - - - NOTE 65 (page 225) - -DISTRIBUTION OF SOME OF THE MOUNTAIN FERNS OF HAWAII THAT ARE NOT FOUND - EITHER IN FIJI OR TAHITI (mainly from Hillebrand) - - +-----------------------------+----------------+--------------------------------+ - | Species. | Altitude of | General distribution. | - | |station in feet.| | - +-----------------------------+----------------+--------------------------------+ - |Schizæa robusta, Bak. | 3,000- 6,000 |Perhaps a form of S. | - | | |australis, Gaud., from the | - | | |Falkland and Auckland Islands. | - | | | | - |Polypodium serrulatum, Mett. | 3,000- 6,000 |Generally diffused in the | - | | |tropics and subtropics. | - | | | | - |Aspidium caryotideum, Wall. | In the forests |Himalayas, South Africa, &c. | - | | | | - |Aspidium filix mas, Sw. |In the highlands|Over four continents, from the | - | | |arctic circle to the higher | - | | |levels of tropical mountains. | - | | | | - |Asplenium trichomanes, L. | 5,000- 8,000 |Temperate zones and the higher | - | | |levels of tropical mountains. | - | | | | - |Asplenium monanthemum, L. | 3,000- 6,000 |American Andes, Madeira, Tristan| - | | |d’Acunha, Azores, Abyssinia, &c.| - | | | | - |Asplenium fragile, Presl. | 4,000- 6,000 |Andes. | - | | | | - |Asplenium contiguum, K. | 2,000- 5,000 |Lord Howe Island, Ceylon, | - | | |Neilgherry Hills. | - | | | | - |Asplenium adiantum nigrum, L.| 4,000-10,000 |Europe, Asia, Africa, | - | | |Atlantic Islands. | - | | | | - |Asplenium aspidioides, Sch. | 1,000- 6,000 |Andes, Africa, India. | - +-----------------------------+----------------+--------------------------------+ - - - NOTE 66 (page 226) - - ENDEMIC GENERA OF FERNS IN HAWAII - -Hillebrand gives two genera of ferns peculiar to Hawaii, one, Sadleria -of Kaulfuss, “scarcely distinct from Blechnum,” and containing four -species; the other, Schizostege, constituted by himself, and represented -by a single species found in only one or two of the islands. - - - NOTE 67 (page 241) - - ON THE DISPERSAL OF COMPOSITÆ BY BIRDS - -The goldfinch’s habit of pecking at the heads of thistles, and pulling -out the achenes in bundles, is well known. Gätke mentions two suggestive -instances of birds feeding on the fruits of a Composite plant. According -to this observer, the Scarlet Grosbeak (Pyrrhula erythrina), when it -alights on Heligoland, always feeds on the achenes of Sonchus oleraceus, -which it picks off the plant; whilst the Parrot Crossbill (Loxia sp.), -feeds in Heligoland on burrs and thistles (_Heligoland as an -Ornithological Observatory_, pp. 407, 409). See Note 91. - - - NOTE 68 (page 264) - -ON SOME OF THE HAWAIIAN ENDEMIC GENERA, EXCLUDING THOSE OF THE COMPOSITÆ - AND LOBELIACEÆ - -_Haplostachys, Phyllostegia, and Stenogyne, all Labiate -Genera._—Phyllostegia is not strictly peculiar to Hawaii, since out of -the 17 species enumerated in the _Index Kewensis_, 15 are Hawaiian, 1 -Tahitian, and 1 is accredited to Unalaska (one of the Aleutian Islands). -The last locality appears to be an error. The species in question is P. -microphylla, Benth.; and on looking up the original authority in -_Linnæa_ (vi. 570, 1831), I find the locality is thus given: “insula -coralligena Romanzoffii,” which is either one of the atolls of the -Paumotu Islands in about lat. 15° S. and long. 144° W., or a coral -island of the Marshall Group, most probably the former.... I paid some -attention to the suitability of the fruits of these three Labiate genera -for dispersal by frugivorous birds, for which the fleshy nucules in the -cases of Phyllostegia and Stenogyne apparently fit them. Out of the -fruits of five species of Phyllostegia examined by me, the -seed-coverings in three species, after the removal of the fleshy -covering of the nucule, were too soft for the protection of the seed in -a bird’s stomach. Hillebrand also observes (p. 347) that the nucules -when dried are wrinkled, and absorb moisture easily, a quality which, if -true of all the species, would make the distribution of the genus by -birds impossible. However, in two species I found the seed-coverings -somewhat harder. It would seem that since birds have largely ceased to -disperse these plants, the soft-skinned nucules would in the absence of -their selective agency more frequently characterise the genus. It is -possible that the dry nucules of Haplostachys, which according to -Hillebrand are not affected by drying, represent the original condition -of those of Phyllostegia, and that the fleshy character has been -acquired in this archipelago. It will be seen in the list on page 263, -that Haplostachys is regarded by Gray as a section of Phyllostegia. The -remarks under Phyllostegia, regarding the softness of the seed-coverings -beneath the fleshy coat of the nucule, also apply to Stenogyne; and -Hillebrand, in contrasting its fleshy nucules with the dry nucules of -Haplostachys, implies that they absorb water, which, I may remark, would -render them quite unfit for dispersal by frugivorous birds. - -_Touchardia (Urticaceæ)._—According to Hillebrand, the solitary species -is by no means common in the group now. In 1897 I found it growing -abundantly some miles up the Waipio gorge, Hawaii. - -_Cheirodendron (Araliaceæ)._—C. Gaudichaudii, the well-known “Olapa” -tree, is common in the forests of all the Hawaiian Islands between 2,000 -and 5,000 feet; but I noticed it occasionally at greater elevations, as -on the south-east slopes of Mauna Kea, where it extends to 7,000 feet. -As described on page 343, the “Olapa” often grows in close contact with -the Lehua (Metrosideros polymorpha), the two trunks appearing as one. -The drupes would attract frugivorous birds and the pyrenes are well -adapted for this mode of dispersal. Mr. Perkins states that the drupes -are much sought after by the various species of Phæornis, a genus of -birds peculiar to Hawaii. - -_Deterioration of Fruits for Purposes of Dispersal._—Among fruits or -endemic genera that have evidently deteriorated in the Hawaiian Group as -far as fitness for dispersal is concerned, may be mentioned, in addition -to those of Phyllostegia and Stenogyne above noticed, those of the -Araliaceous genera, Pterotropia and Triplasandra, and the Amarantaceous -Nototrichium. The pyrenes of the first two genera on account of their -thin covering, and the seed of the last-named genus on account of its -thin testa, seem ill-fitted now for transport in a bird’s stomach, yet -we cannot doubt that their ancestors originally arrived in this fashion. -The same principle is also illustrated by some Hawaiian non-endemic -genera of later eras that possess peculiar species, such, for instance, -as in the case of Elæocarpus discussed in Chapter XXVI. - - - NOTE 69 (page 366) - - ON THE GERMINATION OF CUSCUTA - -My observations were made on the Hawaiian endemic species (C. -sandwichiana) and on a Fijian introduced species. Germination occurs -readily in fresh water, the floating seedling growing rapidly. When the -germinating seed is placed on wet soil in the shade, the seedling grows -at the rate of 3/4 inch (19 mm.) a day. The store of nutriment contained -in the swollen radicular end will support the seedling for a couple of -days, and if it has not then found a host it withers and dies. At first -lying prone the seedling then lifts its upper end into the air, and it -was almost pathetic to notice it moving round and round, endeavouring -vainly to find some object near. The seedlings make no effort to strike -into the soil, and when they are allowed to attach themselves to a plant -they ascend rapidly, growing at the upper end and dying at the lower -end. - - - NOTE 70 (pages 477, 480-1) - - ON BEACH-TEMPERATURE - -My data are rather scanty; but, judging from observations made in -Hawaii, in South America, and in the south of England, the following -scale would probably be true of typical beaches where the sand is found -relatively cool and moist at a depth of four or five inches. This -moisture seems to arise entirely from subsoil drainage seaward. When a -beach fronts an arid, rainless region, few if any plants grow on it; the -sand is loose, hot, and dry at the depth indicated; and the temperature -of the surface half-inch rises to between 130° and 140° F., whilst four -inches down it is 95° to 100°. Salt-marshes situated behind a beach even -in a desert-region change its thermal behaviour, and it would then be -more like a beach skirting a vegetated sea-border in the same latitude. -The method of observation was as follows:—An unmounted thermometer of -the size of a clinical thermometer, but graduated higher, was placed -horizontally in the sand half an inch below the surface and a reading -taken. It was then pushed vertically into the sand until the bulb was -four inches deep and another reading taken. Provided that the sand is -moist beneath, the colour does not seem to make much difference, except -perhaps in very dark sands, none of which were tested. - - _Ordinary Beach-Temperatures with an Unclouded Sky in the Hot Season - during the Early Afternoon._ - - +-----------------------------------+------------------+-----------------+ - | |Surface half-inch.|Four inches deep.| - | +------------------+-----------------+ - |Temperate latitudes about 50-55°| 100-105° F. | 77° F. | - |Sub-tropical latitudes about 30-35 | 105-110 | 80 | - |Tropical latitudes about 10-20 | 110-120 | 85 | - +-----------------------------------+------------------+-----------------+ - -This illustrates only the average condition. On a calm day in the case -of a beach facing south in the South of England, I have obtained exactly -the same readings in July as at Valparaiso in January, 112° at surface, -80° four inches deep. - - - NOTE 71 (page 479) - - On the Buoyancy of the Seeds or Seed-vessels of some Chilian Shore - Plants - -(1) _Nolana_, probably _paradoxa_. Common on the beaches of Southern -Chile. The ripe drupes have a somewhat fleshy outer covering which they -lose when lying on the sand, and present themselves then as dark-brown -angular “stones,” often five to six millimetres across. Inside the outer -hard covering of the stone is a layer of spongy tissue which gives it -buoyancy; but since these coverings are wanting at the scars marking the -basal insertion of the drupe, the embryo seems insufficiently protected -against injury during flotation in sea-water; and the seed-vessel at -first appears to be only fitted for conveyance by the currents over a -limited tract of sea. However, in a preliminary experiment on -seed-vessels that had been kept a few weeks, I found that 30 per cent. -floated after three weeks in sea-water. Subsequently, after drying for a -year, the seed-vessels were again tested in sea-water, nearly all of -them floating after three months’ immersion. Two of them, removed after -six weeks’ flotation, germinated healthily. These fruits are common in -beach-drift between Corral and Valparaiso. - -(2) _Raphanus_, near R. maritimus. Growing near beaches in South Chile, -and not infrequently represented in the stranded beach-drift by the -pods, which in my experiments floated seven to ten days in sea-water, -after drying some weeks. - -(3) _Franseria._ A species common on the beaches of Valparaiso and -Talcahuano. Its prickly fruits, after being kept six weeks, floated only -two to four days. They are well suited for transport in birds’ plumage. - - - NOTE 72 (page 483) - - THE SOUTHERN LIMIT OF THE MANGROVE FORMATION IN ECUADOR. - -... The southern limit of the mangrove formation on the west coast of -South America is usually placed at 4° S. lat.; but it is probable that -the vicinity of Tumbez in lat. 3° 30ʹ S. would be more correct. Baron -von Eggers would place it rather further to the north-east, near the -frontier of Ecuador and Peru in lat. 3° 20ʹ S. I spent eight days in the -locality last named and saw no evidence of the beginning of the -mangrove-formation. - - - NOTE 73 (page 495) - - ADDITIONAL NOTE ON THE TEMPERATURE OF THE DRY COAST OF ECUADOR BETWEEN - PUNA ISLAND AND THE EQUATOR. - -... Baron von Eggers gives the mean annual temperature for El Recreo, -about half a degree south of the equator, at 75° F., which is near that -of Rio de Janeiro in lat 23° S. on the east coast of the continent. Mr. -F. P. Walker has kindly given me the results of temperature-observations -covering a period of ten years, taken in the room for testing cables at -Santa Elena Point (2° 10ʹ S.), usually about 6·30 a.m. The range of the -monthly means was 71° F. (August) to 79·1° (March), and the mean for the -year was 74·8°. In that locality a typical daily range would be 65° to -80°; and Mr. Walker believes that a minimum of 59° has been recorded. - - - NOTE 74 (page 495) - -OBSERVATIONS ON THE TEMPERATURE OF THE HUMBOLDT CURRENT FROM ANTOFAGASTA - NORTHWARD, BETWEEN JANUARY AND MARCH, 1904 (Fahrenheit scale) - -The observations were usually taken at the anchorages, but in some -places, as at Ancon and Puerto Bolivar, they were taken from a boat -outside the roadstead. - -If we wish to ascertain how the Humboldt Current retains its cool -temperature as it advances through the tropics to the equator, a glance -at the following table will show that the surface-temperatures can aid -us but slightly, since they do not vary in accordance with the latitude, -a subject further discussed below. We can, however, obtain some valuable -indications from the deeper temperatures. Let us take for instance the -plane of 60°. Whilst south of Ancon (lat. 11° 45ʹ S.) it was rarely -deeper than four fathoms, north of this latitude it descends rapidly, -being probably about ten fathoms down at Salaverri and Eten and about -twenty fathoms deep at Payta, in latitude 5° S., where the Humboldt -Current leaves the coast. Within the Gulf of Guayaquil it is probable -that the plane of 60° would descend to nearer thirty fathoms, the region -being outside the influence of the current. - -Some interesting facts are also elicited from the variation of the -surface-temperatures. When we were coasting along at a distance of five -or six miles from shore the readings were fairly constant from hour to -hour varying only a degree or so. But nearer the land, for instance, -about two or three miles away, the variation from hour to hour amounted -to two or three degrees, whilst within the limits of the anchorages, a -mile and less from the coast, the change from hour to hour amounted to -three or four degrees. Nor was there any uniformity at the same hour -over the surface of a roadstead. The temperature would often rise or -fall a degree every few boat-lengths. Sometimes the inshore water was -the coolest and sometimes it was the warmest. Thus at Iquique the -inshore water was three degrees warmer than the water half a mile out, -whilst at Mollendo, when the temperature one-third of a mile off the -shore was 70°, it was 63° close to the rocky coast. The same thing was -exhibited at Pisagua, where the surface-water two miles out at sea was -61°, whilst close inshore at the anchorage it was 58°. It was evident -that there was a considerable intermingling of the warmer surface and -the colder, deeper waters on the coasts of Chile and Peru. This was -particularly noticeable on a rocky, steep-to coast, or where there was -an uneven bottom. At some places, indeed, the warm upper layer did not -exist, the cold water welling up all along the coast. This was -especially the case between the 22nd and 19th parallels of latitude, a -tract of coast in which lie Tocopilla, Iquique, and Pisagua, and -probably the coolest part of the sea-border at this season of the year. - -During a fortnight spent at Ancon (11° 45ʹ S.), between January 27 and -February 10, I paid considerable attention to the local climatic -conditions, and especially to the temperature of the inshore water. The -daily range of the air-temperature was only five or six degrees, the -average minimum and maximum being 71° and 75·9°, and the mean for the -period 73·5°. The mean temperature of the surface-water at the head of -the pier, from observations taken at about 7 a.m. and 4 p.m., was 68·6°, -or five degrees cooler than the air, the mean temperature in the morning -being 69·1° and in the afternoon 68°. - - OBSERVATIONS ON THE TEMPERATURE OF THE HUMBOLDT OR PERUVIAN CURRENT - - (Made by H. B. Guppy, January to March, 1904. - Those at Panama are added for the sake of comparison) - - +--------------+----------+----------+--------+--------+---------+----------------------------------------------------------------------------------------------------------------------+ - | | | Distance | | | | Depths in fathoms: temperature in Fahrenheit degrees. | - | | Depth | from | | | +--------+----+----+----+----+----+----+----+----+----+----+----+----+----+----+----+----+----+----+----+----+----+----+ - | Locality. |(fathoms).| shore | S. lat.| Date. | Hour. | | | | | | | | | | | | | | | | | | | | | | | | - | | | (miles). | | | |Surface.| 1.| 2.| 3.| 4.| 5.| 6.| 7.| 8.| 9.| 10.| 11.| 12.| 13.| 14.| 15.| 16.| 17.| 18.| 19.| 20.| 21.| 22.| - +--------------+----------+----------+--------+--------+---------+--------|----+----+----+----+----+----+----+----+----+----+----+----+----+----+----+----+----+----+----+----+----+----+ - | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | - |Antofagasta | 22 | 2/3 | 23°40ʹ |{Jan. 12| 5 p.m. | 71° | 70°| ...| ...| ...| ...| ...| 61°| ...| ...| 57°| ...| ...| ...| ...| ...| 56°| ...| ...| ...| ...| ...| ...| - | | | | |{Jan. 13| 6 a.m. | 70 | ...| ...| ...| ...| ...| ...| ...| ...| ...| ...| ...| ...| ...| ...| ...| ...| ...| ...| ...| 55 | ...| ...| - | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | - |Tocopilla | 18 | 1/2 | 22 0 | Jan. 14| 8 a.m. | 57 | ...| ...| ...| ...| ...| ...| ...| ...| ...| 56 | ...| ...| ...| ...| ...| 55 | ...| ...| ...| ...| ...| ...| - | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | - |Iquique | 9 | 2/3 | 20 15 | Jan. 15|5.30 a.m.| 59 | ...| ...| ...| ...| 58 | ...| ...| ...| 55 | ...| ...| ...| ...| ...| ...| ...| ...| ...| ...| ...| ...| ...| - | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | - |Pisagua | 10 | 1/2 | 19 30 | Jan. 16| 8 a.m. | 57·5 | ...| ...| ...| ...| ...| ...| ...| ...| ...| 56 | ...| ...| ...| ...| ...| ...| ...| ...| ...| ...| ...| ...| - | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | - |Arica | ... | 1/4 | 18 25 | Jan. 16| 7 p.m. | 66 | ...| ...| ...| ...| ...| ...| ...| 57 | ...| ...| ...| ...| ...| ...| ...| ...| ...| ...| ...| ...| ...| ...| - | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | - | | | |{17 0 | Jan. 17| 8 a.m. | 61·5 | ...| ...| ...| ...| ...| ...| ...| 57 | ...| ...| ...| ...| ...| ...| ...| ...| ...| ...| ...| ...| ...| ...| - |Mollendo | 22 | 1/3 |{17 0 | Jan. 17| 6 p.m. | 65 | ...| ...| ...| ...| ...| ...| ...| ...| ...| 59 | ...| ...| ...| ...| ...| ...| ...| ...| ...| ...| ...| ...| - | | | |{17 0 | Jan. 18|9.30 a.m.| 62 | ...| ...| ...| ...|57·5| ...| ...| ...| ...| 57 | ...| ...| ...| ...| 56 | ...| ...| ...| ...| ...| ...| 55 | - | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | - |Callao | 5 | 1 | 12 3 | Jan. 20| 6 a.m. | 60·5 | ...| ...| ...| ...|58·5| ...| ...| ...| ...| ...| ...| ...| ...| ...| ...| ...| ...| ...| ...| ...| ...| ...| - | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | - |Ancon | 17 | 1-3/4 | 11 45 | Jan. 29| 11 a.m. | 66 | ...| ...|6·15| ...|58·5| ...| ...| 57 | ...| 57 | ...| ...| ...| ...| ...| ...|56·5| ...| ...| ...| ...| ...| - | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | - |Salaverri | 6 | 3/4 | 8 15 |Febr. 21|6.30 p.m.| 65 | ...| ...| ...| ...| ...| 62 | ...| ...| ...| ...| ...| ...| ...| ...| ...| ...| ...| ...| ...| ...| ...| ...| - | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | - |Eten | 7 | 1/2 | 7 0 |Febr. 22| 6 a.m. | 66·5 | ...| ...| ...| ...| ...| ...| 63 | ...| ...| ...| ...| ...| ...| ...| ...| ...| ...| ...| ...| ...| ...| ...| - | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | - |Payta | 16 | 1-1/2 | 5 0 |Febr. 23| 9 a.m. | 70 | ...| ...| ...| ...| ...| ...|67·5| ...| ...| ...| ...| ...| ...| ...|62·5| ...| ...| ...| ...| ...| ...| ...| - | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | - |Puerto Bolivar| 10 | 1 | 3 10 | Mar. 7| noon | 78 | ...| ...| ...| ...|71·5| ...| ...| ...| ...| 70 | ...| ...| ...| ...| ...| ...| ...| ...| ...| ...| ...| ...| - | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | - |Panama | 8 | 3 |{(8 50) | Mar. 23| 4 p.m. | 82 | ...| ...| ...| ...| ...| ...|79·5| ...| ...| ...| ...| ...| ...| ...| ...| ...| ...| ...| ...| ...| ...| ...| - | | | |{( N.) | Mar. 24| 7 a.m. | 79·5 | ...| ...| ...| ...| ...| ...| ...|79·5| ...| ...| ...| ...| ...| ...| ...| ...| ...| ...| ...| ...| ...| ...| - +--------------+----------+----------+--------+--------+---------+--------+----+----+----+----+----+----+----+----+----+----+----+----+----+----+----+----+----+----+----+----+----+----+ - -The Ancon climate at this period is full of oddities and abnormalities, -and in this way typifies much of the coast of Peru. Thus, since the heat -of the day is tempered by the cool south-westerly winds which die away -in the evening and give place usually to warm, light, northerly and -north-westerly breezes, there is, as above remarked, but a small -difference between day and night temperatures. The coldest time of the -twenty-four hours is not in the early morning but at sunset. The sea off -the beach is, on the average, much cooler than the air, which is not a -normal state of things; and again, the water is often two or three -degrees colder in the evening than it is in the morning, which is very -unusual. Though the sea-border is practically a desert for the greater -part of the year and has no rain, it is frequently enveloped in -drizzling fogs or “garuas.” Judged from a European standard, things go -by contraries on the coast of Peru; and this is entirely the effect of -the Humboldt Current. - -The temperature of the inshore waters of Ancon Bay varied considerably -during the twenty-four hours. During the day, with the prevailing -southerly wind, the cool waters of the current had free access to the -bay, and swept around its border in their course north; but in the -night, when northerly breezes occurred, the cold waters of the current -were pushed off the coast and their place taken by the warmer inshore -waters from the north; and this sometimes continued for a day or two. -When the current again got mastery and its clean, cool waters filled the -bay, the temperature of the water dropped suddenly five or six degrees, -and the bay was filled with fish. At such times men in boats leave the -beach, and in a few minutes, with hand-nets and baskets, they obtain -thousands of the small fry. Other men, fishing with lines from the -pier-head, seem ill-contented unless they can catch fish of the size of -small mackerel at the rate of one a minute. - -There can be little doubt that on the coasts of Chile and Peru the -instincts of fish often lead them astray, on account of the sudden -changes of temperature arising from the conflict between the warmer -waters of the open sea and the cooler waters of the current. From the -preceding remarks it will be inferred that sometimes the current is -pushed off the coast for a while and its place taken by the warm waters -from the north. At other times it dives down, so to speak, and flows at -a deeper level, and warmer waters prevail both out at sea and inshore. -At other times again, and this must be most disconcerting to the fish, -the cold current suddenly appearing at the coast predominates at the -surface for days together, and we have stretches of coast which, -although lying within tropical latitudes, are washed by waters having -the temperature of the temperate zone. It is to such causes that we must -attribute the reckless habits of fish on these coasts. They are known to -throw themselves on the beaches in thousands, where by their decay they -taint the air long afterwards. Mr. Anderson Smith in his recent book on -_Temperate Chile_ vividly describes what goes on on such occasions at -the port of Valdivia. At times the scene must be indeed a strange one, -since huge octopi are rolled up on the beaches in numbers, and are -regarded by the indigenes as deliberately seeking their death. Whether -they commit suicide or not, “their beaks that blacken the edge of the -sea-wash in places” afford a melancholy proof that their instinct has -blundered. - -_The Mode of Observation._—A thermometer made on the Sixe pattern which -I used several years ago for taking the bottom-temperatures of rivers, -was employed for the deeper temperatures, and at critical depths the -observations were always repeated. This instrument was compared after -each set of observations with an ordinary thermometer graduated on the -stem, which was compared with my standard thermometer provided with a -Kew certificate.... The observations in the Panama Roadstead have been -added for the sake of contrast. - - - NOTE 75 (page 496) - -ON THE STRANDED MASSIVE CORALS APPARENTLY OF THE GENUS PORITES FOUND ON -THE COAST OF PERU AND NORTH CHILE, AT ARICA (18° 25ʹ S.), CALLAO (12° 3ʹ - S.), AND ANCON (11° 45ʹ S.) - -At Arica they occurred on the beach only. At Callao they also extended -inland on the low spit at Punta for about 100 yards. At Ancon they were -found not only on the beach but also twenty or thirty paces inland on -the low adjoining plains. Their size varied from three inches to three -feet. They were all more or less rounded by wave action, and were -extensively burrowed by boring molluscs. Whilst some on the beach still -displayed the dried-up soft parts of the boring mollusc, others inland -were falling to pieces and undergoing chemical change. There was nothing -to indicate that the corals were recently alive; and at Ancon they -appeared to have been torn off a rocky spit of andesite that had become -exposed on the beach during a recent movement of emergence, of which -there is other evidence on this coast. Further particulars are given on -page 496. - - - NOTE 76 (page 429) - - STRANDED PUMICE ON ENGLISH AND SCANDINAVIAN BEACHES - -Sernander, in his description of the Atlantic drift of the Scandinavian -coast, refers to the occurrence of a small amount of true pumice. I have -found solitary fragments of acid pumice well rounded by wave-action at -Croyde Bay on the north coast of Devonshire, at the mouth of Salcombe -Harbour on the south coast of the same county, and at Maenporth, near -Falmouth, in Cornwall. Steamer slag, in some cases rudely simulating -pumice, is common on all the South of England beaches I have examined. -It is also common on the Scandinavian coasts, though seemingly regarded -by Helge Bäckström, who is quoted by Sernander, as derived from the -factories on the east coast of England. (See on these subjects a paper -by Helge Bäckström, “Über angeschwemmte Bimsteine und Schlacken der -nordeuropäischen Küsten”; Bihang till _K. Sv. V. A. Handl._ Bd. 16. Afd. -3, 1890; also a letter in _Nature_, about 1886, by H. B. Guppy.) - - - NOTE 77 (page 21) - - ON THE MODE OF DISPERSAL OF KLEINHOVIA HOSPITA - -This small tree has a very wide distribution in the tropics, ranging -from East Africa and the Mascarene Islands through India, South-eastern -Asia, Malaya, New Guinea, and the Solomon Islands to Fiji and Tahiti. It -is a plant that grows in inland open woods as well as amongst the -littoral trees on the beach; and it is always doubtful (in Malaya, Fiji, -and Samoa) whether to regard it as a shore plant or as an inland plant, -different authors varying on this point. In Vanua Levu I formed the -opinion that it is only an intruder amongst the littoral vegetation. In -accounting for its distribution we have to choose between man, the bird, -and the current. Though it may sometimes be noticed in native -plantations, as I observed in the Solomon Islands, the tree has no -special use; and the Solomon Island natives themselves indicated to me -that the parrots that fed on the fruits of the tree aided in -distributing the plant. The buoyant behaviour of the seeds, which are -freed by the dehiscence of the bladder-capsules on the tree, is not -constant. Whilst in the case of the seeds of littoral trees in Fiji I -found that 30 per cent. floated after ten weeks, Prof. Schimper -ascertained in the case (seemingly) of Malayan seeds that they sank at -once. The seed-structure connected with the buoyancy is, as shown on -page 105, accidental in character, and reference is made on page 20 to -other plants of doubtful littoral reputation, in which the buoyant -qualities are variable. The occasional buoyancy of its seeds will only, -as I think, explain its occasional station at the coast; and I agree -with Prof. Schimper (p. 156) when he attributes its wide distribution to -birds, the seeds being hard, crustaceous, and about three millimetres -across. - - - NOTE 78 (page 436) - - ON THE “SEA”: AN UNIDENTIFIED WILD FRUIT-TREE IN FIJI - -This is a fair-sized forest tree common in places in the lower forests. -I have never been able to identify it; but a “putamen” which was sent to -the Kew Museum was named Spondias with a query. It is to be hoped its -true botanical name will be discovered by one of my successors. Seemann -places it amongst the “desiderata” concerning which further information -is needed. The fruit is a drupe 2 to 2-1/2 inches long possessing a -pleasant fruity odour and inclosing a hard two-celled stone about 1-2/3 -inch long, one cell containing a large fleshy seed covered with tawny -hair, the other filled with the hair only and containing no seed. The -Fijians say that these fruits, large as they are, are swallowed by the -fruit-pigeons, the stones being found in their gullet. The leaves are -distichous, alternate, lanceolate, eight or nine inches long, glabrous -and dark green above, and covered below with a whitish woolly matted -tomentum. The empty stones are not uncommon in the stranded beach-drift. - - - NOTE 79 (page 395) - - ON WILLOW-LEAVED RIVER-SIDE PLANTS - -A number of observers, beginning with Humboldt, in his _Ansichten der -Nature_, and including Seemann, L. H. Grindon, Ridley, Beccari, and -others, have referred to what is called “stenophyllism” in plants. These -willow-leaved river-side plants are found all over the globe, such -plants usually growing close to the water’s edge in situations where -they are liable to be more or less submerged when the river is in flood. -Seemann, Beccari, and Ridley mention more than two dozen genera -belonging to a great variety of orders, and including Acalypha, -Antidesma, Calophyllum, Eulalia, Eugenia, Fagræa, Ficus, Garcinia, -Ixora, Lindenia, Melastoma, Podocarpus, Psychotria, &c., all tropical, -and represented either in Fiji, Borneo, or in the Malay Peninsula; -whilst my readers will recall amongst temperate floras river-side plants -of the genera Epilobium, Lythrum, Salix, &c., possessing the same form -of leaf and the same station. The genus Eugenia comes under this -category in Fiji, Borneo, and the Malay Peninsula, with reference to one -or more of the species. In Fiji, species belonging to the genera -Lindenia and Dolicholobium especially attracted my attention in this -respect. It is noteworthy that several of the Bornean plants and some of -the Fijian plants here concerned are endemic. Just as I have remarked in -the question of the buoyancy of seeds and fruits, that not all -water-side plants have buoyant seeds or fruits, but that nearly all -plants thus endowed are found at the water-side, so we may say of the -willow-leaved plants, that not all river-side plants have the -willow-form of leaf, but that plants thus characterised gather at the -river-side. Beccari and Ridley regard this willow-form of leaf as the -result of adaptation. Seemann remarks that we have here the old question -whether the webbed feet of a duck are the cause or the effect of the -bird’s swimming; and I take the same position. (See Seemann’s _Flora -Vitiensis_; Ridley in _Trans. Linn. Soc. Bot._, vol. iii. 1888-94; and -Beccari’s _Nelle Foreste di Borneo_, 1902, or the English edition of -1904.) - - - NOTE 80 (pages 255, 504) - - MR. PERKINS ON THE HAWAIIAN LOBELIACEÆ (_Fauna hawaiiensis_, vol. I.) - -My view, that the early Hawaiian Lobeliaceæ acquired the monstrous form -of their flowers in the humid forests of a later age, is supported by -the observations of Mr. Perkins on the connection between the -highly-specialised nectar-eating Drepanids of Hawaii and the -highly-specialised flowers of the Tree-Lobelias, a subject further -discussed in Chapter XXXIII. This naturalist ascertained, in the case of -one of the trees, that fertilisation could only be effected by these -birds. So close is the biological connection between the Drepanid and -the Tree-Lobelia, that Mr. Perkins finds here in part the cause of the -development of the most remarkable forms of the birds. The botanist, -also, would not dissociate the plants from this conclusion. There would -be every reason to look for abnormal growth in birds and plants when the -bird depends on the flower for its food, and the flower is dependent on -the bird for its pollenisation. It is through such guises that the -zoologist and the botanist have to penetrate when establishing the -systematic affinity. - - - NOTE 81 - - ON THE VERTICAL RANGE OF SOME OF THE MOST TYPICAL AND MOST CONSPICUOUS -OF THE PLANTS IN THE FORESTS ON THE HAMAKUA SLOPES OF MAUNA KEA, HAWAII - -During a descent of this mountain on its north side to near Ookala, the -conditions were unusually favourable for recording the range of altitude -for some of the plants easily recognisable. - -Acacia koa began at 6,700 feet, and extended down to 2,300 feet. - -Rubus (“akala”) began at 6,500 feet, and extended down to 2,500 feet. - -Cheirodendron (“olapa”) began at 6,400 feet, and extended down to 2,200 -feet. - -Cyanea, a lobeliad growing on trunks of tree-ferns, began at 4,000 feet, -and extended down to 2,300 feet. - -Freycinetia began at 3,850 feet, and extended down to 2,000 feet. - -Asplenium nidus began at 2,800 feet, and extended down to 2,200 feet. - -Aleurites moluccana began at 1,800 feet, and extended down to 50 feet. - -Metrosideros polymorpha, ranging through all the zones. - - - NOTE 82 (page 416) - - ABORIGINAL WEEDS[7] - - (Found by Captain Cook’s Botanists, Banks, Solander, the Forsters, - Nelson, &c., - in the Pacific Islands, 1768-80) - - +-------------------------------+---------------------------------+----------------------------------+ - | | Locality given by Cook’s | General distribution. | - | | botanists. | | - +-------------------------------+---------------------------------+----------------------------------+ - |Cardamine sarmentosa | Tahiti | Polynesia. Introduced into Peru.| - |Sida microphylla | Tonga, New Hebrides | Old World tropics. | - |Sida rhombifolia | New Hebrides. H. | Tropics of Old and New World. | - |Urena lobata | Tahiti | Tropics of Old and New World. | - |Waltheria americana |{ Tahiti. H. |} Tropics of Old and New World. | - | |{ New Caledonia |} | - |Oxalis corniculata | Tahiti | Old and New World. | - |Cardiospermum halicacabum | Tahiti. H. | Tropics of Old and New World. | - |Desmodium polycarpum | Tahiti | Old World. | - |Phaseolus truxillensis | Tahiti. H. | Tropics of Old and New World. | - |Lablab vulgaris | Tahiti | Old World tropics. | - |Abrus precatorius | Tahiti | Tropics of Old and New World. | - |Cassia sophora | Tonga | Tropics of Old and New World. | - |Hydrocotyle asiatica | New Caledonia | Tropics of Old and New World. | - |Oldenlandia tenuifolia | New Hebrides | Old World? | - |Oldenlandia paniculata | Tonga | Old World tropics. | - |Geophila reniformis | Tahiti | Tropics of Old and New World. | - |Ageratum conyzoides | New Hebrides. H. | Tropics of Old and New World. | - |Adenostemma viscosum | Tahiti. H. | Tropics of Old and New World. | - |Eclipta alba | New Hebrides | Tropics of Old and New World. | - |Siegesbeckia orientalis | Tahiti | Tropics of Old and New World. | - |Bidens pilosa | Tonga | Tropics of Old and New World. | - |Dichrocephala latifolia | Tahiti, Tonga, New Hebrides | Old World tropics. | - |Sonchus asper | Tonga, New Zealand | Old World. | - |Ipomœa insularis | New Hebrides, Tonga, Hawaii | Australia and Polynesia. | - |Ipomœa bona-nox | New Hebrides, Tonga, Tahiti. H.| Tropics of Old and New World. | - |Solanum nigrum, var. oleraceum.| Tahiti. H. | Old and New World. | - |Physalis angulata | Tahiti | Tropics of Old and New World. | - |Vandellia crustacea | Tahiti | Tropics of Old and New World. | - |Leucas decemdentata | Tahiti | Old World tropics. | - |Teucrium inflatum | Tonga | New World tropics. | - |Amarantus melancholicus, var. | | | - | tricolor | New Hebrides, Tahiti | Old World. | - |Euxolus caudatus | Tonga, Tahiti | Old World tropics. | - |Achyranthes aspera | Tahiti | Old and New World. | - |Cyathula prostrata | Tahiti | Old World tropics. | - |Fleurya interrupta | Tahiti, Tonga | Old World tropics. | - |Commelina pacifica | Tonga, New Caledonia. H. | Tropics of Old and New World. | - |Eleusine indica | Tahiti. H. | Tropics of Old and New World. | | - +-------------------------------+---------------------------------+----------------------------------+ -Footnote 7: - - Seemann is the principal authority, the results of his examination of - the old collections being given in his _Flora Vitiensis_. Species - regarded by Hillebrand as indigenous in Hawaii or as existing in that - group at the time of its discovery by Cook are indicated by H in the - second column. - - - NOTES 83-89 omitted - - - NOTE 90 (page 29) - -ON THE BUOYANCY OF THE SEEDS OF EUPHORBIA AMYGDALOIDES AND E. SEGETALIS - -The seeds of both species have no proper buoyancy, and display no -structure in their testas suggesting it; though, through the shrinking -of the nucleus, a temporary floating power may be acquired with less -mature or imperfect seeds. They support the general principle indicated -for the British species on page 29. - - - NOTE 91 - - MR. E. KAY ROBINSON ON THE DISPERSAL OF ASTER TRIPOLIUM - -According to this naturalist, the seeds of this plant are eaten in -winter by snow-buntings on the English east coast. In reply to my query -he tells me that the “draggled fluff still containing seeds” might -easily adhere to birds (_The Country-Side_, Sept. 30, 1905). - - - - - GENERAL INDEX - - -_Note._—Several subjects are worked up in this index, which, on account -of the plan of the book, are not dealt with connectedly in the text. As -examples may be cited the entries under the heads of “Hawaiian Flora”; -“Species, their development”; “Fruit-pigeons”; “Polymorphous Species”; -&c. - -The figures in larger type indicate the pages where the subject is -treated at length or where the most important points are discussed. This -sign is not often used where the references can be classed, or where -several references of importance belong to the same subject. - - Abrus precatorius, 531, 605 - - Acacia farnesiana, 478, 552, 555, 556, =557=, =559= - - Acacia heterophylla, 200 - - Acacia koa, 151, =200=, 533, 604 - - Acacia laurifolia, 45, 133, 134, =164=, 166, 200, 529, 551 - - Acacia richii, 531, 549 - - Acæna exigua, 253, 275; - genus, 270-2, =275=, =276= - - Acalypha, 395, 603 - - Acer campestre, 536 - - Achillea millefolium, 536 - - Achras, 373 - - Achyranthes aspera, 605 - - Acorus; - home of the genus, 396 - - Acrostichum squamosum, 593 - - Adaptation in relation to means of dispersal, 11, =99-103=, 105, - =119-129=, 324, 516, 522 - - Adenanthera pavonina, 159, 420 - - Adenostemma viscosum, 240, 417, 568, 605 - - Ægiceras, 470-1 - - Æthusa cynapium, 28, 536 - - Afzelia bijuga, 21, =93=, 107, =170-6=, 436, 529, 563 - - Agapetes, 265 - - Agathis: _see_ Dammara - - Ageratum conyzoides, 417, 531, 605 - - Aglaia, 376 - - Agrostemma, 471 - - Agrostis, 272, 275, 538 - - Aira cæspitosa, 417, 418 - - Ajuga reptans, 28, 537 - - Alchemilla arvensis, 418, 536 - - Alchemilla vulgaris, 417 - - Aleurites moluccana, 59, 61, 361, =418=, 435, 438, 533, 549, 554, - 558-9, 560, 604 - - Alexander, Prof., 587 - - Algaroba, 485, 557 - - Algerian beach-flora, 34 - - Alisma natans, 537 - - Alisma plantago, 38, 92, 537 - - Alisma ranunculoides, 537 - - Alliaria officinalis, 536 - - Alnus glutinosa, 31, 37, 430, 537 - - Alphitonia, 333, =346=, 357, 531 - - Alpine floras, 4, 34, 238; - _see_ under Mountain floras - - Alpinia, 531 - - Alopecurus, 538 - - Alsinidendron, 262-3 - - Alstonia, 381, =384=, 548 - - Alyssum maritimum, 536 - - Alyxia, 334, =344=, 531, 533 - - Amarantus melancholicus, 605 - - Amarouria, 265 - - America, as the home of tropical shore-plants, 67-75; - _see_ under Hawaiian flora and under Tahitian flora for the American - plants in those islands - - Amorphophallus, 412, 414 - - Anagallis arvensis, 537 - - Ancon (Peru), 482, 497; - climate, 492, 598 - - Angelica sylvestris, 28, 536 - - Aniseia uniflora, 530, 563 - - Anona paludosa, 68, 77, 109, 115, 435, 438, 486, 488-9, 498 - - Antarctic flora, represented in the Pacific islands, 271-3, 287, 292, - 294, 305, 503-4, 518 - - Anthyllis vulneraria, 536 - - Antidesma, 371-2, 603 - - Antofagasta (Chile), beach-drift, 480 - - Apetahia, 252, 256-7 - - Apium graveolens, 28, 536 - - Apium inundatum, 28, 536 - - Apium nodiflorum, 28, 37, 536 - - Arabis albida, 568 - - Arabis hirsuta, 536, 568 - - Arabis thaliana, 536, 567 - - Arachis hypogæa, 479 - - Araliaceæ, 261-3 - - Araucaria, 298 - - Arcangeli, Prof., on the Italian species of Medicago, 431 - - Arenaria peploides, 35, 36, 107, =116=, 429-32, 536, =541=, 544 - - Argemone mexicana, 533 - - Argyreia tiliæfolia, 20, 106, 110, 533-4, 558 - - Argyroxiphium, 236-8, 240, 243-4 - - Arica (Chile), 481-2, 497 - - Armeria vulgaris, 33, 34, 36, 511, 537, =540= - - Armeria maritima, 477 - - Artemisia, 238, 240, 269, 272, =278-9=, 540 - - Artemisia absinthium, 279, 536, 540 - - Artemisia maritima, 33 - - Artemisia tridentata, 279 - - Artemisia vulgaris, 279, 536, 540 - - Artocarpus incisa, 531 - - Artocarpus integrifolia, 531 - - Arum maculatum, 537 - - Arundel, Mr., 49, 179 - - Arundo phragmites, 538 - - Asparagus, 538 - - Aspidium aculeatum, 226 - - Aspidium caryotideum, 593 - - Aspidium filix mas, 225, 593 - - Asplenium adiantum nigrum, 225, 593 - - Asplenium aspidioides, 593 - - Asplenium contiguum, 593 - - Asplenium fragile, 593 - - Asplenium monanthemum, 593 - - Asplenium nidus, 554, 604 - - Asplenium trichomanes, 225, 593 - - Astelia, 270-2, 274, 290, =291=, 292-4, 305 - - Aster tripolium, 28, 34-6, 89, 536, 540, 545, 581, 605 - - Astrocaryum, 499 - - Astronia, 376 - - Atlas, Great; - flora, 238, 277 - - Atriplex patula, 537, 544; - genus, 284, 416, 546 - - Avicennia, 68, 69, 77, 78, 438, 484-5, 489, 498 - - Azolla, 488 - - Azores, 505 - - - Bäckström, Helge, 601 - - Baillon, on the Lobeliaceæ, 251 - - Baker, Mr., on Fijian ferns, 224, 592 - - Bakeria, 265 - - Ball, Mr., 238, 476 - - Ballota nigra, 28, 537 - - Bananas: _see_ Musa - - Banyans, 387 - - Barbarea vulgaris, 536 - - Bark, in beach-drift, 430 - - Barratt, Mr., 215 - - Barringtonia formation, 550 - - Barringtonia, genus; - buoyancy and structure of fruits, 17, 18, 121, 160; - mode of dispersal, 161; - seed-structure and vivipary, 132, 168, 573; - relation between coast and inland species, 134, 166 - - Barringtonia edulis, 19, 161, 531, 573 - - Barringtonia excelsa, 19, 108, 574 - - Barringtonia racemosa, buoyancy of fruits, 18, 161, 529; - structure of seeds and fruits, 108, 161, 564, 573-6; - fruits in drift, 76, 78, 435; - station, 43, 47, 551; - distribution and dispersal, 68, 160, 551, 563 - - Barringtonia samoensis, 19 - - Barringtonia speciosa, buoyancy of fruits, 18, 161, 529; - structure of seeds and fruits, 108, 114, 161, 573-6; - fruits in drift, 76, 78, 79, 435-7; - station, 43, 551; - distribution and dispersal, 49, 56, 57, 64, 68, 160, 563 - - Barringtonia, undescribed species; - in Fiji, 19, 161, 531, 574; - in Solomon group, 161 - - Barrows, Prof., 373 - - Bartsia odontites, 537 - - Bassia, 374 - - Batatas edulis, 415; - _see_ under Ipomœa batatas - - Batis maritima, 482, 485, =546=, 557 - - Bats, as dispersing agents, 321, 343, 394, 510, 514 - - Battandier, M., 34 - - Bauhinia, 531 - - Beach flora: _see_ Littoral plants - - Beach formation, 43, 550 - - Beach seed-drift, 31, =429=, 479, 480, 482, 489, 499, =557= - - Beach temperature, 477, 480, 481, =595= - - Beal, Prof., 568 - - Beccari, Dr., on the dispersal of Brackenridgea, 124, 569; - on the Cassowary as a seed-distributor, 152; - on Ficus, 388, 504; - on the retrocession of cultivated plants, 161; - on Sararanga, 156; - on willow-leaved plants, 603 - - Beech-nuts in drift, 429 - - Begonia, 394, 509 - - Begoniaceæ, 263, 394 - - Bell, Mr. Jeffrey, on Peruvian corals, 497 - - Bentham, Mr., 2, 423; - on the Compositæ, 236, 245, 248 - - Berberis vulgaris, 535 - - Berkeley, Mr., 539 - - Bermuda, 348, 351 - - Bernicla sandwicensis, 241, 275, 283 - - Berrya, 379 - - Beta maritima, 35, 537, 542 - - Betula alba, 537 - - Bidens, 379 - - Bidens cernua, 28, 31, 536, 540, 544 - - Bidens pilosa, 379, 533, 605 - - Bidens tripartita, 28, 536, 544 - - Bird, Miss, 584, 587 - - Birds, differentiation of, 5-8, 504-6, 514, 520-2; - crossing oceans, 506; - migrations, 505-6; - Polynesian, 67; - at high altitudes, 241; - biologically connected with plants, 504; - as seed-dispersers, 5, 205, 226, 241, 296, 321; - _see_ under Fruit-pigeons, Ducks, Geese, Sea-birds, Tetraonidæ, &c. - - Bischoffia, 381, =386=, 531 - - Blysmus rufus, 537 - - Bobea, 262-3 - - Boehmeria, 263, 356 - - Boerhaavia, =355=, 552, 568 - - Bonin Islands, 54, 320 - - Boobies (Sula), as seed-dispersers, 188, 347, 356, 511 - - Borago officinalis, 537 - - Bourbon Island: _see_ Mascarene Islands - - Bourne, Mr., 172 - - Brackenridgea, 113, 124, 569 - - Brandis, Dr., on the dispersal of Santalum album, 283 - - Brassica, 536 - - Breadfruit (Artocarpus), 412, 415, 531 - - Breweria, 362-3 - - Brighamia, 252, 255, 258 - - British flora, 23, 31, 115, 432, 535, 539-44 - - Brown, Dr. R., on seed-dispersal by sea-birds, 510 - - Brown, Mr. R., on a drift seed of Cæsalpinia, 189 - - Broussaisia, 263 - - Bruguiera, 43, 441, 551; - dispersal by currents, 48, 55, 77, 94, =461-2=, 467, 529; - distribution, 54, 68, 69, =461=, 563; - in beach-drift, 435, 437, 461; - seed-development and germination, =463-6=, 468-71; - fertilisation, =462= - - Bryonia dioica, 536 - - Buller, Sir W., on the fruits and seeds eaten by New Zealand birds, - 292, 296, 301, 321, 337, 347, 508, 541 - - Buoyancy of seeds and fruits, of Pacific plants, 12-22, 104-15; - of Fijian plants, 44-6, 529, 531-3; - of Hawaiian plants, 57, 533, 552; - of Tahitian plants, 49; - of British plants, 23-39, 115, 535-8, 539-44, 566; - its relation to sea-density, 88-98, 516; - structures concerned, 17, 104-18; - the question of adaptation, 119-29, 516, 569; - effect of inland extension, 121, 568; - the great sorting process, 16, 24, 30, 515; - long flotation experiments, 530, 539; - tables showing results of experiments, 529-38, 552; - effect of drying, 535, 538, 539-44, 571-2; - precautions in testing buoyancy, 566; - the risks of the floating seed in warm seas, 79-87 - - Burkill, Mr., on the Tongan flora, 224, 232, 335, 385 - - Butomus umbellatus, 537 - - Butterflies at high altitudes, 509, 583 - - Buttneria, 379 - - Button, Mr., 172 - - Byronia, 371 - - - Cacao, 489 - - Cacti, 471, 485, 560 - - Cæsalpinia, general account of Pacific species, 183-97 - - Cæsalpinia bonduc, buoyancy of seeds, 21, 192-5, 529; - structures concerned with seed-buoyancy, 106, 191-2; - dispersal by currents, 189; - distribution, 186, 563; - station and extension inland, 49, 186-7 - - Cæsalpinia bonducella, buoyancy of seeds, 21, 192-5, 529, 530-1, 552; - structures concerned with seed-buoyancy, 106, 111, 191-2; - dispersal by currents, 57, 189, 430, 562, 563; - in beach-drift, 189, 430, 437-8, 558; - dispersal by birds, 57, 188, 511, 581; - station and extension inland, 42, 59, 186-8, 551, 552-4, 559; - germination, 191; - distribution, 68, 186, 563; - suggested relation to C. bonduc, 573 - - Cæsalpinia nuga, 183 - - Cæsalpinia, undescribed mountain species of Fiji, 184-5 - - Cakile maritima, 30, 35, 36, 109, =116=, 429-31, =432-3=, 536, 539, - =542=, 544 - - Calamintha officinalis, 28, 537 - - Californian current, 491 - - Calla palustris, 537, 544 - - Callao, 482, 492, 496-7, 599, 601 - - Callitriche, 38, 537 - - Calonyction: _see_ Ipomœa - - Calophyllum; - relation between coast and inland species, 17, 18, 120, 134, 136, 533 - - Calophyllum amœnum, 534 - - Calophyllum burmanni, 18, 136, 531 - - Calophyllum calaba, 534 - - Calophyllum inophyllum; - buoyancy of fruits, 18, 529, 534, 552; - structures concerned in buoyancy, 107, 113, 115, 122; - fruits in beach-drift, 437, 558; - station, 42, 43, 49, 52, 550, 554; - its relation to inland species, 136 - - Calophyllum spectabile, 18, 136, 389, 531 - - Caltha palustris, 85, 535 - - Camelina sativa, 567 - - Campylotheca, 236-8, 240, 243-4, 533 - - Cananga odorata, 159, =393=, 531 - - Canarium, =400=, 532 - - Canavalia, 107, 201-2; - Pacific species, =145=, 578; - relation between the littoral and inland species, 20, 134, 145 - - Canavalia ensiformis, 145, 529, 563, 578-9, 581 (turgida) - - Canavalia galeata, 20, =145-6=, 533 - - Canavalia sericea, 107, 145, 529, 563, =578-9= - - Canavalia obtusifolia; - seed-buoyancy and dispersal by currents, 19, 83, 145, 529, 562, - =579=; - cause of buoyancy, 107, 113; - in beach-drift, 437-8, 489; - station and distribution, 42, 43, 488, 498, 547, 550, 563, =578= - - Candolle, A. de, 25, 62, 80, 239, 418, 562, 573 - - Candolle, C. de, 376 - - Canna indica, 532 - - Canthiopsis, 265 - - Canthium, 355 - - Cape-pigeon (Daption capensis), 511 - - Capercailzie, 282 - - Capparis sandwicensis (sandwichiana), 533, 553 - - Capsella bursa pastoris, 536, 567 - - Carapa, seed-buoyancy, 108, 114, 529; - beach and river drift, 76, 435, 437; - germination, 76, 78, 564; - station, 43, 550-1; - distribution, 68-9, 562 - - Cardamine hirsuta, 536 - - Cardamine pratensis, 536 - - Cardamine sarmentosa, 604 - - Cardiospermum halicacabum, 417, 605; - _see_ Additions and Corrections - - Carduus, 28, 536 - - Carex, 37, 272, 283, 538, 540, 544 - - Careya, 575 - - Carmichael, Captain, on Tristan da Cunha, 276, 286 - - Carpenter, Captain, 490 - - Carruthersia, 265 - - Caryophyllaceæ; - represented in the early flora of Hawaii, 261-3, 518 - - Caspary, R., on the dispersal of water-lilies, 512 - - Cassia gaudichaudii, 533 - - Cassia occidentalis, 533 - - Cassia sophora, 605 - - Cassowaries as seed-dispersers, 152, 160 - - Cassytha filiformis; - fruit-buoyancy and dispersal by currents, 56, 57, 71, 106, 111, - 121-2, 530, 552, 563, 569; - dispersal by birds, 71, 123, 564; - station and distribution, 56, 59, 67, 122, 551-2, 563; - extension inland, 42, 49, 59, 121, 547, 548, 559, 569 - - Castillo (Drake del), on the Tahitian flora, 46, 49, 221, 231, 254, - 285, 318, 347, 504, 551 - - Casuarina, 134, 136, 479 - - Casuarina equisetifolia, 42, 45, 136, 530, 548-9 - - Casuarina nodiflora, 136, 549 - - Cattle Plains of Hawaii, 208 - - Centranthus ruber, 536 - - Centropogon, 251, 259 - - Cerastium vulgatum, 536 - - Ceratophyllum, 38, =398-9=, 408, 537 - - Cerbera odollam, distribution, 64, 563; - station, 551; - extension inland, 41, 42, 49, 121, =547=, 548, 569; - fruit-buoyancy and dispersal by currents, 76, 108, 114, 121, 530; - in river and beach drift, 76, 435, 437 - - Chærophyllum sylvestre, 28, 536 - - Chagres River, 498 - - Chamisso, 367 - - Chancay coast (Peru), 482 - - Charpentiera, 263 - - Cheeseman, Mr., on the flora of Rarotonga, 50, 177, 208, 216, 232, 238, - 256, 291, 293, 295; - on Kermadec plants, 295, 420, 572; - on Polynesian food-plants, 415, 420 - - Cheirodendron, 262, 263, 281, 343, 364, 533, =594=, 604 - - Chelidonium majus, 535 - - Chenopodium, 272, 283, 284, 537 - - Cherry (Cerasus) stones in beach-drift, 429, 431, 479 - - Chestnuts in beach-drift, 429 - - Chile; - the coast plants and beach-drift, 431, 474-80, 596; - the coast climate, 491-3, 598-601 - - Chloridops kona, 275 - - Christmas Island, 422 - - Chrysanthemum leucanthemum, 536, 568 - - Chrysanthemum segetum, 536 - - Chrysodium aureum, 48, 486, 498 - - Chrysophyllum, 362 - - Chrysosplenium, 536 - - Cicuta virosa, 28, 536 - - Citrus aurantium, 125, 532 - - Citrus decumana, 125, 126, 532, 533 - - Citrus, other species, 125, 436, 532 - - Cladium mariscus, 537 - - Clarke, Mr. C. B., on Cyrtandra, 316 - - Cleome, 362 - - Clermontia, 252-6, 258-9, 533 - - Clerodendron, 17, 121 - - Clerodendron inerme, 47, 76, 108, 114, 435, 530, 551, 563 - - Climate and currents, 491-5, 500, 597-601 - - Clouds, altitude of, on mountains; - on the Owen Stanley Range, 215; - observations on the summit of Mauna Loa, 584; - on the Chilian and Peruvian sea-borders, 492, 494 - - Coast and inland species of a genus, relation of, 16, 27, 133-169 - - Cocculus, 362 - - Cochlearia officinalis, 33, 536, 540 - - Coco-de-mer, 61 - - Coco-nut palm (Cocos nucifera), 67, 108, 413, 530, 552, 553-4, 563; - effective dispersal by currents, 436; - viviparous, 472 - - Coix lachryma, 532 - - Collomia, 568 - - Colobanthus, 263 - - Colocasia antiquorum, 412 - - Colon; - shore-plants, 498 - - Colubrina, 134, =137= - - Colubrina asiatica; - station and distribution, 137, 552, 556, 562-4; - inland extension, 49, 547; - seed-buoyancy and dispersal by currents, 56, 57, 105, =137=, 529; - river and beach drift, 436, 559 - - Colubrina oppositifolia, 137, 533-4, 576 - - Columbæ, extinct; - of the Mascarene Islands, 152, 157, 159, 169, 200, 517 - - Comins, Rev. R. B., 379 - - Commelina nudiflora (syn. pacifica), 533, 605 - - Commersonia echinata (syn. platyphylla), 532, 548 - - Commersonia, 376, 380 - - Compositæ, age of, 9, 231-49, 304, 306, 503, 514, 517-20; - endemic Hawaiian and Tahitian genera, 236, 248; - arborescent, 235-49; - dispersal by birds, 241, 593, 605; - fruit-buoyancy of British species, 536 - - Coniferæ, age of, 303-6, 502-3, 514, 519-20; - Fijian, 294, 297-306; - New Zealand, 507-9, 514 - - Conocarpus erectus, 68, 108, 438, 498 - - Convolvulaceæ, seed-buoyancy, 28, 110, 117, 544; - abortive germination of floating seeds, 76, 79, 83, 85, 87 - - Convolvulus arvensis, 28, 110, 537, 544 - - Convolvulus sepium; - station, 29; - distribution, 417-8, 573; - seeding in England, 539; - buoyancy of seeds, 26, 29, 106, 110, 537, 539, 544; - suggested dimorphism, 573 - - Convolvulus soldanella; - seed-buoyancy, 26, 28, 35, 36, 83, 91, 106, 110, 115, 537, 542, 566; - non-germination in sea-water, 35, 544, 546; - seeds in beach-drift, 31, 429-31; - dispersal by currents, 432-3; - distribution, 131, 433, 476-9, 542, 572; - suggested dimorphism, 573 - - Convolvulus tricolor; - germination in sea-water, 546 - - Conway, Sir Martin, 241 - - Coprosma, 270-2, 274-5, 290-3, =294-6=, 305, 315, 321, 331, 533 - - Corals on the coasts of Chile and Peru, 496-7, 601 - - Cordia, 17, 121, 134, 137, 485, 488 - - Cordia aspera, 137 - - Cordia subcordata; - station and distribution, 52, 551-2, 555, 563; - fruit-buoyancy, 79, 108, 114, 530-1; - sea and beach-drift, 79, 437; - becoming extinct in Hawaii, 578 - - Cordyline, 420, 532 - - Coreopsis, 237 - - Coriaria, 270, 290, =291-2=, 305 - - Corks in beach-drift, 479 - - Corral (Chile), 478 - - Corvus tropicus, 321 - - Corylus avellana, 126, 429 (hazel), 431 (hazel), 537, 538, 572 - - Corynocarpus, 508 - - Cotula plumosa, 241 - - Cotyledon umbilicus, 417-8, 536 - - Couthovia, 265-6, =401=, 532 - - Crambe maritima, 35, 536, 542, 581 - - Cratægus oxyacantha, 536 - - Cratæva religiosa, 379 - - Crepis, 536 - - Crinum asiaticum, 530 - - Crithmum maritimum, 28, 35, 109, 116, 429, 433, 536, =542=, 544 - - Crocodile, in Fiji, 65 - - Crosby, Mr., 550 - - Croton, vivipary, 472 - - Cucumis acidus, 532 - - Cucurbita, 125, 479, 532 - - Cupania, 532 - - Curcuma longa, 548 - - Curlews, as seed-dispersers, 355, 356 - - Currents, as seed-dispersers, 4, 12, 44, 49, 57, 58, 61-75, 79-84, 179, - 557, 562, 571; - Gulf-stream, 4, 80, 180, 189, 430, 570, 581; - currents reaching Hawaii, 58, 72-5, 557-9; - Humboldt or Peruvian current, 480, 483, 490-5, 500, 597-601; - _see_ under Climate and currents - - Cuscuta, 58, =366=, 537, 552, 553, 555, 559, =595= - - Cyanea, 252, 254-5, 258, 604 - - Cyathodes, 272, =282=, 284, 285, 290-1, =292=, 305, 533, 553-4 - - Cyathula prostrata, 605 - - Cycas circinalis, 42, 109, 115, 328, 413, 530, 547, 548-9, 563 - - Cynometra, 108, 529, 563 - - Cyperaceæ, dispersal by ducks, 513; - by purple water-hens, 296 - - Cyrtandra, 308-9, =316=, 331, 405, 520 - - Cyrtosperma, 413 - - Cytisus scoparius, 536 - - - Dacrydium, 294, 297-8, =302=, 305-6 - - Dalbergia monosperma, 106, 435, 529, 551, 563 - - Damasonium stellatum, 537 - - Dammara, 294, =298=, 303-6, 532 - - Daption capensis (Cape-pigeon), 511 - - Darwin, Mr., 24, 150, 177, 347, 497, 509, 538, 539, 542, 544, 546, 568 - - Datura stramonium, 537 - - Davis, Prof., 491 - - Davis, Rev. S. H., 213 - - Dead Sea, density of, 89 - - Death and Evolution, 11, 523 - - Delissea, 252, 254-5, 258 - - Density of sea-water and seed-buoyancy, 88, 566 - - Derris uliginosa; - distribution, 68, 552, 563; - station, 44, 551; - inland extension, 42, 547; - fruit-buoyancy, 106, 111, 529, 552; - river-drift, 76, 435; - beach-drift, 437 - - Deschampsia, 272, 275 - - Desmodium umbellatum, 106, 529 - - Desmodium polycarpum, 605 - - Deterioration of capacity for dispersal, 262-3, 337, 350, 365, 507, - 594-5 - - Deyeuxia, 272, 284-5 - - Dianella, =356=, 533 - - Dichrocephala latifolia, 605 - - Dickson, Mr., 493 - - Didunculus, 159, 393 - - Didymocarpus, 318 - - Differentiation of birds and plants, 505-7, 514, 520, 521-2; - of climate, 470, 473, 507, 521-2 - - Dimorphism, suggested, in Rhizophora, 449, 465, 521; - in Cæsalpinia, 573; - in Convolvulus sepium and C. soldanella, 573 - - Dioclea, 82, 107, 113, 436, 529, 531, 563 - - Dioscorea, 412-4, 532 - - Dispersal, agencies of, 5, 61, 502; - suspension, causes of, 5-9, 242-3, 365, =504=, 514, 521-2; - _see_ under Adaptation, Birds, Currents, Deterioration &c. - - Dixon, Mr. C., on differentiation of birds, 505-6; - on seeds in petrels, 581 - - Döderlein, 54 - - Dodo, 8, 159, 522 - - Dodonæa, 67, 71, 106, 333, =338=, 357, 529, 548, 554, 563 - - Dole, Mr., 241, 557 - - Dolicholobium, =394=, 603 - - Douglas, Mr. D., 278, 282, 586 - - Dove: _see_ Pigeon - - Draba verna, 536 - - Dracæna aurea, =367=, 533 - - Dracæna, vivipary in, 471 - - Dracocephalum, 568 - - Dracontomelon, =399=, 532 - - Dragon-flies, dispersed by winds, 510 - - Drepanididæ, 259, 343, 348, 504-5, 603 - - Drepanocarpus, 562 - - Drift: _see_ Beach seed-drift, River seed-drift - - Drift-timber, 58, 72, 557 - - Drosera, 4, 253, 270, 272, 285-8, 536 - - Druce, Mr., on Sea-thrifts, 34 - - Drupe, rubiaceous, its first appearance in the Pacific, 262 - - Drying-winds, 491-4 - - Drymispermum, 45, 133, 134, =164=, 166, 530 - - Dryobalanops, vivipary in, 471 - - Dubautia, 236-8, 240, 243-4 - - Ducie Island, 49, 64 - - Ducks, as dispersers of seeds, 241, 277, 356, =369=, =370=, 375, 399, - 506, =512-3=, 514, 541 - - Dusolier, M., 506 - - Dwarfing of shore-plants growing inland, 547 - - Dysoxylum, 376 - - - Easter Island, 64 - - Ecastaphyllum, 562 - - Eclipta alba, 533, 534, 605 - - Ecuador, climate of sea-border, 476, 483, 489-91, 493-6, 500, 597; - influence of the Humboldt or Peruvian current, 490-1, 493-5, 500, - 599; - mangroves, 3, 77, 445, 448, 474-6, 483-90, 521, 597; - beach-drift and beach-plants, 180, 488-9; - temperature of the Guayaquil estuary, 77, 78, 565; - drift of the Guayaquil River, 91, 435, 488-9; - Santa Rosa River, 486; - Machala plains, 485, 495; - Santa Elena coast, 483, 490, 494, 597 - - Eeka, mountain in Maui, 208, 214, 253 - - Eggers, Baron von, on the climate and mangroves of Ecuador, 449, 450, - 476, 483-4, 487, 490, 493-5, 597 - - Ekstam, O., on seed-dispersal in Spitzbergen and Nova Zembla, 242, 277, - 282, 434, =511=, =512= - - Elæocarpus, =334=, 357-8, 389, 391, 401, 507-9, 532 - - Elatine hydropiper, 536 - - Elatostema, 317, 383, 391, =405= - - Electrical state of the atmosphere, 582, 588 - - Eleusine indica, 605 - - Elizabeth Island, 64 - - Embelia, 362-4, 409, 520 - - Empetrum nigrum, 511 - - Endemic genera and species, tables of, 232, 233, 244, 252, 255, 263, - 265; - _see_ under Fiji, Hawaii, Tahiti - - Endemism: _see_ under Species - - English beach-drift, 429-33; - _see_ Beach seed-drift - - Entada scandens, =176=; - station, 44, 48, 50, =177=, 182; - distribution, 68, =177-9=, 182, 200, 499, 500, 551, 563; - dispersed by currents, =179-80=, 182; - seed-buoyancy, 82, 94, 106, 111, =181-3=, 529, 531; - river and beach drift, =180=, 430, 435, 437-8, 488-9, 499 - - Epilobium, 471, 536, 603 - - Epiphytic plants, 281 (Vaccinium); - 291 (Weinmannia); - 343 (Myoporum and Vaccinium); - 383 (Loranthus); - 402 (Myrmecodia) - - Eranthemum, 532 - - Eriophorum, 537 - - Erodium maritimum, 33 - - Eryngium maritimum, 28, 35, 536, 539, =543=, 544, 581 - - Erythrina; - general discussion of the genus, =577=; - relation between coast and inland species, 19, 134, =141=; - seed-buoyancy, 107; - seeds in river and beach drift, 435, 437-8, 489 - - Erythrina indica, 19, 107, =141-4=, 529, 562, =577= - - Erythrina monosperma, 20, =141-4=, 200, 533, 553-4, 578 - - Erythrina ovalifolia, 107, 577 - - Erythrina aurantiaca, 144 - - Erythrina vespertilio, 144 - - Estuaries, temperature of tropical, 78, 564 - - Etna, shadow of, 586 - - Eucalyptus, 479 - - Eugenia, 61, 134, =163=, 166, 334, =349=, 357, 507, 532, 603 - - Eugenia brackenridgei, 351 - - Eugenia corynocarpa, 350, 351, 532 - - Eugenia grandis, 163 - - Eugenia malaccensis, 349, 532 - - Eugenia monticola, 351 - - Eugenia neurocalyx, 350, 351 - - Eugenia rariflora, 164, 349-52, 532 - - Eugenia richii, 45, =164=, 350, 529 - - Eugenia rivularis, 350, 532 - - Euonymus europæus, 536 - - Eupatorium cannabinum, 536 - - Euphorbia, 29, 478, 537 - - Euphorbia amygdaloides, 537, 605 - - Euphorbia atoto, 106, 530 - - Euphorbia cordata, 556 - - Euphorbia helioscopia, 29, 537 - - Euphorbia paralias, 29, 31, 35, 109, =116=, 429, 431-3, 537, =543=, 544 - - Euphorbia peplus, 537 - - Euphorbia segetalis, 537, 605 - - Eurya, 371-3 - - Euxolus caudatus, 605 - - Everett, Mr., 388 - - Evolution, 11, 227, 522; - _see_ under Adaptation, Natural Selection, and Species - - Excæcaria agallocha, 43, 109, 436, 530, 551, 563 - - Exocarpus, 271, 274-5 - - - Fagræa berteriana, 42, 45, =385=, 532, 548, 603 - - Fagus, 508 - - Fanning Island, 377 - - Fernando Noronha, its flora and its birds, 8, 144, 366, 388 - - Ferns, =220-30=, 509, 517, 592, 593 - - Ficus, 61, 377, =387=, 395, =504=, 532 - - Fiji, the climate, 209, 215-8; - the seed-drift of rivers and beaches, 435-6; - the “talasinga” plains, 42, 43, 215, 386, 547, =548=, 569; - area and altitudes, 207-8 - - Fijian flora, the littoral plants, 13, 40, 528-30, 550-1; - ferns and lycopods, 220-30, 592; - endemic genera and endemic species, 231-5, 264-6; - mountain plants, 269, 293-7, 305; - conifers, 297-306 - The age of Malayan plants comprising first the genera widely - dispersed over the Pacific, 307-58; - and then those that are locally dispersed, 359, 360; - the last comprising Fijian genera found in Hawaii and not in - Tahiti, 359, 371-4; - Fijian genera found in Tahiti and not in Hawaii, 359, 380-8; - and genera found in Fiji, but neither in Hawaii nor in Tahiti, 360, - 399-408 ... - The number of Fijian flowering plants, 528, 592 ... - The Fijian “difficulty,” 155, 166, 169, 517 ... - The buoyancy of the seeds of inland plants, 531 - - Fish in the Humboldt current, 600 - - Fitchia, 236, 237-8, 240, 245, 248 - - Fleurya interrupta, 605 - - Flying-foxes: _see_ Bats - - Focke, W. O., on the dispersal of Leguminosæ, 150, 417 - - Fogs on the Chilian and Peruvian coasts, 475, 481, 490, 492-3, 600 - - Forbes, Dr. H. O., 304, 347, 388 - - Forster on the “Antarctic” flora, 271 - - Fragaria chilensis, 4, 270, 272, 285, 287-8 - - Fragaria vesca, 536 - - Francolins at high altitudes, 241 - - Franseria, 131, 431, 478-9, 597 - - Freycinetia, 308-9, =319=, 331, 504-5, 509-10 - - Frigate-birds (Fregata), as seed-dispersers, 188, 511 - - Fritillaria meleagris, 537 - - Fruits “difficult” or “impossible” from the standpoint of dispersal, - 262, 267, 372-3, 379, 388-9, 391, 409; - _see_ under the Fijian flora and the Hawaiian flora Fruit-pigeons - (Carpophaga, &c.), seeds and fruits eaten by them, limit of size, - 337, 381, 389, 400-1, 410; - Achras, 373; - Areca, 330; - Cananga, 393; - Canarium, 372, 400; - Cassytha, 123, 564; - Coriaria, 292; - Corynocarpus, 508; - Couthovia, 266, 401; - Dracontomelon, 372, 400; - Elæocarpus, 337, 372, 508; - Eugenia, 350; - Ficus, 388, 504; - Gnetum, 404; - Kentia, 330; - Litsea, 508; - Myristica, 403; - Olea, 508; - Oncocarpus, 266; - Phyllanthus, 326; - Pisonia, 347; - Plectronia, 355; - Podocarpus, 301, 508; - Premna, 561; - Psychotria, 314; - Sapota, 373; - Sideroxylon, 373; - Spondias, 602; - Vitex, 564; - Ximenia, 113. - (For additional note on Litsea and Vitex, _see_ Additions and - Corrections) - - Fruits, fleshy, 101 - - Fulmar-petrel, 581 - - - Galapagos Islands, flora and avifauna, 505 - - Galeopsis tetrahit, 28, 537 - - Galium aparine, 27, 536, 539 - - Galium mollugo, 536 - - Galium palustre, 27, 536 - - Garcinia, 575 - - Gardenia, 308-9, =311=, 532-3, 548, 552 - - Garuas, 475, 481, 600 - - Gätke, H., 151, 506, 593 - - Gaudichaud, 385, 434 - - Gay, C., on Chilian plants, 477-8 - - Geese, as seed-dispersers, 241, 275, 283, 356, 511-2, 514 - - Geissois, 343, =393=, 510, 532 - - Geniostoma, 384 - - Geological time, 502, 520 - - Geophila reniformis, 417, 532, 605 - - Geranium, 4, 269, 272, 274-5 - - Germination, of floating seeds, 76; - its connection with vivipary, 78, 84, 87, 191, 468, 521; - effect of previous immersion in sea-water on the germinating capacity - of seeds in soil, 25, 539; - germination in sea-water in temperate latitudes, 35, 544; - germination in sea-water in tropical latitudes, 79-87; - the process in Cæsalpinia, 191, and in Cuscuta, 595 - - Giffard, Mr., 212 - - Gilia, 568 - - Gill, Rev. Wyatt, 48 - - Gizzard-stones in birds, 159 - - Glacial epoch, 503, 509, 513 - - Glaucium luteum, 35, 90, 536, 539, 543 544 - - Glaux maritima, 34, 36, 537, 541 - - Gleichenia, 548 - - Gnetum, 391, =404=, 532 - - Goats as seed-dispersers, 554, 558 - - Goebel, K., 453; - on vivipary, 469-72 - - Gossypium, 352, 533 - - Gossypium tomentosum, 58-9, 529, 533, 552, 556 - - Gouania, 371-3 - - Gouldia, 262-3 - - Goura pigeon, 8, 159, 392 - - Gourds dispersed by currents, 125, 570 - - Græffea, 265 - - Grape-seeds in beach-drift, 429 - - Gray, Asa, 34 - - Grewia, 381, 382, 532 - - Grindon, Mr., 603 - - Grouse-family as seed-dispersers: _see_ Tetraonidæ - - Guavas, 554 - - Guayas or Guayaquil river and estuary: _see_ under Ecuador - - Guettarda, relation between coast and inland species, 17, 121, 134, - =162=, 166, 532, 533; - germination, 79, 132 - - Guettarda speciosa, station at coast, 43, 551; - station inland, 162; - distribution, 64, 68, 563; - buoyancy of fruits and their dispersal by currents, 55-6, 108, 114, - 163, 529; - beach-drift, 163, 437 - - Guilandina (synonym of Cæsalpinia), 188, 562 - - Gulf-stream drift: _see_ under Currents - - Gulls (Laridæ) as seed-dispersers, 241, 511, 514 - - Gum-resins, native names of, 300 - - Gunnera, 269, 271, 272, 274, 275 - - Gunnerus, 570 - - Gyrocarpus jacquini, 2, 49, 68, 106, 111, =422=, 428, 529, 551, 563 - - - Haberlandt, G., on the germination of mangroves, 453, 457, 465 - - Haleakala, 208 - - Hall, Mr. W. L., on the forests of Hawaii, 213 - - Halophily, 581 - - Hamilton, Mr., on the crop-stones of the moa, 159 - - Hann, Prof., on mountain climates, 582 - - Haplopetalon, 264, 265 - - Haplostachys, 263, 594 - - Hawaiian islands; - area and altitude, 207-8; - climate, 209-15, 217-8, 582-92; - temperature and relative humidity, 209-11, 218; - rainfall, 212-15, 218; - meteorology of the summit of Mauna Loa, 582-92 - - Hawaiian flora; - littoral plants, 15, 51, 71, 552, 553-7; - beach-drift, 58, 557; - ferns and lycopods, 220-30, 517, 592, 593; - the eras of the flowering plants, 234-5, 517; - tables of endemic genera and endemic species, 232, 233, 244, 252, - 263; - the age of Compositæ, 235-49; - the age of the Lobeliaceæ, 250-60, 266-7, 603; - the endemic genera belonging neither to the Compositæ nor to the - Lobeliaceæ, 261-4, 594; - mountain flora, 269-88, 518 - The age of Malayan genera, 307-78; - the Malayan genera widely dispersed in the Pacific, 307-58; - the Malayan genera locally dispersed, 359-69; - the residual genera (found only in Hawaii), 359, 361; - the genera occurring in Hawaii and Tahiti but not in Fiji, 359, - 370; - the genera occurring in Hawaii and Fiji but not in Tahiti, 359, - 371; - the absentees from Hawaii, 359, 375 - American plants in Hawaii, 261, 334, 362, 372, 409, 517-8. - They include the following orders and genera:— - Compositæ, 237, 248, 260, 518; - Lobeliaceæ, 254, 260, 266, 518; - Caryophyllaceæ, 261-3, 518; - Sanicula, 272-3, 287; - Sisyrinchium, 272-3, 287; - Fragaria, 285, 288; - Rubus, 273, 285; - Pritchardia, 309, 326; - Lythrum, 362; - Perrottetia, 362; - Sicyos, 362, 365; - Chrysophyllum, 362; - Nama, 362; - Jacquemontia, 362, 365; - Sphacele, 362; - Phytolacca, 362, 364; - Urera, 362 - The Hawaiian difficulty, 140, 165, 168, 517; - Hawaiian plants and birds, 505-6, 603; - buoyancy of fruits and seeds of inland plants, 533-4 - - Haynaldia, 251 - - Hazel: _see_ Corylus avellana - - Hazlewood’s Fijian Dictionary, 400 - - Heather, 511 - - Hedera helix, 536 - - Hedley, Mr., 65, 66, 304 - - Hegelmaier, on Lemna, 408 - - Heilprin, Prof., 506 - - Helianthemum vulgare, 536, 567 - - Heliosciadium, 538 - - Heliotropium anomalum, 49, 56, 58, 365, 370, 528, 552, 554-7 - - Heliotropium curassavicum, 56, 58, 482, 552, 555, 557 - - Hemsley, Mr. W. B., 34, 232, 238, 301, 434, 499, 511, 530, 539, 581; - on dispersal by currents, 62; - on Cæsalpinia, 184; - on insular Leguminosæ, 198-9; - on the Pacific Compositæ, 242, 246-7; - on the Pacific Lobeliaceæ, 251, 257; - on Pisonia, 346; - on Gnetum, 404; - on ten widely spread British plants, 417; - on Rhizophora mangle in the Pacific, 441; - on vivipary in Dracæna, 471; - on Brackenridgea, 569 - - Heptapleurum, 263 - - Heritiera littoralis, station, 43, 551; - distribution, 47, 68, 375-6, 551, 562; - buoyancy of fruits, 106, 112, 529; - their occurrence in river-drift, 79, 435; - and in beach-drift, 437 - - Hernandia peltata, station, 551; - extension inland, 49; - distribution, 54, 64, 68, 562, 563; - buoyancy of fruits, 109, 115, 530; - beach-drift, 437 - - Herpestis monnieria, 552, 557 - - Hesperomannia, 236-7, 243-4 - - Hesselman, H., on plant-dispersal, 282, 511 - - Hibbertia, 402, 548 - - Hibiscus, 21, 134, 137 - - Hibiscus abelmoschus, 21, 124, 532, 533 - - Hibiscus diversifolius, 21, 105, 529 - - Hibiscus esculentus, 21, 532 - - Hibiscus tiliaceus, station, 43, 486, 498, 551, 555-7; - growing inland, 41-2, 419, 547, 557, 560; - distribution, 52, 486, 498, 552, 563; - seed-buoyancy, 21, 105, 529, 552; - beach-drift, 437-8; - river-drift, 435; - sea-drift, 489; - currents, 562 - - Hibiscus youngianus, 21, 533, 534 - - Hillebrand, Dr., on the Hawaiian flora; - introduced littoral trees, 51; - the agency of the currents, 73; - the endemism of the ferns, 224-7, 593; - the Compositæ, 236-47; - the Lobeliaceæ, 250-5; - the absence of Coniferæ, 303; - on Cyrtandra, 317; - other references, 231, 340, 341, 405, 557, 559, 560, 576, 578, 594, - &c. - - Hillebrand, Mr. W. F., 224 - - Hillebrandia, 263, 394 - - Hilo, 53, 213 - - Hippomane mancinella (Manchineel), 68, 109, 498 - - Hochstetter, 336 - - Holland, Mr., 478, 489, 499 - - Holmboe, Jens, on Silene maritima, 280 - - Holmes, Mr., on the Fijian rainfall, 216, 549 - - Honckeneya: _see_ Arenaria peploides - - Hooker, Sir J., 198, 238, 258, 271, 344, 369, 581 - - Horne, Mr., on the Fijian flora, 41, 172, 394, 395, 425, 549, 592 - - Horse-dung containing seeds transported by currents, 558 - - Hottonia palustris, 536 - - Hoya, 377 - - Humboldt on insects in the upper air-currents, 583 - - Humboldt current, 480, 483, 490-5, 500, 597-601 - - Hydnophytum, 402 - - Hydrocharis, 537 - - Hydrocotyle asiatica, 532, 605 - - Hydrocotyle verticillata, 533-4 - - Hydrocotyle vulgaris, 28, 536, 544 - - Hygrophytes, 32, 515 - - Hypericum, perforatum, quadrangulum, elodes, 536 - - - Iceland; - plants, 505 - - Ilex aquifolium, 536 - - Impatiens fulva, 536 - - Impatiens parviflora, 536 - - Incas; - bones exposed on the Ancon plain, 497 - - Inocarpus edulis, 108, =421=, 435, 529, 563 - - Insects in the upper air-currents, 509-10, 514, 582-3 - - Ipomea: _see_ Ipomœa - - Ipomœa, 20, 76, 109, 134, 137, 489, 532-3, 546 - - Ipomœa batatas, 20, 415 (Batatas edulis), 532 - - Ipomœa bona nox (Calonyction speciosum), 20, 106, 110, 533, 534, 554, - 558, 605 - - Ipomœa glaberrima (Calonyction comosperma), 20, 52, 57, 106, 110, 472, - 530, 552, 555, 558, 564 - - Ipomœa grandiflora, 20, 106, 530, 531 - - Ipomœa insularis, 20, 110, 532, 533, 553, 554, 605 - - Ipomœa peltata, 110, 435, 472, 532 - - Ipomœa pentaphylla, 20, 110, 533 - - Ipomœa reptans, 533 - - Ipomœa tuberculata, 20, 110, 533, 556 - - Ipomœa turpethum, 20, 106, 110, 532 - - Ipomœa pes-capræ; - seed-buoyancy, 20, 21, 83, 106, 110, 121, 530, 546, 569; - dispersal by currents, 56-7, 562; - seeds in river and beach drift, 435, 437-8, 489, 558-9; - distribution, 56, 68, 433, 488, 498, 552, 563, 564, 572; - station, 43, 551, 553-6, 560; - growing inland, 21, 41, 42, 121, 547, 560, 569 - - Iris; - home of the genus, 396 - - Iris fœtidissima, 24, 27, 537 - - Iris pseudacorus, 24, 27, 30, 430, 537, 540, 544 - - Isodendrion, 263 - - Isotoma, 255 - - - Jacquemontia, 58, =365=, 533, 552, 553, 555-7 - - Jacquin, on Rhizophora mangle, 457 - - Jambeli Island, 488-9 - - Johnston, Sir H., 241, 252, 510 - - Jouan, H., 187 - - Juan Fernandez, 222, 316 - - Juncus, 89, 471, 537, 545, 568 - - Junghuhn, on the climate of the Java mountains, 211 - - Jussiæa villosa, 533 - - - Kadua, 262, 263 - - Kandavu, 207, 208 - - Karsten, G., on Rhizophora, 453 - - Kauai, 207, 208, 214 - - Kauri pine, 299 - - Keeble, F. W., on the dispersal of Loranthus, 383 - - Keeling atoll, 81, 190, 510 - - Kerguelen, 241, 242, 276 - - Kermadec islands, 258, 276, 295, 420, 572 - - Kerner, Dr., 63, 101, 277, 408, 469, 567 - - Kidder, Dr., 241, 276 - - Kilimanjaro, 251 - - Kinabalu, Mount, in Borneo, 286, 295-6, 301 - - Kirk, T., on the forest-flora of New Zealand, 299, 347, 507 - - Kittlitz, on Nipa fruticans, 66 - - Kiwi, 522 - - Kleinhovia hospita, 21, 105, 376, 529, 562, =602= - - Koebele, Prof., 212, 578 - - Kolpin-Ravn, on seed-buoyancy, 24, 38, 116, 538 - - Krakatoa beach-drift, &c., 180, 190, 206, 221 - - Krämer, Dr., 275, 283 - - Kurz, Mr., 577; - on Scirpodendron, 406 - - - Labiatæ; - station and seed-buoyancy, 28, 537; - mucosity of seeds, 567-8; - Hawaiian endemic genera, 261-3, 518, 594 - - Lablab vulgaris, 413, 417, 605 - - Labordea, 262, 263 - - Labrador current and climate, 493 - - Lagenaria (bottle-gourds), dispersal by currents, 125, 570 - - Lagenophora, 240, 242, 270-2, 274-5, =276=, 293-4, 305, 568 - - Lagerheim, Prof., on the dispersal of Empetrum nigrum, 512 - - Lagopus: _see_ Tetraonidæ - - Laguncularia; - station and distribution, 68-9, 484, 498; - dispersal by currents, 77; - fruit-buoyancy, 108; - fruits in beach-drift, 438; - in river-drift, 489; - in sea-drift, 489; - germination, 77-8, 469, 471 - - Lamium album, 28, 537 - - Lamium galeobdolon, 537 - - Lamium purpureum, 28, 537, 568 - - Lapsana communis, 536 - - Larch, 430 - - Lathyrus maritimus, 35, 107, 116, 430, 432, 536, =543= - - Lathyrus pratensis, 536 - - Layard, Messrs., on birds and seeds, 143, 296, 388, 420 - - Leersia oryzoides, 538 - - Leguminosæ in littoral floras, 13, 19, 68, 76, =79-85=, 87, =107=, 111, - 117, =140-51=, =170-97=, =198=, 438-9; - dispersal by birds, 150, 417, 581 - - Lemnaceæ, =407=, 488, 537 - - Leontodon autumnalis, 536 - - Lepidium sativum, 567 - - Lepinia tahitensis, 378-9 - - Leucæna forsteri, 425, 529 - - Leucas decemdentata, 605 - - Ligustrum vulgare, 537 - - Limnanthemum, =396=, 537 - - Linaria cymbalaria, 537 - - Linaria vulgaris, 537 - - Linden, A., on Pritchardia, 326-7 - - Lindenia, =395=, 532, 603 - - Lindman, C., on the Gulf stream drift of the Scandinavian coast, 180, - 430, 531 - - Linum, 536, 567 - - Lipochæta, 58, 236-7, 239, 243-4, 552-3, 556-7 - - Lister, Mr., 327, 356, 550 - - Lithospermum officinale, 537 - - Litsea, 508; - _see_ Additions and Corrections - - Littoral plants; - distribution in tropics, 68, 516, 562; - causes of the buoyancy of the seeds and fruits, 104, 119, 569; - long flotation experiments, 530; - the littoral plants and the currents of the Pacific, 61; - relation of littoral and inland species of the same genus, 130-70; - the inland extension of beach plants, 34, 40, 49, 59, 121, 547-50, - 559, 568, 579; - Fijian shore plants, 13, 15, 40, 528, 547-50; - Tahitian shore plants, 14, 47, 551; - Hawaiian shore plants, 15, 51, 552-7, 559, 563; - shore plants of west coast of South America, 474-88, 500, 596; - shore plants of the Panama isthmus, 498; - British shore plants, 33-6, 106, 107, 109, 115, 432-3, 540-4; - halophily, 581 - - Lobelia, 252-6, 258, 272-3, 279 - - Lobelia dortmanna, 536 - - Lobeliaceæ, Age of, 250-60, 266-7, 304, 306, 517-20; - arborescent, 250-60; - Hawaiian, 253; - Tahitian and Rarotongan, 256; - affinities with American forms, 251, 267; - relation between the flowers and birds, 504-5, 603; - seed-buoyancy, 533; - capacities of dispersal, 258 - - Locusts as seed-dispersers, 509 - - Loranthus, 377, 381, =383= - - Lotus corniculatus, 536 - - Luffa, =426=, 472, 529, 563, 571 - - Lumnitzera, 43, 69, 108, 435, 437, 529, 551, 563 - - Luzula campestris; - in Hawaii, 272, 286-8; - in Tahiti, 290, 292; - mucosity of the seeds and their dispersal by birds, 286, 417, 568; - seed-buoyancy, 89, 537 - - Lychnis diurna, 536 - - Lycopods, 220-30, 509, 517, 592, 593 - - Lycopsis arvensis, 537 - - Lycopus europæus, 28, 37, 86, 417-8, 537, 545, 568 - - Lyon, Prof., on the Hawaiian rainfall, 213 - - Lysimachia; - in Hawaii, 272, 283-5 - - Lysimachia thyrsiflora, 537 - - Lysimachia vulgaris, 536 - - Lythrum; - in Hawaii, 362 - - Lythrum salicaria, 417-8, 536, 603 - - - - Maba, 371-3, 532-3 - - Macaranga, 381, =384=, 436, 532 - - Macgregor, Sir W., 215 - - Machala plains (Ecuador), 484-5, 495 - - M’Lachlan, Mr., on dragon-flies in ancient drift, 510 - - Magellan Straits, its shore plants, 477 - - Maiden, Mr., on the plants of Pitcairn Island, 345, 355, 418, 562 - - Malayan era of the Pacific floras, 308-9, 330-2, 333, 353, 357, 359, - 519 - - Malva rotundifolia and sylvestris, 536 - - Malvaceæ, the effect of sea-water on the seeds, 544 - - Man in the Pacific, subject to the laws of distribution, 325, 411-2, - 427 - - Manchineel: _see_ Hippomane - - Mandarin orange; - precocious germination of its seeds, 472 - - Mangle chico, 445, 448-9, 498, 501 - - Mangle grande, 445, 448-9, 498, 501 - - Mangrove formation, 9, =43-4=, 47, 50, 53-5, 77, 132, =483-7= - (Ecuador), =551=, 597 - - Marquesas, 208, 529 - - Marsilea, 408 - - Martins, Prof., on the effects of sea-water immersion on seeds, 24, - 538-9, 542-4, 546 - - Mascarene islands; - association of several species of Pandanus with extinct Columbæ, 152, - 157, 169, 517; - linked to the Pacific islands by Afzelia bijuga, 172; - by Acacia heterophylla, 200; - by Naias marina, 368; - by Potamogeton, 369; - by Eugenia, 351; - by Sophora, 148; - by Ochrosia, 153, 580; - by Pandanus, 157-9 - - Mat-names and plant-names in Polynesia, 324, 328 - - Matricaria chamomilla, 536, 568 - - Matricaria inodora; - inland form, 536, 543; - maritime form, 33-4, 109, 116, 536, 540, =543= - - Matricaria maritima: _see_ under M. inodora - - Maui, 207-8 - - Mauna Kea, 207-8, 210-4, 238, 509, 587 - - Mauna Loa, 207-8, 210-4, 238, 509, 587 - - Mauna Loa; - meteorological observations on summit, 582 - - Mauritius: _see_ Mascarene islands - - Maxwell, Dr., on the Hawaiian rainfall, 213 - - Medanos, 482 - - Medicago; - fruits frequent in beach-drift, 429, =431=, 479; - buoyancy of fruits, 431, 536 - - Medinilla, 376 - - Mediterranean beach-drift, 430 - - Megapodes, as probable seed-dispersers, 160, 169, 200, 392 - - Melastoma, 376, 381, =382=, 532, 603 - - Melastoma denticulatum, =382=, 532, 548 - - Melia, 376 - - Meliaceæ, 376 - - Melica nutans, 538 - - Melicope, 263 - - Meliphagidæ (Honey-eaters), as seed-dispersers, 321, 331, 377, 388 - - Melochia, 376 - - Memecylon, 376 - - Mentha aquatica, 28, 537, 545 - - Menyanthes trifoliata, 537 - - Meryta, 381 - - Mesembryanthemum, 131, 478 - - Mesozoic continent in the Western Pacific, 303-6, 503, 509, 514, 519 - - Metrosideros, 259, 333, =341=, 357, 361, 510, 533, 549, 553, 554, 604 - - Meyer, Dr., 388 - - Mez, C., on the genus Embelia, 363 - - Mezoneuron, =146=, 363, 533 - - Micromelum, 393, 532 - - Miers, Mr. J., on the seed-structure of Barringtonia, 575 - - Millett, Mr., on the seed-buoyancy of Convolvulus soldanella, 91, 543, - 566 - - Milner, Sir W., on seeds in petrels, 581 - - Mimosa pudica, 488 - - Mimusops kauki, 374 - - Miquel, on Scirpodendron, 406; - on Carapa, 564 - - Moa, its crop-stones, 159 - - Mœnchia erecta, 536 - - Mollendo (Peru), 482, 598, 599 - - Momordica charantia, 532 - - Montia fontana, 536 - - Moreno Bay (Chile), its beach-drift, 480 - - Morinda; - relation between the shore and inland species, 18, 134, =135=; - inland species, 42, 45, 135, 532, 534 - - Morinda citrifolia; - distribution, 18, 52, 552, 563; - station, 551; - inland extension, 49, 121, 547, 548, 553, 569; - buoyancy of pyrenes, 18, 107, 112, 121, 123, 124, 529, 531, 534, 569; - their occurrence in river-drift, 79, 435 - - Moseley, Prof., 242, 276, 286, 403, 434, 579 - - Mountain bananas, 412, =414=, 427 - - Mountain climates of the Pacific islands, 210-1, 214-5, 218, 582 - - Mountain ferns of the Pacific islands, 225-9, 593 - - Mountain floras of the Pacific islands, 268-306, 518-9 - - Mountain-shadows, 586 - - Mucosity of seeds, 102, 277, 417, 567 - - Mucuna (genus); - abortive germination of seeds in sea-water, 76, 79-82, 202; - long flotation experiments, 80, 81, 531; - cause of seed-buoyancy, 106, 109, 111; - dispersal by currents, 80, 81, 430; - seeds in river-drift, 435, 488; - in beach-drift, 430, 437-8, 489, 499 - - Mucuna gigantea, 44, 81, 94, 109, 115, 529, 531, 552, 554, 562, 563 - - Mucuna urens, 80, 106, 111, 123, 430 499, 531, 533, 534, 562, 563 - - Mueller, Dr. K., 224 - - Mueller, Baron von, 394 - - Musa, 412, 414 - - Musa ensete, 414, 436 - - Mussænda frondosa, 425, 532, 548 - - Mutations, 383 - - Myoporum, 271, 272, 274-5, 343, 533, 553-4 - - Myosotis, 37, 537 - - Myriophyllum, 37, 536 - - Myristica, 391, =402=, 532 - - Myrmecodia, =402=, 472, 532 - - - Nadeaud, Dr. J., on Tahitian plants, 216, 312, 426, 578 - - Naias, 362, 364, =367=, 409, 520, 537, 556; - _see_ Additions and Corrections - - Nama, 362 - - Narthecium, 537 - - Nasturtium, 37, 536 - - Nathorst, Prof., 511 - - Natural Selection, 105, 117, 119-29 - - Nature-Study, 101 - - Nelitris, 381, =382=, 532, 548 - - Nepeta cataria, 28, 537, 567, 568 - - Nepeta glechoma, 28, 537, 567, 568 - - Nephelium, 161, 532 - - Neraudia, 263 - - Nertera depressa, 270, 272, 285, =286=, 288, 290, 292-4, 296-7, 305 - - Nesopanax, 265 - - Nestor meridionalis, its vegetable food, 321, 343 - - New Caledonia, 298, 301-2, 341, 395-6, 420 - - New Zealand birds and seed-dispersal, 296, 321, 337 - - New Zealand flora; - from the standpoint of dispersal, 507-8; - its bearing on the continental theory, 508-9, 512, 514 - - New Zealand plants in the Pacific islands, 271-2, 287, 290, 295, 297, - 305, 315, 336-8, 341, 357, 366, 503-4, 507-9 - - Nicobar-pigeon, as a seed-disperser, 8, 159 - - Nipa fruticans, 66, 68, 108, 472, 550 - - Nolana, 131, 431, 477, 479, 480, =596= - - Norman, J. M., on Scandinavian beach-drift, 430, 432, 539, 543 - - Norwegian beach-drift: _see_ Scandinavia - - Nothocestrum, 262-3 - - Nototrichium, 262-3 - - Nova Zembla; - plant-dispersal, 511 - - Nuphar luteum, 37, 89, 513, 535 - - Nymphæa alba, 37, 89, 513, 535 - - - Oahu, 207-8 - - Oak: _see_ Quercus - - Ochrosia; - relation of coast and inland species, 134, =151=; - dispersal, 152-4; - Schumann’s enumeration of the species, 580 - - Ochrosia parviflora, 49, 108, =152-4=, 530, 563, 580 - - Ochrosia sandwicensis, 153, 580 - - Ocimum basilicum, 568 - - Œnanthe crocata, 28, 536 - - Œnanthe phellandrium, 536 - - Oldenlandia, 605 - - Olea, 364, 508, 533 - - Oliver, Prof., 171, 178, 251 - - Oncocarpus, 265-6; - _see_ Additions and Corrections - - Ononis arvensis, 536 - - Ophiorrhiza, 382, =383=, 532 - - Orchids; - suggested dispersal of their seeds by insects, 509 - - Oregon drift on the Hawaiian coasts, 58, 72, 430, =557= - - Oreobolus, 271, 272, 275 - - Oreodoxa, 489 - - Osmanthus, 364 - - Osteomeles, 353, =354=, 554 - - Owen Stanley range; - climate of the summit, 215 - - Oxalis acetosella, 536 - - Oxalis corniculata, 349, 416, 427, 536, 604 - - Oxyria digyna, 512 - - - Pachyrrhizus, 412, 413, 548 - - Paita (Payta), 482, 599 - - Palms; - specific differentiation in Borneo, 504 - - Panama isthmus; - its shore plants and seed-drift, 180, 498 - - Panax, 263 - - Pandanus, =155=, 508, 517; - relation between coast and inland species, 134, =155-60=, 166, 169, - 517; - mode of dispersal, 157-60, 169; - insular distribution, 156-7, 160, 169, 580 - - Pandanus odoratissimus; - distribution, 52, 53, 156, 552, 563; - station, 551; - inland extension, 41, 42, 361, 548-9, 553-6, 560; - dispersal by currents, 53, 158; - buoyancy of fruits, 109, 530, 552; - their occurrence in river-drift, 435, and in beach-drift, 437, 557-9; - aboriginal food-plant, 413, 427 - - Pangerango, Mount; - rate of decrease of temperature with elevation, 210 - - Panicum, 272, 284 - - Papaver, seed-buoyancy, 535 - - Papaya, 472 - - Paphia, 265 - - Parinarium, 108, 435, 529, 563 - - Parrots, as seed-dispersers, 321, 388, 420, 602 - - Partridges, as seed-dispersers, 284, 356, 367, 416 - - Pastinaca sativa, 28, 536 - - Peach-stones in beach-drift, 429, 431, 479 - - Pea-nuts in beach-drift, 479 - - Peale, Mr., 355 - - Pebble-swallowing by birds, 8, 159 - - Pedicularis palustris, 537 - - Pelagodendron, 265 - - Pelea, 263 - - Pemphis acidula; - station, 42, 43, 551; - distribution, 54, 68, 563; - seed-buoyancy, 108, 114, 529 - - Penzig, Prof., on the beach-drift and flora of Krakatoa, 180, 189, 206, - 571 - - Peperomia, 334, =348=, 357, 417, 509 - - Peplis portula, 536 - - Perkins, Mr., on birds and seeds in Hawaii, 151, 259, 275, 321, 329, - 343, 348, 364, 595; - on the biological connection between the birds and the arborescent - Lobeliaceæ, 255, 504, 603; - on the Hawaiian fauna, 505 - - Perrottetia, 362 - - Peru; - littoral flora, 474-6, 481-2; - on the coast climate and the Humboldt current, 490-4, 500, 598-600; - corals, 496, 601 - - Pes-capræ formation, 550 - - Petrels and seed-dispersal, 242, 511, 581 - - Peucedanum, in Hawaii, 362, 363 - - Peucedanum palustre, 536 - - Phaethon (Tropic-bird) and plant-dispersal, 241 - - Phaseolus truxillensis, 605 - - Philippi, on the shells of Chile, 496 - - Phyllanthus, 309, =325=, 331, 532 - - Phyllostegia, 262, 263, 371, 533, 594 - - Physalis angulata, 605 - - Phytelephas, 489 - - Phytolacca, 362, 364 - - Pigeons, 356, 416, 506; - _see_ Fruit-pigeons - - Pilea, 362 - - Pilger, R., on Podocarpus and Dacrydium, 301-2 - - Pimia, 265, 266 - - Pinguicula lusitanica, 537 - - Pinus sylvestris, 537 - - Piper, 532, 568 - - Pipturus, 356 - - Pisonia, 61, 333, =346=, 357, 568 - - Pistia, 435, 486-9 - - Pitcairn Island, 64, 345, 355, 418, 562 - - Pittosporum, 308, =309=, 313, 509, 532 - - Platydesma, 263 - - Plantago; - on tropical mountains, 269, 270; - Hawaiian species, 271-2; - mode of dispersal, 276; - seed-mucosity, 276, 568; - seed-buoyancy, 537, 540 - - Plantago coronopus, 33 - - Plantago lanceolata, 276, 537, 568 - - Plantago major, 276, 537, 568 - - Plantago maritima, 34, 478, 537, 541, 568 - - Plantago media, 537 - - Plectronia, =355=, 533 - - Plectrophanes nivalis (Snow-bunting), as seed-disperser, 510, 511, 605 - - Pleiosmilax, 371-3, 532 - - Plerandra, 265 - - Plum-stones in beach-drift, 429, 431, 479 - - Poa; - Hawaiian species, 272, 275 - - Poa aquatica, 538 - - Poa fluitans, 538 - - Podocarpus, 294, 297-8, =301=, 306, 508, 603 - - Polycarpon tetraphyllum, 33 - - Polygala vulgaris, 536 - - Polygonum; - dispersal by birds, 356; - buoyancy of fruits, 537, 543 - - Polygonum amphibium, 537 - - Polygonum aviculare, 356, 537 - - Polygonum convolvulus, 356, 537 - - Polygonum glabrum, 354, 356, 435, 487-8 - - Polygonum hydropiper, 37, 537 - - Polygonum lapathifolium, 537 - - Polygonum maritimum, 35, 477, 537, 543 - - Polygonum persicaria, 356, 537 - - Polygonum viviparum, 512 - - Polymorphous species, discussed, 33-4, 353, 357-8, 373, 381, 391, 519, - 520, 522; - independent of insular conditions, 363, 368, 520; - noted under Elæocarpus, 335; - Dodonæa, 339; - Metrosideros, 341; - Alyxia, 345; - Alphitonia, 346; - Pisonia, 346; - Wikstrœmia, 348; - Sicyos, 363, 365; - Naias, 368; - Eurya, 372; - Maba, 372; - Grewia, 382; - Nelitris, 382; - Melastoma, 382; - Loranthus, 383; - Geniostoma, 384; - Macaranga, 384; - Tabernæmontana, 385; - Bischoffia, 386; - Micromelum, 393; - Limnanthemum, 396; - Weinmannia, 291; - Vaccinium, 280-1 - - Polynesians; - their differentiation, 325, 411, 427 - - Polynesian food-plants, 412-4, 427 - - Polynesian weeds, 415, 427, 604 - - Polynesian plant-names, 66, 324, 328, 341, 345, 387, 398, 414, 419, - 421, 424, 425, 441, 578 - - Polypodium, 593 - - Pongamia glabra, 54, 106, 202, 529, 551, 563, 581 - - Pontederia, 435, 486-9 - - Porphyrio (Purple Water-hen), as concerned in plant-dispersal, 296, - 305, 321, 331 - - Portulaca, 532, 552, 553-5 - - Potamogeton, 5, 30, 38, =369=, 513, 537 - - Potentilla; - buoyancy and station, 27 - - Potentilla comarum, 27, 536 - - Potentilla tormentilla, 27, 536 - - Premna; - the genus in the Pacific, =560=; - buoyancy and station, 19, 134, 139; - cause of buoyancy of the fruits of the coast species, 107, 112, 123, - 124, 530, 532, 561, 569; - inland extension of the coast species, 42, 547, 548, 561, 569; - distribution of the species, 561, 563; - occurrence of the fruits in river-drift, 112, 435, 561; - modes of dispersal, 561 - - Prioria copaifera, in Panama beach and river-drift, 499 - - Pritchardia, 124, 308, 309, =326=, 532, 533-4, 554-5 - - Prosopis dulcis, 557 - - Prunella vulgaris, 28, 417, 537, 568 - - Pseudomorus, 371 - - Psittacirostra, 321, 505 - - Psychotria, 308-9, =314=, 331, 391, 532, 603 - - Pteris aquilina, 225, 548 - - Pteropidæ: _see_ Bats - - Pterotropia, 262-3, 595 - - Ptilotus, 263 - - Ptychosperma, 389, 532 - - Puerto Bolivar (Ecuador); - sojourn of the author, 476, 484-8, 494 - - Pumice in beach-drift, 429, 558, 601 - - Puna Island (Ecuador), 489, 494 - - Puna coast (Hawaii), 553 - - - Quercus, 61, 90, 126, 429, 431, 537, 538, =571= - - - Radiola millegrana, 417, 418 - - Raiatea, 250, 252, 257 - - Raillardella, 237 - - Raillardia, 236-8, 240, 243-4 - - Ranunculaceæ, 535, 544 - - Ranunculus; - on tropical mountains, 269, 272; - Hawaiian species, 272-3; - dispersal by birds, 277, 511; - buoyancy of fruits, 535 - - Ranunculus aquatilis, 37, 535 - - Ranunculus repens, 37, 86, 535, 545 - - Ranunculus sceleratus, 37, 535, 544, 545 - - Ranunculus; - other British species, 535 - - Raphanus, 33, 478, 536, 540, 596 - - Rarotonga; - its flora, 48, 177, 291, 295, 309, 317, 320, 323, 336, 374, 419, 425, - 551, 578; - altitude, 208; - rainfall, 216; - absence of mangroves, 50; - ferns, 221-2; - endemic species of flowering plants, 232; - the age of arborescent Compositæ and Lobeliaceæ represented by - Fitchia, 237-8; - and by Sclerotheca, 250, 252, 256-7; - scanty representation of Tahitian mountain plants, 293; - a connection with the Kermadec islands, 295 - - Rauwolfia, 362 - - Reinecke, Dr., on the Samoan flora, 19, 232, 266, 291, 317, 387, 577, - 579; - on Elatostema, 405 - - Remya, 236-7, 243-4 - - Rendle, Mr., on Naias, 368 - - Reseda luteola, 536 - - “Revue Scientifique,” 506 - - Rewa River (Fiji); - seed-drift, 76, 78, =435=; - temperature, 78, 564 - - Reynoldsia, 309, =310=, 371 - - Rhaphidophora, 404, 532 - - Rhinanthus crista galli, 537, 545 - - Rhizophora; - general discussion, =440=, 520-1; - dispersal by currents, 48, 77, 94, =458=; - the “Selala” or seedless form in Fiji, 443; - its representative in Ecuador, 449, 487, 521; - the genus in Ecuador, 483-7; - in Panama, 498-9; - absence from Hawaii, 54-5, and Tahiti, 47; - germination and growth of seedling, =451=, =468=, 575; - river-drift, 435, 460, 499; - beach-drift, 437-8, 460, 499; - polyembryony, 449; - distribution, 54-5, 69, 520 - - Rhizophora mangle; - in Fiji, 43, 441, 520; - in Ecuador, 484-7; - in Panama, 498-9; - discussed in detail in Chapter XXX - - Rhizophora mucronata, 43, 68; - discussed in detail in Chapter XXX - - Rhus, 353, 354 - - Rhynchospora, 272, 283-5 - - Richella, 265 - - Ricinus communis, 142, 533, 558 - - Ridley, Mr. H. N., on the plants of Fernando Noronha and of the Malay - peninsula, 8, 144, 162, 319, 340, 386, 603 - - River seed-drift; - Thames, 37, 85, 91, 430; - Rewa (Fiji), 76, 91, =435-6=; - Guayaquil or Guayas River (Ecuador), 77, 91, 435, =488-9=; - Panama isthmus, 499; - germination in river-drift, 76-8, 84, 85-6, 435, 488-9 - - River temperature, 78, 564 - - Robinson, Mr. E. Kay, on the dispersal of Aster tripolium, 605 - - Rodriguez Island, 157, 351; - _see_ Mascarene Islands - - Rœmeria hybrida, 535 - - Rollandia, 252, 255, 258 - - Rosa arvensis, 536 - - Roxburgh, W., on the seed-structure of Barringtonia, 575 - - Rubiaceæ; - appearance of the order in the Pacific islands, 261, 262 - - Rubus, in Hawaii, 269, 272-4, 285, 533, 604 - - Rumex; - in Hawaii, 366; - modes of dispersal, 367; - fruits in Thames drift, 37, 86, 430; - buoyancy of the fruits of British species, 537, 545 - - Ruppia maritima, 371, 372, =374=, 398, 482, 537, 555 - - Russell, Prof., 493 - - Ruwenzori, Mount, 241, 251 - - - Sachs, Prof., 100 - - Sadleria, 593 - - Sage-brush, 279 - - Sagina procumbens, 536 - - Sagittaria sagittifolia, 38, 537 - - Sagus vitiensis, 413 - - St. Helena, 8, 258 - - Salicornia; - mode of dispersal, 482, 489, 541, 545 - - Salicornia herbacea, 34, 35, 537, 541, 545 - - Salicornia peruviana in Ecuador, 77, 484-6, 489 - - Salicornia; - other species in Chile and Peru, 478, 482 - - Saline deposits of North Chile, suggested origin, 485 - - Salsola kali, 10, 34-6, 429, 431, 477, 479, 537, =541= - - Salvia verbenaca, 28, 537, 567-8 - - Sambucus nigra, 536 - - Salvinia, 488 - - Samoa; - altitude, 208-9; - few peculiar genera, 266; - proportion of peculiar species of flowering plants, 232; - littoral flora, 46; - mountain flora, 290, 297, 305; - peculiar species of - Pandanus, 156, 580, - Elatostema, 405, - Eugenia, 349. - Amongst other genera possessing peculiar species are - Gardenia, 311, - Psychotria, 315, - Cyrtandra, 317, - Macaranga, 384, - Ficus, 387 - - Samolus valerandi, 34, 537, 541 - - Samolus, in Chile, 478 - - Sanicula, in Hawaii, 4, 269, 270, 272, =273=, 274-5 - - San Lorenzo Island (Peru) in the coast-clouds, 492; - decaying shells, 497 - - Santa Elena Point (Ecuador), its vegetation and climate, 490, 494, 597 - - Santa Rosa River (Ecuador), the vegetation of its banks, 486 - - Santalum, 272, 283-4, 285 - - Sapindus, 309, =325=, 332 - - Sapota, 373, 532 - - Sapotaceæ, 372-4 - - Sararanga, 156 - - Saxifraga, British species, 536 - - Saxifragaceæ, in Hawaii, 263 - - Scævola; - relation between littoral and inland species, 18, 134, =135=; - Hawaiian inland species, 18, 135, 533; - Fijian inland species, 18, 532, 576; - distribution, 67, 71, 564; - modes of dispersal, 135, 564; - buoyancy of fruits, 18, 135, 531-3 - - Scævola chamissoniana, Hawaii, 18, 533 - - Scævola gaudichaudii, Hawaii, 18, 533 - - Scævola floribunda, Fiji, 18, 532, 576 - - Scævola koenigii; - distribution, 56, 135, 552, 563; - station at coast, 18, 43, 551, 553-6, 560; - extension inland, 42, 59, 121, 547, 553, 560, 569, 579; - modes of dispersal, 57, 71, 552; - buoyancy of fruits, 108, 114, 121, 122, 529, 531, 569; - their occurrence in beach-drift, 437, 559; - in river-drift, 79; - variety, 579; - synonymy, 71, 564 - - Scævola lobelia, 71 - - Scævola sericea, 579 - - Scandinavia; - Sernander’s “Dispersal-biology,” 24; - Atlantic or Gulf-stream drift, 180,189, =430=, 570, 601 - - Schenck, Dr., 512 - - Scheuchzeria palustris, 537 - - Schiedea, 262-3 - - Schimper, Prof.; - relation between littoral and inland species of a genus and between - the littoral and inland floras, 17, 19, 130-4, 534; - grouping of Malayan littoral plants, 43, 550; - distribution and dispersal of tropical littoral plants, 62, 69; - the littoral Leguminosæ, 201, 581; - structure of the buoyant seeds and fruits of tropical plants and the - question of adaptation, 104, 112-5, 119-29, 569; - on Rhizophora, 441, 446, 453-4, 459; - the essential climatic conditions for mangrove-growth, 470; - the xerophytes and hygrophytes, 32, 40; - epiphytic habit of Vaccinium, 281; - miscellanea, 423, 561, 571, 577, 602; - letters to the author, 43, 121, 440-1, 454; - his indebtedness to Prof. Schimper’s work, 63 - - Schizæa, 593 - - Schizostege, 593 - - Schmidt, J., on Lathyrus maritimus, 543 - - Schönland, S., on the Hawaiian Lobeliaceæ, 255 - - Schumann, K., on Ochrosia, 153, 580; - on Musa, 414, 436 - - “Science Gossip,” 23, 277, 369, 398, 513, 567 - - Scilla nutans, 537 - - Scirpodendron costatum, 47, 108, =405=, 435, 530, 551, 552, 563 - - Scirpus; - buoyancy of the fruits of seven British species, 537; - dispersal by ducks implied in connection with the Cyperaceæ, 513 - - Scirpus maritimus, 34, 92, 109, 116, 537, 541 - - Scirpus palustris, 90, 537 - - Scleria; - dispersal by purple water-hens, 296 - - Sclerotheca, 252, 256-8 - - Scott, J., on Indian parrots, 420 - - Scott-Elliot, Mr., 568 - - Scrophularia aquatica, 89, 537 - - Scrophularia nodosa, 537 - - Scutellaria galericulata, 28, 37, 537, 545 - - Scyphiphora, 109 - - Sea, Fijian tree, 436, 602 - - Sea-aster: _see_ Aster tripolium - - Sea-birds, as seed-dispersers, 188, 241, 242, 347, 506, =510-1=, 514, - 605 - - Sea-eagle, as a seed-disperser, 286 - - Sea-thrifts: _see_ Armeria - - Seed-mucosity: _see_ Mucosity - - Seed-stage, regarded as an adaptation, 11, 85-7, 468, 473, 521 - - Seed-structure; - anomalies connected with a lost viviparous habit, 79, 132, 470, 473, - 521, 575 - - Seemann, Dr., on the Fijian flora, 172, 177, 231, 264, 421, 442, 549, - 576, 577, 579, 592; - on the northern limit of mangroves, 54; - on Polynesian weeds, 415, 604; - on Laguncularia in the Panama isthmus, 498; - on willow-leaved plants, 603 - - Selala, the seedless Rhizophora of Fiji, 441-9, 465-6, 487, 520-1; - _see_ Rhizophora - - Selliera radicans, 477 - - Senecio, in Hawaii, 362 - - Senecio aquaticus, 28, 536, 568 - - Senecio candidans, 477 - - Senecio palustris, 536 - - Senecio vulgaris, 536, 568 - - Serianthes, 108, 389, =424=, 529, 563 - - Sernander, Dr., on the dispersal-biology of Scandinavia, 24; - on the Gulf-stream drift, 180, 189, 430, 570, 601; - miscellanea, 280, 511, 538, 542-3, 571 - - Sesbania, 528, 555 - - Sesuvium zone of west coast of South America, 476, 481 - - Sesuvium portulacastrum; - in Fiji, 529; - in Tahiti, 370, 528; - in Tonga, 528; - in Hawaii, 375, 552, 554-5; - in Ecuador, 485; - in Peru, 482; - seed-buoyancy, 16, 529 - - Shaddock (Citrus decumana), 125-6, 532-3 - - Shadow of mountains, 586 - - Sibthorpia europæa, 417, 418 - - Sicilian beach-drift, 430 - - Sicyos, 362, 363, =365= - - Sideroxylon, 371-4, 507-8 - - Sida, 417, 427, 533, 555, 604 - - Siegesbeckia orientalis, 605 - - Silene, in Hawaii, 272, 279-80 - - Silene cucubalus, 536 - - Silene maritima, 33, 36, 280, 511, 536, 540, 544 - - Simpson, Mr. M.; - precocious germination of the coco-nut, 472 - - Siphocampylus, 251 - - Sisyrinchium, 272, 273, 279, 533 - - Sium; - buoyancy of the fruits of British species, 536 - - Smith, Mr. Jared G., 211, 213 - - Smyrnium olusatrum, 536 - - Smythea pacifica; - distribution, 47, 264-5, 551, 562; - buoyancy of fruits, 106, 112, 529; - their occurrence in river-drift, 435 - - Snow-buntings (Plectrophanes nivalis), as seed-dispersers, 510-1, 605 - - Solanum aculeatissimum, 533 - - Solanum dulcamara, 537 - - Solanum nigrum (and var. oleraceum), 417, 537, 605 - - Solereder, H., 265 - - Solitaire, 8, 159 - - Solomon Islands, 351, 394, 400 - - Sonchus asper, 241, 605 - - Sonchus oleraceus, 536, 593 - - Sonneratia, 69, 108 - - Sophora; - general discussion, =147=; - relation between littoral and inland species, 19, 134, =147=, 165, - 200; - its indication in New Zealand, 508 - - Sophora chrysophylla, 20, =147-51=, 271-2, 278, 533, 580 - - Sophora tetraptera, 64, =148-9=, 271, 431, 478-9, =580= - - Sophora tomentosa; - distribution, 19, 54, 68, 147-8, 563; - station and habit, 201-2, 551, 581; - seed-buoyancy, 107, 113, 529, 531, 579; - seeds in beach-drift, 437-8 - - Soulamea, 265 - - South America; - observations on the littoral flora of the west coast, 474-501; - _see_ Chile, Peru, Ecuador - - Sparganium, 5, 31, 430, 513, 537, 540, 545 - - Species, their development in the Pacific islands; - the views of Hillebrand, 226-7; - the effect of the greater elevation of the Hawaiian islands on the - endemism (production of new species) of the ferns, 227-8; - tables showing the development of new species and new genera of the - flowering plants of the Pacific islands, 232-4, 252, 255, 263, - 265; - the polymorphism of genera, as indicated by their fecundity in - species, illustrated by Cyrtandra, 317; - by Elatostema, 317, 405; - and by Psychotria, 315, 391; - the polymorphism of species, as indicated by their great variability, - 520, 522; - _see_ Polymorphous species; - the question whether the relative fecundity of two genera like - Psychotria and Coprosma is to be connected with difference in - antiquity or with difference in their geographical position, 315, - 331; - the connection between endemism and the suspension of means of - dispersal by birds, 5, 7, 229; - endemism thus produced nearly as pronounced with certain genera like - Cyrtandra, Embelia, and Naias, in continental regions as in - oceanic islands, 229, 317-3, 331, 363-4, 368, 409, 520; - the process concerned in endemism favoured by the deterioration of - plants in their capacity for dispersal, 262-3, 337, 350, 365, 507, - 594-5; - biological connection between plants and birds, 504-5; - differentiation of climate, bird, and plant, the bird being dependent - on the climate, and the plant on the bird, 378, 410, 506-7, 514, - 521-2 - - Spergula arvensis, 536 - - Spergularia rubra and S. marina, 33, 36, 536, 540, 544-5 - - Sphacele, 362 - - Spiræa ulmaria, 37, 92, 536 - - Spitzbergen; - flora and plant dispersal, 511-2, 514 - - Spondias dulcis, 532 - - Spondias lutea, 124, 438, 489, 499 - - Spondias, unidentified Fijian species, 602 - - Spring-centres for the diffusion of aquatic plants, 396, 399 - - Stachys, 27, 28, 537 - - Stapf, Dr., on the flora of Kinabalu, 101, 296, 301 - - Statice, 34 - - Station and seed-buoyancy, 16, 24, 30, 515 - - Stellaria, 37, 89, 536 - - Stenogyne, 262-3, 594-5 - - Sterculia, 11, 375-6, =391=, 532-3 - - Sterculiaceæ, 10, 375 - - Straussia, 262-3, 364 - - Stromboli, its shadow, 586 - - Strongylodon, 80, =82=, 107, 113, 123, =200=, 436, 529, 531, 558, - 562-3, =565= - - Struthious birds, suggested as seed-dispersers, 152, 159-60 - - Stubbs, Dr.; - Agricultural Report on Hawaii, 212, 578 - - Stylocoryne, 532 - - Suæda; - concerning dispersal by birds, 511 - - Suæda fruticosa, 34, 480, 482, 537, 541 - - Suæda maritima, 34, 537, 541 - - Suess, Prof., on the shell-fauna of Chile, 496; - on the salts of ancient seas, 566 - - Sula: _see_ Boobies - - Suriana maritima, 105, 528, 529, 563 - - Swamp-hens: _see_ Porphyrio - - Symphytum officinale, 537 - - - Tabernæmontana, 381, =385=, 532 - - Tacca; - relation between littoral and inland species, 19, 134, =138= - - Tacca maculata, 19, 138, 532, 548 - - Tacca pinnatifida, =19=, =138=; - distribution, 138, 552, 563; - modes of dispersal, 138; - seed-buoyancy, 19, 108, 530, 531, 569; - station, 138; - growing inland, 42, 138, 547, 548, 569; - as a food-plant, 138, 412, 414, 427 - - Tahiti; - area and altitude, 207-8; - temperature, 209; - rainfall, 216, 218 - - Tahitian flora; - littoral plants, 14, =47=, 528, =529=, 551, 552; - ferns and lycopods, 220-230, 592; - endemic genera and endemic species of flowering plants, 231-5; - the age of Compositæ, 236-40, 245, 248; - and of Lobeliaceæ, 250-2, 256-8, 266; - the mountain plants, 269, 290-3, 305, 518-9 - The age of Malayan plants; - the general dispersal of these plants in the Pacific, 307-58; - the local dispersal, - (_a_) genera common to Tahiti and Hawaii, but not found in - Fiji, 359, 370; - (_b_) genera found in Tahiti and Fiji, but not in Hawaii, 359, - 380-8; - (_c_) residual genera (found only in Tahiti), 359, 378. - The absentees from Tahiti, 360, 388 - American genera in Tahiti, 379, 380 - - Talasinga plains: _see_ Fiji - - Tamus communis, 537 - - Tarawau tree of Fiji: _see_ Dracontomelon - - Taraxacum, 536 - - Taro, 412, 415 - - Taubert, P., on Strongylodon, 566 - - Taviuni, 207-8 - - Taxus baccata, 537 - - Teesdalia, 567 - - Temperature: _see_ under Beach-temperature, Estuaries, Humboldt current - - Tephrosia piscatoria; - in Fiji, 42, 45, 547; - in Hawaii, 56, 58, 59, 552, 553-6; - buoyancy of seed and pod, 529, 562; - modes of dispersal, 45, 150, 417, 562 - - Terminalia; - relation between coast and inland species, =17=, 120, 533; - buoyancy, 108, 114 - - Terminalia katappa; - distribution, 52, 54, 551, 552, 563; - buoyancy of fruits, 108, 114, 529; - their occurrence in beach-drift, 437, 559 - - Terminalia litorea or littoralis, 108, 529, 563 - - Tertiary submergence of the Western Pacific archipelagoes, 245, 247, - 249, 260, 267, =304-5=, 306, 503, 518, 520 - - Tetramolopium, 236-7, 243-4 - - Tetraonidæ (Grouse-family), as seed-dispersers, 279, 282, 511, 514 - - Teucrium inflatum, 416, 605 - - Teucrium scorodonia, 28, 537 - - Thacombauia, 265 - - Thalictrum flavum, 535 - - Thames; - the vegetation of the banks, 37; - the seed-drift, 37-8, 85-6, 430, 539; - _see_ Additions and Corrections - - Thauziès, M. A., 506 - - Thespesia populnea, 562; - in Hawaii, 52, 552, 554-6, 558-9; - in Fiji, 547, 550; - seed-buoyancy, 105, 529, 530-1; - beach-drift, 558-9 - - Thlaspi arvense, 536 - - Thomson, Mr. G. M., on the New Zealand flora, 508 - - Thomson, Dr. T., on the seed-structure of Barringtonia, 575 - - Thuret, M., on the buoyancy of seeds and fruits, 24, 63, 538, 542, 544, - 546 - - Thymus, 28, 537, 568 - - Tibet; - its flora, 34, 238 - - Tillandsia, 484, 486 - - Tonga; - included in the Fijian area, 207; - the littoral plants, 46; - the proportion of endemic species, 232; - absence of peculiar genera whether of the Compositæ or Lobeliaceæ or - of any other order, 235, 266; - several peculiar species of Eugenia, 349; - a home for the Sapotaceæ, 374; - Pritchardia pacifica thrives, 327; - vegetation of the leeward plains, 550. - Amongst other genera possessing peculiar species, or in other - respects remarkable, are Podocarpus, 301; - Pittosporum, 309; - Freycinetia, 320; - Elæocarpus, 335; - &c. - - Touchardia, 263, 594 - - Tournefortia argentea; - distribution, 43, 64, 563; - station, 42, 43, 551; - buoyancy of fruits, 108, 114, 530, 531; - their occurrence in beach-drift, 437 - - Tragopogon, fruit-buoyancy of British species, 536 - - Tree-Lobelias, 250-60, 266-7; - _see_ Lobeliaceæ - - Treub, Dr., on the new flora of Krakatoa, 206, 221, 577 - - Trevesia, 309, =310=, 371 - - Tribulus cistoides, 56, 58, 365, 366, 552, 553-6 - - Trichospermum, 265, =392=, 532 - - Trifolium, 150, 536 - - Triglochin, 34, 36, 537, 541, 545 - - Trimenia, 265 - - Triplasandra, 262-3 - - Trisetum, 272, 275 - - Tristan da Cunha, 242, 276, 286, 364, 366 - - Triumfetta procumbens, 42, 43, 45, 529, 550 - - Triumfetta rhomboidea, 529 - - Tropic-bird (Phaethon), 241 - - Tropical beach-drift, 434-9; - _see_ Beach-drift - - Tschudi, Dr., 493 - - Tupa, 251, 258 - - Tussilago farfara, 536, 546 - - Tussilago petasites, 536 - - Typha; - seed-buoyancy, 537 - - - Ulex europæus, 536 - - Ulmus campestris, 537 - - Umbelliferæ; - station and fruit-buoyancy, 28, 537 - - Uncinia, 271-5 - - Urena lobata, 417, 427, 532, 604 - - Urera, 362 - - Urtica dioica, 537 - - Urticaceæ; - represented in the early flora of Hawaii, 261, 263 - - - Vaccinium, 4, 5, 269, 270, 272, 274-5, =280-2=, 290-4, 297, 305, 343 - - Valerianella, 536 - - Vallesia, 154 - - Valparaiso; - beach-plants and beach-drift, 477-9 - - Vandellia crustacea, 605 - - Van Tieghem, on Brackenridgea, 570 - - Vanua Levu; - area, altitude, and rainfall, 207-8, 216 - - Varigny, Dr. H. de, 506 - - Vascular cryptogams in the Pacific islands, 222, 592; - _see_ also under Ferns and Lycopods - - Veitchia, 401, 532 - - Verbena officinalis, 537 - - Veronica; - buoyancy of the fruits of British species, 537 - - Veronica beccabunga, 37, 537, 568 - - Vicia; - in the Hawaiian mountain-flora, 272, 278 - - Vicia faba; - dispersed by pigeons, 150, 417 - - Vicia sativa; - seed-buoyancy, 536 - - Victoria Institute, 66 - - Vigna; - relation between littoral and inland species, 134, 139 - - Vigna lutea; - in Hawaii, 56-7, 139, 552, 554, 558-9, 560; - in Fiji, 550; - distribution, 68, 563; - seeds dispersed by currents, 57; - seed-buoyancy, 56, 106, 529; - seeds in beach-drift, 437-8, 558-9; - seeds in river-drift, 435, 489 (species in last case unidentified) - - Viola; - in Hawaiian mountain-flora, 253, 272; - modes of dispersal, 277; - seed-buoyancy, 533, 536; - seed-mucosity, 278, 567-8 - - Viscum, 355, 358, 377, 536 - - Vitex; - relation between littoral and inland species, 134, 137 - - Vitex trifolia; - station and distribution, 50, 551-2, 563-4; - growing inland, 42, 547-8, 560; - dispersal by currents, 56, 57, 564; - buoyancy of fruits, 108, 122, 530; - their occurrence in river-drift, 435, and in beach-drift, 559; - dispersal by birds, 57, 122, 564; - _see_ Additions and Corrections - Variety, unifoliolata, 108, 547-8, 552, 556, 559, 560 - - Vitex agnus castus, 109, 431 - - Viti Levu, 207-8 - - Vivipary (germination on the plant), 78, 84-7, 132, 191, =468=, =521=, - 564, 574-6; - a lost habit with many plants and only indicated by anomalies in - seed-structure, 79, 132, 470, 473, 521, 575 - - Vries, Prof. H. de, 573 - - - Walker, Mr. F. P., 490, 495, 597 - - Wallace, Mr. A. R., 7, 242, 247, 259, 400 - - Wallis Island, 349, 551 - - Walsh, Canon, on the Cordyline of the Maoris, 420 - - Waltheria americana, 375, 416, 427, 533, 553-4, 604 - - Waltheria pyrolæfolia, 375 - - Warburg, Dr., 54; - on Pandanus, 155-8, 580; - on Freycinetia, 319-22, 510; - on the Samoan species of Ficus, 387 - - Warming, E., on Rhizophora, 450, 453 - - Webster, Mr. H., on Ecuador, 495 - - Wedelia, 108, 116, 529, 551, 563 - - Weed, Prof., 277, 364 - - Weeds of Polynesia, 415, 427, 604 - - Weinmannia, =290-1=, 292-4, 297, 305 - - Whymper, Mr., on insects at great altitudes, 510, 583 - - Wichmann, A., on the submergence of the Western Pacific islands, 304 - - Wiglesworth, Mr., on Polynesian birds, 66, 296 - - Wikstrœmia, 45, 333, =348=, 357, 530 - - Wilkes, Commodore; - meteorological observations on the summit of Mauna Loa, 210, 583, - 585, =586= - - Wilkesia, 236-7, 240, 243-4 - - Willow-leaved river-side plants, 395, 603 - - Wilson, Mr. S. B., on the vegetable food of Hawaiian birds, 151, 321 - - Winds; - in plant-dispersal, 226, 259, 511; - on the summit of Mauna Loa, 583; - _see_ Drying-winds - - Wolff, Dr., on Ecuador, 476, 486, 494-5 - - Wolffia, found by the author in Fiji, 408 - - Wollastonia, 109 - - Woodford, Mr. C. M., 66 - - - Xerophytes, 32, 39, 201, 515 - - Ximenia americana, 107, 113, 115, 122, 128, 529, 563 - - - Yams, 412-4 - - - Zannichellia palustris, 537 - - Zippelius, on Scirpodendron, 406 - - - - - RICHARD CLAY AND SONS, LIMITED - BREAD STREET HILL, E.C., AND - BUNGAY, SUFFOLK. - - - - - Transcriber’s Note - -Some corrections have been made to the original text. In particular, -punctuation has been normalized and ditto marks have been replaced by -the text they represent. Scale bars have been added to the illustrations -where a scale is indicated. Corrections listed in Additions and -Corrections have been made in cases where words or phrases are to be -substituted or removed. The page numbers in the Additions and -Corrections have been corrected and the order has been adjusted -accordingly. - -Further corrections are as follows: - - p. 48 seseedlings will float uninjured -> seedlings will float uninjured - p. 467 in the case of Bruguiera rheedi -> in the case of Bruguiera - rheedii - - - - - -End of the Project Gutenberg EBook of Observations of a Naturalist in the -Pacific Between 1896 and 1899, v. 2, by H. B. Guppy - -*** END OF THIS PROJECT GUTENBERG EBOOK OBSERVATIONS OF A NATURALIST *** - -***** This file should be named 54889-0.txt or 54889-0.zip ***** -This and all associated files of various formats will be found in: - http://www.gutenberg.org/5/4/8/8/54889/ - -Produced by Greg Bergquist, Sonya Schermann and the Online -Distributed Proofreading Team at http://www.pgdp.net (This -file was produced from images generously made available -by The Internet Archive/American Libraries.) - - -Updated editions will replace the previous one--the old editions -will be renamed. - -Creating the works from public domain print editions means that no -one owns a United States copyright in these works, so the Foundation -(and you!) can copy and distribute it in the United States without -permission and without paying copyright royalties. Special rules, -set forth in the General Terms of Use part of this license, apply to -copying and distributing Project Gutenberg-tm electronic works to -protect the PROJECT GUTENBERG-tm concept and trademark. Project -Gutenberg is a registered trademark, and may not be used if you -charge for the eBooks, unless you receive specific permission. If you -do not charge anything for copies of this eBook, complying with the -rules is very easy. You may use this eBook for nearly any purpose -such as creation of derivative works, reports, performances and -research. They may be modified and printed and given away--you may do -practically ANYTHING with public domain eBooks. Redistribution is -subject to the trademark license, especially commercial -redistribution. - - - -*** START: FULL LICENSE *** - -THE FULL PROJECT GUTENBERG LICENSE -PLEASE READ THIS BEFORE YOU DISTRIBUTE OR USE THIS WORK - -To protect the Project Gutenberg-tm mission of promoting the free -distribution of electronic works, by using or distributing this work -(or any other work associated in any way with the phrase "Project -Gutenberg"), you agree to comply with all the terms of the Full Project -Gutenberg-tm License (available with this file or online at -http://gutenberg.org/license). - - -Section 1. General Terms of Use and Redistributing Project Gutenberg-tm -electronic works - -1.A. By reading or using any part of this Project Gutenberg-tm -electronic work, you indicate that you have read, understand, agree to -and accept all the terms of this license and intellectual property -(trademark/copyright) agreement. If you do not agree to abide by all -the terms of this agreement, you must cease using and return or destroy -all copies of Project Gutenberg-tm electronic works in your possession. -If you paid a fee for obtaining a copy of or access to a Project -Gutenberg-tm electronic work and you do not agree to be bound by the -terms of this agreement, you may obtain a refund from the person or -entity to whom you paid the fee as set forth in paragraph 1.E.8. - -1.B. "Project Gutenberg" is a registered trademark. It may only be -used on or associated in any way with an electronic work by people who -agree to be bound by the terms of this agreement. There are a few -things that you can do with most Project Gutenberg-tm electronic works -even without complying with the full terms of this agreement. See -paragraph 1.C below. There are a lot of things you can do with Project -Gutenberg-tm electronic works if you follow the terms of this agreement -and help preserve free future access to Project Gutenberg-tm electronic -works. See paragraph 1.E below. - -1.C. The Project Gutenberg Literary Archive Foundation ("the Foundation" -or PGLAF), owns a compilation copyright in the collection of Project -Gutenberg-tm electronic works. Nearly all the individual works in the -collection are in the public domain in the United States. If an -individual work is in the public domain in the United States and you are -located in the United States, we do not claim a right to prevent you from -copying, distributing, performing, displaying or creating derivative -works based on the work as long as all references to Project Gutenberg -are removed. Of course, we hope that you will support the Project -Gutenberg-tm mission of promoting free access to electronic works by -freely sharing Project Gutenberg-tm works in compliance with the terms of -this agreement for keeping the Project Gutenberg-tm name associated with -the work. You can easily comply with the terms of this agreement by -keeping this work in the same format with its attached full Project -Gutenberg-tm License when you share it without charge with others. - -1.D. The copyright laws of the place where you are located also govern -what you can do with this work. Copyright laws in most countries are in -a constant state of change. If you are outside the United States, check -the laws of your country in addition to the terms of this agreement -before downloading, copying, displaying, performing, distributing or -creating derivative works based on this work or any other Project -Gutenberg-tm work. The Foundation makes no representations concerning -the copyright status of any work in any country outside the United -States. - -1.E. Unless you have removed all references to Project Gutenberg: - -1.E.1. The following sentence, with active links to, or other immediate -access to, the full Project Gutenberg-tm License must appear prominently -whenever any copy of a Project Gutenberg-tm work (any work on which the -phrase "Project Gutenberg" appears, or with which the phrase "Project -Gutenberg" is associated) is accessed, displayed, performed, viewed, -copied or distributed: - -This eBook is for the use of anyone anywhere at no cost and with -almost no restrictions whatsoever. You may copy it, give it away or -re-use it under the terms of the Project Gutenberg License included -with this eBook or online at www.gutenberg.org/license - -1.E.2. If an individual Project Gutenberg-tm electronic work is derived -from the public domain (does not contain a notice indicating that it is -posted with permission of the copyright holder), the work can be copied -and distributed to anyone in the United States without paying any fees -or charges. If you are redistributing or providing access to a work -with the phrase "Project Gutenberg" associated with or appearing on the -work, you must comply either with the requirements of paragraphs 1.E.1 -through 1.E.7 or obtain permission for the use of the work and the -Project Gutenberg-tm trademark as set forth in paragraphs 1.E.8 or -1.E.9. - -1.E.3. If an individual Project Gutenberg-tm electronic work is posted -with the permission of the copyright holder, your use and distribution -must comply with both paragraphs 1.E.1 through 1.E.7 and any additional -terms imposed by the copyright holder. Additional terms will be linked -to the Project Gutenberg-tm License for all works posted with the -permission of the copyright holder found at the beginning of this work. - -1.E.4. Do not unlink or detach or remove the full Project Gutenberg-tm -License terms from this work, or any files containing a part of this -work or any other work associated with Project Gutenberg-tm. - -1.E.5. Do not copy, display, perform, distribute or redistribute this -electronic work, or any part of this electronic work, without -prominently displaying the sentence set forth in paragraph 1.E.1 with -active links or immediate access to the full terms of the Project -Gutenberg-tm License. - -1.E.6. You may convert to and distribute this work in any binary, -compressed, marked up, nonproprietary or proprietary form, including any -word processing or hypertext form. However, if you provide access to or -distribute copies of a Project Gutenberg-tm work in a format other than -"Plain Vanilla ASCII" or other format used in the official version -posted on the official Project Gutenberg-tm web site (www.gutenberg.org), -you must, at no additional cost, fee or expense to the user, provide a -copy, a means of exporting a copy, or a means of obtaining a copy upon -request, of the work in its original "Plain Vanilla ASCII" or other -form. Any alternate format must include the full Project Gutenberg-tm -License as specified in paragraph 1.E.1. - -1.E.7. Do not charge a fee for access to, viewing, displaying, -performing, copying or distributing any Project Gutenberg-tm works -unless you comply with paragraph 1.E.8 or 1.E.9. - -1.E.8. You may charge a reasonable fee for copies of or providing -access to or distributing Project Gutenberg-tm electronic works provided -that - -- You pay a royalty fee of 20% of the gross profits you derive from - the use of Project Gutenberg-tm works calculated using the method - you already use to calculate your applicable taxes. The fee is - owed to the owner of the Project Gutenberg-tm trademark, but he - has agreed to donate royalties under this paragraph to the - Project Gutenberg Literary Archive Foundation. Royalty payments - must be paid within 60 days following each date on which you - prepare (or are legally required to prepare) your periodic tax - returns. Royalty payments should be clearly marked as such and - sent to the Project Gutenberg Literary Archive Foundation at the - address specified in Section 4, "Information about donations to - the Project Gutenberg Literary Archive Foundation." - -- You provide a full refund of any money paid by a user who notifies - you in writing (or by e-mail) within 30 days of receipt that s/he - does not agree to the terms of the full Project Gutenberg-tm - License. You must require such a user to return or - destroy all copies of the works possessed in a physical medium - and discontinue all use of and all access to other copies of - Project Gutenberg-tm works. - -- You provide, in accordance with paragraph 1.F.3, a full refund of any - money paid for a work or a replacement copy, if a defect in the - electronic work is discovered and reported to you within 90 days - of receipt of the work. - -- You comply with all other terms of this agreement for free - distribution of Project Gutenberg-tm works. - -1.E.9. If you wish to charge a fee or distribute a Project Gutenberg-tm -electronic work or group of works on different terms than are set -forth in this agreement, you must obtain permission in writing from -both the Project Gutenberg Literary Archive Foundation and Michael -Hart, the owner of the Project Gutenberg-tm trademark. Contact the -Foundation as set forth in Section 3 below. - -1.F. - -1.F.1. Project Gutenberg volunteers and employees expend considerable -effort to identify, do copyright research on, transcribe and proofread -public domain works in creating the Project Gutenberg-tm -collection. Despite these efforts, Project Gutenberg-tm electronic -works, and the medium on which they may be stored, may contain -"Defects," such as, but not limited to, incomplete, inaccurate or -corrupt data, transcription errors, a copyright or other intellectual -property infringement, a defective or damaged disk or other medium, a -computer virus, or computer codes that damage or cannot be read by -your equipment. - -1.F.2. LIMITED WARRANTY, DISCLAIMER OF DAMAGES - Except for the "Right -of Replacement or Refund" described in paragraph 1.F.3, the Project -Gutenberg Literary Archive Foundation, the owner of the Project -Gutenberg-tm trademark, and any other party distributing a Project -Gutenberg-tm electronic work under this agreement, disclaim all -liability to you for damages, costs and expenses, including legal -fees. YOU AGREE THAT YOU HAVE NO REMEDIES FOR NEGLIGENCE, STRICT -LIABILITY, BREACH OF WARRANTY OR BREACH OF CONTRACT EXCEPT THOSE -PROVIDED IN PARAGRAPH 1.F.3. YOU AGREE THAT THE FOUNDATION, THE -TRADEMARK OWNER, AND ANY DISTRIBUTOR UNDER THIS AGREEMENT WILL NOT BE -LIABLE TO YOU FOR ACTUAL, DIRECT, INDIRECT, CONSEQUENTIAL, PUNITIVE OR -INCIDENTAL DAMAGES EVEN IF YOU GIVE NOTICE OF THE POSSIBILITY OF SUCH -DAMAGE. - -1.F.3. LIMITED RIGHT OF REPLACEMENT OR REFUND - If you discover a -defect in this electronic work within 90 days of receiving it, you can -receive a refund of the money (if any) you paid for it by sending a -written explanation to the person you received the work from. If you -received the work on a physical medium, you must return the medium with -your written explanation. The person or entity that provided you with -the defective work may elect to provide a replacement copy in lieu of a -refund. If you received the work electronically, the person or entity -providing it to you may choose to give you a second opportunity to -receive the work electronically in lieu of a refund. If the second copy -is also defective, you may demand a refund in writing without further -opportunities to fix the problem. - -1.F.4. Except for the limited right of replacement or refund set forth -in paragraph 1.F.3, this work is provided to you 'AS-IS' WITH NO OTHER -WARRANTIES OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO -WARRANTIES OF MERCHANTABILITY OR FITNESS FOR ANY PURPOSE. - -1.F.5. Some states do not allow disclaimers of certain implied -warranties or the exclusion or limitation of certain types of damages. -If any disclaimer or limitation set forth in this agreement violates the -law of the state applicable to this agreement, the agreement shall be -interpreted to make the maximum disclaimer or limitation permitted by -the applicable state law. The invalidity or unenforceability of any -provision of this agreement shall not void the remaining provisions. - -1.F.6. INDEMNITY - You agree to indemnify and hold the Foundation, the -trademark owner, any agent or employee of the Foundation, anyone -providing copies of Project Gutenberg-tm electronic works in accordance -with this agreement, and any volunteers associated with the production, -promotion and distribution of Project Gutenberg-tm electronic works, -harmless from all liability, costs and expenses, including legal fees, -that arise directly or indirectly from any of the following which you do -or cause to occur: (a) distribution of this or any Project Gutenberg-tm -work, (b) alteration, modification, or additions or deletions to any -Project Gutenberg-tm work, and (c) any Defect you cause. - - -Section 2. Information about the Mission of Project Gutenberg-tm - -Project Gutenberg-tm is synonymous with the free distribution of -electronic works in formats readable by the widest variety of computers -including obsolete, old, middle-aged and new computers. It exists -because of the efforts of hundreds of volunteers and donations from -people in all walks of life. - -Volunteers and financial support to provide volunteers with the -assistance they need, are critical to reaching Project Gutenberg-tm's -goals and ensuring that the Project Gutenberg-tm collection will -remain freely available for generations to come. In 2001, the Project -Gutenberg Literary Archive Foundation was created to provide a secure -and permanent future for Project Gutenberg-tm and future generations. -To learn more about the Project Gutenberg Literary Archive Foundation -and how your efforts and donations can help, see Sections 3 and 4 -and the Foundation web page at http://www.pglaf.org. - - -Section 3. Information about the Project Gutenberg Literary Archive -Foundation - -The Project Gutenberg Literary Archive Foundation is a non profit -501(c)(3) educational corporation organized under the laws of the -state of Mississippi and granted tax exempt status by the Internal -Revenue Service. The Foundation's EIN or federal tax identification -number is 64-6221541. Its 501(c)(3) letter is posted at -http://pglaf.org/fundraising. Contributions to the Project Gutenberg -Literary Archive Foundation are tax deductible to the full extent -permitted by U.S. federal laws and your state's laws. - -The Foundation's principal office is located at 4557 Melan Dr. S. -Fairbanks, AK, 99712., but its volunteers and employees are scattered -throughout numerous locations. Its business office is located at -809 North 1500 West, Salt Lake City, UT 84116, (801) 596-1887, email -business@pglaf.org. Email contact links and up to date contact -information can be found at the Foundation's web site and official -page at http://pglaf.org - -For additional contact information: - Dr. Gregory B. Newby - Chief Executive and Director - gbnewby@pglaf.org - - -Section 4. Information about Donations to the Project Gutenberg -Literary Archive Foundation - -Project Gutenberg-tm depends upon and cannot survive without wide -spread public support and donations to carry out its mission of -increasing the number of public domain and licensed works that can be -freely distributed in machine readable form accessible by the widest -array of equipment including outdated equipment. Many small donations -($1 to $5,000) are particularly important to maintaining tax exempt -status with the IRS. - -The Foundation is committed to complying with the laws regulating -charities and charitable donations in all 50 states of the United -States. Compliance requirements are not uniform and it takes a -considerable effort, much paperwork and many fees to meet and keep up -with these requirements. We do not solicit donations in locations -where we have not received written confirmation of compliance. To -SEND DONATIONS or determine the status of compliance for any -particular state visit http://pglaf.org - -While we cannot and do not solicit contributions from states where we -have not met the solicitation requirements, we know of no prohibition -against accepting unsolicited donations from donors in such states who -approach us with offers to donate. - -International donations are gratefully accepted, but we cannot make -any statements concerning tax treatment of donations received from -outside the United States. U.S. laws alone swamp our small staff. - -Please check the Project Gutenberg Web pages for current donation -methods and addresses. Donations are accepted in a number of other -ways including checks, online payments and credit card donations. -To donate, please visit: http://pglaf.org/donate - - -Section 5. General Information About Project Gutenberg-tm electronic -works. - -Professor Michael S. Hart is the originator of the Project Gutenberg-tm -concept of a library of electronic works that could be freely shared -with anyone. For thirty years, he produced and distributed Project -Gutenberg-tm eBooks with only a loose network of volunteer support. - - -Project Gutenberg-tm eBooks are often created from several printed -editions, all of which are confirmed as Public Domain in the U.S. -unless a copyright notice is included. Thus, we do not necessarily -keep eBooks in compliance with any particular paper edition. - - -Most people start at our Web site which has the main PG search facility: - - http://www.gutenberg.org - -This Web site includes information about Project Gutenberg-tm, -including how to make donations to the Project Gutenberg Literary -Archive Foundation, how to help produce our new eBooks, and how to -subscribe to our email newsletter to hear about new eBooks. |