summaryrefslogtreecommitdiff
diff options
context:
space:
mode:
authorRoger Frank <rfrank@pglaf.org>2025-10-14 19:56:58 -0700
committerRoger Frank <rfrank@pglaf.org>2025-10-14 19:56:58 -0700
commitde3ee9e7209d19f6a55d2f55d78fb379d6d0af3b (patch)
tree44cf4bb781cc589750f0d13b8485bfe295216a5d
initial commit of ebook 32063HEADmain
-rw-r--r--.gitattributes3
-rw-r--r--32063-8.txt6172
-rw-r--r--32063-8.zipbin0 -> 113092 bytes
-rw-r--r--32063-h.zipbin0 -> 483786 bytes
-rw-r--r--32063-h/32063-h.htm6917
-rw-r--r--32063-h/images/img861a.jpgbin0 -> 8178 bytes
-rw-r--r--32063-h/images/img861b.jpgbin0 -> 5417 bytes
-rw-r--r--32063-h/images/img862.jpgbin0 -> 4475 bytes
-rw-r--r--32063-h/images/img865.jpgbin0 -> 9217 bytes
-rw-r--r--32063-h/images/img866a.jpgbin0 -> 3541 bytes
-rw-r--r--32063-h/images/img866b.jpgbin0 -> 20927 bytes
-rw-r--r--32063-h/images/img868a.jpgbin0 -> 14997 bytes
-rw-r--r--32063-h/images/img868b.jpgbin0 -> 3567 bytes
-rw-r--r--32063-h/images/img868c.jpgbin0 -> 3253 bytes
-rw-r--r--32063-h/images/img871.jpgbin0 -> 14791 bytes
-rw-r--r--32063-h/images/img872.jpgbin0 -> 14354 bytes
-rw-r--r--32063-h/images/img875.jpgbin0 -> 4336 bytes
-rw-r--r--32063-h/images/img877.jpgbin0 -> 24829 bytes
-rw-r--r--32063-h/images/img881a.jpgbin0 -> 19690 bytes
-rw-r--r--32063-h/images/img881b.jpgbin0 -> 25958 bytes
-rw-r--r--32063-h/images/img883.jpgbin0 -> 63801 bytes
-rw-r--r--32063-h/images/img884a.jpgbin0 -> 21222 bytes
-rw-r--r--32063-h/images/img884b.jpgbin0 -> 32936 bytes
-rw-r--r--32063-h/images/img885.jpgbin0 -> 14910 bytes
-rw-r--r--32063-h/images/img887a.jpgbin0 -> 5701 bytes
-rw-r--r--32063-h/images/img887b.jpgbin0 -> 15317 bytes
-rw-r--r--32063-h/images/img888.jpgbin0 -> 9680 bytes
-rw-r--r--32063-h/images/img889a.jpgbin0 -> 3458 bytes
-rw-r--r--32063-h/images/img889b.jpgbin0 -> 7856 bytes
-rw-r--r--32063-h/images/img890.jpgbin0 -> 5375 bytes
-rw-r--r--32063.txt6178
-rw-r--r--32063.zipbin0 -> 113103 bytes
-rw-r--r--LICENSE.txt11
-rw-r--r--README.md2
34 files changed, 19283 insertions, 0 deletions
diff --git a/.gitattributes b/.gitattributes
new file mode 100644
index 0000000..6833f05
--- /dev/null
+++ b/.gitattributes
@@ -0,0 +1,3 @@
+* text=auto
+*.txt text
+*.md text
diff --git a/32063-8.txt b/32063-8.txt
new file mode 100644
index 0000000..d91cd8d
--- /dev/null
+++ b/32063-8.txt
@@ -0,0 +1,6172 @@
+The Project Gutenberg EBook of Encyclopaedia Britannica, 11th Edition,
+Volume 6, Slice 8, by Various
+
+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
+
+
+Title: Encyclopaedia Britannica, 11th Edition, Volume 6, Slice 8
+ "Conduction, Electric"
+
+Author: Various
+
+Release Date: April 19, 2010 [EBook #32063]
+
+Language: English
+
+Character set encoding: ISO-8859-1
+
+*** START OF THIS PROJECT GUTENBERG EBOOK ENCYC. BRITANNICA, VOL 6, SL 8 ***
+
+
+
+
+Produced by Marius Masi, Don Kretz, Juliet Sutherland, and
+the Online Distributed Proofreading Team at
+https://www.pgdp.net
+
+
+
+
+
+
+
+
+
+Transcriber's notes:
+
+(1) Numbers following letters (without space) like C2 were originally
+ printed in subscript. When letters are subscripted, they are
+ preceded by an underscore, like C_n.
+
+(2) Characters following a carat (^) were originally printed in
+ superscript.
+
+(3) Letters topped by Macron are represented as [=x].
+
+(4) [oo] stands for infinity; [int] for integral; [alpha], [beta], etc.
+ for greek letters.
+
+(5) The following typographical errors have been corrected:
+
+ Page 871: "an equation which is very useful, because it enables us,
+ if we know the distribution." 'because' amended from 'becaus'.
+
+ Page 890: "The colour of the luminosity due to positive rays is not
+ in general the same as that due to anode rays." 'positive' amended
+ from 'postive'.
+
+ Page 890: "Anode Rays.--Gehrcke and Reichenhein (Ann. der Phys. 25,
+ p. 861) have found that when the anode consists of a mixture of
+ sodium and lithium chloride." 'mixture' amended from 'mixure'.
+
+
+
+
+ ENCYCLOPAEDIA BRITANNICA
+
+ A DICTIONARY OF ARTS, SCIENCES, LITERATURE
+ AND GENERAL INFORMATION
+
+ ELEVENTH EDITION
+
+
+ VOLUME VI, SLICE VIII
+
+ Conduction, Electric
+
+
+
+
+Article in This Slice:
+
+ CONDUCTION, ELECTRIC
+
+
+
+
+CONDUCTION, ELECTRIC. The electric conductivity of a substance is that
+property in virtue of which all its parts come spontaneously to the same
+electric potential if the substance is kept free from the operation of
+electric force. Accordingly, the reciprocal quality, electric
+resistivity, may be defined as a quality of a substance in virtue of
+which a difference of potential can exist between different portions of
+the body when these are in contact with some constant source of
+electromotive force, in such a manner as to form part of an electric
+circuit.
+
+All material substances possess in some degree, large or small, electric
+conductivity, and may for the sake of convenience be broadly divided
+into five classes in this respect. Between these, however, there is no
+sharply-marked dividing line, and the classification must therefore be
+accepted as a more or less arbitrary one. These divisions are: (1)
+metallic conductors, (2) non-metallic conductors, (3) dielectric
+conductors, (4) electrolytic conductors, (5) gaseous conductors. The
+first class comprises all metallic substances, and those mixtures or
+combinations of metallic substances known as alloys. The second includes
+such non-metallic bodies as carbon, silicon, many of the oxides and
+peroxides of the metals, and probably also some oxides of the
+non-metals, sulphides and selenides. Many of these substances, for
+instance carbon and silicon, are well-known to have the property of
+existing in several allotropic forms, and in some of these conditions,
+so far from being fairly good conductors, they may be almost perfect
+non-conductors. An example of this is seen in the case of carbon in its
+three allotropic conditions--charcoal, graphite and diamond. As charcoal
+it possesses a fairly well-marked but not very high conductivity in
+comparison with metals; as graphite, a conductivity about
+one-four-hundredth of that of iron; but as diamond so little
+conductivity that the substance is included amongst insulators or
+non-conductors. The third class includes those substances which are
+generally called insulators or non-conductors, but which are better
+denominated dielectric conductors; it comprises such solid substances as
+mica, ebonite, shellac, india-rubber, gutta-percha, paraffin, and a
+large number of liquids, chiefly hydrocarbons. These substances differ
+greatly in insulating power, and according as the conductivity is more
+or less marked, they are spoken of as bad or good insulators. Amongst
+the latter many of the liquid gases hold a high position. Thus, liquid
+oxygen and liquid air have been shown by Sir James Dewar to be almost
+perfect non-conductors of electricity.
+
+The behaviour of substances which fall into these three classes is
+discussed below in section I., dealing with metallic conduction.
+
+The fourth class, namely the electrolytic conductors comprises all those
+substances which undergo chemical decomposition when they form part of
+an electric circuit traversed by an electric current. They are discussed
+in section II., dealing with electrolytic conduction.
+
+The fifth and last class of conductors includes the gases. The
+conditions under which this class of substance becomes possessed of
+electric conductivity are considered in section III., on conduction in
+gases.
+
+In connexion with metallic conductors, it is a fact of great interest
+and considerable practical importance, that, although the majority of
+metals when in a finely divided or powdered condition are practically
+non-conductors, a mass of metallic powder or filings may be made to pass
+suddenly into a conductive condition by being exposed to the influence
+of an electric wave. The same is true of the loose contact of two
+metallic conductors. Thus if a steel point, such as a needle, presses
+very lightly against a metallic plate, say of aluminium, it is found
+that this metallic contact, if carefully adjusted, is non-conductive,
+but that if an electric wave is created anywhere in the neighbourhood,
+this non-conducting contact passes into a conductive state. This fact,
+investigated and discovered independently by D. E. Hughes, C. Onesti, E.
+Branly, O. J. Lodge and others, is applied in the construction of the
+"coherer," or sensitive tube employed as a detector or receiver in that
+form of "wireless telegraphy" chiefly developed by Marconi. Further
+references to it are made in the articles ELECTRIC WAVES and TELEGRAPHY:
+_Wireless_.
+
+ _International Ohm._--The practical unit of electrical resistance was
+ legally defined in Great Britain by the authority of the queen in
+ council in 1894, as the "resistance offered to an invariable electric
+ current by a column of mercury at the temperature of melting ice,
+ 14.4521 grammes in mass, of a constant cross-sectional area, and a
+ length 106.3 centimetres." The same unit has been also legalized as a
+ standard in France, Germany and the United States, and is denominated
+ the "International or Standard Ohm." It is intended to represent as
+ nearly as possible a resistance equal to 10° absolute C.G.S. units of
+ electric resistance. Convenient multiples and subdivisions of the ohm
+ are the microhm and the megohm, the former being a millionth part of
+ an ohm, and the latter a million ohms. The resistivity of substances
+ is then numerically expressed by stating the resistance of one cubic
+ centimetre of the substance taken between opposed faces, and expressed
+ in ohms, microhms or megohms, as may be most convenient. The
+ reciprocal of the ohm is called the mho, which is the unit of
+ conductivity, and is defined as the conductivity of a substance whose
+ resistance is one ohm. The absolute unit of conductivity is the
+ conductivity of a substance whose resistivity is one absolute C.G.S.
+ unit, or one-thousandth-millionth part of an ohm. Resistivity is a
+ quality in which material substances differ very widely. The metals
+ and alloys, broadly speaking, are good conductors, and their
+ resistivity is conveniently expressed in microhms per cubic
+ centimetre, or in absolute C.G.S. units. Very small differences in
+ density and in chemical purity make, however, immense differences in
+ electric resistivity; hence the values given by different
+ experimentalists for the resistivity of known metals differ to a
+ considerable extent.
+
+
+I. CONDUCTION IN SOLIDS
+
+It is found convenient to express the resistivity of metals in two
+different ways: (1) We may state the resistivity of one cubic centimetre
+of the material in microhms or absolute units taken between opposed
+faces. This is called the _volume-resistivity_; (2) we may express the
+resistivity by stating the resistance in ohms offered by a wire of the
+material in question of uniform cross-section one metre in length, and
+one gramme in weight. This numerical measure of the resistivity is
+called the _mass-resistivity_. The mass-resistivity of a body is
+connected with its volume-resistivity and the density of the material in
+the following manner:--The mass-resistivity, expressed in microhms per
+metre-gramme, divided by 10 times the density is numerically equal to
+the volume-resistivity per centimetre-cube in absolute C.G.S. units. The
+mass-resistivity per metre-gramme can always be obtained by measuring
+the resistance and the mass of any wire of uniform cross-section of
+which the length is known, and if the density of the substance is then
+measured, the volume-resistivity can be immediately calculated.
+
+ If R is the resistance in ohms of a wire of length l, uniform
+ cross-section s, and density d, then taking [rho] for the
+ volume-resistivity we have 10^9R = [rho]l/s; but lsd = M, where M is
+ the mass of the wire. Hence 10^9R = [rho]dl²/M. If l = 100 and M = 1,
+ then R = [rho]'= resistivity in ohms per metre-gramme, and 10^9[rho]'
+ = 10,000d[rho], or [rho] = 10^5[rho]'/d, and [rho]' = 10,000MR/l².
+
+ The following rules, therefore, are useful in connexion with these
+ measurements. To obtain the mass-resistivity per metre-gramme of a
+ substance in the form of a uniform metallic wire:--Multiply together
+ 10,000 times the mass in grammes and the total resistance in ohms, and
+ then divide by the square of the length in centimetres. Again, to
+ obtain the volume-resistivity in C.G.S. units per centimetre-cube, the
+ rule is to multiply the mass-resistivity in ohms by 100,000 and divide
+ by the density. These rules, of course, apply only to wires of uniform
+ cross-section. In the following Tables I., II. and III. are given the
+ mass and volume resistivity of ordinary metals and certain alloys
+ expressed in terms of the international ohm or the absolute C.G.S.
+ unit of resistance, the values being calculated from the experiments
+ of A. Matthiessen (1831-1870) between 1860 and 1865, and from later
+ results obtained by J. A. Fleming and Sir James Dewar in 1893.
+
+ TABLE I.--_Electric Mass-Resistivity of Various Metals at 0° C., or
+ Resistance per Metre-gramme in International Ohms at 0° C._
+ (Matthiessen.)
+
+ +---------------------+---------------------+-------------+
+ | | Resistance at 0° C. | |
+ | | in International | Approximate |
+ | Metal. | Ohms of a Wire | Temperature |
+ | | 1 Metre long and | Coefficient |
+ | | Weighing | near 20° C. |
+ | | 1 Gramme. | |
+ +---------------------+---------------------+-------------+
+ |Silver (annealed) | .1523 | 0.00377 |
+ |Silver (hard-drawn) | .1657 | .. |
+ |Copper (annealed) | .1421 | 0.00388 |
+ |Copper (hard-drawn) | .1449 (Matthiessen's Standard) |
+ |Gold (annealed) | .4025 | 0.00365 |
+ |Gold (hard-drawn) | .4094 | .. |
+ |Aluminium (annealed) | .0757 | .. |
+ |Zinc (pressed) | .4013 | .. |
+ |Platinum (annealed) | 1.9337 | .. |
+ |Iron (annealed) | .765 | .. |
+ |Nickel (annealed) | 1.058[1] | .. |
+ |Tin (pressed) | .9618 | 0.00365 |
+ |Lead (pressed) | 2.2268 | 0.00387 |
+ |Antimony (pressed) | 2.3787 | 0.00389 |
+ |Bismuth (pressed) | 12.8554[1] | 0.00354 |
+ |Mercury (liquid) | 12.885[2] | 0.00072 |
+ +---------------------+---------------------+-------------+
+
+ The data commonly used for calculating metallic resistivities were
+ obtained by A. Matthiessen, and his results are set out in the Table
+ II. which is taken from Cantor lectures given by Fleeming Jenkin in
+ 1866 at or about the date when the researches were made. The figures
+ given by Jenkin have, however, been reduced to international ohms and
+ C.G.S. units by multiplying by ([pi]/4) × 0.9866 × 10^5 = 77,485.
+
+ Subsequently numerous determinations of the resistivity of various
+ pure metals were made by Fleming and Dewar, whose results are set out
+ in Table III.
+
+ TABLE II.--_Electric Volume-Resistivity of Various Metals at 0° C.,
+ or Resistance per Centimetre-cube in C.G.S. Units at 0° C._
+
+ +---------------------+---------------------+
+ | | Volume-Resistivity. |
+ | Metal. | at 0° C. in C.G.S. |
+ | | Units |
+ +---------------------+---------------------+
+ |Silver (annealed) | 1,502 |
+ |Silver (hard-drawn) | 1,629 |
+ |Copper (annealed) | 1,594 |
+ |Copper (hard-drawn) | 1,630[3] |
+ |Gold (annealed) | 2,052 |
+ |Gold (hard-drawn) | 2,090 |
+ |Aluminium (annealed) | 3,006 |
+ |Zinc (pressed) | 5,621 |
+ |Platinum (annealed) | 9,035 |
+ |Iron (annealed) | 10,568 |
+ |Nickel (annealed) | 12,429[4] |
+ |Tin (pressed) | 13,178 |
+ |Lead (pressed) | 19,580 |
+ |Antimony (pressed) | 35,418 |
+ |Bismuth (pressed) | 130,872 |
+ |Mercury (liquid) | 94,896[5] |
+ +---------------------+---------------------+
+
+ TABLE III.--_Electric Volume-Resistivity of Various Metals at 0° C.,
+ or Resistance per Centimetre-cube at 0° C. in C.G.S. Units._
+ (Fleming and Dewar, _Phil. Mag._, September 1893.)
+
+ +--------------------------+------------------+------------------+
+ | | Resistance | Mean Temperature |
+ | Metal. | at 0° C. per | Coefficient |
+ | | Centimetre-cube | between 0° C. |
+ | | in C.G.S. Units. | and 100° C. |
+ +--------------------------+------------------+------------------+
+ |Silver (electrolytic and | | |
+ | well annealed)[6] | 1,468 | 0.00400 |
+ |Copper (electrolytic and | | |
+ | well annealed)[6] | 1,561 | 0.00428 |
+ |Gold (annealed) | 2,197 | 0.00377 |
+ |Aluminium (annealed) | 2,665 | 0.00435 |
+ |Magnesium (pressed) | 4,355 | 0.00381 |
+ |Zinc | 5,751 | 0.00406 |
+ |Nickel (electrolytic)[6] | 6,935 | 0.00618 |
+ |Iron (annealed) | 9,065 | 0.00625 |
+ |Cadmium | 10,023 | 0.00419 |
+ |Palladium | 10,219 | 0.00354 |
+ |Platinum (annealed) | 10,917 | 0.003669 |
+ |Tin (pressed) | 13,048 | 0.00440 |
+ |Thallium (pressed) | 17,633 | 0.00398 |
+ |Lead (pressed) | 20,380 | 0.00411 |
+ |Bismuth (electrolytic)[7] | 110,000 | 0.00433 |
+ +--------------------------+------------------+------------------+
+
+ _Resistivity of Mercury._--The volume-resistivity of pure mercury is a
+ very important electric constant, and since 1880 many of the most
+ competent experimentalists have directed their attention to the
+ determination of its value. The experimental process has usually been
+ to fill a glass tube of known dimensions, having large cup-like
+ extensions at the ends, with pure mercury, and determine the absolute
+ resistance of this column of metal. For the practical details of this
+ method the following references may be consulted:--"The Specific
+ Resistance of Mercury," Lord Rayleigh and Mrs Sidgwick, _Phil.
+ Trans._, 1883, part i. p. 173, and R. T. Glazebrook, _Phil. Mag._,
+ 1885, p. 20; "On the Specific Resistance of Mercury," R. T. Glazebrook
+ and T. C. Fitzpatrick, _Phil. Trans._, 1888, p. 179, or _Proc. Roy.
+ Soc._, 1888, p. 44, or _Electrician_, 1888, 21, p. 538; "Recent
+ Determinations of the Absolute Resistance of Mercury," R. T.
+ Glazebrook, _Electrician_, 1890, 25, pp. 543 and 588. Also see J. V.
+ Jones, "On the Determination of the Specific Resistance of Mercury in
+ Absolute Measure," _Phil. Trans._, 1891, A, p. 2. Table IV. gives the
+ values of the volume-resistivity of mercury as determined by various
+ observers, the constant being expressed (a) in terms of the resistance
+ in ohms of a column of mercury one millimetre in cross-section and 100
+ centimetres in length, taken at 0° C.; and (b) in terms of the length
+ in centimetres of a column of mercury one square millimetre in
+ cross-section taken at 0° C. The result of all the most careful
+ determinations has been to show that the resistivity of pure mercury
+ at 0° C. is about 94,070 C.G.S. electromagnetic units of resistance,
+ and that a column of mercury 106.3 centimetres in length having a
+ cross-sectional area of one square millimetre would have a resistance
+ at 0° C. of one international ohm. These values have accordingly been
+ accepted as the official and recognized values for the specific
+ resistance of mercury, and the definition of the ohm. The table also
+ states the methods which have been adopted by the different observers
+ for obtaining the absolute value of the resistance of a known column
+ of mercury, or of a resistance coil afterwards compared with a known
+ column of mercury. A column of figures is added showing the value in
+ fractions of an international ohm of the British Association Unit
+ (B.A.U.), formerly supposed to represent the true ohm. The real value
+ of the B.A.U. is now taken as .9866 of an international ohm.
+
+ TABLE IV.--_Determinations of the Absolute Value of the
+ Volume-Resistivity of Mercury and the Mercury Equivalent of the
+ Ohm._
+
+ +-----------------+------+----------------------+--------+----------+---------+
+ | | | | |Value of |Value of |
+ | | | |Value of|100 Centi-|Ohm in |
+ | Observer. | Date.| Method. |B.A.U. |metres of |Centi- |
+ | | | |in Ohms.|Mercury |metres of|
+ | | | | |in Ohms. |Mercury. |
+ +-----------------+------+----------------------+--------+----------+---------+
+ |Lord Rayleigh | 1882 | Rotating coil | .98651 | .94133 | 106.31 |
+ |Lord Rayleigh | 1883 | Lorenz method | .98677 | .. | 106.27 |
+ |G. Wiedemann | 1884 | Rotation through 180°| .. | .. | 106.19 |
+ |E. E. N. Mascart | 1884 | Induced current | .98611 | .94096 | 106.33 |
+ |H. A. Rowland | 1887 | Mean of several | .98644 | .94071 | 106.32 |
+ | | | methods | | | |
+ |F. Kohlrausch | 1887 | Damping of magnets | .98660 | .94061 | 106.32 |
+ |R. T. Glazebrook | 1882 | Induced currents | .98665 | .94074 | 106.29 |
+ | | 1888 | | | | |
+ |Wuilleumeier | 1890 | | .98686 | .94077 | 106.31 |
+ |Duncan and Wilkes| 1890 | Lorenz | .98634 | .94067 | 106.34 |
+ |J. V. Jones | 1891 | Lorenz | .. | .94067 | 106.31 |
+ | | | +--------+----------+---------+
+ | | | Mean value .98653 | | |
+ | | | | | |
+ |Streker | 1885 | An absolute determin-| | .94056 | 106.32 |
+ |Hutchinson | 1888 | ation of resistance| | .94074 | 106.30 |
+ |E. Salvioni | 1890 | was not made. The | | .94054 | 106.33 |
+ |E. Salvioni | .. | value .98656 has | | .94076 | 106.30 |
+ | | | been used | | | |
+ | | | | +----------+---------+
+ | | | Mean value .94076 | 106.31 |
+ | | | | |
+ |H. F. Weber | 1884 | Induced current | Absolute measure- | 105.37 |
+ |H. F. Weber | .. | Rotating coil | ments compared | 106.16 |
+ |A. Roiti | 1884 | Mean effect of | with German silver| 105.89 |
+ | | | induced current | wire coils issued | |
+ |F. Himstedt | 1885 | | by Siemens and | 105.98 |
+ | | | | Streker | |
+ |F. E. Dorn | 1889 | Damping of a magnet | | 106.24 |
+ |Wild | 1883 | Damping of a magnet | | 106.03 |
+ |L. V. Lorenz | 1885 | Lorenz method | | 105.93 |
+ +-----------------+------+----------------------+-------------------+---------+
+
+ For a critical discussion of the methods which have been adopted in
+ the absolute determination of the resistivity of mercury, and the
+ value of the British Association unit of resistance, the reader may be
+ referred to the _British Association Reports_ for 1890 and 1892
+ (_Report of Electrical Standards Committee_), and to the
+ _Electrician_, 25, p. 456, and 29, p. 462. A discussion of the
+ relative value of the results obtained between 1882 and 1890 was given
+ by R. T. Glazebrook in a paper presented to the British Association at
+ Leeds, 1890.
+
+ _Resistivity of Copper._--In connexion with electro-technical work the
+ determination of the conductivity or resistivity values of annealed
+ and hard-drawn copper wire at standard temperatures is a very
+ important matter. Matthiessen devoted considerable attention to this
+ subject between the years 1860 and 1864 (see _Phil. Trans._, 1860, p.
+ 150), and since that time much additional work has been carried out.
+ Matthiessen's value, known as _Matthiessen's Standard_, for the
+ mass-resistivity of pure hard-drawn copper wire, is the resistance of
+ a wire of pure hard-drawn copper one metre long and weighing one
+ gramme, and this is equal to 0.14493 international ohms at 0° C. For
+ many purposes it is more convenient to express temperature in
+ Fahrenheit degrees, and the recommendation of the 1899 committee on
+ copper conductors[8] is as follows:--"Matthiessen's standard for
+ hard-drawn conductivity commercial copper shall be considered to be
+ the resistance of a wire of pure hard-drawn copper one metre long,
+ weighing one gramme which at 60° F. is 0.153858 international ohms."
+ Matthiessen also measured the mass-resistivity of annealed copper, and
+ found that its conductivity is greater than that of hard-drawn copper
+ by about 2.25% to 2.5% As annealed copper may vary considerably in its
+ state of annealing, and is always somewhat hardened by bending and
+ winding, it is found in practice that the resistivity of commercial
+ annealed copper is about 1¼% less than that of hard-drawn copper. The
+ standard now accepted for such copper, on the recommendation of the
+ 1899 Committee, is a wire of pure annealed copper one metre long,
+ weighing one gramme, whose resistance at 0° C. is 0.1421 international
+ ohms, or at 60° F., 0.150822 international ohms. The specific gravity
+ of copper varies from about 8.89 to 8.95, and the standard value
+ accepted for high conductivity commercial copper is 8.912,
+ corresponding to a weight of 555 lb per cubic foot at 60° F. Hence the
+ volume-resistivity of pure annealed copper at 0° C. is 1.594 microhms
+ per c.c., or 1594 C.G.S. units, and that of pure hard-drawn copper at
+ 0° C. is 1.626 microhms per c.c., or 1626 C.G.S. units. Since
+ Matthiessen's researches, the most careful scientific investigation on
+ the conductivity of copper is that of T. C. Fitzpatrick, carried out
+ in 1890. (_Brit. Assoc. Report_, 1890, Appendix 3, p. 120.)
+ Fitzpatrick confirmed Matthiessen's chief result, and obtained values
+ for the resistivity of hard-drawn copper which, when corrected for
+ temperature variation, are in entire agreement with those of
+ Matthiessen at the same temperature.
+
+The volume resistivity of alloys is, generally speaking, much higher
+than that of pure metals. Table V. shows the volume resistivity at 0° C.
+of a number of well-known alloys, with their chemical composition.
+
+ TABLE V.--_Volume-Resistivity of Alloys of known Composition at 0°
+ C. in C.G.S. Units per Centimetre-cube. Mean Temperature
+ Coefficients taken at 15° C._ (Fleming and Dewar.)
+
+ +-----------------------+-----------+-----------+-------------------+
+ | |Resistivity|Temperature|Composition in per |
+ | Alloys. |at 0° C. |Coefficient| cents. |
+ | | |at 15° C. | |
+ +-----------------------+-----------+-----------+-------------------+
+ |Platinum-silver | 31,582 | .000243 |Pt 33%, Ag 66% |
+ |Platinum-iridium | 30,896 | .000822 |Pt 80%, Ir 20% |
+ |Platinum-rhodium | 21,142 | .00143 |Pt 90%, Rd 10% |
+ |Gold-silver | 6,280 | .00124 |Au 90%, Ag 10% |
+ |Manganese-steel | 67,148 | .00127 |Mn 12%, Fe 78% |
+ |Nickel-steel | 29,452 | .00201 |Ni 4.35%, remaining|
+ | | | | percentage |
+ | | | | chiefly iron, but|
+ | | | | uncertain |
+ |German silver | 29,982 | .000273 |Cu5Zn3Ni2 |
+ |Platinoid[9] | 41,731 | .00031 | |
+ |Manganin | 46,678 | .0000 |Cu 84%, Mn 12%, |
+ | | | |Ni 4% |
+ |Aluminium-silver | 4,641 | .00238 |Al 94%, Ag 6% |
+ |Aluminium-copper | 2,904 | .00381 |Al 94%, Cu 6% |
+ |Copper-aluminium | 8,847 | .000897 |Cu 97%, Al 3% |
+ |Copper-nickel-aluminium| 14,912 | .000643 |Cu 87%, Ni 6.5%, |
+ | | | | Al 6.5% |
+ |Titanium-aluminium | 3,887 | .00290 | |
+ +-----------------------+-----------+-----------+-------------------+
+
+ Generally speaking, an alloy having high resistivity has poor
+ mechanical qualities, that is to say, its tensile strength and
+ ductility are small. It is possible to form alloys having a
+ resistivity as high as 100 microhms per cubic centimetre; but, on the
+ other hand, the value of an alloy for electro-technical purposes is
+ judged not merely by its resistivity, but also by the degree to which
+ its resistivity varies with temperature, and by its capability of
+ being easily drawn into fine wire of not very small tensile strength.
+ Some pure metals when alloyed with a small proportion of another metal
+ do not suffer much change in resistivity, but in other cases the
+ resultant alloy has a much higher resistivity. Thus an alloy of pure
+ copper with 3% of aluminium has a resistivity about 5½ times that of
+ copper; but if pure aluminium is alloyed with 6% of copper, the
+ resistivity of the product is not more than 20% greater than that of
+ pure aluminium. The presence of a very small proportion of a
+ non-metallic element in a metallic mass, such as oxygen, sulphur or
+ phosphorus, has a very great effect in increasing the resistivity.
+ Certain metallic elements also have the same power; thus platinoid has
+ a resistivity 30% greater than German silver, though it differs from
+ it merely in containing a trace of tungsten.
+
+The resistivity of non-metallic conductors is in all cases higher than
+that of any pure metal. The resistivity of carbon, for instance, in the
+forms of charcoal or carbonized organic material and graphite, varies
+from 600 to 6000 microhms per cubic centimetre, as shown in Table VI.:--
+
+ TABLE VI.--_Electric Volume-Resistivity in Microhms per
+ Centimetre-cube of Various Forms of Carbon at 15° C._
+
+ +-----------------------------------------+---------------+
+ | Substance. | Resistivity. |
+ +-----------------------------------------+---------------+
+ | Arc lamp carbon rod | 8000 |
+ | Jablochkoff candle carbon | 4000 |
+ | Carré carbon | 3400 |
+ | Carbonized bamboo | 6000 |
+ | Carbonized parchmentized thread | 4000 to 5000 |
+ | Ordinary carbon filament from glow-lamp | |
+ | "treated" or flashed | 2400 to 2500 |
+ | Deposited or secondary carbon | 600 to 900 |
+ | Graphite | 400 to 500 |
+ +-----------------------------------------+---------------+
+
+The resistivity of liquids is, generally speaking, much higher than that
+of any metals, metallic alloys or non-metallic conductors. Thus fused
+lead chloride, one of the best conducting liquids, has a resistivity in
+its fused condition of 0.376 ohm per centimetre-cube, or 376,000
+microhms per centimetre-cube, whereas that of metallic alloys only in
+few cases exceeds 100 microhms per centimetre-cube. The resistivity of
+solutions of metallic salts also varies very largely with the proportion
+of the diluent or solvent, and in some instances, as in the aqueous
+solutions of mineral acids; there is a maximum conductivity
+corresponding to a certain dilution. The resistivity of many liquids,
+such as alcohol, ether, benzene and pure water, is so high, in other
+words, their conductivity is so small, that they are practically
+insulators, and the resistivity can only be appropriately expressed in
+megohms per centimetre-cube.
+
+In Table VII. are given the names of a few of these badly-conducting
+liquids, with the values of their volume-resistivity in megohms per
+centimetre-cube:--
+
+ TABLE VII.--_Electric Volume-Resistivity of Various Badly-Conducting
+ Liquids in Megohms per Centimetre-cube._
+
+ +---------------------------+----------------+---------------------+
+ | | Resistivity | |
+ | Substance. | in Megohms | Observer. |
+ | | per c.c. | |
+ +---------------------------+----------------+---------------------+
+ | Ethyl alcohol | 0.5 | Pfeiffer. |
+ | Ethyl ether | 1.175 to 3.760 | W. Kohlrausch. |
+ | Benzene | 4.700 | |
+ | Absolutely pure water | 25.0 at 18° C. | Value estimated |
+ | approximates probably to | | by F. Kohlrausch |
+ | | | and A. Heydweiler. |
+ | All very dilute aqueous | 1.00 at 18° C. | From results by |
+ | salt solutions having a | | F. Kohlrausch |
+ | concentration of about | | and others. |
+ | 0.00001 of an equivalent | | |
+ | gramme molecule[10] per | | |
+ | litre approximate to | | |
+ +---------------------------+----------------+---------------------+
+
+The resistivity of all those substances which are generally called
+dielectrics or insulators is also so high that it can only be
+appropriately expressed in millions of megohms per centimetre-cube, or
+in megohms per quadrant-cube, the quadrant being a cube the side of
+which is 10^9 cms. (see Table VIII.).
+
+Effects of Heat.--Temperature affects the resistivity of these different
+classes of conductors in different ways. In all cases, so far as is yet
+known, the resistivity of a pure metal is increased if its temperature is
+raised, and decreased if the temperature is lowered, so that if it could
+be brought to the absolute zero of temperature (-273° C.) its resistivity
+would be reduced to a very small fraction of its resistance at ordinary
+temperatures. With metallic alloys, however, rise of temperature does not
+always increase resistivity: it sometimes diminishes it, so that many
+alloys are known which have a maximum resistivity corresponding to a
+certain temperature, and at or near this point they vary very little in
+resistance with temperature. Such alloys have, therefore, a negative
+temperature-variation of resistance at and above fixed temperatures.
+Prominent amongst these metallic compounds are alloys of iron, manganese,
+nickel and copper, some of which were discovered by Edward Weston, in the
+United States. One well-known alloy of copper, manganese and nickel, now
+called manganin, which was brought to the notice of electricians by the
+careful investigations made at the Berlin Physikalisch-Technische
+Reichsanstalt, is characterized by having a zero temperature coefficient
+at or about a certain temperature in the neighbourhood of 15° C. Hence
+within a certain range of temperature on either side of this critical
+value the resistivity of manganin is hardly affected at all by
+temperature. Similar alloys can be produced from copper and
+ferro-manganese. An alloy formed of 80% copper and 20% manganese in an
+annealed condition has a nearly zero temperature-variation of resistance
+between 20° C. and 100° C. In the case of non-metals the action of
+temperature is generally to diminish the resistivity as temperature
+rises, though this is not universally so. The interesting observation has
+been recorded by J. W. Howell, that "treated" carbon filaments and
+graphite are substances which have a minimum resistance corresponding to
+a certain temperature approaching red heat (_Electrician_, vol. xxxviii.
+p. 835). At and beyond this temperature increased heating appears to
+increase their resistivity; this phenomenon may, however, be accompanied
+by a molecular change and not be a true temperature variation. In the
+case of dielectric conductors and of electrolytes, the action of rising
+temperature is to reduce resistivity. Many of the so-called insulators,
+such as mica, ebonite, indiarubber, and the insulating oils, paraffin,
+&c., decrease in resistivity with great rapidity as the temperature
+rises. With guttapercha a rise in temperature from 0° C. to 24° C. is
+sufficient to reduce the resistivity of one-twentieth part of its value
+at 0° C., and the resistivity of flint glass at 140° C. is only
+one-hundredth of what it is at 60° C.
+
+ TABLE VIII.--_Electric Volume-Resistivity of Dielectrics reckoned in
+ Millions of Megohms (Mega-megohms) per Centimetre-cube, and in Megohms
+ per Quadrant-cube, i.e. a Cube whose Side is 10^9 cms._
+
+ +------------------------+---------------------------+----------+
+ | | Resistivity. | |
+ | +------------+--------------+ Tempera- |
+ | Substance. |Mega-megohms| Megohms per | tura |
+ | | per c.c. |Quadrant-cube.| Cent. |
+ +------------------------+------------+--------------+----------+
+ |Bohemian glass | 61 | .061 | 60° |
+ |Mica | 84 | .084 | 20° |
+ |Gutta-percha | 450 | .45 | 24° |
+ |Flint glass | 1,020 | 1.02 | 60° |
+ |Glover's vulcanized | | | |
+ | indiarubber | 1,630 | 1.63 | 15° |
+ |Siemens' ordinary pure | | | |
+ | vulcanized indiarubber | 2,280 | 2.28 | 15° |
+ |Shellac | 9,000 | 9.0 | 28° |
+ |Indiarubber | 10,900 | 10.9 | 24° |
+ |Siemens' high-insulating| | | |
+ | fibrous material | 11,900 | 11.9 | 15° |
+ |Siemens' special | | | |
+ | high-insulating | | | |
+ | indiarubber. | 16,170 | 16.17 | 15° |
+ |Flint glass | 20,000 | 20.0 | 20° |
+ |Ebonite | 28,000 | 28. | 46° |
+ |Paraffin | 34,000 | 34. | 46° |
+ +------------------------+------------+--------------+----------+
+
+ A definition may here be given of the meaning of the term _Temperature
+ Coefficient_. If, in the first place, we suppose that the resistivity
+ ([rho]t) at any temperature (t) is a simple linear function of the
+ resistivity ([rho]0) at 0° C., then we can write [rho]t = [rho]0(1 +
+ [alpha]t), or [alpha] = ([rho]t - [rho]0)/[rho]0t.
+
+ The quantity [alpha] is then called the temperature-coefficient, and
+ its reciprocal is the temperature at which the resistivity would
+ become zero. By an extension of this notion we can call the quantity
+ d[rho]/[rho]dt the temperature coefficient corresponding to any
+ temperature t at which the resistivity is [rho]. In all cases the
+ relation between the resistivity of a substance and the temperature is
+ best set out in the form of a curve called a temperature-resistance
+ curve. If a series of such curves are drawn for various pure metals,
+ temperature being taken as abscissa and resistance as ordinate, and if
+ the temperature range extends from the absolute zero of temperature
+ upwards, then it is found that these temperature-resistance lines are
+ curved lines having their convexity either upwards or downwards. In
+ other words, the second differential coefficient of resistance with
+ respect to temperature is either a positive or negative quantity. An
+ extensive series of observations concerning the form of the
+ resistivity curves for various pure metals over a range of temperature
+ extending from -200° C. to +200° C. was carried out in 1892 and 1893
+ by Fleming and Dewar (_Phil. Mag._ Oct. 1892 and Sept. 1893). The
+ resistance observations were taken with resistance coils constructed
+ with wires of various metals obtained in a state of great chemical
+ purity. The lengths and mean diameters of the wires were carefully
+ measured, and their resistance was then taken at certain known
+ temperatures obtained by immersing the coils in boiling aniline,
+ boiling water, melting ice, melting carbonic acid in ether, and
+ boiling liquid oxygen, the temperatures thus given being +184°.5 C.,
+ +100° C., 0° C., -78°.2 C. and -182°.5 C. The resistivities of the
+ various metals were then calculated and set out in terms of the
+ temperature. From these data a chart was prepared showing the
+ temperature-resistance curves of these metals throughout a range of
+ 400 degrees. The exact form of these curves through the region of
+ temperature lying between -200° C. and -273° C. is not yet known. As
+ shown on the chart, the curves evidently do not converge to precisely
+ the same point. It is, however, much less probable that the resistance
+ of any metal should vanish at a temperature above the absolute zero
+ than at the absolute zero itself, and the precise path of these curves
+ at their lower ends cannot be delineated until means are found for
+ fixing independently the temperature of some regions in which the
+ resistance of metallic wires can be measured. Sir J. Dewar
+ subsequently showed that for certain pure metals it is clear that the
+ resistance would not vanish at the absolute zero but would be reduced
+ to a finite but small value (see "Electric Resistance Thermometry at
+ the Temperature of Boiling Hydrogen," _Proc. Roy. Soc._ 1904, 73, p.
+ 244).
+
+ The resistivity curves of the magnetic metals are also remarkable for
+ the change of curvature they exhibit at the magnetic critical
+ temperature. Thus J. Hopkinson and D. K. Morris (_Phil. Mag._
+ September 1897, p. 213) observed the remarkable alteration that takes
+ place in the iron resistance temperature curve in the neighbourhood of
+ 780° C. At that temperature the direction of the curvature of the
+ curve changes so that it becomes convex upwards instead of convex
+ downwards, and in addition the value of the temperature coefficient
+ undergoes a great reduction. The mean temperature coefficient of iron
+ in the neighbourhood of 0° C. is 0.0057; at 765° C. it rises to a
+ maximum value 0.0204; but at 1000° C. it falls again to a lower value,
+ 0.00244. A similar rise to a maximum value and subsequent fall are
+ also noted in the case of the specific heat of iron. The changes in
+ the curvature of the resistivity curves are undoubtedly connected with
+ the molecular changes that occur in the magnetic metals at their
+ critical temperatures.
+
+ A fact of considerable interest in connexion with resistivity is the
+ influence exerted by a strong magnetic field in the case of some
+ metals, notably bismuth. It was discovered by A. Righi and confirmed
+ by S. A. Leduc (_Journ. de Phys._ 1886, 5, p. 116, and 1887, 6, p.
+ 189) that if a pure bismuth wire is placed in a magnetic field
+ transversely to the direction of the magnetic field, its resistance is
+ considerably increased. This increase is greatly affected by the
+ temperature of the metal (Dewar and Fleming, _Proc. Roy. Soc._ 1897,
+ 60, p. 427). The temperature coefficient of pure copper is an
+ important constant, and its value as determined by Messrs Clark, Forde
+ and Taylor in terms of Fahrenheit temperature is
+
+ [rho]t = [rho]32 {1 + 0.0023708(t - 32) + 0.0000034548(t - 32)²}.
+
+_Time Effects._--In the case of dielectric conductors, commonly called
+insulators, such as indiarubber, guttapercha, glass and mica, the
+electric resistivity is not only a function of the temperature but also
+of the time during which the electromotive force employed to measure it
+is imposed. Thus if an indiarubber-covered cable is immersed in water
+and the resistance of the dielectric between the copper conductor and
+the water measured by ascertaining the current which can be caused to
+flow through it by an electromotive force, this current is found to vary
+very rapidly with the time during which the electromotive force is
+applied. Apart from the small initial effect due to the electrostatic
+capacity of the cable, the application of an electromotive force to the
+dielectric produces a current through it which rapidly falls in value,
+as if the electric resistance of the dielectric were increasing. The
+current, however, does not fall continuously but tends to a limiting
+value, and it appears that if the electromotive force is kept applied
+to the cable for a prolonged time, a small and nearly constant current
+will ultimately be found flowing through it. It is customary in
+electro-technical work to consider the resistivity of the dielectric as
+the value it has after the electromotive force has been applied for one
+minute, the standard temperature being 75° F. This, however, is a purely
+conventional proceeding, and the number so obtained does not necessarily
+represent the true or ohmic resistance of the dielectric. If the
+electromotive force is increased, in the case of a large number of
+ordinary dielectrics the apparent resistance at the end of one minute's
+electrification decreases as the electromotive force increases.
+
+_Practical Standards._--The practical measurement of resistivity
+involves many processes and instruments (see WHEATSTONE'S BRIDGE and
+OHMMETER). Broadly speaking, the processes are divided into _Comparison
+Methods_ and _Absolute Methods_. In the former a comparison is effected
+between the resistance of a material in a known form and some standard
+resistance. In the _Absolute Methods_ the resistivity is determined
+without reference to any other substance, but with reference only to the
+fundamental standards of length, mass and time. Immense labour has been
+expended in investigations concerned with the production of a standard
+of resistance and its evaluation in absolute measure. In some cases the
+absolute standard is constructed by filling a carefully-calibrated tube
+of glass with mercury, in order to realize in a material form the
+official definition of the ohm; in this manner most of the principal
+national physical laboratories have been provided with standard mercury
+ohms. (For a full description of the standard mercury ohm of the Berlin
+Physikalisch-Technische Reichsanstalt, see the _Electrician_, xxxvii.
+569.) For practical purposes it is more convenient to employ a standard
+of resistance made of wire.
+
+ Opinion is not yet perfectly settled on the question whether a wire
+ made of any alloy can be considered to be a perfectly unalterable
+ standard of resistance, but experience has shown that a platinum
+ silver alloy (66% silver, 33% platinum), and also the alloy called
+ manganin, seem to possess the qualities of permanence essential for a
+ wire-resistance standard. A comparison made in 1892 and 1894 of all
+ the manganin wire copies of the ohm made at the Reichsanstalt in
+ Berlin, showed that these standards had remained constant for two
+ years to within one or two parts in 100,000. It appears, however, that
+ in order that manganin may remain constant in resistivity when used in
+ the manufacture of a resistance coil, it is necessary that the alloy
+ should be _aged_ by heating it to a temperature of 140° C. for ten
+ hours; and to prevent subsequent changes in resistivity, solders
+ containing zinc must be avoided, and a silver solder containing 75% of
+ silver employed in soldering the manganin wire to its connexions.
+
+The authorities of the Berlin Reichsanstalt have devoted considerable
+attention to the question of the best form for a wire standard of
+electric resistance. In that now adopted the resistance wire is
+carefully insulated and wound on a brass cylinder, being doubled on
+itself to annul inductance as much as possible. In the coil two wires
+are wound on in parallel, one being much finer than the other, and the
+final adjustment of the coil to an exact value is made by shortening the
+finer of the two. A standard of resistance for use in a laboratory now
+generally consists of a wire of manganin or platinum-silver carefully
+insulated and enclosed in a brass case. Thick copper rods are connected
+to the terminals of the wire in the interior of the case, and brought to
+the outside, being carefully insulated at the same time from one another
+and from the case. The coil so constructed can be placed under water or
+paraffin oil, the temperature of which can be exactly observed during
+the process of taking a resistance measurement. Equalization of the
+temperature of the surrounding medium is effected by the employment of a
+stirrer, worked by hand or by a small electric motor. The construction
+of a standard of electrical resistance consisting of mercury in a glass
+tube is an operation requiring considerable precautions, and only to be
+undertaken by those experienced in the matter. Opinions are divided on
+the question whether greater permanence in resistance can be secured by
+mercury-in-glass standards of resistance or by wire standards, but the
+latter are at least more portable and less fragile.
+
+ A full description of the construction of a standard wire-resistance
+ coil on the plan adopted by the Berlin Physikalisch-Technische
+ Reichsanstalt is given in the Report of the British Association
+ Committee on Electrical Standards, presented at the Edinburgh Meeting
+ in 1892. For the design and construction of standards of electric
+ resistances adapted for employment in the comparison and measurement
+ of very low or very high resistances, the reader may be referred to
+ standard treatises on electric measurements.
+
+ BIBLIOGRAPHY.--See also J. A. Fleming, _A Handbook for the Electrical
+ Laboratory and Testing Room_, vol. i. (London, 1901); _Reports of the
+ British Association Committee on Electrical Standards_, edited by
+ Fleeming Jenkin (London, 1873); A. Matthiessen and C. Vogt, "On the
+ Influence of Temperature on the Conducting Power of Alloys," _Phil.
+ Trans._, 1864, 154, p. 167, and _Phil. Mag._, 1865, 29, p. 363; A.
+ Matthiessen and M. Holtzmann, "On the Effect of the Presence of Metals
+ and Metalloids upon the Electric Conducting Power of Pure Copper,"
+ _Phil. Trans._, 1860, 150, p. 85; T. C. Fitzpatrick, "On the Specific
+ Resistance of Copper," _Brit. Assoc. Report_, 1890, p. 120, or
+ _Electrician_, 1890, 25, p. 608; R. Appleyard, _The Conductometer and
+ Electrical Conductivity_; Clark, Forde and Taylor, _Temperature
+ Coefficients of Copper_ (London, 1901). (J. A. F.)
+
+
+II. CONDUCTION IN LIQUIDS
+
+Through liquid metals, such as mercury at ordinary temperatures and
+other metals at temperatures above their melting points, the electric
+current flows as in solid metals without changing the state of the
+conductor, except in so far as heat is developed by the electric
+resistance. But another class of liquid conductors exists, and in them
+the phenomena are quite different. The conductivity of fused salts, and
+of solutions of salts and acids, although less than that of metals, is
+very great compared with the traces of conductivity found in so-called
+non-conductors. In fused salts and conducting solutions the passage of
+the current is always accompanied by definite chemical changes; the
+substance of the conductor or electrolyte is decomposed, and the
+products of the decomposition appear at the electrodes, i.e. the
+metallic plates by means of which the current is led into and out of the
+solution. The chemical phenomena are considered in the article
+ELECTROLYSIS; we are here concerned solely with the mechanism of this
+_electrolytic_ conduction of the current.
+
+To explain the appearance of the products of decomposition at the
+electrodes only, while the intervening solution is unaltered, we suppose
+that, under the action of the electric forces, the opposite parts of the
+electrolyte move in opposite directions through the liquid. These
+opposite parts, named ions by Faraday, must therefore be associated with
+electric charges, and it is the convective movement of the opposite
+streams of ions carrying their charges with them that, on this view,
+constitutes the electric current.
+
+In metallic conduction it is found that the current is proportional to
+the applied electromotive force--a relation known by the name of Ohm's
+law. If we place in a circuit with a small electromotive force an
+electrolytic cell consisting of two platinum electrodes and a solution,
+the initial current soon dies away, and we shall find that a certain
+minimum electromotive force must be applied to the circuit before any
+considerable permanent current passes. The chemical changes which are
+initiated on the surfaces of the electrodes set up a reverse
+electromotive force of polarization, and, until this is overcome, only a
+minute current, probably due to the slow but steady removal of the
+products of decomposition from the electrodes by a process of diffusion,
+will pass through the cell. Thus it is evident that, considering the
+electrolytic cell as a whole, the passage of the current through it
+cannot conform to Ohm's law. But the polarization is due to chemical
+changes, which are confined to the surfaces of the electrodes; and it is
+necessary to inquire whether, if the polarization at the electrodes be
+eliminated, the passage of the current through the bulk of the solution
+itself is proportional to the electromotive force actually applied to
+that solution. Rough experiment shows that the current is proportional
+to the excess of the electromotive force over a constant value, and thus
+verifies the law approximately, the constant electromotive force to be
+overcome being a measure of the polarization. A more satisfactory
+examination of the question was made by F. Kohlrausch in the years 1873
+to 1876. Ohm's law states that the current C is proportional to the
+electromotive force E, or C = kR, where k is a constant called the
+conductivity of the circuit. The equation may also be written as C =
+E/R, where R is a constant, the reciprocal of k, known as the resistance
+of the circuit. The essence of the law is the proportionality between C
+and E, which means that the ratio E/C is a constant. But E/C = R, and
+thus the law may be tested by examining the constancy of the measured
+resistance of a conductor when different currents are passing through
+it. In this way Ohm's law has been confirmed in the case of metallic
+conduction to a very high degree of accuracy. A similar principle was
+applied by Kohlrausch to the case of electrolytes, and he was the first
+to show that an electrolyte possesses a definite resistance which has a
+constant value when measured with different currents and by different
+experimental methods.
+
+_Measurement of the Resistance of Electrolytes._--There are two effects
+of the passage of an electric current which prevent the possibility of
+measuring electrolytic resistance by the ordinary methods with the
+direct currents which are used in the case of metals. The products of
+the chemical decomposition of the electrolyte appear at the electrodes
+and set up the opposing electromotive force of polarization, and unequal
+dilution of the solution may occur in the neighbourhood of the two
+electrodes. The chemical and electrolytic aspects of these phenomena are
+treated in the article ELECTROLYSIS, but from our present point of view
+also it is evident that they are again of fundamental importance. The
+polarization at the surface of the electrodes will set up an opposing
+electromotive force, and the unequal dilution of the solution will turn
+the electrolyte into a concentration cell and produce a subsidiary
+electromotive force either in the same direction as that applied or in
+the reverse according as the anode or the cathode solution becomes the
+more dilute. Both effects thus involve internal electromotive forces,
+and prevent the application of Ohm's law to the electrolytic cell as a
+whole. But the existence of a definite measurable resistance as a
+characteristic property of the system depends on the conformity of the
+system to Ohm's law, and it is therefore necessary to eliminate both
+these effects before attempting to measure the resistance.
+
+The usual and most satisfactory method of measuring the resistance of
+electrolytes consists in eliminating the effects of polarization by the
+use of alternating currents, that is, currents that are reversed in
+direction many times a second.[11] The chemical action produced by the
+first current is thus reversed by the second current in the opposite
+direction, and the polarization caused by the first current on the
+surface of the electrodes is destroyed before it rises to an appreciable
+value. The polarization is also diminished in another way. The
+electromotive force of polarization is due to the deposition of films of
+the products of chemical decomposition on the surface of the electrodes,
+and only reaches its full value when a continuous film is formed. If the
+current be stopped before such a film is completed, the reverse
+electromotive force is less than its full value. A given current flowing
+for a given time deposits a definite amount of substance on the
+electrodes, and therefore the amount per unit area is inversely
+proportional to the area of the electrodes--to the area of contact, that
+is, between the electrode and the liquid. Thus, by increasing the area
+of the electrodes, the polarization due to a given current is decreased.
+Now the area of free surface of a platinum plate can be increased
+enormously by coating the plate with platinum black, which is metallic
+platinum in a spongy state, and with such a plate as electrode the
+effects of polarization are diminished to a very marked extent. The
+coating is effected by passing an electric current first one way and
+then the other between two platinum plates immersed in a 3% solution of
+platinum chloride to which a trace of lead acetate is sometimes added.
+The platinized plates thus obtained are quite satisfactory for the
+investigation of strong solutions. They have the power, however, of
+absorbing a certain amount of salt from the solutions and of giving it
+up again when water or more dilute solution is placed in contact with
+them. The measurement of very dilute solutions is thus made difficult,
+but, if the plates be heated to redness after being platinized, a grey
+surface is obtained which possesses sufficient area for use with dilute
+solutions and yet does not absorb an appreciable quantity of salt.
+
+Any convenient source of alternating current may be used. The currents
+from the secondary circuit of a small induction coil are satisfactory,
+or the currents of an alternating electric light supply may be
+transformed down to an electromotive force of one or two volts. With
+such currents it is necessary to consider the effects of self-induction
+in the circuit and of electrostatic capacity. In balancing the
+resistance of the electrolyte, resistance coils may be used in which
+self-induction and the capacity are reduced to a minimum by winding the
+wire of the coil backwards and forwards in alternate layers.
+
+[Illustration: FIG. 1.]
+
+With these arrangements the usual method of measuring resistance by
+means of Wheatstone's bridge may be adapted to the case of electrolytes.
+With alternating currents, however, it is impossible to use a
+galvanometer in the usual way. The galvanometer was therefore replaced
+by Kohlrausch by a telephone, which gives a sound when an alternating
+current passes through it. The most common plan of the apparatus is
+shown diagrammatically in fig. 1. The electrolytic cell and a resistance
+box form two arms of the bridge, and the sliding contact is moved along
+the metre wire which forms the other two arms till no sound is heard in
+the telephone. The resistance of the electrolyte is to that of the box
+as that of the right-hand end of the wire is to that of the left-hand
+end. A more accurate method of using alternating currents, and one more
+pleasant to use, gets rid of the telephone (_Phil. Trans._, 1900, 194,
+p. 321). The current from one or two voltaic cells is led to an ebonite
+drum turned by a motor or a hand-wheel and cord. On the drum are fixed
+brass strips with wire brushes touching them in such a manner that the
+current from the brushes is reversed several times in each revolution of
+the drum. The wires from the brushes are connected with the Wheatstone's
+bridge. A moving coil galvanometer is used as indicator, its connexions
+being reversed in time with those of the battery by a slightly narrower
+set of brass strips fixed on the other end of the ebonite commutator.
+Thus any residual current through the galvanometer is direct and not
+alternating. The high moment of inertia of the coil makes the period of
+swing slow compared with the period of alternation of the current, and
+the slight periodic disturbances are thus prevented from affecting the
+galvanometer. When the measured resistance is not altered by increasing
+the speed of the commutator or changing the ratio of the arms of the
+bridge, the disturbing effects may be considered to be eliminated.
+
+[Illustration: FIG. 2.]
+
+[Illustration: FIG. 3.]
+
+The form of vessel chosen to contain the electrolyte depends on the
+order of resistance to be measured. For dilute solutions the shape of
+cell shown in fig. 2 will be found convenient, while for more
+concentrated solutions, that indicated in fig. 3 is suitable. The
+absolute resistances of certain solutions have been determined by
+Kohlrausch by comparison with mercury, and, by using one of these
+solutions in any cell, the constant of that cell may be found once for
+all. From the observed resistance of any given solution in the cell the
+resistance of a centimetre cube--the so-called specific resistance--may
+be calculated. The reciprocal of this, or the conductivity, is a more
+generally useful constant; it is conveniently expressed in terms of a
+unit equal to the reciprocal of an ohm. Thus Kohlrausch found that a
+solution of potassium chloride, containing one-tenth of a gram
+equivalent (7.46 grams) per litre, has at 18° C. a specific resistance
+of 89.37 ohms per centimetre cube, or a conductivity of 1.119 × 10^-2
+mhos or 1.119 × 10^-11 C.G.S. units. As the temperature variation of
+conductivity is large, usually about 2% per degree, it is necessary to
+place the resistance cell in a paraffin or water bath, and to observe
+its temperature with some accuracy.
+
+Another way of eliminating the effects of polarization and of dilution
+has been used by W. Stroud and J. B. Henderson (_Phil. Mag._, 1897 [5],
+43, p. 19). Two of the arms of a Wheatstone's bridge are composed of
+narrow tubes filled with the solution, the tubes being of equal diameter
+but of different length. The other two arms are made of coils of wire of
+equal resistance, and metallic resistance is added to the shorter tube
+till the bridge is balanced. Direct currents of somewhat high
+electromotive force are used to work the bridge. Equal currents then
+flow through the two tubes; the effects of polarization and dilution
+must be the same in each, and the resistance added to the shorter tube
+must be equal to the resistance of a column of liquid the length of
+which is equal to the difference in length of the two tubes.
+
+A somewhat different principle was adopted by E. Bouty in 1884. If a
+current be passed through two resistances in series by means of an
+applied electromotive force, the electric potential falls from one end
+of the resistances to the other, and, if we apply Ohm's law to each
+resistance in succession, we see that, since for each of them E = CR,
+and C the current is the same through both, E the electromotive force or
+fall of potential between the ends of each resistance must be
+proportional to the resistance between them. Thus by measuring the
+potential difference between the ends of the two resistances
+successively, we may compare their resistances. If, on the other hand,
+we can measure the potential difference in some known units, and
+similarly measure the current flowing, we can determine the resistance
+of a single electrolyte. The details of the apparatus may vary, but its
+principle is illustrated in the following description. A narrow glass
+tube is fixed horizontally into side openings in two glass vessels, and
+an electric current passed through it by means of platinum electrodes
+and a battery of considerable electromotive force. In this way a steady
+fall of electric potential is set up along the length of the tube. To
+measure the potential difference between the ends of the tube, tapping
+electrodes are constructed, e.g. by placing zinc rods in vessels with
+zinc sulphate solution and connecting these vessels (by means of thin
+siphon tubes also filled with solution) with the vessels at the ends of
+the long tube which contains the electrolyte to be examined. Whatever be
+the contact potential difference between zinc and its solution, it is
+the same at both ends, and thus the potential difference between the
+zinc rods is equal to that between the liquid at the two ends of the
+tube. This potential difference may be measured without passing any
+appreciable current through the tapping electrodes, and thus the
+resistance of the liquid deduced.
+
+_Equivalent Conductivity of Solutions._--As is the case in the other
+properties of solutions, the phenomena are much more simple when the
+concentration is small than when it is great, and a study of dilute
+solutions is therefore the best way of getting an insight into the
+essential principles of the subject. The foundation of our knowledge was
+laid by Kohlrausch when he had developed the method of measuring
+electrolyte resistance described above. He expressed his results in
+terms of "equivalent conductivity," that is, the conductivity (k) of the
+solution divided by the number (m) of gram-equivalents of electrolyte
+per litre. He finds that, as the concentration diminishes, the value of
+k/m approaches a limit, and eventually becomes constant, that is to say,
+at great dilution the conductivity is proportional to the concentration.
+Kohlrausch first prepared very pure water by repeated distillation and
+found that its resistance continually increased as the process of
+purification proceeded. The conductivity of the water, and of the slight
+impurities which must always remain, was subtracted from that of the
+solution made with it, and the result, divided by m, gave the equivalent
+conductivity of the substance dissolved. This procedure appears
+justifiable, for as long as conductivity is proportional to
+concentration it is evident that each part of the dissolved matter
+produces its own independent effect, so that the total conductivity is
+the sum of the conductivities of the parts; when this ceases to hold,
+the concentration of the solution has in general become so great that
+the conductivity of the solvent may be neglected. The general result of
+these experiments can be represented graphically by plotting k/m as
+ordinates and [root 3]m as abscissae, [root 3]m being a number
+proportional to the reciprocal of the average distance between the
+molecules, to which it seems likely that the molecular conductivity may
+be related. The general types of curve for a simple neutral salt like
+potassium or sodium chloride and for a caustic alkali or acid are shown
+in fig. 4. The curve for the neutral salt comes to a limiting value;
+that for the acid attains a maximum at a certain very small
+concentration, and falls again when the dilution is carried farther. It
+has usually been considered that this destruction of conductivity is due
+to chemical action between the acid and the residual impurities in the
+water. At such great dilution these impurities are present in quantities
+comparable with the amount of acid which they convert into a less highly
+conducting neutral salt. In the case of acids, then, the maximum must be
+taken as the limiting value. The decrease in equivalent conductivity at
+great dilution is, however, so constant that this explanation seems
+insufficient. The true cause of the phenomenon may perhaps be connected
+with the fact that the bodies in which it occurs, acids and alkalis,
+contain the ions, hydrogen in the one case, hydroxyl in the other, which
+are present in the solvent, water, and have, perhaps because of this
+relation, velocities higher than those of any other ions. The values of
+the molecular conductivities of all neutral salts are, at great
+dilution, of the same order of magnitude, while those of acids at their
+maxima are about three times as large. The influence of increasing
+concentration is greater in the case of salts containing divalent ions,
+and greatest of all in such cases as solutions of ammonia and acetic
+acid, which are substances of very low conductivity.
+
+[Illustration: FIG. 4.]
+
+_Theory of Moving Ions._--Kohlrausch found that, when the polarization
+at the electrodes was eliminated, the resistance of a solution was
+constant however determined, and thus established Ohm's Law for
+electrolytes. The law was confirmed in the case of strong currents by G.
+F. Fitzgerald and F. T. Trouton (_B.A. Report_, 1886, p. 312). Now,
+Ohm's Law implies that no work is done by the current in overcoming
+reversible electromotive forces such as those of polarization. Thus the
+molecular interchange of ions, which must occur in order that the
+products may be able to work their way through the liquid and appear at
+the electrodes, continues throughout the solution whether a current is
+flowing or not. The influence of the current on the ions is merely
+directive, and, when it flows, streams of electrified ions travel in
+opposite directions, and, if the applied electromotive force is enough
+to overcome the local polarization, give up their charges to the
+electrodes. We may therefore represent the facts by considering the
+process of electrolysis to be a kind of convection. Faraday's classical
+experiments proved that when a current flows through an electrolyte the
+quantity of substance liberated at each electrode is proportional to its
+chemical equivalent weight, and to the total amount of electricity
+passed. Accurate determinations have since shown that the mass of an ion
+deposited by one electromagnetic unit of electricity, i.e. its
+electro-chemical equivalent, is 1.036 × 10^-4 × its chemical equivalent
+weight. Thus the amount of electricity associated with one
+gram-equivalent of any ion is 10^4/1.036 = 9653 units. Each monovalent
+ion must therefore be associated with a certain definite charge, which
+we may take to be a natural unit of electricity; a divalent ion carries
+two such units, and so on. A cation, i.e. an ion giving up its charge at
+the cathode, as the electrode at which the current leaves the solution
+is called, carries a positive charge of electricity; an anion,
+travelling in the opposite direction, carries a negative charge. It will
+now be seen that the quantity of electricity flowing per second, i.e.
+the current through the solution, depends on (1) the number of the ions
+concerned, (2) the charge on each ion, and (3) the velocity with which
+the ions travel past each other. Now, the number of ions is given by the
+concentration of the solution, for even if all the ions are not actively
+engaged in carrying the current at the same instant, they must, on any
+dynamical idea of chemical equilibrium, be all active in turn. The
+charge on each, as we have seen, can be expressed in absolute units, and
+therefore the velocity with which they move past each other can be
+calculated. This was first done by Kohlrausch (_Göttingen Nachrichten_,
+1876, p. 213, and _Das Leitvermögen der Elektrolyte_, Leipzig, 1898)
+about 1879.
+
+ In order to develop Kohlrausch's theory, let us take, as an example,
+ the case of an aqueous solution of potassium chloride, of
+ concentration n gram-equivalents per cubic centimetre. There will then
+ be n gram-equivalents of potassium ions and the same number of
+ chlorine ions in this volume. Let us suppose that on each
+ gram-equivalent of potassium there reside +e units of electricity, and
+ on each gram-equivalent of chlorine ions -e units. If u denotes the
+ average velocity of the potassium ion, the positive charge carried per
+ second across unit area normal to the flow is n e u. Similarly, if v
+ be the average velocity of the chlorine ions, the negative charge
+ carried in the opposite direction is n e v. But positive electricity
+ moving in one direction is equivalent to negative electricity moving
+ in the other, so that, before changes in concentration sensibly
+ supervene, the total current, C, is ne(u + v). Now let us consider the
+ amounts of potassium and chlorine liberated at the electrodes by this
+ current. At the cathode, if the chlorine ions were at rest, the excess
+ of potassium ions would be simply those arriving in one second,
+ namely, nu. But since the chlorine ions move also, a further
+ separation occurs, and nv potassium ions are left without partners.
+ The total number of gram-equivalents liberated is therefore n(u + v).
+ By Faraday's law, the number of grams liberated is equal to the
+ product of the current and the electro-chemical equivalent of the ion;
+ the number of gram-equivalents therefore must be equal to [eta]C,
+ where [eta] denotes the electro-chemical equivalent of hydrogen in
+ C.G.S. units. Thus we get
+
+ n(u + v) = [eta]C = [eta]ne(u + v),
+
+ and it follows that the charge, e, on 1 gram-equivalent of each kind
+ of ion is equal to 1/[eta]. We know that Ohm's Law holds good for
+ electrolytes, so that the current C is also given by k·dP/dx, where k
+ denotes the conductivity of the solution, and dP/dx the potential
+ gradient, i.e. the change in potential per unit length along the lines
+ of current flow. Thus
+
+ n
+ -----(u + v) = kdP/dx;
+ [eta]
+
+ therefore
+
+ k dP
+ u + v = [eta] - --.
+ n dx
+
+ Now [eta] is 1.036 × 10^-4, and the concentration of a solution is
+ usually expressed in terms of the number, m, of gram-equivalents per
+ litre instead of per cubic centimetre. Therefore
+
+ k dP
+ u + v = 1.036 × 10^-1 - -- --.
+ m dx
+
+ When the potential gradient is one volt (10^8 C.G.S. units) per
+ centimetre this becomes
+
+ u + v = 1.036 × 10^-7 × k/m.
+
+ Thus by measuring the value of k/m, which is known as the equivalent
+ conductivity of the solution, we can find u + v, the velocity of the
+ ions relative to each other. For instance, the equivalent conductivity
+ of a solution of potassium chloride containing one-tenth of a
+ gram-equivalent per litre is 1119 × 10^-13 C.G.S. units at 18° C.
+ Therefore
+
+ u + v = 1.036 × 10^7 × 1119 × 10^-13
+ = 1.159 × 10^-3 = 0.001159 cm. per sec.
+
+ In order to obtain the absolute velocities u and v, we must find some
+ other relation between them. Let us resolve u into ½(u + v) in one
+ direction, say to the right, and ½(u - v) to the left. Similarly v can
+ be resolved into ½(v+u) to the left and ½(v-u) to the right. On
+ pairing these velocities we have a combined movement of the ions to
+ the right, with a speed of ½(u - v) and a drift right and left, past
+ each other, each ion travelling with a speed of ½(u + v), constituting
+ the electrolytic separation. If u is greater than v, the combined
+ movement involves a concentration of salt at the cathode, and a
+ corresponding dilution at the anode, and _vice versa_. The rate at
+ which salt is electrolysed, and thus removed from the solution at each
+ electrode, is ½(u + v). Thus the total loss of salt at the cathode is
+ ½(u + v) - ½(u - v) or v, and at the anode, ½(v + u) - ½(v - u), or u.
+ Therefore, as is explained in the article ELECTROLYSIS, by measuring
+ the dilution of the liquid round the electrodes when a current passed,
+ W. Hittorf (_Pogg. Ann._, 1853-1859, 89, p. 177; 98, p. 1; 103, p. 1;
+ 106, pp. 337 and 513) was able to deduce the ratio of the two
+ velocities, for simple salts when no complex ions are present, and
+ many further experiments have been made on the subject (see _Das
+ Leitvermögen der Elektrolyte_).
+
+ By combining the results thus obtained with the sum of the velocities,
+ as determined from the conductivities, Kohlrausch calculated the
+ absolute velocities of different ions under stated conditions. Thus,
+ in the case of the solution of potassium chloride considered above,
+ Hittorf's experiments show us that the ratio of the velocity of the
+ anion to that of the cation in this solution is .51 : .49. The
+ absolute velocity of the potassium ion under unit potential gradient
+ is therefore 0.000567 cm. per sec., and that of the chlorine ion
+ 0.000592 cm. per sec. Similar calculations can be made for solutions
+ of other concentrations, and of different substances.
+
+Table IX. shows Kohlrausch's values for the ionic velocities of three
+chlorides of alkali metals at 18° C, calculated for a potential gradient
+of 1 volt per cm.; the numbers are in terms of a unit equal to 10^-6 cm.
+per sec.:--
+
+ TABLE IX.
+
+ +--------------------------------------------+----------------+
+ | | KCl | NaCl | LiCl |
+ +---------+-----------------+----------------+----------------+
+ | m | u + v u v | u + v u v |u + v u v |
+ +---------+-----------------+----------------+----------------+
+ | 0 | 1350 660 690 | 1140 450 690 | 1050 360 690 |
+ | 0.0001 | 1335 654 681 | 1129 448 681 | 1037 356 681 |
+ | .001 | 1313 643 670 | 1110 440 670 | 1013 343 670 |
+ | .01 | 1263 619 644 | 1059 415 644 | 962 318 644 |
+ | .03 | 1218 597 621 | 1013 390 623 | 917 298 619 |
+ | .1 | 1153 564 589 | 952 360 592 | 853 259 594 |
+ | .3 | 1088 531 557 | 876 324 552 | 774 217 557 |
+ | 1.0 | 1011 491 520 | 765 278 487 | 651 169 482 |
+ | 3.0 | 911 442 469 | 582 206 376 | 463 115 348 |
+ | 5.0 | | 438 153 285 | 334 80 254 |
+ | 10.0 | | | 117 25 92 |
+ +---------+-----------------+----------------+----------------+
+
+These numbers show clearly that there is an increase in ionic velocity
+as the dilution proceeds. Moreover, if we compare the values for the
+chlorine ion obtained from observations on these three different salts,
+we see that as the concentrations diminish the velocity of the chlorine
+ion becomes the same in all of them. A similar relation appears in other
+cases, and, in general, we may say that at great dilution the velocity
+of an ion is independent of the nature of the other ion present. This
+introduces the conception of specific ionic velocities, for which some
+values at 18° C. are given by Kohlrausch in Table X.:--
+
+ Table X.
+
+ +------------------------------------+
+ | K 66 × 10^-5 cms. per sec. |
+ | Na 45 " " |
+ | Li 36 " " |
+ | NH4 66 " " |
+ | H 320 " " |
+ | Ag 57 " " |
+ | Cl 69 " " |
+ | I 69 " " |
+ | NO3 64 " " |
+ | OH 162 " " |
+ | C2H3O2 36 " " |
+ | C3H5O2 33 " " |
+ +------------------------------------+
+
+Having obtained these numbers we can deduce the conductivity of the
+dilute solution of any salt, and the comparison of the calculated with
+the observed values furnished the first confirmation of Kohlrausch's
+theory. Some exceptions, however, are known. Thus acetic acid and
+ammonia give solutions of much lower conductivity than is indicated by
+the sum of the specific ionic velocities of their ions as determined
+from other compounds. An attempt to find in Kohlrausch's theory some
+explanation of this discrepancy shows that it could be due to one of two
+causes. Either the velocities of the ions must be much less in these
+solutions than in others, or else only a fractional part of the number
+of molecules present can be actively concerned in conveying the current.
+We shall return to this point later.
+
+ _Friction on the Ions._--It is interesting to calculate the magnitude
+ of the forces required to drive the ions with a certain velocity. If
+ we have a potential gradient of 1 volt per centimetre the electric
+ force is 10^8 in C.G.S. units. The charge of electricity on 1
+ gram-equivalent of any ion is 1/.0001036 = 9653 units, hence the
+ mechanical force acting on this mass is 9653 × 10^8 dynes. This, let
+ us say, produces a velocity u; then the force required to produce unit
+ velocity is PA = 9.653 × 10^11/u dynes = 9.84 × 10^5/u kilograms-weight.
+ If the ion have an equivalent weight A, the force producing unit
+ velocity when acting on 1 gram is P1 = 9.84 × 10^5/Au kilograms-weight.
+ Thus the aggregate force required to drive 1 gram of potassium ions
+ with a velocity of 1 centimetre per second through a very dilute
+ solution must be equal to the weight of 38 million kilograms.
+
+ Table XI.
+
+ +-------------------------------+
+ | Kilograms-weight. |
+ +-------------------------------+
+ | PA P1 |
+ |K 15 × 10^8 38 × 10^6 |
+ |Na 22 " 95 " |
+ |Li 27 " 390 " |
+ |NH4 15 " 83 " |
+ |H 3.1 " 310 " |
+ |Ag 17 " 16 " |
+ |Cl 14 " 40 " |
+ |I 14 " 11 " |
+ |NO3 15 " 25 " |
+ |OH 5.4 " 32 " |
+ |C2H8O2 27 " 46 " |
+ |C3H5O2 30 " 41 " |
+ +------------- -----------------+
+
+ Since the ions move with uniform velocity, the frictional resistances
+ brought into play must be equal and opposite to the driving forces,
+ and therefore these numbers also represent the ionic friction
+ coefficients in very dilute solutions at 18° C.
+
+_Direct Measurement of Ionic Velocities._--Sir Oliver Lodge was the
+first to directly measure the velocity of an ion (_B.A. Report_, 1886,
+p. 389). In a horizontal glass tube connecting two vessels filled with
+dilute sulphuric acid he placed a solution of sodium chloride in solid
+agar-agar jelly. This solid solution was made alkaline with a trace of
+caustic soda in order to bring out the red colour of a little
+phenol-phthalein added as indicator. An electric current was then passed
+from one vessel to the other. The hydrogen ions from the anode vessel of
+acid were thus carried along the tube, and, as they travelled,
+decolourized the phenol-phthalein. By this method the velocity of the
+hydrogen ion through a jelly solution under a known potential gradient
+was observed to about 0.0026 cm. per sec, a number of the same order as
+that required by Kohlrausch's theory. Direct determinations of the
+velocities of a few other ions have been made by W. C. D. Whetham
+(_Phil. Trans._ vol. 184, A, p. 337; vol. 186, A, p. 507; _Phil. Mag._,
+October 1894). Two solutions having one ion in common, of equivalent
+concentrations, different densities, different colours, and nearly equal
+specific resistances, were placed one over the other in a vertical glass
+tube. In one case, for example, decinormal solutions of potassium
+carbonate and potassium bichromate were used. The colour of the latter
+is due to the presence of the bichromate group, Cr2O7. When a current
+was passed across the junction, the anions CO3 and Cr2O7 travelled in
+the direction opposite to that of the current, and their velocity could
+be determined by measuring the rate at which the colour boundary moved.
+Similar experiments were made with alcoholic solutions of cobalt salts,
+in which the velocities of the ions were found to be much less than in
+water. The behaviour of agar jelly was then investigated, and the
+velocity of an ion through a solid jelly was shown to be very little
+less than in an ordinary liquid solution. The velocities could therefore
+be measured by tracing the change in colour of an indicator or the
+formation of a precipitate. Thus decinormal jelly solutions of barium
+chloride and sodium chloride, the latter containing a trace of sodium
+sulphate, were placed in contact. Under the influence of an
+electromotive force the barium ions moved up the tube, disclosing their
+presence by the trace of insoluble barium sulphate formed. Again, a
+measurement of the velocity of the hydrogen ion, when travelling through
+the solution of an acetate, showed that its velocity was then only about
+the one-fortieth part of that found during its passage through
+chlorides. From this, as from the measurements on alcohol solutions, it
+is clear that where the equivalent conductivities are very low the
+effective velocities of the ions are reduced in the same proportion.
+
+Another series of direct measurements has been made by Orme Masson
+(_Phil. Trans._ vol. 192, A, p. 331). He placed the gelatine solution of
+a salt, potassium chloride, for example, in a horizontal glass tube, and
+found the rate of migration of the potassium and chlorine ions by
+observing the speed at which they were replaced when a coloured anion,
+say, the Cr2O7 from a solution of potassium bichromate, entered the tube
+at one end, and a coloured cation, say, the Cu from copper sulphate, at
+the other. The coloured ions are specifically slower than the colourless
+ions which they follow, and in this case it follows that the coloured
+solution has a higher resistance than the colourless. For the same
+current, therefore, the potential gradient is higher in the coloured
+solution and lower in the colourless one. Thus a coloured ion which gets
+in front of the advancing boundary finds itself acted on by a smaller
+force and falls back into line, while a straggling colourless ion is
+pushed forward again. Hence a sharp boundary is preserved. B. D. Steele
+has shown that with these sharp boundaries the use of coloured ions is
+unnecessary, the junction line being visible owing to the difference in
+the optical refractive indices of two colourless solutions. Once the
+boundary is formed, too, no gelatine is necessary, and the motion can be
+watched through liquid aqueous solutions (see R. B. Denison and B. D.
+Steele, _Phil. Trans._, 1906).
+
+All the direct measurements which have been made on simple binary
+electrolytes agree with Kohlrausch's results within the limits of
+experimental error. His theory, therefore, probably holds good in such
+cases, whatever be the solvent, if the proper values are given to the
+ionic velocities, i.e. the values expressing the velocities with which
+the ions actually move in the solution of the strength taken, and under
+the conditions of the experiment. If we know the specific velocity of
+any one ion, we can deduce, from the conductivity of very dilute
+solutions, the velocity of any other ion with which it may be
+associated, a proceeding which does not involve the difficult task of
+determining the migration constant of the compound. Thus, taking the
+specific ionic velocity of hydrogen as 0.00032 cm. per second, we can
+find, by determining the conductivity of dilute solutions of any acid,
+the specific velocity of the acid radicle involved. Or again, since we
+know the specific velocity of silver, we can find the velocities of a
+series of acid radicles at great dilution by measuring the conductivity
+of their silver salts.
+
+ By such methods W. Ostwald, G. Bredig and other observers have found
+ the specific velocities of many ions both of inorganic and organic
+ compounds, and examined the relation between constitution and ionic
+ velocity. The velocity of elementary ions is found to be a periodic
+ function of the atomic weight, similar elements lying on corresponding
+ portions of a curve drawn to express the relation between these two
+ properties. Such a curve much resembles that giving the relation
+ between atomic weight and viscosity in solution. For complex ions the
+ velocity is largely an additive property; to a continuous additive
+ change in the composition of the ion corresponds a continuous but
+ decreasing change in the velocity. The following table gives Ostwald's
+ results for the formic acid series:--
+
+ Table XII.
+
+ +----------------------+----------+---------------------+
+ | | Velocity.| Difference for CH2. |
+ +----------------------+----------+---------------------+
+ | Formic acid HCO2 | 51.2 | .. |
+ | Acetic " H3C2O2 | 38.3 | -12.9 |
+ | Propionic " H5C3O2 | 34.3 | - 4.0 |
+ | Butyric " H7C4O2 | 30.8 | - 3.5 |
+ | Valeric " H9C5O2 | 28.8 | - 2.0 |
+ | Caprionic " H11C6O2 | 27.4 | - 1.4 |
+ +----------------------+----------+---------------------+
+
+_Nature of Electrolytes._--We have as yet said nothing about the
+fundamental cause of electrolytic activity, nor considered why, for
+example, a solution of potassium chloride is a good conductor, while a
+solution of sugar allows practically no current to pass.
+
+All the preceding account of the subject is, then, independent of any
+view we may take of the nature of electrolytes, and stands on the basis
+of direct experiment. Nevertheless, the facts considered point to a very
+definite conclusion. The specific velocity of an ion is independent of
+the nature of the opposite ion present, and this suggests that the ions
+themselves, while travelling through the liquid, are dissociated from
+each other. Further evidence, pointing in the same direction, is
+furnished by the fact that since the conductivity is proportional to the
+concentration at great dilution, the equivalent-conductivity, and
+therefore the ionic velocity, is independent of it. The importance of
+this relation will be seen by considering the alternative to the
+dissociation hypothesis. If the ions are not permanently free from each
+other their mobility as parts of the dissolved molecules must be secured
+by continual interchanges. The velocity with which they work their way
+through the liquid must then increase as such molecular rearrangements
+become more frequent, and will therefore depend on the number of solute
+molecules, i.e. on the concentration. On this supposition the observed
+constancy of velocity would be impossible. We shall therefore adopt as a
+wording hypothesis the theory, confirmed by other phenomena (see
+ELECTROLYSIS), that an electrolyte consists of dissociated ions.
+
+It will be noticed that neither the evidence in favour of the
+dissociation theory which is here considered, nor that described in the
+article ELECTROLYSIS, requires more than the effective dissociation of
+the ions from each other. They may well be connected in some way with
+solvent molecules, and there are several indications that an ion
+consists of an electrified part of the molecule of the dissolved salt
+with an attendant atmosphere of solvent round it. The conductivity of a
+salt solution depends on two factors--(1) the fraction of the salt
+ionized; (2) the velocity with which the ions, when free from each
+other, move under the electric forces.[12] When a solution is heated,
+both these factors may change. The coefficient of ionization usually,
+though not always, decreases; the specific ionic velocities increase.
+Now the rate of increase with temperature of these ionic velocities is
+very nearly identical with the rate of decrease of the viscosity of the
+liquid. If the curves obtained by observations at ordinary temperatures
+be carried on they indicate a zero of fluidity and a zero of ionic
+velocity about the same point, 38.5° C. below the freezing point of
+water (Kohlrausch, _Sitz. preuss. Akad. Wiss._, 1901, 42, p. 1026). Such
+relations suggest that the frictional resistance to the motion of an ion
+is due to the ordinary viscosity of the liquid, and that the ion is
+analogous to a body of some size urged through a viscous medium rather
+than to a particle of molecular dimensions finding its way through a
+crowd of molecules of similar magnitude. From this point of view W. K.
+Bousfield has calculated the sizes of ions on the assumption that
+Stokes's theory of the motion of a small sphere through a viscous medium
+might be applied (_Zeits. phys. Chem._, 1905, 53, p. 257; _Phil. Trans._
+A, 1906, 206, p. 101). The radius of the potassium or chlorine ion with
+its envelope of water appears to be about 1.2 × 10^-8 centimetres.
+
+ For the bibliography of electrolytic conduction see ELECTROLYSIS. The
+ books which deal more especially with the particular subject of the
+ present article are _Das Leitvermögen der Elektrolyte_, by F.
+ Kohlrausch and L. Holborn (Leipzig, 1898), and _The Theory of Solution
+ and Electrolysis_, by W. C. D. Whetham (Cambridge, 1902).
+ (W. C. D. W.)
+
+
+III. ELECTRIC CONDUCTION THROUGH GASES
+
+A gas such as air when it is under normal conditions conducts
+electricity to a small but only to a very small extent, however small
+the electric force acting on the gas may be. The electrical conductivity
+of gases not exposed to special conditions is so small that it was only
+definitely established in the early years of the 20th century, although
+it had engaged the attention of physicists for more than a hundred
+years. It had been known for a long time that a body charged with
+electricity slowly lost its charge even when insulated with the greatest
+care, and though long ago some physicists believed that part of the leak
+of electricity took place through the air, the general view seems to
+have been that it was due to almost unavoidable defects in the
+insulation or to dust in the air, which after striking the charged body
+was repelled from it and went off with some of the charge. C. A.
+Coulomb, who made some very careful experiments which were published in
+1785 (_Mém. de l'Acad. des Sciences_, 1785, p. 612), came to the
+conclusion that after allowing for the leakage along the threads which
+supported the charged body there was a balance over, which he attributed
+to leakage through the air. His view was that when the molecules of air
+come into contact with a charged body some of the electricity goes on to
+the molecules, which are then repelled from the body carrying their
+charge with them. We shall see later that this explanation is not
+tenable. C. Matteucci (_Ann. chim. phys._, 1850, 28, p. 390) in 1850
+also came to the conclusion that the electricity from a charged body
+passes through the air; he was the first to prove that the rate at
+which electricity escapes is less when the pressure of the gas is low
+than when it is high. He found that the rate was the same whether the
+charged body was surrounded by air, carbonic acid or hydrogen.
+Subsequent investigations have shown that the rate in hydrogen is in
+general much less than in air. Thus in 1872 E. G. Warburg (_Pogg. Ann._,
+1872, 145, p. 578) found that the leak through hydrogen was only about
+one-half of that through air: he confirmed Matteucci's observations on
+the effect of pressure on the rate of leak, and also found that it was
+the same whether the gas was dry or damp. He was inclined to attribute
+the leak to dust in the air, a view which was strengthened by an
+experiment of J. W. Hittorf's (_Wied. Ann._, 1879, 7, p. 595), in which
+a small carefully insulated electroscope, placed in a small vessel
+filled with carefully filtered gas, retained its charge for several
+days; we know now that this was due to the smallness of the vessel and
+not to the absence of dust, as it has been proved that the rate of leak
+in small vessels is less than in large ones.
+
+Great light was thrown on this subject by some experiments on the rates
+of leak from charged bodies in closed vessels made almost simultaneously
+by H. Geitel (_Phys. Zeit._, 1900, 2, p. 116) and C. T. R. Wilson
+(_Proc. Camb. Phil._ Soc., 1900, 11, p. 32). These observers established
+that (1) the rate of escape of electricity in a closed vessel is much
+smaller than in the open, and the larger the vessel the greater is the
+rate of leak; and (2) the rate of leak does not increase in proportion
+to the differences of potential between the charged body and the walls
+of the vessel: the rate soon reaches a limit beyond which it does not
+increase, however much the potential difference may be increased,
+provided, of course, that this is not great enough to cause sparks to
+pass from the charged body. On the assumption that the maximum leak is
+proportional to the volume, Wilson's experiments, which were made in
+vessels less than 1 litre in volume, showed that in dust-free air at
+atmospheric pressure the maximum quantity of electricity which can
+escape in one second from a charged body in a closed volume of V cubic
+centimetres is about 10^-8V electrostatic units. E. Rutherford and S. T.
+Allan (_Phys. Zeit._, 1902, 3, p. 225), working in Montreal, obtained
+results in close agreement with this. Working between pressures of from
+43 to 743 millimetres of mercury, Wilson showed that the maximum rate of
+leak is very approximately proportional to the pressure; it is thus
+exceedingly small when the pressure is low--a result illustrated in a
+striking way by an experiment of Sir W. Crookes (_Proc. Roy. Soc._,
+1879, 28, p. 347) in which a pair of gold leaves retained an electric
+charge for several months in a very high vacuum. Subsequent experiments
+have shown that it is only in very small vessels that the rate of leak
+is proportional to the volume and to the pressure; in large vessels the
+rate of leak per unit volume is considerably smaller than in small ones.
+In small vessels the maximum rate of leak in different gases, is, with
+the exception of hydrogen, approximately proportional to the density of
+the gas. Wilson's results on this point are shown in the following table
+(Proc. Roy. Soc., 1901, 60, p. 277):--
+
+ +---------+------------------------+-----------------+
+ | Gas. | Relative Rate of Leak. | _Rate of Leak._ |
+ | | | Sp. Gr. |
+ +---------+------------------------+-----------------+
+ | Air | 1.00 | 1 |
+ | H2 | .184 | 2.7 |
+ | CO2 | 1.69 | 1.10 |
+ | SO2 | 2.64 | 1.21 |
+ | CH3Cl | 4.7 | 1.09 |
+ | Ni(CO)4 | 5.1 | .867 |
+ +---------+------------------------+-----------------+
+
+The rate of leak of electricity through gas contained in a closed vessel
+depends to some extent on the material of which the walls of the vessel
+are made; thus it is greater, other circumstances being the same, when
+the vessel is made of lead than when it is made of aluminium. It also
+varies, as Campbell and Wood (_Phil. Mag._ [6], 13, p. 265) have shown,
+with the time of the day, having a well-marked minimum at about 3
+o'clock in the morning: it also varies from month to month. Rutherford
+(_Phys. Rev._, 1903, 16, p. 183), Cooke (_Phil. Mag._, 1903 [6], 6, p.
+403) and M'Clennan and Burton (_Phys. Rev._, 1903, 16, p. 184) have
+shown that the leak in a closed vessel can be reduced by about 30% by
+surrounding the vessel with sheets of thick lead, but that the reduction
+is not increased beyond this amount, however thick the lead sheets may
+be. This result indicates that part of the leak is due to a very
+penetrating kind of radiation, which can get through the thin walls of
+the vessel but is stopped by the thick lead. A large part of the leak we
+are describing is due to the presence of radioactive substances such as
+radium and thorium in the earth's crust and in the walls of the vessel,
+and to the gaseous radioactive emanations which diffuse from them into
+the atmosphere. This explains the very interesting effect discovered by
+J. Elster and H. Geitel (_Phys. Zeit._, 1901, 2, p. 560), that the rate
+of leak in caves and cellars when the air is stagnant and only renewed
+slowly is much greater than in the open air. In some cases the
+difference is very marked; thus they found that in the cave called the
+Baumannshöhle in the Harz mountains the electricity escaped at seven
+times the rate it did in the air outside. In caves and cellars the
+radioactive emanations from the walls can accumulate and are not blown
+away as in the open air.
+
+The electrical conductivity of gases in the normal state is, as we have
+seen, exceedingly small, so small that the investigation of its
+properties is a matter of considerable difficulty; there are, however,
+many ways by which the electrical conductivity of a gas can be increased
+so greatly that the investigation becomes comparatively easy. Among such
+methods are raising the temperature of the gas above a certain point.
+Gases drawn from the neighbourhood of flames, electric arcs and sparks,
+or glowing pieces of metal or carbon are conductors, as are also gases
+through which Röntgen or cathode rays or rays of positive electricity
+are passing; the rays from the radioactive metals, radium, thorium,
+polonium and actinium, produce the same effect, as does also
+ultra-violet light of exceedingly short wave-length. The gas, after
+being made a conductor of electricity by any of these means, is found to
+possess certain properties; thus it retains its conductivity for some
+little time after the agent which made it a conductor has ceased to act,
+though the conductivity diminishes very rapidly and finally gets too
+small to be appreciable.
+
+[Illustration: FIG. 5.]
+
+This and several other properties of conducting gas may readily be
+proved by the aid of the apparatus represented in fig. 5. V is a testing
+vessel in which an electroscope is placed. Two tubes A and C are fitted
+into the vessel, A being connected with a water pump, while the far end
+of C is in the region where the gas is exposed to the agent which makes
+it a conductor of electricity. Let us suppose that the gas is made
+conducting by Röntgen rays produced by a vacuum tube which is placed in
+a box, covered except for a window at B with lead so as to protect the
+electroscope from the direct action of the rays. If a slow current of
+air is drawn by the water pump through the testing vessel, the charge on
+the electroscope will gradually leak away. The leak, however, ceases
+when the current of air is stopped. This result shows that the gas
+retains its conductivity during the time taken by it to pass from one
+end to the other of the tube C.
+
+The gas loses its conductivity when filtered through a plug of
+glass-wool, or when it is made to bubble through water. This can readily
+be proved by inserting in the tube C a plug of glass-wool or a water
+trap; then if by working the pump a little harder the same current of
+air is produced as before, it will be found that the electroscope will
+now retain its charge, showing that the conductivity can, as it were, be
+filtered out of the gas. The conductivity can also be removed from the
+gas by making the gas traverse a strong electric field. We can show this
+by replacing the tube C by a metal tube with an insulated wire passing
+down the axis of the tube. If there is no potential difference between
+the wire and the tube then the electroscope will leak when a current of
+air is drawn through the vessel, but the leak will stop if a
+considerable difference of potential is maintained between the wire and
+the tube: this shows that a strong electric field removes the
+conductivity from the gas.
+
+The fact that the conductivity of the gas is removed by filtering shows
+that it is due to something mixed with the gas which is removed from it
+by filtration, and since the conductivity is also removed by an electric
+field, the cause of the conductivity must be charged with electricity so
+as to be driven to the sides of the tube by the electric force. Since
+the gas as a whole is not electrified either positively or negatively,
+there must be both negative and positive charges in the gas, the amount
+of electricity of one sign being equal to that of the other. We are thus
+led to the conclusion that the conductivity of the gas is due to
+electrified particles being mixed up with the gas, some of these
+particles having charges of positive electricity, others of negative.
+These electrified particles are called _ions_, and the process by which
+the gas is made a conductor is called the ionization of the gas. We
+shall show later that the charges and masses of the ions can be
+determined, and that the gaseous ions are not identical with those met
+with in the electrolysis of solutions.
+
+[Illustration: FIG. 6.]
+
+One very characteristic property of conduction of electricity through a
+gas is the relation between the current through the gas and the electric
+force which gave rise to it. This relation is not in general that
+expressed by Ohm's law, which always, as far as our present knowledge
+extends, expresses the relation for conduction through metals and
+electrolytes. With gases, on the other hand, it is only when the current
+is very small that Ohm's law is true. If we represent graphically by
+means of a curve the relation between the current passing between two
+parallel metal plates separated by ionized gas and the difference of
+potential between the plates, the curve is of the character shown in
+fig. 6 when the ordinates represent the current and the abscissae the
+difference of potential between the plates. We see that when the
+potential difference is very small, i.e. close to the origin, the curve
+is approximately straight, but that soon the current increases much less
+rapidly than the potential difference, and that a stage is reached when
+no appreciable increase of current is produced when the potential
+difference is increased; when this stage is reached the current is
+constant, and this value of the current is called the "saturation"
+value. When the potential difference approaches the value at which
+sparks would pass through the gas, the current again increases with the
+potential difference; thus the curve representing the relation between
+the current and potential difference over very wide ranges of potential
+difference has the shape shown in fig. 7; curves of this kind have been
+obtained by von Schweidler (_Wien. Ber._, 1899, 108, p. 273), and J. E.
+S. Townsend (_Phil. Mag._, 1901 [6], 1, p. 198). We shall discuss later
+the causes of the rise in the current with large potential differences,
+when we consider ionization by collision.
+
+ The general features of the earlier part of the curve are readily
+ explained on the ionization hypothesis. On this view the Röntgen rays
+ or other ionizing agent acting on the gas between the plates, produces
+ positive and negative ions at a definite rate. Let us suppose that q
+ positive and q negative ions are by this means produced per second
+ between the plates; these under the electric force will tend to move,
+ the positive ones to the negative plate, the negative ones to the
+ positive. Some of these ions will reach the plate, others before
+ reaching the plate will get so near one of the opposite sign that the
+ attraction between them will cause them to unite and form an
+ electrically neutral system; when they do this they end their
+ existence as ions. The current between the plates is proportional to
+ the number of ions which reach the plates per second. Now it is
+ evident that we cannot go on taking more ions out of the gas than are
+ produced; thus we cannot, when the current is steady, have more than q
+ positive ions driven to the negative plate per second, and the same
+ number of negative ions to the positive. If each of the positive ions
+ carries a charge of e units of positive electricity, and if there is
+ an equal and opposite charge on each negative ion, then the maximum
+ amount of electricity which can be given to the plates per second is
+ qe, and this is equal to the saturation current. Thus if we measure
+ the saturation current, we get a direct measure of the ionization, and
+ this does not require us to know the value of any quantity except the
+ constant charge on the ion. If we attempted to deduce the amount of
+ ionization by measurements of the current before it was saturated, we
+ should require to know in addition the velocity with which the ions
+ move under a given electric force, the time that elapses between the
+ liberation of an ion and its combination with one of the opposite
+ sign, and the potential difference between the plates. Thus if we wish
+ to measure the amount of ionization in a gas we should be careful to
+ see that the current is saturated.
+
+ [Illustration: FIG. 7.]
+
+The difference between conduction through gases and through metals is
+shown in a striking way when we use potential differences large enough
+to produce the saturation current. Suppose we have got a potential
+difference between the plates more than sufficient to produce the
+saturation current, and let us increase the distance between the plates.
+If the gas were to act like a metallic conductor this would diminish the
+current, because the greater length would involve a greater resistance
+in the circuit. In the case we are considering the separation of the
+plates will _increase_ the current, because now there is a larger volume
+of gas exposed to the rays; there are therefore more ions produced, and
+as the saturation current is proportional to the number of ions the
+saturation current is increased. If the potential difference between the
+plates were much less than that required to saturate the current, then
+increasing the distance would diminish the current; the gas for such
+potential differences obeys Ohm's law and the behaviour of the gaseous
+resistance is therefore similar to that of a metallic one.
+
+In order to produce the saturation current the electric field must be
+strong enough to drive each ion to the electrode before it has time to
+enter into combination with one of the opposite sign. Thus when the
+plates in the preceding example are far apart, it will take a larger
+potential difference to produce this current than when the plates are
+close together. The potential difference required to saturate the
+current will increase as the square of the distance between the plates,
+for if the ions are to be delivered in a given time to the plates their
+speed must be proportional to the distance between the plates. But the
+speed is proportional to the electric force acting on the ion; hence the
+electric force must be proportional to the distance between the plates,
+and as in a uniform field the potential difference is equal to the
+electric force multiplied by the distance between the plates, the
+potential difference will vary as the square of this distance.
+
+The potential difference required to produce saturation will, other
+circumstances being the same, increase with the amount of ionization,
+for when the number of ions is large and they are crowded together, the
+time which will elapse before a positive one combines with a negative
+will be smaller than when the number of ions is small. The ions have
+therefore to be removed more quickly from the gas when the ionization is
+great than when it is small; thus they must move at a higher speed and
+must therefore be acted upon by a larger force.
+
+
+When the ions are not removed from the gas, they will increase until the
+number of ions of one sign which combine with ions of the opposite sign
+in any time is equal to the number produced by the ionizing agent in
+that time. We can easily calculate the number of free ions at any time
+after the ionizing agent has commenced to act.
+
+ Let q be the number of ions (positive or negative) produced in one
+ cubic centimetre of the gas per second by the ionizing agent, n1, n2,
+ the number of free positive and negative ions respectively per cubic
+ centimetre of the gas. The number of collisions between positive and
+ negative ions per second in one cubic centimetre of the gas is
+ proportional to n1n2. If a certain fraction of the collisions between
+ the positive and negative ions result in the formation of an
+ electrically neutral system, the number of ions which disappear per
+ second on a cubic centimetre will be equal to [alpha]n1 n2, where
+ [alpha] is a quantity which is independent of n1, n2; hence if t is
+ the time since the ionizing agent was applied to the gas, we have
+
+ dn1/dt = q - [alpha]n1 n2, dn2/dt = q - [alpha]n1 n2.
+
+ Thus n1 - n2 is constant, so if the gas is uncharged to begin with, n1
+ will always equal n2. Putting n1 = n2 = n we have
+
+ dn/dt = q - [alpha]n² (1),
+
+ the solution of which is, since n = 0 when t = 0,
+
+ k([epsilon]^{2k[alpha]t} - 1)
+ n = ---------------------------- (2)
+ [epsilon]^{2k[alpha]t} + 1
+
+ if k² = q/[alpha]. Now the number of ions when the gas has reached a
+ steady state is got by putting t equal to infinity in the preceding
+ equation, and is therefore given by the equation
+
+ n0 = k = [root](q/[alpha]).
+
+ We see from equation (1) that the gas will not approximate to its
+ steady state until 2k[alpha]t is large, that is until t is large
+ compared with 1/2k[alpha] or with 1/2[root](q[alpha]). We may thus
+ take 1/2[root](q[alpha]) as a measure of the time taken by the gas to
+ reach a steady state when exposed to an ionizing agent; as this time
+ varies inversely as [root]q we see that when the ionization is feeble
+ it may take a very considerable time for the gas to reach a steady
+ state. Thus in the case of our atmosphere where the production of ions
+ is only at the rate of about 30 per cubic centimetre per second, and
+ where, as we shall see, [alpha] is about 10^-6, it would take some
+ minutes for the ionization in the air to get into a steady state if
+ the ionizing agent were suddenly applied.
+
+ We may use equation (1) to determine the rate at which the ions
+ disappear when the ionizing agent is removed. Putting q=0 in that
+ equation we get dn/[alpha]t = -[alpha]n².
+
+ Hence n = n0/(1 + n0[alpha]t) (3),
+
+ where n0 is the number of ions when t = 0. Thus the number of ions
+ falls to one-half its initial value in the time 1/n0[alpha]. The
+ quantity [alpha] is called the _coefficient of recombination_, and its
+ value for different gases has been determined by Rutherford (_Phil.
+ Mag._ 1897 [5], 44, p. 422), Townsend (_Phil. Trans._, 1900, 193, p.
+ 129), McClung (_Phil. Mag._, 1902 [6], 3, p. 283), Langevin (_Ann.
+ chim. phys._ [7], 28, p. 289), Retschinsky (_Ann. d. Phys._, 1905, 17,
+ p. 518), Hendred (_Phys. Rev._, 1905, 21, p. 314). The values of
+ [alpha]/e, e being the charge on an ion in electrostatic measure as
+ determined by these observers for different gases, is given in the
+ following table:--
+
+ +-----+----------+----------+----------+------------+----------+
+ | | Townsend.| McClung. | Langevin.|Retschinsky.| Hendred. |
+ +-----+----------+----------+----------+------------+----------+
+ | Air | 3420 | 3380 | 3200 | 4140 | 3500 |
+ | O2 | 3380 | | | | |
+ | CO2 | 3500 | 3490 | 3400 | | |
+ | H2 | 3020 | 2940 | | | |
+ +-----+----------+----------+----------+------------+----------+
+
+ The gases in these experiments were carefully dried and free from
+ dust; the apparent value of [alpha] is much increased when dust or
+ small drops of water are present in the gas, for then the ions get
+ caught by the dust particles, the mass of a particle is so great
+ compared with that of an ion that they are practically immovable under
+ the action of the electric field, and so the ions clinging to them
+ escape detection when electrical methods are used. Taking e as 3.5 ×
+ 10^-10, we see that [alpha] is about 1.2 × 10^-6, so that the number
+ of recombinations in unit time between n positive and n negative ions
+ in unit volume is 1.2 × 10^-6n². The kinetic theory of gases shows
+ that if we have n molecules of air per cubic centimetre, the number of
+ collisions per second is 1.2 × 10^-10n² at a temperature of 0° C.
+ Thus we see that the number of recombinations between oppositely
+ charged ions is enormously greater than the number of collisions
+ between the same number of neutral molecules. We shall see that the
+ difference in size between the ion and the molecule is not nearly
+ sufficient to account for the difference between the collisions in the
+ two cases; the difference is due to the force between the oppositely
+ charged ions, which drags ions into collisions which but for this
+ force would have missed each other.
+
+ Several methods have been used to measure [alpha]. In one method air,
+ exposed to some ionizing agent at one end of a long tube, is slowly
+ sucked through the tube and the saturation current measured at
+ different points along the tube. These currents are proportional to
+ the values of n at the place of observation: if we know the distance
+ of this place from the end of the tube when the gas was ionized and
+ the velocity of the stream of gas, we can find t in equation (3), and
+ knowing the value of n we can deduce the value of [alpha] from the
+ equation
+
+ 1/n1 - 1/n2 = [alpha](t1 - t2),
+
+ where n1, n2 are the values of n at the times t1, t2 respectively. In
+ this method the tubes ought to be so wide that the loss of ions by
+ diffusion to the sides of the tube is negligible. There are other
+ methods which involve the knowledge of the speed with which the ions
+ move under the action of known electric forces; we shall defer the
+ consideration of these methods until we have discussed the question of
+ these speeds.
+
+ In measuring the value of [alpha] it should be remembered that the
+ theory of the methods supposes that the ionization is uniform
+ throughout the gas. If the total ionization throughout a gas remains
+ constant, but instead of being uniformly distributed is concentrated
+ in patches, it is evident that the ions will recombine more quickly in
+ the second case than in the first, and that the value of [alpha] will
+ be different in the two cases. This probably explains the large values
+ of [alpha] obtained by Retschinsky, who ionized the gas by the [alpha]
+ rays from radium, a method which produces very patchy ionization.
+
+ _Variation of [alpha] with the Pressure of the Gas._--All observers
+ agree that there is little variation in [alpha] with the pressures for
+ pressures of between 5 and 1 atmospheres; at lower pressures, however,
+ the value of [alpha] seems to diminish with the pressure: thus
+ Langevin (_Ann. chim. phys._, 1903, 28, p. 287) found that at a
+ pressure of 1/5 of an atmosphere the value of [alpha] was about 1/5 of
+ its value at atmospheric pressure.
+
+ _Variation of [alpha] with the Temperature._--Erikson (_Phil. Mag._,
+ Aug. 1909) has shown that the value of [alpha] for air increases as
+ the temperature diminishes, and that at the temperature of liquid air
+ -180° C., it is more than twice as great as at +12° C.
+
+ Since, as we have seen, the recombination is due to the coming
+ together of the positive and negative ions under the influence of the
+ electrical attraction between them, it follows that a large electric
+ force sufficient to overcome this attraction would keep the ions apart
+ and hence diminish the coefficient of recombination. Simple
+ considerations, however, will show that it would require exceedingly
+ strong electric fields to produce an appreciable effect. The value of
+ [alpha] indicates that for two oppositely charged ions to unite they
+ must come within a distance of about 1.5 × 10^-6 centimetres; at this
+ distance the attraction between them is e² × 10^12/2.25, and if X is
+ the external electric force, the force tending to pull them apart
+ cannot be greater than Xe; if this is to be comparable with the
+ attraction, X must be comparable with e × 10^12/2.25, or putting e = 4
+ × 10^-10, with 1.8 × 10²; this is 54,000 volts per centimetre, a
+ force which could not be applied to gas at atmospheric pressure
+ without producing a spark.
+
+ _Diffusion of the Ions._--The ionized gas acts like a mixture of
+ gases, the ions corresponding to two different gases, the non-ionized
+ gas to a third. If the concentration of the ions is not uniform, they
+ will diffuse through the non-ionized gas in such a way as to produce a
+ more uniform distribution. A very valuable series of determinations of
+ the coefficient of diffusion of ions through various gases has been
+ made by Townsend (_Phil. Trans._, 1900, A, 193, p. 129). The method
+ used was to suck the ionized gas through narrow tubes; by measuring
+ the loss of both the positive and negative ions after the gases had
+ passed through a known length of tube, and allowing for the loss by
+ recombination, the loss by diffusion and hence the coefficient of
+ diffusion could be determined. The following tables give the values of
+ the coefficients of diffusion D on the C.G.S. system of units as
+ determined by Townsend:--
+
+ Table I.--_Coefficients of Diffusion (D) in Dry Gases._
+
+ +-----+------------+------------+----------+------------------+
+ |Gas. |D for +ions.|D for -ions.|Mean Value| Ratio of D for |
+ | | | | of D. | - to D for +ions.|
+ +-----+------------+------------+----------+------------------+
+ | Air | .028 | .043 | .0347 | 1.54 |
+ | O2 | .025 | .0396 | .0323 | 1.58 |
+ | CO2 | .023 | .026 | .0245 | 1.13 |
+ | H2 | .123 | .190 | .156 | 1.54 |
+ +-----+------------+------------+----------+------------------+
+
+ Table II.--Coefficients of Diffusion in Moist Gases.
+
+ +-----+------------+------------+----------+------------------+
+ |Gas. |D for +ions.|D for -ions.|Mean Value| Ratio of D for |
+ | | | | of D. | - to D for +ions.|
+ +-----+------------+------------+----------+------------------+
+ | Air | .032 | .037 | .0335 | 1.09 |
+ | O2 | .0288 | .0358 | .0323 | 1.24 |
+ | CO2 | .0245 | .0255 | .025 | 1.04 |
+ | H2 | .128 | .142 | .135 | 1.11 |
+ +-----+------------+------------+----------+------------------+
+
+ It is interesting to compare with these coefficients the values of D
+ when various gases diffuse through each other. D for hydrogen through
+ air is .634, for oxygen through air .177, for the vapour of isobutyl
+ amide through air .042. We thus see that the velocity of diffusion of
+ ions through air is much less than that of the simple gas, but that it
+ is quite comparable with that of the vapours of some complex organic
+ compounds.
+
+ The preceding tables show that the negative ions diffuse more rapidly
+ than the positive, especially in dry gases. The superior mobility of
+ the negative ions was observed first by Zeleny (_Phil. Mag._, 1898
+ [5], 46, p. 120), who showed that the velocity of the negative ions
+ under an electric force is greater than that of the positive. It will
+ be noticed that the difference between the mobility of the negative
+ and the positive ions is much more pronounced in dry gases than in
+ moist. The difference in the rates of diffusion of the positive and
+ negative ions is the reason why ionized gas, in which, to begin with,
+ the positive and negative charges were of equal amounts, sometimes
+ becomes electrified even although the gas is not acted upon by
+ electric forces. Thus, for example, if such gas be blown through
+ narrow tubes, it will be positively electrified when it comes out, for
+ since the negative ions diffuse more rapidly than the positive, the
+ gas in its passage through the tubes will lose by diffusion more
+ negative than positive ions and hence will emerge positively
+ electrified. Zeleny snowed that this effect does not occur when, as in
+ carbonic acid gas, the positive and negative ions diffuse at the same
+ rates. Townsend (loc. cit.) showed that the coefficient of diffusion
+ of the ions is the same whether the ionization is produced by Röntgen
+ rays, radioactive substances, ultra-violet light, or electric sparks.
+ The ions produced by chemical reactions and in flames are much less
+ mobile; thus, for example, Bloch (_Ann. chim. phys._, 1905 [8], 4, p.
+ 25) found that for the ions produced by drawing air over phosphorus
+ the value of [alpha]/e was between 1 and 6 instead of over 3000, the
+ value when the air was ionized by Röntgen rays.
+
+_Velocity of Ions in an Electric Field._--The velocity of ions in an
+electric field, which is of fundamental importance in conduction, is
+very closely related to the coefficient of diffusion. Measurements of
+this velocity for ions produced by Röntgen rays have been made by
+Rutherford (_Phil. Mag._ [5], 44, p. 422), Zeleny (_Phil. Mag._ [5], 46,
+p. 120), Langevin (_Ann. Chim. Phys._, 1903, 28, p. 289), Phillips
+(_Proc. Roy. Soc._ 78, A, p. 167), and Wellisch (_Phil. Trans._, 1909,
+209, p. 249). The ions produced by radioactive substance have been
+investigated by Rutherford (_Phil. Mag._ [5], 47, p. 109) and by Franck
+and Pohl (_Verh. deutsch. phys. Gesell._, 1907, 9, p. 69), and the
+negative ions produced when ultra-violet light falls on a metal plate by
+Rutherford (_Proc. Camb. Phil. Soc._ 9, p. 401). H. A. Wilson (_Phil.
+Trans._ 192, p. 4O9), Marx (_Ann. de Phys._ 11, p. 765), Moreau (_Journ.
+de Phys._ 4, 11, p. 558; _Ann. Chim. Phys._ 7, 30, p. 5) and Gold
+(_Proc. Roy. Soc._ 79, p. 43) have investigated the velocities of ions
+produced by putting various salts into flames; McClelland (_Phil. Mag._
+46, p. 29) the velocity of the ions in gases sucked from the
+neighbourhood of flames and arcs; Townsend (_Proc. Camb. Phil. Soc._ 9,
+p. 345) and Bloch (_loc. cit._) the velocity of ions produced by
+chemical reaction; and Chattock (_Phil. Mag._ [5], 48, p. 401) the
+velocity of the ions produced when electricity escapes from a sharp
+needle point into a gas.
+
+Several methods have been employed to determine these velocities. The
+one most frequently employed is to find the electromotive intensity
+required to force an ion against the stream of gas moving with a known
+velocity parallel to the lines of electric force. Thus, of two
+perforated plane electrodes vertically over each other, suppose the
+lower to be positively, the upper negatively electrified, and suppose
+that the gas is streaming vertically downwards with the velocity V; then
+unless the upward velocity of the positive ion is greater than V, no
+positive electricity will reach the upper plate. If we increase the
+strength of the field between the plates, and hence the upward velocity
+of the positive ion, until the positive ions just begin to reach the
+upper plate, we know that with this strength of field the velocity of
+the positive ion is equal to V. By this method, which has been used by
+Rutherford, Zeleny and H. A. Wilson, the velocity of ions in fields of
+various strengths has been determined.
+
+ The arrangement used by Zeleny is represented in fig. 8. P and Q are
+ square brass plates. They are bored through their centres, and to the
+ openings the tubes R and S are attached, the space between the plates
+ being covered in so as to form a closed box. K is a piece of wire
+ gauze completely covering the opening in Q; T is an insulated piece of
+ wire gauze nearly but not quite filling the opening in the plate P,
+ and connected with one pair of quadrants of an electrometer E. A plug
+ of glass wool G filters out the dust from a stream of gas which enters
+ the vessel by the tube D and leaves it by F; this plug also makes the
+ velocity of the flow of the gas uniform across the section of the
+ tube. The Röntgen rays to ionize the gas were produced by a bulb at
+ O, the bulb and coil being in a lead-covered box, with an aluminium
+ window through which the rays passed. Q is connected with one pole of
+ a battery of cells, P and the other pole of the battery are put to
+ earth. The changes in the potential of T are due to ions giving up
+ their charges to it. With a given velocity of air-blast the potential
+ of T was found not to change unless the difference of potential
+ between P and Q exceeded a critical value. The field corresponding to
+ this critical value thus made the ions move with the known velocity of
+ the blast.
+
+ [Illustration: FIG. 8.]
+
+ [Illustration: FIG. 9.]
+
+ Another method which has been employed by Rutherford and McClelland is
+ based on the action of an electric field in destroying the
+ conductivity of gas streaming through it. Suppose that BAB, DCD (fig.
+ 9) are a system of parallel plates boxed in so that a stream of gas,
+ after flowing between BB, passes between DD without any loss of gas in
+ the interval. Suppose the plates DD are insulated, and connected with
+ one pair of quadrants of an electrometer, by charging up C to a
+ sufficiently high potential we can drive all the positive ions which
+ enter the system DCD against the plates D; this will cause a deflexion
+ of the electrometer, which in one second will be proportional to the
+ number of positive ions which have entered the system in that time. If
+ we charge A up to a high potential, B being put to earth, we shall
+ find that the deflexion of the electrometer connected with DD is less
+ than it was when A and B were at the same potential, because some of
+ the positive ions in their passage through BAB are driven against the
+ plates B. If u is the velocity along the lines of force in the uniform
+ electric field between A and B, and t the time it takes for the gas to
+ pass through BAB, then all the positive ions within a distance ut of
+ the plates B will be driven up against these plates, and thus if the
+ positive ions are equally distributed through the gas, the number of
+ positive ions which emerge from the system when the electric field is
+ on will bear to the number which emerge when the field is off the
+ ratio of 1 - ut/l to unity, where l is the distance between A and B.
+ This ratio is equal to the ratio of the deflexions in one second of
+ the electrometer attached to D, hence the observations of this
+ instrument give 1 - ut/l. If we know the velocity of the gas and the
+ length of the plates A and B, we can determine t, and since l can be
+ easily measured, we can find u, the velocity of the positive ion in a
+ field of given strength. By charging A and C negatively instead of
+ positively we can arrive at the velocity of the negative ion. In
+ practice it is more convenient to use cylindrical tubes with coaxial
+ wires instead of the systems of parallel plates, though in this case
+ the calculation of the velocity of the ions from the observations is a
+ little more complicated, inasmuch as the electric field is not uniform
+ between the tubes.
+
+ [Illustration: FIG. 10.]
+
+ A method which gives very accurate results, though it is only
+ applicable in certain cases, is the one used by Rutherford to measure
+ the velocity of the negative ions produced close to a metal plate by
+ the incidence on the plate of ultra-violet light. The principle of the
+ method is as follows:--AB (fig. 10) is an insulated horizontal plate
+ of well-polished zinc, which can be moved vertically up and down by
+ means of a screw; it is connected with one pair of quadrants of an
+ electrometer, the other pair of quadrants being put to earth. CD is a
+ base-plate with a hole EF in it; this hole is covered with fine wire
+ gauze, through which ultra-violet light passes and falls on the plate
+ AB. The plate CD is connected with an alternating current dynamo,
+ which produces a simply-periodic potential difference between AB and
+ CD, the other pole being put to earth. Suppose that at any instant the
+ plate CD is at a higher potential than AB, then the negative ions from
+ AB will move towards CD, and will continue to do so as long as the
+ potential of CD is higher than that of AB. If, however, the potential
+ difference changes sign before the negative ions reach CD, these ions
+ will go back to AB. Thus AB will not lose any negative charge unless
+ the distance between the plates AB and CD is less than the distance
+ traversed by the negative ion during the time the potential of CD is
+ higher than that of AB. By altering the distance between the plates
+ until CD just begins to lose a negative charge, we find the velocity
+ of the negative ion under unit electromotive intensity. For suppose
+ the difference of potential between AB and CD is equal to a sin pt,
+ then if d is the distance between the plates, the electric intensity
+ is equal to a sin pt/d; if we suppose the velocity of the ion is
+ proportional to the electric intensity, and if u is the velocity for
+ unit electric intensity, the velocity of the negative ion will be ua
+ sin pt/d. Hence if x represent the distance of the ion from AB
+
+ dx ua
+ --- = --- sin pt
+ dT d
+
+ ua
+ x = ----(1 - cos pt), if x = 0 when t = 0.
+ pd
+
+ Thus the greatest distance the ion can get from the plate is equal to
+ 2au/pd, and if the distance between the plates is gradually reduced to
+ this value, the plate AB will begin to lose a negative charge; hence
+ when this happens
+
+ d = 2au/pd, or u = pd²/2a,
+
+ an equation by means of which we can find u.
+
+ In this form the method is not applicable when ions of both signs are
+ present. Franck and Pohl (_Verh. deutsch. physik. Gesell._ 1907, 9, p.
+ 69) have by a slight modification removed this restriction. The
+ modification consists in confining the ionization to a layer of gas
+ below the gauze EF. If the velocity of the positive ions is to be
+ determined, these ions are forced through the gauze by applying to the
+ ionized gas a small constant electric force acting upwards; if
+ negative ions are required, the constant force is reversed. After
+ passing through the gauze the ions are acted upon by alternating
+ forces as in Rutherford's method.
+
+ Langevin (_Ann. chim. phys._, 1903, 28, p. 289) devised a method of
+ measuring the velocity of the ions which has been extensively used; it
+ has the advantage of not requiring the rate of ionization to remain
+ uniform. The general idea is as follows. Suppose that we expose the
+ gas between two parallel plates A, B to Röntgen rays or some other
+ ionizing agent, then stop the rays and apply a uniform electric field
+ to the region between the plates. If the force on the positive ion is
+ from A to B, the plate B will receive a positive charge of
+ electricity. After the electric force has acted for a time T reverse
+ it. B will now begin to receive negative electricity and will go on
+ doing so until the supply of negative ions is exhausted. Let us
+ consider how the quantity of positive electricity received by B will
+ vary with T. To fix our ideas, suppose the positive ions move more
+ slowly than the negative; let T2 and T1 be respectively the times
+ taken by the positive and negative ions to move under the electric
+ field through a distance equal to AB, the distance between the planes.
+ Then if T is greater than T2 all the ions will have been driven from
+ between the plates before the field is reversed, and therefore the
+ positive charge received by B will not depend upon T. Next let T be
+ less than T2 but greater than T1; then at the time when the field is
+ reversed all the negative ions will have been driven from between the
+ plates, so that the positive charge received by B will not be
+ neutralized by the arrival of fresh ions coming to it after the
+ reversal of the field. The number of positive ions driven against the
+ plate B will be proportional to T. Thus if we measure the value of the
+ positive charge on B for a series of values of T, each value being
+ less than the preceding, we shall find that until T reaches a certain
+ value the charge remains constant, but as soon as we reduce the time
+ below this value the charge diminishes. The value of T when the
+ diminution in the field begins is T2, the time taken for a positive
+ ion to cross from A to B under the electric field; thus from T2 we can
+ calculate the velocity of the positive ion in this field. If we still
+ further diminish T, we shall find that we reach a value when the
+ diminution of the positive charge on B with the time suddenly becomes
+ much more rapid; this change occurs when T falls below T1 the time
+ taken for the negative ions to go from one plate to the other, for now
+ when the field is reversed there are still some negative ions left
+ between the plates, and these will be driven against B and rob it of
+ some of the positive charge it had acquired before the field was
+ reversed. By observing the time when the increase in the rate of
+ diminution of the positive charge with the time suddenly sets in we
+ can determine T1, and hence the velocity of the negative ions.
+
+ The velocity of the ions produced by the discharge of electricity from
+ a fine point was determined by Chattock by an entirely different
+ method. In this case the electric field is so strong and the velocity
+ of the ion so great that the preceding methods are not applicable.
+ Suppose P represents a vertical needle discharging electricity into
+ air, consider the force acting on the ions included between two
+ horizontal planes A, B. If P is the density of the electrification,
+ and Z the vertical component of the electric intensity, F the
+ resultant force on the ions between A and B is vertical and equal to
+ _ _ _
+ / / /
+ | | | Z[rho]dxdydz.
+ _/_/_/
+
+ Let us suppose that the velocity of the ion is proportional to the
+ electric intensity, so that if w is the vertical velocity of the ions,
+ which are supposed all to be of one sign, w = RZ.
+
+ Substituting this value of Z, the vertical force on the ions between A
+ and B is equal to
+ _ _ _
+ 1 / / /
+ - | | | w[rho]dxdydz.
+ R _/_/_/
+
+ But [integral][integral]w[rho]dxdy = [iota], where [iota] is the
+ current streaming from the point. This current, which can be easily
+ measured by putting a galvanometer in series with the discharging
+ point, is independent of z, the vertical distance of a plane between A
+ and B below the charging point. Hence we have
+ _
+ [iota] / [iota]
+ F = ------ | dz = ------·z.
+ R _/ R
+
+ This force must be counterbalanced by the difference of gaseous
+ pressures over the planes A and B; hence if pB and pA denote
+ respectively the pressures over B and A, we have
+
+ [iota]
+ pB - pA = ------ z.
+ R
+
+ Hence by the measurement of these pressures we can determine R, and
+ hence the velocity with which an ion moves under a given electric
+ intensity.
+
+ There are other methods of determining the velocities of the ions, but
+ as these depend on the theory of the conduction of electricity through
+ a gas containing charged ions, we shall consider them in our
+ discussion of that theory.
+
+ By the use of these methods it has been shown that the velocities of
+ the ions in a given gas are the same whether the ionization is
+ produced by Röntgen rays, radioactive substances, ultra-violet light,
+ or by the discharge of electricity from points. When the ionization is
+ produced by chemical action the ions are very much less mobile, moving
+ in the same electric field with a velocity less than one-thousandth
+ part of the velocity of the first kind of ions. On the other hand, as
+ we shall see later, the velocity of the negative ions in flames is
+ enormously greater than that of even the first kind of ion under
+ similar electric fields and at the same pressure. But when these
+ negative ions get into the cold part of the flame, they move
+ sluggishly with velocities of the order of those possessed by the
+ second kind. The results of the various determinations of the
+ velocities of the ions are given in the following table. The
+ velocities are in centimetres per second under an electric force of
+ one volt per centimetre, the pressure of the gas being 1 atmosphere.
+ V+ denotes the velocity of the positive ion, V- that of the negative.
+ V is the mean velocity of the positive and negative ions.
+
+ _Velocities of Ions.--Ions produced by Röntgen Rays._
+
+ +----------------------+------+------+------+-----------------+
+ | Gas. | V+. | V-. | V. | Observer. |
+ +----------------------+------+------+------+-----------------+
+ | Air | .. | .. | 1.6 | Rutherford |
+ | Air (dry) | 1.36 | 1.87 | .. | Zeleny |
+ | " | 1.60 | 1.70 | .. | Langevin |
+ | " | 1.39 | 1.78 | .. | Phillips |
+ | " | 1.54 | 1.78 | .. | Wellisch |
+ | Air (moist) | 1.37 | 1.81 | .. | Zeleny |
+ | Oxygen (dry) | 1.36 | 1.80 | .. | " |
+ | Oxygen (moist) | 1.29 | 1.52 | .. | " |
+ | Carbonic acid (dry) | 0.76 | 0.81 | .. | " |
+ | " | 0.86 | 0.90 | .. | Langevin |
+ | " | 0.81 | 0.85 | .. | Wellisch |
+ | Carbonic acid (moist)| 0.82 | 0.75 | .. | Zeleny |
+ | Hydrogen (dry) | 6.70 | 7.95 | .. | " |
+ | Nitrogen | .. | .. | 1.6 | Rutherford |
+ | Sulphur dioxide | 0.44 | 0.41 | .. | Wellisch |
+ | Hydrochloric acid | .. | .. | 1.27 | Rutherford |
+ | Chlorine | .. | .. | 1.0 | " |
+ | Helium (dry) | 5.09 | 6.31 | .. | Franck and Pohl |
+ | Carbon monoxide | 1.10 | 1.14 | .. | Wellisch |
+ | Nitrous oxide | 0.82 | 0.90 | .. | " |
+ | Ammonia | 0.74 | 0.80 | .. | " |
+ | Aldehyde | 0.31 | 0.30 | .. | " |
+ | Ethyl alcohol | 0.34 | 0.27 | .. | " |
+ | Acetone | 0.31 | 0.29 | .. | " |
+ | Ethyl chloride | 0.33 | 0.31 | .. | " |
+ | Pentane | 0.36 | 0.35 | .. | " |
+ | Methyl acetate | 0.33 | 0.36 | .. | " |
+ | Ethyl formate | 0.30 | 0.31 | .. | " |
+ | Ethyl ether | 0.29 | 0.31 | .. | " |
+ | Ethyl acetate | 0.31 | 0.28 | .. | " |
+ | Methyl bromide | 0.29 | 0.28 | .. | " |
+ | Methyl iodide | 0.21 | 0.22 | .. | " |
+ | Carbon tetrachloride | 0.30 | 0.31 | .. | " |
+ | Ethyl iodide | 0.17 | 0.16 | .. | " |
+ +----------------------+------+------+------+-----------------+
+
+ _Ions produced by Ultra-Violet Light._
+
+ Air 1.4 Rutherford
+ Hydrogen 3.9 Rutherford
+ Carbonic acid 0.78 Rutherford
+
+ _Ions in Gases sucked from Flames._
+
+ Velocities varying from .04 to .23 McClelland
+
+ _Ions in Flames containing Salts._
+
+ Negative ions 12.9 cm./sec. Gold
+ +ions for salts of Li, Na,
+ K, Rb, Cs 62 H. A. Wilson
+ " 200 Marx
+ " 80 Moreau
+
+ _Ions liberated by Chemical Action._
+
+ Velocities of the order of 0.0005 cm./sec. Bloch
+
+ _Ions from Point Discharge._
+
+ +---------------+------+-------+------+----------+
+ | Hydrogen | 5.4 | 7.43 | 6.41 | Chattock |
+ | Carbonic acid | 0.83 | 0.925 | 0.88 | Chattock |
+ | Air | 1.32 | 1.80 | 1.55 | Chattock |
+ | Oxygen | 1.30 | 1.85 | 1.57 | Chattock |
+ +---------------+------+-------+------+----------+
+
+ It will be seen from this table that the greater mobility of the
+ negative ions is very much more marked in the case of the lighter and
+ simpler gases than in that of the heavier and more complicated ones;
+ with the vapours of organic substances there seems but little
+ difference between the mobilities of the positive and negative ions,
+ indeed in one or two cases the positive one seems slightly but very
+ slightly the more mobile of the two. In the case of the simple gases
+ the difference is much greater when the gases are dry than when they
+ are moist. It has been shown by direct experiment that the velocities
+ are directly proportional to the electric force.
+
+ _Variation of Velocities with Pressure._--Until the pressure gets low
+ the velocities of the ions, negative as well as positive, vary
+ inversely as the pressure. Langevin (loc. cit.) was the first to show
+ that at very low pressures the velocity of the negative ions increases
+ more rapidly as the pressure is diminished than this law indicates. If
+ the nature of the ion did not change with the pressure, the kinetic
+ theory of gases indicates that the velocity would vary inversely as
+ the pressure, so that Langevin's results indicate a change in the
+ nature of the negative ion when the pressure is diminished below a
+ certain value. Langevin's results are given in the following table,
+ where p represents the pressure measured in centimetres of mercury, V+
+ and V- the velocities of the positive and negative ions in air under
+ unit electrostatic force, i.e. 300 volts per centimetre:--
+
+ +----------------------+----------------------+
+ | Negative Ions. | Positive Ions. |
+ +-------+------+-------+-------+------+-------+
+ | p. | V-. |pV-/76.| p. | V+. |pV+/76.|
+ +-------+------+-------+-------+------+-------+
+ | 7.5 | 6560 | 647 | 7.5 | 4430 | 437 |
+ | 20.0 | 2204 | 580 | 20.0 | 1634 | 430 |
+ | 41.5 | 994 | 530 | 41.5 | 782 | 427 |
+ | 76.0 | 510 | 510 | 76.0 | 480 | 420 |
+ | 142.0 | 270 | 505 | 142.0 | 225 | 425 |
+ +-------+------+-------+-------+------+-------+
+
+ The increase in the case of pV- indicates that the structure of the
+ negative ion gets simpler as the pressure is reduced. Wallisch in some
+ experiments made at the Cavendish Laboratory found that the diminution
+ in the value of pV- at low pressures is much more marked in some gases
+ than in others, and in some gases he failed to detect it; but it must
+ be remembered that it is difficult to get measurements at pressures of
+ only a few millimetres, as the amount of ionization is so exceedingly
+ small at such pressures that the quantities to be observed are hardly
+ large enough to admit of accurate measurements by the methods
+ available at higher pressures.
+
+ _Effect of Temperature on the Velocity of the Ions._--Phillips (_Proc.
+ Roy. Soc._, 1906, 78, p. 167) investigated, using Langevin's method,
+ the velocities of the + and - ions through air at atmospheric pressure
+ at temperatures ranging from that of boiling liquid air to 411° C.; R1
+ and R2 are the velocities of the + and - ions respectively when the
+ force is a volt per centimetre.
+
+ +-------+-------+---------------------+
+ | R1. | R2. |Temperature Absolute.|
+ +-------+-------+---------------------+
+ | 2.00 | 2.495 | 411° |
+ | 1.95 | 2.40 | 399° |
+ | 1.85 | 2.30 | 383° |
+ | 1.81 | 2.21 | 373° |
+ | 1.67 | 2.125 | 348° |
+ | 1.60 | 2.00 | 333° |
+ | 1.39 | 1.785 | 285° |
+ | 0.945 | 1.23 | 209° |
+ | 0.235 | 0.235 | 94° |
+ +-------+-------+---------------------+
+
+ We see that except in the case of the lowest temperature, that of
+ liquid air, where there is a great drop in the velocity, the
+ velocities of the ions are proportional to the absolute temperature.
+ On the hypothesis of an ion of constant size we should, from the
+ kinetic theory of gases, expect the velocity to be proportional to the
+ square root of the absolute temperature, if the charge on the ion did
+ not affect the number of collisions between the ion and the molecules
+ of the gas through which it is moving. If the collisions were brought
+ about by the electrical attraction between the ions and the molecules,
+ the velocity would be proportional to the absolute temperature. H. A.
+ Wilson (_Phil. Trans._ 192, p. 499), in his experiments on the
+ conduction of flames and hot gases into which salts had been put,
+ found that the velocity of the positive ions in flames at a
+ temperature of 2000° C. containing the salts of the alkali metals was
+ 62 cm./sec. under an electric force of one volt per centimetre, while
+ the velocity of the positive ions in a stream of hot air at 1000° C.
+ containing the same salts was only 7 cm./sec. under the same force.
+ The great effect of temperature is also shown in some experiments of
+ McClelland (_Phil. Mag._ [5], 46, p. 29) on the velocities of the ions
+ in gases drawn from Bunsen flames and arcs; he found that these
+ depended upon the distance the gas had travelled from the flame. Thus,
+ the velocity of the ions at a distance of 5.5 cm. from the Bunsen
+ flame when the temperature was 230° C. was .23 cm./sec. for a volt per
+ centimetre; at a distance of 10 cm. from the flame when the
+ temperature was 160° C. the velocity was .21 cm./sec; while at a
+ distance of 14.5 cm. from the flame when the temperature was 105° C.
+ the velocity was only .04 cm./sec. If the temperature of the gas at
+ this distance from the flame was raised by external means, the
+ velocity of the ions increased.
+
+ We can derive some information as to the constitution of the ions by
+ calculating the velocity with which a molecule of the gas would move
+ in the electric field if it carried the same charge as the ion. From
+ the theory of the diffusion of gases, as developed by Maxwell, we know
+ that if the particles of a gas A are surrounded by a gas B, then, if
+ the partial pressure of A is small, the velocity u with which its
+ particles will move when acted upon by a force Xe is given by the
+ equation
+
+ Xe
+ u = ------- D,
+ (p1/N1)
+
+ where D represents the coefficient of inter-diffusion of A into B, and
+ N1 the number of particles of A per cubic centimetre when the pressure
+ due to A is p1. Let us calculate by this equation the velocity with
+ which a molecule of hydrogen would move through hydrogen if it carried
+ the charge carried by an ion, which we shall prove shortly to be equal
+ to the charge carried by an atom of hydrogen in the electrolysis of
+ solutions. Since p1/N1 is independent of the pressure, it is equal to
+ [Pi]/N, where [Pi] is the atmospheric pressure and N the number of
+ molecules in a cubic centimetre of gas at atmospheric pressure. Now Ne
+ = 1.22 × 10^10, if e is measured in electrostatic units; [Pi] = 10^6
+ and D in this case is the coefficient of diffusion of hydrogen into
+ itself, and is equal to 1.7. Substituting these values we find
+
+ u = 1.97 × 10^4X.
+
+ If the potential gradient is 1 volt per centimetre, X = 1/300.
+ Substituting this value for X, we find u = 66 cm./sec, for the
+ velocity of a hydrogen molecule. We have seen that the velocity of the
+ ion in hydrogen is only about 5 cm./sec, so that the ion moves more
+ slowly than it would if it were a single molecule. One way of
+ explaining this is to suppose that the ion is bigger than the
+ molecule, and is in fact an aggregation of molecules, the charged ion
+ acting as a nucleus around which molecules collect like dust round a
+ charged body. This view is supported by the effect produced by
+ moisture in diminishing the velocity of the negative ion, for, as C.
+ T. R. Wilson (_Phil. Trans._ 193, p. 289) has shown, moisture tends to
+ collect round the ions, and condenses more easily on the negative than
+ on the positive ion. In connexion with the velocities of ions in the
+ gases drawn from flames, we find other instances which suggest that
+ condensation takes place round the ions. An increase in the size of
+ the system is not, however, the only way by which the velocity might
+ fall below that calculated for the hydrogen molecule, for we must
+ remember that the hydrogen molecule, whose coefficient of diffusion is
+ 1.7, is not charged, while the ion is. The forces exerted by the ion
+ on the other molecules of hydrogen are not the same as those which
+ would be exerted by a molecule of hydrogen, and as the coefficient of
+ diffusion depends on the forces between the molecules, the coefficient
+ of diffusion of a charged molecule into hydrogen might be very
+ different from that of an uncharged one.
+
+ Wellisch (_loc. cit._) has shown that the effect of the charge on the
+ ion is sufficient in many cases to explain the small velocity of the
+ ions, even if there were no aggregation.
+
+ _Mixture of Gases._--The ionization of a mixture of gases raises some
+ very interesting questions. If we ionize a mixture of two very
+ different gases, say hydrogen and carbonic acid, and investigate the
+ nature of the ions by measuring their velocities, the question arises,
+ shall we find two kinds of positive and two kinds of negative ions
+ moving with different velocities, as we should do if some of the
+ positive ions were positively charged hydrogen molecules, while others
+ were positively charged molecules of carbonic acid; or shall we find
+ only one velocity for the positive ions and one for the negative? Many
+ experiments have been made on the velocity of ions in mixtures of two
+ gases, but as yet no evidence has been found of the existence of two
+ different kinds of either positive or negative ions in such mixtures,
+ although some of the methods for determining the velocities of the
+ ions, especially Langevin's, ought to give evidence of this effect, if
+ it existed. The experiments seem to show that the positive (and the
+ same is true for the negative) ions in a mixture of gases are all of
+ the same kind. This conclusion is one of considerable importance, as
+ it would not be true if the ions consisted of single molecules of the
+ gas from which they are produced.
+
+ _Recombination._--Several methods enable us to deduce the coefficient
+ of recombination of the ions when we know their velocities. Perhaps
+ the simplest of these consists in determining the relation between the
+ current passing between two parallel plates immersed in ionized gas
+ and the potential difference between the plates. For let q be the
+ amount of ionization, i.e. the number of ions produced per second per
+ unit volume of the gas, A the area of one of the plates, and d the
+ distance between them; then if the ionization is constant through the
+ volume, the number of ions of one sign produced per second in the gas
+ is qAd. Now if i is the current per unit area of the plate, e the
+ charge on an ion, iA/e ions of each sign are driven out of the gas by
+ the current per second. In addition to this source of loss of ions
+ there is the loss due to the recombination; if n is the number of
+ positive or negative ions per unit volume, then the number which
+ recombine per second is [alpha]n² per cubic centimetre, and if n is
+ constant through the volume of the gas, as will approximately be the
+ case if the current through the gas is only a small fraction of the
+ saturation current, the number of ions which disappear per second
+ through recombination is [alpha]n²·Ad. Hence, since when the gas is
+ in a steady state the number of ions produced must be equal to the
+ number which disappear, we have
+
+ qAd = iA/e + [alpha]n²·Ad,
+ q = i/ed + [alpha]^n2.
+
+ If u1 and u2 are the velocities with which the positive and negative
+ ions move, nu1e and nu2e are respectively the quantities of positive
+ electricity passing in one direction through unit area of the gas per
+ second, and of negative in the opposite direction, hence
+
+ i = nu1e + nu2e.
+
+ If X is the electric force acting on the gas, k1 and k2 the velocities
+ of the positive and negative ions under unit force, u1 = k1X, u2 =
+ k2X; hence
+
+ n = i/(k1 + k2)Xe,
+
+ and we have
+
+ i [alpha]i²
+ q = -- + --------------.
+ ed (k1 + k2)²e²X²
+
+ But qed is the saturation current per unit area of the plate; calling
+ this I, we have
+
+ d[alpha]i²
+ I - i = -------------
+ e(k1 + k2)²X²
+
+ or
+
+ i²·d[alpha]
+ X² = ------------------.
+ e(I - i)(k1 + k2)²
+
+ Hence if we determine corresponding values of X and i we can deduce
+ the value of [alpha]/e if we also know (k1 + k2). The value of I is
+ easily determined, as it is the current when X is very large. The
+ preceding result only applies when i is small compared with I, as it
+ is only in this case that the values of n and X are uniform throughout
+ the volume of the gas. Another method which answers the same purpose
+ is due to Langevin (_Ann. Chim. Phys._, 1903, 28, p. 289); it is as
+ follows. Let A and B be two parallel planes immersed in a gas, and let
+ a slab of the gas bounded by the planes a, b parallel to A and B be
+ ionized by an instantaneous flash of Röntgen rays. If A and B are at
+ different electric potentials, then all the positive ions produced by
+ the rays will be attracted by the negative plate and all the negative
+ ions by the positive, if the electric field were exceedingly large
+ they would reach these plates before they had time to recombine, so
+ that each plate would receive N0 ions if the flash of Röntgen rays
+ produced N0 positive and N0 negative ions. With weaker fields the
+ number of ions received by the plates will be less as some of them
+ will recombine before they can reach the plates. We can find the
+ number of ions which reach the plates in this case in the following
+ way:--In consequence of the movement of the ions the slab of ionized
+ gas will broaden out and will consist of three portions, one in which
+ there are nothing but positive ions,--this is on the side of the
+ negative plate,--another on the side of the positive plate in which
+ there are nothing but negative ions, and a portion between these in
+ which there are both positive and negative ions; it is in this layer
+ that recombination takes place, and here if n is the number of
+ positive or negative ions at the time t after the flash of Röntgen
+ rays,
+
+ n = n0/(1 + [alpha]n0t).
+
+ With the same notation as before, the breadth of either of the outer
+ layers will in time dt increase by X(k1 + k2)dt, and the number of ions
+ in it by X(k1 + k2)ndt; these ions will reach the plate, the outer
+ layers will receive fresh ions until the middle one disappears, which
+ it will do after a time l/X(k1 + k2), where l is the thickness of the
+ slab ab of ionized gas; hence N, the number of ions reaching either
+ plate, is given by the equation
+ _
+ / l/X(k1+k2) n0X(k1 + k2) X(k1 + k2) / n0[alpha]l \
+ N = | --------------dt = ---------- log( 1 + ---------- ).
+ _/ 0 1 + n0[alpha]t [alpha] \ X(k1 + k2) /
+
+
+ If Q is the charge received by the plate,
+
+ X / Q0[epsilon]\
+ Q = Ne = -------------- log ( 1 + ----------- ),
+ 4[pi][epsilon] \ 4[pi]X /
+
+ where Q0 = n0le is the charge received by the plate when the electric
+ force is large enough to prevent recombination, and [epsilon] =
+ [alpha]4[pi]e(R1 + R2). We can from this result deduce the value of
+ [epsilon] and hence the value of [alpha] when R1+R2 is known.
+
+ _Distribution of Electric Force when a Current is passing through an
+ Ionized Gas._--Let the two plates be at right angles to the axis of x;
+ then we may suppose that between the plates the electric intensity X
+ is everywhere parallel to the axis of x. The velocities of both the
+ positive and negative ions are assumed to be proportional to X. Let
+ k1X, k2X represent these velocities respectively; let n1, n2 be
+ respectively the number of positive and negative ions per unit volume
+ at a point fixed by the co-ordinate x; let q be the number of positive
+ or negative ions produced in unit time per unit volume at this point;
+ and let the number of ions which recombine in unit volume in unit time
+ be [alpha]n1n2; then if e is the charge on the ion, the volume density
+ of the electrification is (n1 - n2)e, hence
+
+ dX
+ -- = 4[pi](n1 - n2)e (1).
+ dx
+
+ If I is the current through unit area of the gas and if we neglect any
+ diffusion except that caused by the electric field,
+
+ n1ek1X + n2ek2X = I (2).
+
+ From equations (1) and (2) we have
+
+ 1 / I k2 dX \
+ n1e = ------- ( - + ----- -- ) (3),
+ k1 + k2 \ X 4[pi] dx /
+
+ 1 / I k1 dX \
+ n2e = ------- ( - - ----- -- ) (4),
+ k1 + k2 \ X 4[pi] dx /
+
+ and from these equations we can, if we know the distribution of
+ electric intensity between the plates, calculate the number of
+ positive and negative ions.
+
+ In a steady state the number of positive and negative ions in unit
+ volume at a given place remains constant, hence neglecting the loss by
+ diffusion, we have
+
+ d
+ --(k1n1X) = q - [alpha]n1n2 (5).
+ dx
+
+ d
+ - --(k2n2X) = q - [alpha]n1n2 (6).
+ dx
+
+ If k1 and K2 are constant, we have from (1), (5) and (6)
+
+ d²X² / 1 1 \
+ ---- = 8[pi]e(q - [alpha]n1n2)( --- + --- ) (7),
+ dx² \ k1 k2 /
+
+ an equation which is very useful, because it enables us, if we know
+ the distribution of X², to find whether at any point in the gas the
+ ionization is greater or less than the recombination of the ions. We
+ see that q - [alpha]n1n2, which is the excess of ionization over
+ recombination, is proportional to d²X²/dx². Thus when the ionization
+ exceeds the recombination, i.e. when q - [alpha]n1n2 is positive, the
+ curve for X² is convex to the axis of x, while when the recombination
+ exceeds the ionization the curve for X² will be concave to the axis of
+ x. Thus, for example, fig. 11 represents the curve for X² observed by
+ Graham (_Wied. Ann._ 64, p. 49) in a tube through which a steady
+ current is passing. Interpreting it by equation (7), we infer that
+ ionization was much in excess of recombination at A and B, slightly so
+ along C, while along D the recombination exceeded the ionization.
+ Substituting in equation (7) the values of n1, n2 given in (3), (4),
+ we get
+ _ _
+ d²X² | [alpha] / k² dX²\ / k2 dX²\ | / 1 1 \
+ ---- = 8[pi]e |q - -------------- (1 + ----- --- ) (1 - ----- --- )| ( --- + --- ) (8).
+ dx² |_ e²X²(k1 + k2)² \ 8[pi] dx / \ 8[pi] dx /_| \k1 k2 /
+
+ [Illustration: Fig. 11.]
+
+ This equation can be solved (see Thomson, _Phil. Mag._ xlvii. P. 253),
+ when q is constant and k1 = k2. From the solution it appears that if
+ X1 be the value of x close to one of the plates, and X0 the value
+ midway between them,
+
+ 1
+ X1/X0 = ------------------
+ [beta]² - 2/[beta]
+
+ where [beta] = 8[pi]ek1/[alpha].
+
+
+ Since e = 4 × 10^-10, [alpha] = 2 × 10^-6, and k1 for air at
+ atmospheric pressure = 450, [beta] is about 2.3 for air at atmospheric
+ pressure and it becomes much greater at lower pressures.
+
+ Thus X1/X0 is always greater than unity, and the value of the ratio
+ increases from unity to infinity as [beta] increases from zero to
+ infinity. As [beta] does not involve either q or I, the ratio of X1 to
+ X0 is independent of the strength of the current and of the intensity
+ of the ionization.
+
+ No general solution of equation (8) has been found when k1 is not
+ equal to k2, but we can get an approximation to the solution when q is
+ constant. The equations (1), (2), (3), (4) are satisfied by the
+ values--
+
+ n1 = n2 = (q / [alpha])^½
+
+ k1
+ k1n1Xe = ------- I,
+ k1 + k2
+
+ k2
+ k2n2Xe = ------- I,
+ k1 + k2
+
+
+ /[alpha]\^½ I
+ X = ( ------- ) ----------.
+ \ q / e(k1 + k2)
+
+ These solutions cannot, however, hold right up to the surface of the
+ plates, for across each unit of area, at a point P, k1I/(k1+k2)e
+ positive ions pass in unit time, and these must all come from the
+ region between P and the positive plate. If [lambda] is the distance
+ of P from this plate, this region cannot furnish more than q[lambda]
+ positive ions, and only this number if there are no recombinations.
+ Hence the solution cannot hold when q[lambda] is less than k1I/(k1 +
+ k2)e, or where [lambda] is less than k1I/(k1 + k2)qe.
+
+ Similarly the solution cannot hold nearer to the negative plate than
+ the distance k2I/(k1 + k2)qe.
+
+ [Illustration: FIG. 12.]
+
+ The force in these layers will be greater than that in the middle of
+ the gas, and so the loss of ions by recombination will be smaller in
+ comparison with the loss due to the removal of the ions by the
+ current. If we assume that in these layers the loss of ions by
+ recombination can be neglected, we can by the method of the next
+ article find an expression for the value of the electric force at any
+ point in the layer. This, in conjunction with the value
+
+ /[alpha]\^½ I
+ X0 = ( ------- ) ----------
+ \ q / e(k1 + k2)
+
+ for the gas outside the layer, will give the value of X at any point
+ between the plates. It follows from this investigation that if X1 and
+ X2 are the values of X at the positive and negative plates
+ respectively, and X0 the value of X outside the layer,
+
+ / k1 I \^½ / k2 I \^½
+ X1 = X0 ( I + -- --------- ) , X2 = X0 ( I + -- --------- ) ,
+ \ k2 [epsilon]/ \ k1 [epsilon]/
+
+ where [epsilon] = [alpha]/4[pi]e(k1 + k2). Langevin found that for air
+ at a pressure of 152 mm. [epsilon] = 0.01, at 375 mm. [epsilon] =
+ 0.06, and at 760 mm. [epsilon] = 0.27. Thus at fairly low pressures
+ 1/[epsilon] is large, and we have approximately
+
+ /k1\^½ I /k2\^½ I
+ X1 = X0 ( -- ) ---------------, X2 = X0 ( -- ) ---------------.
+ \k2/ [root][epsilon] \k1/ [root][epsilon]
+
+ Therefore X1/X2 = k1/k2,
+
+ or the force at the positive plate is to that at the negative plate as
+ the velocity of the positive ion is to that of the negative ion. Thus
+ the force at the negative plate is greater than that at the positive.
+ The falls of potential V1, V2 at the two layers when 1/[epsilon] is
+ large can be shown to be given by the equations
+
+ /[epsilon]\^3/2 /k1\^1/2
+ V1 = 8[pi]²( --------- ) k1 ( -- ) i²,
+ \q [alpha]/ \k2/
+
+ /[epsilon]\^3/2 /k2\^1/2
+ V2 = 8[pi]²( --------- ) k2 ( -- ) i²,
+ \q [alpha]/ \k1/
+
+ hence V1/V2 = k1²/k2²,
+
+ so that the potential falls at the electrodes are proportional to the
+ squares of the velocities of the ions. The change in potential across
+ the layers is proportional to the square of the current, while the
+ potential change between the layers is proportional to the current,
+ the total potential difference between the plates is the sum of these
+ changes, hence the relation between V and i will be of the form
+
+ V = Ai + Bi².
+
+ Mie (_Ann. der. Phys._, 1904, 13, P. 857) has by the method of
+ successive approximations obtained solutions of equation (8) (i.) when
+ the current is only a small fraction of the saturation current, (ii.)
+ when the current is nearly saturated. The results of his
+ investigations are represented in fig. 12, which represents the
+ distribution of electric force along the path of the current for
+ various values of the current expressed as fractions of the saturation
+ current. It will be seen that until the current amounts to about
+ one-fifth of the maximum current, the type of solution is the one just
+ indicated, i.e. the electric force is constant except in the
+ neighbourhood of the electrodes when it increases rapidly.
+
+ Though we are unable to obtain a general solution of the equation (8),
+ there are some very important special cases in which that equation can
+ be solved without difficulty. We shall consider two of these, the
+ first being that when the current is saturated. In this case there is
+ no loss of ions by recombination, so that using the same notation as
+ before we have
+
+ d
+ --(n1k1X) = q,
+ dx
+
+ d
+ --(n2k2X) = -q.
+ dx
+
+ The solutions of which if q is constant are
+
+ n1k1X = qx,
+
+ n2k2X = I/e - qx = q(l - x),
+
+ if l is the distance between the plates, and x = 0 at the positive
+ electrode. Since
+
+ dX/dx = 4[pi](n1 - n2)e,
+
+ we get
+
+ 1 dX² / 1 1 \ l
+ ----- --- = qx ( -- + -- ) - q --,
+ 8[pi] d²x \k1 k2 / k2
+
+ or
+
+ X² x² / 1 1 \ lx
+ ----- = q -- ( -- + -- ) - q -- + C,
+ 8[pi] 2 \k1 k2 / k2
+
+ where C is a quantity to be determined by the condition that
+
+ _
+ / l
+ | Xdx = V,
+ _/0
+
+
+ where V is the given potential difference between the plates. When the
+ force is a minimum dX/dx = 0, hence at this point
+
+ lk1 lk2
+ x = -------, l - x = -------.
+ k1 + k2 k1 + k2
+
+ Hence the ratio of the distances of this point from the positive and
+ negative plates respectively is equal to the ratio of the velocities
+ of the positive and negative ions.
+
+ The other case we shall consider is the very important one in which
+ the velocity of the negative ion is exceedingly large compared with
+ the positive; this is the case in flames where, as Gold (_Proc. Roy.
+ Soc._ 97, p. 43) has shown, the velocity of the negative ion is many
+ thousand times the velocity of the positive; it is also very probably
+ the case in all gases when the pressure is low. We may get the
+ solution of this case either by putting k1/k2 = 0 in equation (8), or
+ independently as follows:--Using the same notation as before, we have
+
+ i = n1k1Xe + n2k2Xe,
+
+ d
+ --(n2k2X) = q - [alpha]n1n2,
+ dx
+
+ dX
+ -- = 4[pi](n1 - n2)e.
+ dx
+
+ In this case practically all the current is carried by the negative
+ ions so that i = n2k2Xe, and therefore q = [alpha]n1n2.
+
+ Thus
+
+ n2 = i/k2Xe, n1 = qk2Xe/[alpha]i.
+
+ Thus
+
+ dX 4[pi]e²k2qX 4[pi]i
+ -- = ----------- - ------,
+ dx [alpha]i k2X
+
+ or
+
+ dX² 8[pi]e²k2qX² 8[pi]i
+ -- - ------------ = - ------.
+ dx [alpha]i k2
+
+ The solution of this equation is
+
+ [alpha] i²
+ X² = ------- ----- + C[epsilon]^(8[pi]e²k2qx/[alpha]i)
+ q k2²e²
+
+ Here x is measured from the positive electrode; it is more convenient
+ in this case, however, to measure it from the negative electrode. If x
+ be the distance from the negative electrode at which the electric
+ force is X, we have from equation (7)
+
+ [alpha] i²
+ X² = ------- ----- + C¹[epsilon]^(8[pi]e²k2qx/[alpha]i)
+ q k2²e²
+
+ To find the value of C¹ we see by equation (7) that
+
+ d²X² k1k2 1
+ --- ------- ------ = q - [alpha]n1n2;
+ dX² k1 + k2 8[pi]e
+
+ hence
+ _ _ _
+ | dX² k1k2 1 |^x1 / x1
+ | --- ------- ------ | = | (q - [alpha]n1n2)dx.
+ |_dX k1 + k2 8[pi]e_| _/0
+
+ The right hand side of this equation is the excess of ionization over
+ recombination in the region extending from the cathode to x1; it must
+ therefore, when things are in a steady state, equal the excess of the
+ number of negative ions which leave this region over those which enter
+ it. The number which leave is i/e and the number which enter is i0/e,
+ if it is the current of negative ions coming from unit area of the
+ cathode, as hot metal cathodes emit large quantities of negative
+ electricity i0 may in some cases be considerable, thus the right hand
+ side of equation is (i - i0)/e. When x1 is large dX²/dx = 0; hence we
+ have from equation
+
+ [alpha]i(i - i0) k1 + k2
+ C¹ = ---------------- -------,
+ qk1k2e² k2
+
+ and since k1 is small compared with k2, we have
+
+ [alpha]i² / k2 i - i0 \
+ X² = --------- (1 + -- ------ [epsilon]^{-8[pi]e²k2·qx/[alpha]·i} ).
+ qk2²e² \ k1 i /
+
+ From the values which have been found for k2 and [alpha], we know that
+ 8[pi]ek2/[alpha] is a large quantity, hence the second term inside the
+ bracket will be very small when eqx is equal to or greater than i;
+ thus this term will be very small outside a layer of gas next the
+ cathode of such thickness that the number of ions produced on it would
+ be sufficient, if they were all utilized for the purpose, to carry the
+ current; in the case of flames this layer is exceedingly thin unless
+ the current is very large. The value of the electric force in the
+ uniform part of the field is equal to i/k2e·[root]([alpha]/q), while
+ when i0 = 0, the force at the cathode itself bears to the uniform
+ force the ratio of (k1 + k2)^½ to k1^½. As k1 is many thousand times
+ k2 the force increases with great rapidity as we approach the cathode;
+ this is a very characteristic feature of the passage of electricity
+ through flames and hot gases. Thus in an experiment made by H. A.
+ Wilson with a flame 18 cm. long, the drop of potential within 1
+ centimetre of the cathode was about five times the drop in the other
+ 17 cm. of the tube. The relation between the current and the potential
+ difference when the velocity of the negative ion is much greater than
+ the positive is very easily obtained. Since the force is uniform and
+ equal to i/k2e·[root]([alpha]/q), until we get close to the cathode
+ the fall of potential in this part of the discharge will be very
+ approximately equal to i/k2e·[root]([alpha]l/q), where l is the
+ distance between the electrodes. Close to the cathode, the electric
+ force when i0 is not nearly equal to i is approximately given by the
+ equation
+
+ i /[alpha]\^½
+ X = --------- (---------) [epsilon]^{-4[pi]e²k2qx/[alpha]i},
+ e(k1k2)^½ \ q / ,
+
+ and the fall of potential at the cathode is equal approximately to
+
+ _[oo]
+ /
+ | X dx,
+ _/0
+
+ that is to
+
+ i /[alpha]\^½ [alpha] i
+ --------- (---------) ----------.
+ e(k1k2)^½ \ q / 4[pi]e²k2q
+
+ The potential difference between the plates is the sum of the fall of
+ potential in the uniform part of the discharge plus the fall at the
+ cathode, hence
+
+ /[alpha]\^½ i / i[alpha]² 1 \
+ V = (---------) --- ( il + --------- ------------ ).
+ \ q / ek2 \ 4[pi]e²q [root](k1k2)/
+
+ The fall of potential at the cathode is proportional to the square of
+ the current, while the fall in the rest of the circuit is directly
+ proportional to the current. In the case of flames or hot gases, the
+ fall of potential at the cathode is much greater than that in the rest
+ of the circuit, so that in such cases the current through the gas
+ varies nearly as the square root of the potential difference. The
+ equation we have just obtained is of the form
+
+ V = Ai + Bi²,
+
+ and H. A. Wilson has shown that a relation of this form represents the
+ results of his experiments on the conduction of electricity through
+ flames.
+
+ The expression for the fall of potential at the cathode is inversely
+ proportional to q^(3/2), q being the number of ions produced per cubic
+ centimetre per second close to the cathode; thus any increase in the
+ ionization at the cathode will diminish the potential fall at the
+ cathode, and as practically the whole potential difference between the
+ electrodes occurs at the cathode, a diminution in the potential fall
+ there will be much more important than a diminution in the electric
+ force in the uniform part of the discharge, when the force is
+ comparatively insignificant. This consideration explains a very
+ striking phenomenon discovered many years ago by Hittorf, who found
+ that if he put a wire carrying a bead of a volatile salt into the
+ flame, it produced little effect upon the current, unless it were
+ placed close to the cathode where it gave rise to an enormous increase
+ in the current, sometimes increasing the current more than a
+ hundredfold. The introduction of the salt increases very largely the
+ number of ions produced, so that q is much greater for a salted flame
+ than for a plain one. Thus Hittorf's result coincides with the
+ conclusions we have drawn from the theory of this class of conduction.
+
+ The fall of potential at the cathode is proportional to i - i0, where
+ i0 is the stream of negative electricity which comes from the cathode
+ itself, thus as i0 increases the fall of potential at the cathode
+ diminishes and the current sent by a given potential difference
+ through the gas increases. Now all metals give out negative particles
+ when heated, at a rate which increases very rapidly with the
+ temperature, but at the same temperature some metals give out more
+ than others. If the cathode is made of a metal which emits large
+ quantities of negative particles, (i - i0) will for a given value of i
+ be smaller than if the metal only emitted a small number of
+ particles; thus the cathode fall will be smaller for the metal with
+ the greater emissitivity, and the relation between the potential
+ difference and the current will be different in the two cases. These
+ considerations are confirmed by experience, for it has been found that
+ the current between electrodes immersed in a flame depends to a great
+ extent upon the metal of which the electrodes are made. Thus
+ Pettinelli (_Acc. dei Lincei_ [5], v. p. 118) found that, _ceteris
+ paribus_, the current between two carbon electrodes was about 500
+ times that between two iron ones. If one electrode was carbon and the
+ other iron, the current when the carbon was cathode and the iron anode
+ was more than 100 times the current when the electrodes were reversed.
+ The emission of negative particles by some metallic oxides, notably
+ those of calcium and barium, has been shown by Wehnelt (_Ann. der
+ Phys._ 11, p. 425) to be far greater than that of any known metal, and
+ the increase of current produced by coating the cathodes with these
+ oxides is exceedingly large; in some cases investigated by Tufts and
+ Stark (_Physik. Zeits._, 1908, 5, p. 248) the current was increased
+ many thousand times by coating the cathode with lime. No appreciable
+ effect is produced by putting lime on the anode.
+
+ _Conduction when all the Ions are of one Sign._--There are many
+ important cases in which the ions producing the current come from one
+ electrode or from a thin layer of gas close to the electrode, no
+ ionization occurring in the body of the gas or at the other electrode.
+ Among such cases may be mentioned those where one of the electrodes is
+ raised to incandescence while the other is cold, or when the negative
+ electrode is exposed to ultra-violet light. In such cases if the
+ electrode at which the ionization occurs is the positive electrode,
+ all the ions will be positively charged, while if it is the negative
+ electrode the ions will all be charged negatively. The theory of this
+ case is exceedingly simple. Suppose the electrodes are parallel planes
+ at right angles to the axis of x; let X be the electric force at a
+ distance x from the electrode where the ionization occurs, n the
+ number of ions (all of which are of one sign) at this place per cubic
+ centimetre, k the velocity of the ion under unit electric force, e the
+ charge on an ion, and i the current per unit area of the electrode.
+ Then we have dX/dx = 4[pi]ne, and if u is the velocity of the ion neu
+ = i. But u = kX, hence we have kX/4[pi] · dX/dx = i, and since the
+ right hand side of this equation does not depend upon x, we get
+ kX²/8[pi] = ix + C, where C is a constant to be determined. If l is
+ the distance between the plates, and V the potential difference
+ between them,
+ _ _____ _ _
+ / l 1 /8[pi] | |
+ V = | Xdx = --- / ----- | ( il + C )^3/2 - C^3/2 |.
+ _/0 i \/ k |_ _|
+
+ We shall show that when the current is far below the saturation value,
+ C is very small compared with il, so that the preceding equation
+ becomes
+
+ V² = 8[pi]l³i/k (1).
+
+ To show that for small currents C is small compared with il, consider
+ the case when the ionization is confined to a thin layer, thickness d
+ close to the electrode, in that layer let n0 be the value of n, then
+ we have q = [alpha]n0² + i/ed. If X0 be the value of X when x = 0,
+ kX0n0e = i, and,
+
+ kX0² i² [alpha] i²
+ C = ----- = ----------- = -------- · -------- (2).
+ 8[pi] n0²ke·8[pi] 8[pi]ke² q + i/ed
+
+ Since [alpha]/8[pi]ke is, as we have seen, less than unity, C will be
+ small compared with il, if i/(eq + i/d) is small compared with l. If
+ I0 is the saturation current, q = I0/ed, so that the former expression
+ = id/(I0 + i), if i is small compared with I0, this expression is
+ small compared with d, and therefore _a fortiori_ compared with l, so
+ that we are justified in this case in using equation (1).
+
+ From equation (2) we see that the current increases as the square of
+ the potential difference. Here an increase in the potential difference
+ produces a much greater percentage increase than in conduction through
+ metals, where the current is proportional to the potential difference.
+ When the ionization is distributed through the gas, we have seen that
+ the current is approximately proportional to the square root of the
+ potential, and so increases more slowly with the potential difference
+ than currents through metals. From equation (1) the current is
+ inversely proportional to the cube of the distance between the
+ electrodes, so that it falls off with great rapidity as this distance
+ is increased. We may note that for a given potential difference the
+ expression for the current does not involve q, the rate of production
+ of the ions at the electrode, in other words, if we vary the
+ ionization the current will not begin to be affected by the strength
+ of the ionization until this falls so low that the current is a
+ considerable fraction of the saturation current. For the same
+ potential difference the current is proportional to k, the velocity
+ under unit electric force of the ion which carries the current. As the
+ velocity of the negative ion is greater than that of the positive, the
+ current when the ionization is confined to the neighbourhood of one of
+ the electrodes will be greater when that electrode is made cathode
+ than when it is anode. Thus the current will appear to pass more
+ easily in one direction than in the opposite.
+
+ Since the ions which carry the current have to travel all the way from
+ one electrode to the other, any obstacle which is impervious to these
+ ions will, if placed between the electrodes, stop the current to the
+ electrode where there is no ionization. A plate of metal will be as
+ effectual as one made of a non-conductor, and thus we get the
+ remarkable result that by interposing a plate of an excellent
+ conductor like copper or silver between the electrode, we can entirely
+ stop the current. This experiment can easily be tried by using a hot
+ plate as the electrode at which the ionization takes place: then if
+ the other electrode is cold the current which passes when the hot
+ plate is cathode can be entirely stopped by interposing a cold metal
+ plate between the electrodes.
+
+_Methods of counting the Number of Ions._--The detection of the ions and
+the estimation of their number in a given volume is much facilitated by
+the property they possess of promoting the condensation of water-drops
+in dust-free air supersaturated with water vapour. If such air contains
+no ions, then it requires about an eightfold supersaturation before any
+water-drops are formed; if, however, ions are present C. T. R. Wilson
+(_Phil. Trans._ 189, p. 265) has shown that a sixfold supersaturation is
+sufficient to cause the water vapour to condense round the ions and to
+fall down as raindrops. The absence of the drops when no ions are
+present is due to the curvature of the drop combined with the surface
+tension causing, as Lord Kelvin showed, the evaporation from a small
+drop to be exceeding rapid, so that even if a drop of water were formed
+the evaporation would be so great in its early stages that it would
+rapidly evaporate and disappear. It has been shown, however (J. J.
+Thomson, _Application of Dynamics to Physics and Chemistry_, p. 164;
+_Conduction of Electricity through Gases_, 2nd ed. p. 179), that if a
+drop of water is charged with electricity the effect of the charge is to
+diminish the evaporation; if the drop is below a certain size the effect
+the charge has in promoting condensation more than counterbalances the
+effect of the surface tension in promoting evaporation. Thus the
+electric charge protects the drop in the most critical period of its
+growth. The effect is easily shown experimentally by taking a bulb
+connected with a piston arranged so as to move with great rapidity. When
+the piston moves so as to increase the volume of the air contained in
+the bulb the air is cooled by expansion, and if it was saturated with
+water vapour before it is supersaturated after the expansion. By
+altering the throw of the piston the amount of supersaturation can be
+adjusted within very wide limits. Let it be adjusted so that the
+expansion produces about a sixfold supersaturation; then if the gas is
+not exposed to any ionizing agents very few drops (and these probably
+due to the small amount of ionization which we have seen is always
+present in gases) are formed. If, however, the bulb is exposed to strong
+Röntgen rays expansion produces a dense cloud which gradually falls down
+and disappears. If the gas in the bulb at the time of its exposure to
+the Röntgen rays is subject to a strong electric field hardly any cloud
+is formed when the gas is suddenly expanded. The electric field removes
+the charged ions from the gas as soon as they are formed so that the
+number of ions present is greatly reduced. This experiment furnishes a
+very direct proof that the drops of water which form the cloud are only
+formed round the ions.
+
+This method gives us an exceedingly delicate test for the presence of
+ions, for there is no difficulty in detecting ten or so raindrops per
+cubic centimetre; we are thus able to detect the presence of this number
+of ions. This result illustrates the enormous difference between the
+delicacy of the methods of detecting ions and those for detecting
+uncharged molecules; we have seen that we can easily detect ten ions per
+cubic centimetre, but there is no known method, spectroscopic or
+chemical, which would enable us to detect a billion (10^12) times this
+number of uncharged molecules. The formation of the water-drops round
+the charged ions gives us a means of counting the number of ions present
+in a cubic centimetre of gas; we cool the gas by sudden expansion until
+the supersaturation produced by the cooling is sufficient to cause a
+cloud to be formed round the ions, and the problem of finding the number
+of ions per cubic centimetre of gas is thus reduced to that of finding
+the number of drops per cubic centimetre in the cloud. Unless the drops
+are very few and far between we cannot do this by direct counting; we
+can, however, arrive at the result in the following way. From the amount
+of expansion of the gas we can calculate the lowering produced in its
+temperature and hence the total quantity of water precipitated. The
+water is precipitated as drops, and if all the drops are the same size
+the number per cubic centimetre will be equal to the volume of water
+deposited per cubic centimetre, divided by the volume of one of the
+drops. Hence we can calculate the number of drops if we know their size,
+and this can be determined by measuring the velocity with which they
+fall under gravity through the air.
+
+ The theory of the fall of a heavy drop of water through a viscous
+ fluid shows that v = (2/9)ga²/[mu], where a is the radius of the drop,
+ g the acceleration due to gravity, and [mu] the coefficient of
+ viscosity of the gas through which the drop falls. Hence if we know v
+ we can deduce the value of a and hence the volume of each drop and the
+ number of drops.
+
+ _Charge on Ion._--By this method we can determine the number of ions
+ per unit volume of an ionized gas. Knowing this number we can proceed
+ to determine the charge on an ion. To do this let us apply an electric
+ force so as to send a current of electricity through the gas, taking
+ care that the current is only a small fraction of the saturating
+ current. Then if u is the sum of the velocities of the positive and
+ negative ions produced in the electric field applied to the gas, the
+ current through unit area of the gas is neu, where n is the number of
+ positive or negative ions per cubic centimetre, and e the charge on an
+ ion. We can easily measure the current through the gas and thus
+ determine neu; we can determine n by the method just described, and u,
+ the velocity of the ions under the given electric field, is known from
+ the experiments of Zeleny and others. Thus since the product neu, and
+ two of the factors n, u are known, we can determine the other factor
+ e, the charge on the ion. This method was used by J. J. Thomson, and
+ details of the method will be found in _Phil. Mag._ [5], 46, p. 528;
+ [5], 48, p. 547; [6], 5, p. 346. The result of these measurements
+ shows that the charge on the ion is the same whether the ionization is
+ by Röntgen rays or by the influence of ultra-violet light on a metal
+ plate. It is the same whether the gas ionized is hydrogen, air or
+ carbonic acid, and thus is presumably independent of the nature of the
+ gas. The value of e formed by this method was 3.4 × 10^-10
+ electrostatic units.
+
+ H. A. Wilson (_Phil. Mag._ [6], 5, p. 429) used another method. Drops
+ of water, as we have seen, condense more easily on negative than on
+ positive ions. It is possible, therefore, to adjust the expansion so
+ that a cloud is formed on the negative but not on the positive ions.
+ Wilson arranged the experiments so that such a cloud was formed
+ between two horizontal plates which could be maintained at different
+ potentials. The charged drops between the plates were acted upon by a
+ uniform vertical force which affected their rate of fall. Let X be the
+ vertical electric force, e the charge on the drop, v1 the rate of fall
+ of the drop when this force acts, and v the rate of fall due to
+ gravity alone. Then since the rate of fall is proportionate to the
+ force on the drop, if a is the radius of the drop, and [rho] its
+ density, then
+
+ Xe + (4/3)[pi][rho]ga³ v1
+ ---------------------- = ---,
+ 4/3[pi][rho]ga³ v
+
+ or Xe = (4/3)[pi][rho]ga³(v1 - v)/v.
+
+ But v = 2/9ga²[rho]/[mu],
+
+ so that
+
+ / / [mu]³ v^(3/2)(v1 - v)
+ Xe = \/ 2.9[pi] - / ------ · ---------------.
+ \/ g[rho] v
+
+ Thus if X, v, v1 are known e can be determined. Wilson by this method
+ found that e was 3.1 × 10^-10 electrostatic units. A few of the ions
+ carried charges 2e or 3e.
+
+ Townsend has used the following method to compare the charge carried
+ by a gaseous ion with that carried by an atom of hydrogen in the
+ electrolysis of solution. We have
+
+ u/D = Ne/[Pi],
+
+ where D is the coefficient of diffusion of the ions through the gas, u
+ the velocity of the ion in the same gas when acted on by unit electric
+ force, N the number of molecules in a cubic centimetre of the gas when
+ the pressure is [Pi] dynes per square centimetre, and e the charge in
+ electrostatic units. This relation is obtained on the hypothesis that
+ N ions in a cubic centimetre produce the same pressure as N uncharged
+ molecules.
+
+ We know the value of D from Townsend's experiments and the values of u
+ from those of Zeleny. We get the following values for Ne × 10^-10:--
+
+ +---------------+---------------------+---------------------+
+ | | Moist Gas. | Moist Gas. |
+ +---------------+----------+----------+----------+----------+
+ | Gas. | Positive | Negative | Positive | Negative |
+ | | Ions. | Ions. | Ions. | Ions. |
+ +---------------+----------+----------+----------+----------+
+ | Air | 1.28 | 1.29 | 1.46 | 1.31 |
+ | Oxygen | 1.34 | 1.27 | 1.63 | 1.36 |
+ | Carbonic acid | 1.01 | .87 | .99 | .93 |
+ | Hydrogen | 1.24 | 1.18 | 1.63 | 1.25 |
+ +---------------+----------+----------+----------+----------+
+ | Mean | 1.22 | 1.15 | 1.43 | 1.21 |
+ +---------------+----------+----------+----------+----------+
+
+
+ Since 1.22 cubic centimetres of hydrogen at the temperature 15° C. and
+ pressure 760 mm. of mercury are liberated by the passage through
+ acidulated water of one electromagnetic unit of electricity or 3 ×
+ 10^10 electrostatic units, and since in one cubic centimetre of the
+ gas there are 2.46 N atoms of hydrogen, we have, if E is the charge in
+ electrostatic units, on the atom of hydrogen in the electrolysis of
+ solutions
+
+ 2.46NE = 3 × 10^10,
+
+ or
+
+ NE = 1.22 × 10^10.
+
+ The mean of the values of Ne in the preceding table is 1.24 × 10^10.
+ Hence we may conclude that the charge of electricity carried by a
+ gaseous ion is equal to the charge carried by the hydrogen atom in the
+ electrolysis of solutions. The values of Ne for the different gases
+ differ more than we should have expected from the probable accuracy of
+ the determination of D and the velocity of the ions: Townsend (_Proc.
+ Roy. Soc._ 80, p. 207) has shown that when the ionization is produced
+ by Röntgen rays some of the positive ions carry a double charge and
+ that this accounts for the values of Ne being greater for the positive
+ than for the negative ions. Since we know the value of e, viz. 3.5 ×
+ 10^-10, and, also Ne, = 1.24 × 10^10, we find N the number of
+ molecules in a cubic centimetre of gas at standard temperature and
+ pressure to be equal to 3.5 × 10^19. This method of obtaining N is the
+ only one which does not involve any assumption as to the shape of the
+ molecules and the forces acting between them.
+
+ Another method of determining the charge carried by an ion has been
+ employed by Rutherford (_Proc. Roy. Soc._ 81, pp. 141, 162), in which
+ the positively electrified particles emitted by radium are made use
+ of. The method consists of: (1) Counting the number of [alpha]
+ particles emitted by a given quantity of radium in a known time. (2)
+ Measuring the electric charge emitted by this quantity in the same
+ time. To count the number of the [alpha] particles the radium was so
+ arranged that it shot into an ionization chamber a small number of
+ [alpha] particles per minute; the interval between the emission of
+ individual particles was several seconds. When an [alpha] particle
+ passed into the vessel it ionized the gas inside and so greatly
+ increased its conductivity; thus, if the gas were kept exposed to an
+ electric field, the current through the gas would suddenly increase
+ when an [alpha] particle passed into the vessel. Although each [alpha]
+ particle produces about thirty thousand ions, this is hardly large
+ enough to produce the conductivity appreciable without the use of very
+ delicate apparatus; to increase the conductivity Rutherford took
+ advantage of the fact that ions, especially negative ones, when
+ exposed to a strong electric field, produce other ions by collision
+ against the molecules of the gas through which they are moving. By
+ suitably choosing the electric field and the pressure in the
+ ionization chamber, the 30,000 ions produced by each [alpha] particle
+ can be multiplied to such an extent that an appreciable current passes
+ through the ionization chamber on the arrival of each [alpha]
+ particle. An electrometer placed in series with this vessel will show
+ by its deflection when an [alpha] particle enters the chamber, and by
+ counting the number of deflections per minute we can determine the
+ number of [alpha] particles given out by the radium in that time.
+ Another method of counting this number is to let the particles fall on
+ a phosphorescent screen, and count the number of scintillations on the
+ screen in a certain time. Rutherford has shown that these two methods
+ give concordant results.
+
+ The charge of positive electricity given out by the radium was
+ measured by catching the [alpha] particles in a Faraday cylinder
+ placed in a very highly exhausted vessel, and measuring the charge per
+ minute received by this cylinder. In this way Rutherford showed that
+ the charge on the [alpha] particle was 9.4 × 10^-10 electrostatic
+ units. Now e/m for the [alpha] particle = 5 × 10³, and there is
+ evidence that the [alpha] particle is a charged atom of helium; since
+ the atomic weight of helium is 4 and e/m for hydrogen is 10^4, it
+ follows that the charge on the helium atom is twice that on the
+ hydrogen, so that the charge on the hydrogen atom is 4.7 × 10^-10
+ electrostatic units.
+
+_Calculation of the Mass of the Ions at Low Pressures._--Although at
+ordinary pressures the ion seems to have a very complex structure and to
+be the aggregate of many molecules, yet we have evidence that at very
+low pressures the structure of the ion, and especially of the negative
+one, becomes very much simpler. This evidence is afforded by
+determination of the mass of the atom. We can measure the ratio of the
+mass of an ion to the charge on the ion by observing the deflections
+produced by magnetic and electric forces on a moving ion. If an ion
+carrying a charge e is moving with a velocity v, at a point where the
+magnetic force is H, a mechanical force acts on the ion, whose direction
+is at right angles both to the direction of motion of the ion and to the
+magnetic force, and whose magnitude is evH sin [theta], where [theta] is
+the angle between v and H. Suppose then that we have an ion moving
+through a gas whose pressure is so low that the free path of the ion is
+long compared with the distance through which it moves whilst we are
+experimenting upon it; in this case the motion of the ion will be free,
+and will not be affected by the presence of the gas.
+
+ Since the force is always at right angles to the direction of motion
+ of the ion, the speed of the ion will not be altered by the action of
+ this force; and if the ion is projected with a velocity v in a
+ direction at right angles to the magnetic force, and if the magnetic
+ force is constant in magnitude and direction, the ion will describe a
+ curve in a plane at right angles to the magnetic force. If [rho] is
+ the radius of curvature of this curve, m the mass of the ion,
+ mv²/[rho] must equal the normal force acting on the ion, i.e. it must
+ be equal to Hev, or [rho] = mv/He. Thus the radius of curvature is
+ constant; the path is therefore a circle, and if we can measure the
+ radius of this circle we know the value of mv/He. In the case of the
+ rapidly moving negative ions projected from the cathode in a highly
+ exhausted tube, which are known as _cathode rays_, the path of the
+ ions can be readily determined since they make many substances
+ luminous when they impinge against them. Thus by putting a screen of
+ such a substance in the path of the rays the shape of the path will be
+ determined. Let us now suppose that the ion is acted upon by a
+ vertical electric force X and is free from magnetic force, if it be
+ projected with a horizontal velocity v, the vertical deflection y
+ after a time t is ½ × et²/m, or if l is the horizontal distance
+ travelled over by the ion in this time we have since l = vt,
+
+ Xe l²
+ y = ½ -- --.
+ m v²
+
+ Thus if we measure y and l we can deduce e/mv². From the effect of the
+ magnetic force we know e/mv. Combining these results we can find both
+ e/m and v.
+
+ [Illustration: FIG. 13.]
+
+ The method by which this determination is carried out in practice is
+ illustrated in fig. 13. The cathode rays start from the electrode C in
+ a highly exhausted tube, pass through two small holes in the plugs A
+ and B, the holes being in the same horizontal line. Thus a pencil of
+ rays emerging from B is horizontal and produces a bright spot at the
+ far end of the tube. In the course of their journey to the end of the
+ tube they pass between the horizontal plates E and D, by connecting
+ these plates with an electric battery a vertical electric field is
+ produced between E and D and the phosphorescent spot is deflected. By
+ measuring this deflection we determine e/mv². The tube is now placed
+ in a uniform magnetic field, the lines of magnetic force being
+ horizontal and at right angles to the plane of the paper. The magnetic
+ force makes the rays describe a circle in the plane of the paper, and
+ by measuring the vertical deflection of the phosphorescent patch at
+ the end of the tube we can determine the radius of this circle, and
+ hence the value of e/mv. From the two observations the value of e/m
+ and v can be calculated.
+
+ Another method of finding e/m for the negative ion which is applicable
+ in many cases to which the preceding one is not suitable, is as
+ follows: Let us suppose that the ion starts from rest and moves in a
+ field where the electric and magnetic forces are both uniform, the
+ electric force X being parallel to the axis of x, and the magnetic
+ force Z parallel to the axis of z; then if x, y, are the co-ordinates
+ of the ion at the time t, the equations of motion of the ion are--
+
+ d²x dy
+ m --- = Xe - He -- ,
+ dt² dt
+
+ d²y dx
+ m --- = He --.
+ dt² dt
+
+ The solution of these equations, if x, y, dx/dt, dy/dt all vanish when
+ t = 0, is
+
+ Xm / / e \ \
+ x = --- {1 - cos( -- Ht ) }
+ eH² \ \ m / /
+
+ Xm /e / e \ \
+ y = --- {-- Ht - sin( -- Ht ) }.
+ eH² \m \ m / /
+
+ These equations show that the path of the ion is a cycloid, the
+ generating circle of which has a diameter equal to 2Xm/eH², and rolls
+ on the line x = 0.
+
+ Suppose now that we have a number of ions starting from the plane x =
+ 0, and moving towards the plane x = a. The particles starting from x =
+ 0 describe cycloids, and the greatest distance they can get from the
+ plane is equal to the diameter of the generating circle of the
+ cycloid, i.e. to 2Xm/eH². (After reaching this distance they begin to
+ approach the plane.) Hence if a is less than the diameter of the
+ generating circle, all the particles starting from x = 0 will reach
+ the plane x = a, if this is unlimited in extent; while if a is greater
+ than the diameter of the generating circle none of the particles which
+ start from x = 0 will reach the plane x = a. Thus, if x = 0 is a plane
+ illuminated by ultra-violet light, and consequently the seat of a
+ supply of negative ions, and x = a a plane connected with an
+ electrometer, then if a definite electric intensity is established
+ between the planes, i.e. if X be fixed, so that the rate of emission
+ of negative ions from the illuminated plate is given, and if a is less
+ than 2Xm/eH², all the ions which start from x = 0 will reach x = a.
+ That is, the rate at which this plane receives an electric charge
+ will be the same whether there is a magnetic field between the plate
+ or not, but if a is greater than 2Xm/eH², then no particle which
+ starts from the plate x = 0 will reach the plate x = a, and this plate
+ will receive no charge. Thus the supply of electricity to the plate
+ has been entirely stopped by the magnetic field. Thus, on this theory,
+ if the distance between the plates is less than a certain value, the
+ magnetic force should produce no effect on the rate at which the
+ electrometer plate receives a charge, while if the distance is greater
+ than this value the magnetic force would completely stop the supply of
+ electricity to the plate. The actual phenomena are not so abrupt as
+ this theory indicates. We find that when the plates are very near
+ together the magnetic force produces a very slight effect, and this an
+ increase in the rate of charging of the plate. On increasing the
+ distance we come to a stage where the magnetic force produces a great
+ diminution in the rate of charging. It does not, however, stop it
+ abruptly, there being a considerable range of distance, in which the
+ magnetic force diminishes but does not destroy the current. At still
+ greater distances the current to the plate under the magnetic force is
+ quite inappreciable compared with that when there is no magnetic
+ force. We should get this gradual instead of abrupt decay of the
+ current if some of the particles, instead of all starting from rest,
+ started with a finite velocity; in that case the first particles
+ stopped would be those which started from rest. This would be when a =
+ 2Xm/eH². Thus if we measure the value of a when the magnetic force
+ first begins to affect the leak to the electrometer we determine
+ 2Xm/eH², and as we can easily measure X and H, we can deduce the value
+ of m/e.
+
+By these methods Thomson determined the value of e/m for the negative
+ions produced when ultra-violet light falls on a metal plate, as well as
+for the negative ions produced by an incandescent carbon filament in an
+atmosphere of hydrogen (_Phil. Mag._ [5], 48, p. 547) as well as for the
+cathode rays. It was found that the value of e/m for the negative ions
+was the same in all these cases, and that it was a constant quantity
+independent of the nature of the gas from which the ions are produced
+and the means used to produce them. It was found, too, that this value
+was more than a thousand times the value of e/M, where e is the charge
+carried by an atom of hydrogen in the electrolysis of solutions, and M
+the mass of an atom of hydrogen. We have seen that this charge is the
+same as that carried by the negative ion in gases; thus since e/m is
+more than a thousand times e/M, it follows that M must be more than a
+thousand times m. Thus the mass of the negative ion is exceedingly small
+compared with the mass of the atom of hydrogen, the smallest mass
+recognized in chemistry. The production of negative ions thus involves
+the splitting up of the atom, as from a collection of atoms something is
+detached whose mass is less than that of a single atom. It is important
+to notice in connexion with this subject that an entirely different line
+of argument, based on the Zeeman effect (see MAGNETO-OPTICS), leads to
+the recognition of negatively electrified particles for which e/m is of
+the same order as that deduced from the consideration of purely
+electrical phenomena. These small negatively electrified particles are
+called corpuscles. The latest determinations of e/m for corpuscles
+available are the following:--
+
+ Observer. e/m.
+
+ Classen (_Ber. deut. phys. Ges._ 6, p. 700) 1.7728 × 10^7
+ Bucherer (_Ann. der Phys._, 28, p. 513) 1.763 × 10^7
+
+It follows from electrical theory that when the corpuscles are moving
+with a velocity comparable with that of light their masses increase
+rapidly with their velocity. This effect has been detected by Kauffmann
+(_Gött. Nach._, Nov. 8, 1901), who used the corpuscles shot out from
+radium, some of which move with velocities only a few per cent less than
+that of light. Other experiments on this point have been made by
+Bucherer (_Ann. der Phys._ 28, p. 513).
+
+_Conductivity Produced by Ultra-Violet Light._--So much use has been
+made in recent times of ultra-violet light for producing ions that it is
+desirable to give some account of the electrical effects produced by
+light. The discovery by Hertz (_Wied. Ann._ 31, p. 983) in 1887, that
+the incidence of ultra-violet light on a spark gap facilitates the
+passage of a spark, led to a series of investigations by Hallwachs,
+Hoor, Righi and Stoletow, on the effect of ultra-violet light on
+electrified bodies. These researches have shown that a freshly cleaned
+metal surface, charged with negative electricity, rapidly loses its
+charge, however small, when exposed to ultra-violet light, and that if
+the surface is insulated and without charge initially, it acquires a
+positive charge under the influence of the light. The magnitude of this
+positive charge may be very much increased by directing a blast of air
+on the plate. This, as Zeleny (_Phil. Mag._ [5], 45, p. 272) showed, has
+the effect of blowing from the neighbourhood of the plate negatively
+electrified gas, which has similar properties to the charged gas
+obtained by the separation of ions from a gas exposed to Röntgen rays or
+uranium radiation. If the metal plate is positively electrified, there
+is no loss of electrification caused by ultra-violet light. This has
+been questioned, but a very careful examination of the question by
+Elster and Geitel (_Wied. Ann._ 57, p. 24) has shown that the apparent
+exceptions are due to the accidental exposure to reflected ultra-violet
+light of metal surfaces in the neighbourhood of the plate negatively
+electrified by induction, so that the apparent loss of charge is due to
+negative electricity coming up to the plate, and not to positive
+electricity going away from it. The ultra-violet light may be obtained
+from an arc-lamp, the effectiveness of which is increased if one of the
+terminals is made of zinc or aluminium, the light from these substances
+being very rich in ultra-violet rays; it may also be got very
+conveniently by sparking with an induction coil between zinc or cadmium
+terminals. Sunlight is not rich in ultra-violet light, and does not
+produce anything like so great an effect as the arc light. Elster and
+Geitel, who have investigated with great success the effects of light on
+electrified bodies, have shown that the more electro-positive metals
+lose negative charges when exposed to ordinary light, and do not need
+the presence of the ultra-violet rays. Thus they found that amalgams of
+sodium or potassium enclosed in a glass vessel lose a negative charge
+when exposed to daylight, though the glass stops the small amount of
+ultra-violet light left in sunlight after its passage through the
+atmosphere. If sodium or potassium be employed, or, what is more
+convenient, the mercury-like liquid obtained by mixing sodium and
+potassium in the proportion of their combining weights, they found that
+negative electricity was discharged by an ordinary petroleum lamp. If
+the still more electro-positive metal rubidium is used, the discharge
+can be produced by the light from a glass rod just heated to redness;
+but there is no discharge till the glass is luminous. Elster and Geitel
+arrange the metals in the following order for the facility with which
+negative electrification is discharged by light: rubidium, potassium,
+alloy of sodium and potassium, sodium, lithium, magnesium, thallium,
+zinc. With copper, platinum, lead, iron, cadmium, carbon and mercury the
+effects with ordinary light are too small to be appreciable. The order
+is the same as that in Volta's electro-chemical series. With
+ultra-violet light the different metals show much smaller differences in
+their power of discharging negative electricity than they do with
+ordinary light. Elster and Geitel found that the ratio of the
+photo-electric effects of two metals exposed to approximately
+monochromatic light depended upon the wave-length of the light,
+different metals showing a maximum sensitiveness in different parts of
+the spectrum. This is shown by the following table for the alkaline
+metals. The numbers in the table are the rates of emission of negative
+electricity under similar circumstances. The rate of emission under the
+light from a petroleum lamp was taken as unity:--
+
+ Blue. Yellow. Orange. Red.
+ Rb .16 .64 .33 .039
+ Na .37 .36 .14 .009
+ K .57 .07 .04 .002
+
+The table shows that the absorption of light by the metal has great
+influence on the photo-electric effect, for while potassium is more
+sensitive in blue light than sodium, the strong absorption of yellow
+light by sodium makes it more than five times more sensitive to this
+light than potassium. Stoletow, at an early period, called attention to
+the connexion between strong absorption and photo-electric effects. He
+showed that water, which does not absorb to any great extent either the
+ultra-violet or visible rays, does not show any photo-electric effect,
+while strongly coloured solutions, and especially solutions of
+fluorescent substances such as methyl green or methyl violet, do so to a
+very considerable extent; indeed, a solution of methyl green is more
+sensitive than zinc. Hallwachs (_Wied. Ann._ 37, p. 666) proved that in
+liquids showing photo-electric effects there is always strong
+absorption; we may, however, have absorption without these effects.
+Phosphorescent substances, such as calcium sulphide show this effect, as
+also do various specimens of fluor-spar. As phosphorescence and
+fluorescence are probably accompanied by a very intense absorption by
+the surface layers, the evidence is strong that to get the
+photo-electric effects we must have strong absorption of some kind of
+light, either visible or ultra-violet.
+
+[Illustration: FIG. 14.]
+
+If a conductor A is placed near a conductor B exposed to ultra-violet
+light, and if B is made the negative electrode and a difference of
+potential established between A and B, a current of electricity will
+flow between the conductors. The relation between the magnitude of the
+current and the difference of potential when A and B are parallel plates
+has been investigated by Stoletow (_Journal de physique_, 1890, 11, p.
+469), von Schweidler (_Wien. Ber._, 1899, 108, p. 273) and Varley
+(_Phil. Trans. A._, 1904, 202, p. 439). The results of some of Varley's
+experiments are represented in the curves shown in fig. 14, in which the
+ordinates are the currents and the abscissae the potentials. It will be
+seen that when the pressure is exceedingly low the current is
+independent of the potential difference and is equal to the negative
+charge carried off in unit time by the corpuscles emitted from the
+surface exposed to the light. At higher pressures the current rises far
+above these values and increases rapidly with the potential difference.
+This is due to the corpuscles emitted by the illuminated surface
+acquiring under the electric field such high velocities that when they
+strike against the molecules of the gas through which they are passing
+they ionize them, producing fresh ions which can carry on additional
+current. The relation between the current and the potential difference
+in this case is in accordance with the results of the theory of
+ionization by collision. The corpuscles emitted from a body under the
+action of ultra-violet light start from the surface with a finite
+velocity. The velocity is not the same for all the corpuscles, nor
+indeed could we expect that it should be: for as Ladenburg has shown
+(_Ann. der Phys._, 1903, 12, p. 558) the seat of their emission is not
+confined to the surface layer of the illuminated metal but extends to a
+layer of finite, though small, thickness. Thus the particles which start
+deep down will have to force their way through a layer of metal before
+they reach the surface, and in doing so will have their velocities
+retarded by an amount depending on the thickness of this layer. The
+variation in the velocity of the corpuscles is shown in the following
+table, due to Lenard (_Ann. der Phys._, 1902, 8, p. 149).
+
+ +------------------------------------+--------+----------+-----------+
+ | | Carbon.| Platinum.| Aluminium.|
+ +------------------------------------+--------+----------+-----------+
+ | Corpuscles emitted with velocities | | | |
+ | between 12 and 8 × 10^7 cm sec. | 0.000 | 0.000 | 0.004 |
+ | between 8 and 4 × 10^7 cm sec. | 0.049 | 0.155 | 0.151 |
+ | between 4 and 0 × 10^7 cm sec. | 0.67 | 0.65 | 0.49 |
+ | | | | |
+ | Corpuscles only emitted with the | | | |
+ | help of an external electric | 0.28 | 0.21 | 0.35 |
+ | field. +--------+----------+-----------|
+ | | 1.00 | 1.00 | 1.00 |
+ +------------------------------------+--------+----------+-----------+
+
+If the illuminated surface is completely surrounded by an envelope of
+the same metal insulated from and completely shielded from the light,
+the emission of the negative corpuscles from the illuminated surface
+would go on until the potential difference V between this surface and
+the envelope became so great that the corpuscles with the greatest
+velocity lost their energy before reaching the envelope, i.e. if m is
+the mass, e the charge on a corpuscle, v the greatest velocity of
+projection, until Ve = ½mv². The values found for V by different
+observers are not very consistent. Lenard found that V for aluminium was
+about 3 volts and for platinum 2. Millikan and Winchester (_Phil. Mag._,
+July 1907) found for aluminium V = .738. The apparatus used by them was
+so complex that the interpretation of their results is difficult.
+
+An extremely interesting fact discovered by Lenard is that the velocity
+with which the corpuscles are emitted from the metal is independent of
+the intensity of the incident light. The quantity of corpuscles
+increases with the intensity, but the velocity of the individual
+corpuscles does not. It is worthy of notice that in other cases when
+negative corpuscles are emitted from metals, as for example when the
+metals are exposed to cathode rays, Canal-strahlen, or Röntgen rays, the
+velocity of the emitted corpuscles is independent of the intensity of
+the primary radiation which excites them. The velocity is not, however,
+independent of the nature of the primary rays. Thus when light is used
+to produce the emission of corpuscles the velocity, as Ladenburg has
+shown, depends on the wave length of the light, increasing as the wave
+length diminishes. The velocity of corpuscles emitted under the action
+of cathode rays is greater than that of those ejected by light, while
+the incidence of Röntgen rays produces the emission of corpuscles moving
+much more rapidly than those in the cases already mentioned, and the
+harder the primary rays the greater is the velocity of the corpuscles.
+
+The importance of the fact that the velocity and therefore the energy of
+the corpuscles emitted from the metal is independent of the intensity of
+the incident light can hardly be overestimated. It raises the most
+fundamental questions as to the nature of light and the constitution of
+the molecules. What is the source of the energy possessed by these
+corpuscles? Is it the light, or in the stores of internal energy
+possessed by the molecule? Let us follow the consequences of supposing
+that the energy comes from the light. Then, since the energy is
+independent of the intensity of the light, the electric forces which
+liberate the corpuscles must also be independent of that intensity. But
+this cannot be the case if, as is usually assumed in the electromagnetic
+theory, the wave front consists of a uniform distribution of electric
+force without structure, for in this case the magnitude of the electric
+force is proportional to the square root of the intensity. On the
+emission theory of light a difficulty of this kind would not arise, for
+on that theory the energy in a luminiferous particle remains constant as
+the particle pursues its flight through space. Thus any process which a
+single particle is able to effect by virtue of its energy will be done
+just as well a thousand miles away from the source of light as at the
+source itself, though of course in a given space there will not be
+nearly so many particles to do this process far from the source as there
+are close in. Thus, if one of the particles when it struck against a
+piece of metal caused the ejection of a corpuscle with a given velocity,
+the velocity of emission would not depend on the intensity of the light.
+There does not seem any reason for believing that the electromagnetic
+theory is inconsistent with the idea that on this theory, as on the
+emission theory, the energy in the light wave may instead of being
+uniformly distributed through space be concentrated in bundles which
+occupy only a small fraction of the volume traversed by the light, and
+that as the wave travels out the bundles get farther apart, the energy
+in each remaining undiminished. Some such view of the structure of light
+seems to be required to account for the fact that when a plate of metal
+is struck by a wave of ultra-violet light, it would take years before
+the corpuscles emitted from the metal would equal in number the
+molecules on the surface of the metal plate, and yet on the ordinary
+theory of light each one of these is without interruption exposed to the
+action of the light. The fact discovered by E. Ladenburg (_Verh. d.
+deutsch. physik. Ges._ 9, p. 504) that the velocity with which the
+corpuscles are emitted depends on the wave length of the light suggests
+that the energy in each bundle depends upon the wave length and
+increases as the wave length diminishes.
+
+These considerations illustrate the evidence afforded by photo-electric
+effects on the nature of light; these effects may also have a deep
+significance with regard to the structure of matter. The fact that the
+energy of the individual corpuscles is independent of the intensity of
+the light might be explained by the hypothesis that the energy of the
+corpuscles does not come from the light but from the energy stored up in
+the molecules of the metal exposed to the light. We may suppose that
+under the action of the light some of the molecules are thrown into an
+unstable state and explode, ejecting corpuscles; the light in this case
+acts only as a trigger to liberate the energy in the atom, and it is
+this energy and not that of the light which goes into the corpuscles. In
+this way the velocity of the corpuscles would be independent of the
+intensity of the light. But it may be asked, is this view consistent
+with the result obtained by Ladenburg that the velocity of the
+corpuscles depends upon the nature of the light? If light of a definite
+wave length expelled corpuscles with a definite and uniform velocity, it
+would be very improbable that the emission of the corpuscles is due to
+an explosion of the atoms. The experimental facts as far as they are
+known at present do not allow us to say that the connexion between the
+velocity of the corpuscles and the wave length of the light is of this
+definite character, and a connexion such as a gradual increase of
+average velocity as the wave length of the light diminishes, would be
+quite consistent with the view that the corpuscles are ejected by the
+explosion of the atom. For in a complex thing like an atom there may be
+more than one system which becomes unstable when exposed to light. Let
+us suppose that there are two such systems, A and B, of which B ejects
+the corpuscles with the greater velocity. If B is more sensitive to the
+short waves, and A to the long ones, then as the wave length of the
+light diminishes the proportion of the corpuscles which come from B will
+increase, and as these are the faster, the average velocity of the
+corpuscles emitted will also increase. And although the potential
+acquired by a perfectly insulated piece of metal when exposed to
+ultra-violet light would depend only on the velocity of the fastest
+corpuscles and not upon their number, in practice perfect insulation is
+unattainable, and the potential actually acquired is determined by the
+condition that the gain of negative electricity by the metal through
+lack of insulation, is equal to the loss by the emission of negatively
+electrified corpuscles. The potential acquired will fall below that
+corresponding to perfect insulation by an amount depending on the number
+of the faster corpuscles emitted, and the potential will rise if the
+proportion of the rapidly moving corpuscles is increased, even though
+there is no increase in their velocity. It is interesting to compare
+other cases in which corpuscles are emitted with the case of
+ultra-violet light. When a metal or gas is bombarded by cathode rays it
+emits corpuscles and the velocity of these is found to be independent of
+the velocity of the cathode rays which excite them; the velocity is
+greater than for corpuscles emitted under ultra-violet light. Again,
+when bodies are exposed to Röntgen rays they emit corpuscles moving with
+a much greater velocity than those excited by cathode rays, but again
+the velocity does not depend upon the intensity of the rays although it
+does to some extent on their hardness. In the case of cathode and
+Röntgen rays, the velocity with which the corpuscles are emitted seems,
+as far as we know at present, to vary slightly, but only slightly, with
+the nature of the substance on which the rays fall. May not this
+indicate that the first effect of the primary rays is to detach a
+neutral doublet, consisting of a positive and negative charge, this
+doublet being the same from whatever system it is detached? And that the
+doublet is unstable and explodes, expelling the negative charge with a
+high velocity, and the positive one, having a much larger charge, with a
+much smaller velocity, the momentum of the negative charge being equal
+to that of the positive.
+
+Up to now we have been considering the effects produced when light is
+incident on metals. Lenard found (and the result has been confirmed by
+the experiments of J. J. Thomson and Lyman) that certain kinds of
+ultra-violet light ionize a gas when they pass through. The type of
+ultra-violet light which produces this effect is so easily absorbed that
+it is stopped by a layer a few millimetres thick of air at atmospheric
+pressure.
+
+_Ionization by Collision._--When the ionization of the gas is produced
+by external agents such as Röntgen rays or ultra-violet light, the
+electric field produces a current by setting the positive ions moving in
+one direction, and the negative ones in the opposite; it makes use of
+ions already made and does not itself give rise to ionization. In many
+cases, however, such as in electric sparks, there are no external agents
+to produce ionization and the electric field has to produce the ions as
+well as set them in motion. When the ionization is produced by external
+means the smallest electric field is able to produce a current through
+the gas; when, however, these external means are absent no current is
+produced unless the strength of the electric field exceeds a certain
+critical value, which depends not merely upon the nature of the gas but
+also upon the pressure and the dimensions of the vessel in which it is
+contained. The variation of the electric field required to produce
+discharge can be completely explained if we suppose that the ionization
+of the gas is produced by the impact with its molecules of corpuscles,
+and in certain cases of positive ions, which under the influence of the
+electric field have acquired considerable kinetic energy. We have direct
+evidence that rapidly moving corpuscles are able to ionize molecules
+against which they strike, for the cathode rays consist of such
+corpuscles, and these when they pass through a gas produce large amounts
+of ionization. Suppose then that we have in a gas exposed to an electric
+field a few corpuscles. These will be set in motion by the field and
+will acquire an amount of energy in proportion to the product of the
+electric force, their charge, and the distance travelled in the
+direction of the electric field between two collisions with the
+molecules of the gas. If this energy is sufficient to give them the
+ionizing property possessed by cathode rays, then when a corpuscle
+strikes against a molecule it will detach another corpuscle; this under
+the action of the electric field will acquire enough energy to produce
+corpuscles on its own account, and so as the corpuscles move through the
+gas their number will increase in geometrical progression. Thus, though
+there were but few corpuscles to begin with, there may be great
+ionization after these have been driven some distance through the gas by
+the electric field.
+
+ The number of ions produced by collisions can be calculated by the
+ following method. Let the electric force be parallel to the axis of x,
+ and let n be the number of corpuscles per unit volume at a place fixed
+ by the co-ordinate x; then in unit time these corpuscles will make
+ nu/[lambda] collisions with the molecules, if u is the velocity of a
+ corpuscle and [lambda] the mean free path of a corpuscle. When the
+ corpuscles are moving fast enough to produce ions by collision their
+ velocities are very much greater than those they would possess at the
+ same temperature if they were not acted on by electrical force, and so
+ we may regard the velocities as being parallel to the axis of x and
+ determined by the electric force and the mean free path of the
+ corpuscles. We have to consider how many of the nu/[lambda] collisions
+ which take place per second will produce ions. We should expect that
+ the ionization of a molecule would require a certain amount of energy,
+ so that if the energy of the corpuscle fell below this amount no
+ ionization would take place, while if the energy of the corpuscle were
+ exceedingly large, every collision would result in ionization. We
+ shall suppose that a certain fraction of the number of collisions
+ result in ionization and that this fraction is a function of the
+ energy possessed by the corpuscle when it collides against the
+ molecules. This energy is proportional to Xe[lambda] when X is the
+ electric force, e the charge on the corpuscle, and [lambda] the mean
+ free path. If the fraction of collisions which produce ionization is
+ [int](Xe[lambda]), then the number of ions produced per cubic
+ centimetre per second is [int](Xe[lambda])nu/[lambda]. If the
+ collisions follow each other with great rapidity so that a molecule
+ has not had time to recover from one collision before it is struck
+ again, the effect of collisions might be cumulative, so that a
+ succession of collisions might give rise to ionization, though none of
+ the collisions would produce an ion by itself. In this case [int]
+ would involve the frequency of the collisions as well as the energy of
+ the corpuscle; in other words, it might depend on the current through
+ the gas as well as upon the intensity of the electric field. We
+ shall, however, to begin with, assume that the current is so small
+ that this cumulative effect may be neglected.
+
+ Let us now consider the rate of increase, dn/dt, in the number of
+ corpuscles per unit volume. In consequence of the collisions,
+ [int](Xe[lambda])nu/[lambda] corpuscles are produced per second; in
+ consequence of the motion of the corpuscles, the number which leave
+ unit volume per second is greater than those which enter it by
+ (d/dx)(nu); while in a certain number of collisions a corpuscle will
+ stick to the molecule and will thus cease to be a free corpuscle. Let
+ the fraction of the number of collisions in which this occurs be
+ [beta]. Thus the gain in the number of corpuscles is
+ [int](Xe[lambda])nu/[lambda], while the loss is (d/dx)(nu) +
+ [beta](nu)/[lambda]; hence
+
+ dn nu d [beta]nu
+ -- = [int](Xe[lambda]) -------- - --(nu) - --------.
+ dt [lambda] dx [lambda]
+
+ When things are in a steady state dn/dt = 0, and we have
+
+ d 1 / \
+ --(nu) = --------( [int](Xe[lambda]) - [beta] )nu.
+ dx [lambda] \ /
+
+ If the current is so small that the electrical charges in the gas are
+ not able to produce any appreciable variations in the field, X will be
+ constant and we get nu = C[epsilon]^{[alpha]x}, where [alpha] =
+ {[int](Xe[lambda]) - [beta]}/[lambda]. If we take the origin from
+ which we measure x at the cathode, C is the value of nu at the
+ cathode, i.e. it is the number of corpuscles emitted per unit area of
+ the cathode per unit time; this is equal to i0/e if i0 is the quantity
+ of negative electricity coming from unit area of the cathode per
+ second, and e the electric charge carried by a corpuscle. Hence we
+ have nue = i0[epsilon]^{[alpha]x}. If l is the distance between the
+ anode and the cathode, the value of nue, when x = l, is the current
+ passing through unit area of the gas, if we neglect the electricity
+ carried by negatively electrified carriers other than corpuscles.
+ Hence i = i0[epsilon]^{[alpha]l}. Thus the current between the plates
+ increases in geometrical progression with the distance between the
+ plates.
+
+ By measuring the variation of the current as the distance between the
+ plates is increased, Townsend, to whom we owe much of our knowledge on
+ this subject, determined the values of [alpha] for different values of
+ X and for different pressures for air, hydrogen and carbonic acid gas
+ (_Phil. Mag._ [6], 1, p. 198). Since [lambda] varies inversely as the
+ pressure, we see that [alpha] may be written in the form p[phi](X/p)
+ or [alpha]/X = F(X/p). The following are some of the values of [alpha]
+ found by Townsend for air.
+
+ +---------+----------+----------+----------+----------+----------+
+ | X Volts | Pressure | Pressure | Pressure | Pressure | Pressure |
+ | per cm. | .17 mm. | .38 mm. | 1.10 mm. | 2.1 mm. | 4.1 mm. |
+ +---------+----------+----------+----------+----------+----------+
+ | 20 | .24 | | | | |
+ | 40 | .65 | .34 | | | |
+ | 80 | 1.35 | 1.3 | .45 | .13 | |
+ | 120 | 1.8 | 2.0 | 1.1 | .42 | .13 |
+ | 160 | 2.1 | 2.8 | 2.0 | .9 | .28 |
+ | 200 | | 3.4 | 2.8 | 1.6 | .5 |
+ | 240 | 2.45 | 3.8 | 4.0 | 2.35 | .99 |
+ | 320 | 2.7 | 4.5 | 5.5 | 4.0 | 2.1 |
+ | 400 | | 5.0 | 6.8 | 6.0 | 3.6 |
+ | 480 | 3.15 | 5.4 | 8.0 | 7.8 | 5.3 |
+ | 560 | | 5.8 | 9.3 | 9.4 | 7.1 |
+ | 640 | 3.25 | 6.2 | 10.6 | 10.8 | 8.9 |
+ +---------+----------+----------+----------+----------+----------+
+
+ We see from this table that for a given value of X, [alpha] for small
+ pressures increases as the pressure increases; it attains a maximum at
+ a particular pressure, and then diminishes as the pressure increases.
+ The increase in the pressure increases the number of collisions, but
+ diminishes the energy acquired by the corpuscle in the electric field,
+ and thus diminishes the change of any one collision resulting in
+ ionization. If we suppose the field is so strong that at some
+ particular pressure the energy acquired by the corpuscle is well above
+ the value required to ionize at each collision, then it is evident
+ that increasing the number of collisions will increase the amount of
+ ionization, and therefore [alpha], and [alpha] cannot begin to
+ diminish until the pressure has increased to such an extent that the
+ mean free path of a corpuscle is so small that the energy acquired by
+ the corpuscle from the electric field falls below the value when each
+ collision results in ionization.
+
+ The value of p, when X is given, for which [alpha] is a maximum, is
+ proportional to X; this follows at once from the fact that [alpha] is
+ of the form X·F(X/p). The value of X/p for which F(X/p) is a maximum is
+ seen from the preceding table to be about 420, when X is expressed in
+ volts per centimetre and p in millimetres of mercury. The maximum value
+ of F(X/p) is about 1/60. Since the current passing between two planes
+ at a distance l apart is i0[epsilon]^{[alpha]l} or
+ i0[epsilon]^{XlF(X/p)}, and since the force between the plates is
+ supposed to be uniform, Xl is equal to V, the potential between the
+ plates; hence the current between the plates is i0[epsilon]^{VlF(X/p)},
+ and the greatest value it can have is i0[epsilon]^{V/60}. Thus the
+ ratio between the current between the plates when there is ionization
+ and when there is none cannot be greater than [epsilon]^{V/60}, when V
+ is measured in volts. This result is based on Townsend's experiments
+ with very weak currents; we must remember, however, that when the
+ collisions are so frequent that the effects of collisions can
+ accumulate, [alpha] may have much larger values than when the current
+ is small. In some experiments made by J. J. Thomson with intense
+ currents from cathodes covered with hot lime, the increase in the
+ current when the potential difference was 60 volts, instead of being e
+ times the current when there was no ionization, as the preceding theory
+ indicates, was several hundred times that value, thus indicating a
+ great increase in [alpha] with the strength of the current.
+
+ Townsend has shown that we can deduce from the values of [alpha] the
+ mean free path of a corpuscle. For if the ionization is due to the
+ collisions with the corpuscles, then unless one collision detaches
+ more than one corpuscle the maximum number of corpuscles produced will
+ be equal to the number of collisions. When each collision results in
+ the production of a corpuscle, [alpha] = 1/[lambda] and is independent
+ of the strength of the electric field. Hence we see that the value of
+ [alpha], when it is independent of the electric field, is equal to the
+ reciprocal of the free path. Thus from the table we infer that at a
+ pressure of 17 mm. the mean free path is 1/325 cm.; hence at 1 mm. the
+ mean free path of a corpuscle is 1/19 cm. Townsend has shown that this
+ value of the mean free path agrees well with the value 1/21 cm.
+ deduced from the kinetic theory of gases for a corpuscle moving
+ through air. By measuring the values of [alpha] for hydrogen and
+ carbonic acid gas Townsend and Kirby (_Phil. Mag._ [6], 1, p. 630)
+ showed that the mean free paths for corpuscles in these gases are
+ respectively 1/11.5 and 1/29 cm. at a pressure of 1 mm. These results
+ again agree well with the values given by the kinetic theory of gases.
+
+ If the number of positive ions per unit volume is m and v is the
+ velocity, we have nue+mve = i, where i is the current through unit
+ area of the gas. Since nue = i0[epsilon]^nx and i = i0[epsilon]^nl,
+ when l is the distance between the plates, we see that
+
+ nu / mv = [epsilon]^(nx) / ([epsilon]^(nl) - [epsilon]^(nx)),
+
+ n v [epsilon]^(nx)
+ -- = -- · -------------------------------.
+ m u [epsilon]^(ne) - [epsilon]^(nx)
+
+ Since v/u is a very small quantity we see that n will be less than m
+ except when [epsilon]^nl - [epsilon]^nx is small, i.e. except close to
+ the anode. Thus there will be an excess of positive electricity from
+ the cathode almost up to the anode, while close to the anode there
+ will be an excess of negative. This distribution of electricity will
+ make the electric force diminish from the cathode to the place where
+ there is as much positive as negative electricity, where it will have
+ its minimum value, and then increase up to the anode.
+
+ The expression i = i0[epsilon]^[alpha]l applies to the case when there
+ is no source of ionization in the gas other than the collisions; if in
+ addition to this there is a source of uniform ionization producing q
+ ions per cubic centimetre, we can easily show that
+
+ qe
+ i = i0[epsilon]^{[alpha]l} + -------(e^{[alpha]l} - 1).
+ [alpha]
+
+ With regard to the minimum energy which must be possessed by a
+ corpuscle to enable it to produce ions by collision, Townsend (loc.
+ cit.) came to the conclusion that to ionize air the corpuscle must
+ possess an amount of energy equal to that acquired by the fall of its
+ charge through a potential difference of about 2 volts. This is also
+ the value arrived at by H. A. Wilson by entirely different
+ considerations. Stark, however, gives 17 volts as the minimum for
+ ionization. The energy depends upon the nature of the gas; recent
+ experiments by Dawes and Gill and Pedduck (_Phil. Mag._, Aug. 1908)
+ have shown that it is smaller for helium than for air, hydrogen, or
+ carbonic acid gas.
+
+If there is no external source of ionization and no emission of
+corpuscles from the cathode, then it is evident that even if some
+corpuscles happened to be present in the gas when the electric field
+were applied, we could not get a permanent current by the aid of
+collisions made by these corpuscles. For under the electric field, the
+corpuscles would be driven from the cathode to the anode, and in a short
+time all the corpuscles originally present in the gas and those produced
+by them would be driven from the gas against the anode, and if there was
+no source from which fresh corpuscles could be introduced into the gas
+the current would cease. The current, however, could be maintained
+indefinitely if the positive ions in their journey back to the cathode
+also produced ions by collisions, for then we should have a kind of
+regenerative process by which the supply of corpuscles could be
+continually renewed. To maintain the current it is not necessary that
+the ionization resulting from the positive ions should be anything like
+as great as that from the negative, as the investigation given below
+shows a very small amount of ionization by the positive ions will
+suffice to maintain the current. The existence of ionization by
+collision with positive ions has been proved by Townsend. Another method
+by which the current could be and is maintained is by the anode emitting
+corpuscles under the impact of the positive ions driven against it by
+the electric field. J. J. Thomson has shown by direct experiment that
+positively electrified particles when they strike against a metal plate
+cause the metal to emit corpuscles (J. J. Thomson, _Proc. Camb. Phil.
+Soc._ 13, p. 212; Austin, _Phys. Rev._ 22, p. 312). If we assume that
+the number of corpuscles emitted by the plate in one second is
+proportional to the energy in the positive ions which strike the plate
+in that second, we can readily find an expression for the difference of
+potential which will maintain without any external ionization a current
+of electricity through the gas. As this investigation brings into
+prominence many of the most important features of the electric
+discharge, we shall consider it in some detail.
+
+ Let us suppose that the electrodes are parallel plates of metal at
+ right angles to the axis of x, and that at the cathode x = 0 and at the
+ anode x = d, d being thus the distance between the plates. Let us also
+ suppose that the current of electricity flowing between the plates is
+ so small that the electrification between the plates due to the
+ accumulation of ions is not sufficient to disturb appreciably the
+ electric field, which we regard as uniform between the plates, the
+ electric force being equal to V/d, where V is the potential difference
+ between the plates. The number of positive ions produced per second in
+ a layer of gas between the planes x and x+dx is [alpha]nu·dx. Here n is
+ the number of corpuscles per unit volume, [alpha] the coefficient of
+ ionization (for strong electric field [alpha] = 1/[lambda]', where
+ [lambda]' is the mean free path of a corpuscle), and u the velocity of
+ a corpuscle parallel to x. We have seen that nu = i0[epsilon]^[alpha]x,
+ where i0 is the number of corpuscles emitted per second by unit area of
+ the cathode. Thus the number of positive ions produced in the layer is
+ [alpha]i0[epsilon]^[alpha]x dx. If these went straight to the cathode
+ without a collision, each of them would have received an amount of
+ kinetic energy Vex/d when they struck the cathode, and the energy of
+ the group of ions would be Vex/d·[alpha]i0[epsilon]^dx dx. The positive
+ ions will, however, collide with the molecules of the gas through which
+ they are passing, and this will diminish the energy they possess when
+ they reach the cathode.
+
+ The diminution in the energy will increase in geometrical proportion
+ with the length of path travelled by the ion and will thus be
+ proportional to [epsilon]^-[beta]x, [beta] will be proportional to the
+ number of collisions and will thus be proportional to the pressure of
+ the gas. Thus the kinetic energy possessed by the ions when they reach
+ the cathode will be
+
+ [epsilon]^{-[beta]x} · V(ex/d) · [alpha]i0[epsilon]^{[alpha]x} dx,
+
+ and E, the total amount of energy in the positive ions which reach the
+ cathode in unit time, will be given by the equation
+ _
+ /d
+ E = | [epsilon]^{-[beta]x} · V(ex/d) · [alpha]i0[epsilon]^{[alpha]x} dx
+ _/0
+ _
+ Ve[alpha]i0 /d
+ = ----------- | [epsilon]^{-([beta]-[alpha])x}·x·dx
+ d _/0
+
+ Ve[alpha]i0 / 1 / 1 d \ \
+ = ----------- { ---------------- - [epsilon]^{-([beta]-[alpha])d} { ----------------- + ---------------- } } (1).
+ d \([beta]-[alpha])² \([beta]-[alpha])² ([beta]-[alpha])/ /
+
+
+ If the number of corpuscles emitted by the cathode in unit time is
+ proportional to this energy we have i0 = kE, where k is a constant;
+ hence by equation (1) we have
+
+ ([beta]-[alpha])² d
+ V = ----------------- · --,
+ ke[alpha] I
+
+ where
+
+ I = 1 - [epsilon]^{-([beta]-[alpha])d} (1 + d([beta] - [alpha])).
+
+ Since both [beta] and [alpha] are proportional to the pressure, I and
+ ([beta] - [alpha])²d/[alpha] are both functions of pd, the product of
+ the pressure and the spark length, hence we see that V is expressed by
+ an equation of the form
+
+ 1
+ V = -- [int](pd) (2),
+ ke
+
+ where [int](pd) denotes a function of pd, and neither p nor d enter
+ into the expression for V except in this product. Thus the potential
+ difference required to produce discharge is constant as long as the
+ product of the pressure and spark length remains constant; in other
+ words, the spark potential is constant as long as the mass of the gas
+ between the electrodes is constant. Thus, for example, if we halve the
+ pressure the same potential difference will produce a spark of twice
+ the length. This law, which was discovered by Paschen for fairly long
+ sparks (_Annalen_, 37, p. 79), and has been shown by Carr (_Phil.
+ Trans._, 1903) to hold for short ones, is one of the most important
+ properties of the electric discharge.
+
+ We see from the expression for V that when ([beta] - [alpha])d is very
+ large
+
+ V = ([beta] - [alpha])²d/ke[alpha].
+
+ Thus V becomes infinite when d is infinite. Again when ([beta] -
+ [alpha])d is very small we find
+
+ V = 1/ke[alpha]d;
+
+ thus V is again infinite when d is nothing. There must therefore be
+ some value of d intermediate between zero and infinity for which V is
+ a minimum. This value is got by finding in the usual way the value of
+ d, which makes the expression for V given in equation (1) a minimum.
+ We find that d must satisfy the equation
+
+ / \
+ 1 = [epsilon]^{-([beta]-[alpha])d} {1 + ([beta] - [alpha])d + ([beta] - [alpha]·d)²}.
+ \ /
+
+ We find by a process of trial and error that ([beta]-[alpha])d = 1.8
+ is approximately a solution of this equation; hence the distance for
+ minimum potential is 1.8/([beta] - [alpha]). Since [beta] and [alpha]
+ are both proportional to the pressure, we see that the critical spark
+ length varies inversely as the pressure. If we substitute this value
+ in the expression for V we find that [=V], the minimum spark
+ potential, is given by
+
+ _ [beta] - [alpha] 2.2
+ V = ---------------- · ---.
+ [alpha] ke
+
+ Since [beta] and [alpha] are each proportional to the pressure, the
+ minimum potential is independent of the pressure of the gas. On this
+ view the minimum potential depends upon the metal of which the cathode
+ is made, since k measures the number of corpuscles emitted per unit
+ time by the cathode when struck by positive ions carrying unit energy,
+ and unless [beta] bears the same ratio to [alpha] for all gases the
+ minimum potential will also vary with the gas. The measurements which
+ have been made of the "cathode fall of potential," which as we shall
+ see is equal to the minimum potential required to produce a spark,
+ show that this quantity varies with the material of which the cathode
+ is made and also with the nature of the gas. Since a metal plate, when
+ bombarded by positive ions, emits corpuscles, the effect we have been
+ considering must play a part in the discharge; it is not, however, the
+ only effect which has to be considered, for as Townsend has shown,
+ positive ions when moving above a certain speed ionize the gas, and
+ cause it to emit corpuscles. It is thus necessary to take into account
+ the ionization of the positive ions.
+
+ Let m be the number of positive ions per unit volume, and w their
+ velocity, the number of collisions which occur in one second in one
+ cubic centimetre of the gas will be proportional to mwp, where p is
+ the pressure of the gas. Let the number of ions which result from
+ these collisions be [gamma]mw; [gamma] will be a function of p and of
+ the strength of the electric field. Let as before n be the number of
+ corpuscles per cubic centimetre, u their velocity, and [alpha]nu the
+ number of ions which result in one second from the collisions between
+ the corpuscles and the gas. The number of ions produced per second per
+ cubic centimetre is equal to [alpha]nu + [gamma]mw; hence when things
+ are in a steady state
+
+ d
+ --(nu) = [alpha]nu + [gamma]mw ,
+ dx
+
+ and
+
+ e(nu + mw) = i,
+
+ where e is the charge on the ion and i the current through the gas.
+ The solution of these equations when the field is uniform between the
+ plates, is
+
+ enu = C[epsilon]^{([alpha]-[gamma])x} - [gamma]i/([alpha] - [gamma]),
+
+ emw = -C[epsilon]^{([alpha]-[gamma])x} + [alpha]i/([alpha] - [gamma]),
+
+ where C is a constant of integration. If there is no emission of
+ positive ions from the anode enu = i, when x = d. Determining C from
+ this condition we find
+
+ i / \
+ enu = ----------------- {[alpha][epsilon]^{([alpha]-[gamma])(x-d)} - [gamma] },
+ [alpha] - [gamma] \ /
+
+ [alpha]i / \
+ emw = ----------------- {1 - [epsilon]^{([alpha]-[gamma])(x-d)} }.
+ [alpha] - [gamma] \ /
+
+ If the cathode did not emit any corpuscles owing to the bombardment by
+ positive ions, the condition that the charge should be maintained is
+ that there should be enough positive ions at the cathode to carry the
+ current i.e. that emw = i; when x = 0, the condition gives
+
+ i / \
+ ----------------- {[alpha][epsilon]^{-([alpha]-[gamma])d} - [gamma] } = 0,
+ [alpha] - [gamma] \ /
+
+ or
+
+ [epsilon]^{[alpha]d}/[alpha] = [epsilon]^{[gamma]d}/[gamma].
+
+ Since [alpha] and [gamma] are both of the form pf(X/p) and X = V/d, we
+ see that V will be a function of pd, in agreement with Paschen's law.
+ If we take into account both the ionization of the gas and the
+ emission of corpuscles by the metal we can easily show that
+
+ _
+ [alpha]-[gamma][epsilon]^{([alpha]-[gamma])d} k[alpha]Ve | 1
+ --------------------------------------------- = ---------- | ------------------------- -
+ [alpha] - [gamma] d |_ ([beta]+[gamma]-[alpha])²
+ _
+ / 1 d \ |
+ [epsilon]^{-([beta]+[gamma]-[alpha])d} { ------------------------ + ---------------------- } |,
+ \([beta]+[gamma]-[alpha])² [beta]+[gamma]-[alpha]/ _|
+
+ where k and [beta] have the same meaning as in the previous
+ investigation. When d is large, [epsilon]^{([alpha]-[gamma])d} is also
+ large; hence in order that the left-hand side of this equation should
+ not be negative [gamma] must be less than [alpha]/[epsilon]^
+ {([alpha]-[gamma])d}; as this diminishes as d increases we see that when
+ the sparks are very long discharge will take place, practically as soon
+ as [gamma] has a finite value, i.e. as soon as the positive ions begin
+ to produce fresh ions by their collisions.
+
+In the preceding investigation we have supposed that the electric field
+between the plates was uniform; if it were not uniform we could get
+discharges produced by very much smaller differences of potential than
+are necessary in a uniform field. For to maintain the discharge it is
+not necessary that the positive ions should act as ionizers all along
+their path; it is sufficient that they should do so in the neighbourhood
+of cathode. Thus if we have a strong field close to the cathode we might
+still get the discharge though the rest of the field were comparatively
+weak. Such a distribution of electric force requires, however, a great
+accumulation of charged ions near the cathode; until these ions
+accumulate the field will be uniform. If the uniform field existing in
+the gas before the discharge begins were strong enough to make the
+corpuscles produce ions by collision, but not strong enough to make the
+positive ions act as ionizers, there would be some accumulation of ions,
+and the amount of this accumulation would depend upon the number of free
+corpuscles originally present in the gas, and upon the strength of the
+electric field. If the accumulation were sufficient to make the field
+near the cathode so strong that the positive ions could produce fresh
+ions either by collision with the cathode or with the gas, the discharge
+would pass though the gas; if not, there will be no continuous
+discharge. As the amount of the accumulation depends on the number of
+corpuscles present in the gas, we can understand how it is that after a
+spark has passed, leaving for a time a supply of corpuscles behind it,
+it is easier to get a discharge to pass through the gas than it was
+before.
+
+[Illustration: Fig. 15.]
+
+The inequality of the electric field in the gas when a continuous
+discharge is passing through it is very obvious when the pressure of the
+gas is low. In this case the discharge presents a highly differentiated
+appearance of which a type is represented in fig. 15. Starting from the
+cathode we have a thin velvety luminous glow in contact with the
+surface; this glow is often called the "first cathode layer." Next this
+we have a comparatively dark space whose thickness increases as the
+pressure diminishes; this is called the "Crookes's dark space," or the
+"second cathode layer." Next this we have a luminous position called the
+"negative glow" or the "third cathode layer." The boundary between the
+second and third layers is often very sharply defined. Next to the third
+layer we have another dark space called the "Faraday dark space." Next
+to this and reaching up to the anode is another region of luminosity,
+called the "positive column," sometimes (as in fig. 15, a) continuous,
+sometimes (as in fig. 15, b) broken up into light or dark patches called
+"striations." The dimensions of the Faraday dark space and the positive
+column vary greatly with the current passing through the gas and with
+its pressure; sometimes one or other of them is absent. These
+differences in appearances are accompanied by great difference in the
+strength of the electric field. The magnitude of the electric force at
+different parts of the discharge is represented in fig. 16, where the
+ordinates represent the electric force at different parts of the tube,
+the cathode being on the right. We see that the electric force is very
+large indeed between the negative glow and the cathode, much larger than
+in any other part of the tube. It is not constant in this region, but
+increases as we approach the cathode. The force reaches a minimum either
+in the negative glow itself or in the part of the Faraday dark space
+just outside, after which it increases towards the positive column. In
+the case of a uniform positive column the electric force along it is
+constant until we get quite close to the anode, when a sudden change,
+called the "anode fall," takes place in the potential.
+
+[Illustration:
+ _Discharge in Hydrogen
+ Pressure 2.25 m.m. Current 0.568·10^-3 ampere_
+FIG. 16.]
+
+The difference of potential between the cathode and the negative glow is
+called the "cathode potential fall" and is found to be constant for wide
+variations in the pressure of the gas and the current passing through.
+It increases, however, considerably when the current through the gas
+exceeds a certain critical value, depending among other things on the
+size of the cathode. This cathode fall of potential is shown by
+experiment to be very approximately equal to the minimum potential
+difference. The following table contains a comparison of the
+measurements of the cathode fall of potentials in various gases made by
+Warburg (_Wied. Ann._, 1887, 31, p. 545, and 1890, 40, p. 1), Capstick
+(_Proc. Roy. Society_, 1898, 63, p. 356), and Strutt (_Phil. Trans._,
+1900, 193, p. 377), and the measurements by Strutt of the smallest
+difference of potential which will maintain a spark through these gases.
+
+ +---------+-----------------------------------------+-----------------+
+ | | Cathode fall in Volts. |Least potential |
+ | Gas. +-----------------------------+-----------+ difference |
+ | | Platinum Electrodes. |Aluminium | required to |
+ | | |Electrodes.|maintain a Spark.|
+ +---------+-----------+---------+-------+-----------+-----------------+
+ | | Warburg. |Capstick.|Strutt.| Warburg. | Strutt. |
+ +---------+-----------+---------+-------+-----------+-----------------+
+ |Air | 340-350 | .. | .. | .. | 341 |
+ |H2 | about 300 | 298 | .. | 168 | 302-308 |
+ |O2 | .. | 369 | .. | .. | .. |
+ |N2 |230 if free| 232 | .. | 207 | 251 |
+ | |from oxygen| | | | |
+ |Hg vapour| 340 | .. | .. | .. | .. |
+ |Helium | .. | .. | 226 | .. | 261-326 |
+ |H2O | .. | 469 | .. | .. | .. |
+ |NH3 | .. | 582 | .. | .. | .. |
+ +---------+-----------+---------+-------+-----------+-----------------+
+
+Thus in the cases in which the measurements could be made with the
+greatest accuracy the agreement between the cathode fall and the minimum
+potential difference is very close. The cathode fall depends on the
+material of which the terminals are made, as is shown by the following
+table due to Mey (_Verh. deutsch. physik. Gesell._, 1903, 5, p. 72).
+
+ +------+---------------------------------------------+
+ | Gas. | Electrode. |
+ +------+---+---+---+---+---+---+---+---+---+-----+---+
+ | | Pt| Hg| Ag| Cu| Fe| Zn| Al| Mg| Na| Na-K| K |
+ +------+---+---+---+---+---+---+---+---+---+-----+---+
+ |O2 |369| ..| ..| ..| ..| ..| ..| ..| ..| .. | ..|
+ |H2 |300| ..|295|280|230|213|190|168|185|169 |172|
+ |N2 |232|226| ..| ..| ..| ..| ..|207|178|125 |170|
+ |He |226| ..| ..| ..| ..| ..| ..| ..| 80| 78.5| 69|
+ |Argon |167| ..| ..| ..| ..| ..|100| ..| ..| .. | ..|
+ +------+---+---+---+---+---+---+---+---+---+-----+---+
+
+The dependence of the minimum potential required to produce a spark upon
+the metal of which the cathode is made has not been clearly established,
+some observers being unable to detect any difference between the
+potential required to spark between electrodes of aluminium and those of
+brass, while others thought they had detected such a difference. It is
+only with sparks not much longer than the critical spark length that we
+could hope to detect this difference. When the current through the gas
+exceeds a certain critical value depending among other things on the
+size of the cathode, the cathode fall of potential increases rapidly and
+at the same time the thickness of the dark spaces diminishes. We may
+regard the part of the discharge between the cathode and the negative
+glow as a discharge taking place under minimum potential difference
+through a distance equal to the critical spark length. An inspection of
+fig. 16 will show that we cannot regard the electric field as constant
+even for this small distance; it thus becomes a matter of interest to
+know what would be the effect on the minimum potential difference
+required to produce a spark if there were sufficient ions present to
+produce variations in the electric field analogous to those represented
+in fig. 16. If the electric force at a distance x from the cathode were
+proportional to [epsilon]^-px we should have a state of things much
+resembling the distribution of electric force near the cathode. If we
+apply to this distribution the methods used above for the case when the
+force was uniform, we shall find that the minimum potential is less and
+the critical spark length greater than when the electric force is
+uniform.
+
+_Potential Difference required to produce a Spark of given Length._--We
+may regard the region between the cathode and the negative glow as a
+place for the production of corpuscles, these corpuscles finding their
+way from this region through the negative glow. The parts of this glow
+towards the anode we may regard as a cathode, from which, as from a hot
+lime cathode, corpuscles are emitted. Let us now consider what will
+happen to these corpuscles shot out from the negative glow with a
+velocity depending on the cathode fall of potential and independent of
+the pressure. These corpuscles will collide with the molecules of the
+gas, and unless there is an external electric field to maintain their
+velocity they will soon come to rest and accumulate in front of the
+negative glow. The electric force exerted by this cloud of corpuscles
+will diminish the strength of the electric field in the region between
+the cathode and the negative glow, and thus tend to stop the discharge.
+To keep up the discharge we must have a sufficiently strong electric
+field between the negative glow and the anode to remove the corpuscles
+from this region as fast as they are sent into it from the cathode. If,
+however, there is no production of ions in the region between the
+negative glow and the anode, all the ions in this region will have come
+from near the cathode and will be negatively charged; this negative
+electrification will diminish the electric force on the cathode side of
+it and thus tend to stop the discharge. This back electric field could,
+however, be prevented by a little ionization in the region between the
+anode and glow, for this would afford a supply of positive ions, and
+thus afford an opportunity for the gas in this region to have in it as
+many positive as negative ions; in this case it would not give rise to
+any back electromotive force. The ionization which produces these
+positive ions may, if the field is intense, be due to the collisions of
+corpuscles, or it may be due to radiation analogous to ultra-violet, or
+soft Röntgen rays, which have been shown by experiment to accompany the
+discharge. Thus in the most simple conditions for discharge we should
+have sufficient ionization to keep up the supply of positive ions, and
+an electric field strong enough to keep the velocity of the negative
+corpuscle equal to the value it has when it emerges from the negative
+glow. Thus the force must be such as to give a constant velocity to the
+corpuscle, and since the force required to move an ion with a given
+velocity is proportional to the pressure, this force will be
+proportional to the pressure of the gas. Let us call this force ap; then
+if l is the distance of the anode from the negative glow the potential
+difference between these points will be alp. The potential difference
+between the negative glow and the cathode is constant and equals c;
+hence if V is the potential difference between the anode and cathode,
+then V = c + alp, a relation which expresses the connexion between the
+potential difference and spark length for spark lengths greater than the
+critical distance. It is to be remembered that the result we have
+obtained applies only to such a case as that indicated above, where the
+electric force is constant along the positive column. Experiments with
+the discharge through gases at low pressure show the discharge may take
+other forms. Thus the positive column may be striated when the force
+along it is no longer uniform, or the positive column may be absent;
+the discharge may be changed from one of these forms to another by
+altering the current. The relation between the potential and the
+distance between the electrodes varies greatly, as we might expect, with
+the current passing through the gas.
+
+The connexion between the potential difference and the spark length has
+been made the subject of a large number of experiments. The first
+measurements were made by Lord Kelvin in 1860 (_Collected Papers on
+Electrostatics and Magnetism_, p. 247); subsequent experiments have been
+made by Baille (_Ann. de chimie et de physique_, 5, 25, p. 486), Liebig
+(_Phil. Mag._ [5], 24, p. 106), Paschen (_Wied. Ann._ 37, p. 79), Peace
+(_Proc. Roy. Soc._, 1892, 52, p. 99), Orgler (_Ann. der Phys._ 1, p.
+159), Strutt (_Phil. Trans._ 193, p. 377), Bouty (_Comptes rendus_, 131,
+pp. 469, 503), Earhart (_Phil. Mag._ [6], 1, p. 147), Carr (_Phil.
+Trans._, 1903), Russell (_Phil. Mag._ [5], 64, p. 237), Hobbs (_Phil.
+Mag._ [6], 10, p. 617), Kinsley (_Phil. Mag._ [6], 9, 692), Ritter
+(_Ann. der Phys._ 14, p. 118). The results of their experiments show
+that for sparks considerably longer than the critical spark length, the
+relation between the potential difference V and the spark length l may
+be expressed when the electrodes are large with great accuracy by the
+linear relation V = c + blp, where p is the pressure and c and b are
+constants depending on the nature of the gas. When the sparks are long
+the term blp is the most important and the sparking potential is
+proportional to the spark length. Though there are considerable
+discrepancies between the results obtained by different observers, these
+indicate that the production of a long spark between large electrodes in
+air at atmospheric pressure requires a potential difference of 30,000
+volts for each centimetre of spark length. In hydrogen only about half
+this potential difference is required, in carbonic acid gas the
+potential difference is about the same as in air, while Ritter's
+experiments show that in helium only about one-tenth of this potential
+difference is required.
+
+In the case when the electric field is not uniform, as for example when
+the discharge takes place between spherical electrodes, Russell's
+experiments show that the discharge takes place as soon as the maximum
+electric force in the field between the electrodes reaches a definite
+value, which he found was for air at atmospheric pressure about 38,000
+volts per centimetre.
+
+_Very Short Sparks._--Some very interesting experiments on the potential
+difference required to produce exceedingly short sparks have been made
+by Earhart, Hobbs and Kinsley; the length of these sparks was comparable
+with the wave length of sodium light. With sparks of these lengths it
+was found that it was possible to get a discharge with less than 330
+volts, the minimum potential difference in air. The results of these
+observers show that there is no diminution in the minimum potential
+difference required to produce discharge until the spark length gets so
+small that the average electric force between the electrodes amounts to
+about one million volts per centimetre. When the force rises to this
+value a discharge takes place even though the potential difference is
+much less than 330 volts; in some of Earhart's experiments it was only
+about 2 volts. This kind of discharge is determined not by the condition
+that the potential difference should have a given value, but that the
+electric force should have a given value. Another point in which this
+discharge differs from the ordinary one is that it is influenced
+entirely by the nature of the electrodes and not by the nature or
+pressure of the gas between them, whereas the ordinary discharge is in
+many cases not affected appreciably by changes in the metal of the
+electrodes, but is always affected by changes in the pressure and
+character of the gas between them. Kinsley found that when one of these
+small sparks passed between the electrodes a kind of metallic bridge was
+formed between them, so that they were in metallic connexion, and that
+the distance between them had to be considerably increased before the
+bridge was broken. Almy (_Phil. Mag._, Sept. 1908), who used very small
+electrodes, was unable to get a discharge with less than the minimum
+spark potential even when the spark length was reduced to one-third of
+the wave length of sodium light. He suggests that the discharges
+obtained with larger electrodes for smaller voltages are due to the
+electrodes being dragged together by the electrostatic attraction
+between them.
+
+_Constitution of the Electric Spark._--Schuster and Hemsalech (_Phil.
+Trans._ 193, p. 189), Hemsalech (_Comptes Rendus_, 130, p. 898; 132, p.
+917; _Jour. de Phys._ 3. 9, p. 43, and Schenck, _Astrophy. Jour._ 14, p.
+116) have by spectroscopic methods obtained very interesting results
+about the constitution of the spark. The method employed by Schuster and
+Hemsalech was as follows: Suppose we photograph the spectrum of a
+horizontal spark on a film which is on the rim of a wheel rotating about
+a horizontal axis with great velocity. If the luminosity travelled with
+infinite speed from one electrode to the other, the image on the film
+would be a horizontal line. If, however, the speed with which the
+luminosity travelled between the electrodes was comparable with the
+speed of the film, the line would be inclined to the horizontal, and by
+measuring the inclinations we could find the speed at which the
+luminosity travelled. In this way Schuster and Hemsalech showed that
+when an oscillating discharge passed between metallic terminals in air,
+the first spark passes through the air alone, no lines of the metal
+appearing in its spectrum. This first spark vaporizes some of the metal
+and the subsequent sparks passing mainly through the metallic vapour;
+the appearance of the lines in the film shows that the velocity of the
+luminous part of the vapour was finite. The velocity of the vapour of
+metals of low atomic weight was in general greater than that of the
+vapour of heavier metals. Thus the velocity of aluminium vapour was 1890
+metres per second, that of zinc and cadmium only about 545. Perhaps the
+most interesting point in the investigation was the discovery that the
+velocities corresponding to different lines in the spectrum of the same
+metal were in some cases different. Thus with bismuth some of the lines
+indicated a velocity of 1420 metres per second, others a velocity of
+only 550, while one ([lambda] = 3793) showed a still smaller velocity.
+These results are in accordance with a view suggested by other phenomena
+that many of the lines in a spectrum produced by an electrical discharge
+originate from systems formed during the discharge and not from the
+normal atom or molecule. Schuster and Hemsalech found that by inserting
+a coil with large self induction in the primary circuit they could
+obliterate the air lines in the discharge.
+
+Schenck, by observing the appearance presented when an alternating
+current, produced by discharging Leyden jars, was examined in a rapidly
+rotating mirror, found it showed the following stages: (1) a thin bright
+line, followed in some cases at intervals of half the period of the
+discharge by fainter lines; (2) bright curved streamers starting from
+the negative terminal, and diminishing rapidly in speed as they receded
+from the cathode; (3) a diffused glow lasting for a much longer period
+than either of the preceding. These constituents gave out quite
+different spectra.
+
+The structure of the discharge is much more easily studied when the
+pressure of the gas is low, as the various parts which make up the
+discharge are more widely separated from each other. We have already
+described the general appearance of the discharge through gases at low
+pressures (see p. 657). There is, however, one form of discharge which
+is so striking and beautiful that it deserves more detailed
+consideration. In this type of discharge, known as the striated
+discharge, the positive column is made up of alternate bright and dark
+patches known as _striations_. Some of these are represented in fig. 17,
+which is taken from a paper by De la Rue and Müller (_Phil. Trans._,
+1878, Pt. 1). This type of discharge only occurs when the current and
+the pressure of the gas are between certain limits. It is most
+beautifully shown when a Wehnelt cathode is used and the current is
+produced by storage cells, as this allows us to use large currents and
+to maintain a steady potential difference between the electrodes. The
+striations are in consequence very bright and steady. The facts which
+have been established about these striations are as follows: The
+distance between the bright parts of the striations is greater at low
+pressures than at high; it depends also upon the diameter of the tube,
+increasing as the diameter of the tube increases. If the discharge tube
+is wide at one place and narrow in another the striations will be
+closer together in the narrow parts than in the wide. The distance
+between the striations depends on the current through the tube. The
+relation is not a very simple one, as an increase of current sometimes
+increases while under other circumstances it decreases the distance
+between the striations (see Willows, _Proc. Camb. Phil. Soc._ 10, p.
+302). The electric force is not uniform along the striated discharge,
+but is greater in the bright than in the dark parts of the striation. An
+example is shown in fig. 16, due to H. A. Wilson, which shows the
+distribution of electric force at every place in a striated discharge.
+In experiments made by J. J. Thomson (_Phil. Mag._, Oct. 1909), using a
+Wehnelt cathode, the variations in the electric force were more
+pronounced than those shown in fig. 16. The electric force in this case
+changed so greatly that it actually became negative just on the cathode
+side of the bright part of the striation. Just inside the striation on
+the anode side it rose to a very high value, then continually diminished
+towards the bright side of the next striation when it again increased.
+This distribution of electric force implies that there is great excess
+of negative electricity at the bright head of the striation, and a small
+excess of positive everywhere else. The temperature of the gas is higher
+in the bright than in the dark parts of the striations. Wood (_Wied.
+Ann._ 49, p. 238), who has made a very careful study of the distribution
+of temperature in a discharge tube, finds that in those tubes the
+temperature varies in the same way as the electric force, but that this
+temperature (which it must be remembered is the average temperature of
+all the molecules and not merely of those which are taking part in the
+discharge) is by no means high; in no part of the discharge did the
+temperature in his experiments exceed 100° C.
+
+[Illustration: FIG. 17.]
+
+_Theory of the Striations._--We may regard the heaping up of the
+negative charges at intervals along the discharge as the fundamental
+feature in the striations, and this heaping up may be explained as
+follows. Imagine a corpuscle projected with considerable velocity from a
+place where the electric field is strong, such as the neighbourhood of
+the cathode; as it moves towards the anode through the gas it will
+collide with the molecules, ionize them and lose energy and velocity.
+Thus unless the corpuscle is acted on by a field strong enough to supply
+it with the energy it loses by collision, its speed will gradually
+diminish. Further, when its energy falls below a certain value it will
+unite with a molecule and become part of a negative ion, instead of a
+corpuscle; at this stage there will be a sudden and very large
+diminution in its velocity. Let us now follow the course of a stream of
+corpuscles starting from the cathode and approaching the anode. If the
+speed falls off as the stream proceeds, the corpuscles in the rear will
+gain on those in front and the density of the stream in the front will
+be increased. If at a certain place the velocity receives a sudden check
+by the corpuscles becoming loaded with a molecule, the density of the
+negative electricity will increase at this place with great rapidity,
+and here there will be a great accumulation of negative electricity, as
+at the bright head on the cathode side of a striation. Now this
+accumulation of negative electricity will produce a large electric force
+on the anode side; this will drive corpuscles forward with great
+velocity and ionize the gas. These corpuscles will behave like those
+shot from the cathode and will accumulate again at some distance from
+their origin, forming the bright head of the next striation, when the
+process will be repeated. On this view the bright heads of the
+striations act like electrodes, and the discharge passes from one bright
+head to the next as by a number of stepping stones, and not directly
+from cathode to anode. The luminosity at the head of the striations is
+due to the recombination of the ions. These ions have acquired
+considerable energy from the electric field, and this energy will be
+available for supplying the energy radiated away as light. The
+recombination of ions which do not possess considerable amounts of
+energy does not seem to give rise to luminosity. Thus, in an ionized gas
+not exposed to an electric field, although we have recombination between
+the ions, we need not have luminosity. We have at present no exact data
+as to the amount of energy which must be given to an ion to make it
+luminous on recombination; it also certainly varies with the nature of
+the ion; thus even with hot Wehnelt cathodes J. J. Thomson has never
+been able to make the discharge through air luminous with a potential
+less than from 16 to 17 volts. The mercury lamps, however, in which the
+discharge passes through mercury vapour are luminous with a potential
+difference of about 12 volts. It follows that if the preceding theory be
+right the potential difference between two bright striations must be
+great enough to make the corpuscles ionize by collision and also to give
+enough energy to the ions to make them luminous when they recombine. The
+difference of potential between the bright parts of successive
+striations has been measured by Hohn (_Phys. Zeit._ 9, p. 558); it
+varies with the pressure and with the gas. The smallest value given by
+Hohn is about 15 volts. In some experiments made by J. J. Thomson, when
+the pressure of the gas was very low, the difference of potential
+between two adjacent dark spaces was as low as 3.75 volts.
+
+_The Arc Discharge._--The discharges we have hitherto considered have
+been characterized by large potential differences and small currents. In
+the arc discharge we get very large currents with comparatively small
+potential differences. We may get the arc discharge by taking a battery
+of cells large enough to give a potential difference of 60 to 80 volts,
+and connecting the cells with two carbon terminals, which are put in
+contact, so that a current of electricity flows round the circuit. If
+the terminals, while the current is on, are drawn apart, a bright
+discharge, which may carry a current of many amperes, passes from one to
+the other. This arc discharge, as it is called, is characterized by
+intense heat and by the brilliant luminosity of the terminals. This
+makes it a powerful source of light. The temperature of the positive
+terminal is much higher than that of the negative. According to Violle
+(_Comptes Rendus_, 115, p. 1273) the temperature of the tip of the
+former is about 3500° C, and that of the latter 2700° C. The temperature
+of the arc itself he found to be higher than that of either of its
+terminals. As the arc passes, the positive terminal gets hollowed out
+into a crater-like shape, but the negative terminal remains pointed.
+Both terminals lose weight.
+
+ The appearance of the terminals is shown in fig. 18, given by Mrs
+ Ayrton (_Proc. Inst. Elec. Eng._ 28, p. 400); a, b represent the
+ terminals when the arc is quiet, and c when it is accompanied by a
+ hissing sound. The intrinsic brightness of the positive crater does
+ not increase with an increase in the current; an increased current
+ produces an increase in the area of the luminous crater, but the
+ amount of light given out by each unit of area of luminous surface is
+ unaltered. This indicates that the temperature of the crater is
+ constant; it is probably that at which carbon volatilizes. W. E.
+ Wilson (_Proc. Roy. Soc._ 58, p. 174; 60, p. 377) has shown that at
+ pressures of several atmospheres the intrinsic brightness of the
+ crater is considerably diminished.
+
+ [Illustration: FIG. 18.]
+
+ [Illustration: FIG. 19.]
+
+ The connexion between V, the potential difference between the
+ terminals, and l, the length of the arc, is somewhat analogous to that
+ which holds for the spark discharge. Fröhlich (_Electrotech. Zeit._ 4,
+ p. 150) gives for this connexion the relation V = m + nl, where m and
+ n are constants. Mrs Ayrton (_The Electric Arc_, chap. iv.) finds that
+ both m and n depend upon the current passing between the terminals,
+ and gives as the relation between V and l, V = [alpha] + [beta]/I +
+ ([gamma] + [delta]/I)l, where [alpha], [beta], [gamma], [delta] are
+ constants and I the current. The relation between current and
+ potential difference was made the subject of a series of experiments
+ by Ayrton (_Electrician_, 1, p. 319; xi. p. 418), some of whose
+ results are represented in fig. 19. For a quiet arc an increase in
+ current is accompanied by a fall in potential difference, while for
+ the hissing arc the potential difference is independent of the
+ current. The quantities m and n which occur in Fröhlich's equation
+ have been determined by several experimenters. For carbon electrodes
+ in air at atmospheric pressure m is about 39 volts, varying somewhat
+ with the size and purity of the carbons; it is diminished by soaking
+ the terminals in salt solution. The value of n given by different
+ observers varies considerably, ranging from .76 to 2 volts when l is
+ measured in millimetres; it depends upon the current, diminishing as
+ the current increases. When metallic terminals are used instead of
+ carbons, the value of m depends upon the nature of the metal, m in
+ general being larger the higher the temperature at which the metal
+ volatilizes. Thus v. Lang (_Wied. Ann._ 31, p. 384) found the
+ following values for m in air at atmospheric pressure:--C = 35; Pt =
+ 27.4; Fe = 25; Ni = 26.18; Cu = 23.86; Ag = 15.23; Zn = 19.86; Cd =
+ 10.28. Lecher (_Wied. Ann._ 33, p. 609) gives Pt = 28, Fe = 20, Ag =
+ 8, while Arons (_Wied. Ann._ 31, p. 384) found for Hg the value 12.8;
+ in this case the fall of potential along the arc itself was abnormally
+ small. In comparing these values it is important to remember that
+ Lecher (loc. cit.) has shown that with Fe or Pt terminals the arc
+ discharge is intermittent. Arons has shown that this is also the case
+ with Hg terminals, but no intermittence has been detected with
+ terminals of C, Ag or Cu. The preceding measurements refer to mean
+ potentials, and no conclusions as to the actual potential differences
+ at any time can be drawn when the discharge is discontinuous, unless
+ we know the law of discontinuity. The ease with which an arc is
+ sustained depends greatly on the nature of the electrodes; when they
+ are brass, zinc, cadmium, or magnesium it is exceedingly difficult to
+ get the arc.
+
+ [Illustration: FIG. 20.]
+
+ [Illustration: FIG. 21.]
+
+ The potential difference between the terminals is affected by the
+ pressure of the gas. The most extensive series of experiments on this
+ point is that made by Duncan, Rowland, and Tod (_Electrician_, 31, p.
+ 60), whose results are represented in fig. 20. We see from these
+ curves that for very short arcs the potential difference increases
+ continuously with the pressure, but for longer ones there is a
+ critical pressure at which the potential difference is a minimum, and
+ that this critical pressure seems to increase with the length of arc.
+ The nature of the gas also affects the potential difference. The
+ magnitude of this effect may be gathered from the following values
+ given by Arons (_Ann. der Phys._ 1, p. 700) for the potential
+ difference required to produce an arc 1.5 mm. long, carrying a current
+ of 4.5 amperes, between terminals of different metals in air and pure
+ nitrogen.
+
+ +-----------+------+-----------+
+ | Terminal. | Air. | Nitrogen. |
+ +-----------+------+-----------+
+ | Ag | 21 | ? |
+ | Zn | 23 | 21 |
+ | Cd | 25 | 21 |
+ | Cu | 27 | 30 |
+ | Fe | 29 | 20 |
+ | Pt | 36 | 30 |
+ | Al | 39 | 27 |
+ | Pb | .. | 18 |
+ | Mg | .. | 22 |
+ +-----------+------+-----------+
+
+ Thus, with the discharge for an arc of given length and current, the
+ nature of the terminals is the most important factor in determining
+ the potential difference. The effects produced by the pressure and
+ nature of the surrounding gas, although quite appreciable, are not of
+ so much importance, while in the spark discharge the nature of the
+ terminals is of no importance, everything depending upon the nature
+ and pressure of the gas.
+
+ The potential gradient in the arc is very far from being uniform. With
+ carbon terminals Luggin (_Wien. Ber._ 98, p. 1192) found that, with a
+ current of 15 amperes, there was a fall of potential of 33.7 close to
+ the anode, and one 8.7 close to the cathode, so that the curve
+ representing the distribution of potential between the terminals would
+ be somewhat like that shown in fig. 21. We have seen that a somewhat
+ analogous distribution of potential holds in the case of conduction
+ through flames, though in that case the greatest drop of potential is
+ in general at the cathode and not at the anode. The difference between
+ the changes of potential at the anode and cathode is not so large with
+ Fe and Cu terminals as with carbon ones; with mercury terminals, Arons
+ (_Wied. Ann._ 58, p. 73) found the anode fall to be 7.4 volts, the
+ cathode fall 5.4 volts.
+
+The case of the arc when the cathode is a pool of mercury and the anode
+a metal wire placed in a vessel from which the air has been exhausted is
+one which has attracted much attention, and important investigations on
+this point have been made by Hewitt (_Electrician_, 52, p. 447), Wills
+(_Electrician_, 54, p. 26), Stark, Retschinsky and Schnaposnikoff (_Ann.
+der Phys._ 18, p. 213) and Pollak (_Ann. der Phys._ 19, p. 217). In this
+arrangement the mercury is vaporized by the heat, and the discharge
+which passes through the mercury vapour gives an exceedingly bright
+light, which has been largely used for lighting factories, &c. The
+arrangement can also be used as a rectifier, for a current will only
+pass through it when the mercury pool is the cathode. Thus if such a
+lamp is connected with an alternating current circuit, it lets through
+the current in one direction and stops that in the other, thus
+furnishing a current which is always in one direction.
+
+_Theory of the Arc Discharge._--An incandescent body such as a piece of
+carbon even when at a temperature far below that of the terminals in an
+arc, emits corpuscles at a rate corresponding to a current of the order
+of 1 ampere per square centimetre of incandescent surface, and as the
+rate of increase of emission with the temperature is very rapid, it is
+probably at the rate of many amperes per square centimetre at the
+temperature of the negative carbon in the arc. If then a piece of carbon
+were maintained at this temperature by some external means, and used as
+a cathode, a current could be sent from it to another electrode whether
+the second electrode were cold or hot. If, however, these negatively
+electrified corpuscles did not produce other ions either by collision
+with the gas through which they move or with the anode, the spaces
+between cathode and anode would have a negative charge, which would tend
+to stop the corpuscles leaving the cathode and would require a large
+potential difference between anode and cathode to produce any
+considerable current. If, however, there is ionization either in the gas
+or at the anode, the positive ions will diffuse into the region of the
+discharge until they are sensibly equal in number to the negative ions.
+When this is the case the back electromotive force is destroyed and the
+same potential difference will carry a much larger current. The arc
+discharge may be regarded as analogous to the discharge between
+incandescent terminals, the only difference being that in the arc the
+terminals are maintained in the state of incandescence by the current
+and not by external means. On this view the cathode is bombarded by
+positive ions which heat it to such a temperature that negative
+corpuscles sufficient to carry the current are emitted by it. These
+corpuscles bombard the anode and keep it incandescent. They ionize also,
+either directly by collision or indirectly by heating the anode, the gas
+and vapour of the metal of which the anode is made, and produce in this
+way the supply of positive ions which keep the cathode hot.
+
+_Discharge from a Point._--A very interesting case of electric discharge
+is that between a sharply pointed electrode, such as a needle, and a
+metal surface of considerable area. At atmospheric pressures the
+luminosity is confined to the immediate neighbourhood of the point. If
+the sign of the potential of the point does not change, the discharge is
+carried by ions of one sign--that of the charge on the pointed
+electrode. The velocity of these ions under a given potential gradient
+has been measured by Chattock (_Phil. Mag._ 32, p. 285), and found to
+agree with that of the ions produced by Röntgen or uranium radiation,
+while Townsend (_Phil. Trans._ 195, p. 259) has shown that the charge on
+these ions is the same as that on the ions streaming from the point. If
+the pointed electrode be placed at right angles to a metal plane serving
+as the other electrode, the discharge takes place when, for a given
+distance of the point from the plane, the potential difference between
+the electrodes exceeds a definite value depending upon the pressure and
+nature of the gas through which the discharge passes; its value also
+depends upon whether, beginning with a small potential difference, we
+gradually increase it until discharge commences, or, beginning with a
+large potential difference, we decrease it until the discharge stops.
+The value found by the latter method is less than that by the former.
+According to Chattock's measurements the potential difference V for
+discharge between the point and the plate is given by the linear
+relation V = a + bl, where l is the distance of the point from the plate
+and a and b are constants. From v. Obermayer's (_Wien. Ber._ 100, 2, p.
+127) experiments, in which the distance l was greater than in
+Chattock's, it would seem that the potential for larger distances does
+not increase quite so rapidly with l as is indicated by Chattock's
+relation. The potential required to produce this discharge is much less
+than that required to produce a spark of length l between parallel
+plates; thus from Chattock's experiments to produce the point discharge
+when l = .5 cm. in air at atmospheric pressure requires a potential
+difference of about 3800 volts when the pointed electrode is positive,
+while to produce a spark at the same distance between plane electrodes
+would require a potential difference of about 15,000 volts. Chattock
+showed that with the same pointed electrode the value of the electric
+intensity at the point was the same whatever the distance of the point
+from the plane. The value of the electric intensity depended upon the
+sharpness of the point. When the end of the pointed electrode is a
+hemisphere of radius a, Chattock showed that for the same gas at the
+same pressure the electric intensity f when discharge takes place is
+roughly proportioned to a^-0.8. The value of the electric intensity at
+the pointed electrode is much greater than its value at a plane
+electrode for long sparks; but we must remember that at a distance from
+a pointed electrode equal to a small multiple of the radius of curvature
+of its extremity the electric intensity falls very far below that
+required to produce discharge in a uniform field, so that the discharge
+from a pointed electrode ought to be compared with a spark whose length
+is comparable with the radius of curvature of the point. For such short
+sparks the electric intensity is very high. The electric intensity
+required to produce the discharge from a gas diminishes as the pressure
+of the gas diminishes, but not nearly so rapidly as the electric
+intensity for long sparks. Here again the discharge from a point is
+comparable with short sparks, which, as we have seen, are much less
+sensitive to pressure changes than longer ones. The minimum potential at
+which the electricity streams from the point does not depend upon the
+material of which the point is made; it varies, however, considerably
+with the nature of the gas. The following are the results of some
+experiments on this point. Those in the first two columns are due to
+Röntgen, those in the third and fourth to Precht:--
+
+ +------+-----------------------------+--------------------+
+ | |Discharge Potential. Point +.| Pressure 760. |
+ | Gas. +--------------+--------------+----------+---------+
+ | | Pressure 205.| Pressure 110.| Point +. | Point -.|
+ +------+--------------+--------------+----------+---------+
+ | | Volts. | Volts. | Volts. | Volts. |
+ | H2 | 1296 | 1174 | 2125 | 1550 |
+ | O2 | 2402 | 1975 | 2800 | 2350 |
+ | CO | 2634 | 2100 | .. | .. |
+ | CH4 | 2777 | 2317 | .. | .. |
+ | NO | 3188 | 2543 | .. | .. |
+ | CO2 | 3287 | 2655 | 3475 | 2100 |
+ | N2 | .. | .. | 2600 | 2000 |
+ | Air | .. | .. | 2750 | 2050 |
+ +------+--------------+--------------+----------+---------+
+
+We see from this table that in the case of the discharge from a
+positively electrified point the greater the molecular weight of the gas
+the greater the potential required for discharge. Röntgen concluded from
+his experiments that the discharging potential from a positive point in
+different gases at the same pressure varies inversely as the mean free
+path of the molecules of the gas. In the same gas, however, at different
+pressures the discharging potential does not vary so quickly with the
+pressure as does the mean free path. In Precht's experiments, in which
+different gases were used, the variations in the discharging potential
+are not so great as the variations in the mean free path of the gases.
+
+The current of electrified air flowing from the point when the
+electricity is escaping--the well-known "electrical wind"--is
+accompanied by a reaction on the point which tends to drive it
+backwards. This reaction has been measured by Arrhenius (_Wied. Ann._
+63, p. 305), who finds that when positive electricity is escaping from a
+point in air the reaction on the point for a given current varies
+inversely as the pressure of the gas, and for different gases (air,
+hydrogen and carbonic acid) inversely as the square root of the
+molecular weight of the gas. The reaction when negative electricity is
+escaping is much less. The proportion between the reactions for positive
+and negative currents depends on the pressure of the gas. Thus for equal
+positive and negative currents in air at a pressure of 70 cm. the
+reaction for a positive point was 1.9 times that of a negative one, at
+40 cm. pressure 2.6 times, at 20 cm. pressure 3.2 times, at 10.3 cm.
+pressure 7 times, and at 5.1 cm. pressure 15 times the reaction for the
+negative point. Investigation shows that the reaction should be
+proportional to the quotient of the current by the velocity acquired by
+an ion under unit potential gradient. Now this velocity is inversely
+proportional to the pressure, so that the reaction should on this view
+be directly proportional to the pressure. This agrees with Arrhenius'
+results when the point is positive. Again, the velocities of an ion in
+hydrogen, air and carbonic acid at the same pressure are approximately
+inversely proportional to the square roots of their molecular weights,
+so that the reaction should be directly proportional to this quantity.
+This also agrees with Arrhenius' results for the discharge from a
+positive point. The velocity of the negative ion is greater than that of
+a positive one under the same potential gradient, so that the reaction
+for the negative point should be less than that for a positive one, but
+the excess of the positive reaction over the negative is much greater
+than that of the velocity of the negative ion over the velocity of the
+positive. There is, however, reason to believe that a considerable
+condensation takes place around the negative ion as a nucleus after it
+is formed, so that the velocity of the negative ion under a given
+potential gradient will be greater immediately after the ion is formed
+than when it has existed for some time. The measurements which have been
+made of the velocities of the ions relate to those which have been some
+time in existence, but a large part of the reaction will be due to the
+newly-formed ions moving with a greater velocity, and thus giving a
+smaller reaction than that calculated from the observed velocity.
+
+With a given potential difference between the point and the neighbouring
+conductor the current issuing from the point is greater when the point
+is negative than when it is positive, except in oxygen, when it is less.
+Warburg (_Sitz. Akad. d. Wissensch. zu Berlin_, 1899, 50, p. 770) has
+shown that the addition of a small quantity of oxygen to nitrogen
+produces a great diminution in the current from a negative point, but
+has very little effect on the discharge from a positive point. Thus the
+removal of a trace of oxygen made a leak from a negative point 50 times
+what it was before. Experiments with hydrogen and helium showed that
+impurities in these gases had a great effect on the current when the
+point was negative, and but little when it was positive. This suggests
+that the impurities, by condensing round the negative ions as nuclei,
+seriously diminish their velocity. If a point is charged up to a high
+and rapidly alternating potential, such as can be produced by the
+electric oscillations started when a Leyden jar is discharged, then in
+hydrogen, nitrogen, ammonia and carbonic acid gas a conductor placed in
+the neighbourhood of the point gets a negative charge, while in air and
+oxygen it gets a positive one. There are two considerations which are of
+importance in connexion with this effect. The first is the velocity of
+the ions in the electric field, and the second the ease with which the
+ions can give up their charges to the metal point. The greater velocity
+of the negative ions would, if the potential were rapidly alternating,
+cause an excess of negative ions to be left in the surrounding gas. This
+is the case in hydrogen. If, however, the metal had a much greater
+tendency to unite with negative than with positive ions, such as we
+should expect to be the case in oxygen, this would act in the opposite
+direction, and tend to leave an excess of positive ions in the gas.
+
+_The Characteristic Curve for Discharge through Gases._--When a current
+of electricity passes through a metallic conductor the relation between
+the current and the potential difference is the exceedingly simple one
+expressed by Ohm's law; the current is proportional to the potential
+difference. When the current passes through a gas there is no such
+simple relation. Thus we have already mentioned cases where the current
+increased as the potential increased although not in the same
+proportion, while as we have seen in certain stages of the arc discharge
+the potential difference diminishes as the current increases. Thus the
+problem of finding the current which a given battery will produce when
+part of the circuit consists of a gas discharge is much more complicated
+than when the circuit consists entirely of metallic conductors. If,
+however, we measure the potential difference between the electrodes in
+the gas when different currents are sent through it, we can plot a
+curve, called the "characteristic curve," whose ordinates are the
+potential differences between the electrodes in the gas and the
+abscissae the corresponding currents. By the aid of this curve we can
+calculate the current produced when a given battery is connected up to
+the gas by leads of known resistance.
+
+ For let E0 be the electromotive force of the battery, R the resistance
+ of the leads, i the current, the potential difference between the
+ terms in the gas will be E0 - Ri. Let ABC (fig. 22) be the
+ "characteristic curve," the ordinates being the potential difference
+ between the terminals in the gas, and the abscissae the current. Draw
+ the line LM whose equation is E = E0 - Ri, then the points where this
+ line cuts the characteristic curves will give possible values of i and
+ E, the current through the discharge tube and the potential difference
+ between the terminals. Some of these points may, however, correspond
+ to an unstable position and be impossible to realize. The following
+ method gives us a criterion by which we can distinguish the stable
+ from the unstable positions. If the current is increased by [delta]i,
+ the electromotive force which has to be overcome by the battery is
+ R[delta]i + dE/di · [delta]i. If R + dE/di is positive there will be
+ an unbalanced electromotive force round the circuit tending to stop
+ the current. Thus the increase in the current will be stopped and the
+ condition will be a stable one. If, however, R + dE/di is negative
+ there will be an unbalanced electromotive force tending to increase
+ the current still further; thus the current will go on increasing and
+ the condition will be unstable. Thus for stability R + dE/di must be
+ positive, a condition first given by Kaufmann (_Ann. der Phys._ 11, p.
+ 158). The geometrical interpretation of this condition is that the
+ straight line LM must, at the point where it cuts the characteristic
+ curve, be steeper than the tangent to characteristic curve. Thus of
+ the points ABC where the line cuts the curve in fig. 22, A and C
+ correspond to stable states and B to an unstable one. The state of
+ things represented by a point P on the characteristic curve when the
+ slope is downward cannot be stable unless there is in the external
+ circuit a resistance greater than that represented by the tangent of
+ the inclination of the tangent to the curve at P to the horizontal
+ axis.
+
+ [Illustration: FIG. 22.]
+
+ If we keep the external electromotive force the same and gradually
+ increase the resistance in the leads, the line LM will become steeper
+ and steeper. C will move to the left so that the current will
+ diminish; when the line gets so steep that it touches the curve at C',
+ any further increase in the resistance will produce an abrupt change
+ in the current; for now the state of things represented by a point
+ near A' is the only stable state. Thus if the BC part of the curve
+ corresponded to a luminous discharge and the A part to a dark
+ discharge, we see that if the electromotive force is kept constant
+ there is a minimum value of the current for the luminous discharge. If
+ the current is reduced below this value, the discharge ceases to be
+ luminous, and there is an abrupt diminution in the current.
+
+_Cathode Rays._--When the gas in the discharge tube is at a very low
+pressure some remarkable phenomena occur in the neighbourhood of the
+cathode. These seem to have been first observed by Plücker (_Pogg. Ann._
+107, p. 77; 116, p. 45) who noticed on the walls of the glass tube near
+the cathode a greenish phosphorescence, which he regarded as due to rays
+proceeding from the cathode, striking against the sides of the tube, and
+then travelling back to the cathode. He found that the action of a
+magnet on these rays was not the same as the action on the part of the
+discharge near the positive electrode. Hittorf (_Pogg. Ann._ 136, p. 8)
+showed that the agent producing the phosphorescence was intercepted by a
+solid, whether conductor or insulator, placed between the cathode and
+the sides of the tube. He regarded the phosphorescence as caused by a
+motion starting from the cathode and travelling in straight lines
+through the gas. Goldstein (_Monat. der Berl. Akad._, 1876, p. 24)
+confirmed this discovery of Hittorf's, and further showed that a
+distinct, though not very sharp, shadow is cast by a small object placed
+near a large plane cathode. This is a proof that the rays producing the
+phosphorescence must be emitted almost normally from the cathode, and
+not, like the rays of light from a luminous surface, in all directions,
+for such rays would not produce a perceptible shadow if a small body
+were placed near the plane. Goldstein regarded the phosphorescence as
+due to waves in the ether, for whose propagation the gas was not
+necessary. Crookes (_Phil. Trans._, 1879, pt. i. p. 135; pt. ii. pp.
+587, 661), who made many remarkable researches in this subject, took a
+different view. He regarded the rays as streams of negatively
+electrified particles projected normally from the cathode with great
+velocity, and, when the pressure is sufficiently low, reaching the sides
+of the tube, and by their impact producing phosphorescence and heat. The
+rays on this view are deflected by a magnet, because a magnet exerts a
+force on a charged moving body.
+
+These rays striking against glass make it phosphorescent. The colour of
+the phosphorescence depends on the kind of glass; thus the light from soda
+glass is a yellowish green, and that from lead glass blue. Many other
+bodies phosphoresce when exposed to these rays, and in particular the
+phosphorescence of some gems, such as rubies and diamonds, is exceedingly
+vivid. The spectrum of the phosphorescent light is generally continuous,
+but Crookes showed that the phosphorescence of some of the rare earths,
+such as yttrium, gives a spectrum of bright bands, and he founded on this
+fact a spectroscopic method of great importance. Goldstein (_Wied. Ann._
+54, p. 371) discovered that the haloid salts of the alkali metals change
+colour under the rays, sodium chloride, for example, becoming violet. The
+coloration is a surface one, and has been traced by E. Wiedemann and
+Schmidt (_Wied. Ann._ 54, p. 618) to the formation of a subchloride.
+Chlorides of tin, mercury and lead also change colour in the same way. E.
+Wiedemann (_Wied. Ann._ 56, p. 201) discovered another remarkable effect,
+which he called thermo-luminescence; he found that many bodies after being
+exposed to the cathode rays possess for some time the power of becoming
+luminous when their temperature is raised to a point far below that at
+which they become luminous in the normal state. Substances belonging to
+the class called by van 't Hoff solid solutions exhibit this property of
+thermo-luminescence to a remarkable extent. They are formed when two
+salts, one greatly in excess of the other, are simultaneously precipitated
+from a solution. A trace of MnSO4 in CaSO4 shows very brilliant
+thermo-luminescence. The impact of cathode rays produces after a time
+perceptible changes in the glass. Crookes (_Phil. Trans._ pt. ii. 1879, p.
+645) found that after glass has been phosphorescing for some time under
+the cathode rays it seems to get tired, and the phosphorescence is not so
+bright as it was initially. Thus, for example, when the shadow of a
+Maltese cross is thrown on the walls of the tube as in fig. 23, if after
+the discharge has been going on for some time the cross is shaken down or
+a new cathode used whose line of fire does not cut the cross, the pattern
+of the cross will still be seen on the glass, but it will now be brighter
+instead of darker than the surrounding portion. The portions shielded by
+the cross, not being tired by being made to phosphoresce for a long time,
+respond more vigorously to the stimulus than those portions which have not
+been protected. Skinner (_Proc. Camb. Phil. Soc._ ix. p. 371) and Thomson
+found on the glass which had been exposed to the rays gelatinous
+filaments, apparently silica, resulting from the reduction of the glass. A
+reducing action was also noticed by Villard (_Journ. de phys._ 3, viii. p.
+140) and Wehnelt (_Wied. Ann._ 67, p. 421). It can be well shown by
+letting the rays fall on a plate of oxidized copper, when the part struck
+by the rays will become bright. The rays heat bodies on which they fall,
+and if they are concentrated by using as a cathode a portion of a
+spherical surface, the heat at the centre becomes so great that a piece of
+platinum wire can be melted or a diamond charred. Measurements of the
+heating effects of the rays have been made by Thomson (_Phil. Mag._ [5],
+44, p. 293) and Cady (_Ann. der Phys._ 1, p. 678). Crookes (_Phil.
+Trans._, 1879, pt. i. p. 152) showed that a vane mounted as in a
+radiometer is set in rotation by the rays, the direction of the rotation
+being the same as would be produced by a stream of particles proceeding
+from the cathode. The movement is not due to the momentum imparted to the
+vanes by the rays, but to the difference in temperature between the sides
+of the vanes, the rays making the side against which they strike hotter
+than the other.
+
+[Illustration: FIG. 23.]
+
+_Effect of a Magnet._--The rays are deflected by a magnet, so that the
+distribution of phosphorescence over the glass and the shape and
+position of the shadows cast by bodies in the tube are altered by the
+proximity of a magnet. The laws of magnetic deflection of these rays
+have been investigated by Plücker (_Pogg._ _Ann._ 103, p. 88), Hittorf
+(_Pogg. Ann._ 136, p. 213), Crookes (_Phil. Trans._, 1879, pt. 1, p.
+557), and Schuster (_Proc. Roy. Soc._ 47, p. 526). The deflection is the
+same as that of negatively electrified particles travelling along the
+path of the rays. Such particles would in a magnetic field be acted on
+by a force at right angles to the direction of motion of the particle
+and also to the magnetic force, the magnitude of the force being
+proportional to the product of the velocity of the particle, the
+magnetic force, and the sine of the angle between these vectors. In this
+case we have seen that if the particle is not acted on by an
+electrostatic field, the path in a uniform magnetic field is a spiral,
+which, if the magnetic force is at right angles to the direction of
+projection of the particle, becomes a circle in the plane at right
+angles to the magnetic force, the radius being mv/He, where m, v, e are
+respectively the mass, velocity and charge on the particle, and H is the
+magnetic force. The smaller the difference of potential between the
+electrodes of the discharge tube the greater the deflection produced by
+a magnetic field of given strength, and as the difference of potential
+rapidly increases with diminution of pressure, after a certain pressure
+has been passed, the higher the exhaustion of the tube the less the
+magnetic deflection of the rays. Birkeland (_Comptes rendus_, 1896, p.
+492) has shown that when the discharge is from an induction coil the
+cathode rays produced in the tube at any one time are not equally
+deflected by a magnet, but that a narrow patch of phosphorescence when
+deflected by a magnet is split up into several distinct patches, giving
+rise to what Birkeland calls the "magnetic spectrum." Strutt (_Phil.
+Mag._ 48, p. 478) has shown that this magnetic spectrum does not occur
+if the discharge of a large number of cells is employed instead of the
+coil. Thomson (_Proc. Camb. Phil. Soc._ 9, p. 243) has shown that if the
+potential difference between the electrodes is kept the same the
+magnetic deflection is independent of the nature of the gas filling the
+discharge tube; this was tested with gases so different as air,
+hydrogen, carbonic acid and methyl iodide.
+
+_Charge of Negative Electricity carried by the Rays._--We have seen that
+the rays are deflected by a magnet, as if they were particles charged
+with negative electricity. Perrin (_Comptes rendus_, 121, p. 1130)
+showed by direct experiment that a stream of negative electricity is
+associated with the rays. A modification made by Thomson of Perrin's
+experiment is sketched in fig. 24 (_Phil. Mag._ 48, p. 478).
+
+ [Illustration: FIG. 24.]
+
+ The rays start from the cathode A, and pass through a slit in a solid
+ brass rod B fitting tightly into the neck of the tube. This rod is
+ connected with earth and used as the anode. The rays after passing
+ through the slit travel through the vessel C. D and E are two
+ insulated metal cylinders insulated from each other, and each having a
+ slit cut in its face so as to enable the rays to pass into the inside
+ of the inner cylinder, which is connected with an electrometer, the
+ outer cylinder being connected with the earth. The two cylinders are
+ placed on the far side of the vessel, but out of the direct line of
+ fire of the rays. When the rays go straight through the slit there is
+ only a very small negative charge communicated to the inner cylinder,
+ but when they are deflected by a magnet so that the phosphorescent
+ patch falls on the slit in the outer cylinder the inner cylinder
+ receives a very large negative charge, the increase coinciding very
+ sharply with the appearance of the phosphorescent patch on the slit.
+ When the patch is so much deflected by the magnet that it falls below
+ the slit, the negative charge in the cylinder again disappears. This
+ experiment shows that the cathode rays are accompanied by a stream of
+ negative electrification. The same apparatus can be used to show that
+ the passage of cathode rays through a gas makes it a conductor of
+ electricity. For if the induction coil is kept running and a stream of
+ the rays kept steadily going into the inner cylinder, the potential
+ of the inner cylinder reaches a definite negative value below which it
+ does not fall, however long the rays may be kept going. The cylinder
+ reaches a steady state in which the gain of negative electricity from
+ the cathode rays is equal to the loss by leakage through the
+ conducting gas, the conductivity being produced by the passage of the
+ rays through it. If the inner cylinder is charged up initially with a
+ greater negative charge than corresponds to the steady state, on
+ turning the rays on to the cylinder the negative charge will decrease
+ and not increase until it reaches the steady state. The conductivity
+ produced by the passage of cathode rays through a gas diminishes
+ rapidly with the pressure. When rays pass through a gas at a low
+ pressure, they are deflected by an electric field; when the pressure
+ of the gas is higher the conductivity it acquires when the cathode
+ rays pass through it is so large that the potential gradient cannot
+ reach a sufficiently high value to produce an appreciable deflection.
+
+Thus the cathode rays carry a charge of negative electricity; the
+experiment described on page 875 (fig. 13) shows that they are deflected
+by an electric field as if they were negatively electrified, and are
+acted on by a magnetic force in just the way this force would act on a
+negatively electrified body moving along the path of the rays. There is
+therefore every reason for believing that they are charges of negative
+electricity in rapid motion. By measuring the deflection produced by
+magnetic and electric fields we can determine the velocity with which
+these particles moved and the ratio of the mass of the particle to the
+charge carried by it.
+
+We may conclude from the experiments that the value of m/e for the
+particles constituting the cathode rays is of the order 1/1.7 × 10^7,
+and we have seen that m/e has the same value in all the other cases of
+negative ions in a gas at low pressure for which it has been
+measured--viz. for the ions produced when ultra-violet light falls on a
+metal plate, or when an incandescent carbon filament is surrounded by a
+gas at a low pressure, and for the [beta] particles given out by
+radio-active bodies. We have also seen that the value of the charge on
+the gaseous ion, in all cases in which it has been measured--viz. the
+ions produced by Röntgen and uranium radiation, by ultra-violet light,
+and by the discharge of electrification from a point--is the same in
+magnitude as the charge carried by the hydrogen atom in the electrolysis
+of solutions. The mass of the hydrogen alone is, however, 10^-4 times
+this charge, while the mass of the carriers of negative electrification
+is only 1/1.7 × 10^7 times the charge; hence the mass of the carriers of
+the negative electrification is only 1/1700 of the mass of the hydrogen
+atom. We are thus, by the study of the electric discharge, forced to
+recognize the existence of masses very much smaller than the smallest
+mass hitherto recognized.
+
+ Direct determinations of the velocity of the cathode rays have been
+ made by J. J. Thomson (_Phil. Mag._ 38, p. 358), who measured the
+ interval between the appearance of phosphorescence on two pieces of
+ glass placed at a known distance apart, and by Maiorana (_Nuovo
+ Cimento_, 4, 6, p. 336) and Battelli and Stefanini (_Phys. Zeit._ 1,
+ p. 51), who measured the interval between the arrival of the negative
+ charge carried by the rays at two places separated by a known
+ distance. The values of the velocity got in this way are much smaller
+ than the values got by the indirect methods previously described: thus
+ J. J. Thomson at a fairly high pressure found the velocity to be 2 ×
+ 10^7 cm./sec. Maiorana found values ranging between 10^7 and 6 × 10^7
+ cm./sec, and Battelli and Stefanini values ranging from 6 × 10^6 to
+ 1.2 × 10^7. In these methods it is very difficult to eliminate the
+ effect of the interval which elapses between the arrival of the rays
+ and the attainment by the means of detection, such as the
+ phosphorescence of the glass or the deflection of the electrometer, of
+ sufficient intensity to affect the senses.
+
+[Illustration: FIG. 25.]
+
+_Transmission of Cathode Rays through Solids--Lenard Rays._--It was for
+a long time believed that all solids were absolutely opaque to these
+rays, as Crookes and Goldstein had proved that very thin glass, and even
+a film of collodion, cast intensely black shadows. Hertz (_Wied. Ann._
+45, p. 28), however, showed that behind a piece of gold-leaf or
+aluminium foil an appreciable amount of phosphorescence occurred on the
+glass, and that the phosphorescence moved when a magnet was brought
+near. A most important advance was next made by Lenard (_Wied. Ann._ 51,
+p. 225), who got the cathode rays to pass from the inside of a discharge
+tube to the air outside. For this purpose he used a tube like that shown
+in fig. 25. The cathode K is an aluminium disc 1.2 cm. in diameter
+fastened to a stiff wire, which is surrounded by a glass tube. The anode
+A is a brass strip partly surrounding the cathode. The end of the tube
+in front of the cathode is closed by a strong metal cap, fastened in
+with marine glue, in the middle of which a hole 1.7 mm. in diameter is
+bored, and covered with a piece of very thin aluminium foil about .0026
+mm. in thickness. The aluminium window is in metallic contact with the
+cap, and this and the anode are connected with the earth. The tube is
+then exhausted until the cathode rays strike against the window. Diffuse
+light spreads from the window into the air outside the tube, and can be
+traced in a dark room for a distance of several centimetres. From the
+window, too, proceed rays which, like the cathode rays, can produce
+phosphorescence, for certain bodies phosphoresce when placed in the
+neighbourhood of the window. This effect is conveniently observed by the
+platino-cryanide screens used to detect Röntgen radiation. The
+properties of the rays outside the tube resemble in all respects those
+of cathode rays; they are deflected by a magnet and by an electric
+field, they ionize the gas through which they pass and make it a
+conductor of electricity, and they affect a photographic plate and
+change the colour of the haloid salts of the alkali metals. As, however,
+it is convenient to distinguish between cathode rays outside and inside
+the tube, we shall call the former Lenard rays. In air at atmospheric
+pressure the Lenard rays spread out very diffusely. If the aluminium
+window, instead of opening into the air, opens into another tube which
+can be exhausted, it is found that the lower the pressure of the gas in
+this tube the farther the rays travel and the less diffuse they are. By
+filling the tube with different gases Lenard showed that the greater the
+density of the gas the greater is the absorption of these rays. Thus
+they travel farther in hydrogen than in any other gas at the same
+pressure. Lenard showed, too, that if he adjusted the pressure so that
+the density of the gas in this tube was the same--if, for example, the
+pressure when the tube was filled with oxygen was 1/16 of the pressure
+when it was filled with hydrogen--the absorption was constant whatever
+the nature of the gas. Becker (_Ann. der Phys._ 17, p. 381) has shown
+that this law is only approximately true, the absorption by hydrogen
+being abnormally large, and by the inert monatomic gases, such as helium
+and argon, abnormally small. The distance to which the Lenard rays
+penetrate into this tube depends upon the pressure in the discharge
+tube; if the exhaustion in the latter is very high, so that there is a
+large potential difference between the cathode and the anode, and
+therefore a high velocity for the cathode rays, the Lenard rays will
+penetrate farther than when the pressure in the discharge tube is higher
+and the velocity of the cathode rays smaller. Lenard showed that the
+greater the penetrating power of his rays the smaller was their magnetic
+deflection, and therefore the greater their velocity; thus the greater
+the velocity of the cathode rays the greater is the velocity of the
+Lenard rays to which they give rise. For very slow cathode rays the
+absorption by different gases departs altogether from the density law,
+so much so that the absorption of these rays by hydrogen is greater than
+that by air (Lenard, _Ann. der Phys._ 12, p. 732). Lenard (_Wied. Ann._
+56, p. 255) studied the passage of his rays through solids as well as
+through gases, and arrived at the very interesting result that the
+absorption of a substance depends only upon its density, and not upon
+its chemical composition or physical state; in other words, the amount
+of absorption of the rays when they traverse a given distance depends
+only on the quantity of matter they cut through in the distance.
+McClelland (_Proc. Roy. Soc._ 61, p. 227) showed that the rays carry a
+charge of negative electricity, and M'Lennan measured the amount of
+ionization rays of given intensity produced in different gases, finding
+that if the pressure is adjusted so that the density of the different
+gases is the same the number of ions per cubic centimetre is also the
+same. In this case, as Lenard has shown, the absorption is the same, so
+that with the Lenard rays, as with uranium and probably with Röntgen
+rays, equal absorption corresponds to equal ionization. A convenient
+method for producing Lenard rays of great intensity has been described
+by Des Coudres (_Wied. Ann._ 62, p. 134).
+
+_Diffuse Reflection of Cathode Rays._--When cathode rays fall upon a
+surface, whether of an insulator or a conductor, cathode rays start from
+the surface in all directions. This phenomenon, which was discovered by
+Goldstein (_Wied. Ann._ 62, p. 134), has been investigated by Starke
+(_Wied. Ann._ 66, p. 49; _Ann. der Phys._ 111, p. 75), Austin and Starke
+(_Ann. der Phys._ 9, p. 271), Campbell-Swinton (_Proc. Roy. Soc._ 64, p.
+377), Merritt (_Phys. Rev._ 7, p. 217) and Gehrcke (_Ann. der Phys._ 8,
+p. 81); it is often regarded as analogous to the diffuse reflection of
+light from such a surface as gypsum, and is spoken of as the diffuse
+reflection of the cathode rays. According to Merritt and Austin and
+Starke the deviation in a magnetic field of these reflected rays is the
+same as that of the incident rays. The experiments, however, were
+confined to rays reflected so that the angle of reflection was nearly
+equal to that of incidence. Gehrcke showed that among the reflected rays
+there were a large number which had a much smaller velocity than the
+incident ones. According to Campbell-Swinton the "diffuse" reflection is
+accompanied by a certain amount of "specular" reflection. Lenard, who
+used slower cathode rays than Austin and Starke, could not detect in the
+scattered rays any with velocities comparable with that of the incident
+rays; he obtained copious supplies of slow rays whose speed did not
+depend on the angle of incidence of the primary rays (_Ann. der Phys._
+15, p. 485). When the angle of incidence is very oblique the surface
+struck by the rays gets positively charged, showing that the secondary
+rays are more numerous than the primary.
+
+_Repulsion of two Cathode Streams._--Goldstein discovered that if in a
+tube there are two cathodes connected together, the cathodic rays from
+one cathode are deflected when they pass near the other. Experiments
+bearing on this subject have been made by Crookes and Wiedemann and
+Ebert. The phenomena may be described by saying that the repulsion of
+the rays from a cathode A by a cathode B is only appreciable when the
+rays from A pass through the Crookes dark space round B. This is what we
+should expect if we remember that the electric field in the dark space
+is far stronger than in the rest of the discharge, and that the gas in
+the other parts of the tube is rendered a conductor by the passage
+through it of the cathode rays, and therefore incapable of transmitting
+electrostatic repulsion.
+
+Scattering of the Negative Electrodes.--In addition to the cathode rays,
+portions of metal start normally from the cathode and form a metallic
+deposit on the walls of the tube. The amount of this deposit varies very
+much with the metal. Crookes (_Proc. Roy. Soc._ 50, p. 88) found that
+the quantities of metal torn from electrodes of the same size, in equal
+times, by the same current, are in the order Pd, Au, Ag, Pb, Sn, Pt, Cu,
+Cd, Ni, In, Fe.... In air there is very little deposit from an Al
+cathode, but it is abundant in tubes filled with the monatomic gases,
+mercury vapour, argon or helium. The scattering increases as the density
+of the gas diminishes. The particles of metal are at low pressures
+deflected by a magnet, though not nearly to the same extent as the
+cathode rays. According to Grandquist, the loss of weight of the cathode
+in a given time is proportional to the square of the current; it is
+therefore not, like the loss of the cathode in ordinary electrolysis,
+proportional to the quantity of current which passes through it.
+
+[Illustration: FIG. 26.]
+
+_Positive Rays or "Canalstrahlen."_--Goldstein (_Berl. Sitzungsb._ 39,
+p. 691) found that with a perforated cathode certain rays occurred
+behind the cathode which were not appreciably deflected by a magnet;
+these he called Canalstrahlen, but we shall, for reasons which will
+appear later, call them "positive rays."
+
+Their appearance is well shown in fig. 26, taken from a paper by Wehnelt
+(_Wied. Ann._ 67, p. 421) in which they are represented at B. Goldstein
+found that their colour depends on the gas in which they are formed,
+being gold-colour in air and nitrogen, rose-colour in hydrogen,
+yellowish rose in oxygen, and greenish gray in carbonic acid.
+
+The colour of the luminosity due to positive rays is not in general the
+same as that due to anode rays; the difference is exceptionally well
+marked in helium, where the cathode ray luminosity is blue while that
+due to the positive rays is red. The luminosity produced when the rays
+strike against solids is also quite distinct. The cathode rays make the
+body emit a continuous spectrum, while the spectrum produced by the
+positive rays often shows bright lines. Thus lithium chloride under
+cathode rays gives out a steely blue light and the spectrum is
+continuous, while under the positive rays the salt gives out a brilliant
+red light and the spectrum shows the red helium line. It is remarkable
+that the lines on the spectra of the alkali metals are much more easily
+produced when the positive rays fall on the oxide of the metal than when
+they fall on the metal itself. Thus when the positive rays fall on a
+pool of the liquid alloy of sodium and potassium the specks of oxide on
+the surface shine with a bright yellow light while the untarnished part
+of the surface is quite dark.
+
+W. Wien (_Wied. Ann._ 65, p. 445) measured the values of e/m for the
+particles forming the positive rays. Other measurements have been made
+by Ewers (_Wied. Ann._ 69, p. 167) and J. J. Thomson (_Phil. Mag._ 13,
+p. 561). The differences between the values of e/m for the cathode and
+positive rays are very remarkable. For cathode rays whose velocity does
+not approach that of light, e/m is always equal to 1.7 × 10^8, while for
+the positive rays the greatest value of this quantity yet observed is
+10^4, which is also the value of e/m for the hydrogen ions in the
+electrolysis of dilute solutions. In some experiments made by J. J.
+Thomson (_Phil. Mag._, 14, p. 359) it was found that when the pressure
+of the gas was not too low the bright spot produced by the impact of a
+pencil of these rays on a phosphorescent screen is deflected by electric
+and magnetic forces into a continuous band extending on both sides of
+the undeflected position. The portion on one side is in general much
+fainter than that on the other. The direction of this deflection shows
+that it is produced by particles charged with negative electricity,
+while the brighter band is due to particles charged with positive
+electricity. The negatively electrified particles which produce the band
+c.c are not corpuscles, for from the electric and magnetic deflections
+we can find the value of e/m. As this proves to be equal to 10^4, we see
+that the mass of the carrier of the negative charge is comparable with
+that of an atom, and so very much greater than that of a corpuscle. At
+very low pressures part of the phosphorescence disappears, while the
+upper portion breaks up into two patches (fig. 27). For one of these the
+maximum value of e/m is 10^4 and for the other 5 × 10³. At low pressures
+the appearance of the patches and the values of e/m are the same whether
+the tube is filled originally with air, hydrogen or helium. In some of
+the experiments the tube was exhausted until the pressure was too low to
+allow the discharge to pass. A very small quantity of the gas under
+investigation was then admitted into the tube, just sufficient to allow
+the discharge to pass, and the deflection of the phosphorescent patch
+measured. The following gases were admitted into the tube, air, carbonic
+oxide, oxygen, hydrogen, helium, argon and neon, but whatever the gas
+the appearance of the phosphorescence was the same; in every case there
+were two patches, for one of which e/m = 10^4 and for the other e/m =
+5 × 10³. In helium at higher pressures another patch was observed, for
+which e/m = 2.5 × 10^8. The continuous band into which the
+phosphorescent spot is drawn out when the pressure is not exceedingly
+low, which involves the existence of particles for which the mean value
+of e/m varies from zero to 10^4, can be explained as follows. The rays
+on their way to the phosphorescent screen have to pass through gas which
+is ionized by the passage through it of the positive rays; this gas will
+therefore contain free corpuscles. The particles which constitute the
+rays start with a charge of positive electricity. Some of these
+particles in their journey through the gas attract a corpuscle whose
+negative charge neutralizes the positive charge on the particle. The
+particles when in this neutral state may be ionized by collision and
+reacquire a positive charge, or by attracting another particle may
+become negatively charged, and this process may be repeated several
+times on their journey to the phosphorescent screen. Thus some of the
+particles, instead of being positively charged for the whole of the time
+they are exposed to the electric and magnetic forces, may be for a part
+of that time without a charge or even have a negative charge. The
+deflection of a particle is proportional to the average value of its
+charge whilst under the influence of the deflecting forces. Thus if a
+particle is without a charge for a part of the time, its deflection will
+be less than that of a particle which has retained its positive charge
+for the whole of its journey, while the few particles which have a
+negative charge for a longer time than they have a positive will be
+deflected in the opposite direction to the main portion and will produce
+the tail (fig. 27).
+
+[Illustration: Fig. 27.]
+
+A similar explanation will apply to the positive rays discovered by
+Villard (_Comptes rendus_, 143, p. 674) and J. J. Thomson (_Phil. Mag._
+13, p. 359), which travel in the opposite direction to the rays we have
+been considering, i.e. they travel away from the cathode and in the
+direction of the cathode's rays; these rays are sometimes called
+"retrograde" rays. These as far as has been observed have always the
+same maximum value of e/m, i.e. 10^4, and there are a considerable
+number of negative ones always mixed with them. The maximum velocity of
+both the positive and retrograde rays is about 2 × 10^8 cm./sec. and
+varies very little with the potential difference between the electrodes
+in the tube in which they are produced (J. J. Thomson, _Phil. Mag._,
+Dec. 1909).
+
+The positive rays show, when the pressure is not very low, the line
+spectrum of the gas through which they pass. An exceedingly valuable set
+of observations on this point have been made by Stark and his pupils
+(_Physik. Zeit._ 6, p. 892; _Ann. der Phys._ 21, pp. 40, 457). Stark has
+shown that in many gases, notably hydrogen, the spectrum shows the
+Doppler effect, and he has been able to calculate in this way the
+velocity of the positive rays.
+
+_Anode Rays._--Gehrcke and Reichenhein (_Ann. der Phys._ 25, p. 861)
+have found that when the anode consists of a mixture of sodium and
+lithium chloride raised to a high temperature either by the discharge
+itself or by an independent heating circuit, very conspicuous rays come
+from the anode when the pressure of the gas in the discharge tube is
+very low, and a large coil is used to produce the discharge. The
+determination of e/m for these rays showed that they are positively
+charged atoms of sodium or lithium, moving with very considerable
+velocity; in some of Gehrcke's experiments the maximum velocity was as
+great as 1.8 × 10^7 cm./sec. though the average was about 10^7 cm./sec.
+These velocities are less than those of the positive rays whose maximum
+velocity is about 2 × 10^8 cm./sec. (J. J. T.)
+
+
+FOOTNOTES:
+
+ [1] The values for nickel and bismuth given in the table are much
+ higher than later values obtained with pure electrolytic nickel and
+ bismuth.
+
+ [2] The value here given, namely 12.885, for the electric
+ mass-resistivity of liquid mercury as determined by Matthiessen is
+ now known to be too high by nearly 1%. The value at present accepted
+ is 12.789 ohms per metre-gramme at 0° C.
+
+ [3] The value (1630) here given for hard-drawn copper is about ¼%
+ higher than the value now adopted, namely, 1626. The difference is
+ due to the fact that either Jenkin or Matthiessen did not employ
+ precisely the value at present employed for the density of hard-drawn
+ and annealed copper in calculating the volume-resistivities from the
+ mass-resistivities.
+
+ [4] Matthiessen's value for nickel is much greater than that obtained
+ in more recent researches. (See Matthiessen and Vogt, _Phil. Trans._,
+ 1863, and J. A. Fleming, _Proc. Roy. Soc._, December 1899.)
+
+ [5] Matthiessen's value for mercury is nearly 1% greater than the
+ value adopted at present as the mean of the best results, namely
+ 94,070.
+
+ [6] The samples of silver, copper and nickel employed for these tests
+ were prepared electrolytically by Sir J. W. Swan, and were
+ exceedingly pure and soft. The value for volume-resistivity of nickel
+ as given in the above table (from experiments by J. A. Fleming,
+ _Proc. Roy. Soc._, December 1899) is much less (nearly 40%) than the
+ value given by Matthiessen's researches.
+
+ [7] The electrolytic bismuth here used was prepared by Hartmann and
+ Braun, and the resistivity taken by J. A. Fleming. The value is
+ nearly 20% less than that given by Matthiessen.
+
+ [8] In 1899 a committee was formed of representatives from eight of
+ the leading manufacturers of insulated copper cables with delegates
+ from the Post Office and Institution of Electrical Engineers, to
+ consider the question of the values to be assigned to the resistivity
+ of hard-drawn and annealed copper. The sittings of the committee were
+ held in London, the secretary being A. H. Howard. The values given in
+ the above paragraphs are in accordance with the decision of this
+ committee, and its recommendations have been accepted by the General
+ Post Office and the leading manufacturers of insulated copper wire
+ and cables.
+
+ [9] Platinoid is an alloy introduced by Martino, said to be similar
+ in composition to German silver, but with a little tungsten added. It
+ varies a good deal in composition according to manufacture, and the
+ resistivity of different specimens is not identical. Its electric
+ properties were first made known by J. T. Bottomley, in a paper read
+ at the Royal Society, May 5, 1885.
+
+ [10] An equivalent gramme molecule is a weight in grammes equal
+ numerically to the chemical equivalent of the salt. For instance, one
+ equivalent gramme molecule of sodium chloride is a mass of 58.5
+ grammes. NaCl = 58.5.
+
+ [11] F. Kohlrausch and L. Holborn, _Das Leitvermögen der Elektrolyte_
+ (Leipzig, 1898).
+
+ [12] It should be noticed that the velocities calculated in
+ Kohlrausch's theory and observed experimentally are the average
+ velocities, and involve both the factors mentioned above; they
+ include the time wasted by the ions in combination with each other,
+ and, except at great dilution, are less than the velocity with which
+ the ions move when free from each other.
+
+
+
+
+
+End of the Project Gutenberg EBook of Encyclopaedia Britannica, 11th
+Edition, Volume 6, Slice 8, by Various
+
+*** END OF THIS PROJECT GUTENBERG EBOOK ENCYC. BRITANNICA, VOL 6, SL 8 ***
+
+***** This file should be named 32063-8.txt or 32063-8.zip *****
+This and all associated files of various formats will be found in:
+ https://www.gutenberg.org/3/2/0/6/32063/
+
+Produced by Marius Masi, Don Kretz, Juliet Sutherland, and
+the Online Distributed Proofreading Team at
+https://www.pgdp.net
+
+
+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
+https://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
+
+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 F3. 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 MERCHANTIBILITY 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 https://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
+https://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 https://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 https://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 including checks, online payments and credit card
+donations. To donate, please visit: https://pglaf.org/donate
+
+
+Section 5. General Information About Project Gutenberg-tm electronic
+works.
+
+Professor Michael S. Hart was 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:
+
+ https://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.
diff --git a/32063-8.zip b/32063-8.zip
new file mode 100644
index 0000000..579f1bc
--- /dev/null
+++ b/32063-8.zip
Binary files differ
diff --git a/32063-h.zip b/32063-h.zip
new file mode 100644
index 0000000..b3fac78
--- /dev/null
+++ b/32063-h.zip
Binary files differ
diff --git a/32063-h/32063-h.htm b/32063-h/32063-h.htm
new file mode 100644
index 0000000..818cb7d
--- /dev/null
+++ b/32063-h/32063-h.htm
@@ -0,0 +1,6917 @@
+<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Strict//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-strict.dtd">
+<html xmlns="http://www.w3.org/1999/xhtml" xml:lang="en" lang="en">
+
+ <head>
+ <meta http-equiv="Content-Type" content=
+ "text/html; charset=iso-8859-1" />
+
+ <title>
+ The Project Gutenberg eBook of Encyclop&aelig;dia Britannica, Volume VI slice VIII - Conduction, Electric.
+ </title>
+
+ <style type="text/css">
+
+ body { margin-left: 12%; margin-right: 12%; text-align: justify; }
+ p { margin-top: .75em; margin-bottom: .75em; text-indent: 1em; line-height: 1.4em;}
+ p.tab { margin-left: 2em; text-indent: -2em; }
+ p.c { margin-top: .25em; margin-bottom: .25em; text-indent: 1em; padding-left: 1em; line-height: 1.4em;}
+ p.noind { margin-top: .75em; margin-bottom: .75em; text-indent: 0; }
+
+ h2,h3 { text-align: center; }
+ hr { margin-left: auto; margin-right: auto; text-align: center; width: 70%; height: 5px; background-color: #dcdcdc; border:none; }
+ hr.art { margin-left: auto; margin-right: auto; width: 40%; height: 5px; background-color: #778899;
+ margin-top: 2em; margin-bottom: 6em }
+ hr.foot {margin-left: 2em; width: 16%; background-color: black; margin-top: 1em; margin-bottom: 0; height: 1px; }
+ hr.full {width: 100%}
+
+ table.ws {white-space: nowrap; border-collapse: collapse; margin-left: auto; margin-right: auto;
+ margin-top: 2em; margin-bottom: 2em;}
+ table.reg { margin-left: auto; margin-right: auto; clear: both; }
+ table.nobctr { margin-left: auto; margin-right: auto; border-collapse: collapse; }
+ table.pic { margin-left: auto; margin-right: auto; }
+ table.math0 { vertical-align: middle; margin-left: auto; margin-right: auto; border-collapse: collapse;}
+ table.math0 td {text-align: center;}
+ table.math0 td.np {text-align: center; padding-left: 0; padding-right: 0;}
+
+ table.reg td { padding-left: 3em; text-indent: -2em; white-space: normal;}
+ table.reg td.tc5p { padding-left: 2em; text-indent: 0em; white-space: normal;}
+ table.nobctr td { white-space: normal; }
+ table.pic td { white-space: normal; text-indent: 1em; padding-left: 2em; padding-right: 1em;}
+ table.nobctr p {text-indent: -1em; margin-left: 1em;}
+
+ td { white-space: nowrap; padding-right: 0.3em; padding-left: 0.3em;}
+ td.norm { white-space: normal; }
+ td.denom { border-top: 1px solid black; text-align: center; padding-right: 0.3em; padding-left: 0.3em;}
+
+ td.tcc { padding-right: 0.5em; padding-left: 0.5em; text-align: center; vertical-align: top;}
+ td.tccm { padding-right: 0.5em; padding-left: 0.5em; text-align: center; vertical-align: middle;}
+ td.tccb { padding-right: 0.5em; padding-left: 0.5em; text-align: center; vertical-align: bottom;}
+ td.tcr { padding-right: 0.5em; padding-left: 0.5em; text-align: right; vertical-align: top;}
+ td.tcrm { padding-right: 0.5em; padding-left: 0.5em; text-align: right; vertical-align: middle;}
+ td.tcl { padding-right: 0.5em; padding-left: 0.5em; text-align: left; vertical-align: top;}
+ td.tclm { padding-right: 0.5em; padding-left: 0.5em; text-align: left; vertical-align: middle;}
+ td.vb { vertical-align: bottom; }
+
+ .caption { font-size: 0.9em; text-align: center; padding-bottom: 1em; padding-left: 1em; padding-right: 1em;}
+ .caption1 { font-size: 0.9em; text-align: left; padding-bottom: 1em; padding-left: 3em; padding-right: 2em;}
+
+ td.lb {border-left: black 1px solid;}
+ td.ltb {border-left: black 1px solid; border-top: black 1px solid;}
+ td.rb {border-right: black 1px solid;}
+ td.rb2 {border-right: black 2px solid;}
+ td.tb, span.tb {border-top: black 1px solid;}
+ td.bb {border-bottom: black 1px solid;}
+ td.rlb {border-right: black 1px solid; border-left : black 1px solid;}
+ td.allb {border: black 1px solid;}
+
+ table p { margin: 0;}
+
+ a:link, a:visited, link {text-decoration:none}
+
+ .author {text-align: right; margin-top: -1em; margin-right: 1em; font-variant: small-caps;}
+ .center {text-align: center; text-indent: 0;}
+ .center1 {text-align: center; text-indent: 0; margin-top: 1em; margin-bottom: 1em;}
+ .grk {font-style: normal; font-family:"Palatino Linotype","New Athena Unicode",Gentium,"Lucida Grande", Galilee, "Arial Unicode MS", sans-serif;}
+
+ .f80 {font-size: 80%}
+ .f90 {font-size: 90%}
+ .f150 {font-size: 150%; font-family: 'Courier New'}
+ .f200 {font-size: 200%; font-family: 'Courier New'}
+ .f250 {font-size: 250%; font-family: 'Courier New'; letter-spacing: -0.1em;}
+
+ .sp {position: relative; bottom: 0.5em; font-size: 0.75em;}
+ .sp1 {position: relative; bottom: 0.6em; font-size: 0.75em;}
+ .su {position: relative; top: 0.3em; font-size: 0.75em;}
+ .su1 {position: relative; top: 0.5em; font-size: 0.75em; margin-left: -1.2ex;}
+ .su2 {position: relative; top: 0.8em; font-size: 0.75em; margin-left: -1.9ex;}
+ .spp {position: relative; bottom: 0.5em; font-size: 0.6em;}
+ .suu {position: relative; top: 0.2em; font-size: 0.6em;}
+ .sc {font-variant: small-caps;}
+ .scs {font-variant: small-caps; font-size: 70%;}
+ .ov {text-decoration: overline}
+ .cl {background-color: #f5f5f5;}
+
+ .pagenum {position: absolute; right: 5%; text-align: right; font-size: 10pt;
+ background-color: #f5f5f5; color: #778899; text-indent: 0;
+ padding-left: 0.5em; padding-right: 0.5em; font-style: normal; }
+ span.sidenote {width: 8em; margin-bottom: 1em; margin-top: 1.7em; margin-right: 2em;
+ font-size: 85%; float: left; clear: left; font-weight: bold;
+ font-style: italic; text-align: left; text-indent: 0;
+ background-color: #f5f5f5; color: black; }
+ .note {margin-left: 2em; margin-right: 2em; font-size: 0.9em; }
+ .fn { position: absolute; left: 12%; text-align: left; background-color: #f5f5f5;
+ text-indent: 0; padding-left: 0.2em; padding-right: 0.2em; }
+ span.correction {border-bottom: 1px dashed red;}
+
+ div.poemr { margin-top: .75em; margin-bottom: .75em;}
+ div.poemr p { margin-left: 0; padding-left: 3em; text-indent: -3em; margin-top: 0em; margin-bottom: 0em; }
+ div.poemr p.s { margin-top: 1.5em; }
+ div.poemr p.i05 { margin-left: 0.4em; }
+ div.poemr p.i2 { margin-left: 2em; }
+ div.poemr p.i4 { margin-left: 4em; }
+
+ .figright1 { padding-right: 1em; padding-left: 2em; padding-top: 1.5em; text-align: center; }
+ .figleft1 { padding-right: 2em; padding-left: 1em; padding-top: 1.5em; text-align: center; }
+ .figcenter {text-align: center; margin: auto; margin-left: auto; margin-right: auto; padding-top: 1.5em;}
+ .figcenter1 {text-align: center; margin-left: auto; margin-right: auto; padding-top: 2em; padding-bottom: 2em;}
+ .figure {text-align: center; padding-left: 1.5em; padding-right: 1.5em; padding-top: 1.5em; padding-bottom: 0;}
+ .bold {font-weight: bold; }
+
+ div.minind {text-align: justify;}
+ div.condensed, div.condensed1 { line-height: 1.3em; margin-left: 3%; margin-right: 3%; font-size: 95%; }
+ div.condensed1 p {margin-left: 0; padding-left: 2em; text-indent: -2em;}
+ div.condensed span.sidenote {font-size: 90%}
+
+ div.list {margin-left: 0;}
+ div.list p {padding-left: 6em; text-indent: -3em;}
+
+ .pt05 {padding-top: 0.5em;}
+ .pt2 {padding-top: 2em;}
+ .ptb1 {padding-top: 1em; padding-bottom: 1em;}
+
+ </style>
+ </head>
+<body>
+
+
+<pre>
+
+The Project Gutenberg EBook of Encyclopaedia Britannica, 11th Edition,
+Volume 6, Slice 8, by Various
+
+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
+
+
+Title: Encyclopaedia Britannica, 11th Edition, Volume 6, Slice 8
+ "Conduction, Electric"
+
+Author: Various
+
+Release Date: April 19, 2010 [EBook #32063]
+
+Language: English
+
+Character set encoding: ISO-8859-1
+
+*** START OF THIS PROJECT GUTENBERG EBOOK ENCYC. BRITANNICA, VOL 6, SL 8 ***
+
+
+
+
+Produced by Marius Masi, Don Kretz, Juliet Sutherland, and
+the Online Distributed Proofreading Team at
+https://www.pgdp.net
+
+
+
+
+
+
+</pre>
+
+
+
+<table border="0" cellpadding="10" style="background-color: #dcdcdc; color: #696969; " summary="Transcriber's note">
+<tr>
+<td style="width:25%; vertical-align:top">
+Transcriber's note:
+</td>
+<td class="norm">
+A few typographical errors have been corrected. They
+appear in the text <span class="correction" title="explanation will pop up">like this</span>, and the
+explanation will appear when the mouse pointer is moved over the marked
+passage. Sections in Greek will yield a transliteration
+when the pointer is moved over them, and words using diacritic characters in the
+Latin Extended Additional block, which may not display in some fonts or browsers, will
+display an unaccented version. <br /><br />
+<a name="artlinks">Links to other EB articles:</a> Links to articles residing in other EB volumes will
+be made available when the respective volumes are introduced online.
+</td>
+</tr>
+</table>
+<div style="padding-top: 3em; ">&nbsp;</div>
+
+
+<h2>THE ENCYCLOP&AElig;DIA BRITANNICA</h2>
+
+<h2>A DICTIONARY OF ARTS, SCIENCES, LITERATURE AND GENERAL INFORMATION</h2>
+
+<h3>ELEVENTH EDITION</h3>
+<div style="padding-top: 3em; ">&nbsp;</div>
+
+
+<hr class="full" />
+<h3>VOLUME VI SLICE VIII<br /><br />
+Conduction, Electric</h3>
+<hr class="full" />
+<div style="padding-top: 3em; ">&nbsp;</div>
+
+<p class="center1" style="font-size: 180%;">Articles in This Slice</p>
+<p class="pt2"><a href="#ar1">CONDUCTION, ELECTRIC</a></p>
+<div style="padding-top: 3em; ">&nbsp;</div>
+
+<hr class="art" />
+<p><span class="pagenum"><a name="page855" id="page855"></a>855</span></p>
+
+<p><span class="bold"> <a name="ar1" id="ar1">CONDUCTION, ELECTRIC.</a></span> The electric conductivity of a
+substance is that property in virtue of which all its parts come
+spontaneously to the same electric potential if the substance is
+kept free from the operation of electric force. Accordingly, the
+reciprocal quality, electric resistivity, may be defined as a
+quality of a substance in virtue of which a difference of potential
+can exist between different portions of the body when these are
+in contact with some constant source of electromotive force, in
+such a manner as to form part of an electric circuit.</p>
+
+<p>All material substances possess in some degree, large or small,
+electric conductivity, and may for the sake of convenience be
+broadly divided into five classes in this respect. Between these,
+however, there is no sharply-marked dividing line, and the
+classification must therefore be accepted as a more or less
+arbitrary one. These divisions are: (1) metallic conductors,
+(2) non-metallic conductors, (3) dielectric conductors, (4) electrolytic
+conductors, (5) gaseous conductors. The first class comprises
+all metallic substances, and those mixtures or combinations
+of metallic substances known as alloys. The second includes
+such non-metallic bodies as carbon, silicon, many of the oxides
+and peroxides of the metals, and probably also some oxides of
+the non-metals, sulphides and selenides. Many of these substances,
+for instance carbon and silicon, are well-known to have
+the property of existing in several allotropic forms, and in some
+of these conditions, so far from being fairly good conductors,
+they may be almost perfect non-conductors. An example of
+this is seen in the case of carbon in its three allotropic conditions&mdash;charcoal,
+graphite and diamond. As charcoal it possesses a
+fairly well-marked but not very high conductivity in comparison
+with metals; as graphite, a conductivity about one-four-hundredth
+of that of iron; but as diamond so little conductivity
+that the substance is included amongst insulators or non-conductors.
+The third class includes those substances which are
+generally called insulators or non-conductors, but which are
+better denominated dielectric conductors; it comprises such
+solid substances as mica, ebonite, shellac, india-rubber, gutta-percha,
+paraffin, and a large number of liquids, chiefly hydrocarbons.
+These substances differ greatly in insulating power,
+and according as the conductivity is more or less marked, they
+are spoken of as bad or good insulators. Amongst the latter
+many of the liquid gases hold a high position. Thus, liquid
+oxygen and liquid air have been shown by Sir James Dewar
+to be almost perfect non-conductors of electricity.</p>
+
+<p>The behaviour of substances which fall into these three classes
+is discussed below in section I., dealing with metallic conduction.</p>
+
+<p>The fourth class, namely the electrolytic conductors comprises
+all those substances which undergo chemical decomposition
+when they form part of an electric circuit traversed by an
+electric current. They are discussed in section II., dealing with
+electrolytic conduction.</p>
+
+<p>The fifth and last class of conductors includes the gases. The
+conditions under which this class of substance becomes possessed
+of electric conductivity are considered in section III., on conduction
+in gases.</p>
+
+<p>In connexion with metallic conductors, it is a fact of great
+interest and considerable practical importance, that, although
+the majority of metals when in a finely divided or powdered
+condition are practically non-conductors, a mass of metallic
+powder or filings may be made to pass suddenly into a conductive
+condition by being exposed to the influence of an electric wave.
+The same is true of the loose contact of two metallic conductors.
+Thus if a steel point, such as a needle, presses very lightly
+against a metallic plate, say of aluminium, it is found that this
+metallic contact, if carefully adjusted, is non-conductive, but
+that if an electric wave is created anywhere in the neighbourhood,
+this non-conducting contact passes into a conductive state.
+This fact, investigated and discovered independently by D. E.
+Hughes, C. Onesti, E. Branly, O. J. Lodge and others, is applied
+in the construction of the &ldquo;coherer,&rdquo; or sensitive tube employed
+as a detector or receiver in that form of &ldquo;wireless telegraphy&rdquo;
+chiefly developed by Marconi. Further references to it are
+made in the articles <span class="sc"><a href="#artlinks">Electric Waves</a></span> and <span class="sc"><a href="#artlinks">Telegraphy</a></span>:
+<i>Wireless</i>.</p>
+
+<div class="condensed">
+<p><i>International Ohm.</i>&mdash;The practical unit of electrical resistance
+was legally defined in Great Britain by the authority of the queen
+in council in 1894, as the &ldquo;resistance offered to an invariable electric
+current by a column of mercury at the temperature of melting ice,
+14.4521 grammes in mass, of a constant cross-sectional area, and a
+length 106.3 centimetres.&rdquo; The same unit has been also legalized
+as a standard in France, Germany and the United States, and is
+denominated the &ldquo;International or Standard Ohm.&rdquo; It is intended
+to represent as nearly as possible a resistance equal to 10° absolute
+C.G.S. units of electric resistance. Convenient multiples and subdivisions
+of the ohm are the microhm and the megohm, the former
+being a millionth part of an ohm, and the latter a million ohms.
+The resistivity of substances is then numerically expressed by stating
+the resistance of one cubic centimetre of the substance taken between
+opposed faces, and expressed in ohms, microhms or megohms, as
+may be most convenient. The reciprocal of the ohm is called the
+mho, which is the unit of conductivity, and is defined as the conductivity
+of a substance whose resistance is one ohm. The absolute
+unit of conductivity is the conductivity of a substance whose resistivity
+is one absolute C.G.S. unit, or one-thousandth-millionth part
+of an ohm. Resistivity is a quality in which material substances
+differ very widely. The metals and alloys, broadly speaking, are
+good conductors, and their resistivity is conveniently expressed in
+microhms per cubic centimetre, or in absolute C.G.S. units. Very
+small differences in density and in chemical purity make, however,
+immense differences in electric resistivity; hence the values given
+by different experimentalists for the resistivity of known metals
+differ to a considerable extent.</p>
+</div>
+
+<p class="center1">I. <span class="sc">Conduction in Solids</span></p>
+
+<p>It is found convenient to express the resistivity of metals in two
+different ways: (1) We may state the resistivity of one cubic
+centimetre of the material in microhms or absolute units taken
+between opposed faces. This is called the <i>volume-resistivity</i>; (2)
+we may express the resistivity by stating the resistance in ohms
+offered by a wire of the material in question of uniform cross-section
+one metre in length, and one gramme in weight. This
+numerical measure of the resistivity is called the <i>mass-resistivity</i>.
+The mass-resistivity of a body is connected with its volume-resistivity
+and the density of the material in the following
+manner:&mdash;The mass-resistivity, expressed in microhms per metre-gramme,
+divided by 10 times the density is numerically equal to
+the volume-resistivity per centimetre-cube in absolute C.G.S.
+units. The mass-resistivity per metre-gramme can always be
+obtained by measuring the resistance and the mass of any wire of
+<span class="pagenum"><a name="page856" id="page856"></a>856</span>
+uniform cross-section of which the length is known, and if the
+density of the substance is then measured, the volume-resistivity
+can be immediately calculated.</p>
+
+<div class="condensed">
+<p>If R is the resistance in ohms of a wire of length l, uniform cross-section
+s, and density d, then taking &rho; for the volume-resistivity we
+have 10<span class="sp">9</span>R = &rho;l/s; but lsd = M, where M is the mass of the wire.
+Hence 10<span class="sp">9</span>R = &rho;dl<span class="sp">2</span>/M. If l = 100 and M = 1, then R = &rho;&prime;= resistivity in
+ohms per metre-gramme, and 10<span class="sp">9</span>&rho;&prime; = 10,000d&rho;, or &rho; = 10<span class="sp">5</span>&rho;&prime;/d, and
+&rho;&prime; = 10,000MR/l<span class="sp">2</span>.</p>
+
+<p>The following rules, therefore, are useful in connexion with
+these measurements. To obtain the mass-resistivity per metre-gramme
+of a substance in the form of a uniform metallic wire:&mdash;Multiply
+together 10,000 times the mass in grammes and the total
+resistance in ohms, and then divide by the square of the length in
+centimetres. Again, to obtain the volume-resistivity in C.G.S. units
+per centimetre-cube, the rule is to multiply the mass-resistivity in
+ohms by 100,000 and divide by the density. These rules, of course,
+apply only to wires of uniform cross-section. In the following
+Tables I., II. and III. are given the mass and volume resistivity of
+ordinary metals and certain alloys expressed in terms of the international
+ohm or the absolute C.G.S. unit of resistance, the values
+being calculated from the experiments of A. Matthiessen (1831-1870)
+between 1860 and 1865, and from later results obtained by
+J. A. Fleming and Sir James Dewar in 1893.</p>
+
+<p class="tab pt05"><span class="sc">Table I.</span>&mdash;<i>Electric Mass-Resistivity of Various Metals at 0° C., or
+Resistance per Metre-gramme in International Ohms at 0° C.</i>
+(Matthiessen.)</p>
+
+<table class="ws" summary="Contents">
+<tr><td class="tccm allb">Metal.</td>
+ <td class="tccm allb">Resistance at 0° C.<br />in International Ohms<br />of a Wire 1 Metre long<br />and Weighing 1 Gramme.</td>
+ <td class="tccm allb">Approximate Temperature<br />Coefficient near 20° C.</td></tr>
+
+<tr><td class="tcl lb rb">Silver (annealed)</td> <td class="tcr rb">.1523&emsp;&emsp;&emsp;</td> <td class="tcc rb">0.00377</td></tr>
+<tr><td class="tcl lb rb">Silver (hard-drawn)</td> <td class="tcr rb">.1657&emsp;&emsp;&emsp;</td> <td class="tcc rb">..</td></tr>
+<tr><td class="tcl lb rb">Copper (annealed)</td> <td class="tcr rb">.1421&emsp;&emsp;&emsp;</td> <td class="tcc rb">0.00388</td></tr>
+<tr><td class="tcl lb rb">Copper (hard-drawn)</td> <td class="tcl rb">.1449 (Matthiessen&rsquo;s Standard)</td> <td class="tcc rb">..</td></tr>
+<tr><td class="tcl lb rb">Gold (annealed)</td> <td class="tcr rb">.4025&emsp;&emsp;&emsp;</td> <td class="tcc rb">0.00365</td></tr>
+<tr><td class="tcl lb rb">Gold (hard-drawn)</td> <td class="tcr rb">.4094&emsp;&emsp;&emsp;</td> <td class="tcc rb">..</td></tr>
+<tr><td class="tcl lb rb">Aluminium (annealed)</td> <td class="tcr rb">.0757&emsp;&emsp;&emsp;</td> <td class="tcc rb">..</td></tr>
+<tr><td class="tcl lb rb">Zinc (pressed)</td> <td class="tcr rb">.4013&emsp;&emsp;&emsp;</td> <td class="tcc rb">..</td></tr>
+<tr><td class="tcl lb rb">Platinum (annealed)</td> <td class="tcr rb">1.9337&emsp;&emsp;&emsp;</td> <td class="tcc rb">..</td></tr>
+<tr><td class="tcl lb rb">Iron (annealed)</td> <td class="tcr rb">.765&emsp;&emsp;&emsp;</td> <td class="tcc rb">..</td></tr>
+<tr><td class="tcl lb rb">Nickel (annealed)</td> <td class="tcr rb">1.058<a name="FnAnchor_1" id="FnAnchor_1" href="#Footnote_1"><span class="sp">1</span></a>&emsp;&emsp;&emsp;</td> <td class="tcc rb">..</td></tr>
+<tr><td class="tcl lb rb">Tin (pressed)</td> <td class="tcr rb">.9618&emsp;&emsp;&emsp;</td> <td class="tcc rb">0.00365</td></tr>
+<tr><td class="tcl lb rb">Lead (pressed)</td> <td class="tcr rb">2.2268&emsp;&emsp;&emsp;</td> <td class="tcc rb">0.00387</td></tr>
+<tr><td class="tcl lb rb">Antimony (pressed)</td> <td class="tcr rb">2.3787&emsp;&emsp;&emsp;</td> <td class="tcc rb">0.00389</td></tr>
+<tr><td class="tcl lb rb">Bismuth (pressed)</td> <td class="tcr rb">12.8554<a href="#Footnote_1"><span class="sp">1</span></a>&emsp;&emsp;&emsp;</td> <td class="tcc rb">0.00354</td></tr>
+<tr><td class="tcl lb rb bb">Mercury (liquid)</td> <td class="tcr rb bb">12.885<a name="FnAnchor_2" id="FnAnchor_2" href="#Footnote_2"><span class="sp">2</span></a>&emsp;&emsp;&emsp;</td> <td class="tcc rb bb">0.00072</td></tr>
+</table>
+
+<p>The data commonly used for calculating metallic resistivities
+were obtained by A. Matthiessen, and his results are set out in the
+Table II. which is taken from Cantor lectures given by Fleeming
+Jenkin in 1866 at or about the date when the researches were made.
+The figures given by Jenkin have, however, been reduced to international
+ohms and C.G.S. units by multiplying by (&pi;/4)&times;0.9866 &times;
+10<span class="sp">5</span> = 77,485.</p>
+
+<p>Subsequently numerous determinations of the resistivity of various
+pure metals were made by Fleming and Dewar, whose results are
+set out in Table III.</p>
+
+<p class="tab pt05"><span class="sc">Table II.</span>&mdash;<i>Electric Volume-Resistivity of Various Metals at 0° C.,
+or Resistance per Centimetre-cube in C.G.S. Units at 0° C.</i></p>
+
+<table class="ws" summary="Contents">
+<tr><td class="tccm allb">Metal.</td>
+ <td class="tccm allb">Volume-Resistivity.<br />at 0° C. in C.G.S.<br />Units</td></tr>
+
+<tr><td class="tcl lb rb">Silver (annealed)</td> <td class="tcr rb">1,502&emsp;&emsp;</td></tr>
+<tr><td class="tcl lb rb">Silver (hard-drawn)</td> <td class="tcr rb">1,629&emsp;&emsp;</td></tr>
+<tr><td class="tcl lb rb">Copper (annealed)</td> <td class="tcr rb">1,594&emsp;&emsp;</td></tr>
+<tr><td class="tcl lb rb">Copper (hard-drawn)</td> <td class="tcr rb">1,630<a name="FnAnchor_3" id="FnAnchor_3" href="#Footnote_3"><span class="sp">3</span></a>&emsp;&emsp;</td></tr>
+<tr><td class="tcl lb rb">Gold (annealed)</td> <td class="tcr rb">2,052&emsp;&emsp;</td></tr>
+<tr><td class="tcl lb rb">Gold (hard-drawn)</td> <td class="tcr rb">2,090&emsp;&emsp;</td></tr>
+<tr><td class="tcl lb rb">Aluminium (annealed)</td> <td class="tcr rb">3,006&emsp;&emsp;</td></tr>
+<tr><td class="tcl lb rb">Zinc (pressed)</td> <td class="tcr rb">5,621&emsp;&emsp;</td></tr>
+<tr><td class="tcl lb rb">Platinum (annealed)</td> <td class="tcr rb">9,035&emsp;&emsp;</td></tr>
+<tr><td class="tcl lb rb">Iron (annealed)</td> <td class="tcr rb">10,568&emsp;&emsp;</td></tr>
+<tr><td class="tcl lb rb">Nickel (annealed)</td> <td class="tcr rb">12,429<a name="FnAnchor_4" id="FnAnchor_4" href="#Footnote_4"><span class="sp">4</span></a>&emsp;&emsp;</td></tr>
+<tr><td class="tcl lb rb">Tin (pressed)</td> <td class="tcr rb">13,178&emsp;&emsp;</td></tr>
+<tr><td class="tcl lb rb">Lead (pressed)</td> <td class="tcr rb">19,580&emsp;&emsp;</td></tr>
+<tr><td class="tcl lb rb">Antimony (pressed)</td> <td class="tcr rb">35,418&emsp;&emsp;</td></tr>
+<tr><td class="tcl lb rb">Bismuth (pressed)</td> <td class="tcr rb">130,872&emsp;&emsp;</td></tr>
+<tr><td class="tcl lb rb bb">Mercury (liquid)</td> <td class="tcr rb bb">94,896<a name="FnAnchor_5" id="FnAnchor_5" href="#Footnote_5"><span class="sp">5</span></a>&emsp;&emsp;</td></tr>
+</table>
+
+<p><i>Resistivity of Mercury.</i>&mdash;The volume-resistivity of pure mercury
+is a very important electric constant, and since 1880 many of the
+most competent experimentalists have directed their attention to
+the determination of its value. The experimental process has
+usually been to fill a glass tube of known dimensions, having large
+cup-like extensions at the ends, with pure mercury, and determine
+the absolute resistance of this column of metal. For the practical
+details of this method the following references may be consulted:&mdash;&ldquo;The
+Specific Resistance of Mercury,&rdquo; Lord Rayleigh and Mrs Sidgwick,
+<i>Phil. Trans.</i>, 1883, part i. p. 173, and R. T. Glazebrook, <i>Phil.
+Mag.</i>, 1885, p. 20; &ldquo;On the Specific Resistance of Mercury,&rdquo; R. T.
+Glazebrook and T. C. Fitzpatrick, <i>Phil. Trans.</i>, 1888, p. 179, or <i>Proc.
+Roy. Soc.</i>, 1888, p. 44, or <i>Electrician</i>, 1888, 21, p. 538; &ldquo;Recent
+Determinations of the Absolute Resistance of Mercury,&rdquo; R. T. Glazebrook,
+<i>Electrician</i>, 1890, 25, pp. 543 and 588. Also see J. V. Jones,
+&ldquo;On the Determination of the Specific Resistance of Mercury in
+Absolute Measure,&rdquo; <i>Phil. Trans.</i>, 1891, A, p. 2. Table IV. gives
+the values of the volume-resistivity of mercury as determined by
+various observers, the constant being expressed (a) in terms of the
+resistance in ohms of a column of mercury one millimetre in cross-section
+and 100 centimetres in length, taken at 0° C.; and (b) in terms
+of the length in centimetres of a column of mercury one square millimetre
+in cross-section taken at 0° C. The result of all the most careful
+determinations has been to show that the resistivity of pure
+mercury at 0° C. is about 94,070 C.G.S. electromagnetic units of
+resistance, and that a column of mercury 106.3 centimetres in length
+having a cross-sectional area of one square millimetre would have a
+resistance at 0° C. of one international ohm. These values have
+accordingly been accepted as the official and recognized values for
+the specific resistance of mercury, and the definition of the ohm.
+The table also states the methods which have been adopted by the
+different observers for obtaining the absolute value of the resistance
+of a known column of mercury, or of a resistance coil afterwards
+compared with a known column of mercury. A column of figures
+is added showing the value in fractions of an international ohm of
+the British Association Unit (B.A.U.), formerly supposed to represent
+the true ohm. The real value of the B.A.U. is now taken as .9866
+of an international ohm.</p>
+
+<p class="tab pt05"><span class="sc">Table III.</span>&mdash;<i>Electric Volume-Resistivity of Various Metals at 0° C.,
+or Resistance per Centimetre-cube at 0° C. in C.G.S. Units.</i>
+(Fleming and Dewar, <i>Phil. Mag.</i>, September 1893.)</p>
+
+<table class="ws" summary="Contents">
+<tr><td class="tccm allb">Metal.</td>
+ <td class="tccm allb">Resistance at 0° C.<br />per Centimetre-cube<br />in C.G.S. Units.</td>
+ <td class="tccm allb">Mean Temperature<br />Coefficient between<br />0° C. and 100° C.</td></tr>
+
+<tr><td class="tcl lb rb">Silver (electrolytic and well annealed)<a name="FnAnchor_6" id="FnAnchor_6" href="#Footnote_6"><span class="sp">6</span></a></td> <td class="tcr rb">1,468&emsp;&emsp;</td> <td class="tcl rb">&emsp;&emsp;0.00400</td></tr>
+<tr><td class="tcl lb rb">Copper (electrolytic and well annealed)<a href="#Footnote_6"><span class="sp">6</span></a></td> <td class="tcr rb">1,561&emsp;&emsp;</td> <td class="tcl rb">&emsp;&emsp;0.00428</td></tr>
+<tr><td class="tcl lb rb">Gold (annealed)</td> <td class="tcr rb">2,197&emsp;&emsp;</td> <td class="tcl rb">&emsp;&emsp;0.00377</td></tr>
+<tr><td class="tcl lb rb">Aluminium (annealed)</td> <td class="tcr rb">2,665&emsp;&emsp;</td> <td class="tcl rb">&emsp;&emsp;0.00435</td></tr>
+<tr><td class="tcl lb rb">Magnesium (pressed)</td> <td class="tcr rb">4,355&emsp;&emsp;</td> <td class="tcl rb">&emsp;&emsp;0.00381</td></tr>
+<tr><td class="tcl lb rb">Zinc</td> <td class="tcr rb">5,751&emsp;&emsp;</td> <td class="tcl rb">&emsp;&emsp;0.00406</td></tr>
+<tr><td class="tcl lb rb">Nickel (electrolytic)<a href="#Footnote_6"><span class="sp">6</span></a></td> <td class="tcr rb">6,935&emsp;&emsp;</td> <td class="tcl rb">&emsp;&emsp;0.00618</td></tr>
+<tr><td class="tcl lb rb">Iron (annealed)</td> <td class="tcr rb">9,065&emsp;&emsp;</td> <td class="tcl rb">&emsp;&emsp;0.00625</td></tr>
+<tr><td class="tcl lb rb">Cadmium</td> <td class="tcr rb">10,023&emsp;&emsp;</td> <td class="tcl rb">&emsp;&emsp;0.00419</td></tr>
+<tr><td class="tcl lb rb">Palladium</td> <td class="tcr rb">10,219&emsp;&emsp;</td> <td class="tcl rb">&emsp;&emsp;0.00354</td></tr>
+<tr><td class="tcl lb rb">Platinum (annealed)</td> <td class="tcr rb">10,917&emsp;&emsp;</td> <td class="tcl rb">&emsp;&emsp;0.003669</td></tr>
+<tr><td class="tcl lb rb">Tin (pressed)</td> <td class="tcr rb">13,048&emsp;&emsp;</td> <td class="tcl rb">&emsp;&emsp;0.00440</td></tr>
+<tr><td class="tcl lb rb">Thallium (pressed)</td> <td class="tcr rb">17,633&emsp;&emsp;</td> <td class="tcl rb">&emsp;&emsp;0.00398</td></tr>
+<tr><td class="tcl lb rb">Lead (pressed)</td> <td class="tcr rb">20,380&emsp;&emsp;</td> <td class="tcl rb">&emsp;&emsp;0.00411</td></tr>
+<tr><td class="tcl lb rb bb">Bismuth (electrolytic)<a name="FnAnchor_7" id="FnAnchor_7" href="#Footnote_7"><span class="sp">7</span></a></td> <td class="tcr rb bb">110,000&emsp;&emsp;</td> <td class="tcl rb bb">&emsp;&emsp;0.00433</td></tr>
+</table>
+
+<p><span class="pagenum"><a name="page857" id="page857"></a>857</span></p>
+
+<p class="tab pt05"><span class="sc">Table IV.</span>&mdash;<i>Determinations of the Absolute Value of the Volume-Resistivity of
+Mercury and the Mercury Equivalent of the Ohm.</i></p>
+
+<table class="ws" summary="Contents">
+<tr><td class="tccm allb">Observer.</td> <td class="tccm allb">Date.</td> <td class="tccm allb">Method.</td>
+ <td class="tccm allb">Value of B.A.U.<br />in Ohms.</td>
+ <td class="tccm allb">Value of 100<br />Centimetres of<br />Mercury in<br />Ohms.</td>
+ <td class="tccm allb">Value of Ohm<br />in Centimetres<br />of Mercury.</td></tr>
+
+<tr><td class="tcl lb rb">Lord Rayleigh</td> <td class="tcc rb">1882</td> <td class="tcl rb">Rotating coil</td> <td class="tcc rb">.98651</td> <td class="tcc rb">.94133</td> <td class="tcc rb">106.31</td></tr>
+<tr><td class="tcl lb rb">Lord Rayleigh</td> <td class="tcc rb">1883</td> <td class="tcl rb">Lorenz method</td> <td class="tcc rb">.98677</td> <td class="tcc rb">..</td> <td class="tcc rb">106.27</td></tr>
+<tr><td class="tcl lb rb">G. Wiedemann</td> <td class="tcc rb">1884</td> <td class="tcl rb">Rotation through 180°</td> <td class="tcc rb">..</td> <td class="tcc rb">..</td> <td class="tcc rb">106.19</td></tr>
+<tr><td class="tcl lb rb">E. E. N. Mascart</td> <td class="tcc rb">1884</td> <td class="tcl rb">Induced current</td> <td class="tcc rb">.98611</td> <td class="tcc rb">.94096</td> <td class="tcc rb">106.33</td></tr>
+<tr><td class="tcl lb rb">H. A. Rowland</td> <td class="tcc rb">1887</td> <td class="tcl rb">Mean of several methods</td> <td class="tcc rb">.98644</td> <td class="tcc rb">.94071</td> <td class="tcc rb">106.32</td></tr>
+<tr><td class="tcl lb rb">F. Kohlrausch</td> <td class="tcc rb">1887</td> <td class="tcl rb">Damping of magnets</td> <td class="tcc rb">.98660</td> <td class="tcc rb">.94061</td> <td class="tcc rb">106.32</td></tr>
+<tr><td class="tcl lb rb">R. T. Glazebrook</td> <td class="tcc rb">1882/8</td> <td class="tcl rb">Induced currents</td> <td class="tcc rb">.98665</td> <td class="tcc rb">.94074</td> <td class="tcc rb">106.29</td></tr>
+<tr><td class="tcl lb rb">Wuilleumeier</td> <td class="tcc rb">1890</td> <td class="tcl rb">&nbsp;</td> <td class="tcc rb">.98686</td> <td class="tcc rb">.94077</td> <td class="tcc rb">106.31</td></tr>
+<tr><td class="tcl lb rb">Duncan and Wilkes</td> <td class="tcc rb">1890</td> <td class="tcl rb">Lorenz</td> <td class="tcc rb">.98634</td> <td class="tcc rb">.94067</td> <td class="tcc rb">106.34</td></tr>
+<tr><td class="tcl lb rb">J. V. Jones</td> <td class="tcc rb">1891</td> <td class="tcl rb">Lorenz</td> <td class="tcc rb bb">..</td> <td class="tcc rb bb">.94067</td> <td class="tcc rb bb">106.31</td></tr>
+
+<tr><td class="tcl lb rb ptb1">&nbsp;</td> <td class="tcc rb ptb1">&nbsp;</td> <td class="tcc ptb1">Mean value</td> <td class="tcc ptb1">.98653</td> <td class="tcc rb">&nbsp;</td> <td class="tcc rb ptb1">&nbsp;</td></tr>
+<tr><td class="tcl lb rb">Streker</td> <td class="tcc rb">1885</td> <td class="tcl rb">An absolute determination</td> <td class="tcc rb">&nbsp;</td> <td class="tcc rb">.94056</td> <td class="tcc rb">106.32</td></tr>
+<tr><td class="tcl lb rb">Hutchinson</td> <td class="tcc rb">1888</td> <td class="tcl rb">of resistance was not</td> <td class="tcc rb">&nbsp;</td> <td class="tcc rb">.94074</td> <td class="tcc rb">106.30</td></tr>
+<tr><td class="tcl lb rb">E. Salvioni</td> <td class="tcc rb">1890</td> <td class="tcl rb">made. The value .98656</td> <td class="tcc rb">&nbsp;</td> <td class="tcc rb">.94054</td> <td class="tcc rb">106.33</td></tr>
+<tr><td class="tcl lb rb">E. Salvioni</td> <td class="tcc rb">..</td> <td class="tcl rb">value .98656 has been used</td> <td class="tcc rb">&nbsp;</td> <td class="tcc rb bb">.94076</td> <td class="tcc rb bb">106.30</td></tr>
+
+<tr><td class="tcl lb rb ptb1">&nbsp;</td> <td class="tcc rb ptb1">&nbsp;</td> <td class="tcc ptb1" colspan="2">Mean value</td> <td class="tcc rb ptb1">.94076</td> <td class="tcc rb ptb1">106.31</td></tr>
+<tr><td class="tcl lb rb">H. F. Weber</td> <td class="tcc rb">1884</td> <td class="tcl rb">Induced current</td> <td class="tcl rb" colspan="2">Absolute measurements</td> <td class="tcc rb">105.37</td></tr>
+<tr><td class="tcl lb rb">H. F. Weber</td> <td class="tcc rb">..</td> <td class="tcl rb">Rotating coil</td> <td class="tcl rb" colspan="2">&emsp;compared with German</td> <td class="tcc rb">106.16</td></tr>
+<tr><td class="tcl lb rb">A. Roiti</td> <td class="tcc rb">1884</td> <td class="tcl rb">Mean effect of induced current</td> <td class="tcl rb" colspan="2">&emsp;silver wire coils issued by</td> <td class="tcc rb">105.89</td></tr>
+<tr><td class="tcl lb rb">F. Himstedt</td> <td class="tcc rb">1885</td> <td class="tcl rb">&nbsp;</td> <td class="tcl rb" colspan="2">&emsp;Siemens and Streker</td> <td class="tcc rb">105.98</td></tr>
+<tr><td class="tcl lb rb">F. E. Dorn</td> <td class="tcc rb">1889</td> <td class="tcl rb">Damping of a magnet</td> <td class="tcl rb" colspan="2">&nbsp;</td> <td class="tcc rb">106.24</td></tr>
+<tr><td class="tcl lb rb">Wild</td> <td class="tcc rb">1883</td> <td class="tcl rb">Damping of a magnet</td> <td class="tcl rb" colspan="2">&nbsp;</td> <td class="tcc rb">106.03</td></tr>
+<tr><td class="tcl lb rb bb">L. V. Lorenz</td> <td class="tcc rb bb">1885</td> <td class="tcl rb bb">Lorenz method</td> <td class="tcl rb bb" colspan="2">&nbsp;</td> <td class="tcc rb bb">105.93</td></tr>
+</table>
+
+<p>For a critical discussion of the methods which have
+been adopted in the absolute determination of the
+resistivity of mercury, and the value of the British
+Association unit of resistance, the reader may be referred
+to the <i>British Association Reports</i> for 1890 and
+1892 (<i>Report of Electrical Standards Committee</i>), and to
+the <i>Electrician</i>, 25, p. 456, and 29, p. 462. A discussion
+of the relative value of the results obtained between
+1882 and 1890 was given by R. T. Glazebrook
+in a paper presented to the British Association at
+Leeds, 1890.</p>
+
+<p><i>Resistivity of Copper.</i>&mdash;In connexion with electro-technical
+work the determination of the conductivity
+or resistivity values of annealed and hard-drawn copper
+wire at standard temperatures is a very important
+matter. Matthiessen devoted considerable attention
+to this subject between the years 1860 and 1864 (see
+<i>Phil. Trans.</i>, 1860, p. 150), and since that time much
+additional work has been carried out. Matthiessen&rsquo;s
+value, known as <i>Matthiessen&rsquo;s Standard</i>, for the mass-resistivity
+of pure hard-drawn copper wire, is the
+resistance of a wire of pure hard-drawn copper one
+metre long and weighing one gramme, and this is
+equal to 0.14493 international ohms at 0° C. For
+many purposes it is more convenient to express temperature
+in Fahrenheit degrees, and the recommendation
+of the 1899 committee on copper conductors<a name="FnAnchor_8" id="FnAnchor_8" href="#Footnote_8"><span class="sp">8</span></a> is as
+follows:&mdash;&ldquo;Matthiessen&rsquo;s standard for hard-drawn conductivity
+commercial copper shall be considered to be
+the resistance of a wire of pure hard-drawn copper one
+metre long, weighing one gramme which at
+60° F. is 0.153858 international ohms.&rdquo;
+Matthiessen also measured the mass-resistivity
+of annealed copper, and found that its
+conductivity is greater than that of hard-drawn
+copper by about 2.25% to 2.5%
+As annealed copper may vary considerably
+in its state of annealing, and is always
+somewhat hardened by bending and winding,
+it is found in practice that the resistivity of
+commercial annealed copper is about 1¼%
+less than that of hard-drawn copper. The
+standard now accepted for such copper, on
+the recommendation of the 1899 Committee,
+is a wire of pure annealed copper one metre
+long, weighing one gramme, whose resistance
+at 0° C. is 0.1421 international ohms, or at
+60° F., 0.150822 international ohms. The
+specific gravity of copper varies from about
+8.89 to 8.95, and the standard value accepted
+for high conductivity commercial copper is
+8.912, corresponding to a weight of 555 lb
+per cubic foot at 60° F. Hence the volume-resistivity
+of pure annealed copper at 0° C. is
+1.594 microhms per c.c., or 1594 C.G.S. units,
+and that of pure hard-drawn copper at 0° C. is
+1.626 microhms per c.c., or 1626 C.G.S. units.
+Since Matthiessen&rsquo;s researches, the most careful
+scientific investigation on the conductivity
+of copper is that of T. C. Fitzpatrick,
+carried out in 1890. (<i>Brit. Assoc. Report</i>, 1890,
+Appendix 3, p. 120.) Fitzpatrick confirmed
+Matthiessen&rsquo;s chief result, and obtained values
+for the resistivity of hard-drawn copper which,
+when corrected for temperature variation, are
+in entire agreement with those of Matthiessen
+at the same temperature.</p>
+</div>
+
+<p>The volume resistivity of alloys is, generally
+speaking, much higher than that of pure
+metals. Table V. shows the volume resistivity
+at 0° C. of a number of well-known
+alloys, with their chemical composition.</p>
+
+<div class="condensed">
+
+<p class="tab pt05"><span class="sc">Table V.</span>&mdash;<i>Volume-Resistivity of Alloys of known Composition at 0° C. in C.G.S.
+Units per Centimetre-cube. Mean Temperature Coefficients taken at 15° C.</i>
+(Fleming and Dewar.)</p>
+
+<table class="ws" summary="Contents">
+<tr><td class="tccm allb">Alloys.</td>
+ <td class="tccm allb">Resistivity<br />at 0° C.</td>
+ <td class="tccm allb">Temperature<br />Coefficient<br />at 15° C.</td>
+ <td class="tccm allb">Composition in per cents.</td></tr>
+
+<tr><td class="tcl lb rb">Platinum-silver</td> <td class="tcr rb">31,582</td> <td class="tcl rb">&emsp;.000243</td> <td class="tcl rb">Pt 33%, Ag 66%</td></tr>
+<tr><td class="tcl lb rb">Platinum-iridium</td> <td class="tcr rb">30,896</td> <td class="tcl rb">&emsp;.000822</td> <td class="tcl rb">Pt 80%, Ir 20%</td></tr>
+<tr><td class="tcl lb rb">Platinum-rhodium</td> <td class="tcr rb">21,142</td> <td class="tcl rb">&emsp;.00143</td> <td class="tcl rb">Pt 90%, Rd 10%</td></tr>
+<tr><td class="tcl lb rb">Gold-silver</td> <td class="tcr rb">6,280</td> <td class="tcl rb">&emsp;.00124</td> <td class="tcl rb">Au 90%, Ag 10%</td></tr>
+<tr><td class="tcl lb rb">Manganese-steel</td> <td class="tcr rb">67,148</td> <td class="tcl rb">&emsp;.00127</td> <td class="tcl rb">Mn 12%, Fe 78%</td></tr>
+<tr><td class="tcl lb rb">Nickel-steel</td> <td class="tcr rb">29,452</td> <td class="tcl rb">&emsp;.00201</td> <td class="tcl rb">Ni 4.35%, remaining percentage</td></tr>
+<tr><td class="tcl lb rb">&nbsp;</td> <td class="tcr rb">&nbsp;</td> <td class="tcl rb">&nbsp;</td> <td class="tcl rb">&emsp;chiefly iron, but uncertain</td></tr>
+<tr><td class="tcl lb rb">German silver</td> <td class="tcr rb">29,982</td> <td class="tcl rb">&emsp;.000273</td> <td class="tcl rb">Cu<span class="su">5</span>Zn<span class="su">3</span>Ni<span class="su">2</span></td></tr>
+<tr><td class="tcl lb rb">Platinoid<a name="FnAnchor_9" id="FnAnchor_9" href="#Footnote_9"><span class="sp">9</span></a></td> <td class="tcr rb">41,731</td> <td class="tcl rb">&emsp;.00031</td> <td class="tcl rb">&nbsp;</td></tr>
+<tr><td class="tcl lb rb">Manganin</td> <td class="tcr rb">46,678</td> <td class="tcl rb">&emsp;.0000</td> <td class="tcl rb">Cu 84%, Mn 12%, Ni 4%</td></tr>
+<tr><td class="tcl lb rb">Aluminium-silver</td> <td class="tcr rb">4,641</td> <td class="tcl rb">&emsp;.00238</td> <td class="tcl rb">Al 94%, Ag 6%</td></tr>
+<tr><td class="tcl lb rb">Aluminium-copper</td> <td class="tcr rb">2,904</td> <td class="tcl rb">&emsp;.00381</td> <td class="tcl rb">Al 94%, Cu 6%</td></tr>
+<tr><td class="tcl lb rb">Copper-aluminium</td> <td class="tcr rb">8,847</td> <td class="tcl rb">&emsp;.000897</td> <td class="tcl rb">Cu 97%, Al 3%</td></tr>
+<tr><td class="tcl lb rb">Copper-nickel-aluminium</td> <td class="tcr rb">14,912</td> <td class="tcl rb">&emsp;.000643</td> <td class="tcl rb">Cu 87%, Ni 6.5%, Al 6.5%</td></tr>
+<tr><td class="tcl lb rb bb">Titanium-aluminium</td> <td class="tcr rb bb">3,887</td> <td class="tcl rb bb">&emsp;.00290</td> <td class="tcl rb bb">&nbsp;</td></tr>
+</table>
+
+<p>Generally speaking, an alloy having high resistivity has poor
+mechanical qualities, that is to say, its tensile strength and ductility
+are small. It is possible to form alloys having a resistivity as high
+as 100 microhms per cubic centimetre; but, on the other hand, the
+value of an alloy for electro-technical purposes is judged not merely
+by its resistivity, but also by the degree to which its resistivity varies
+with temperature, and by its capability of being easily drawn into
+fine wire of not very small tensile strength. Some pure metals when
+alloyed with a small proportion of another metal do not suffer much
+<span class="pagenum"><a name="page858" id="page858"></a>858</span>
+change in resistivity, but in other cases the resultant alloy has a
+much higher resistivity. Thus an alloy of pure copper with 3% of
+aluminium has a resistivity about 5½ times that of copper; but if
+pure aluminium is alloyed with 6% of copper, the resistivity of the
+product is not more than 20% greater than that of pure aluminium.
+The presence of a very small proportion of a non-metallic element in
+a metallic mass, such as oxygen, sulphur or phosphorus, has a very
+great effect in increasing the resistivity. Certain metallic elements
+also have the same power; thus platinoid has a resistivity 30%
+greater than German silver, though it differs from it merely in
+containing a trace of tungsten.</p>
+</div>
+
+<p>The resistivity of non-metallic conductors is in all cases higher
+than that of any pure metal. The resistivity of carbon, for
+instance, in the forms of charcoal or carbonized organic material
+and graphite, varies from 600 to 6000 microhms per cubic
+centimetre, as shown in Table VI.:&mdash;</p>
+
+<p class="tab pt05"><span class="sc">Table VI.</span>&mdash;<i>Electric Volume-Resistivity in Microhms per
+Centimetre-cube of Various Forms of Carbon at 15° C.</i></p>
+
+<table class="ws" summary="Contents">
+<tr><td class="tccm allb">Substance.</td> <td class="tccm allb">Resistivity.</td></tr>
+
+<tr><td class="tcl lb rb">Arc lamp carbon rod</td> <td class="tcl rb">8000</td></tr>
+<tr><td class="tcl lb rb">Jablochkoff candle carbon</td> <td class="tcl rb">4000</td></tr>
+<tr><td class="tcl lb rb">Carré carbon</td> <td class="tcl rb">3400</td></tr>
+<tr><td class="tcl lb rb">Carbonized bamboo</td> <td class="tcl rb">6000</td></tr>
+<tr><td class="tcl lb rb">Carbonized parchmentized thread</td> <td class="tcl rb">4000 to 5000</td></tr>
+<tr><td class="tcl lb rb">Ordinary carbon filament from glow-lamp</td> <td class="tcl rb">&nbsp;</td></tr>
+<tr><td class="tcl lb rb">&emsp;&ldquo;treated&rdquo; or flashed</td> <td class="tcl rb">2400 to 2500</td></tr>
+<tr><td class="tcl lb rb">Deposited or secondary carbon</td> <td class="tcl rb">&ensp;600 to &ensp;900</td></tr>
+<tr><td class="tcl lb rb bb">Graphite</td> <td class="tcl rb bb">&ensp;400 to &ensp;500</td></tr>
+</table>
+
+<p>The resistivity of liquids is, generally speaking, much higher
+than that of any metals, metallic alloys or non-metallic conductors.
+Thus fused lead chloride, one of the best conducting
+liquids, has a resistivity in its fused condition of 0.376 ohm per
+centimetre-cube, or 376,000 microhms per centimetre-cube,
+whereas that of metallic alloys only in few cases exceeds 100
+microhms per centimetre-cube. The resistivity of solutions of
+metallic salts also varies very largely with the proportion of the
+diluent or solvent, and in some instances, as in the aqueous
+solutions of mineral acids; there is a maximum conductivity
+corresponding to a certain dilution. The resistivity of many
+liquids, such as alcohol, ether, benzene and pure water, is so high,
+in other words, their conductivity is so small, that they are
+practically insulators, and the resistivity can only be appropriately
+expressed in megohms per centimetre-cube.</p>
+
+<p>In Table VII. are given the names of a few of these badly-conducting
+liquids, with the values of their volume-resistivity in
+megohms per centimetre-cube:&mdash;</p>
+
+<p class="tab pt05"><span class="sc">Table VII.</span>&mdash;<i>Electric Volume-Resistivity of Various Badly-Conducting
+Liquids in Megohms per Centimetre-cube.</i></p>
+
+<table class="ws" summary="Contents">
+<tr><td class="tccm allb">Substance.</td>
+ <td class="tccm allb">Resistivity<br />in Megohms<br />per c.c.</td>
+ <td class="tccm allb">Observer.</td></tr>
+
+<tr><td class="tcl lb rb">Ethyl alcohol</td> <td class="tcc rb">0.5</td> <td class="tcl rb">Pfeiffer.</td></tr>
+<tr><td class="tcl lb rb">Ethyl ether</td> <td class="tcc rb">1.175 to 3.760</td> <td class="tcl rb">W. Kohlrausch.</td></tr>
+<tr><td class="tcl lb rb">Benzene</td> <td class="tcc rb">4.700</td> <td class="tcl rb">&nbsp;</td></tr>
+<tr><td class="tcl lb rb">Absolutely pure water approximates probably to</td> <td class="tcc rb">25.0 at 18° C.</td> <td class="tcl rb">Value estimated by F. Kohlrausch</td></tr>
+<tr><td class="tcl lb rb">&nbsp;</td> <td class="rb">&nbsp;</td> <td class="tcl rb">&emsp;and A. Heydweiler.</td></tr>
+<tr><td class="tcl lb rb">All very dilute aqueous salt solutions having a</td> <td class="tcc rb">1.00 at 18° C.</td> <td class="rb tcl">From results by F. Kohlrausch</td></tr>
+<tr><td class="tcl lb rb">&emsp;concentration of about 0.00001 of an equivalent</td> <td class="rb">&nbsp;</td> <td class="rb tcl">&emsp;and others.</td></tr>
+<tr><td class="tcl lb rb bb">&emsp;gramme molecule<a name="FnAnchor_10" id="FnAnchor_10" href="#Footnote_10"><span class="sp">10</span></a> per litre approximate to</td>
+ <td class="rb bb">&nbsp;</td> <td class="rb bb">&nbsp;</td></tr>
+</table>
+
+<p>The resistivity of all those substances which are generally
+called dielectrics or insulators is also so high that it can only be
+appropriately expressed in millions of megohms per centimetre-cube,
+or in megohms per quadrant-cube, the quadrant being a
+cube the side of which is 10<span class="sp">9</span> cms. (see Table VIII.).</p>
+
+<p class="tab pt05"><span class="sc">Table VIII.</span>&mdash;<i>Electric Volume-Resistivity of Dielectrics reckoned in
+ Millions of Megohms (Mega-megohms) per Centimetre-cube, and in
+ Megohms per Quadrant-cube, i.e. a Cube whose Side is</i> 10<span class="sp">9</span> <i>cms.</i></p>
+
+<table class="ws" summary="Contents">
+<tr><td class="tccm allb" rowspan="2">Substance.</td>
+ <td class="tccm allb" colspan="2">Resistivity.</td>
+ <td class="tccm allb" rowspan="2">Temperature<br />Cent.</td></tr>
+<tr><td class="tccm allb">Mega-megohms<br />per c.c.</td>
+ <td class="tccm allb">Megohms per<br />Quadrant-cube.</td></tr>
+
+<tr><td class="tcl lb rb">Bohemian glass</td> <td class="tcr rb">61&emsp;</td> <td class="tcl rb">&emsp;&emsp;.061</td> <td class="tcc rb">60°</td></tr>
+<tr><td class="tcl lb rb">Mica</td> <td class="tcr rb">84&emsp;</td> <td class="tcl rb">&emsp;&emsp;.084</td> <td class="tcc rb">20°</td></tr>
+<tr><td class="tcl lb rb">Gutta-percha</td> <td class="tcr rb">450&emsp;</td> <td class="tcl rb">&emsp;&emsp;.45</td> <td class="tcc rb">24°</td></tr>
+<tr><td class="tcl lb rb">Flint glass</td> <td class="tcr rb">1,020&emsp;</td> <td class="tcl rb">&emsp;&ensp;1.02</td> <td class="tcc rb">60°</td></tr>
+<tr><td class="tcl lb rb">Glover&rsquo;s vulcanized indiarubber</td> <td class="tcr rb">1,630&emsp;</td> <td class="tcl rb">&emsp;&ensp;1.63</td> <td class="tcc rb">15°</td></tr>
+<tr><td class="tcl lb rb">Siemens&rsquo; ordinary pure vulcanized indiarubber</td> <td class="tcr rb">2,280&emsp;</td> <td class="tcl rb">&emsp;&ensp;2.28</td> <td class="tcc rb">15°</td></tr>
+<tr><td class="tcl lb rb">Shellac</td> <td class="tcr rb">9,000&emsp;</td> <td class="tcl rb">&emsp;&ensp;9.0</td> <td class="tcc rb">28°</td></tr>
+<tr><td class="tcl lb rb">Indiarubber</td> <td class="tcr rb">10,900&emsp;</td> <td class="tcl rb">&emsp;10.9</td> <td class="tcc rb">24°</td></tr>
+<tr><td class="tcl lb rb">Siemens&rsquo; high-insulating fibrous material</td> <td class="tcr rb">11,900&emsp;</td> <td class="tcl rb">&emsp;11.9</td> <td class="tcc rb">15°</td></tr>
+<tr><td class="tcl lb rb">Siemens&rsquo; special high-insulating indiarubber</td> <td class="tcr rb">16,170&emsp;</td> <td class="tcl rb">&emsp;16.17</td> <td class="tcc rb">15°</td></tr>
+<tr><td class="tcl lb rb">Flint glass</td> <td class="tcr rb">20,000&emsp;</td> <td class="tcl rb">&emsp;20.0</td> <td class="tcc rb">20°</td></tr>
+<tr><td class="tcl lb rb">Ebonite</td> <td class="tcr rb">28,000&emsp;</td> <td class="tcl rb">&emsp;28.</td> <td class="tcc rb">46°</td></tr>
+<tr><td class="tcl lb rb bb">Paraffin</td> <td class="tcr rb bb">34,000&emsp;</td> <td class="tcl rb bb">&emsp;34.</td> <td class="tcc rb bb">46°</td></tr>
+</table>
+
+<p>Effects of Heat.&mdash;Temperature affects the resistivity of these
+different classes of conductors in different ways. In all cases, so
+far as is yet known, the resistivity of a pure metal is increased if
+its temperature is raised, and decreased if the temperature is
+lowered, so that if it could be brought to the absolute zero of
+temperature (-273° C.) its resistivity would be reduced to a very
+small fraction of its resistance at ordinary temperatures. With
+metallic alloys, however, rise of temperature does not always
+increase resistivity: it sometimes diminishes it, so that many
+alloys are known which have a maximum resistivity corresponding
+to a certain temperature, and at or near this point they vary very
+little in resistance with temperature. Such alloys have, therefore,
+a negative temperature-variation of resistance at and above
+fixed temperatures. Prominent amongst these metallic compounds
+are alloys of iron, manganese, nickel and copper, some
+of which were discovered by Edward Weston, in the United
+States. One well-known alloy of copper, manganese and nickel,
+now called manganin, which was brought to the notice of
+electricians by the careful investigations made at the Berlin
+Physikalisch-Technische Reichsanstalt, is characterized by
+having a zero temperature coefficient at or about a certain
+temperature in the neighbourhood of 15° C. Hence within a
+certain range of temperature on either side of this critical value
+the resistivity of manganin is hardly affected at all by temperature.
+Similar alloys can be produced from copper and ferro-manganese.
+An alloy formed of 80% copper and 20%
+manganese in an annealed condition has a nearly zero temperature-variation
+of resistance between 20° C. and 100° C. In
+the case of non-metals the action of temperature is generally
+to diminish the resistivity as temperature rises, though this is not
+universally so. The interesting observation has been recorded by
+J. W. Howell, that &ldquo;treated&rdquo; carbon filaments and graphite are
+substances which have a minimum resistance corresponding to a
+certain temperature approaching red heat (<i>Electrician</i>, vol.
+xxxviii. p. 835). At and beyond this temperature increased
+heating appears to increase their resistivity; this phenomenon
+may, however, be accompanied by a molecular change and not be
+a true temperature variation. In the case of dielectric conductors
+and of electrolytes, the action of rising temperature is to reduce
+resistivity. Many of the so-called insulators, such as mica,
+ebonite, indiarubber, and the insulating oils, paraffin, &amp;c.,
+decrease in resistivity with great rapidity as the temperature
+rises. With guttapercha a rise in temperature from 0° C. to
+24° C. is sufficient to reduce the resistivity of one-twentieth part
+of its value at 0° C., and the resistivity of flint glass at 140° C.
+is only one-hundredth of what it is at 60° C.</p>
+
+<div class="condensed">
+<p>A definition may here be given of the meaning of the term <i>Temperature
+Coefficient</i>. If, in the first place, we suppose that the resistivity
+(&rho;<span class="su">t</span>) at any temperature (t) is a simple linear function of the resistivity
+(&rho;<span class="su">0</span>) at 0° C., then we can write &rho;<span class="su">t</span> = &rho;<span class="su">0</span>(1 + &alpha;t), or &alpha; = (&rho;<span class="su">t</span> &minus; &rho;<span class="su">0</span>)/&rho;<span class="su">0</span>t.</p>
+
+<p>The quantity &alpha; is then called the temperature-coefficient, and its
+reciprocal is the temperature at which the resistivity would become
+<span class="pagenum"><a name="page859" id="page859"></a>859</span>
+zero. By an extension of this notion we can call the quantity
+d&rho;/&rho;dt the temperature coefficient corresponding to any temperature
+t at which the resistivity is &rho;. In all cases the relation between the
+resistivity of a substance and the temperature is best set out in the
+form of a curve called a temperature-resistance curve. If a series of
+such curves are drawn for various pure metals, temperature being
+taken as abscissa and resistance as ordinate, and if the temperature
+range extends from the absolute zero of temperature upwards, then
+it is found that these temperature-resistance lines are curved lines
+having their convexity either upwards or downwards. In other
+words, the second differential coefficient of resistance with respect
+to temperature is either a positive or negative quantity. An extensive
+series of observations concerning the form of the resistivity
+curves for various pure metals over a range of temperature extending
+from -200° C. to +200° C. was carried out in 1892 and 1893 by
+Fleming and Dewar (<i>Phil. Mag.</i> Oct. 1892 and Sept. 1893).
+The resistance observations were taken with resistance coils constructed
+with wires of various metals obtained in a state of great
+chemical purity. The lengths and mean diameters of the wires were
+carefully measured, and their resistance was then taken at certain
+known temperatures obtained by immersing the coils in boiling
+aniline, boiling water, melting ice, melting carbonic acid in ether,
+and boiling liquid oxygen, the temperatures thus given being
++184°.5 C., +100° C., 0° C., -78°.2 C. and -182°.5 C. The
+resistivities of the various metals were then calculated and set out
+in terms of the temperature. From these data a chart was prepared
+showing the temperature-resistance curves of these metals
+throughout a range of 400 degrees. The exact form of these curves
+through the region of temperature lying between -200° C. and
+-273° C. is not yet known. As shown on the chart, the curves
+evidently do not converge to precisely the same point. It is, however,
+much less probable that the resistance of any metal should
+vanish at a temperature above the absolute zero than at the absolute
+zero itself, and the precise path of these curves at their lower ends
+cannot be delineated until means are found for fixing independently
+the temperature of some regions in which the resistance of metallic
+wires can be measured. Sir J. Dewar subsequently showed that for
+certain pure metals it is clear that the resistance would not vanish
+at the absolute zero but would be reduced to a finite but small value
+(see &ldquo;Electric Resistance Thermometry at the Temperature of
+Boiling Hydrogen,&rdquo; <i>Proc. Roy. Soc.</i> 1904, 73, p. 244).</p>
+
+<p>The resistivity curves of the magnetic metals are also remarkable
+for the change of curvature they exhibit at the magnetic critical
+temperature. Thus J. Hopkinson and D. K. Morris (<i>Phil. Mag.</i>
+September 1897, p. 213) observed the remarkable alteration that
+takes place in the iron resistance temperature curve in the neighbourhood
+of 780° C. At that temperature the direction of the
+curvature of the curve changes so that it becomes convex upwards
+instead of convex downwards, and in addition the value of the
+temperature coefficient undergoes a great reduction. The mean
+temperature coefficient of iron in the neighbourhood of 0° C. is
+0.0057; at 765° C. it rises to a maximum value 0.0204; but at
+1000° C. it falls again to a lower value, 0.00244. A similar rise to
+a maximum value and subsequent fall are also noted in the case of
+the specific heat of iron. The changes in the curvature of the resistivity
+curves are undoubtedly connected with the molecular changes
+that occur in the magnetic metals at their critical temperatures.</p>
+
+<p>A fact of considerable interest in connexion with resistivity is the
+influence exerted by a strong magnetic field in the case of some
+metals, notably bismuth. It was discovered by A. Righi and confirmed
+by S. A. Leduc (<i>Journ. de Phys.</i> 1886, 5, p. 116, and 1887,
+6, p. 189) that if a pure bismuth wire is placed in a magnetic field
+transversely to the direction of the magnetic field, its resistance is
+considerably increased. This increase is greatly affected by the
+temperature of the metal (Dewar and Fleming, <i>Proc. Roy. Soc.</i> 1897,
+60, p. 427). The temperature coefficient of pure copper is an important
+constant, and its value as determined by Messrs Clark,
+Forde and Taylor in terms of Fahrenheit temperature is</p>
+
+<p class="center1">&rho;<span class="su">t</span> = &rho;<span class="su">32</span> {1 + 0.0023708(t &minus; 32) + 0.0000034548(t &minus; 32)²}.</p>
+</div>
+
+<p><i>Time Effects.</i>&mdash;In the case of dielectric conductors, commonly
+called insulators, such as indiarubber, guttapercha, glass and
+mica, the electric resistivity is not only a function of the temperature
+but also of the time during which the electromotive
+force employed to measure it is imposed. Thus if an indiarubber-covered
+cable is immersed in water and the resistance of the
+dielectric between the copper conductor and the water measured
+by ascertaining the current which can be caused to flow through
+it by an electromotive force, this current is found to vary very
+rapidly with the time during which the electromotive force is
+applied. Apart from the small initial effect due to the electrostatic
+capacity of the cable, the application of an electromotive
+force to the dielectric produces a current through it which
+rapidly falls in value, as if the electric resistance of the dielectric
+were increasing. The current, however, does not fall continuously
+but tends to a limiting value, and it appears that if the
+electromotive force is kept applied to the cable for a prolonged
+time, a small and nearly constant current will ultimately be
+found flowing through it. It is customary in electro-technical
+work to consider the resistivity of the dielectric as the value it has
+after the electromotive force has been applied for one minute, the
+standard temperature being 75° F. This, however, is a purely
+conventional proceeding, and the number so obtained does not
+necessarily represent the true or ohmic resistance of the dielectric.
+If the electromotive force is increased, in the case of a large
+number of ordinary dielectrics the apparent resistance at the end
+of one minute&rsquo;s electrification decreases as the electromotive
+force increases.</p>
+
+<p><i>Practical Standards.</i>&mdash;The practical measurement of resistivity
+involves many processes and instruments (see <span class="sc"><a href="#artlinks">Wheatstone&rsquo;s
+Bridge</a></span> and <span class="sc"><a href="#artlinks">Ohmmeter</a></span>). Broadly speaking, the
+processes are divided into <i>Comparison Methods</i> and <i>Absolute
+Methods</i>. In the former a comparison is effected between the
+resistance of a material in a known form and some standard
+resistance. In the <i>Absolute Methods</i> the resistivity is determined
+without reference to any other substance, but with reference
+only to the fundamental standards of length, mass and time.
+Immense labour has been expended in investigations concerned
+with the production of a standard of resistance and its evaluation
+in absolute measure. In some cases the absolute standard is
+constructed by filling a carefully-calibrated tube of glass with
+mercury, in order to realize in a material form the official definition
+of the ohm; in this manner most of the principal national
+physical laboratories have been provided with standard mercury
+ohms. (For a full description of the standard mercury ohm of
+the Berlin Physikalisch-Technische Reichsanstalt, see the
+<i>Electrician</i>, xxxvii. 569.) For practical purposes it is more convenient
+to employ a standard of resistance made of wire.</p>
+
+<div class="condensed">
+<p>Opinion is not yet perfectly settled on the question whether a
+wire made of any alloy can be considered to be a perfectly unalterable
+standard of resistance, but experience has shown that a platinum
+silver alloy (66% silver, 33% platinum), and also the alloy called
+manganin, seem to possess the qualities of permanence essential for
+a wire-resistance standard. A comparison made in 1892 and 1894
+of all the manganin wire copies of the ohm made at the Reichsanstalt
+in Berlin, showed that these standards had remained constant for
+two years to within one or two parts in 100,000. It appears, however,
+that in order that manganin may remain constant in resistivity when
+used in the manufacture of a resistance coil, it is necessary that the
+alloy should be <i>aged</i> by heating it to a temperature of 140° C. for ten
+hours; and to prevent subsequent changes in resistivity, solders
+containing zinc must be avoided, and a silver solder containing 75%
+of silver employed in soldering the manganin wire to its connexions.</p>
+</div>
+
+<p>The authorities of the Berlin Reichsanstalt have devoted
+considerable attention to the question of the best form for a wire
+standard of electric resistance. In that now adopted the resistance
+wire is carefully insulated and wound on a brass cylinder,
+being doubled on itself to annul inductance as much as possible.
+In the coil two wires are wound on in parallel, one being much
+finer than the other, and the final adjustment of the coil to an
+exact value is made by shortening the finer of the two. A
+standard of resistance for use in a laboratory now generally
+consists of a wire of manganin or platinum-silver carefully
+insulated and enclosed in a brass case. Thick copper rods are
+connected to the terminals of the wire in the interior of the case,
+and brought to the outside, being carefully insulated at the same
+time from one another and from the case. The coil so constructed
+can be placed under water or paraffin oil, the temperature of
+which can be exactly observed during the process of taking a
+resistance measurement. Equalization of the temperature of
+the surrounding medium is effected by the employment of a
+stirrer, worked by hand or by a small electric motor. The
+construction of a standard of electrical resistance consisting of
+mercury in a glass tube is an operation requiring considerable
+precautions, and only to be undertaken by those experienced
+in the matter. Opinions are divided on the question whether
+greater permanence in resistance can be secured by mercury-in-glass
+standards of resistance or by wire standards, but the latter
+are at least more portable and less fragile.</p>
+
+<div class="condensed">
+<p>A full description of the construction of a standard wire-resistance
+coil on the plan adopted by the Berlin Physikalisch-Technische
+<span class="pagenum"><a name="page860" id="page860"></a>860</span>
+Reichsanstalt is given in the Report of the British Association Committee
+on Electrical Standards, presented at the Edinburgh Meeting
+in 1892. For the design and construction of standards of electric
+resistances adapted for employment in the comparison and measurement
+of very low or very high resistances, the reader may be referred
+to standard treatises on electric measurements.</p>
+
+<p><span class="sc">Bibliography.</span>&mdash;See also J. A. Fleming, <i>A Handbook for the
+Electrical Laboratory and Testing Room</i>, vol. i. (London, 1901);
+<i>Reports of the British Association Committee on Electrical Standards</i>,
+edited by Fleeming Jenkin (London, 1873); A. Matthiessen and C.
+Vogt, &ldquo;On the Influence of Temperature on the Conducting Power
+of Alloys,&rdquo; <i>Phil. Trans.</i>, 1864, 154, p. 167, and <i>Phil. Mag.</i>, 1865,
+29, p. 363; A. Matthiessen and M. Holtzmann, &ldquo;On the Effect of
+the Presence of Metals and Metalloids upon the Electric Conducting
+Power of Pure Copper,&rdquo; <i>Phil. Trans.</i>, 1860, 150, p. 85; T. C. Fitzpatrick,
+&ldquo;On the Specific Resistance of Copper,&rdquo; <i>Brit. Assoc. Report</i>,
+1890, p. 120, or <i>Electrician</i>, 1890, 25, p. 608; R. Appleyard, <i>The
+Conductometer and Electrical Conductivity</i>; Clark, Forde and Taylor,
+<i>Temperature Coefficients of Copper</i> (London, 1901).</p>
+</div>
+<div class="author">(J. A. F.)</div>
+
+<p class="center1">II. <span class="sc">Conduction in Liquids</span></p>
+
+<p>Through liquid metals, such as mercury at ordinary temperatures
+and other metals at temperatures above their melting
+points, the electric current flows as in solid metals without
+changing the state of the conductor, except in so far as heat is
+developed by the electric resistance. But another class of liquid
+conductors exists, and in them the phenomena are quite
+different. The conductivity of fused salts, and of solutions of
+salts and acids, although less than that of metals, is very great
+compared with the traces of conductivity found in so-called non-conductors.
+In fused salts and conducting solutions the passage
+of the current is always accompanied by definite chemical
+changes; the substance of the conductor or electrolyte is
+decomposed, and the products of the decomposition appear at the
+electrodes, <i>i.e.</i> the metallic plates by means of which the current
+is led into and out of the solution. The chemical phenomena are
+considered in the article <span class="sc"><a href="#artlinks">Electrolysis</a></span>; we are here concerned
+solely with the mechanism of this <i>electrolytic</i> conduction of the
+current.</p>
+
+<p>To explain the appearance of the products of decomposition at
+the electrodes only, while the intervening solution is unaltered,
+we suppose that, under the action of the electric forces, the
+opposite parts of the electrolyte move in opposite directions
+through the liquid. These opposite parts, named ions by
+Faraday, must therefore be associated with electric charges, and
+it is the convective movement of the opposite streams of ions
+carrying their charges with them that, on this view, constitutes
+the electric current.</p>
+
+<p>In metallic conduction it is found that the current is proportional
+to the applied electromotive force&mdash;a relation known by
+the name of Ohm&rsquo;s law. If we place in a circuit with a small
+electromotive force an electrolytic cell consisting of two platinum
+electrodes and a solution, the initial current soon dies away, and
+we shall find that a certain minimum electromotive force must be
+applied to the circuit before any considerable permanent current
+passes. The chemical changes which are initiated on the surfaces
+of the electrodes set up a reverse electromotive force of polarization,
+and, until this is overcome, only a minute current, probably
+due to the slow but steady removal of the products of decomposition
+from the electrodes by a process of diffusion, will pass
+through the cell. Thus it is evident that, considering the
+electrolytic cell as a whole, the passage of the current through it
+cannot conform to Ohm&rsquo;s law. But the polarization is due to
+chemical changes, which are confined to the surfaces of the
+electrodes; and it is necessary to inquire whether, if the polarization
+at the electrodes be eliminated, the passage of the current
+through the bulk of the solution itself is proportional to the
+electromotive force actually applied to that solution. Rough
+experiment shows that the current is proportional to the excess of
+the electromotive force over a constant value, and thus verifies
+the law approximately, the constant electromotive force to be
+overcome being a measure of the polarization. A more satisfactory
+examination of the question was made by F. Kohlrausch
+in the years 1873 to 1876. Ohm&rsquo;s law states that the current C
+is proportional to the electromotive force E, or C = kR, where k is
+a constant called the conductivity of the circuit. The equation
+may also be written as C = E/R, where R is a constant, the
+reciprocal of k, known as the resistance of the circuit. The
+essence of the law is the proportionality between C and E, which
+means that the ratio E/C is a constant. But E/C = R, and thus
+the law may be tested by examining the constancy of the
+measured resistance of a conductor when different currents are
+passing through it. In this way Ohm&rsquo;s law has been confirmed in
+the case of metallic conduction to a very high degree of accuracy.
+A similar principle was applied by Kohlrausch to the case of
+electrolytes, and he was the first to show that an electrolyte
+possesses a definite resistance which has a constant value when
+measured with different currents and by different experimental
+methods.</p>
+
+<p><i>Measurement of the Resistance of Electrolytes.</i>&mdash;There are two
+effects of the passage of an electric current which prevent the
+possibility of measuring electrolytic resistance by the ordinary
+methods with the direct currents which are used in the case of
+metals. The products of the chemical decomposition of the
+electrolyte appear at the electrodes and set up the opposing
+electromotive force of polarization, and unequal dilution of the
+solution may occur in the neighbourhood of the two electrodes.
+The chemical and electrolytic aspects of these phenomena are
+treated in the article <span class="sc"><a href="#artlinks">Electrolysis</a></span>, but from our present point
+of view also it is evident that they are again of fundamental
+importance. The polarization at the surface of the electrodes
+will set up an opposing electromotive force, and the unequal
+dilution of the solution will turn the electrolyte into a concentration
+cell and produce a subsidiary electromotive force either in
+the same direction as that applied or in the reverse according
+as the anode or the cathode solution becomes the more dilute.
+Both effects thus involve internal electromotive forces, and
+prevent the application of Ohm&rsquo;s law to the electrolytic cell as a
+whole. But the existence of a definite measurable resistance as a
+characteristic property of the system depends on the conformity
+of the system to Ohm&rsquo;s law, and it is therefore necessary to
+eliminate both these effects before attempting to measure the
+resistance.</p>
+
+<p>The usual and most satisfactory method of measuring the
+resistance of electrolytes consists in eliminating the effects of
+polarization by the use of alternating currents, that is, currents
+that are reversed in direction many times a second.<a name="FnAnchor_11" id="FnAnchor_11" href="#Footnote_11"><span class="sp">11</span></a> The
+chemical action produced by the first current is thus reversed by
+the second current in the opposite direction, and the polarization
+caused by the first current on the surface of the electrodes is
+destroyed before it rises to an appreciable value. The polarization
+is also diminished in another way. The electromotive force
+of polarization is due to the deposition of films of the products of
+chemical decomposition on the surface of the electrodes, and
+only reaches its full value when a continuous film is formed. If
+the current be stopped before such a film is completed, the
+reverse electromotive force is less than its full value. A given
+current flowing for a given time deposits a definite amount of
+substance on the electrodes, and therefore the amount per unit
+area is inversely proportional to the area of the electrodes&mdash;to
+the area of contact, that is, between the electrode and the liquid.
+Thus, by increasing the area of the electrodes, the polarization due
+to a given current is decreased. Now the area of free surface of a
+platinum plate can be increased enormously by coating the plate
+with platinum black, which is metallic platinum in a spongy
+state, and with such a plate as electrode the effects of polarization
+are diminished to a very marked extent. The coating is effected
+by passing an electric current first one way and then the other
+between two platinum plates immersed in a 3% solution of
+platinum chloride to which a trace of lead acetate is sometimes
+added. The platinized plates thus obtained are quite satisfactory
+for the investigation of strong solutions. They have the power,
+however, of absorbing a certain amount of salt from the solutions
+and of giving it up again when water or more dilute solution is
+placed in contact with them. The measurement of very dilute
+solutions is thus made difficult, but, if the plates be heated to
+<span class="pagenum"><a name="page861" id="page861"></a>861</span>
+redness after being platinized, a grey surface is obtained which
+possesses sufficient area for use with dilute solutions and yet does
+not absorb an appreciable quantity of salt.</p>
+
+<p>Any convenient source of alternating current may be used.
+The currents from the secondary circuit of a small induction coil
+are satisfactory, or the currents of an alternating electric light
+supply may be transformed down to an electromotive force of one
+or two volts. With such currents it is necessary to consider the
+effects of self-induction in the circuit and of electrostatic capacity.
+In balancing the resistance of the electrolyte, resistance coils may
+be used in which self-induction and the capacity are reduced
+to a minimum by winding the wire of the coil backwards and
+forwards in alternate layers.</p>
+
+<table class="nobctr" style="float: left; width: 290px;" summary="Illustration">
+<tr><td class="figleft1"><img style="width:237px; height:185px" src="images/img861a.jpg" alt="" /></td></tr>
+<tr><td class="caption"><span class="sc">Fig. 1.</span></td></tr></table>
+
+<p>With these arrangements the usual method of measuring
+resistance by means of Wheatstone&rsquo;s bridge may be adapted to
+the case of electrolytes. With alternating currents, however,
+it is impossible to use a galvanometer in the usual way. The
+galvanometer was therefore replaced by Kohlrausch by a
+telephone, which gives a sound
+when an alternating current
+passes through it. The most common
+plan of the apparatus is
+shown diagrammatically in fig. 1.
+The electrolytic cell and a resistance
+box form two arms of the
+bridge, and the sliding contact is
+moved along the metre wire which
+forms the other two arms till no
+sound is heard in the telephone.
+The resistance of the electrolyte is to that of the box as that
+of the right-hand end of the wire is to that of the left-hand
+end. A more accurate method of using alternating currents,
+and one more pleasant to use, gets rid of the telephone
+(<i>Phil. Trans.</i>, 1900, 194, p. 321). The current from one or two
+voltaic cells is led to an ebonite drum turned by a motor or
+a hand-wheel and cord. On the drum are fixed brass strips
+with wire brushes touching them in such a manner that the
+current from the brushes is reversed several times in each
+revolution of the drum. The wires from the brushes are connected
+with the Wheatstone&rsquo;s bridge. A moving coil galvanometer
+is used as indicator, its connexions being reversed in time
+with those of the battery by a slightly narrower set of brass
+strips fixed on the other end of the ebonite commutator. Thus
+any residual current through the galvanometer is direct and not
+alternating. The high moment of inertia of the coil makes the
+period of swing slow compared with the period of alternation of
+the current, and the slight periodic disturbances are thus prevented
+from affecting the galvanometer. When the measured
+resistance is not altered by increasing the speed of the commutator
+or changing the ratio of the arms of the bridge, the
+disturbing effects may be considered to be eliminated.</p>
+
+<table class="nobctr" style="float: left; width: 310px;" summary="Illustration">
+<tr><td class="figleft1" colspan="2"><img style="width:264px; height:117px" src="images/img861b.jpg" alt="" /></td></tr>
+<tr><td class="caption"><span class="sc">Fig. 2.&emsp;</span></td>
+ <td class="caption"><span class="sc">Fig. 3.&emsp;</span></td></tr></table>
+
+<p>The form of vessel chosen to contain the electrolyte depends
+on the order of resistance to be measured. For dilute solutions
+the shape of cell shown in
+fig. 2 will be found convenient,
+while for more concentrated
+solutions, that indicated in fig.
+3 is suitable. The absolute
+resistances of certain solutions
+have been determined by
+Kohlrausch by comparison
+with mercury, and, by using one of these solutions in any
+cell, the constant of that cell may be found once for all.
+From the observed resistance of any given solution in the
+cell the resistance of a centimetre cube&mdash;the so-called specific
+resistance&mdash;may be calculated. The reciprocal of this, or
+the conductivity, is a more generally useful constant;
+it is conveniently expressed in terms of a unit equal to the
+reciprocal of an ohm. Thus Kohlrausch found that a solution of
+potassium chloride, containing one-tenth of a gram equivalent
+(7.46 grams) per litre, has at 18° C. a specific resistance of 89.37
+ohms per centimetre cube, or a conductivity of 1.119&times;10<span class="sp">-2</span>
+mhos or 1.119&times;10<span class="sp">-11</span> C.G.S. units. As the temperature variation
+of conductivity is large, usually about 2% per degree, it is
+necessary to place the resistance cell in a paraffin or water bath,
+and to observe its temperature with some accuracy.</p>
+
+<p>Another way of eliminating the effects of polarization and of
+dilution has been used by W. Stroud and J. B. Henderson
+(<i>Phil. Mag.</i>, 1897 [5], 43, p. 19). Two of the arms of a Wheatstone&rsquo;s
+bridge are composed of narrow tubes filled with the
+solution, the tubes being of equal diameter but of different
+length. The other two arms are made of coils of wire of equal
+resistance, and metallic resistance is added to the shorter tube
+till the bridge is balanced. Direct currents of somewhat high
+electromotive force are used to work the bridge. Equal currents
+then flow through the two tubes; the effects of polarization and
+dilution must be the same in each, and the resistance added to the
+shorter tube must be equal to the resistance of a column of liquid
+the length of which is equal to the difference in length of the two
+tubes.</p>
+
+<p>A somewhat different principle was adopted by E. Bouty in
+1884. If a current be passed through two resistances in series by
+means of an applied electromotive force, the electric potential
+falls from one end of the resistances to the other, and, if we apply
+Ohm&rsquo;s law to each resistance in succession, we see that, since for
+each of them E = CR, and C the current is the same through both,
+E the electromotive force or fall of potential between the ends of
+each resistance must be proportional to the resistance between
+them. Thus by measuring the potential difference between the
+ends of the two resistances successively, we may compare their
+resistances. If, on the other hand, we can measure the potential
+difference in some known units, and similarly measure the current
+flowing, we can determine the resistance of a single electrolyte.
+The details of the apparatus may vary, but its principle is
+illustrated in the following description. A narrow glass tube is
+fixed horizontally into side openings in two glass vessels, and an
+electric current passed through it by means of platinum electrodes
+and a battery of considerable electromotive force. In this way a
+steady fall of electric potential is set up along the length of the
+tube. To measure the potential difference between the ends of
+the tube, tapping electrodes are constructed, <i>e.g.</i> by placing zinc
+rods in vessels with zinc sulphate solution and connecting these
+vessels (by means of thin siphon tubes also filled with solution)
+with the vessels at the ends of the long tube which contains the
+electrolyte to be examined. Whatever be the contact potential
+difference between zinc and its solution, it is the same at both
+ends, and thus the potential difference between the zinc rods is
+equal to that between the liquid at the two ends of the tube.
+This potential difference may be measured without passing any
+appreciable current through the tapping electrodes, and thus the
+resistance of the liquid deduced.</p>
+
+<p><i>Equivalent Conductivity of Solutions.</i>&mdash;As is the case in the
+other properties of solutions, the phenomena are much more
+simple when the concentration is small than when it is great, and
+a study of dilute solutions is therefore the best way of getting an
+insight into the essential principles of the subject. The foundation
+of our knowledge was laid by Kohlrausch when he had developed
+the method of measuring electrolyte resistance described above.
+He expressed his results in terms of &ldquo;equivalent conductivity,&rdquo;
+that is, the conductivity (k) of the solution divided by the number
+(m) of gram-equivalents of electrolyte per litre. He finds that, as
+the concentration diminishes, the value of k/m approaches a
+limit, and eventually becomes constant, that is to say, at great
+dilution the conductivity is proportional to the concentration.
+Kohlrausch first prepared very pure water by repeated distillation
+and found that its resistance continually increased as the process
+of purification proceeded. The conductivity of the water, and of
+the slight impurities which must always remain, was subtracted
+from that of the solution made with it, and the result, divided
+by m, gave the equivalent conductivity of the substance dissolved.
+This procedure appears justifiable, for as long as conductivity is
+proportional to concentration it is evident that each part of the
+dissolved matter produces its own independent effect, so that the
+total conductivity is the sum of the conductivities of the parts;
+<span class="pagenum"><a name="page862" id="page862"></a>862</span>
+when this ceases to hold, the concentration of the solution has in
+general become so great that the conductivity of the solvent may
+be neglected. The general result of these experiments can be
+represented graphically by plotting <i>k/m</i> as ordinates and <span class="sp">3</span>&radic;<span class="ov">m</span>
+as abscissae, <span class="sp">3</span>&radic;<span class="ov">m</span> being a number proportional to the reciprocal
+of the average distance between the molecules, to which it seems
+likely that the molecular conductivity may be related. The
+general types of curve for a simple neutral salt like potassium or
+sodium chloride and for a caustic alkali or acid are shown in fig. 4.
+The curve for the neutral salt comes to a limiting value; that for
+the acid attains a maximum at a certain very small concentration,
+and falls again when the dilution
+is carried farther. It has usually
+been considered that this destruction
+of conductivity is due to
+chemical action between the acid
+and the residual impurities in the
+water. At such great dilution these
+impurities are present in quantities
+comparable with the amount of acid
+which they convert into a less
+highly conducting neutral salt. In
+the case of acids, then, the maximum
+must be taken as the limiting
+value. The decrease in equivalent conductivity at great dilution
+is, however, so constant that this explanation seems insufficient.
+The true cause of the phenomenon may perhaps be connected
+with the fact that the bodies in which it occurs, acids and
+alkalis, contain the ions, hydrogen in the one case, hydroxyl in
+the other, which are present in the solvent, water, and have,
+perhaps because of this relation, velocities higher than those of any
+other ions. The values of the molecular conductivities of all
+neutral salts are, at great dilution, of the same order of magnitude,
+while those of acids at their maxima are about three times as
+large. The influence of increasing concentration is greater in the
+case of salts containing divalent ions, and greatest of all in such
+cases as solutions of ammonia and acetic acid, which are substances
+of very low conductivity.</p>
+
+<table class="nobctr" style="float: left; width: 275px;" summary="Illustration">
+<tr><td class="figleft1"><img style="width:225px; height:217px" src="images/img862.jpg" alt="" /></td></tr>
+<tr><td class="caption"><span class="sc">Fig. 4.</span></td></tr></table>
+
+
+<p><i>Theory of Moving Ions.</i>&mdash;Kohlrausch found that, when the
+polarization at the electrodes was eliminated, the resistance of a
+solution was constant however determined, and thus established
+Ohm&rsquo;s Law for electrolytes. The law was confirmed in the case
+of strong currents by G. F. Fitzgerald and F. T. Trouton (<i>B.A.
+Report</i>, 1886, p. 312). Now, Ohm&rsquo;s Law implies that no work is
+done by the current in overcoming reversible electromotive
+forces such as those of polarization. Thus the molecular interchange
+of ions, which must occur in order that the products may
+be able to work their way through the liquid and appear at the
+electrodes, continues throughout the solution whether a current is
+flowing or not. The influence of the current on the ions is
+merely directive, and, when it flows, streams of electrified ions
+travel in opposite directions, and, if the applied electromotive
+force is enough to overcome the local polarization, give up their
+charges to the electrodes. We may therefore represent the facts
+by considering the process of electrolysis to be a kind of convection.
+Faraday&rsquo;s classical experiments proved that when a current
+flows through an electrolyte the quantity of substance liberated
+at each electrode is proportional to its chemical equivalent
+weight, and to the total amount of electricity passed. Accurate
+determinations have since shown that the mass of an ion deposited
+by one electromagnetic unit of electricity, <i>i.e.</i> its electro-chemical
+equivalent, is 1.036&times;10<span class="sp">-4</span>&times;its chemical equivalent
+weight. Thus the amount of electricity associated with one
+gram-equivalent of any ion is 10<span class="sp">4</span>/1.036 = 9653 units. Each
+monovalent ion must therefore be associated with a certain
+definite charge, which we may take to be a natural unit of
+electricity; a divalent ion carries two such units, and so on.
+A cation, <i>i.e.</i> an ion giving up its charge at the cathode, as the
+electrode at which the current leaves the solution is called, carries
+a positive charge of electricity; an anion, travelling in the
+opposite direction, carries a negative charge. It will now be seen
+that the quantity of electricity flowing per second, <i>i.e.</i> the current
+through the solution, depends on (1) the number of the ions
+concerned, (2) the charge on each ion, and (3) the velocity with
+which the ions travel past each other. Now, the number of ions
+is given by the concentration of the solution, for even if all the
+ions are not actively engaged in carrying the current at the same
+instant, they must, on any dynamical idea of chemical equilibrium,
+be all active in turn. The charge on each, as we have
+seen, can be expressed in absolute units, and therefore the
+velocity with which they move past each other can be calculated.
+This was first done by Kohlrausch (<i>Göttingen Nachrichten</i>, 1876,
+p. 213, and <i>Das Leitvermögen der Elektrolyte</i>, Leipzig, 1898)
+about 1879.</p>
+
+<div class="condensed">
+<p>In order to develop Kohlrausch&rsquo;s theory, let us take, as an example,
+the case of an aqueous solution of potassium chloride, of concentration
+n gram-equivalents per cubic centimetre. There will then
+be n gram-equivalents of potassium ions and the same number of
+chlorine ions in this volume. Let us suppose that on each gram-equivalent
+of potassium there reside +e units of electricity, and on
+each gram-equivalent of chlorine ions -e units. If u denotes the
+average velocity of the potassium ion, the positive charge carried
+per second across unit area normal to the flow is n e u. Similarly, if
+v be the average velocity of the chlorine ions, the negative charge
+carried in the opposite direction is n e v. But positive electricity
+moving in one direction is equivalent to negative electricity moving
+in the other, so that, before changes in concentration sensibly supervene,
+the total current, C, is ne(u + v). Now let us consider the
+amounts of potassium and chlorine liberated at the electrodes by
+this current. At the cathode, if the chlorine ions were at rest, the
+excess of potassium ions would be simply those arriving in one second,
+namely, nu. But since the chlorine ions move also, a further separation
+occurs, and nv potassium ions are left without partners. The
+total number of gram-equivalents liberated is therefore n(u + v).
+By Faraday&rsquo;s law, the number of grams liberated is equal to the
+product of the current and the electro-chemical equivalent of the
+ion; the number of gram-equivalents therefore must be equal to
+&eta;C, where &eta; denotes the electro-chemical equivalent of hydrogen in
+C.G.S. units. Thus we get</p>
+
+<p class="center1">n(u + v) = &eta;C = &eta;ne(u + v),</p>
+
+<p class="noind">and it follows that the charge, e, on 1 gram-equivalent of each kind
+of ion is equal to 1/&eta;. We know that Ohm&rsquo;s Law holds good for
+electrolytes, so that the current C is also given by k·dP/dx, where
+k denotes the conductivity of the solution, and dP/dx the potential
+gradient, <i>i.e.</i> the change in potential per unit length along the lines of
+current flow. Thus</p>
+
+<table class="math0" summary="math">
+<tr> <td>n</td> <td rowspan="2">(u + v) = kdP/dx;</td></tr>
+<tr> <td class="denom">&eta;</td></tr></table>
+
+<p class="noind">therefore</p>
+
+<table class="math0" summary="math">
+<tr><td rowspan="2">u + v = &eta;</td>
+ <td>k</td> <td rowspan="2">&nbsp;</td> <td>dP</td> <td rowspan="2">.</td></tr>
+<tr> <td class="denom">n</td> <td class="denom">dx</td></tr></table>
+
+<p class="noind">Now &eta; is 1.036&times;10<span class="sp">-4</span>, and the concentration of a solution is usually
+expressed in terms of the number, m, of gram-equivalents per litre
+instead of per cubic centimetre. Therefore</p>
+
+<table class="math0" summary="math">
+<tr><td rowspan="2">u + v = 1.036&times;10<span class="sp">-1</span></td>
+ <td>k</td> <td rowspan="2">&nbsp;</td> <td>dP</td> <td rowspan="2">.</td></tr>
+<tr> <td class="denom">m</td> <td class="denom">dx</td></tr></table>
+
+<p>When the potential gradient is one volt (10<span class="sp">8</span> C.G.S. units) per
+centimetre this becomes</p>
+
+<p class="center1">u + v = 1.036&times;10<span class="sp">-7</span>&times;k/m.</p>
+
+<p>Thus by measuring the value of k/m, which is known as the
+equivalent conductivity of the solution, we can find u + v, the
+velocity of the ions relative to each other. For instance, the equivalent
+conductivity of a solution of potassium chloride containing one-tenth
+of a gram-equivalent per litre is 1119&times;10<span class="sp">-13</span> C.G.S. units at
+18° C. Therefore</p>
+
+<p class="center1">u + v = 1.036&times;10<span class="sp">7</span>&times;1119&times;10<span class="sp">-13</span>
+= 1.159&times;10<span class="sp">-3</span> = 0.001159 cm. per sec.</p>
+
+<p>In order to obtain the absolute velocities u and v, we must find
+some other relation between them. Let us resolve u into ½(u + v)
+in one direction, say to the right, and ½(u &minus; v) to the left. Similarly
+v can be resolved into ½(v + u) to the left and ½(v &minus; u) to the right.
+On pairing these velocities we have a combined movement of the
+ions to the right, with a speed of ½(u &minus; v) and a drift right and left,
+past each other, each ion travelling with a speed of ½(u + v), constituting
+the electrolytic separation. If u is greater than v, the combined
+movement involves a concentration of salt at the cathode, and a
+corresponding dilution at the anode, and <i>vice versa</i>. The rate at
+which salt is electrolysed, and thus removed from the solution at
+each electrode, is ½(u + v). Thus the total loss of salt at the cathode
+is ½(u + v) &minus; ½(u &minus; v) or v, and at the anode, ½(v + u) &minus; ½(v &minus; u), or u.
+Therefore, as is explained in the article <span class="sc"><a href="#artlinks">Electrolysis</a></span>, by measuring
+the dilution of the liquid round the electrodes when a current passed,
+W. Hittorf (<i>Pogg. Ann.</i>, 1853-1859, 89, p. 177; 98, p. 1; 103, p. 1; 106,
+pp. 337 and 513) was able to deduce the ratio of the two velocities,
+for simple salts when no complex ions are present, and many further
+<span class="pagenum"><a name="page863" id="page863"></a>863</span>
+experiments have been made on the subject (see <i>Das Leitvermögen
+der Elektrolyte</i>).</p>
+
+<p>By combining the results thus obtained with the sum of the
+velocities, as determined from the conductivities, Kohlrausch calculated
+the absolute velocities of different ions under stated conditions.
+Thus, in the case of the solution of potassium chloride
+considered above, Hittorf&rsquo;s experiments show us that the ratio of
+the velocity of the anion to that of the cation in this solution is
+.51 : .49. The absolute velocity of the potassium ion under unit
+potential gradient is therefore 0.000567 cm. per sec., and that of
+the chlorine ion 0.000592 cm. per sec. Similar calculations can
+be made for solutions of other concentrations, and of different
+substances.</p>
+</div>
+
+<p>Table IX. shows Kohlrausch&rsquo;s values for the ionic velocities of
+three chlorides of alkali metals at 18° C, calculated for a potential
+gradient of 1 volt per cm.; the numbers are in terms of a unit
+equal to 10<span class="sp">-6</span> cm. per sec.:&mdash;</p>
+
+<p class="center1"><span class="sc">Table IX.</span></p>
+
+<table class="ws" summary="Contents">
+<tr><td class="tcc allb">&nbsp;</td>
+ <td class="tcc allb" colspan="3">KCl</td> <td class="tcc allb" colspan="3">NaCl</td> <td class="tcc allb" colspan="3">LiCl</td></tr>
+<tr><td class="tcc allb">m</td>
+ <td class="tcc bb">u + v</td> <td class="tcc bb">u</td> <td class="tcc rb bb">v</td>
+ <td class="tcc bb">u + v</td> <td class="tcc bb">u</td> <td class="tcc rb bb">v</td>
+ <td class="tcc bb">u + v</td> <td class="tcc bb">u</td> <td class="tcc rb bb">v</td></tr>
+
+<tr><td class="tcl lb rb">&ensp;0</td> <td class="tcr">1350</td> <td class="tcr">660</td> <td class="tcr rb">690</td> <td class="tcr">1140</td> <td class="tcr">450</td> <td class="tcr rb">690</td> <td class="tcr">1050</td> <td class="tcr">360</td> <td class="tcr rb">690</td></tr>
+<tr><td class="tcl lb rb">&ensp;0.0001</td> <td class="tcr">1335</td> <td class="tcr">654</td> <td class="tcr rb">681</td> <td class="tcr">1129</td> <td class="tcr">448</td> <td class="tcr rb">681</td> <td class="tcr">1037</td> <td class="tcr">356</td> <td class="tcr rb">681</td></tr>
+<tr><td class="tcl lb rb">&emsp;.001</td> <td class="tcr">1313</td> <td class="tcr">643</td> <td class="tcr rb">670</td> <td class="tcr">1110</td> <td class="tcr">440</td> <td class="tcr rb">670</td> <td class="tcr">1013</td> <td class="tcr">343</td> <td class="tcr rb">670</td></tr>
+<tr><td class="tcl lb rb">&emsp;.01</td> <td class="tcr">1263</td> <td class="tcr">619</td> <td class="tcr rb">644</td> <td class="tcr">1059</td> <td class="tcr">415</td> <td class="tcr rb">644</td> <td class="tcr">962</td> <td class="tcr">318</td> <td class="tcr rb">644</td></tr>
+<tr><td class="tcl lb rb">&emsp;.03</td> <td class="tcr">1218</td> <td class="tcr">597</td> <td class="tcr rb">621</td> <td class="tcr">1013</td> <td class="tcr">390</td> <td class="tcr rb">623</td> <td class="tcr">917</td> <td class="tcr">298</td> <td class="tcr rb">619</td></tr>
+<tr><td class="tcl lb rb">&emsp;.1</td> <td class="tcr">1153</td> <td class="tcr">564</td> <td class="tcr rb">589</td> <td class="tcr">952</td> <td class="tcr">360</td> <td class="tcr rb">592</td> <td class="tcr">853</td> <td class="tcr">259</td> <td class="tcr rb">594</td></tr>
+<tr><td class="tcl lb rb">&emsp;.3</td> <td class="tcr">1088</td> <td class="tcr">531</td> <td class="tcr rb">557</td> <td class="tcr">876</td> <td class="tcr">324</td> <td class="tcr rb">552</td> <td class="tcr">774</td> <td class="tcr">217</td> <td class="tcr rb">557</td></tr>
+<tr><td class="tcl lb rb">&ensp;1.0</td> <td class="tcr">1011</td> <td class="tcr">491</td> <td class="tcr rb">520</td> <td class="tcr">765</td> <td class="tcr">278</td> <td class="tcr rb">487</td> <td class="tcr">651</td> <td class="tcr">169</td> <td class="tcr rb">482</td></tr>
+<tr><td class="tcl lb rb">&ensp;3.0</td> <td class="tcr">911</td> <td class="tcr">442</td> <td class="tcr rb">469</td> <td class="tcr">582</td> <td class="tcr">206</td> <td class="tcr rb">376</td> <td class="tcr">463</td> <td class="tcr">115</td> <td class="tcr rb">348</td></tr>
+<tr><td class="tcl lb rb">&ensp;5.0</td> <td class="tcr">&nbsp;</td> <td class="tcr">&nbsp;</td> <td class="tcr rb">&nbsp;</td> <td class="tcr">438</td> <td class="tcr">153</td> <td class="tcr rb">285</td> <td class="tcr">334</td> <td class="tcr">80</td> <td class="tcr rb">254</td></tr>
+<tr><td class="tcl lb rb bb">10.0</td> <td class="tcr bb">&nbsp;</td> <td class="tcr bb">&nbsp;</td> <td class="tcr rb bb">&nbsp;</td> <td class="tcr bb">&nbsp;</td> <td class="tcr bb">&nbsp;</td> <td class="tcr rb bb">&nbsp;</td> <td class="tcr bb">117</td> <td class="tcr bb">25</td> <td class="tcr rb bb">92</td></tr>
+</table>
+
+<p>These numbers show clearly that there is an increase in ionic
+velocity as the dilution proceeds. Moreover, if we compare the
+values for the chlorine ion obtained from observations on these
+three different salts, we see that as the concentrations diminish
+the velocity of the chlorine ion becomes the same in all of them.
+A similar relation appears in other cases, and, in general, we may
+say that at great dilution the velocity of an ion is independent of
+the nature of the other ion present. This introduces the conception
+of specific ionic velocities, for which some values at 18° C.
+are given by Kohlrausch in Table X.:&mdash;</p>
+
+<p class="center1">Table X.</p>
+
+<table class="ws" summary="Contents">
+
+<tr><td class="tcl lb tb">K</td> <td class="tcr tb">66</td> <td class="tcc tb">&times; 10<span class="sp">-5</span> cms.</td> <td class="tcc rb tb">per sec.</td>
+ <td class="tcl lb tb">Cl</td> <td class="tcr tb">69</td> <td class="tcc tb">&times; 10<span class="sp">-5</span> cms.</td> <td class="tcc rb tb">per sec.</td></tr>
+
+<tr><td class="tcl lb">Na</td> <td class="tcr">45</td> <td class="tcc">&rdquo;</td> <td class="tcc rb">&rdquo;</td>
+ <td class="tcl lb">I</td> <td class="tcr">69</td> <td class="tcc">&rdquo;</td> <td class="tcc rb">&rdquo;</td></tr>
+
+<tr><td class="tcl lb">Li</td> <td class="tcr">36</td> <td class="tcc">&rdquo;</td> <td class="tcc rb">&rdquo;</td>
+ <td class="tcl lb">NO<span class="su">3</span></td> <td class="tcr">64</td> <td class="tcc">&rdquo;</td> <td class="tcc rb">&rdquo;</td></tr>
+
+<tr><td class="tcl lb">NH<span class="su">4</span></td> <td class="tcr">66</td> <td class="tcc">&rdquo;</td> <td class="tcc rb">&rdquo;</td>
+ <td class="tcl lb">OH</td> <td class="tcr">162</td> <td class="tcc">&rdquo;</td> <td class="tcc rb">&rdquo;</td></tr>
+
+<tr><td class="tcl lb">H</td> <td class="tcr">320</td> <td class="tcc">&rdquo;</td> <td class="tcc rb">&rdquo;</td>
+ <td class="tcl lb">C<span class="su">2</span>H<span class="su">3</span>O<span class="su">2</span></td> <td class="tcr">36</td> <td class="tcc">&rdquo;</td> <td class="tcc rb">&rdquo;</td></tr>
+
+<tr><td class="tcl lb bb">Ag</td> <td class="tcr bb">57</td> <td class="tcc bb">&rdquo;</td> <td class="tcc rb bb">&rdquo;</td>
+ <td class="tcl lb bb">C<span class="su">3</span>H<span class="su">5</span>O<span class="su">2</span></td> <td class="tcr bb">33</td> <td class="tcc bb">&rdquo;</td> <td class="tcc rb bb">&rdquo;</td></tr>
+</table>
+
+<p>Having obtained these numbers we can deduce the conductivity
+of the dilute solution of any salt, and the comparison of the
+calculated with the observed values furnished the first confirmation
+of Kohlrausch&rsquo;s theory. Some exceptions, however, are
+known. Thus acetic acid and ammonia give solutions of much
+lower conductivity than is indicated by the sum of the specific
+ionic velocities of their ions as determined from other compounds.
+An attempt to find in Kohlrausch&rsquo;s theory some explanation of
+this discrepancy shows that it could be due to one of two causes.
+Either the velocities of the ions must be much less in these
+solutions than in others, or else only a fractional part of the
+number of molecules present can be actively concerned in conveying
+the current. We shall return to this point later.</p>
+
+<div class="condensed">
+<p><i>Friction on the Ions.</i>&mdash;It is interesting to calculate the magnitude
+of the forces required to drive the ions with a certain velocity. If
+we have a potential gradient of 1 volt per centimetre the electric
+force is 10<span class="sp">8</span> in C.G.S. units. The charge of electricity on 1 gram-equivalent
+of any ion is 1/.0001036 = 9653 units, hence the mechanical
+force acting on this mass is 9653&times;10<span class="sp">8</span> dynes. This, let us say,
+produces a velocity u; then the force required to produce unit
+velocity is P<span class="su">A</span> = 9.653&times;10<span class="sp">11</span>/u dynes = 9.84&times;10<span class="sp">5</span>/u kilograms-weight.
+If the ion have an equivalent weight A, the force producing unit velocity
+when acting on 1 gram is P<span class="su">1</span> = 9.84&times;10<span class="sp">5</span>/Au kilograms-weight.
+Thus the aggregate force required to drive 1 gram of potassium ions with
+a velocity of 1 centimetre per second through a very dilute solution
+must be equal to the weight of 38 million kilograms.</p>
+
+<p class="center1">Table XI.</p>
+
+<table class="ws" summary="Contents">
+<tr><td class="tcc allb" colspan="3">Kilograms-weight.</td> <td class="tcc allb" colspan="3">Kilograms-weight.</td></tr>
+
+<tr><td class="lb">&nbsp;</td> <td class="tcc">P<span class="su">A</span></td> <td class="tcc rb">P<span class="su">1</span></td>
+ <td>&nbsp;</td> <td class="tcc">P<span class="su">A</span></td> <td class="tcc rb">P<span class="su">1</span></td></tr>
+
+<tr><td class="tcl lb">K</td> <td class="tcl">15&times;10<span class="sp">8</span></td> <td class="tcl rb">&ensp;38&times;10<span class="sp">6</span></td>
+ <td class="tcl">Cl</td> <td class="tcl">14&emsp;10<span class="sp">8</span></td> <td class="tcl rb">40&times;10<span class="sp">6</span></td></tr>
+
+<tr><td class="tcl lb">Na</td> <td class="tcl">22&emsp;&rdquo;</td> <td class="tcl rb">&ensp;95&emsp;&rdquo;</td>
+ <td class="tcl">I</td> <td class="tcl">14&emsp;&rdquo;</td> <td class="tcl rb">11&emsp;&rdquo;</td></tr>
+
+<tr><td class="tcl lb">Li</td> <td class="tcl">27&emsp;&rdquo;</td> <td class="tcl rb">390&emsp;&rdquo;</td>
+ <td class="tcl">NO<span class="su">3</span></td> <td class="tcl">15&emsp;&rdquo;</td> <td class="tcl rb">25&emsp;&rdquo;</td></tr>
+
+<tr><td class="tcl lb">NH<span class="su">4</span></td> <td class="tcl">15&emsp;&rdquo;</td> <td class="tcl rb">&ensp;83&emsp;&rdquo;</td>
+ <td class="tcl">OH</td> <td class="tcl">&ensp;5.4&ensp;&rdquo;</td> <td class="tcl rb">32&emsp;&rdquo;</td></tr>
+
+<tr><td class="tcl lb">H</td> <td class="tcl">&ensp;3.1&ensp;&rdquo;</td> <td class="tcl rb">310&emsp;&rdquo;</td>
+ <td class="tcl">C<span class="su">2</span>H<span class="su">8</span>O<span class="su">2</span></td> <td class="tcl">27&emsp;&rdquo;</td> <td class="tcl rb">46&emsp;&rdquo;</td></tr>
+
+<tr><td class="tcl lb bb">Ag</td> <td class="tcl bb">17&emsp;&rdquo;</td> <td class="tcl rb bb">&ensp;16&emsp;&rdquo;</td>
+ <td class="tcl bb">C<span class="su">3</span>H<span class="su">5</span>O<span class="su">2</span></td> <td class="tcl bb">30&emsp;&rdquo;</td> <td class="tcl rb bb">41&emsp;&rdquo;</td></tr>
+</table>
+
+<p>Since the ions move with uniform velocity, the frictional resistances
+brought into play must be equal and opposite to the driving
+forces, and therefore these numbers also represent the ionic friction
+coefficients in very dilute solutions at 18° C.</p>
+</div>
+
+<p><i>Direct Measurement of Ionic Velocities.</i>&mdash;Sir Oliver Lodge was
+the first to directly measure the velocity of an ion (<i>B.A. Report</i>,
+1886, p. 389). In a horizontal glass tube connecting two vessels
+filled with dilute sulphuric acid he placed a solution of sodium
+chloride in solid agar-agar jelly. This solid solution was made
+alkaline with a trace of caustic soda in order to bring out the red
+colour of a little phenol-phthalein added as indicator. An
+electric current was then passed from one vessel to the other. The
+hydrogen ions from the anode vessel of acid were thus carried
+along the tube, and, as they travelled, decolourized the phenol-phthalein.
+By this method the velocity of the hydrogen ion
+through a jelly solution under a known potential gradient was
+observed to about 0.0026 cm. per sec, a number of the same
+order as that required by Kohlrausch&rsquo;s theory. Direct determinations
+of the velocities of a few other ions have been made by
+W. C. D. Whetham (<i>Phil. Trans.</i> vol. 184, A, p. 337; vol. 186, A,
+p. 507; <i>Phil. Mag.</i>, October 1894). Two solutions having one
+ion in common, of equivalent concentrations, different densities,
+different colours, and nearly equal specific resistances, were
+placed one over the other in a vertical glass tube. In one case,
+for example, decinormal solutions of potassium carbonate and
+potassium bichromate were used. The colour of the latter is due
+to the presence of the bichromate group, Cr<span class="su">2</span>O<span class="su">7</span>. When a current
+was passed across the junction, the anions CO<span class="su">3</span> and Cr<span class="su">2</span>O<span class="su">7</span>
+travelled in the direction opposite to that of the current, and
+their velocity could be determined by measuring the rate at which
+the colour boundary moved. Similar experiments were made
+with alcoholic solutions of cobalt salts, in which the velocities of
+the ions were found to be much less than in water. The behaviour
+of agar jelly was then investigated, and the velocity of an ion
+through a solid jelly was shown to be very little less than in
+an ordinary liquid solution. The velocities could therefore be
+measured by tracing the change in colour of an indicator or the
+formation of a precipitate. Thus decinormal jelly solutions of
+barium chloride and sodium chloride, the latter containing a trace
+of sodium sulphate, were placed in contact. Under the influence
+of an electromotive force the barium ions moved up the
+tube, disclosing their presence by the trace of insoluble barium
+sulphate formed. Again, a measurement of the velocity of
+the hydrogen ion, when travelling through the solution of an
+acetate, showed that its velocity was then only about the
+one-fortieth part of that found during its passage through
+chlorides. From this, as from the measurements on alcohol
+solutions, it is clear that where the equivalent conductivities are
+very low the effective velocities of the ions are reduced in the
+same proportion.</p>
+
+<p>Another series of direct measurements has been made by Orme
+Masson (<i>Phil. Trans.</i> vol. 192, A, p. 331). He placed the gelatine
+solution of a salt, potassium chloride, for example, in a horizontal
+glass tube, and found the rate of migration of the potassium and
+chlorine ions by observing the speed at which they were replaced
+when a coloured anion, say, the Cr<span class="su">2</span>O<span class="su">7</span> from a solution of potassium
+bichromate, entered the tube at one end, and a coloured cation,
+say, the Cu from copper sulphate, at the other. The coloured
+ions are specifically slower than the colourless ions which they
+follow, and in this case it follows that the coloured solution has a
+<span class="pagenum"><a name="page864" id="page864"></a>864</span>
+higher resistance than the colourless. For the same current,
+therefore, the potential gradient is higher in the coloured solution
+and lower in the colourless one. Thus a coloured ion which gets
+in front of the advancing boundary finds itself acted on by a
+smaller force and falls back into line, while a straggling colourless
+ion is pushed forward again. Hence a sharp boundary is preserved.
+B. D. Steele has shown that with these sharp boundaries
+the use of coloured ions is unnecessary, the junction line being
+visible owing to the difference in the optical refractive indices of
+two colourless solutions. Once the boundary is formed, too, no
+gelatine is necessary, and the motion can be watched through
+liquid aqueous solutions (see R. B. Denison and B. D. Steele,
+<i>Phil. Trans.</i>, 1906).</p>
+
+<p>All the direct measurements which have been made on simple
+binary electrolytes agree with Kohlrausch&rsquo;s results within the
+limits of experimental error. His theory, therefore, probably
+holds good in such cases, whatever be the solvent, if the proper
+values are given to the ionic velocities, <i>i.e.</i> the values expressing
+the velocities with which the ions actually move in the solution
+of the strength taken, and under the conditions of the experiment.
+If we know the specific velocity of any one ion, we can deduce,
+from the conductivity of very dilute solutions, the velocity of any
+other ion with which it may be associated, a proceeding which
+does not involve the difficult task of determining the migration
+constant of the compound. Thus, taking the specific ionic
+velocity of hydrogen as 0.00032 cm. per second, we can find, by
+determining the conductivity of dilute solutions of any acid, the
+specific velocity of the acid radicle involved. Or again, since we
+know the specific velocity of silver, we can find the velocities of a
+series of acid radicles at great dilution by measuring the conductivity
+of their silver salts.</p>
+
+<div class="condensed">
+<p>By such methods W. Ostwald, G. Bredig and other observers have
+found the specific velocities of many ions both of inorganic and
+organic compounds, and examined the relation between constitution
+and ionic velocity. The velocity of elementary ions is found to
+be a periodic function of the atomic weight, similar elements lying
+on corresponding portions of a curve drawn to express the relation
+between these two properties. Such a curve much resembles that
+giving the relation between atomic weight and viscosity in solution.
+For complex ions the velocity is largely an additive property; to
+a continuous additive change in the composition of the ion corresponds
+a continuous but decreasing change in the velocity. The
+following table gives Ostwald&rsquo;s results for the formic acid series:&mdash;</p>
+
+<p class="center1">Table XII.</p>
+
+<table class="ws" summary="Contents">
+<tr><td class="tcc allb" colspan="2">&nbsp;</td>
+ <td class="tcc allb">Velocity.</td> <td class="tcc allb">Difference for CH<span class="su">2</span>.</td></tr>
+
+<tr><td class="tcl lb">Formic acid</td> <td class="tcl rb">HCO<span class="su">2</span></td> <td class="tcc rb">51.2</td> <td class="tcc rb">..</td></tr>
+<tr><td class="tcl lb">Acetic acid</td> <td class="tcl rb">H<span class="su">3</span>C<span class="su">2</span>O<span class="su">2</span></td> <td class="tcc rb">38.3</td> <td class="tcc rb">-12.9</td></tr>
+<tr><td class="tcl lb">Propionic acid</td> <td class="tcl rb">H<span class="su">5</span>C<span class="su">3</span>O<span class="su">2</span></td> <td class="tcc rb">34.3</td> <td class="tcc rb">-&ensp;4.0</td></tr>
+<tr><td class="tcl lb">Butyric acid</td> <td class="tcl rb">H<span class="su">7</span>C<span class="su">4</span>O<span class="su">2</span></td> <td class="tcc rb">30.8</td> <td class="tcc rb">-&ensp;3.5</td></tr>
+<tr><td class="tcl lb">Valeric acid</td> <td class="tcl rb">H<span class="su">9</span>C<span class="su">5</span>O<span class="su">2</span></td> <td class="tcc rb">28.8</td> <td class="tcc rb">-&ensp;2.0</td></tr>
+<tr><td class="tcl lb bb">Caprionic acid</td> <td class="tcl rb bb">H<span class="su">11</span>C<span class="su">6</span>O<span class="su">2</span></td> <td class="tcc rb bb">27.4</td> <td class="tcc rb bb">-&ensp;1.4</td></tr>
+</table>
+</div>
+
+<p><i>Nature of Electrolytes.</i>&mdash;We have as yet said nothing about the
+fundamental cause of electrolytic activity, nor considered why,
+for example, a solution of potassium chloride is a good conductor,
+while a solution of sugar allows practically no current to pass.</p>
+
+<p>All the preceding account of the subject is, then, independent
+of any view we may take of the nature of electrolytes, and stands
+on the basis of direct experiment. Nevertheless, the facts
+considered point to a very definite conclusion. The specific
+velocity of an ion is independent of the nature of the opposite ion
+present, and this suggests that the ions themselves, while
+travelling through the liquid, are dissociated from each other.
+Further evidence, pointing in the same direction, is furnished by
+the fact that since the conductivity is proportional to the
+concentration at great dilution, the equivalent-conductivity, and
+therefore the ionic velocity, is independent of it. The importance
+of this relation will be seen by considering the alternative to the
+dissociation hypothesis. If the ions are not permanently free
+from each other their mobility as parts of the dissolved molecules
+must be secured by continual interchanges. The velocity with
+which they work their way through the liquid must then increase
+as such molecular rearrangements become more frequent, and will
+therefore depend on the number of solute molecules, <i>i.e.</i> on the
+concentration. On this supposition the observed constancy of
+velocity would be impossible. We shall therefore adopt as a
+wording hypothesis the theory, confirmed by other phenomena
+(see <span class="sc"><a href="#artlinks">Electrolysis</a></span>), that an electrolyte consists of dissociated ions.</p>
+
+<p>It will be noticed that neither the evidence in favour of the
+dissociation theory which is here considered, nor that described
+in the article <span class="sc">Electrolysis</span>, requires more than the effective
+dissociation of the ions from each other. They may well be
+connected in some way with solvent molecules, and there are
+several indications that an ion consists of an electrified part of the
+molecule of the dissolved salt with an attendant atmosphere of
+solvent round it. The conductivity of a salt solution depends on
+two factors&mdash;(1) the fraction of the salt ionized; (2) the velocity
+with which the ions, when free from each other, move under the
+electric forces.<a name="FnAnchor_12" id="FnAnchor_12" href="#Footnote_12"><span class="sp">12</span></a> When a solution is heated, both these factors may
+change. The coefficient of ionization usually, though not always,
+decreases; the specific ionic velocities increase. Now the rate of
+increase with temperature of these ionic velocities is very nearly
+identical with the rate of decrease of the viscosity of the liquid.
+If the curves obtained by observations at ordinary temperatures
+be carried on they indicate a zero of fluidity and a zero of ionic
+velocity about the same point, 38.5° C. below the freezing point of
+water (Kohlrausch, <i>Sitz. preuss. Akad. Wiss.</i>, 1901, 42, p. 1026).
+Such relations suggest that the frictional resistance to the motion
+of an ion is due to the ordinary viscosity of the liquid, and that the
+ion is analogous to a body of some size urged through a viscous
+medium rather than to a particle of molecular dimensions finding
+its way through a crowd of molecules of similar magnitude.
+From this point of view W. K. Bousfield has calculated the sizes
+of ions on the assumption that Stokes&rsquo;s theory of the motion of a
+small sphere through a viscous medium might be applied (<i>Zeits.
+phys. Chem.</i>, 1905, 53, p. 257; <i>Phil. Trans.</i> A, 1906, 206, p. 101).
+The radius of the potassium or chlorine ion with its envelope of
+water appears to be about 1.2&times;10<span class="sp">-8</span> centimetres.</p>
+
+<div class="condensed">
+<p>For the bibliography of electrolytic conduction see <span class="sc"><a href="#artlinks">Electrolysis</a></span>.
+The books which deal more especially with the particular subject
+of the present article are <i>Das Leitvermögen der Elektrolyte</i>, by
+F. Kohlrausch and L. Holborn (Leipzig, 1898), and <i>The Theory
+of Solution and Electrolysis</i>, by W. C. D. Whetham (Cambridge,
+1902).</p>
+</div>
+<div class="author">(W. C. D. W.)</div>
+
+<p class="center1">III. <span class="sc">Electric Conduction through Gases</span></p>
+
+<p>A gas such as air when it is under normal conditions conducts
+electricity to a small but only to a very small extent, however
+small the electric force acting on the gas may be. The electrical
+conductivity of gases not exposed to special conditions is so
+small that it was only definitely established in the early years
+of the 20th century, although it had engaged the attention of
+physicists for more than a hundred years. It had been known
+for a long time that a body charged with electricity slowly lost
+its charge even when insulated with the greatest care, and though
+long ago some physicists believed that part of the leak of
+electricity took place through the air, the general view seems to
+have been that it was due to almost unavoidable defects in the
+insulation or to dust in the air, which after striking the charged
+body was repelled from it and went off with some of the charge.
+C. A. Coulomb, who made some very careful experiments which
+were published in 1785 (<i>Mém. de l&rsquo;Acad. des Sciences</i>, 1785, p.
+612), came to the conclusion that after allowing for the leakage
+along the threads which supported the charged body there was
+a balance over, which he attributed to leakage through the air.
+His view was that when the molecules of air come into contact
+with a charged body some of the electricity goes on to the molecules,
+which are then repelled from the body carrying their
+charge with them. We shall see later that this explanation is
+not tenable. C. Matteucci (<i>Ann. chim. phys.</i>, 1850, 28, p. 390)
+in 1850 also came to the conclusion that the electricity from a
+charged body passes through the air; he was the first to prove
+<span class="pagenum"><a name="page865" id="page865"></a>865</span>
+that the rate at which electricity escapes is less when the pressure
+of the gas is low than when it is high. He found that the rate
+was the same whether the charged body was surrounded by air,
+carbonic acid or hydrogen. Subsequent investigations have
+shown that the rate in hydrogen is in general much less than in
+air. Thus in 1872 E. G. Warburg (<i>Pogg. Ann.</i>, 1872, 145, p. 578)
+found that the leak through hydrogen was only about one-half
+of that through air: he confirmed Matteucci&rsquo;s observations on
+the effect of pressure on the rate of leak, and also found that it
+was the same whether the gas was dry or damp. He was inclined
+to attribute the leak to dust in the air, a view which was
+strengthened by an experiment of J. W. Hittorf&rsquo;s (<i>Wied. Ann.</i>,
+1879, 7, p. 595), in which a small carefully insulated electroscope,
+placed in a small vessel filled with carefully filtered gas, retained
+its charge for several days; we know now that this was due to
+the smallness of the vessel and not to the absence of dust, as it
+has been proved that the rate of leak in small vessels is less than
+in large ones.</p>
+
+<p>Great light was thrown on this subject by some experiments
+on the rates of leak from charged bodies in closed vessels made
+almost simultaneously by H. Geitel (<i>Phys. Zeit.</i>, 1900, 2, p. 116)
+and C. T. R. Wilson (<i>Proc. Camb. Phil.</i> Soc., 1900, 11, p. 32).
+These observers established that (1) the rate of escape of electricity
+in a closed vessel is much smaller than in the open, and
+the larger the vessel the greater is the rate of leak; and (2) the rate
+of leak does not increase in proportion to the differences of
+potential between the charged body and the walls of the vessel:
+the rate soon reaches a limit beyond which it does not increase,
+however much the potential difference may be increased, provided,
+of course, that this is not great enough to cause sparks to pass
+from the charged body. On the assumption that the maximum
+leak is proportional to the volume, Wilson&rsquo;s experiments, which
+were made in vessels less than 1 litre in volume, showed that in
+dust-free air at atmospheric pressure the maximum quantity
+of electricity which can escape in one second from a charged
+body in a closed volume of V cubic centimetres is about 10<span class="sp">-8</span>V
+electrostatic units. E. Rutherford and S. T. Allan (<i>Phys. Zeit.</i>,
+1902, 3, p. 225), working in Montreal, obtained results in close
+agreement with this. Working between pressures of from
+43 to 743 millimetres of mercury, Wilson showed that the
+maximum rate of leak is very approximately proportional to
+the pressure; it is thus exceedingly small when the pressure
+is low&mdash;a result illustrated in a striking way by an experiment
+of Sir W. Crookes (<i>Proc. Roy. Soc.</i>, 1879, 28, p. 347) in which a
+pair of gold leaves retained an electric charge for several months
+in a very high vacuum. Subsequent experiments have shown
+that it is only in very small vessels that the rate of leak is proportional
+to the volume and to the pressure; in large vessels
+the rate of leak per unit volume is considerably smaller than in
+small ones. In small vessels the maximum rate of leak in different
+gases, is, with the exception of hydrogen, approximately proportional
+to the density of the gas. Wilson&rsquo;s results on this point are
+shown in the following table (Proc. Roy. Soc., 1901, 60, p. 277):&mdash;</p>
+
+<table class="ws" summary="Contents">
+<tr><td class="tccm allb">Gas.</td> <td class="tccm allb">Relative Rate of Leak.</td>
+ <td class="tccm allb"><i>Rate of Leak.</i><br />Sp. Gr.</td></tr>
+
+<tr><td class="tcl lb rb">Air</td> <td class="tcc rb">1.00&ensp;</td> <td class="tcc rb">1&emsp;</td></tr>
+<tr><td class="tcl lb rb">H<span class="su">2</span></td> <td class="tcc rb">&ensp;.184</td> <td class="tcc rb">2.7&emsp;</td></tr>
+<tr><td class="tcl lb rb">CO<span class="su">2</span></td> <td class="tcc rb">1.69&ensp;</td> <td class="tcc rb">1.10&ensp;</td></tr>
+<tr><td class="tcl lb rb">SO<span class="su">2</span></td> <td class="tcc rb">2.64&ensp;</td> <td class="tcc rb">1.21&ensp;</td></tr>
+<tr><td class="tcl lb rb">CH<span class="su">3</span>Cl</td> <td class="tcc rb">4.7&emsp;</td> <td class="tcc rb">1.09&ensp;</td></tr>
+<tr><td class="tcl lb rb bb">Ni(CO)<span class="su">4</span></td> <td class="tcc rb bb">5.1&emsp;</td> <td class="tcc rb bb">&ensp;.867</td></tr>
+</table>
+
+<p>The rate of leak of electricity through gas contained in a closed
+vessel depends to some extent on the material of which the walls
+of the vessel are made; thus it is greater, other circumstances
+being the same, when the vessel is made of lead than when it is
+made of aluminium. It also varies, as Campbell and Wood
+(<i>Phil. Mag.</i> [6], 13, p. 265) have shown, with the time of the day,
+having a well-marked minimum at about 3 o&rsquo;clock in the morning:
+it also varies from month to month. Rutherford (<i>Phys. Rev.</i>,
+1903, 16, p. 183), Cooke (<i>Phil. Mag.</i>, 1903 [6], 6, p. 403) and
+M&rsquo;Clennan and Burton (<i>Phys. Rev.</i>, 1903, 16, p. 184) have shown
+that the leak in a closed vessel can be reduced by about 30%
+by surrounding the vessel with sheets of thick lead, but that the
+reduction is not increased beyond this amount, however thick
+the lead sheets may be. This result indicates that part of the
+leak is due to a very penetrating kind of radiation, which can get
+through the thin walls of the vessel but is stopped by the thick
+lead. A large part of the leak we are describing is due to the
+presence of radioactive substances such as radium and thorium
+in the earth&rsquo;s crust and in the walls of the vessel, and to the
+gaseous radioactive emanations which diffuse from them into
+the atmosphere. This explains the very interesting effect
+discovered by J. Elster and H. Geitel (<i>Phys. Zeit.</i>, 1901, 2, p. 560),
+that the rate of leak in caves and cellars when the air is stagnant
+and only renewed slowly is much greater than in the open air.
+In some cases the difference is very marked; thus they found
+that in the cave called the Baumannshöhle in the Harz mountains
+the electricity escaped at seven times the rate it did in the air
+outside. In caves and cellars the radioactive emanations from
+the walls can accumulate and are not blown away as in the
+open air.</p>
+
+<p>The electrical conductivity of gases in the normal state is,
+as we have seen, exceedingly small, so small that the investigation
+of its properties is a matter of considerable difficulty; there
+are, however, many ways by which the electrical conductivity
+of a gas can be increased so greatly that the investigation
+becomes comparatively easy. Among such methods are raising
+the temperature of the gas above a certain point. Gases drawn
+from the neighbourhood of flames, electric arcs and sparks, or
+glowing pieces of metal or carbon are conductors, as are also
+gases through which Röntgen or cathode rays or rays of positive
+electricity are passing; the rays from the radioactive metals,
+radium, thorium, polonium and actinium, produce the same
+effect, as does also ultra-violet light of exceedingly short wave-length.
+The gas, after being made a conductor of electricity
+by any of these means, is found to possess certain properties;
+thus it retains its conductivity for some little time after the agent
+which made it a conductor has ceased to act, though the conductivity
+diminishes very rapidly and finally gets too small
+to be appreciable.</p>
+
+<table class="nobctr" summary="Illustration">
+<tr><td class="figcenter"><img style="width:474px; height:186px" src="images/img865.jpg" alt="" /></td></tr>
+<tr><td class="caption"><span class="sc">Fig. 5.</span></td></tr></table>
+
+<p>This and several other properties of conducting gas may
+readily be proved by the aid of the apparatus represented in fig. 5.
+V is a testing vessel in which an electroscope is placed. Two tubes
+A and C are fitted into the vessel, A being connected with a water
+pump, while the far end of C is in the region where the gas is
+exposed to the agent which makes it a conductor of electricity.
+Let us suppose that the gas is made conducting by Röntgen rays
+produced by a vacuum tube which is placed in a box, covered
+except for a window at B with lead so as to protect the electroscope
+from the direct action of the rays. If a slow current of air
+is drawn by the water pump through the testing vessel, the charge
+on the electroscope will gradually leak away. The leak, however,
+ceases when the current of air is stopped. This result shows that
+the gas retains its conductivity during the time taken by it to pass
+from one end to the other of the tube C.</p>
+
+<p>The gas loses its conductivity when filtered through a plug of
+glass-wool, or when it is made to bubble through water. This
+can readily be proved by inserting in the tube C a plug of glass-wool
+or a water trap; then if by working the pump a little
+harder the same current of air is produced as before, it will be
+found that the electroscope will now retain its charge, showing
+that the conductivity can, as it were, be filtered out of the gas.
+<span class="pagenum"><a name="page866" id="page866"></a>866</span>
+The conductivity can also be removed from the gas by making
+the gas traverse a strong electric field. We can show this by
+replacing the tube C by a metal tube with an insulated wire
+passing down the axis of the tube. If there is no potential
+difference between the wire and the tube then the electroscope
+will leak when a current of air is drawn through the vessel, but
+the leak will stop if a considerable difference of potential is
+maintained between the wire and the tube: this shows that a
+strong electric field removes the conductivity from the gas.</p>
+
+<p>The fact that the conductivity of the gas is removed by
+filtering shows that it is due to something mixed with the gas
+which is removed from it by filtration, and since the conductivity
+is also removed by an electric field, the cause of the conductivity
+must be charged with electricity so as to be driven to the sides
+of the tube by the electric force. Since the gas as a whole is not
+electrified either positively or negatively, there must be both
+negative and positive charges in the gas, the amount of electricity
+of one sign being equal to that of the other. We are thus led to
+the conclusion that the conductivity of the gas is due to electrified
+particles being mixed up with the gas, some of these particles
+having charges of positive electricity, others of negative. These
+electrified particles are called <i>ions</i>, and the process by which the
+gas is made a conductor is called the ionization of the gas. We
+shall show later that the charges and masses of the ions can be
+determined, and that the gaseous ions are not identical with
+those met with in the electrolysis of solutions.</p>
+
+<table class="nobctr" style="float: left; width: 280px;" summary="Illustration">
+<tr><td class="figleft1"><img style="width:228px; height:225px" src="images/img866a.jpg" alt="" /></td></tr>
+<tr><td class="caption"><span class="sc">Fig. 6.</span></td></tr></table>
+
+<p>One very characteristic property of conduction of electricity
+through a gas is the relation between the current through the
+gas and the electric force which gave rise to it. This relation
+is not in general that expressed by Ohm&rsquo;s law, which always,
+as far as our present knowledge extends, expresses the relation
+for conduction through metals and electrolytes. With gases, on
+the other hand, it is only when the current is very small that
+Ohm&rsquo;s law is true. If we represent graphically by means of a
+curve the relation between the current passing between two
+parallel metal plates separated by ionized gas and the difference
+of potential between the plates, the curve is of the character
+shown in fig. 6 when the ordinates represent the current and
+the abscissae the difference of potential between the plates.
+We see that when the potential difference is very small, <i>i.e.</i>
+close to the origin, the curve is approximately straight, but that
+soon the current increases much
+less rapidly than the potential
+difference, and that a stage is
+reached when no appreciable
+increase of current is produced
+when the potential difference is
+increased; when this stage is
+reached the current is constant,
+and this value of the current is
+called the &ldquo;saturation&rdquo; value.
+When the potential difference
+approaches the value at which
+sparks would pass through the
+gas, the current again increases with the potential difference;
+thus the curve representing the relation between the current
+and potential difference over very wide ranges of potential
+difference has the shape shown in fig. 7; curves of this kind
+have been obtained by von Schweidler (<i>Wien. Ber.</i>, 1899,
+108, p. 273), and J. E. S. Townsend (<i>Phil. Mag.</i>, 1901 [6], 1,
+p. 198). We shall discuss later the causes of the rise in the
+current with large potential differences, when we consider
+ionization by collision.</p>
+
+<div class="condensed">
+<p>The general features of the earlier part of the curve are readily
+explained on the ionization hypothesis. On this view the Röntgen
+rays or other ionizing agent acting on the gas between the plates,
+produces positive and negative ions at a definite rate. Let us suppose
+that q positive and q negative ions are by this means produced
+per second between the plates; these under the electric force will
+tend to move, the positive ones to the negative plate, the negative
+ones to the positive. Some of these ions will reach the plate, others
+before reaching the plate will get so near one of the opposite sign that
+the attraction between them will cause them to unite and form an
+electrically neutral system; when they do this they end their
+existence as ions. The current between the plates is proportional
+to the number of ions which reach the plates per second. Now it is
+evident that we cannot go on taking more ions out of the gas than
+are produced; thus we cannot, when the current is steady, have
+more than q positive ions driven to the negative plate per second,
+and the same number of negative ions to the positive. If each of the
+positive ions carries a charge of e units of positive electricity, and
+if there is an equal and opposite charge on each negative ion, then
+the maximum amount of electricity which can be given to the plates
+per second is qe, and this is equal to the saturation current. Thus
+if we measure the saturation current, we get a direct measure of the
+ionization, and this does not require us to know the value of any
+quantity except the constant charge on the ion. If we attempted
+to deduce the amount of ionization by measurements of the current
+before it was saturated, we should require to know in addition the
+velocity with which the ions move under a given electric force, the
+time that elapses between the liberation of an ion and its combination
+with one of the opposite sign, and the potential difference
+between the plates. Thus if we wish to measure the amount of
+ionization in a gas we should be careful to see that the current is
+saturated.</p>
+</div>
+
+<table class="nobctr" summary="Illustration">
+<tr><td class="figcenter"><img style="width:484px; height:228px" src="images/img866b.jpg" alt="" /></td></tr>
+<tr><td class="caption"><span class="sc">Fig. 7.</span></td></tr></table>
+
+<p>The difference between conduction through gases and through
+metals is shown in a striking way when we use potential differences
+large enough to produce the saturation current. Suppose we
+have got a potential difference between the plates more than
+sufficient to produce the saturation current, and let us increase
+the distance between the plates. If the gas were to act like a
+metallic conductor this would diminish the current, because the
+greater length would involve a greater resistance in the circuit.
+In the case we are considering the separation of the plates will
+<i>increase</i> the current, because now there is a larger volume of gas
+exposed to the rays; there are therefore more ions produced,
+and as the saturation current is proportional to the number of
+ions the saturation current is increased. If the potential difference
+between the plates were much less than that required to
+saturate the current, then increasing the distance would diminish
+the current; the gas for such potential differences obeys Ohm&rsquo;s
+law and the behaviour of the gaseous resistance is therefore
+similar to that of a metallic one.</p>
+
+<p>In order to produce the saturation current the electric field
+must be strong enough to drive each ion to the electrode before
+it has time to enter into combination with one of the opposite
+sign. Thus when the plates in the preceding example are far
+apart, it will take a larger potential difference to produce this
+current than when the plates are close together. The potential
+difference required to saturate the current will increase as the
+square of the distance between the plates, for if the ions are to
+be delivered in a given time to the plates their speed must be
+proportional to the distance between the plates. But the speed
+is proportional to the electric force acting on the ion; hence the
+electric force must be proportional to the distance between the
+plates, and as in a uniform field the potential difference is equal
+to the electric force multiplied by the distance between the plates,
+the potential difference will vary as the square of this distance.</p>
+
+<p>The potential difference required to produce saturation will,
+other circumstances being the same, increase with the amount
+of ionization, for when the number of ions is large and they are
+crowded together, the time which will elapse before a positive
+one combines with a negative will be smaller than when the
+number of ions is small. The ions have therefore to be removed
+more quickly from the gas when the ionization is great than
+when it is small; thus they must move at a higher speed and
+must therefore be acted upon by a larger force.</p>
+
+<p><span class="pagenum"><a name="page867" id="page867"></a>867</span></p>
+
+<p>When the ions are not removed from the gas, they will increase
+until the number of ions of one sign which combine with ions
+of the opposite sign in any time is equal to the number produced
+by the ionizing agent in that time. We can easily calculate the
+number of free ions at any time after the ionizing agent has
+commenced to act.</p>
+
+<div class="condensed">
+<p>Let q be the number of ions (positive or negative) produced in
+one cubic centimetre of the gas per second by the ionizing agent,
+n<span class="su">1</span>, n<span class="su">2</span>, the number of free positive and negative ions respectively per
+cubic centimetre of the gas. The number of collisions between
+positive and negative ions per second in one cubic centimetre of the
+gas is proportional to n<span class="su">1</span>n<span class="su">2</span>. If a certain fraction of the collisions
+between the positive and negative ions result in the formation of an
+electrically neutral system, the number of ions which disappear per
+second on a cubic centimetre will be equal to &alpha;n<span class="su">1</span> n<span class="su">2</span>, where &alpha; is a
+quantity which is independent of n<span class="su">1</span>, n<span class="su">2</span>; hence if t is the time since
+the ionizing agent was applied to the gas, we have</p>
+
+<p class="center1">dn<span class="su">1</span>/dt = q &minus; &alpha;n<span class="su">1</span> n<span class="su">2</span>, dn<span class="su">2</span>/dt = q &minus; &alpha;n<span class="su">1</span> n<span class="su">2</span>.</p>
+
+<p class="noind">Thus n<span class="su">1</span> &minus; n<span class="su">2</span> is constant, so if the gas is uncharged to begin with, n<span class="su">1</span>
+will always equal n<span class="su">2</span>. Putting n<span class="su">1</span> = n<span class="su">2</span> = n we have</p>
+
+<p class="center1">dn/dt = q &minus; &alpha;n<span class="sp">2</span>&emsp;&emsp;(1),</p>
+
+<p class="noind">the solution of which is, since n = 0 when t = 0,</p>
+
+<table class="math0" summary="math">
+<tr> <td rowspan="2">n =</td> <td>k(&epsilon;<span class="sp">2k&alpha;t</span> &minus; 1)</td>
+ <td rowspan="2">&emsp;&emsp;(2),</td></tr>
+<tr> <td class="denom">&epsilon;<span class="sp">2k&alpha;t</span> + 1</td></tr></table>
+
+<p class="noind">if k<span class="sp">2</span> = q/&alpha;. Now the number of ions when the gas has reached a
+steady state is got by putting t equal to infinity in the preceding
+equation, and is therefore given by the equation</p>
+
+<p class="center1">n<span class="su">0</span> = k = &radic; (q/&alpha;).</p>
+
+<p class="noind">We see from equation (1) that the gas will not approximate to its
+steady state until 2k&alpha;t is large, that is until t is large compared with
+½k&alpha; or with ½&radic; (q&alpha;). We may thus take ½&radic; (q&alpha;) as a measure of
+the time taken by the gas to reach a steady state when exposed to
+an ionizing agent; as this time varies inversely as &radic;q we see that
+when the ionization is feeble it may take a very considerable time for
+the gas to reach a steady state. Thus in the case of our atmosphere
+where the production of ions is only at the rate of about 30 per cubic
+centimetre per second, and where, as we shall see, &alpha; is about 10<span class="sp">-6</span>,
+it would take some minutes for the ionization in the air to get into
+a steady state if the ionizing agent were suddenly applied.</p>
+
+<p>We may use equation (1) to determine the rate at which the ions
+disappear when the ionizing agent is removed. Putting q=0 in
+that equation we get dn/&alpha;t = -&alpha;n<span class="sp">2</span>.</p>
+
+<p class="noind">Hence</p>
+
+<p class="center1">n = n<span class="su">0</span>/(1 + n<span class="su">0</span> &alpha;t)&emsp;&emsp;(3),</p>
+
+<p class="noind">where n<span class="su">0</span> is the number of ions when t = 0. Thus the number of ions
+falls to one-half its initial value in the time 1/n<span class="su">0</span>&alpha;. The quantity &alpha; is
+called the <i>coefficient of recombination</i>, and its value for different gases
+has been determined by Rutherford (<i>Phil. Mag.</i> 1897 [5], 44, p. 422),
+Townsend (<i>Phil. Trans.</i>, 1900, 193, p. 129), McClung (<i>Phil. Mag.</i>,
+1902 [6], 3, p. 283), Langevin (<i>Ann. chim. phys.</i> [7], 28, p. 289),
+Retschinsky (<i>Ann. d. Phys.</i>, 1905, 17, p. 518), Hendred (<i>Phys. Rev.</i>,
+1905, 21, p. 314). The values of &alpha;/e, e being the charge on an ion in
+electrostatic measure as determined by these observers for different
+gases, is given in the following table:&mdash;</p>
+
+<table class="ws" summary="Contents">
+<tr><td class="tcc allb">&nbsp;</td> <td class="tcc allb">Townsend.</td> <td class="tcc allb">McClung.</td> <td class="tcc allb">Langevin.</td> <td class="tcc allb">Retschinsky.</td> <td class="tcc allb">Hendred.</td></tr>
+
+<tr><td class="tcl lb rb">Air</td> <td class="tcc rb">3420</td> <td class="tcc rb">3380</td> <td class="tcc rb">3200</td> <td class="tcc rb">4140</td> <td class="tcc rb">3500</td></tr>
+<tr><td class="tcl lb rb">O<span class="su">2</span></td> <td class="tcc rb">3380</td> <td class="tcc rb">&nbsp;</td> <td class="tcc rb">&nbsp;</td> <td class="tcc rb">&nbsp;</td> <td class="tcc rb">&nbsp;</td></tr>
+<tr><td class="tcl lb rb">CO<span class="su">2</span></td> <td class="tcc rb">3500</td> <td class="tcc rb">3490</td> <td class="tcc rb">3400</td> <td class="tcc rb">&nbsp;</td> <td class="tcc rb">&nbsp;</td></tr>
+<tr><td class="tcl lb rb bb">H<span class="su">2</span></td> <td class="tcc rb bb">3020</td> <td class="tcc rb bb">2940</td> <td class="tcc rb bb">&nbsp;</td> <td class="tcc rb bb">&nbsp;</td> <td class="tcc rb bb">&nbsp;</td></tr>
+</table>
+
+<p class="noind">The gases in these experiments were carefully dried and free from
+dust; the apparent value of &alpha; is much increased when dust or small
+drops of water are present in the gas, for then the ions get caught
+by the dust particles, the mass of a particle is so great compared
+with that of an ion that they are practically immovable under the
+action of the electric field, and so the ions clinging to them escape
+detection when electrical methods are used. Taking e as 3.5&times;10<span class="sp">-10</span>,
+we see that &alpha; is about 1.2&times;10<span class="sp">-6</span>, so that the number of recombinations
+in unit time between n positive and n negative ions in unit
+volume is 1.2&times;10<span class="sp">-6</span>n<span class="sp">2</span>. The kinetic theory of gases shows that
+if we have n molecules of air per cubic centimetre, the number of
+collisions per second is 1.2&times;10<span class="sp">-10</span>n<span class="sp">2</span> at a temperature of 0° C. Thus
+we see that the number of recombinations between oppositely
+charged ions is enormously greater than the number of collisions
+between the same number of neutral molecules. We shall see that
+the difference in size between the ion and the molecule is not nearly
+sufficient to account for the difference between the collisions in the
+two cases; the difference is due to the force between the oppositely
+charged ions, which drags ions into collisions which but for this force
+would have missed each other.</p>
+
+<p>Several methods have been used to measure &alpha;. In one method
+air, exposed to some ionizing agent at one end of a long tube, is
+slowly sucked through the tube and the saturation current measured
+at different points along the tube. These currents are proportional
+to the values of n at the place of observation: if we know the
+distance of this place from the end of the tube when the gas was
+ionized and the velocity of the stream of gas, we can find t in equation
+(3), and knowing the value of n we can deduce the value of &alpha; from
+the equation</p>
+
+<p class="center1">1/n<span class="su">1</span> &minus; 1/n<span class="su">2</span> = &alpha;(t<span class="su">1</span> &minus; t<span class="su">2</span>),</p>
+
+<p class="noind">where n<span class="su">1</span>, n<span class="su">2</span> are the values of n at the times t<span class="su">1</span>, t<span class="su">2</span> respectively. In this
+method the tubes ought to be so wide that the loss of ions by diffusion
+to the sides of the tube is negligible. There are other methods which
+involve the knowledge of the speed with which the ions move under
+the action of known electric forces; we shall defer the consideration
+of these methods until we have discussed the question of these
+speeds.</p>
+
+<p>In measuring the value of &alpha; it should be remembered that the
+theory of the methods supposes that the ionization is uniform
+throughout the gas. If the total ionization throughout a gas remains
+constant, but instead of being uniformly distributed is concentrated
+in patches, it is evident that the ions will recombine more quickly
+in the second case than in the first, and that the value of &alpha; will be
+different in the two cases. This probably explains the large values
+of &alpha; obtained by Retschinsky, who ionized the gas by the &alpha; rays
+from radium, a method which produces very patchy ionization.</p>
+
+<p><i>Variation of &alpha; with the Pressure of the Gas.</i>&mdash;All observers agree
+that there is little variation in &alpha; with the pressures for pressures of
+between 5 and 1 atmospheres; at lower pressures, however, the
+value of &alpha; seems to diminish with the pressure: thus Langevin
+(<i>Ann. chim. phys.</i>, 1903, 28, p. 287) found that at a pressure of <span class="spp">1</span>&frasl;<span class="suu">5</span>
+of an atmosphere the value of &alpha; was about <span class="spp">1</span>&frasl;<span class="suu">5</span> of its value at
+atmospheric pressure.</p>
+
+<p><i>Variation of &alpha; with the Temperature.</i>&mdash;Erikson (<i>Phil. Mag.</i>, Aug.
+1909) has shown that the value of &alpha; for air increases as the temperature
+diminishes, and that at the temperature of liquid air -180° C.,
+it is more than twice as great as at +12° C.</p>
+
+<p>Since, as we have seen, the recombination is due to the coming
+together of the positive and negative ions under the influence of the
+electrical attraction between them, it follows that a large electric
+force sufficient to overcome this attraction would keep the ions apart
+and hence diminish the coefficient of recombination. Simple considerations,
+however, will show that it would require exceedingly
+strong electric fields to produce an appreciable effect. The value of
+&alpha; indicates that for two oppositely charged ions to unite they must
+come within a distance of about 1.5&times;10<span class="sp">-6</span> centimetres; at this
+distance the attraction between them is e<span class="sp">2</span>&times;10<span class="sp">12</span>/2.25, and if X is the
+external electric force, the force tending to pull them apart cannot
+be greater than Xe; if this is to be comparable with the attraction,
+X must be comparable with e&times;10<span class="sp">12</span>/2.25, or putting e = 4&times;10<span class="sp">-10</span>,
+with 1.8&times;10<span class="sp">2</span>; this is 54,000 volts per centimetre, a force which
+could not be applied to gas at atmospheric pressure without producing
+a spark.</p>
+
+<p><i>Diffusion of the Ions.</i>&mdash;The ionized gas acts like a mixture of gases,
+the ions corresponding to two different gases, the non-ionized gas
+to a third. If the concentration of the ions is not uniform, they will
+diffuse through the non-ionized gas in such a way as to produce a
+more uniform distribution. A very valuable series of determinations
+of the coefficient of diffusion of ions through various gases has been
+made by Townsend (<i>Phil. Trans.</i>, 1900, A, 193, p. 129). The method
+used was to suck the ionized gas through narrow tubes; by measuring
+the loss of both the positive and negative ions after the gases
+had passed through a known length of tube, and allowing for the loss
+by recombination, the loss by diffusion and hence the coefficient of
+diffusion could be determined. The following tables give the values
+of the coefficients of diffusion D on the C.G.S. system of units as
+determined by Townsend:&mdash;</p>
+
+<p class="center1">Table I.&mdash;<i>Coefficients of Diffusion (D) in Dry Gases.</i></p>
+
+<table class="ws" summary="Contents">
+<tr><td class="tcc allb">Gas.</td> <td class="tcc allb">D for +ions.</td> <td class="tcc allb">D for -ions.</td> <td class="tcc allb">Mean Value<br />of D.</td> <td class="tcc allb">Ratio of D for<br />- to D for +ions.</td></tr>
+
+<tr><td class="tcl lb rb">Air</td> <td class="tcc rb">.028</td> <td class="tcc rb">.043&ensp;</td> <td class="tcc rb">.0347</td> <td class="tcc rb">1.54</td></tr>
+<tr><td class="tcl lb rb">O<span class="su">2</span></td> <td class="tcc rb">.025</td> <td class="tcc rb">.0396</td> <td class="tcc rb">.0323</td> <td class="tcc rb">1.58</td></tr>
+<tr><td class="tcl lb rb">CO<span class="su">2</span></td> <td class="tcc rb">.023</td> <td class="tcc rb">.026</td> <td class="tcc rb">.0245</td> <td class="tcc rb">1.13</td></tr>
+<tr><td class="tcl lb rb bb">H<span class="su">2</span></td> <td class="tcc rb bb">.123</td> <td class="tcc rb bb">.190</td> <td class="tcc rb bb">.156&ensp;</td> <td class="tcc rb bb">1.54</td></tr>
+</table>
+
+<p class="center1">Table II.&mdash;<i>Coefficients of Diffusion in Moist Gases.</i></p>
+
+<table class="ws" summary="Contents">
+<tr><td class="tcc allb">Gas.</td> <td class="tcc allb">D for +ions.</td> <td class="tcc allb">D for -ions.</td> <td class="tcc allb">Mean Value<br />of D.</td> <td class="tcc allb">Ratio of D for<br />- to D for +ions.</td></tr>
+
+<tr><td class="tcl lb rb">Air</td> <td class="tcc rb">.032&ensp;</td> <td class="tcc rb">.037&ensp;</td> <td class="tcc rb">.0335</td> <td class="tcc rb">1.09</td></tr>
+<tr><td class="tcl lb rb">O<span class="su">2</span></td> <td class="tcc rb">.0288</td> <td class="tcc rb">.0358</td> <td class="tcc rb">.0323</td> <td class="tcc rb">1.24</td></tr>
+<tr><td class="tcl lb rb">CO<span class="su">2</span></td> <td class="tcc rb">.0245</td> <td class="tcc rb">.0255</td> <td class="tcc rb">.025&ensp;</td> <td class="tcc rb">1.04</td></tr>
+<tr><td class="tcl lb rb bb">H<span class="su">2</span></td> <td class="tcc rb bb">.128&ensp;</td> <td class="tcc rb bb">.142&ensp;</td> <td class="tcc rb bb">.135&ensp;</td> <td class="tcc rb bb">1.11</td></tr>
+</table>
+
+<p class="noind">It is interesting to compare with these coefficients the values of D
+when various gases diffuse through each other. D for hydrogen
+through air is .634, for oxygen through air .177, for the vapour of
+<span class="pagenum"><a name="page868" id="page868"></a>868</span>
+isobutyl amide through air .042. We thus see that the velocity
+of diffusion of ions through air is much less than that of the simple
+gas, but that it is quite comparable with that of the vapours of some
+complex organic compounds.</p>
+
+<p>The preceding tables show that the negative ions diffuse more
+rapidly than the positive, especially in dry gases. The superior
+mobility of the negative ions was observed first by Zeleny (<i>Phil. Mag.</i>,
+1898 [5], 46, p. 120), who showed that the velocity of the negative
+ions under an electric force is greater than that of the positive. It
+will be noticed that the difference between the mobility of the
+negative and the positive ions is much more pronounced in dry
+gases than in moist. The difference in the rates of diffusion of the
+positive and negative ions is the reason why ionized gas, in which,
+to begin with, the positive and negative charges were of equal
+amounts, sometimes becomes electrified even although the gas is not
+acted upon by electric forces. Thus, for example, if such gas be
+blown through narrow tubes, it will be positively electrified when
+it comes out, for since the negative ions diffuse more rapidly than
+the positive, the gas in its passage through the tubes will lose by
+diffusion more negative than positive ions and hence will emerge
+positively electrified. Zeleny snowed that this effect does not occur
+when, as in carbonic acid gas, the positive and negative ions diffuse
+at the same rates. Townsend (loc. cit.) showed that the coefficient
+of diffusion of the ions is the same whether the ionization is produced
+by Röntgen rays, radioactive substances, ultra-violet light, or
+electric sparks. The ions produced by chemical reactions and in
+flames are much less mobile; thus, for example, Bloch (<i>Ann. chim.
+phys.</i>, 1905 [8], 4, p. 25) found that for the ions produced by drawing
+air over phosphorus the value of &alpha;/e was between 1 and 6 instead
+of over 3000, the value when the air was ionized by Röntgen rays.</p>
+</div>
+
+<p><i>Velocity of Ions in an Electric Field.</i>&mdash;The velocity of ions in an
+electric field, which is of fundamental importance in conduction,
+is very closely related to the coefficient of diffusion. Measurements
+of this velocity for ions produced by Röntgen rays have
+been made by Rutherford (<i>Phil. Mag.</i> [5], 44, p. 422), Zeleny
+(<i>Phil. Mag.</i> [5], 46, p. 120), Langevin (<i>Ann. Chim. Phys.</i>, 1903,
+28, p. 289), Phillips (<i>Proc. Roy. Soc.</i> 78, A, p. 167), and Wellisch
+(<i>Phil. Trans.</i>, 1909, 209, p. 249). The ions produced by radioactive
+substance have been investigated by Rutherford (<i>Phil.
+Mag.</i> [5], 47, p. 109) and by Franck and Pohl (<i>Verh. deutsch. phys.
+Gesell.</i>, 1907, 9, p. 69), and the negative ions produced when ultra-violet
+light falls on a metal plate by Rutherford (<i>Proc. Camb. Phil.
+Soc.</i> 9, p. 401). H. A. Wilson (<i>Phil. Trans.</i> 192, p. 4O9), Marx (<i>Ann.
+de Phys.</i> 11, p. 765), Moreau (<i>Journ. de Phys.</i> 4, 11, p. 558; <i>Ann.
+Chim. Phys.</i> 7, 30, p. 5) and Gold (<i>Proc. Roy. Soc.</i> 79, p. 43) have
+investigated the velocities of ions produced by putting various
+salts into flames; McClelland (<i>Phil. Mag.</i> 46, p. 29) the velocity
+of the ions in gases sucked from the neighbourhood of flames and
+arcs; Townsend (<i>Proc. Camb. Phil. Soc.</i> 9, p. 345) and Bloch
+(<i>loc. cit.</i>) the velocity of ions produced by chemical reaction; and
+Chattock (<i>Phil. Mag.</i> [5], 48, p. 401) the velocity of the ions produced
+when electricity escapes from a sharp needle point into a gas.</p>
+
+<p>Several methods have been employed to determine these
+velocities. The one most frequently employed is to find the
+electromotive intensity required to force an ion against the
+stream of gas moving with a known velocity parallel to the lines
+of electric force. Thus, of two perforated plane electrodes
+vertically over each other, suppose the lower to be positively,
+the upper negatively electrified, and suppose that the gas is
+streaming vertically downwards with the velocity V; then unless
+the upward velocity of the positive ion is greater than V, no
+positive electricity will reach the upper plate. If we increase
+the strength of the field between the plates, and hence the upward
+velocity of the positive ion, until the positive ions just begin to
+reach the upper plate, we know that with this strength of field the
+velocity of the positive ion is equal to V. By this method, which
+has been used by Rutherford, Zeleny and H. A. Wilson, the
+velocity of ions in fields of various strengths has been determined.</p>
+
+<div class="condensed">
+<p>The arrangement used by Zeleny is represented in fig. 8. P and
+Q are square brass plates. They are bored through their centres,
+and to the openings the tubes R and S are attached, the space
+between the plates being covered in so as to form a closed box.
+K is a piece of wire gauze completely covering the opening in Q;
+T is an insulated piece of wire gauze nearly but not quite filling the
+opening in the plate P, and connected with one pair of quadrants of an
+electrometer E. A plug of glass wool G filters out the dust from a
+stream of gas which enters the vessel by the tube D and leaves it by
+F; this plug also makes the velocity of the flow of the gas uniform
+across the section of the tube. The Röntgen rays to ionize the gas
+were produced by a bulb at O, the bulb and coil being in a lead-covered
+box, with an aluminium window through which the rays
+passed. Q is connected with one pole of a battery of cells, P and the
+other pole of the battery are put to earth. The changes in the
+potential of T are due to ions giving up their charges to it. With a
+given velocity of air-blast the potential of T was found not to change
+unless the difference of potential between P and Q exceeded a critical
+value. The field corresponding to this critical value thus made the
+ions move with the known velocity of the blast.</p>
+
+<table class="nobctr" summary="Illustration">
+<tr><td class="figcenter"><img style="width:447px; height:250px" src="images/img868a.jpg" alt="" /></td></tr>
+<tr><td class="caption"><span class="sc">Fig. 8.</span></td></tr></table>
+
+<table class="nobctr" style="float: right; width: 356px;" summary="Illustration">
+<tr><td class="figright1"><img style="width:306px; height:68px" src="images/img868b.jpg" alt="" /></td></tr>
+<tr><td class="caption"><span class="sc">Fig. 9.</span></td></tr></table>
+
+<p>Another method which has been employed by Rutherford and
+McClelland is based on the action of an electric field in destroying
+the conductivity of gas streaming through it. Suppose that BAB,
+DCD (fig. 9) are a system of parallel plates boxed in so that a stream
+of gas, after flowing between BB, passes between DD without any
+loss of gas in the interval. Suppose the plates DD are insulated, and
+connected with one pair of quadrants of an electrometer, by charging
+up C to a sufficiently high potential we can drive all the positive
+ions which enter the system DCD against the plates D; this will
+cause a deflexion of the electrometer, which in one second will be
+proportional to the number of positive ions which have entered the
+system in that time. If we charge A up to a high potential, B being
+put to earth, we shall find
+that the deflexion of the electrometer
+connected with
+DD is less than it was when
+A and B were at the same
+potential, because some of
+the positive ions in their
+passage through BAB are driven against the plates B. If u is the
+velocity along the lines of force in the uniform electric field between
+A and B, and t the time it takes for the gas to pass through BAB, then
+all the positive ions within a distance ut of the plates B will be driven
+up against these plates, and thus if the positive ions are equally distributed
+through the gas, the number of positive ions which emerge
+from the system when the electric field is on will bear to the number
+which emerge when the field is off the ratio of 1 &minus; ut/l to unity, where
+l is the distance between A and B. This ratio is equal to the ratio of
+the deflexions in one second of the electrometer attached to D, hence
+the observations of this instrument give 1 &minus; ut/l. If we know the
+velocity of the gas and the length of the plates A and B, we can
+determine t, and since l can be easily measured, we can find u, the
+velocity of the positive ion in a field of given strength. By charging
+A and C negatively instead of positively we can arrive at the
+velocity of the negative ion. In practice it is more convenient to use
+cylindrical tubes with coaxial wires instead of the systems of parallel
+plates, though in this case the calculation of the velocity of the ions
+from the observations is a little more complicated, inasmuch as the
+electric field is not uniform between the tubes.</p>
+
+<table class="nobctr" style="float: right; width: 260px;" summary="Illustration">
+<tr><td class="figright1"><img style="width:212px; height:152px" src="images/img868c.jpg" alt="" /></td></tr>
+<tr><td class="caption"><span class="sc">Fig. 10.</span></td></tr></table>
+
+<p>A method which gives very accurate results, though it is only
+applicable in certain cases, is the one used by Rutherford to measure
+the velocity of the negative ions produced close to a metal plate by
+the incidence on the plate of ultra-violet light. The principle of the
+method is as follows:&mdash;AB (fig. 10) is an insulated horizontal plate
+of well-polished zinc, which can be
+moved vertically up and down by
+means of a screw; it is connected with
+one pair of quadrants of an electrometer,
+the other pair of quadrants being
+put to earth. CD is a base-plate with a
+hole EF in it; this hole is covered with
+fine wire gauze, through which ultra-violet
+light passes and falls on the plate
+AB. The plate CD is connected with
+an alternating current dynamo, which
+produces a simply-periodic potential
+difference between AB and CD, the other pole being put to earth.
+Suppose that at any instant the plate CD is at a higher potential than
+AB, then the negative ions from AB will move towards CD, and will
+continue to do so as long as the potential of CD is higher than that of
+AB. If, however, the potential difference changes sign before the negative
+ions reach CD, these ions will go back to AB. Thus AB will not
+<span class="pagenum"><a name="page869" id="page869"></a>869</span>
+lose any negative charge unless the distance between the plates AB and
+CD is less than the distance traversed by the negative ion during the
+time the potential of CD is higher than that of AB. By altering the
+distance between the plates until CD just begins to lose a negative
+charge, we find the velocity of the negative ion under unit electromotive
+intensity. For suppose the difference of potential between
+AB and CD is equal to a sin pt, then if d is the distance between the
+plates, the electric intensity is equal to a sin pt/d; if we suppose the
+velocity of the ion is proportional to the electric intensity, and if u
+is the velocity for unit electric intensity, the velocity of the negative
+ion will be ua sin pt/d. Hence if x represent the distance of the ion
+from AB</p>
+
+<table class="math0" summary="math">
+<tr><td>dx</td> <td rowspan="2">=</td> <td>ua</td> <td rowspan="2">sin pt</td></tr>
+<tr> <td class="denom">dT</td> <td class="denom">d</td></tr></table>
+
+<table class="math0" summary="math">
+<tr><td rowspan="2">x =</td> <td>ua</td> <td rowspan="2">(1 &minus; cos pt), if x = 0 when t = 0.</td></tr>
+<tr> <td class="denom">pd</td></tr></table>
+
+<p class="noind">Thus the greatest distance the ion can get from the plate is equal
+to 2au/pd, and if the distance between the plates is gradually reduced
+to this value, the plate AB will begin to lose a negative charge; hence
+when this happens</p>
+
+<p class="center1">d = 2au/pd, &emsp;or u = pd<span class="sp">2</span>/2a,</p>
+
+<p class="noind">an equation by means of which we can find u.</p>
+
+<p>In this form the method is not applicable when ions of both signs
+are present. Franck and Pohl (<i>Verh. deutsch. physik. Gesell.</i> 1907,
+9, p. 69) have by a slight modification removed this restriction.
+The modification consists in confining the ionization to a layer of gas
+below the gauze EF. If the velocity of the positive ions is to be
+determined, these ions are forced through the gauze by applying
+to the ionized gas a small constant electric force acting upwards;
+if negative ions are required, the constant force is reversed. After
+passing through the gauze the ions are acted upon by alternating
+forces as in Rutherford&rsquo;s method.</p>
+
+<p>Langevin (<i>Ann. chim. phys.</i>, 1903, 28, p. 289) devised a method
+of measuring the velocity of the ions which has been extensively
+used; it has the advantage of not requiring the rate of ionization
+to remain uniform. The general idea is as follows. Suppose that
+we expose the gas between two parallel plates A, B to Röntgen rays
+or some other ionizing agent, then stop the rays and apply a uniform
+electric field to the region between the plates. If the force on the
+positive ion is from A to B, the plate B will receive a positive charge
+of electricity. After the electric force has acted for a time T reverse
+it. B will now begin to receive negative electricity and will go on
+doing so until the supply of negative ions is exhausted. Let us
+consider how the quantity of positive electricity received by B will
+vary with T. To fix our ideas, suppose the positive ions move more
+slowly than the negative; let T<span class="su">2</span> and T<span class="su">1</span> be respectively the times
+taken by the positive and negative ions to move under the electric
+field through a distance equal to AB, the distance between the
+planes. Then if T is greater than T<span class="su">2</span> all the ions will have been
+driven from between the plates before the field is reversed, and therefore
+the positive charge received by B will not depend upon T.
+Next let T be less than T<span class="su">2</span> but greater than T<span class="su">1</span>; then at the time
+when the field is reversed all the negative ions will have been driven
+from between the plates, so that the positive charge received by B
+will not be neutralized by the arrival of fresh ions coming to it after
+the reversal of the field. The number of positive ions driven against
+the plate B will be proportional to T. Thus if we measure the value
+of the positive charge on B for a series of values of T, each value being
+less than the preceding, we shall find that until T reaches a certain
+value the charge remains constant, but as soon as we reduce the
+time below this value the charge diminishes. The value of T when
+the diminution in the field begins is T<span class="su">2</span>, the time taken for a positive
+ion to cross from A to B under the electric field; thus from T<span class="su">2</span> we
+can calculate the velocity of the positive ion in this field. If we still
+further diminish T, we shall find that we reach a value when the
+diminution of the positive charge on B with the time suddenly
+becomes much more rapid; this change occurs when T falls below T<span class="su">1</span>
+the time taken for the negative ions to go from one plate to the other,
+for now when the field is reversed there are still some negative ions
+left between the plates, and these will be driven against B and rob it
+of some of the positive charge it had acquired before the field was
+reversed. By observing the time when the increase in the rate of
+diminution of the positive charge with the time suddenly sets in
+we can determine T<span class="su">1</span>, and hence the velocity of the negative ions.</p>
+
+<p>The velocity of the ions produced by the discharge of electricity
+from a fine point was determined by Chattock by an entirely different
+method. In this case the electric field is so strong and the velocity
+of the ion so great that the preceding methods are not applicable.
+Suppose P represents a vertical needle discharging electricity into
+air, consider the force acting on the ions included between two
+horizontal planes A, B. If P is the density of the electrification,
+and Z the vertical component of the electric intensity, F the resultant
+force on the ions between A and B is vertical and equal to</p>
+
+<p class="center1"><span class="f150">&int;&int;&int;</span> Z&rho; dxdydz.</p>
+
+<p class="noind">Let us suppose that the velocity of the ion is proportional to the
+electric intensity, so that if w is the vertical velocity of the ions,
+which are supposed all to be of one sign, w = RZ.</p>
+
+<p>Substituting this value of Z, the vertical force on the ions between
+A and B is equal to</p>
+
+<table class="math0" summary="math">
+<tr><td>1</td> <td rowspan="2"><span class="f150">&int;&int;&int;</span> w&rho; dxdydz.</td></tr>
+<tr> <td class="denom">R</td></tr></table>
+
+<p class="noind">But &int;&int; w&rho;dxdy = &iota;, where &iota; is the current streaming from the point.
+This current, which can be easily measured by putting a galvanometer
+in series with the discharging point, is independent of z,
+the vertical distance of a plane between A and B below the charging
+point. Hence we have</p>
+
+<table class="math0" summary="math">
+<tr> <td rowspan="2">F =</td> <td>&iota;</td> <td rowspan="2"><span class="f150">&int;</span>dz &emsp;=</td>
+ <td>&iota;</td> <td rowspan="2"> · z.</td></tr>
+<tr> <td class="denom">R</td> <td class="denom">R</td></tr></table>
+
+<p class="noind">This force must be counterbalanced by the difference of gaseous
+pressures over the planes A and B; hence if p<span class="su">B</span> and p<span class="su">A</span> denote
+respectively the pressures over B and A, we have</p>
+
+<table class="math0" summary="math">
+<tr> <td rowspan="2">p<span class="su">B</span> &minus; p<span class="su">A</span> =</td> <td>&iota;</td>
+ <td rowspan="2">z.</td></tr>
+<tr> <td class="denom">R</td></tr></table>
+
+<p class="noind">Hence by the measurement of these pressures we can determine
+R, and hence the velocity with which an ion moves under a given
+electric intensity.</p>
+
+<p>There are other methods of determining the velocities of the
+ions, but as these depend on the theory of the conduction of electricity
+through a gas containing charged ions, we shall consider them in our
+discussion of that theory.</p>
+
+<p>By the use of these methods it has been shown that the velocities
+of the ions in a given gas are the same whether the ionization is
+produced by Röntgen rays, radioactive substances, ultra-violet
+light, or by the discharge of electricity from points. When the
+ionization is produced by chemical action the ions are very much
+less mobile, moving in the same electric field with a velocity less
+than one-thousandth part of the velocity of the first kind of ions.
+On the other hand, as we shall see later, the velocity of the negative
+ions in flames is enormously greater than that of even the first kind
+of ion under similar electric fields and at the same pressure. But
+when these negative ions get into the cold part of the flame, they
+move sluggishly with velocities of the order of those possessed by
+the second kind. The results of the various determinations of the
+velocities of the ions are given in the following table. The velocities
+are in centimetres per second under an electric force of one volt per
+centimetre, the pressure of the gas being 1 atmosphere. V+
+denotes the velocity of the positive ion, V- that of the negative.
+V is the mean velocity of the positive and negative ions.</p>
+
+<p class="center1"><i>Velocities of Ions.&mdash;Ions produced by Röntgen Rays.</i></p>
+
+<table class="ws" summary="Contents">
+<tr><td class="tcc allb">Gas.</td> <td class="tcc allb">V+.</td> <td class="tcc allb">V-.</td> <td class="tcc allb">V.</td> <td class="tcc allb">Observer.</td></tr>
+
+<tr><td class="tcl lb rb">Air</td> <td class="tcc rb">..</td> <td class="tcc rb">..</td> <td class="tcc rb">1.6</td> <td class="tcl rb">Rutherford</td></tr>
+<tr><td class="tcl lb rb">Air (dry)</td> <td class="tcc rb">1.36</td> <td class="tcc rb">1.87</td> <td class="tcc rb">..</td> <td class="tcl rb">Zeleny</td></tr>
+<tr><td class="tcl lb rb">&emsp;&emsp;&rdquo;</td> <td class="tcc rb">1.60</td> <td class="tcc rb">1.70</td> <td class="tcc rb">..</td> <td class="tcl rb">Langevin</td></tr>
+<tr><td class="tcl lb rb">&emsp;&emsp;&rdquo;</td> <td class="tcc rb">1.39</td> <td class="tcc rb">1.78</td> <td class="tcc rb">..</td> <td class="tcl rb">Phillips</td></tr>
+<tr><td class="tcl lb rb">&emsp;&emsp;&rdquo;</td> <td class="tcc rb">1.54</td> <td class="tcc rb">1.78</td> <td class="tcc rb">..</td> <td class="tcl rb">Wellisch</td></tr>
+<tr><td class="tcl lb rb">Air (moist)</td> <td class="tcc rb">1.37</td> <td class="tcc rb">1.81</td> <td class="tcc rb">..</td> <td class="tcl rb">Zeleny</td></tr>
+<tr><td class="tcl lb rb">Oxygen (dry)</td> <td class="tcc rb">1.36</td> <td class="tcc rb">1.80</td> <td class="tcc rb">..</td> <td class="tcl rb">&emsp;&rdquo;</td></tr>
+<tr><td class="tcl lb rb">Oxygen (moist)</td> <td class="tcc rb">1.29</td> <td class="tcc rb">1.52</td> <td class="tcc rb">..</td> <td class="tcl rb">&emsp;&rdquo;</td></tr>
+<tr><td class="tcl lb rb">Carbonic acid (dry)</td> <td class="tcc rb">0.76</td> <td class="tcc rb">0.81</td> <td class="tcc rb">..</td> <td class="tcl rb">&emsp;&rdquo;</td></tr>
+<tr><td class="tcl lb rb">&emsp;&emsp;&rdquo;&emsp;&emsp;&rdquo;</td> <td class="tcc rb">0.86</td> <td class="tcc rb">0.90</td> <td class="tcc rb">..</td> <td class="tcl rb">Langevin</td></tr>
+<tr><td class="tcl lb rb">&emsp;&emsp;&rdquo;&emsp;&emsp;&rdquo;</td> <td class="tcc rb">0.81</td> <td class="tcc rb">0.85</td> <td class="tcc rb">..</td> <td class="tcl rb">Wellisch</td></tr>
+<tr><td class="tcl lb rb">Carbonic acid (moist)</td> <td class="tcc rb">0.82</td> <td class="tcc rb">0.75</td> <td class="tcc rb">..</td> <td class="tcl rb">Zeleny</td></tr>
+<tr><td class="tcl lb rb">Hydrogen (dry)</td> <td class="tcc rb">6.70</td> <td class="tcc rb">7.95</td> <td class="tcc rb">..</td> <td class="tcl rb">&emsp;&rdquo;</td></tr>
+<tr><td class="tcl lb rb">Nitrogen</td> <td class="tcc rb">..</td> <td class="tcc rb">..</td> <td class="tcc rb">1.6</td> <td class="tcl rb">Rutherford</td></tr>
+<tr><td class="tcl lb rb">Sulphur dioxide</td> <td class="tcc rb">0.44</td> <td class="tcc rb">0.41</td> <td class="tcc rb">..</td> <td class="tcl rb">Wellisch</td></tr>
+<tr><td class="tcl lb rb">Hydrochloric acid</td> <td class="tcc rb">..</td> <td class="tcc rb">..</td> <td class="tcc rb">1.27</td> <td class="tcl rb">Rutherford</td></tr>
+<tr><td class="tcl lb rb">Chlorine</td> <td class="tcc rb">..</td> <td class="tcc rb">..</td> <td class="tcc rb">1.0</td> <td class="tcl rb">&emsp;&rdquo;</td></tr>
+<tr><td class="tcl lb rb">Helium (dry)</td> <td class="tcc rb">5.09</td> <td class="tcc rb">6.31</td> <td class="tcc rb">..</td> <td class="tcl rb">Franck and Pohl</td></tr>
+<tr><td class="tcl lb rb">Carbon monoxide</td> <td class="tcc rb">1.10</td> <td class="tcc rb">1.14</td> <td class="tcc rb">..</td> <td class="tcl rb">Wellisch</td></tr>
+<tr><td class="tcl lb rb">Nitrous oxide</td> <td class="tcc rb">0.82</td> <td class="tcc rb">0.90</td> <td class="tcc rb">..</td> <td class="tcl rb">&emsp;&rdquo;</td></tr>
+<tr><td class="tcl lb rb">Ammonia</td> <td class="tcc rb">0.74</td> <td class="tcc rb">0.80</td> <td class="tcc rb">..</td> <td class="tcl rb">&emsp;&rdquo;</td></tr>
+<tr><td class="tcl lb rb">Aldehyde</td> <td class="tcc rb">0.31</td> <td class="tcc rb">0.30</td> <td class="tcc rb">..</td> <td class="tcl rb">&emsp;&rdquo;</td></tr>
+<tr><td class="tcl lb rb">Ethyl alcohol</td> <td class="tcc rb">0.34</td> <td class="tcc rb">0.27</td> <td class="tcc rb">..</td> <td class="tcl rb">&emsp;&rdquo;</td></tr>
+<tr><td class="tcl lb rb">Acetone</td> <td class="tcc rb">0.31</td> <td class="tcc rb">0.29</td> <td class="tcc rb">..</td> <td class="tcl rb">&emsp;&rdquo;</td></tr>
+<tr><td class="tcl lb rb">Ethyl chloride</td> <td class="tcc rb">0.33</td> <td class="tcc rb">0.31</td> <td class="tcc rb">..</td> <td class="tcl rb">&emsp;&rdquo;</td></tr>
+<tr><td class="tcl lb rb">Pentane</td> <td class="tcc rb">0.36</td> <td class="tcc rb">0.35</td> <td class="tcc rb">..</td> <td class="tcl rb">&emsp;&rdquo;</td></tr>
+<tr><td class="tcl lb rb">Methyl acetate</td> <td class="tcc rb">0.33</td> <td class="tcc rb">0.36</td> <td class="tcc rb">..</td> <td class="tcl rb">&emsp;&rdquo;</td></tr>
+<tr><td class="tcl lb rb">Ethyl formate</td> <td class="tcc rb">0.30</td> <td class="tcc rb">0.31</td> <td class="tcc rb">..</td> <td class="tcl rb">&emsp;&rdquo;</td></tr>
+<tr><td class="tcl lb rb">Ethyl ether</td> <td class="tcc rb">0.29</td> <td class="tcc rb">0.31</td> <td class="tcc rb">..</td> <td class="tcl rb">&emsp;&rdquo;</td></tr>
+<tr><td class="tcl lb rb">Ethyl acetate</td> <td class="tcc rb">0.31</td> <td class="tcc rb">0.28</td> <td class="tcc rb">..</td> <td class="tcl rb">&emsp;&rdquo;</td></tr>
+<tr><td class="tcl lb rb">Methyl bromide</td> <td class="tcc rb">0.29</td> <td class="tcc rb">0.28</td> <td class="tcc rb">..</td> <td class="tcl rb">&emsp;&rdquo;</td></tr>
+<tr><td class="tcl lb rb">Methyl iodide</td> <td class="tcc rb">0.21</td> <td class="tcc rb">0.22</td> <td class="tcc rb">..</td> <td class="tcl rb">&emsp;&rdquo;</td></tr>
+<tr><td class="tcl lb rb">Carbon tetrachloride</td> <td class="tcc rb">0.30</td> <td class="tcc rb">0.31</td> <td class="tcc rb">..</td> <td class="tcl rb">&emsp;&rdquo;</td></tr>
+<tr><td class="tcl lb rb bb">Ethyl iodide</td> <td class="tcc rb bb">0.17</td> <td class="tcc rb bb">0.16</td> <td class="tcc rb bb">..</td> <td class="tcl rb bb">&emsp;&rdquo;</td></tr>
+</table>
+
+<table class="nobctr" style="width: 70%;" summary="Contents">
+<tr><td class="pt2 tcc" colspan="3"><i>Ions produced by Ultra-Violet Light.</i></td></tr>
+
+<tr><td class="tcl">Air</td> <td class="tcc">1.4&ensp;</td> <td class="tcl">Rutherford</td></tr>
+<tr><td class="tcl">Hydrogen</td> <td class="tcc">3.9&ensp;</td> <td class="tcl">Rutherford</td></tr>
+<tr><td class="tcl">Carbonic acid</td> <td class="tcc">0.78&ensp;</td> <td class="tcl">Rutherford<span class="pagenum"><a name="page870" id="page870"></a>870</span></td></tr>
+
+<tr><td class="pt2 tcc" colspan="3"><i>Ions in Gases sucked from Flames.</i></td></tr>
+
+<tr><td class="tcl" colspan="2">Velocities varying from .04 to .23</td> <td class="tcl">McClelland</td></tr>
+
+<tr><td class="pt2 tcc" colspan="3"><i>Ions in Flames containing Salts.</i></td></tr>
+
+<tr><td class="tcl">Negative ions</td> <td class="tcc">12.9 cm./sec.</td> <td class="tcl">Gold</td></tr>
+<tr><td class="tcl">+ions for salts of Li, Na, K, Rb, Cs</td> <td class="tcc">&ensp;62</td> <td class="tcl">H. A. Wilson</td></tr>
+<tr><td class="tcc">&rdquo;</td> <td class="tcc">200</td> <td class="tcl">Marx</td></tr>
+<tr><td class="tcc">&rdquo;</td> <td class="tcc">&ensp;80</td> <td class="tcl">Moreau</td></tr>
+
+<tr><td class="pt2 tcc" colspan="3"><i>Ions liberated by Chemical Action.</i></td></tr>
+
+<tr><td class="tcl" colspan="2">Velocities of the order of 0.0005 cm./sec.</td> <td class="tcl">Bloch</td></tr>
+</table>
+
+<p class="center pt2"><i>Ions from Point Discharge.</i></p>
+
+<table class="ws" summary="Contents">
+
+<tr><td class="tcl lb rb tb">Hydrogen</td> <td class="tcl rb tb">5.4</td> <td class="tcl rb tb">7.43</td> <td class="tcc rb tb">6.41</td> <td class="tcc rb tb">Chattock</td></tr>
+<tr><td class="tcl lb rb">Carbonic acid</td> <td class="tcl rb">0.83</td> <td class="tcl rb">0.925</td> <td class="tcc rb">0.88</td> <td class="tcc rb">Chattock</td></tr>
+<tr><td class="tcl lb rb">Air</td> <td class="tcl rb">1.32</td> <td class="tcl rb">1.80</td> <td class="tcc rb">1.55</td> <td class="tcc rb">Chattock</td></tr>
+<tr><td class="tcl lb rb bb">Oxygen</td> <td class="tcl rb bb">1.30</td> <td class="tcl rb bb">1.85</td> <td class="tcc rb bb">1.57</td> <td class="tcc rb bb">Chattock</td></tr>
+</table>
+
+<p>It will be seen from this table that the greater mobility of the
+negative ions is very much more marked in the case of the lighter
+and simpler gases than in that of the heavier and more complicated
+ones; with the vapours of organic substances there seems but little
+difference between the mobilities of the positive and negative ions,
+indeed in one or two cases the positive one seems slightly but very
+slightly the more mobile of the two. In the case of the simple gases
+the difference is much greater when the gases are dry than when they
+are moist. It has been shown by direct experiment that the velocities
+are directly proportional to the electric force.</p>
+
+<p><i>Variation of Velocities with Pressure.</i>&mdash;Until the pressure gets low
+the velocities of the ions, negative as well as positive, vary inversely
+as the pressure. Langevin (loc. cit.) was the first to show that at very
+low pressures the velocity of the negative ions increases more
+rapidly as the pressure is diminished than this law indicates. If the
+nature of the ion did not change with the pressure, the kinetic theory
+of gases indicates that the velocity would vary inversely as the
+pressure, so that Langevin&rsquo;s results indicate a change in the nature
+of the negative ion when the pressure is diminished below a certain
+value. Langevin&rsquo;s results are given in the following table, where p
+represents the pressure measured in centimetres of mercury, V+ and
+V- the velocities of the positive and negative ions in air under
+unit electrostatic force, <i>i.e.</i> 300 volts per centimetre:&mdash;</p>
+
+<table class="ws" summary="Contents">
+<tr><td class="tcc allb" colspan="3">Negative Ions.</td> <td class="tcc allb" colspan="3">Positive Ions.</td></tr>
+<tr><td class="tcc allb">p.</td> <td class="tcc allb">V-.</td> <td class="tcc allb">pV-/76.</td>
+ <td class="tcc allb">p.</td> <td class="tcc allb">V+.</td> <td class="tcc allb">pV+/76.</td></tr>
+
+<tr><td class="tcr lb rb">7.5</td> <td class="tcr rb">6560</td> <td class="tcc rb">647</td> <td class="tcr rb">7.5</td> <td class="tcr rb">4430</td> <td class="tcc rb">437</td></tr>
+<tr><td class="tcr lb rb">20.0</td> <td class="tcr rb">2204</td> <td class="tcc rb">580</td> <td class="tcr rb">20.0</td> <td class="tcr rb">1634</td> <td class="tcc rb">430</td></tr>
+<tr><td class="tcr lb rb">41.5</td> <td class="tcr rb">994</td> <td class="tcc rb">530</td> <td class="tcr rb">41.5</td> <td class="tcr rb">782</td> <td class="tcc rb">427</td></tr>
+<tr><td class="tcr lb rb">76.0</td> <td class="tcr rb">510</td> <td class="tcc rb">510</td> <td class="tcr rb">76.0</td> <td class="tcr rb">480</td> <td class="tcc rb">420</td></tr>
+<tr><td class="tcr lb rb bb">142.0</td> <td class="tcr rb bb">270</td> <td class="tcc rb bb">505</td> <td class="tcr rb bb">142.0</td> <td class="tcr rb bb">225</td> <td class="tcc rb bb">425</td></tr>
+</table>
+
+<p class="noind">The increase in the case of pV- indicates that the structure of the
+negative ion gets simpler as the pressure is reduced. Wallisch in
+some experiments made at the Cavendish Laboratory found that the
+diminution in the value of pV- at low pressures is much more marked
+in some gases than in others, and in some gases he failed to detect
+it; but it must be remembered that it is difficult to get measurements
+at pressures of only a few millimetres, as the amount of ionization
+is so exceedingly small at such pressures that the quantities to be
+observed are hardly large enough to admit of accurate measurements
+by the methods available at higher pressures.</p>
+
+<p><i>Effect of Temperature on the Velocity of the Ions.</i>&mdash;Phillips (<i>Proc.
+Roy. Soc.</i>, 1906, 78, p. 167) investigated, using Langevin&rsquo;s method,
+the velocities of the + and &minus; ions through air at atmospheric
+pressure at temperatures ranging from that of boiling liquid air to
+411° C.; R<span class="su">1</span> and R<span class="su">2</span> are the velocities of the + and &minus; ions respectively
+when the force is a volt per centimetre.</p>
+
+<table class="ws" summary="Contents">
+
+<tr><td class="tcc allb">R<span class="su">1</span>.</td> <td class="tcc allb">R<span class="su">2</span>.</td> <td class="tcc allb">Temperature Absolute.</td></tr>
+
+<tr><td class="tcl lb rb">2.00</td> <td class="tcl rb">2.495</td> <td class="tcc rb">411°</td></tr>
+<tr><td class="tcl lb rb">1.95</td> <td class="tcl rb">2.40</td> <td class="tcc rb">399°</td></tr>
+<tr><td class="tcl lb rb">1.85</td> <td class="tcl rb">2.30</td> <td class="tcc rb">383°</td></tr>
+<tr><td class="tcl lb rb">1.81</td> <td class="tcl rb">2.21</td> <td class="tcc rb">373°</td></tr>
+<tr><td class="tcl lb rb">1.67</td> <td class="tcl rb">2.125</td> <td class="tcc rb">348°</td></tr>
+<tr><td class="tcl lb rb">1.60</td> <td class="tcl rb">2.00</td> <td class="tcc rb">333°</td></tr>
+<tr><td class="tcl lb rb">1.39</td> <td class="tcl rb">1.785</td> <td class="tcc rb">285°</td></tr>
+<tr><td class="tcl lb rb">0.945</td> <td class="tcl rb">1.23</td> <td class="tcc rb">209°</td></tr>
+<tr><td class="tcl lb rb bb">0.235</td> <td class="tcl rb bb">0.235</td> <td class="tcc rb bb">&ensp;94°</td></tr>
+</table>
+
+<p>We see that except in the case of the lowest temperature, that of
+liquid air, where there is a great drop in the velocity, the velocities
+of the ions are proportional to the absolute temperature. On the
+hypothesis of an ion of constant size we should, from the kinetic
+theory of gases, expect the velocity to be proportional to the square
+root of the absolute temperature, if the charge on the ion did not
+affect the number of collisions between the ion and the molecules of
+the gas through which it is moving. If the collisions were brought
+about by the electrical attraction between the ions and the molecules,
+the velocity would be proportional to the absolute temperature.
+H. A. Wilson (<i>Phil. Trans.</i> 192, p. 499), in his experiments on the
+conduction of flames and hot gases into which salts had been put,
+found that the velocity of the positive ions in flames at a temperature
+of 2000° C. containing the salts of the alkali metals was 62 cm./sec.
+under an electric force of one volt per centimetre, while the velocity
+of the positive ions in a stream of hot air at 1000° C. containing the
+same salts was only 7 cm./sec. under the same force. The great effect
+of temperature is also shown in some experiments of McClelland
+(<i>Phil. Mag.</i> [5], 46, p. 29) on the velocities of the ions in gases drawn
+from Bunsen flames and arcs; he found that these depended upon
+the distance the gas had travelled from the flame. Thus, the velocity
+of the ions at a distance of 5.5 cm. from the Bunsen flame when the
+temperature was 230° C. was .23 cm./sec. for a volt per centimetre;
+at a distance of 10 cm. from the flame when the temperature was
+160° C. the velocity was .21 cm./sec; while at a distance of 14.5
+cm. from the flame when the temperature was 105° C. the velocity
+was only .04 cm./sec. If the temperature of the gas at this distance
+from the flame was raised by external means, the velocity of the ions
+increased.</p>
+
+<p>We can derive some information as to the constitution of the
+ions by calculating the velocity with which a molecule of the gas
+would move in the electric field if it carried the same charge as the
+ion. From the theory of the diffusion of gases, as developed by
+Maxwell, we know that if the particles of a gas A are surrounded
+by a gas B, then, if the partial pressure of A is small, the velocity u
+with which its particles will move when acted upon by a force Xe
+is given by the equation</p>
+
+<table class="math0" summary="math">
+<tr> <td rowspan="2">u =</td> <td>Xe</td> <td rowspan="2">D,</td></tr>
+<tr> <td class="denom">(p<span class="su">1</span>/N<span class="su">1</span>)</td></tr></table>
+
+<p class="noind">where D represents the coefficient of inter-diffusion of A into B,
+and N<span class="su">1</span> the number of particles of A per cubic centimetre when the
+pressure due to A is p<span class="su">1</span>. Let us calculate by this equation the
+velocity with which a molecule of hydrogen would move through
+hydrogen if it carried the charge carried by an ion, which we shall
+prove shortly to be equal to the charge carried by an atom of hydrogen
+in the electrolysis of solutions. Since p<span class="su">1</span>/N<span class="su">1</span> is independent of the
+pressure, it is equal to &Pi;/N, where &Pi; is the atmospheric pressure and
+N the number of molecules in a cubic centimetre of gas at atmospheric
+pressure. Now N<span class="su">e</span> = 1.22&times;10<span class="sp">10</span>, if e is measured in electrostatic
+units; &Pi; = 10<span class="sp">6</span> and D in this case is the coefficient of diffusion
+of hydrogen into itself, and is equal to 1.7. Substituting these values
+we find</p>
+
+<p class="center1">u = 1.97&times;10<span class="sp">4</span>X.</p>
+
+<p class="noind">If the potential gradient is 1 volt per centimetre, X = <span class="spp">1</span>&frasl;<span class="suu">300</span>. Substituting
+this value for X, we find u = 66 cm./sec, for the velocity of
+a hydrogen molecule. We have seen that the velocity of the ion in
+hydrogen is only about 5 cm./sec, so that the ion moves more slowly
+than it would if it were a single molecule. One way of explaining
+this is to suppose that the ion is bigger than the molecule, and is
+in fact an aggregation of molecules, the charged ion acting as a
+nucleus around which molecules collect like dust round a charged
+body. This view is supported by the effect produced by moisture in
+diminishing the velocity of the negative ion, for, as C. T. R. Wilson
+(<i>Phil. Trans.</i> 193, p. 289) has shown, moisture tends to collect
+round the ions, and condenses more easily on the negative than on
+the positive ion. In connexion with the velocities of ions in the
+gases drawn from flames, we find other instances which suggest
+that condensation takes place round the ions. An increase in the
+size of the system is not, however, the only way by which the velocity
+might fall below that calculated for the hydrogen molecule, for we
+must remember that the hydrogen molecule, whose coefficient of
+diffusion is 1.7, is not charged, while the ion is. The forces exerted
+by the ion on the other molecules of hydrogen are not the same as
+those which would be exerted by a molecule of hydrogen, and as the
+coefficient of diffusion depends on the forces between the molecules,
+the coefficient of diffusion of a charged molecule into hydrogen might
+be very different from that of an uncharged one.</p>
+
+<p>Wellisch (<i>loc. cit.</i>) has shown that the effect of the charge on the
+ion is sufficient in many cases to explain the small velocity of the ions,
+even if there were no aggregation.</p>
+
+<p><i>Mixture of Gases.</i>&mdash;The ionization of a mixture of gases raises
+some very interesting questions. If we ionize a mixture of two
+very different gases, say hydrogen and carbonic acid, and investigate
+the nature of the ions by measuring their velocities, the question
+arises, shall we find two kinds of positive and two kinds of negative
+ions moving with different velocities, as we should do if some of the
+positive ions were positively charged hydrogen molecules, while
+others were positively charged molecules of carbonic acid; or shall
+we find only one velocity for the positive ions and one for the negative?
+Many experiments have been made on the velocity of ions
+in mixtures of two gases, but as yet no evidence has been found of
+the existence of two different kinds of either positive or negative
+ions in such mixtures, although some of the methods for determining
+the velocities of the ions, especially Langevin&rsquo;s, ought to give
+evidence of this effect, if it existed. The experiments seem to show
+<span class="pagenum"><a name="page871" id="page871"></a>871</span>
+that the positive (and the same is true for the negative) ions in a
+mixture of gases are all of the same kind. This conclusion is one of
+considerable importance, as it would not be true if the ions consisted
+of single molecules of the gas from which they are produced.</p>
+
+<p><i>Recombination.</i>&mdash;Several methods enable us to deduce the coefficient
+of recombination of the ions when we know their velocities.
+Perhaps the simplest of these consists in determining the relation
+between the current passing between two parallel plates immersed
+in ionized gas and the potential difference between the plates. For
+let q be the amount of ionization, <i>i.e.</i> the number of ions produced
+per second per unit volume of the gas, A the area of one of the plates,
+and d the distance between them; then if the ionization is constant
+through the volume, the number of ions of one sign produced per
+second in the gas is qAd. Now if i is the current per unit area of
+the plate, e the charge on an ion, iA/e ions of each sign are driven
+out of the gas by the current per second. In addition to this source
+of loss of ions there is the loss due to the recombination; if n is the
+number of positive or negative ions per unit volume, then the
+number which recombine per second is &alpha;n<span class="sp">2</span> per cubic centimetre,
+and if n is constant through the volume of the gas, as will approximately
+be the case if the current through the gas is only a small
+fraction of the saturation current, the number of ions which disappear
+per second through recombination is &alpha;n<span class="sp">2</span>·Ad. Hence, since when
+the gas is in a steady state the number of ions produced must be
+equal to the number which disappear, we have</p>
+
+<p class="center1">qAd = iA/e + &alpha;n<span class="sp">2</span>·Ad,<br />
+q = i/ed + &alpha;n<span class="sp">2</span>.</p>
+
+<p class="noind">If u<span class="su">1</span> and u<span class="su">2</span> are the velocities with which the positive and negative
+ions move, nu<span class="su">1</span>e and nu<span class="su">2</span>e are respectively the quantities of positive
+electricity passing in one direction through unit area of the gas per
+second, and of negative in the opposite direction, hence</p>
+
+<p class="center1">i = nu<span class="su">1</span>e + nu<span class="su">2</span>e.</p>
+
+<p class="noind">If X is the electric force acting on the gas, k<span class="su">1</span> and k<span class="su">2</span> the velocities
+of the positive and negative ions under unit force, u<span class="su">1</span> = k<span class="su">1</span>X,
+u<span class="su">2</span> = k<span class="su">2</span>X; hence</p>
+
+<p class="center1">n = i/(k<span class="su">1</span> + k<span class="su">2</span>)Xe,</p>
+
+<p class="noind">and we have</p>
+
+<table class="math0" summary="math">
+<tr> <td rowspan="2">q =</td> <td>i</td> <td rowspan="2">+</td> <td>&alpha;i<span class="sp">2</span></td> <td rowspan="2">.</td></tr>
+<tr> <td class="denom">ed</td> <td class="denom">(k<span class="su">1</span> + k<span class="su">2</span>)<span class="sp">2</span> e<span class="sp">2</span>X<span class="sp">2</span></td></tr></table>
+
+<p class="noind">But qed is the saturation current per unit area of the plate; calling
+this I, we have</p>
+
+<table class="math0" summary="math">
+<tr> <td rowspan="2">I &minus; i =</td> <td>d&alpha;i<span class="sp">2</span></td></tr>
+<tr> <td class="denom">e(k<span class="su">1</span> + k<span class="su">2</span>)<span class="sp">2</span>X<span class="sp">2</span></td></tr></table>
+
+<p class="noind">or</p>
+
+<table class="math0" summary="math">
+<tr> <td rowspan="2">X<span class="sp">2</span> =</td> <td>i<span class="sp">2</span>·d&alpha;</td> <td rowspan="2">.</td></tr>
+<tr> <td class="denom">e(I &minus; i) (k<span class="su">1</span> + k<span class="su">2</span>)<span class="sp">2</span></td></tr></table>
+
+<p class="noind">Hence if we determine corresponding values of X and i we can
+deduce the value of &alpha;/e if we also know (k<span class="su">1</span> + k<span class="su">2</span>). The value of I
+is easily determined, as it is the current when X is very large. The
+preceding result only applies when i is small compared with I,
+as it is only in this case that the values of n and X are uniform
+throughout the volume of the gas. Another method which answers
+the same purpose is due to Langevin (<i>Ann. Chim. Phys.</i>, 1903, 28,
+p. 289); it is as follows. Let A and B be two parallel planes immersed
+in a gas, and let a slab of the gas bounded by the planes a, b parallel
+to A and B be ionized by an instantaneous flash of Röntgen rays.
+If A and B are at different electric potentials, then all the positive
+ions produced by the rays will be attracted by the negative plate
+and all the negative ions by the positive, if the electric field were
+exceedingly large they would reach these plates before they had time
+to recombine, so that each plate would receive N<span class="su">0</span> ions if the flash of
+Röntgen rays produced N<span class="su">0</span> positive and N<span class="su">0</span> negative ions. With
+weaker fields the number of ions received by the plates will be less
+as some of them will recombine before they can reach the plates.
+We can find the number of ions which reach the plates in this case
+in the following way:&mdash;In consequence of the movement of the ions
+the slab of ionized gas will broaden out and will consist of three
+portions, one in which there are nothing but positive ions,&mdash;this is
+on the side of the negative plate,&mdash;another on the side of the positive
+plate in which there are nothing but negative ions, and a portion
+between these in which there are both positive and negative ions;
+it is in this layer that recombination takes place, and here if n is the
+number of positive or negative ions at the time t after the flash of
+Röntgen rays,</p>
+
+<p class="center1">n = n<span class="su">0</span>/(1 + &alpha;n<span class="su">0</span>t).</p>
+
+<p class="noind">With the same notation as before, the breadth of either of the outer
+layers will in time dt increase by X(k<span class="su">1</span> + k<span class="su">2</span>)dt, and the number of
+ions in it by X(k<span class="su">1</span> + k<span class="su">2</span>)ndt; these ions will reach the plate, the outer
+layers will receive fresh ions until the middle one disappears, which
+it will do after a time l/X(k<span class="su">1</span> + k<span class="su">2</span>), where l is the thickness of the
+slab ab of ionized gas; hence N, the number of ions reaching either
+plate, is given by the equation</p>
+
+<table class="math0" summary="math">
+<tr> <td rowspan="2">N = <span class="f200">&int;</span><span class="su1">0</span><span class="sp1">l/X(k1+k2)</span></td>
+ <td>n<span class="su">0</span>X(k<span class="su">1</span> + k<span class="su">2</span>)</td>
+ <td rowspan="2">dt =</td> <td>X(k<span class="su">1</span> + k<span class="su">2</span>)</td>
+ <td rowspan="2">log <span class="f250">(</span>1 +</td> <td>n<span class="su">0</span>&alpha;l</td>
+ <td rowspan="2"><span class="f250">)</span>.</td></tr>
+<tr> <td class="denom">1 + n<span class="su">0</span>&alpha;t</td>
+ <td class="denom">&alpha;</td>
+ <td class="denom">X(k<span class="su">1</span> + k<span class="su">2</span>)</td></tr></table>
+
+<p class="noind">If Q is the charge received by the plate,</p>
+
+<table class="math0" summary="math">
+<tr> <td rowspan="2">Q = Ne =</td> <td>X</td> <td rowspan="2">log<span class="f250">(</span>1 +</td>
+ <td>Q<span class="su">0</span>&epsilon;</td> <td rowspan="2"><span class="f250">)</span>,</td></tr>
+<tr> <td class="denom">4&pi;&epsilon;</td> <td class="denom">4&pi;X</td></tr></table>
+
+<p class="noind">where Q<span class="su">0</span> = n<span class="su">0</span>le is the charge received by the plate when the electric
+force is large enough to prevent recombination, and &epsilon; = &alpha;4&pi;e(R<span class="su">1</span> + R<span class="su">2</span>).
+We can from this result deduce the value of &epsilon; and hence the value
+of &alpha; when R<span class="su">1</span> + R<span class="su">2</span> is known.</p>
+
+<p><i>Distribution of Electric Force when a Current is passing through an
+Ionized Gas.</i>&mdash;Let the two plates be at right angles to the axis of x;
+then we may suppose that between the plates the electric intensity
+X is everywhere parallel to the axis of x. The velocities of both the
+positive and negative ions are assumed to be proportional to X. Let
+k<span class="su">1</span>X, k<span class="su">2</span>X represent these velocities respectively; let n<span class="su">1</span>, n<span class="su">2</span> be respectively
+the number of positive and negative ions per unit volume at
+a point fixed by the co-ordinate x; let q be the number of positive
+or negative ions produced in unit time per unit volume at this
+point; and let the number of ions which recombine in unit volume
+in unit time be &alpha;n<span class="su">1</span>n<span class="su">2</span>; then if e is the charge on the ion, the volume
+density of the electrification is (n<span class="su">1</span> &minus; n<span class="su">2</span>)e, hence</p>
+
+<table class="math0" summary="math">
+<tr> <td>dX</td> <td rowspan="2">= 4&pi;(n<span class="su">1</span> &minus; n<span class="su">2</span>)e &emsp;&emsp;(1).</td></tr>
+<tr> <td class="denom">dx</td></tr></table>
+
+<p class="noind">If I is the current through unit area of the gas and if we neglect
+any diffusion except that caused by the electric field,</p>
+
+<p class="center1">n<span class="su">1</span>ek<span class="su">1</span>X + n<span class="su">2</span>ek<span class="su">2</span>X = I &emsp;&emsp;(2).</p>
+
+<p class="noind">From equations (1) and (2) we have</p>
+
+<table class="math0" summary="math">
+<tr><td rowspan="2">n<span class="su">1</span>e =</td> <td>1</td> <td rowspan="2"><span class="f250">(</span></td>
+ <td>I</td> <td rowspan="2">+</td> <td>k<span class="su">2</span></td> <td rowspan="2">&nbsp;</td> <td>dX</td>
+ <td rowspan="2"><span class="f250">)</span> &emsp;&emsp;(3),</td></tr>
+<tr><td class="denom">k<span class="su">1</span> + k<span class="su">2</span></td> <td class="denom">X</td>
+ <td class="denom">4&pi;</td> <td class="denom">dx</td></tr></table>
+
+<table class="math0" summary="math">
+<tr><td rowspan="2">n<span class="su">1</span>e =</td> <td>1</td> <td rowspan="2"><span class="f250">(</span></td>
+ <td>I</td> <td rowspan="2">-</td> <td>k<span class="su">2</span></td> <td rowspan="2">&nbsp;</td> <td>dX</td>
+ <td rowspan="2"><span class="f250">)</span> &emsp;&emsp;(4),</td></tr>
+<tr><td class="denom">k<span class="su">1</span> + k<span class="su">2</span></td> <td class="denom">X</td>
+ <td class="denom">4&pi;</td> <td class="denom">dx</td></tr></table>
+
+<p class="noind">and from these equations we can, if we know the distribution of
+electric intensity between the plates, calculate the number of positive
+and negative ions.</p>
+
+<p>In a steady state the number of positive and negative ions in
+unit volume at a given place remains constant, hence neglecting
+the loss by diffusion, we have</p>
+
+<table class="math0" summary="math">
+<tr><td>d</td> <td rowspan="2">(k<span class="su">1</span>n<span class="su">1</span>X) = q &minus; &alpha;n<span class="su">1</span>n<span class="su">2</span> &emsp;&emsp;(5).</td></tr>
+<tr><td class="denom">dx</td></tr></table>
+
+<table class="math0" summary="math">
+<tr><td rowspan="2">-</td> <td>d</td> <td rowspan="2">(k<span class="su">2</span>n<span class="su">2</span>X) = q &minus; &alpha;n<span class="su">1</span>n<span class="su">2</span> &emsp;&emsp;(6).</td></tr>
+<tr><td class="denom">dx</td></tr></table>
+
+<p class="noind">If k<span class="su">1</span> and k<span class="su">2</span> are constant, we have from (1), (5) and (6)</p>
+
+<table class="math0" summary="math">
+<tr><td>d²X²</td> <td rowspan="2">= 8&pi;e(q &minus; &alpha;n<span class="su">1</span>n<span class="su">2</span>)<span class="f250">(</span></td>
+ <td>1</td> <td rowspan="2">+</td> <td>1</td> <td rowspan="2"><span class="f250">)</span> &emsp;&emsp;(7).</td></tr>
+<tr><td class="denom">dx²</td> <td class="denom">k<span class="su">1</span></td> <td class="denom">k<span class="su">2</span></td></tr></table>
+
+<p class="noind">an equation which is very useful, <span class="correction" title="amended from becaus">because</span> it enables us, if we know
+the distribution of X², to find whether at any point in the gas
+the ionization is greater or less than the recombination of the ions.
+We see that q &minus; &alpha;n<span class="su">1</span>n<span class="su">2</span>, which is the excess of ionization over recombination,
+is proportional to d²X²/dx². Thus when the ionization
+exceeds the recombination, <i>i.e.</i> when q &minus; &alpha;n<span class="su">1</span>n<span class="su">2</span> is positive, the curve
+for X² is convex to the axis of x, while when the recombination
+exceeds the ionization the curve for X² will be concave to the axis of x.
+Thus, for example, fig. 11 represents the curve for X² observed by
+Graham (<i>Wied. Ann.</i> 64, p. 49) in a tube through which a steady
+current is passing. Interpreting it by equation (7), we infer that
+ionization was much in excess of recombination at A and B, slightly
+so along C, while along D the recombination exceeded the ionization.
+Substituting in equation (7) the values of n<span class="su">1</span>, n<span class="su">2</span> given in (3), (4),
+we get</p>
+
+<table class="math0" summary="math">
+<tr><td>d²X²</td> <td rowspan="2">8&pi;e<span class="f250">[</span>q -</td>
+ <td>&alpha;</td> <td rowspan="2"><span class="f250">(</span>I +</td>
+ <td>k<span class="su">2</span></td> <td rowspan="2">&nbsp;</td> <td>dX²</td> <td rowspan="2"><span class="f250">)(</span>I -</td>
+ <td>k<span class="su">2</span></td> <td rowspan="2">&nbsp;</td> <td>dX²</td> <td rowspan="2"><span class="f250">)](</span></td>
+ <td>1</td> <td rowspan="2">+</td> <td>1</td> <td rowspan="2"><span class="f250">)</span>&emsp;&emsp;(8).</td></tr>
+<tr><td class="denom">dx²</td> <td class="denom">e²X²(k<span class="su">1</span> + k<span class="su">2</span>)²</td>
+ <td class="denom">8&pi;</td> <td class="denom">dx</td> <td class="denom">8&pi;</td> <td class="denom">dx</td>
+ <td class="denom">k<span class="su">1</span></td> <td class="denom">k<span class="su">2</span></td></tr></table>
+
+<table class="nobctr" summary="Illustration">
+<tr><td class="figcenter"><img style="width:422px; height:262px" src="images/img871.jpg" alt="" /></td></tr>
+<tr><td class="caption sc">Fig. 11.</td></tr></table>
+
+<p>This equation can be solved (see Thomson, <i>Phil. Mag.</i> xlvii.
+P. 253), when q is constant and k<span class="su">1</span> = k<span class="su">2</span>. From the solution it appears
+that if X<span class="su">1</span> be the value of x close to one of the plates, and X<span class="su">0</span> the
+value midway between them,</p>
+
+<table class="math0" summary="math">
+<tr><td rowspan="2">X<span class="su">1</span>/X<span class="su">0</span> =</td> <td>1</td></tr>
+<tr><td class="denom">&beta;<span class="sp">2</span> &minus; 2/&beta;</td></tr></table>
+
+<p class="noind">where &beta; = 8&pi;ek<span class="su">1</span>/&alpha;.</p>
+
+<p><span class="pagenum"><a name="page872" id="page872"></a>872</span></p>
+
+<p>Since e = 4&times;10<span class="sp">-10</span>, &alpha; = 2&times;10<span class="sp">-6</span>, and k<span class="su">1</span> for air at atmospheric
+pressure = 450, &beta; is about 2.3 for air at atmospheric pressure and it
+becomes much greater at lower pressures.</p>
+
+<p>Thus X<span class="su">1</span>/X<span class="su">0</span> is always greater than unity, and the value of the
+ratio increases from unity to infinity as &beta; increases from zero to
+infinity. As &beta; does not involve either q or I, the ratio of X<span class="su">1</span> to X<span class="su">0</span>
+is independent of the strength of the current and of the intensity
+of the ionization.</p>
+
+<p>No general solution of equation (8) has been found when k<span class="su">1</span> is
+not equal to k<span class="su">2</span>, but we can get an approximation to the solution
+when q is constant. The equations (1), (2), (3), (4) are satisfied
+by the values&mdash;</p>
+
+<p class="center1">n<span class="su">1</span> = n<span class="su">2</span> = (q/&alpha;)<span class="sp">1/2</span></p>
+
+<table class="math0" summary="math">
+<tr><td rowspan="2">k<span class="su">1</span>n<span class="su">1</span>Xe =</td>
+ <td>k<span class="su">1</span></td> <td rowspan="2">I,</td></tr>
+<tr><td class="denom">k<span class="su">1</span> + k<span class="su">2</span></td></tr></table>
+
+<table class="math0" summary="math">
+<tr><td rowspan="2">k<span class="su">2</span>n<span class="su">2</span>Xe =</td>
+ <td>k<span class="su">2</span></td> <td rowspan="2">I,</td></tr>
+<tr><td class="denom">k<span class="su">1</span> + k<span class="su">2</span></td></tr></table>
+
+<table class="math0" summary="math">
+<tr><td rowspan="2">X =<span class="f250">(</span></td>
+ <td>&alpha;</td> <td rowspan="2"><span class="f250">)</span><span class="sp" style="bottom: 1.5em;">1/2</span></td>
+ <td>I</td> <td rowspan="2">.</td></tr>
+<tr><td class="denom">q</td> <td class="denom">e(k<span class="su">1</span> + k<span class="su">2</span>)</td></tr></table>
+
+<p class="noind">These solutions cannot, however, hold right up to the surface of
+the plates, for across each unit of area, at a point P, k<span class="su">1</span>I/(k<span class="su">1</span> + k<span class="su">2</span>)e
+positive ions pass in unit time, and these must all come from the
+region between P and the positive plate. If &lambda; is the distance of P
+from this plate, this region cannot furnish more than q&lambda; positive
+ions, and only this number if there are no recombinations. Hence
+the solution cannot hold when q&lambda; is less than k<span class="su">1</span>I/(k<span class="su">1</span> + k<span class="su">2</span>)e, or where
+&lambda; is less than k<span class="su">1</span>I/(k<span class="su">1</span> + k<span class="su">2</span>)qe.</p>
+
+<p>Similarly the solution cannot hold nearer to the negative plate
+than the distance k<span class="su">2</span>I/(k<span class="su">1</span> + k<span class="su">2</span>)qe.</p>
+
+<p>The force in these layers will be greater than that in the middle
+of the gas, and so the loss of ions by recombination will be smaller
+in comparison with the loss due to the removal of the ions by the
+current. If we assume that in these layers the loss of ions by
+recombination can be neglected, we can by the method of the
+next article find an expression for the value of the electric force at
+any point in the layer. This, in conjunction with the value
+X<span class="su">0</span> = (&alpha;/q)<span class="sp">1/2</span> · I/e(k<span class="su">1</span> + k<span class="su">2</span>)
+for the gas outside the layer, will give the value
+of X at any point between the plates. It follows from this investigation
+that if X<span class="su">1</span> and X<span class="su">2</span> are the values of X at the positive and negative
+plates respectively, and X<span class="su">0</span> the value of X outside the layer,</p>
+
+<table class="math0" summary="math">
+<tr><td rowspan="2">X<span class="su">1</span> = X<span class="su">0</span><span class="f250">(</span>1 +</td>
+ <td>k<span class="su">1</span></td> <td rowspan="2">&nbsp;</td> <td>1</td>
+ <td rowspan="2"><span class="f250">)</span><span class="sp" style="bottom: 1.5em;">1/2</span>,&emsp;</td>
+ <td rowspan="2">X<span class="su">2</span> = X<span class="su">0</span><span class="f250">(</span>1 +</td>
+ <td>k<span class="su">2</span></td> <td rowspan="2">&nbsp;</td> <td>1</td>
+ <td rowspan="2"><span class="f250">)</span><span class="sp" style="bottom: 1.5em;">1/2</span>,</td></tr>
+<tr><td class="denom">k<span class="su">2</span></td> <td class="denom">&epsilon;</td>
+ <td class="denom">k<span class="su">1</span></td> <td class="denom">&epsilon;</td></tr></table>
+
+<p class="noind">where &epsilon; = &alpha;/4&pi;e(k<span class="su">1</span> + k<span class="su">2</span>). Langevin found that for air at a pressure
+of 152 mm. &epsilon; = 0.01, at 375 mm. &epsilon; = 0.06, and at 760 mm. &epsilon; = 0.27.
+Thus at fairly low pressures 1/&epsilon; is large, and we have approximately</p>
+
+<table class="math0" summary="math">
+<tr><td rowspan="2">X<span class="su">1</span> = X<span class="su">0</span><span class="f250">(</span></td>
+ <td>k<span class="su">1</span></td> <td rowspan="2"><span class="f250">)</span><span class="sp" style="bottom: 1.5em;">1/2</span></td>
+ <td>1</td> <td rowspan="2">,&emsp;</td>
+ <td rowspan="2">X<span class="su">2</span> = X<span class="su">0</span><span class="f250">(</span></td>
+ <td>k<span class="su">2</span></td> <td rowspan="2"><span class="f250">)</span><span class="sp" style="bottom: 1.5em;">1/2</span></td>
+ <td>1</td> <td rowspan="2">.</td></tr>
+<tr><td class="denom">k<span class="su">2</span></td> <td class="denom">&radic;&epsilon;</td>
+ <td class="denom">k<span class="su">1</span></td> <td class="denom">&radic;&epsilon;</td></tr></table>
+
+<p class="noind">Therefore</p>
+
+<p class="center1">X<span class="su">1</span>/X<span class="su">2</span> = k<span class="su">1</span>/k<span class="su">2</span>,</p>
+
+<table class="nobctr" style="float: left; width: 330px;" summary="Illustration">
+<tr><td class="figleft1"><img style="width:282px; height:457px" src="images/img872.jpg" alt="" /></td></tr>
+<tr><td class="caption sc">Fig. 12.</td></tr></table>
+
+<p class="noind">or the force at the positive plate is to that at the negative plate as
+the velocity of the positive ion is to that of the negative ion. Thus
+the force at the negative plate
+is greater than that at the positive.
+The falls of potential
+V<span class="su">1</span>, V<span class="su">2</span> at the two layers when
+1/&epsilon; is large can be shown to be
+given by the equations</p>
+
+<table class="math0" summary="math">
+<tr><td rowspan="2">V<span class="su">1</span> = 8&pi;²<span class="f250">(</span></td> <td>&epsilon;</td>
+ <td rowspan="2"><span class="f250">)</span><span class="sp" style="bottom: 1.5em;">3/2</span>k<span class="su">1</span><span class="f250">(</span></td>
+ <td>k<span class="su">1</span></td> <td rowspan="2"><span class="f250">)</span><span class="sp" style="bottom: 1.5em;">1/2</span>i²,</td></tr>
+<tr><td class="denom">q&alpha;</td> <td class="denom">k<span class="su">2</span></td></tr></table>
+
+<table class="math0" summary="math">
+<tr><td rowspan="2">V<span class="su">2</span> = 8&pi;²<span class="f250">(</span></td> <td>&epsilon;</td>
+ <td rowspan="2"><span class="f250">)</span><span class="sp" style="bottom: 1.5em;">3/2</span>k<span class="su">2</span><span class="f250">(</span></td>
+ <td>k<span class="su">2</span></td> <td rowspan="2"><span class="f250">)</span><span class="sp" style="bottom: 1.5em;">1/2</span>i²,</td></tr>
+<tr><td class="denom">q&alpha;</td> <td class="denom">k<span class="su">1</span></td></tr></table>
+
+<p class="noind">hence</p>
+
+<p class="center1">V<span class="su">1</span>/V<span class="su">2</span> = k<span class="su">1</span>²/k<span class="su">2</span>²,</p>
+
+<p class="noind">so that the potential falls at the
+electrodes are proportional to
+the squares of the velocities
+of the ions. The change in
+potential across the layers is
+proportional to the square of
+the current, while the potential
+change between the layers is
+proportional to the current,
+the total potential difference
+between the plates is the sum
+of these changes, hence the
+relation between V and i will
+be of the form</p>
+
+<p class="center1">V = Ai + Bi².</p>
+
+<p>Mie (<i>Ann. der. Phys.</i>, 1904,
+13, P. 857) has by the method
+of successive approximations obtained solutions of equation (8) (i.)
+when the current is only a small fraction of the saturation current,
+(ii.) when the current is nearly saturated. The results of his investigations
+are represented in fig. 12, which represents the distribution of
+electric force along the path of the current for various values of the
+current expressed as fractions of the saturation current. It will
+be seen that until the current amounts to about one-fifth of the
+maximum current, the type of solution is the one just indicated, <i>i.e.</i>
+the electric force is constant except in the neighbourhood of the electrodes
+when it increases rapidly.</p>
+
+<p>Though we are unable to obtain a general solution of the equation
+(8), there are some very important special cases in which that
+equation can be solved without difficulty. We shall consider two
+of these, the first being that when the current is saturated. In this
+case there is no loss of ions by recombination, so that using the same
+notation as before we have</p>
+
+<table class="math0" summary="math">
+<tr><td>d</td> <td rowspan="2">(n<span class="su">1</span>k<span class="su">1</span>X) = q,</td></tr>
+<tr><td class="denom">dx</td></tr></table>
+
+<table class="math0" summary="math">
+<tr><td>d</td> <td rowspan="2">(n<span class="su">2</span>k<span class="su">2</span>X) = -q.</td></tr>
+<tr><td class="denom">dx</td></tr></table>
+
+<p class="noind">The solutions of which if q is constant are</p>
+
+<table class="ws" summary="Contents">
+<tr><td class="tcl">n<span class="su">1</span>k<span class="su">1</span>X = qx,</td></tr>
+<tr><td class="tcl">n<span class="su">2</span>k<span class="su">2</span>X = I/e &minus; qx = q(l &minus; x),</td></tr></table>
+
+<p class="noind">if l is the distance between the plates, and x = 0 at the positive
+electrode. Since</p>
+
+<p class="center1">dX/dx = 4&pi;(n<span class="su">1</span> &minus; n<span class="su">2</span>)e,</p>
+
+<p class="noind">we get</p>
+
+<table class="math0" summary="math">
+<tr><td>1</td> <td rowspan="2">&nbsp;</td> <td>dX²</td>
+ <td rowspan="2">= qx<span class="f250">{</span></td> <td>1</td> <td rowspan="2">+</td> <td>1</td>
+ <td rowspan="2"><span class="f250">}</span>- q</td> <td>l</td> <td rowspan="2">,</td></tr>
+<tr><td class="denom">8&pi;</td> <td class="denom">d²x</td> <td class="denom">k<span class="su">1</span></td>
+ <td class="denom">k<span class="su">2</span></td> <td class="denom">k<span class="su">2</span></td></tr></table>
+
+<p class="noind">or</p>
+
+<table class="math0" summary="math">
+<tr><td>X²</td> <td rowspan="2">= q</td> <td>x²</td> <td rowspan="2"><span class="f250">(</span></td>
+ <td>1</td> <td rowspan="2">+</td> <td>1</td> <td rowspan="2"><span class="f250">)</span>- q</td>
+ <td>lx</td> <td rowspan="2">+ C,</td></tr>
+<tr><td class="denom">8&pi;</td> <td class="denom">2</td> <td class="denom">k<span class="su">1</span></td>
+ <td class="denom">k<span class="su">2</span></td> <td class="denom">k<span class="su">2</span></td></tr></table>
+
+<p class="noind">where C is a quantity to be determined by the condition that
+<span class="f150">&int;</span><span class="sp1">l</span><span class="su1">0</span> Xdx = V,
+where V is the given potential difference between the
+plates. When the force is a minimum dX/dx = 0, hence at this point</p>
+
+<table class="math0" summary="math">
+<tr><td rowspan="2">x =</td> <td>l k<span class="su">1</span></td>
+ <td rowspan="2">, &emsp;l &minus; x =</td> <td>l k<span class="su">2</span></td> <td rowspan="2">.</td></tr>
+<tr><td class="denom">k<span class="su">1</span> + k<span class="su">2</span></td> <td class="denom">k<span class="su">1</span> + k<span class="su">2</span></td></tr></table>
+
+<p class="noind">Hence the ratio of the distances of this point from the positive and
+negative plates respectively is equal to the ratio of the velocities of
+the positive and negative ions.</p>
+
+<p>The other case we shall consider is the very important one in
+which the velocity of the negative ion is exceedingly large compared
+with the positive; this is the case in flames where, as Gold (<i>Proc.
+Roy. Soc.</i> 97, p. 43) has shown, the velocity of the negative ion is
+many thousand times the velocity of the positive; it is also very
+probably the case in all gases when the pressure is low. We may get
+the solution of this case either by putting k<span class="su">1</span>/k<span class="su">2</span> = 0 in equation (8),
+or independently as follows:&mdash;Using the same notation as before,
+we have</p>
+
+<p class="center1">i = n<span class="su">1</span>k<span class="su">1</span>Xe + n<span class="su">2</span>k<span class="su">2</span>Xe,</p>
+
+<table class="math0" summary="math">
+<tr><td>d</td> <td rowspan="2">(n<span class="su">2</span>k<span class="su">2</span>X) = q &minus; &alpha;n<span class="su">1</span>n<span class="su">2</span>,</td></tr>
+<tr><td class="denom">dx</td></tr></table>
+
+<table class="math0" summary="math">
+<tr><td>dX</td> <td rowspan="2">= 4&pi; (n<span class="su">1</span> &minus; n<span class="su">2</span>)e.</td></tr>
+<tr><td class="denom">dx</td></tr></table>
+
+<p class="noind">In this case practically all the current is carried by the negative
+ions so that i = n<span class="su">2</span>k<span class="su">2</span>Xe, and therefore q = &alpha;n<span class="su">1</span>n<span class="su">2</span>.</p>
+
+<p class="noind">Thus</p>
+
+<p class="center1">n<span class="su">2</span> = i/k<span class="su">2</span>Xe, &emsp;n<span class="su">1</span> = qk<span class="su">2</span>Xe/&alpha;i.</p>
+
+<p class="noind">Thus</p>
+
+<table class="math0" summary="math">
+<tr><td>dX</td> <td rowspan="2">=</td>
+ <td>4&pi;e²k<span class="su">2</span>qX</td> <td rowspan="2">-</td>
+ <td>4&pi;i</td> <td rowspan="2">,</td></tr>
+<tr><td class="denom">dx</td> <td class="denom">&alpha;i</td> <td class="denom">k<span class="su">2</span>X</td></tr></table>
+
+<p class="noind">or</p>
+
+<table class="math0" summary="math">
+<tr><td>dX²</td> <td rowspan="2">-</td>
+ <td>8&pi;e²k<span class="su">2</span>qX²</td> <td rowspan="2">= -</td>
+ <td>8&pi;i</td> <td rowspan="2">.</td></tr>
+<tr><td class="denom">dx</td> <td class="denom">&alpha;i</td> <td class="denom">k<span class="su">2</span></td></tr></table>
+
+<p class="noind">The solution of this equation is</p>
+
+<table class="math0" summary="math">
+<tr><td rowspan="2">X² =</td> <td>&alpha;</td> <td rowspan="2">&nbsp;</td> <td>i²</td>
+ <td rowspan="2">+ C&epsilon; <span class="sp">8&pi;e² k2qx/&alpha;i</span>.</td></tr>
+<tr><td class="denom">q</td> <td class="denom">k²<span class="su">2</span>e²</td></tr></table>
+
+<p class="noind">Here x is measured from the positive electrode; it is more convenient
+in this case, however, to measure it from the negative electrode.
+If x be the distance from the negative electrode at which the electric
+force is X, we have from equation (7)</p>
+
+<table class="math0" summary="math">
+<tr><td rowspan="2">X² =</td> <td>&alpha;</td> <td rowspan="2">&nbsp;</td> <td>i²</td>
+ <td rowspan="2">+ C¹&epsilon; <span class="sp">8&pi;e² k2qx/&alpha;i</span>.</td></tr>
+<tr><td class="denom">q</td> <td class="denom">k²<span class="su">2</span>e²</td></tr></table>
+
+<p class="noind">To find the value of C¹ we see by equation (7) that</p>
+
+<table class="math0" summary="math">
+<tr><td>d²X²</td> <td rowspan="2">&nbsp;</td> <td>k<span class="su">1</span>k<span class="su">2</span></td>
+ <td rowspan="2">&nbsp;</td> <td>1</td> <td rowspan="2">= q &minus; &alpha;n<span class="su">1</span>n<span class="su">2</span>;</td></tr>
+<tr><td class="denom">dX²</td> <td class="denom">k<span class="su">1</span> + k<span class="su">2</span></td> <td class="denom">8&pi;e</td></tr></table>
+
+<p class="noind">hence</p>
+
+<table class="math0" summary="math">
+<tr><td rowspan="2"><span class="f250">[</span></td> <td>dX²</td> <td rowspan="2">&nbsp;</td> <td>k<span class="su">1</span>k<span class="su">2</span></td>
+ <td rowspan="2">&nbsp;</td> <td>1</td>
+ <td rowspan="2"><span class="f250">]</span><span class="sp" style="bottom: 1.5em;">x1</span> =
+ <span class="f200">&int;</span><span class="sp1">x1</span><span class="su1">0</span>
+ (q &minus; &alpha;n<span class="su">1</span>n<span class="su">2</span>)dx.</td></tr>
+<tr><td class="denom">dX</td> <td class="denom">k<span class="su">1</span> + k<span class="su">2</span></td>
+ <td class="denom">8&pi;e</td></tr></table>
+
+<p class="noind">The right hand side of this equation is the excess of ionization
+over recombination in the region extending from the cathode to x<span class="su">1</span>;
+it must therefore, when things are in a steady state, equal the excess
+of the number of negative ions which leave this region over those
+which enter it. The number which leave is i/e and the number which
+enter is i<span class="su">0</span>/e, if it is the current of negative ions coming from unit area
+<span class="pagenum"><a name="page873" id="page873"></a>873</span>
+of the cathode, as hot metal cathodes emit large quantities of
+negative electricity i<span class="su">0</span> may in some cases be considerable, thus the
+right hand side of equation is (i &minus; i<span class="su">0</span>)/e. When x<span class="su">1</span> is large dX²/dx = 0;
+hence we have from equation</p>
+
+<table class="math0" summary="math">
+<tr><td rowspan="2">C¹ =</td> <td>&alpha;i(i &minus; i<span class="su">0</span>)</td>
+ <td rowspan="2">&nbsp;</td> <td>k<span class="su">1</span> + k<span class="su">2</span></td> <td rowspan="2">,</td></tr>
+<tr><td class="denom">qk<span class="su">1</span>k<span class="su">2</span>e²</td> <td class="denom">k<span class="su">2</span></td></tr></table>
+
+<p class="noind">and since k<span class="su">1</span> is small compared with k<span class="su">2</span>, we have</p>
+
+<table class="math0" summary="math">
+<tr><td rowspan="2">X² =</td> <td>&alpha; i²</td>
+ <td rowspan="2"><span class="f250">(</span>1 +</td> <td>k<span class="su">2</span></td>
+ <td rowspan="2">&nbsp;</td> <td>i &minus; i<span class="su">0</span></td>
+ <td rowspan="2">&epsilon; <span class="sp">-8&pi;e² k2 · qx/&alpha; · i</span>
+ <span class="f250">)</span>.</td></tr>
+<tr><td class="denom">qk²<span class="su1">2</span> e²</td> <td class="denom">k<span class="su">1</span></td> <td class="denom">i</td></tr></table>
+
+<p class="noind">From the values which have been found for k<span class="su">2</span> and &alpha;, we know that
+8&pi;ek<span class="su">2</span>/&alpha; is a large quantity, hence the second term inside the bracket
+will be very small when eqx is equal to or greater than i; thus this
+term will be very small outside a layer of gas next the cathode of
+such thickness that the number of ions produced on it would be
+sufficient, if they were all utilized for the purpose, to carry the
+current; in the case of flames this layer is exceedingly thin unless
+the current is very large. The value of the electric force in the
+uniform part of the field is equal to i/k<span class="su">2</span>e · &radic;<span class="ov">a/q</span>, while when i<span class="su">0</span> = 0,
+the force at the cathode itself bears to the uniform force the ratio of
+(k<span class="su">1</span> + k<span class="su">2</span>)<span class="sp">1/2</span> to k<span class="su">1</span><span class="sp">1/2</span>. As k<span class="su">1</span> is many thousand times k<span class="su">2</span> the force increases
+with great rapidity as we approach the cathode; this is a very
+characteristic feature of the passage of electricity through flames
+and hot gases. Thus in an experiment made by H. A. Wilson with a
+flame 18 cm. long, the drop of potential within 1 centimetre of the
+cathode was about five times the drop in the other 17 cm. of the tube.
+The relation between the current and the potential difference when
+the velocity of the negative ion is much greater than the positive is
+very easily obtained. Since the force is uniform and equal to i/k<span class="su">2</span>e · &radic;<span class="ov">a/q</span>,
+until we get close to the cathode the fall of potential in this part
+of the discharge will be very approximately equal to i/k<span class="su">2</span>e · &radic;<span class="ov">(a/q) l</span>,
+where l is the distance between the electrodes. Close to the cathode,
+the electric force when i<span class="su">0</span> is not nearly equal to i is approximately
+given by the equation</p>
+
+<table class="math0" summary="math">
+<tr><td rowspan="2">X =</td> <td>i</td>
+ <td rowspan="2"><span class="f250">(</span></td> <td>&alpha;</td>
+ <td rowspan="2"><span class="f250">)</span><span class="sp" style="bottom: 1.5em;">1/2</span>
+ &epsilon; <span class="sp">&minus;4&pi; e²k2 qx / &alpha;i</span>,</td></tr>
+<tr><td class="denom">e(k<span class="su">1</span>k<span class="su">2</span>)<span class="sp">1/2</span></td> <td class="denom">q</td></tr></table>
+
+<p class="noind">and the fall of potential at the cathode is equal approximately to
+<span class="f150">&int;</span><span class="sp1">&infin;</span><span class="su1">0</span> Xdx,
+that is to</p>
+
+<table class="math0" summary="math">
+<tr><td>i</td> <td rowspan="2"><span class="f250">(</span></td>
+ <td>&alpha;</td> <td rowspan="2"><span class="f250">)</span><span class="sp" style="bottom: 1.5em;">1/2</span></td>
+ <td>&alpha; i</td> <td rowspan="2">.</td></tr>
+<tr><td class="denom">e(k<span class="su">1</span>k<span class="su">2</span>)<span class="sp">1/2</span></td> <td class="denom">q</td>
+ <td class="denom">4&pi;e²k<span class="su">2</span>q</td></tr></table>
+
+<p class="noind">The potential difference between the plates is the sum of the fall of
+potential in the uniform part of the discharge plus the fall at the
+cathode, hence</p>
+
+<table class="math0" summary="math">
+<tr><td rowspan="2">V =<span class="f250">(</span></td> <td>&alpha;</td>
+ <td rowspan="2"><span class="f250">)</span><span class="sp" style="bottom: 1.5em;">1/2</span></td> <td>i</td>
+ <td rowspan="2"><span class="f250">(</span>il +</td> <td>i &alpha;²</td>
+ <td rowspan="2">&nbsp;</td> <td>1</td>
+ <td rowspan="2"><span class="f250">)</span>.</td></tr>
+<tr><td class="denom">q</td> <td class="denom">ek<span class="su">2</span></td>
+ <td class="denom">4&pi;e²q</td> <td class="denom">&radic;(k<span class="su">1</span>k<span class="su">2</span>)</td></tr></table>
+
+<p class="noind">The fall of potential at the cathode is proportional to the square of
+the current, while the fall in the rest of the circuit is directly proportional
+to the current. In the case of flames or hot gases, the fall
+of potential at the cathode is much greater than that in the rest of the
+circuit, so that in such cases the current through the gas varies nearly
+as the square root of the potential difference. The equation we have
+just obtained is of the form</p>
+
+<p class="center1">V = Ai + Bi²,</p>
+
+<p class="noind">and H. A. Wilson has shown that a relation of this form represents
+the results of his experiments on the conduction of electricity through
+flames.</p>
+
+<p>The expression for the fall of potential at the cathode is inversely
+proportional to q<span class="sp">3/2</span>, q being the number of ions produced per cubic
+centimetre per second close to the cathode; thus any increase in
+the ionization at the cathode will diminish the potential fall at the
+cathode, and as practically the whole potential difference between
+the electrodes occurs at the cathode, a diminution in the potential
+fall there will be much more important than a diminution in the
+electric force in the uniform part of the discharge, when the force is
+comparatively insignificant. This consideration explains a very
+striking phenomenon discovered many years ago by Hittorf, who
+found that if he put a wire carrying a bead of a volatile salt into the
+flame, it produced little effect upon the current, unless it were placed
+close to the cathode where it gave rise to an enormous increase in
+the current, sometimes increasing the current more than a hundredfold.
+The introduction of the salt increases very largely the number
+of ions produced, so that q is much greater for a salted flame than
+for a plain one. Thus Hittorf&rsquo;s result coincides with the conclusions
+we have drawn from the theory of this class of conduction.</p>
+
+<p>The fall of potential at the cathode is proportional to i &minus; i<span class="su">0</span>,
+where i<span class="su">0</span> is the stream of negative electricity which comes from the
+cathode itself, thus as i<span class="su">0</span> increases the fall of potential at the cathode
+diminishes and the current sent by a given potential difference
+through the gas increases. Now all metals give out negative particles
+when heated, at a rate which increases very rapidly with the temperature,
+but at the same temperature some metals give out more than
+others. If the cathode is made of a metal which emits large quantities
+of negative particles, (i &minus; i<span class="su">0</span>) will for a given value of i be smaller
+than if the metal only emitted a small number of particles; thus the
+cathode fall will be smaller for the metal with the greater emissitivity,
+and the relation between the potential difference and the current
+will be different in the two cases. These considerations are confirmed
+by experience, for it has been found that the current between
+electrodes immersed in a flame depends to a great extent upon the
+metal of which the electrodes are made. Thus Pettinelli (<i>Acc. dei
+Lincei</i> [5], v. p. 118) found that, <i>ceteris paribus</i>, the current between
+two carbon electrodes was about 500 times that between two iron
+ones. If one electrode was carbon and the other iron, the current
+when the carbon was cathode and the iron anode was more than
+100 times the current when the electrodes were reversed. The
+emission of negative particles by some metallic oxides, notably
+those of calcium and barium, has been shown by Wehnelt (<i>Ann. der
+Phys.</i> 11, p. 425) to be far greater than that of any known metal,
+and the increase of current produced by coating the cathodes with
+these oxides is exceedingly large; in some cases investigated by
+Tufts and Stark (<i>Physik. Zeits.</i>, 1908, 5, p. 248) the current was
+increased many thousand times by coating the cathode with lime.
+No appreciable effect is produced by putting lime on the anode.</p>
+
+<p><i>Conduction when all the Ions are of one Sign.</i>&mdash;There are many
+important cases in which the ions producing the current come from
+one electrode or from a thin layer of gas close to the electrode, no
+ionization occurring in the body of the gas or at the other electrode.
+Among such cases may be mentioned those where one of the electrodes
+is raised to incandescence while the other is cold, or when the
+negative electrode is exposed to ultra-violet light. In such cases if
+the electrode at which the ionization occurs is the positive electrode,
+all the ions will be positively charged, while if it is the negative
+electrode the ions will all be charged negatively. The theory of
+this case is exceedingly simple. Suppose the electrodes are parallel
+planes at right angles to the axis of x; let X be the electric force
+at a distance x from the electrode where the ionization occurs, n
+the number of ions (all of which are of one sign) at this place per
+cubic centimetre, k the velocity of the ion under unit electric force,
+e the charge on an ion, and i the current per unit area of the electrode.
+Then we have dX/dx = 4&pi;ne, and if u is the velocity of the ion
+neu = i. But u = kX, hence we have kX/4&pi; · dX/dx = i, and since the right
+hand side of this equation does not depend upon x, we get kX²/8&pi;
+= ix + C, where C is a constant to be determined. If l is the distance
+between the plates, and V the potential difference between them,</p>
+
+<table class="math0" summary="math">
+<tr><td rowspan="2">V =<span class="f250">&int;</span><span class="sp" style="bottom: 1.5em;">l</span>
+ <span class="su1" style="margin-left: -1em;">0</span> Xdx =</td> <td>1</td>
+ <td rowspan="2"><span class="f250">&radic;</span></td> <td>8&pi;</td>
+ <td rowspan="2"><span class="f250">[</span>(il + C)<span class="sp">3/2</span> &minus; C<span class="sp">3/2</span>
+ <span class="f250">]</span>.</td></tr>
+<tr><td class="denom">i</td> <td class="denom">k</td></tr></table>
+
+<p class="noind">We shall show that when the current is far below the saturation
+value, C is very small compared with il, so that the preceding
+equation becomes</p>
+
+<p class="center1">V² = 8&pi;l³ i/k &emsp;&emsp;(1).</p>
+
+<p class="noind">To show that for small currents C is small compared with il, consider
+the case when the ionization is confined to a thin layer, thickness d
+close to the electrode, in that layer let n<span class="su">0</span> be the value of n, then
+we have q = &alpha;n<span class="su">0</span>² + i/ed. If X<span class="su">0</span> be the value of X when x = 0,
+kX<span class="su">0</span>n<span class="su">0</span>e = i, and,</p>
+
+<table class="math0" summary="math">
+<tr><td rowspan="2">C =</td> <td>kX<span class="su">0</span>²</td>
+ <td rowspan="2">=</td> <td>i²</td>
+ <td rowspan="2">=</td> <td>&alpha;</td>
+ <td rowspan="2">·</td> <td>i²</td> <td rowspan="2">&emsp;&emsp;(2).</td></tr>
+<tr><td class="denom">8&pi;</td> <td class="denom">n<span class="su">0</span>²ke·8&pi;</td>
+ <td class="denom">8&pi;ke²</td> <td class="denom">q + i/ed</td></tr></table>
+
+<p class="noind">Since &alpha;/8&pi;ke is, as we have seen, less than unity, C will be small
+compared with il, if i/(eq + i/d) is small compared with l. If I<span class="su">0</span> is
+the saturation current, q = I<span class="su">0</span>/ed, so that the former expression
+= id/(I<span class="su">0</span> + i), if i is small compared with I<span class="su">0</span>, this expression is small
+compared with d, and therefore <i>a fortiori</i> compared with l, so that we
+are justified in this case in using equation (1).</p>
+
+<p>From equation (2) we see that the current increases as the square
+of the potential difference. Here an increase in the potential
+difference produces a much greater percentage increase than in
+conduction through metals, where the current is proportional to the
+potential difference. When the ionization is distributed through
+the gas, we have seen that the current is approximately proportional
+to the square root of the potential, and so increases more slowly
+with the potential difference than currents through metals. From
+equation (1) the current is inversely proportional to the cube of the
+distance between the electrodes, so that it falls off with great rapidity
+as this distance is increased. We may note that for a given
+potential difference the expression for the current does not involve q,
+the rate of production of the ions at the electrode, in other words,
+if we vary the ionization the current will not begin to be affected
+by the strength of the ionization until this falls so low that the current
+is a considerable fraction of the saturation current. For the same
+potential difference the current is proportional to k, the velocity
+under unit electric force of the ion which carries the current. As the
+velocity of the negative ion is greater than that of the positive,
+the current when the ionization is confined to the neighbourhood of
+one of the electrodes will be greater when that electrode is made
+cathode than when it is anode. Thus the current will appear to
+pass more easily in one direction than in the opposite.</p>
+
+<p>Since the ions which carry the current have to travel all the way
+from one electrode to the other, any obstacle which is impervious
+to these ions will, if placed between the electrodes, stop the current
+<span class="pagenum"><a name="page874" id="page874"></a>874</span>
+to the electrode where there is no ionization. A plate of metal will
+be as effectual as one made of a non-conductor, and thus we get the
+remarkable result that by interposing a plate of an excellent conductor
+like copper or silver between the electrode, we can entirely
+stop the current. This experiment can easily be tried by using a
+hot plate as the electrode at which the ionization takes place: then
+if the other electrode is cold the current which passes when the hot
+plate is cathode can be entirely stopped by interposing a cold metal
+plate between the electrodes.</p>
+</div>
+
+<p><i>Methods of counting the Number of Ions.</i>&mdash;The detection of the
+ions and the estimation of their number in a given volume is
+much facilitated by the property they possess of promoting the
+condensation of water-drops in dust-free air supersaturated with
+water vapour. If such air contains no ions, then it requires about
+an eightfold supersaturation before any water-drops are formed;
+if, however, ions are present C. T. R. Wilson (<i>Phil. Trans.</i>
+189, p. 265) has shown that a sixfold supersaturation is sufficient
+to cause the water vapour to condense round the ions and to fall
+down as raindrops. The absence of the drops when no ions
+are present is due to the curvature of the drop combined with the
+surface tension causing, as Lord Kelvin showed, the evaporation
+from a small drop to be exceeding rapid, so that even if a drop of
+water were formed the evaporation would be so great in its early
+stages that it would rapidly evaporate and disappear. It has
+been shown, however (J. J. Thomson, <i>Application of Dynamics
+to Physics and Chemistry</i>, p. 164; <i>Conduction of Electricity
+through Gases</i>, 2nd ed. p. 179), that if a drop of water is charged
+with electricity the effect of the charge is to diminish the evaporation;
+if the drop is below a certain size the effect the charge has
+in promoting condensation more than counterbalances the effect
+of the surface tension in promoting evaporation. Thus the electric
+charge protects the drop in the most critical period of its growth.
+The effect is easily shown experimentally by taking a bulb connected
+with a piston arranged so as to move with great rapidity.
+When the piston moves so as to increase the volume of the air
+contained in the bulb the air is cooled by expansion, and if it was
+saturated with water vapour before it is supersaturated after the
+expansion. By altering the throw of the piston the amount of
+supersaturation can be adjusted within very wide limits. Let
+it be adjusted so that the expansion produces about a sixfold
+supersaturation; then if the gas is not exposed to any ionizing
+agents very few drops (and these probably due to the small
+amount of ionization which we have seen is always present in
+gases) are formed. If, however, the bulb is exposed to strong
+Röntgen rays expansion produces a dense cloud which gradually
+falls down and disappears. If the gas in the bulb at the time of
+its exposure to the Röntgen rays is subject to a strong electric
+field hardly any cloud is formed when the gas is suddenly
+expanded. The electric field removes the charged ions from the
+gas as soon as they are formed so that the number of ions present
+is greatly reduced. This experiment furnishes a very direct
+proof that the drops of water which form the cloud are only
+formed round the ions.</p>
+
+<p>This method gives us an exceedingly delicate test for the
+presence of ions, for there is no difficulty in detecting ten or so
+raindrops per cubic centimetre; we are thus able to detect the
+presence of this number of ions. This result illustrates the enormous
+difference between the delicacy of the methods of detecting
+ions and those for detecting uncharged molecules; we have seen
+that we can easily detect ten ions per cubic centimetre, but there
+is no known method, spectroscopic or chemical, which would
+enable us to detect a billion (10<span class="sp">12</span>) times this number of uncharged
+molecules. The formation of the water-drops round the charged
+ions gives us a means of counting the number of ions present
+in a cubic centimetre of gas; we cool the gas by sudden expansion
+until the supersaturation produced by the cooling is sufficient
+to cause a cloud to be formed round the ions, and the problem
+of finding the number of ions per cubic centimetre of gas is thus
+reduced to that of finding the number of drops per cubic centimetre
+in the cloud. Unless the drops are very few and far between
+we cannot do this by direct counting; we can, however, arrive
+at the result in the following way. From the amount of expansion
+of the gas we can calculate the lowering produced in its
+temperature and hence the total quantity of water precipitated.
+The water is precipitated as drops, and if all the drops are the
+same size the number per cubic centimetre will be equal to the
+volume of water deposited per cubic centimetre, divided by the
+volume of one of the drops. Hence we can calculate the number
+of drops if we know their size, and this can be determined by
+measuring the velocity with which they fall under gravity through
+the air.</p>
+
+<div class="condensed">
+<p>The theory of the fall of a heavy drop of water through a viscous
+fluid shows that v = <span class="spp">2</span>&frasl;<span class="suu">9</span>ga²/&mu;, where a is the radius of the drop, g the
+acceleration due to gravity, and &mu; the coefficient of viscosity of the
+gas through which the drop falls. Hence if we know v we can deduce
+the value of a and hence the volume of each drop and the number
+of drops.</p>
+
+<p><i>Charge on Ion.</i>&mdash;By this method we can determine the number of
+ions per unit volume of an ionized gas. Knowing this number we
+can proceed to determine the charge on an ion. To do this let us
+apply an electric force so as to send a current of electricity through
+the gas, taking care that the current is only a small fraction of the
+saturating current. Then if u is the sum of the velocities of the
+positive and negative ions produced in the electric field applied to
+the gas, the current through unit area of the gas is neu, where n is
+the number of positive or negative ions per cubic centimetre, and e
+the charge on an ion. We can easily measure the current through
+the gas and thus determine neu; we can determine n by the method
+just described, and u, the velocity of the ions under the given
+electric field, is known from the experiments of Zeleny and others.
+Thus since the product neu, and two of the factors n, u are known,
+we can determine the other factor e, the charge on the ion. This
+method was used by J. J. Thomson, and details of the method
+will be found in <i>Phil. Mag.</i> [5], 46, p. 528; [5], 48, p. 547; [6], 5,
+p. 346. The result of these measurements shows that the charge
+on the ion is the same whether the ionization is by Röntgen rays or
+by the influence of ultra-violet light on a metal plate. It is the
+same whether the gas ionized is hydrogen, air or carbonic acid,
+and thus is presumably independent of the nature of the gas. The
+value of e formed by this method was 3.4&times;10<span class="sp">-10</span> electrostatic units.</p>
+
+<p>H. A. Wilson (<i>Phil. Mag.</i> [6], 5, p. 429) used another method.
+Drops of water, as we have seen, condense more easily on negative
+than on positive ions. It is possible, therefore, to adjust the expansion
+so that a cloud is formed on the negative but not on the
+positive ions. Wilson arranged the experiments so that such a cloud
+was formed between two horizontal plates which could be maintained
+at different potentials. The charged drops between the plates were
+acted upon by a uniform vertical force which affected their rate of
+fall. Let X be the vertical electric force, e the charge on the drop,
+v<span class="su">1</span> the rate of fall of the drop when this force acts, and v the rate of
+fall due to gravity alone. Then since the rate of fall is proportionate
+to the force on the drop, if a is the radius of the drop, and &rho; its
+density, then</p>
+
+<table class="math0" summary="math">
+<tr><td>Xe + <span class="spp">4</span>&frasl;<span class="suu">3</span>&pi;&rho; ga³</td> <td rowspan="2">=</td>
+ <td>v<span class="su">1</span></td> <td rowspan="2">,</td></tr>
+<tr><td class="denom"><span class="spp">4</span>&frasl;<span class="suu">3</span>&pi; &rho;ga³</td> <td class="denom">v</td></tr></table>
+
+<p class="noind">or</p>
+
+<p class="center1">Xe = <span class="spp">4</span>&frasl;<span class="suu">3</span>&pi; &rho;ga³ (v<span class="su">1</span> &minus; v)/v.</p>
+
+<p class="noind">But</p>
+
+<p class="center1">v = <span class="spp">2</span>&frasl;<span class="suu">9</span>ga²&rho;/&mu;,</p>
+
+<p class="noind">so that</p>
+
+<table class="math0" summary="math">
+<tr><td rowspan="2">Xe = &radic; 2 · 9&pi;<span class="f250">&radic;</span></td> <td>&mu;<span class="sp1">3</span></td>
+ <td rowspan="2">·</td> <td>v<span class="sp1">3</span><span class="su1">2</span>(v<span class="su">1</span> &minus; v)</td> <td rowspan="2">.</td></tr>
+<tr><td class="denom">g &rho;</td> <td class="denom">v</td></tr></table>
+
+<p class="noind">Thus if X, v, v<span class="su">1</span> are known e can be determined. Wilson by this
+method found that e was 3.1&times;10<span class="sp">-10</span> electrostatic units. A few of
+the ions carried charges 2e or 3e.</p>
+
+<p>Townsend has used the following method to compare the charge
+carried by a gaseous ion with that carried by an atom of hydrogen
+in the electrolysis of solution. We have</p>
+
+<p class="center1">u/D = Ne/&Pi;,</p>
+
+<p class="noind">where D is the coefficient of diffusion of the ions through the gas,
+u the velocity of the ion in the same gas when acted on by unit
+electric force, N the number of molecules in a cubic centimetre of
+the gas when the pressure is &Pi; dynes per square centimetre, and e the
+charge in electrostatic units. This relation is obtained on the
+hypothesis that N ions in a cubic centimetre produce the same
+pressure as N uncharged molecules.</p>
+
+<p>We know the value of D from Townsend&rsquo;s experiments and the
+values of u from those of Zeleny. We get the following values for
+Ne&times;10<span class="sp">-10</span>:&mdash;</p>
+
+<table class="ws" summary="Contents">
+<tr><td class="tccm allb" rowspan="2">Gas.</td>
+ <td class="tccm allb" colspan="2">Moist Gas.</td> <td class="tccm allb" colspan="2">Moist Gas.</td></tr>
+<tr><td class="tccm allb">Positive<br />Ions.</td> <td class="tccm allb">Positive<br />Ions.</td>
+ <td class="tccm allb">Positive<br />Ions.</td> <td class="tccm allb">Positive<br />Ions.</td></tr>
+
+<tr><td class="tcl lb rb">Air</td> <td class="tcc rb">1.28</td> <td class="tcc rb">1.29</td> <td class="tcc rb">1.46</td> <td class="tcc rb">1.31</td></tr>
+<tr><td class="tcl lb rb">Oxygen</td> <td class="tcc rb">1.34</td> <td class="tcc rb">1.27</td> <td class="tcc rb">1.63</td> <td class="tcc rb">1.36</td></tr>
+<tr><td class="tcl lb rb">Carbonic acid</td> <td class="tcc rb">1.01</td> <td class="tcc rb">&ensp;.87</td> <td class="tcc rb">&ensp;.99</td> <td class="tcc rb">&ensp;.93</td></tr>
+<tr><td class="tcl lb rb">Hydrogen</td> <td class="tcc rb">1.24</td> <td class="tcc rb">1.18</td> <td class="tcc rb">1.63</td> <td class="tcc rb">1.25</td></tr>
+<tr><td class="tcc lb bb">Mean</td> <td class="tcc allb">1.22</td> <td class="tcc allb">1.15</td> <td class="tcc allb">1.43</td> <td class="tcc allb">1.21</td></tr>
+</table>
+
+<p><span class="pagenum"><a name="page875" id="page875"></a>875</span></p>
+
+<p>Since 1.22 cubic centimetres of hydrogen at the temperature 15° C.
+and pressure 760 mm. of mercury are liberated by the passage
+through acidulated water of one electromagnetic unit of electricity
+or 3&times;10<span class="sp">10</span> electrostatic units, and since in one cubic centimetre of
+the gas there are 2.46 N atoms of hydrogen, we have, if E is the
+charge in electrostatic units, on the atom of hydrogen in the electrolysis
+of solutions</p>
+
+<p class="center1">2.46NE = 3&times;10<span class="sp">10</span>,</p>
+
+<p class="noind">or</p>
+
+<p class="center1">NE = 1.22&times;10<span class="sp">10</span>.</p>
+
+<p class="noind">The mean of the values of Ne in the preceding table is 1.24&times;10<span class="sp">10</span>.
+Hence we may conclude that the charge of electricity carried by a
+gaseous ion is equal to the charge carried by the hydrogen atom in
+the electrolysis of solutions. The values of Ne for the different gases
+differ more than we should have expected from the probable accuracy
+of the determination of D and the velocity of the ions: Townsend
+(<i>Proc. Roy. Soc.</i> 80, p. 207) has shown that when the ionization is
+produced by Röntgen rays some of the positive ions carry a double
+charge and that this accounts for the values of Ne being greater for
+the positive than for the negative ions. Since we know the value
+of e, viz. 3.5&times;10<span class="sp">-10</span>, and, also Ne, = 1.24&times;10<span class="sp">10</span>, we find N the number
+of molecules in a cubic centimetre of gas at standard temperature and
+pressure to be equal to 3.5&times;10<span class="sp">19</span>. This method of obtaining N is
+the only one which does not involve any assumption as to the shape
+of the molecules and the forces acting between them.</p>
+
+<p>Another method of determining the charge carried by an ion has
+been employed by Rutherford (<i>Proc. Roy. Soc.</i> 81, pp. 141, 162),
+in which the positively electrified particles emitted by radium are
+made use of. The method consists of: (1) Counting the number of
+&alpha; particles emitted by a given quantity of radium in a known time.
+(2) Measuring the electric charge emitted by this quantity in the
+same time. To count the number of the &alpha; particles the radium
+was so arranged that it shot into an ionization chamber a small
+number of &alpha; particles per minute; the interval between the emission
+of individual particles was several seconds. When an &alpha; particle
+passed into the vessel it ionized the gas inside and so greatly increased
+its conductivity; thus, if the gas were kept exposed to an electric
+field, the current through the gas would suddenly increase when an
+&alpha; particle passed into the vessel. Although each &alpha; particle produces
+about thirty thousand ions, this is hardly large enough to produce
+the conductivity appreciable without the use of very delicate
+apparatus; to increase the conductivity Rutherford took advantage
+of the fact that ions, especially negative ones, when exposed to a
+strong electric field, produce other ions by collision against the
+molecules of the gas through which they are moving. By suitably
+choosing the electric field and the pressure in the ionization chamber,
+the 30,000 ions produced by each &alpha; particle can be multiplied to
+such an extent that an appreciable current passes through the
+ionization chamber on the arrival of each &alpha; particle. An electrometer
+placed in series with this vessel will show by its deflection when an &alpha;
+particle enters the chamber, and by counting the number of deflections
+per minute we can determine the number of &alpha; particles given
+out by the radium in that time. Another method of counting this
+number is to let the particles fall on a phosphorescent screen, and
+count the number of scintillations on the screen in a certain time.
+Rutherford has shown that these two methods give concordant
+results.</p>
+
+<p>The charge of positive electricity given out by the radium was
+measured by catching the &alpha; particles in a Faraday cylinder placed
+in a very highly exhausted vessel, and measuring the charge per
+minute received by this cylinder. In this way Rutherford showed
+that the charge on the &alpha; particle was 9.4&times;10<span class="sp">-10</span> electrostatic units.
+Now e/m for the &alpha; particle = 5 &times;10<span class="sp">3</span>, and there is evidence that the
+&alpha; particle is a charged atom of helium; since the atomic weight of
+helium is 4 and e/m for hydrogen is 10<span class="sp">4</span>, it follows that the charge
+on the helium atom is twice that on the hydrogen, so that the charge
+on the hydrogen atom is 4.7&times;10<span class="sp">-10</span> electrostatic units.</p>
+</div>
+
+<p><i>Calculation of the Mass of the Ions at Low Pressures.</i>&mdash;Although
+at ordinary pressures the ion seems to have a very complex
+structure and to be the aggregate of many molecules, yet we have
+evidence that at very low pressures the structure of the ion, and
+especially of the negative one, becomes very much simpler.
+This evidence is afforded by determination of the mass of the
+atom. We can measure the ratio of the mass of an ion to the
+charge on the ion by observing the deflections produced by magnetic
+and electric forces on a moving ion. If an ion carrying a
+charge e is moving with a velocity v, at a point where the magnetic
+force is H, a mechanical force acts on the ion, whose direction
+is at right angles both to the direction of motion of the ion and
+to the magnetic force, and whose magnitude is evH sin &theta;, where
+&theta; is the angle between v and H. Suppose then that we have an
+ion moving through a gas whose pressure is so low that the free
+path of the ion is long compared with the distance through which
+it moves whilst we are experimenting upon it; in this case the
+motion of the ion will be free, and will not be affected by the
+presence of the gas.</p>
+
+<div class="condensed">
+<p>Since the force is always at right angles to the direction of motion
+of the ion, the speed of the ion will not be altered by the action
+of this force; and if the ion is projected with a velocity v in a direction
+at right angles to the magnetic force, and if the magnetic force is
+constant in magnitude and direction, the ion will describe a curve in
+a plane at right angles to the magnetic force. If &rho; is the radius of
+curvature of this curve, m the mass of the ion, mv²/&rho; must equal
+the normal force acting on the ion, <i>i.e.</i> it must be equal to Hev, or
+&rho; = mv/He. Thus the radius of curvature is constant; the path is
+therefore a circle, and if we can measure the radius of this circle we
+know the value of mv/He. In the case of the rapidly moving negative
+ions projected from the cathode in a highly exhausted tube, which
+are known as <i>cathode rays</i>, the path of the ions can be readily determined
+since they make many substances luminous when they
+impinge against them. Thus by putting a screen of such a substance
+in the path of the rays the shape of the path will be determined.
+Let us now suppose that the ion is acted upon by a vertical electric
+force X and is free from magnetic force, if it be projected with a
+horizontal velocity v, the vertical deflection y after a time t is ½&times;et²/m,
+or if l is the horizontal distance travelled over by the ion in this time
+we have since l = vt,</p>
+
+<table class="math0" summary="math">
+<tr> <td rowspan="2">y = ½</td> <td>Xe</td>
+ <td rowspan="2">&nbsp;</td> <td>l²</td> <td rowspan="2">.</td></tr>
+<tr> <td class="denom">m</td> <td class="denom">v²</td></tr></table>
+
+<p class="noind">Thus if we measure y and l we can deduce e/mv². From the effect
+of the magnetic force we know e/mv. Combining these results we
+can find both e/m and v.</p>
+
+<table class="nobctr" style="float: right; width: 370px;" summary="Illustration">
+<tr><td class="figright1"><img style="width:323px; height:88px" src="images/img875.jpg" alt="" /></td></tr>
+<tr><td class="caption sc">Fig. 13.</td></tr></table>
+
+<p>The method by which this determination is carried out in practice
+is illustrated in fig. 13. The cathode rays start from the electrode
+C in a highly exhausted tube, pass through two small holes in the
+plugs A and B, the holes being in the same horizontal line. Thus a
+pencil of rays emerging from B is horizontal and produces a bright
+spot at the far end of the
+tube. In the course of
+their journey to the end
+of the tube they pass
+between the horizontal
+plates E and D, by connecting
+these plates with
+an electric battery a vertical
+electric field is produced between E and D and the phosphorescent
+spot is deflected. By measuring this deflection we determine
+e/mv². The tube is now placed in a uniform magnetic field, the lines
+of magnetic force being horizontal and at right angles to the plane
+of the paper. The magnetic force makes the rays describe a circle in
+the plane of the paper, and by measuring the vertical deflection of
+the phosphorescent patch at the end of the tube we can determine
+the radius of this circle, and hence the value of e/mv. From the two
+observations the value of e/m and v can be calculated.</p>
+
+<p>Another method of finding e/m for the negative ion which is
+applicable in many cases to which the preceding one is not suitable,
+is as follows: Let us suppose that the ion starts from rest and moves
+in a field where the electric and magnetic forces are both uniform,
+the electric force X being parallel to the axis of x, and the magnetic
+force Z parallel to the axis of z; then if x, y, are the co-ordinates of
+the ion at the time t, the equations of motion of the ion are&mdash;</p>
+
+<table class="math0" summary="math">
+<tr> <td rowspan="2">m</td> <td>d²x</td>
+ <td rowspan="2">= Xe &minus; He</td> <td>dy</td> <td rowspan="2">,</td></tr>
+<tr> <td class="denom">dt²</td> <td class="denom">dt</td></tr></table>
+
+<table class="math0" summary="math">
+<tr> <td rowspan="2">m</td> <td>d²y</td>
+ <td rowspan="2">= He</td> <td>dx</td> <td rowspan="2">.</td></tr>
+<tr> <td class="denom">dt²</td> <td class="denom">dt</td></tr></table>
+
+<p>The solution of these equations, if x, y, dx/dt, dy/dt all vanish
+when t = 0, is</p>
+
+<table class="math0" summary="math">
+<tr><td rowspan="2">x =</td> <td>Xm</td>
+ <td rowspan="2"><span class="f250">{</span>1 &minus; cos<span class="f250">(</span></td> <td>e</td>
+ <td rowspan="2">Ht<span class="f250">)}</span></td></tr>
+<tr><td class="denom">eH²</td> <td class="denom">m</td></tr></table>
+
+<table class="math0" summary="math">
+<tr><td rowspan="2">y =</td> <td>Xm</td>
+ <td rowspan="2"><span class="f250">{</span></td> <td>e</td>
+ <td rowspan="2">Ht &minus; sin<span class="f250">(</span></td> <td>e</td>
+ <td rowspan="2">Ht<span class="f250">)}</span>.</td></tr>
+<tr><td class="denom">eH²</td> <td class="denom">m</td> <td class="denom">m</td></tr></table>
+
+<p class="noind">These equations show that the path of the ion is a cycloid, the
+generating circle of which has a diameter equal to 2Xm/eH², and
+rolls on the line x = 0.</p>
+
+<p>Suppose now that we have a number of ions starting from the
+plane x = 0, and moving towards the plane x = a. The particles
+starting from x = 0 describe cycloids, and the greatest distance they
+can get from the plane is equal to the diameter of the generating
+circle of the cycloid, <i>i.e.</i> to 2Xm/eH². (After reaching this distance
+they begin to approach the plane.) Hence if a is less than the
+diameter of the generating circle, all the particles starting from
+x = 0 will reach the plane x = a, if this is unlimited in extent; while
+if a is greater than the diameter of the generating circle none of
+the particles which start from x = 0 will reach the plane x = a. Thus,
+if x = 0 is a plane illuminated by ultra-violet light, and consequently
+the seat of a supply of negative ions, and x = a a plane connected
+with an electrometer, then if a definite electric intensity is established
+between the planes, <i>i.e.</i> if X be fixed, so that the rate of emission of
+negative ions from the illuminated plate is given, and if a is less than
+2Xm/eH², all the ions which start from x = 0 will reach x = a. That
+<span class="pagenum"><a name="page876" id="page876"></a>876</span>
+is, the rate at which this plane receives an electric charge will be the
+same whether there is a magnetic field between the plate or not,
+but if a is greater than 2Xm/eH², then no particle which starts from
+the plate x = 0 will reach the plate x = a, and this plate will receive
+no charge. Thus the supply of electricity to the plate has been entirely
+stopped by the magnetic field. Thus, on this theory, if the
+distance between the plates is less than a certain value, the magnetic
+force should produce no effect on the rate at which the electrometer
+plate receives a charge, while if the distance is greater than this value
+the magnetic force would completely stop the supply of electricity
+to the plate. The actual phenomena are not so abrupt as this theory
+indicates. We find that when the plates are very near together the
+magnetic force produces a very slight effect, and this an increase in
+the rate of charging of the plate. On increasing the distance we come
+to a stage where the magnetic force produces a great diminution in
+the rate of charging. It does not, however, stop it abruptly, there
+being a considerable range of distance, in which the magnetic force
+diminishes but does not destroy the current. At still greater distances
+the current to the plate under the magnetic force is quite
+inappreciable compared with that when there is no magnetic force.
+We should get this gradual instead of abrupt decay of the current
+if some of the particles, instead of all starting from rest, started
+with a finite velocity; in that case the first particles stopped would
+be those which started from rest. This would be when a = 2Xm/eH².
+Thus if we measure the value of a when the magnetic force first
+begins to affect the leak to the electrometer we determine 2Xm/eH²,
+and as we can easily measure X and H, we can deduce the value of m/e.</p>
+</div>
+
+<p>By these methods Thomson determined the value of e/m for
+the negative ions produced when ultra-violet light falls on a
+metal plate, as well as for the negative ions produced by an
+incandescent carbon filament in an atmosphere of hydrogen
+(<i>Phil. Mag.</i> [5], 48, p. 547) as well as for the cathode rays. It was
+found that the value of e/m for the negative ions was the same
+in all these cases, and that it was a constant quantity independent
+of the nature of the gas from which the ions are produced and the
+means used to produce them. It was found, too, that this value
+was more than a thousand times the value of e/M, where e is the
+charge carried by an atom of hydrogen in the electrolysis of
+solutions, and M the mass of an atom of hydrogen. We have
+seen that this charge is the same as that carried by the negative
+ion in gases; thus since e/m is more than a thousand times e/M,
+it follows that M must be more than a thousand times m. Thus
+the mass of the negative ion is exceedingly small compared with
+the mass of the atom of hydrogen, the smallest mass recognized in
+chemistry. The production of negative ions thus involves the
+splitting up of the atom, as from a collection of atoms something
+is detached whose mass is less than that of a single atom. It is
+important to notice in connexion with this subject that an entirely
+different line of argument, based on the Zeeman effect (see <span class="sc"><a href="#artlinks">Magneto-Optics</a></span>),
+leads to the recognition of negatively electrified
+particles for which e/m is of the same order as that deduced from
+the consideration of purely electrical phenomena. These small
+negatively electrified particles are called corpuscles. The latest
+determinations of e/m for corpuscles available are the following:&mdash;</p>
+
+<table class="nobctr" style="width: 70%;" summary="Contents">
+
+<tr><td class="tcc">Observer.</td> <td class="tcc">e/m.</td></tr>
+
+<tr><td class="tcl">Classen (<i>Ber. deut. phys. Ges.</i> 6, p. 700)</td> <td class="tcl">1.7728&times;10<span class="sp">7</span></td></tr>
+<tr><td class="tcl">Bucherer (<i>Ann. der Phys.</i>, 28, p. 513)</td> <td class="tcl">1.763&times;10<span class="sp">7</span></td></tr>
+</table>
+
+<p>It follows from electrical theory that when the corpuscles
+are moving with a velocity comparable with that of light their
+masses increase rapidly with their velocity. This effect has been
+detected by Kauffmann (<i>Gött. Nach.</i>, Nov. 8, 1901), who used the
+corpuscles shot out from radium, some of which move with
+velocities only a few per cent less than that of light. Other
+experiments on this point have been made by Bucherer (<i>Ann. der
+Phys.</i> 28, p. 513).</p>
+
+<p><i>Conductivity Produced by Ultra-Violet Light.</i>&mdash;So much use has
+been made in recent times of ultra-violet light for producing
+ions that it is desirable to give some account of the electrical
+effects produced by light. The discovery by Hertz (<i>Wied. Ann.</i>
+31, p. 983) in 1887, that the incidence of ultra-violet light on a
+spark gap facilitates the passage of a spark, led to a series of
+investigations by Hallwachs, Hoor, Righi and Stoletow, on the
+effect of ultra-violet light on electrified bodies. These researches
+have shown that a freshly cleaned metal surface, charged with
+negative electricity, rapidly loses its charge, however small, when
+exposed to ultra-violet light, and that if the surface is insulated
+and without charge initially, it acquires a positive charge under
+the influence of the light. The magnitude of this positive charge
+may be very much increased by directing a blast of air on the plate.
+This, as Zeleny (<i>Phil. Mag.</i> [5], 45, p. 272) showed, has the effect
+of blowing from the neighbourhood of the plate negatively
+electrified gas, which has similar properties to the charged gas
+obtained by the separation of ions from a gas exposed to Röntgen
+rays or uranium radiation. If the metal plate is positively
+electrified, there is no loss of electrification caused by ultra-violet
+light. This has been questioned, but a very careful examination
+of the question by Elster and Geitel (<i>Wied. Ann.</i> 57, p. 24) has
+shown that the apparent exceptions are due to the accidental
+exposure to reflected ultra-violet light of metal surfaces in the
+neighbourhood of the plate negatively electrified by induction,
+so that the apparent loss of charge is due to negative electricity
+coming up to the plate, and not to positive electricity going away
+from it. The ultra-violet light may be obtained from an arc-lamp,
+the effectiveness of which is increased if one of the terminals
+is made of zinc or aluminium, the light from these substances
+being very rich in ultra-violet rays; it may also be got very
+conveniently by sparking with an induction coil between zinc
+or cadmium terminals. Sunlight is not rich in ultra-violet light,
+and does not produce anything like so great an effect as the arc
+light. Elster and Geitel, who have investigated with great success
+the effects of light on electrified bodies, have shown that the more
+electro-positive metals lose negative charges when exposed to
+ordinary light, and do not need the presence of the ultra-violet
+rays. Thus they found that amalgams of sodium or potassium
+enclosed in a glass vessel lose a negative charge when exposed to
+daylight, though the glass stops the small amount of ultra-violet
+light left in sunlight after its passage through the atmosphere.
+If sodium or potassium be employed, or, what is more convenient,
+the mercury-like liquid obtained by mixing sodium and potassium
+in the proportion of their combining weights, they found that
+negative electricity was discharged by an ordinary petroleum
+lamp. If the still more electro-positive metal rubidium is used,
+the discharge can be produced by the light from a glass rod just
+heated to redness; but there is no discharge till the glass is luminous.
+Elster and Geitel arrange the metals in the following order
+for the facility with which negative electrification is discharged
+by light: rubidium, potassium, alloy of sodium and potassium,
+sodium, lithium, magnesium, thallium, zinc. With copper,
+platinum, lead, iron, cadmium, carbon and mercury the effects
+with ordinary light are too small to be appreciable. The order
+is the same as that in Volta&rsquo;s electro-chemical series. With
+ultra-violet light the different metals show much smaller differences
+in their power of discharging negative electricity than they
+do with ordinary light. Elster and Geitel found that the ratio of
+the photo-electric effects of two metals exposed to approximately
+monochromatic light depended upon the wave-length of the light,
+different metals showing a maximum sensitiveness in different
+parts of the spectrum. This is shown by the following table for
+the alkaline metals. The numbers in the table are the rates of
+emission of negative electricity under similar circumstances. The
+rate of emission under the light from a petroleum lamp was
+taken as unity:&mdash;</p>
+
+<table class="nobctr" style="width: 60%;" summary="Contents">
+<tr><td class="tcl">&nbsp;</td> <td class="tcc">Blue.</td> <td class="tcc">Yellow.</td> <td class="tcc">Orange.</td> <td class="tcc">Red.</td></tr>
+<tr><td class="tcl">Rb</td> <td class="tcc">.16</td> <td class="tcc">.64</td> <td class="tcc">.33</td> <td class="tcc">.039</td></tr>
+<tr><td class="tcl">Na</td> <td class="tcc">.37</td> <td class="tcc">.36</td> <td class="tcc">.14</td> <td class="tcc">.009</td></tr>
+<tr><td class="tcl">K</td> <td class="tcc">.57</td> <td class="tcc">.07</td> <td class="tcc">.04</td> <td class="tcc">.002</td></tr>
+</table>
+
+<p class="noind">The table shows that the absorption of light by the metal has
+great influence on the photo-electric effect, for while potassium
+is more sensitive in blue light than sodium, the strong absorption
+of yellow light by sodium makes it more than five times more
+sensitive to this light than potassium. Stoletow, at an early
+period, called attention to the connexion between strong absorption
+and photo-electric effects. He showed that water, which
+does not absorb to any great extent either the ultra-violet or
+visible rays, does not show any photo-electric effect, while
+strongly coloured solutions, and especially solutions of fluorescent
+substances such as methyl green or methyl violet, do so to a very
+considerable extent; indeed, a solution of methyl green is more
+sensitive than zinc. Hallwachs (<i>Wied. Ann.</i> 37, p. 666) proved
+<span class="pagenum"><a name="page877" id="page877"></a>877</span>
+that in liquids showing photo-electric effects there is always strong
+absorption; we may, however, have absorption without these
+effects. Phosphorescent substances, such as calcium sulphide
+show this effect, as also do various specimens of fluor-spar. As
+phosphorescence and fluorescence are probably accompanied by
+a very intense absorption by the surface layers, the evidence is
+strong that to get the photo-electric effects we must have strong
+absorption of some kind of light, either visible or ultra-violet.</p>
+
+<table class="nobctr" style="float: left; width: 356px;" summary="Illustration">
+<tr><td class="figleft1"><img style="width:306px; height:466px" src="images/img877.jpg" alt="" /></td></tr>
+<tr><td class="caption sc">Fig. 14.</td></tr></table>
+
+<p>If a conductor A is placed near a conductor B exposed to ultra-violet
+light, and if B is made the negative electrode and a difference
+of potential established between A and B, a current of
+electricity will flow between the conductors. The relation between
+the magnitude of the current and the difference of potential
+when A and B are parallel
+plates has been investigated
+by Stoletow (<i>Journal
+de physique</i>, 1890, 11,
+p. 469), von Schweidler
+(<i>Wien. Ber.</i>, 1899, 108, p.
+273) and Varley (<i>Phil.
+Trans. A.</i>, 1904, 202, p.
+439). The results of some
+of Varley&rsquo;s experiments are
+represented in the curves
+shown in fig. 14, in which
+the ordinates are the currents
+and the abscissae the
+potentials. It will be seen
+that when the pressure is
+exceedingly low the current
+is independent of the
+potential difference and
+is equal to the negative
+charge carried off in unit
+time by the corpuscles
+emitted from the surface
+exposed to the light. At
+higher pressures the current
+rises far above these values and increases rapidly with the
+potential difference. This is due to the corpuscles emitted by the
+illuminated surface acquiring under the electric field such high
+velocities that when they strike against the molecules of the gas
+through which they are passing they ionize them, producing fresh
+ions which can carry on additional current. The relation between
+the current and the potential difference in this case is in accordance
+with the results of the theory of ionization by collision.
+The corpuscles emitted from a body under the action of ultra-violet
+light start from the surface with a finite velocity. The
+velocity is not the same for all the corpuscles, nor indeed could
+we expect that it should be: for as Ladenburg has shown
+(<i>Ann. der Phys.</i>, 1903, 12, p. 558) the seat of their emission is not
+confined to the surface layer of the illuminated metal but extends
+to a layer of finite, though small, thickness. Thus the particles
+which start deep down will have to force their way through a
+layer of metal before they reach the surface, and in doing so will
+have their velocities retarded by an amount depending on the
+thickness of this layer. The variation in the velocity of the
+corpuscles is shown in the following table, due to Lenard (<i>Ann.
+der Phys.</i>, 1902, 8, p. 149).</p>
+
+<table class="ws" summary="Contents">
+<tr><td class="allb">&nbsp;</td> <td class="tcc allb">Carbon.</td>
+ <td class="tcc allb">Platinum.</td> <td class="tcc allb">Aluminium.</td></tr>
+
+<tr><td class="tcl lb rb">Corpuscles emitted with velocities</td> <td class="tcc rb">&nbsp;</td> <td class="tcc rb">&nbsp;</td> <td class="tcc rb">&nbsp;</td></tr>
+<tr><td class="tcl lb rb">&emsp;between 12 and 8&times;10<span class="sp">7</span> cm sec.</td> <td class="tcc rb">0.000</td> <td class="tcc rb">0.000</td> <td class="tcc rb">0.004</td></tr>
+<tr><td class="tcl lb rb">&emsp;between 8 and 4&times;10<span class="sp">7</span> cm sec.</td> <td class="tcc rb">0.049</td> <td class="tcc rb">0.155</td> <td class="tcc rb">0.151</td></tr>
+<tr><td class="tcl lb rb">&emsp;between 4 and 0&times;10<span class="sp">7</span> cm sec.</td> <td class="tcc rb">0.67&ensp;</td> <td class="tcc rb">0.65&ensp;</td> <td class="tcc rb">0.49&ensp;</td></tr>
+<tr><td class="tcl lb rb">Corpuscles only emitted with the help</td> <td class="tcc rb">&nbsp;</td> <td class="tcc rb">&nbsp;</td> <td class="tcc rb">&nbsp;</td></tr>
+<tr><td class="tcl lb rb">&emsp;of an external electric field.</td> <td class="tcc rb">0.28&ensp;</td> <td class="tcc rb">0.21&ensp;</td> <td class="tcc rb">0.35&ensp;</td></tr>
+<tr><td class="tcl lb rb bb">&nbsp;</td> <td class="tcc allb">1.00&ensp;</td> <td class="tcc allb">1.00&ensp;</td> <td class="tcc allb">1.00&ensp;</td></tr>
+</table>
+
+<p class="noind">If the illuminated surface is completely surrounded by an envelope
+of the same metal insulated from and completely shielded from
+the light, the emission of the negative corpuscles from the illuminated
+surface would go on until the potential difference V
+between this surface and the envelope became so great that the
+corpuscles with the greatest velocity lost their energy before
+reaching the envelope, <i>i.e.</i> if m is the mass, e the charge on a
+corpuscle, v the greatest velocity of projection, until Ve = ½mv².
+The values found for V by different observers are not very
+consistent. Lenard found that V for aluminium was about 3
+volts and for platinum 2. Millikan and Winchester (<i>Phil.
+Mag.</i>, July 1907) found for aluminium V = .738. The apparatus
+used by them was so complex that the interpretation of their
+results is difficult.</p>
+
+<p>An extremely interesting fact discovered by Lenard is that the
+velocity with which the corpuscles are emitted from the metal is
+independent of the intensity of the incident light. The quantity
+of corpuscles increases with the intensity, but the velocity of the
+individual corpuscles does not. It is worthy of notice that in
+other cases when negative corpuscles are emitted from metals,
+as for example when the metals are exposed to cathode rays,
+Canal-strahlen, or Röntgen rays, the velocity of the emitted
+corpuscles is independent of the intensity of the primary radiation
+which excites them. The velocity is not, however, independent
+of the nature of the primary rays. Thus when light is used
+to produce the emission of corpuscles the velocity, as Ladenburg
+has shown, depends on the wave length of the light, increasing
+as the wave length diminishes. The velocity of corpuscles
+emitted under the action of cathode rays is greater than that
+of those ejected by light, while the incidence of Röntgen rays
+produces the emission of corpuscles moving much more rapidly
+than those in the cases already mentioned, and the harder the
+primary rays the greater is the velocity of the corpuscles.</p>
+
+<p>The importance of the fact that the velocity and therefore the
+energy of the corpuscles emitted from the metal is independent
+of the intensity of the incident light can hardly be overestimated.
+It raises the most fundamental questions as to the nature of light
+and the constitution of the molecules. What is the source of
+the energy possessed by these corpuscles? Is it the light, or in the
+stores of internal energy possessed by the molecule? Let us
+follow the consequences of supposing that the energy comes from
+the light. Then, since the energy is independent of the intensity
+of the light, the electric forces which liberate the corpuscles must
+also be independent of that intensity. But this cannot be the
+case if, as is usually assumed in the electromagnetic theory, the
+wave front consists of a uniform distribution of electric force
+without structure, for in this case the magnitude of the electric
+force is proportional to the square root of the intensity. On the
+emission theory of light a difficulty of this kind would not arise,
+for on that theory the energy in a luminiferous particle remains
+constant as the particle pursues its flight through space. Thus any
+process which a single particle is able to effect by virtue of its
+energy will be done just as well a thousand miles away from the
+source of light as at the source itself, though of course in a given
+space there will not be nearly so many particles to do this process
+far from the source as there are close in. Thus, if one of the
+particles when it struck against a piece of metal caused the
+ejection of a corpuscle with a given velocity, the velocity of
+emission would not depend on the intensity of the light. There
+does not seem any reason for believing that the electromagnetic
+theory is inconsistent with the idea that on this theory, as on the
+emission theory, the energy in the light wave may instead of being
+uniformly distributed through space be concentrated in bundles
+which occupy only a small fraction of the volume traversed by
+the light, and that as the wave travels out the bundles get farther
+apart, the energy in each remaining undiminished. Some such
+view of the structure of light seems to be required to account for
+the fact that when a plate of metal is struck by a wave of ultra-violet
+light, it would take years before the corpuscles emitted
+from the metal would equal in number the molecules on the
+surface of the metal plate, and yet on the ordinary theory of light
+each one of these is without interruption exposed to the action of
+<span class="pagenum"><a name="page878" id="page878"></a>878</span>
+the light. The fact discovered by E. Ladenburg (<i>Verh. d.
+deutsch. physik. Ges.</i> 9, p. 504) that the velocity with which
+the corpuscles are emitted depends on the wave length of the
+light suggests that the energy in each bundle depends upon the
+wave length and increases as the wave length diminishes.</p>
+
+<p>These considerations illustrate the evidence afforded by photo-electric
+effects on the nature of light; these effects may also
+have a deep significance with regard to the structure of matter.
+The fact that the energy of the individual corpuscles is independent
+of the intensity of the light might be explained by the
+hypothesis that the energy of the corpuscles does not come from
+the light but from the energy stored up in the molecules of the
+metal exposed to the light. We may suppose that under the
+action of the light some of the molecules are thrown into an
+unstable state and explode, ejecting corpuscles; the light in this
+case acts only as a trigger to liberate the energy in the atom, and
+it is this energy and not that of the light which goes into the
+corpuscles. In this way the velocity of the corpuscles would be
+independent of the intensity of the light. But it may be asked,
+is this view consistent with the result obtained by Ladenburg
+that the velocity of the corpuscles depends upon the nature of
+the light? If light of a definite wave length expelled corpuscles
+with a definite and uniform velocity, it would be very improbable
+that the emission of the corpuscles is due to an explosion of the
+atoms. The experimental facts as far as they are known at
+present do not allow us to say that the connexion between the
+velocity of the corpuscles and the wave length of the light is of
+this definite character, and a connexion such as a gradual increase
+of average velocity as the wave length of the light diminishes,
+would be quite consistent with the view that the corpuscles are
+ejected by the explosion of the atom. For in a complex thing like
+an atom there may be more than one system which becomes unstable
+when exposed to light. Let us suppose that there are
+two such systems, A and B, of which B ejects the corpuscles with
+the greater velocity. If B is more sensitive to the short waves,
+and A to the long ones, then as the wave length of the light
+diminishes the proportion of the corpuscles which come from B
+will increase, and as these are the faster, the average velocity of
+the corpuscles emitted will also increase. And although the
+potential acquired by a perfectly insulated piece of metal when
+exposed to ultra-violet light would depend only on the velocity
+of the fastest corpuscles and not upon their number, in practice
+perfect insulation is unattainable, and the potential actually
+acquired is determined by the condition that the gain of negative
+electricity by the metal through lack of insulation, is equal to the
+loss by the emission of negatively electrified corpuscles. The
+potential acquired will fall below that corresponding to perfect
+insulation by an amount depending on the number of the faster
+corpuscles emitted, and the potential will rise if the proportion of
+the rapidly moving corpuscles is increased, even though there is
+no increase in their velocity. It is interesting to compare other
+cases in which corpuscles are emitted with the case of ultra-violet
+light. When a metal or gas is bombarded by cathode rays it
+emits corpuscles and the velocity of these is found to be independent
+of the velocity of the cathode rays which excite them; the
+velocity is greater than for corpuscles emitted under ultra-violet
+light. Again, when bodies are exposed to Röntgen rays they emit
+corpuscles moving with a much greater velocity than those
+excited by cathode rays, but again the velocity does not depend
+upon the intensity of the rays although it does to some extent
+on their hardness. In the case of cathode and Röntgen rays, the
+velocity with which the corpuscles are emitted seems, as far as we
+know at present, to vary slightly, but only slightly, with the
+nature of the substance on which the rays fall. May not this
+indicate that the first effect of the primary rays is to detach a
+neutral doublet, consisting of a positive and negative charge,
+this doublet being the same from whatever system it is detached?
+And that the doublet is unstable and explodes, expelling the
+negative charge with a high velocity, and the positive one,
+having a much larger charge, with a much smaller velocity,
+the momentum of the negative charge being equal to that of the
+positive.</p>
+
+<p>Up to now we have been considering the effects produced when
+light is incident on metals. Lenard found (and the result has
+been confirmed by the experiments of J. J. Thomson and
+Lyman) that certain kinds of ultra-violet light ionize a gas
+when they pass through. The type of ultra-violet light
+which produces this effect is so easily absorbed that it is
+stopped by a layer a few millimetres thick of air at atmospheric
+pressure.</p>
+
+<p><i>Ionization by Collision.</i>&mdash;When the ionization of the gas is
+produced by external agents such as Röntgen rays or ultra-violet
+light, the electric field produces a current by setting the
+positive ions moving in one direction, and the negative ones in the
+opposite; it makes use of ions already made and does not itself
+give rise to ionization. In many cases, however, such as in
+electric sparks, there are no external agents to produce ionization
+and the electric field has to produce the ions as well as set them in
+motion. When the ionization is produced by external means the
+smallest electric field is able to produce a current through the
+gas; when, however, these external means are absent no current
+is produced unless the strength of the electric field exceeds a
+certain critical value, which depends not merely upon the nature
+of the gas but also upon the pressure and the dimensions of
+the vessel in which it is contained. The variation of the electric
+field required to produce discharge can be completely explained
+if we suppose that the ionization of the gas is produced by the
+impact with its molecules of corpuscles, and in certain cases of
+positive ions, which under the influence of the electric field
+have acquired considerable kinetic energy. We have direct
+evidence that rapidly moving corpuscles are able to ionize
+molecules against which they strike, for the cathode rays consist
+of such corpuscles, and these when they pass through a gas
+produce large amounts of ionization. Suppose then that we
+have in a gas exposed to an electric field a few corpuscles. These
+will be set in motion by the field and will acquire an amount
+of energy in proportion to the product of the electric force,
+their charge, and the distance travelled in the direction of the
+electric field between two collisions with the molecules of the
+gas. If this energy is sufficient to give them the ionizing property
+possessed by cathode rays, then when a corpuscle strikes against
+a molecule it will detach another corpuscle; this under the action
+of the electric field will acquire enough energy to produce
+corpuscles on its own account, and so as the corpuscles move
+through the gas their number will increase in geometrical progression.
+Thus, though there were but few corpuscles to begin
+with, there may be great ionization after these have been
+driven some distance through the gas by the electric field.</p>
+
+<div class="condensed">
+<p>The number of ions produced by collisions can be calculated by
+the following method. Let the electric force be parallel to the axis
+of x, and let n be the number of corpuscles per unit volume at a place
+fixed by the co-ordinate x; then in unit time these corpuscles will
+make nu/&lambda; collisions with the molecules, if u is the velocity of a
+corpuscle and &lambda; the mean free path of a corpuscle. When the
+corpuscles are moving fast enough to produce ions by collision their
+velocities are very much greater than those they would possess at
+the same temperature if they were not acted on by electrical force,
+and so we may regard the velocities as being parallel to the axis of x
+and determined by the electric force and the mean free path of the
+corpuscles. We have to consider how many of the nu/&lambda; collisions
+which take place per second will produce ions. We should expect
+that the ionization of a molecule would require a certain amount of
+energy, so that if the energy of the corpuscle fell below this amount
+no ionization would take place, while if the energy of the corpuscle
+were exceedingly large, every collision would result in ionization.
+We shall suppose that a certain fraction of the number of collisions
+result in ionization and that this fraction is a function of the energy
+possessed by the corpuscle when it collides against the molecules.
+This energy is proportional to Xe&lambda; when X is the electric force,
+e the charge on the corpuscle, and &lambda; the mean free path. If the
+fraction of collisions which produce ionization is &int; (Xe&lambda;), then
+the number of ions produced per cubic centimetre per second is
+&int; (Xe&lambda;)nu/&lambda;. If the collisions follow each other with great rapidity
+so that a molecule has not had time to recover from one collision
+before it is struck again, the effect of collisions might be cumulative,
+so that a succession of collisions might give rise to ionization, though
+none of the collisions would produce an ion by itself. In this case &int;
+would involve the frequency of the collisions as well as the energy
+of the corpuscle; in other words, it might depend on the current
+through the gas as well as upon the intensity of the electric field.
+<span class="pagenum"><a name="page879" id="page879"></a>879</span>
+We shall, however, to begin with, assume that the current is so small
+that this cumulative effect may be neglected.</p>
+
+<p>Let us now consider the rate of increase, dn/dt, in the number of
+corpuscles per unit volume. In consequence of the collisions,
+&int; (Xe&lambda;)nu/&lambda; corpuscles are produced per second; in consequence
+of the motion of the corpuscles, the number which leave unit volume
+per second is greater than those which enter it by d/dx · (nu); while in a
+certain number of collisions a corpuscle will stick to the molecule and
+will thus cease to be a free corpuscle. Let the fraction of the number
+of collisions in which this occurs be &beta;. Thus the gain in the number
+of corpuscles is &int; (Xe&lambda;)nu/&lambda;, while the loss is d/dx·(nu) + &beta;·nu/&lambda; hence</p>
+
+<table class="math0" summary="math">
+<tr><td>dn</td> <td rowspan="2"><span class="f150">&int;</span>(Xe&lambda;)</td> <td>nu</td>
+ <td rowspan="2">&minus;</td> <td>d</td>
+ <td rowspan="2">(nu) &minus;</td> <td>&beta;nu</td> <td rowspan="2">.</td></tr>
+<tr><td class="denom">dt</td> <td class="denom">&lambda;</td>
+ <td class="denom">dx</td> <td class="denom">&lambda;</td></tr></table>
+
+<p class="noind">When things are in a steady state dn/dt = 0, and we have</p>
+
+<table class="math0" summary="math">
+<tr><td>d</td> <td rowspan="2">(nu) =</td> <td>1</td>
+ <td rowspan="2">(<span class="f150">&int;</span>(Xe&lambda;) &minus; &beta;) nu.</td></tr>
+<tr><td class="denom">dx</td> <td class="denom">&lambda;</td></tr></table>
+
+<p class="noind">If the current is so small that the electrical charges in the gas are
+not able to produce any appreciable variations in the field, X will be
+constant and we get nu = C&epsilon;<span class="sp">&alpha;x</span>, where &alpha; = {&int; (Xe&lambda;) &minus; &beta;}/&lambda;. If we take
+the origin from which we measure x at the cathode, C is the value
+of nu at the cathode, <i>i.e.</i> it is the number of corpuscles emitted per
+unit area of the cathode per unit time; this is equal to i<span class="su">0</span>/e if i<span class="su">0</span> is
+the quantity of negative electricity coming from unit area of the
+cathode per second, and e the electric charge carried by a corpuscle.
+Hence we have nue = i<span class="su">0</span>&epsilon;<span class="sp">&alpha;x</span>. If l is the distance between the anode
+and the cathode, the value of nue, when x = l, is the current passing
+through unit area of the gas, if we neglect the electricity carried by
+negatively electrified carriers other than corpuscles. Hence i = i<span class="su">0</span>&epsilon;<span class="sp">&alpha; l</span>.
+Thus the current between the plates increases in geometrical
+progression with the distance between the plates.</p>
+
+<p>By measuring the variation of the current as the distance between
+the plates is increased, Townsend, to whom we owe much of our
+knowledge on this subject, determined the values of &alpha; for different
+values of X and for different pressures for air, hydrogen and carbonic
+acid gas (<i>Phil. Mag.</i> [6], 1, p. 198). Since &lambda; varies inversely as the
+pressure, we see that &alpha; may be written in the form p&phi;(X/p) or
+&alpha;/X = F(X/p). The following are some of the values of &alpha; found by
+Townsend for air.</p>
+
+<table class="ws" summary="Contents">
+<tr><td class="tcc allb">X Volts<br />per cm.</td> <td class="tcc allb">Pressure<br />.17 mm.</td>
+ <td class="tcc allb">Pressure<br />.38 mm.</td> <td class="tcc allb">Pressure<br />1.10 mm.</td>
+ <td class="tcc allb">Pressure<br />2.1 mm.</td> <td class="tcc allb">Pressure<br />4.1 mm.</td></tr>
+
+<tr><td class="tcc lb rb">&ensp;20</td> <td class="tcc rb">&ensp;.24</td> <td class="tcc rb">&nbsp;</td> <td class="tcc rb">&nbsp;</td> <td class="tcc rb">&nbsp;</td> <td class="tcc rb">&nbsp;</td></tr>
+<tr><td class="tcc lb rb">&ensp;40</td> <td class="tcc rb">&ensp;.65</td> <td class="tcc rb">&ensp;.34</td> <td class="tcc rb">&nbsp;</td> <td class="tcc rb">&nbsp;</td> <td class="tcc rb">&nbsp;</td></tr>
+<tr><td class="tcc lb rb">&ensp;80</td> <td class="tcc rb">1.35</td> <td class="tcc rb">1.3&ensp;</td> <td class="tcc rb">&emsp;.45</td> <td class="tcc rb">&ensp;.13</td> <td class="tcc rb">&nbsp;</td></tr>
+<tr><td class="tcc lb rb">120</td> <td class="tcc rb">1.8&ensp;</td> <td class="tcc rb">2.0&ensp;</td> <td class="tcc rb">&ensp;1.1&ensp;</td> <td class="tcc rb">&emsp;.42</td> <td class="tcc rb">&ensp;.13</td></tr>
+<tr><td class="tcc lb rb">160</td> <td class="tcc rb">2.1&ensp;</td> <td class="tcc rb">2.8&ensp;</td> <td class="tcc rb">&ensp;2.0&ensp;</td> <td class="tcc rb">&emsp;.9&ensp;</td> <td class="tcc rb">&ensp;.28</td></tr>
+<tr><td class="tcc lb rb">200</td> <td class="tcc rb">&nbsp;</td> <td class="tcc rb">3.4&ensp;</td> <td class="tcc rb">&ensp;2.8&ensp;</td> <td class="tcc rb">&ensp;1.6&ensp;</td> <td class="tcc rb">&ensp;.5&ensp;</td></tr>
+<tr><td class="tcc lb rb">240</td> <td class="tcc rb">2.45</td> <td class="tcc rb">3.8&ensp;</td> <td class="tcc rb">&ensp;4.0&ensp;</td> <td class="tcc rb">&ensp;2.35</td> <td class="tcc rb">&ensp;.99</td></tr>
+<tr><td class="tcc lb rb">320</td> <td class="tcc rb">2.7&ensp;</td> <td class="tcc rb">4.5&ensp;</td> <td class="tcc rb">&ensp;5.5&ensp;</td> <td class="tcc rb">&ensp;4.0&ensp;</td> <td class="tcc rb">2.1 &ensp;</td></tr>
+<tr><td class="tcc lb rb">400</td> <td class="tcc rb">&nbsp;</td> <td class="tcc rb">5.0&ensp;</td> <td class="tcc rb">&ensp;6.8&ensp;</td> <td class="tcc rb">&ensp;6.0&ensp;</td> <td class="tcc rb">3.6&ensp;</td></tr>
+<tr><td class="tcc lb rb">480</td> <td class="tcc rb">3.15</td> <td class="tcc rb">5.4&ensp;</td> <td class="tcc rb">&ensp;8.0&ensp;</td> <td class="tcc rb">&ensp;7.8&ensp;</td> <td class="tcc rb">5.3&ensp;</td></tr>
+<tr><td class="tcc lb rb">560</td> <td class="tcc rb">&nbsp;</td> <td class="tcc rb">5.8&ensp;</td> <td class="tcc rb">&ensp;9.3&ensp;</td> <td class="tcc rb">&ensp;9.4&ensp;</td> <td class="tcc rb">7.1&ensp;</td></tr>
+<tr><td class="tcc lb rb bb">640</td> <td class="tcc rb bb">3.25</td> <td class="tcc rb bb">6.2&ensp;</td> <td class="tcc rb bb">10.6&ensp;</td> <td class="tcc rb bb">10.8&ensp;</td> <td class="tcc rb bb">8.9&ensp;</td></tr>
+</table>
+
+<p class="noind">We see from this table that for a given value of X, &alpha; for small pressures
+increases as the pressure increases; it attains a maximum at a
+particular pressure, and then diminishes as the pressure increases.
+The increase in the pressure increases the number of collisions, but
+diminishes the energy acquired by the corpuscle in the electric
+field, and thus diminishes the change of any one collision resulting
+in ionization. If we suppose the field is so strong that at some
+particular pressure the energy acquired by the corpuscle is well
+above the value required to ionize at each collision, then it is evident
+that increasing the number of collisions will increase the amount
+of ionization, and therefore &alpha;, and &alpha; cannot begin to diminish until
+the pressure has increased to such an extent that the mean free
+path of a corpuscle is so small that the energy acquired by the
+corpuscle from the electric field falls below the value when each
+collision results in ionization.</p>
+
+<p>The value of p, when X is given, for which &alpha; is a maximum, is
+proportional to X; this follows at once from the fact that &alpha; is of the
+form X·F(X/p). The value of X/p for which F(X/p) is a maximum
+is seen from the preceding table to be about 420, when X is expressed
+in volts per centimetre and p in millimetres of mercury. The
+maximum value of F(X/p) is about <span class="spp">1</span>&frasl;<span class="suu">60</span>. Since the current passing
+between two planes at a distance l apart is i<span class="su">0</span>&epsilon;<span class="sp">&alpha;l</span> or i<span class="su">0</span>&epsilon;<span class="sp">XlF(X/p)</span>,
+and since the force between the plates is supposed to be uniform,
+Xl is equal to V, the potential between the plates; hence the
+current between the plates is i<span class="su">0</span>&epsilon;<span class="sp">V·F(X/p)</span>, and the greatest value
+it can have is i<span class="su">0</span>&epsilon;<span class="sp">V/60</span>. Thus the ratio between the current between
+the plates when there is ionization and when there is none cannot
+be greater than &epsilon;<span class="sp">V/60</span>, when V is measured in volts. This result is
+based on Townsend&rsquo;s experiments with very weak currents; we
+must remember, however, that when the collisions are so frequent
+that the effects of collisions can accumulate, &alpha; may have much larger
+values than when the current is small. In some experiments made
+by J. J. Thomson with intense currents from cathodes covered
+with hot lime, the increase in the current when the potential difference
+was 60 volts, instead of being e times the current when there was no
+ionization, as the preceding theory indicates, was several hundred
+times that value, thus indicating a great increase in &alpha; with the
+strength of the current.</p>
+
+<p>Townsend has shown that we can deduce from the values of &alpha; the
+mean free path of a corpuscle. For if the ionization is due to the
+collisions with the corpuscles, then unless one collision detaches
+more than one corpuscle the maximum number of corpuscles produced
+will be equal to the number of collisions. When each collision
+results in the production of a corpuscle, &alpha; = 1/&lambda; and is independent
+of the strength of the electric field. Hence we see that the value of
+&alpha;, when it is independent of the electric field, is equal to the reciprocal
+of the free path. Thus from the table we infer that at a pressure
+of 17 mm. the mean free path is <span class="spp">1</span>&frasl;<span class="suu">325</span> cm.; hence at 1 mm. the mean
+free path of a corpuscle is <span class="spp">1</span>&frasl;<span class="suu">19</span> cm. Townsend has shown that this
+value of the mean free path agrees well with the value <span class="spp">1</span>&frasl;<span class="suu">21</span> cm.
+deduced from the kinetic theory of gases for a corpuscle moving
+through air. By measuring the values of &alpha; for hydrogen and carbonic
+acid gas Townsend and Kirby (<i>Phil. Mag.</i> [6], 1, p. 630) showed
+that the mean free paths for corpuscles in these gases are respectively
+<span class="spp">1</span>&frasl;<span class="suu">11.5</span> and <span class="spp">1</span>&frasl;<span class="suu">29</span> cm. at a pressure of 1 mm. These results again agree
+well with the values given by the kinetic theory of gases.</p>
+
+<p>If the number of positive ions per unit volume is m and v is the
+velocity, we have nue + mve = i, where i is the current through unit
+area of the gas. Since nue = i<span class="su">0</span>&epsilon;<span class="sp">nx</span> and i = i<span class="su">0</span>&epsilon;<span class="sp">nl</span>, when l is the distance
+between the plates, we see that</p>
+
+<p class="center1">nu / mv = &epsilon;<span class="sp">nx</span> / (&epsilon;<span class="sp">nl</span> &minus; &epsilon;<span class="sp">nx</span>),</p>
+
+<table class="math0" summary="math">
+<tr><td>n</td> <td rowspan="2">=</td> <td>v</td> <td rowspan="2">·</td>
+ <td>&epsilon;<span class="sp">nx</span></td> <td rowspan="2">.</td></tr>
+<tr><td class="denom">m</td> <td class="denom">u</td> <td class="denom">&epsilon;<span class="sp">ne</span> &minus; &epsilon;<span class="sp">nx</span></td></tr></table>
+
+<p class="noind">Since v/u is a very small quantity we see that n will be less than m
+except when &epsilon;<span class="sp">nl</span> - &epsilon;<span class="sp">nx</span> is small, <i>i.e.</i> except close to the anode. Thus
+there will be an excess of positive electricity from the cathode almost
+up to the anode, while close to the anode there will be an excess of
+negative. This distribution of electricity will make the electric
+force diminish from the cathode to the place where there is as much
+positive as negative electricity, where it will have its minimum
+value, and then increase up to the anode.</p>
+
+<p>The expression i = i<span class="su">0</span>&epsilon;<span class="sp">&alpha;l</span> applies to the case when there is no source
+of ionization in the gas other than the collisions; if in addition to
+this there is a source of uniform ionization producing q ions per cubic
+centimetre, we can easily show that</p>
+
+<table class="math0" summary="math">
+<tr><td rowspan="2">i = i<span class="su">0</span>&epsilon;<span class="sp">&alpha;l</span> +</td> <td>qe</td>
+ <td rowspan="2">(e<span class="sp">&alpha;l</span> &minus; 1).</td></tr>
+<tr><td class="denom">&alpha;</td></tr></table>
+
+<p class="noind">With regard to the minimum energy which must be possessed by a
+corpuscle to enable it to produce ions by collision, Townsend (loc.
+cit.) came to the conclusion that to ionize air the corpuscle must
+possess an amount of energy equal to that acquired by the fall of its
+charge through a potential difference of about 2 volts. This is also
+the value arrived at by H. A. Wilson by entirely different considerations.
+Stark, however, gives 17 volts as the minimum for ionization.
+The energy depends upon the nature of the gas; recent experiments by
+Dawes and Gill and Pedduck (<i>Phil. Mag.</i>, Aug. 1908) have shown that
+it is smaller for helium than for air, hydrogen, or carbonic acid gas.</p>
+</div>
+
+<p>If there is no external source of ionization and no emission of
+corpuscles from the cathode, then it is evident that even if some
+corpuscles happened to be present in the gas when the electric
+field were applied, we could not get a permanent current by
+the aid of collisions made by these corpuscles. For under the
+electric field, the corpuscles would be driven from the cathode
+to the anode, and in a short time all the corpuscles originally
+present in the gas and those produced by them would be driven
+from the gas against the anode, and if there was no source from
+which fresh corpuscles could be introduced into the gas the
+current would cease. The current, however, could be maintained
+indefinitely if the positive ions in their journey back to the cathode
+also produced ions by collisions, for then we should have a kind
+of regenerative process by which the supply of corpuscles could
+be continually renewed. To maintain the current it is not necessary
+that the ionization resulting from the positive ions should be
+anything like as great as that from the negative, as the investigation
+given below shows a very small amount of ionization by the
+positive ions will suffice to maintain the current. The existence
+of ionization by collision with positive ions has been proved by
+Townsend. Another method by which the current could be
+and is maintained is by the anode emitting corpuscles under the
+impact of the positive ions driven against it by the electric field.
+J. J. Thomson has shown by direct experiment that positively
+<span class="pagenum"><a name="page880" id="page880"></a>880</span>
+electrified particles when they strike against a metal plate cause
+the metal to emit corpuscles (J. J. Thomson, <i>Proc. Camb. Phil.
+Soc.</i> 13, p. 212; Austin, <i>Phys. Rev.</i> 22, p. 312). If we assume
+that the number of corpuscles emitted by the plate in one second
+is proportional to the energy in the positive ions which strike
+the plate in that second, we can readily find an expression for
+the difference of potential which will maintain without any
+external ionization a current of electricity through the gas.
+As this investigation brings into prominence many of the most
+important features of the electric discharge, we shall consider it
+in some detail.</p>
+
+<div class="condensed">
+<p>Let us suppose that the electrodes are parallel plates of metal at
+right angles to the axis of x, and that at the cathode x = 0 and at the
+anode x = d, d being thus the distance between the plates. Let us
+also suppose that the current of electricity flowing between the plates
+is so small that the electrification between the plates due to the
+accumulation of ions is not sufficient to disturb appreciably the
+electric field, which we regard as uniform between the plates, the
+electric force being equal to V/d, where V is the potential difference
+between the plates. The number of positive ions produced per
+second in a layer of gas between the planes x and x + dx is &alpha;nu·dx.
+Here n is the number of corpuscles per unit volume, &alpha; the coefficient
+of ionization (for strong electric field &alpha; = 1/&lambda;&prime;, where &lambda;&prime; is the mean
+free path of a corpuscle), and u the velocity of a corpuscle parallel
+to x. We have seen that nu = i<span class="su">0</span>&epsilon;<span class="sp">&alpha;x</span>, where i<span class="su">0</span> is the number of
+corpuscles emitted per second by unit area of the cathode. Thus
+the number of positive ions produced in the layer is &alpha;i<span class="su">0</span>&epsilon;<span class="sp">&alpha;x</span> dx. If
+these went straight to the cathode without a collision, each of them
+would have received an amount of kinetic energy Vex/d when
+they struck the cathode, and the energy of the group of ions would
+be Vex/d·&alpha;i<span class="su">0</span>&epsilon;<span class="sp">dx</span> dx. The positive ions will, however, collide with
+the molecules of the gas through which they are passing, and this
+will diminish the energy they possess when they reach the cathode.</p>
+
+<p>The diminution in the energy will increase in geometrical proportion
+with the length of path travelled by the ion and will thus
+be proportional to &epsilon;<span class="sp">&minus;&beta;x</span>, &beta; will be proportional to the number of
+collisions and will thus be proportional to the pressure of the gas.
+Thus the kinetic energy possessed by the ions when they reach the
+cathode will be</p>
+
+<p class="center1">&epsilon;<span class="sp">&minus;&beta;x</span>·V(ex/d) · &alpha; i<span class="su">0</span>&epsilon;<span class="sp">&alpha;x</span> dx,</p>
+
+<p class="noind">and E, the total amount of energy in the positive ions which reach
+the cathode in unit time, will be given by the equation</p>
+
+<table class="math0" summary="math">
+<tr><td rowspan="2">E = <span class="f200">&int;</span><span class="sp" style="bottom: 1.5em;">d</span><span class="su2">0</span>
+ &epsilon;<span class="sp">&minus;&beta;x</span> · V(ex/d) · &alpha; i<span class="su">0</span>&epsilon;<span class="sp">&alpha;x</span> dx =</td>
+ <td>Ve&alpha; i<span class="su">0</span></td>
+ <td rowspan="2"><span class="f200">&int;</span><span class="sp" style="bottom: 1.5em;">d</span><span class="su2">0</span>
+ &epsilon;<span class="sp">&minus;(&beta;&minus;&alpha;)x</span> x dx</td></tr>
+<tr><td class="denom">d</td></tr></table>
+
+<table class="math0" summary="math">
+<tr><td rowspan="2">=</td> <td>Ve&alpha; i<span class="su">0</span></td>
+ <td rowspan="2"><span class="f250">{</span></td> <td>1</td>
+ <td rowspan="2"> &minus; &epsilon;<span class="sp">&minus;(&beta;&minus;&alpha;)d</span><span class="f200">{</span></td> <td>1</td>
+ <td rowspan="2">+</td> <td>d</td>
+ <td rowspan="2"><span class="f200">}</span><span class="f250">}</span>&emsp;&emsp;(1).</td></tr>
+<tr><td class="denom">d</td> <td class="denom">(&beta; &minus; &alpha;)²</td>
+ <td class="denom">(&beta; &minus; &alpha;)²</td> <td class="denom">(&beta; &minus; &alpha;)</td></tr></table>
+
+<p class="noind">If the number of corpuscles emitted by the cathode in unit time is
+proportional to this energy we have i<span class="su">0</span> = kE, where k is a constant;
+hence by equation (1) we have</p>
+
+<table class="math0" summary="math">
+<tr><td rowspan="2">V =</td> <td>(&beta; &minus; &alpha;)²</td>
+ <td rowspan="2">·</td> <td>d</td> <td rowspan="2">,</td></tr>
+<tr><td class="denom">ke &alpha;</td> <td class="denom">I</td></tr></table>
+
+<p class="noind">where</p>
+
+<p class="center1">I = 1 &minus; &epsilon;<span class="sp">&minus;(&beta;&minus;&alpha;)d</span> (1 + d (&beta; &minus; &alpha;)).</p>
+
+<p class="noind">Since both &beta; and &alpha; are proportional to the pressure, I and (&beta; &minus; &alpha;)²d/&alpha;
+are both functions of pd, the product of the pressure and the spark
+length, hence we see that V is expressed by an equation of the form</p>
+
+<table class="math0" summary="math">
+<tr><td rowspan="2">V =</td> <td>1</td> <td rowspan="2"><span class="f150">&int;</span> (pd) &emsp;&emsp;(2),</td></tr>
+<tr><td class="denom">ke</td></tr></table>
+
+<p class="noind">where &int; (pd) denotes a function of pd, and neither p nor d enter into
+the expression for V except in this product. Thus the potential
+difference required to produce discharge is constant as long as the
+product of the pressure and spark length remains constant; in
+other words, the spark potential is constant as long as the mass
+of the gas between the electrodes is constant. Thus, for example,
+if we halve the pressure the same potential difference will produce
+a spark of twice the length. This law, which was discovered by
+Paschen for fairly long sparks (<i>Annalen</i>, 37, p. 79), and has been
+shown by Carr (<i>Phil. Trans.</i>, 1903) to hold for short ones, is one of
+the most important properties of the electric discharge.</p>
+
+<p>We see from the expression for V that when (&beta; &minus; &alpha;)d is very large</p>
+
+<p class="center1">V = (&beta; &minus; &alpha;)²d/ke&alpha;.</p>
+
+<p class="noind">Thus V becomes infinite when d is infinite. Again when (&beta; &minus; &alpha;)d
+is very small we find</p>
+
+<p class="center1">V = 1/ke&alpha;d;</p>
+
+<p class="noind">thus V is again infinite when d is nothing. There must therefore
+be some value of d intermediate between zero and infinity for which
+V is a minimum. This value is got by finding in the usual way the
+value of d, which makes the expression for V given in equation (1)
+a minimum. We find that d must satisfy the equation</p>
+
+<p class="center1">1 = &epsilon;<span class="sp">&minus;(&beta;&minus;&alpha;)d</span> {1 + (&beta; &minus; &alpha;)d + (&beta; &minus; &alpha;·d)²}.</p>
+
+<p class="noind">We find by a process of trial and error that (&beta; &minus; &alpha;)d = 1.8 is approximately
+a solution of this equation; hence the distance for minimum
+potential is 1.8/(&beta; &minus; &alpha;). Since &beta; and &alpha; are both proportional to the
+pressure, we see that the critical spark length varies inversely as
+the pressure. If we substitute this value in the expression for V
+we find that <span class="ov">V</span>, the minimum spark potential, is given by</p>
+
+<table class="math0" summary="math">
+<tr><td rowspan="2"><span class="ov">V</span> =</td> <td>&beta; &minus; &alpha;</td>
+ <td rowspan="2">·</td> <td>2.2</td> <td rowspan="2">.</td></tr>
+<tr><td class="denom">&alpha;</td> <td class="denom">ke</td></tr></table>
+
+<p class="noind">Since &beta; and &alpha; are each proportional to the pressure, the minimum
+potential is independent of the pressure of the gas. On this view
+the minimum potential depends upon the metal of which the cathode
+is made, since k measures the number of corpuscles emitted per unit
+time by the cathode when struck by positive ions carrying unit
+energy, and unless &beta; bears the same ratio to &alpha; for all gases the
+minimum potential will also vary with the gas. The measurements
+which have been made of the &ldquo;cathode fall of potential,&rdquo; which as
+we shall see is equal to the minimum potential required to produce a
+spark, show that this quantity varies with the material of which the
+cathode is made and also with the nature of the gas. Since a metal
+plate, when bombarded by positive ions, emits corpuscles, the effect
+we have been considering must play a part in the discharge; it is
+not, however, the only effect which has to be considered, for as
+Townsend has shown, positive ions when moving above a certain
+speed ionize the gas, and cause it to emit corpuscles. It is thus
+necessary to take into account the ionization of the positive ions.</p>
+
+<p>Let m be the number of positive ions per unit volume, and w
+their velocity, the number of collisions which occur in one second
+in one cubic centimetre of the gas will be proportional to mwp,
+where p is the pressure of the gas. Let the number of ions which
+result from these collisions be &gamma;mw; &gamma; will be a function of p and
+of the strength of the electric field. Let as before n be the number
+of corpuscles per cubic centimetre, u their velocity, and &alpha;nu the
+number of ions which result in one second from the collisions between
+the corpuscles and the gas. The number of ions produced per
+second per cubic centimetre is equal to &alpha;nu + &gamma;mw; hence when
+things are in a steady state</p>
+
+<table class="math0" summary="math">
+<tr><td>d</td> <td rowspan="2">(nu) = &alpha;nu + &gamma; mw,</td></tr>
+<tr><td class="denom">dx</td></tr></table>
+
+<p class="noind">and</p>
+
+<p class="center1">e(nu + mw) = i,</p>
+
+<p class="noind">where e is the charge on the ion and i the current through the gas.
+The solution of these equations when the field is uniform between the
+plates, is</p>
+
+<p class="center1">enu = C&epsilon;<span class="sp">(&alpha;&minus;&gamma;)x</span> &minus; &gamma;i / (&alpha; &minus; &gamma;),&emsp;&emsp;
+ emw = -C&epsilon;<span class="sp">(&alpha;&minus;&gamma;)x</span> + &alpha;i / (&alpha; &minus; &gamma;),</p>
+
+<p class="noind">where C is a constant of integration. If there is no emission of
+positive ions from the anode enu = i, when x = d. Determining C
+from this condition we find</p>
+
+<table class="math0" summary="math">
+<tr><td rowspan="2">enu =</td> <td>i</td>
+ <td rowspan="2"><span class="f250">{</span>&alpha;&epsilon; <span class="sp">(&alpha;&minus;&gamma;) (x&minus;d)</span> &minus; &gamma;
+ <span class="f250">}</span>,&emsp;&emsp;emw =</td> <td>&alpha;i</td>
+ <td rowspan="2"><span class="f250">{</span>1 &minus; &epsilon; <span class="sp">(&alpha;&minus;&gamma;) (x&minus;d)</span>
+ <span class="f250">}</span>.</td></tr>
+<tr><td class="denom">&alpha; &minus; &gamma;</td> <td class="denom">&alpha; &minus; &gamma;</td></tr></table>
+
+<p class="noind">If the cathode did not emit any corpuscles owing to the bombardment
+by positive ions, the condition that the charge should be
+maintained is that there should be enough positive ions at the cathode
+to carry the current <i>i.e.</i> that emw = i; when x = 0, the condition
+gives</p>
+
+<table class="math0" summary="math">
+<tr><td>i</td> <td rowspan="2"><span class="f250">{</span>
+ &alpha;&epsilon;<span class="sp">&minus;(&alpha;&minus;&gamma;)d</span> &minus; &gamma;<span class="f250">}</span> = 0</td></tr>
+<tr><td class="denom">&alpha; &minus; &gamma;</td></tr></table>
+
+<p class="noind">or</p>
+
+<p class="center1">&epsilon; <span class="sp">&alpha;d</span>/&alpha; = &epsilon; <span class="sp">&gamma;d</span>/&gamma;.</p>
+
+<p class="noind">Since &alpha; and &gamma; are both of the form p&int; (X/p) and X = V/d, we see that
+V will be a function of pd, in agreement with Paschen&rsquo;s law. If we
+take into account both the ionization of the gas and the emission
+of corpuscles by the metal we can easily show that</p>
+
+<table class="math0" summary="math">
+<tr><td>&alpha; &minus; &gamma;&epsilon;<span class="sp">(&alpha;&minus;&gamma;)d</span></td> <td rowspan="2">=</td>
+ <td>k&alpha;Ve</td> <td rowspan="2"><span class="f250">[</span></td>
+ <td>1</td> <td rowspan="2">&minus; &epsilon;<span class="sp">&minus;(&beta;+&gamma;&minus;&alpha;)d</span><span class="f250">{</span></td>
+ <td>1</td> <td rowspan="2">+</td>
+ <td>d</td> <td rowspan="2"><span class="f250">}]</span>,</td></tr>
+<tr><td class="denom">&alpha; &minus; &gamma;</td> <td class="denom">d</td>
+ <td class="denom">(&beta; + &gamma; &minus; &alpha;)²</td> <td class="denom">(&beta; + &gamma; &minus; &alpha;)²</td>
+ <td class="denom">&beta; + &gamma; &minus; &alpha;</td></tr></table>
+
+<p class="noind">where k and &beta; have the same meaning as in the previous investigation.
+When d is large, &epsilon;<span class="sp">(&alpha;&minus;&gamma;)d</span> is also large; hence in order that the left-hand
+side of this equation should not be negative &gamma; must be less
+than &alpha;/&epsilon;<span class="sp">(&alpha;&minus;&gamma;)d</span>; as this diminishes as d increases we see that when
+the sparks are very long discharge will take place, practically as
+soon as &gamma; has a finite value, <i>i.e.</i> as soon as the positive ions begin to
+produce fresh ions by their collisions.</p>
+</div>
+
+<p>In the preceding investigation we have supposed that the
+electric field between the plates was uniform; if it were not
+uniform we could get discharges produced by very much smaller
+differences of potential than are necessary in a uniform field.
+For to maintain the discharge it is not necessary that the positive
+ions should act as ionizers all along their path; it is sufficient
+that they should do so in the neighbourhood of cathode. Thus
+if we have a strong field close to the cathode we might still get
+<span class="pagenum"><a name="page881" id="page881"></a>881</span>
+the discharge though the rest of the field were comparatively
+weak. Such a distribution of electric force requires, however,
+a great accumulation of charged ions near the cathode; until
+these ions accumulate the field will be uniform. If the uniform
+field existing in the gas before the discharge begins were strong
+enough to make the corpuscles produce ions by collision, but not
+strong enough to make the positive ions act as ionizers, there
+would be some accumulation of ions, and the amount of this
+accumulation would depend upon the number of free corpuscles
+originally present in the gas, and upon the strength of the electric
+field. If the accumulation were sufficient to make the field
+near the cathode so strong that the positive ions could produce
+fresh ions either by collision with the cathode or with the gas,
+the discharge would pass though the gas; if not, there will be no
+continuous discharge. As the amount
+of the accumulation depends on the
+number of corpuscles present in the gas,
+we can understand how it is that after
+a spark has passed, leaving for a time
+a supply of corpuscles behind it, it is
+easier to get a discharge to pass through
+the gas than it was before.</p>
+
+<table class="nobctr" style="float: left; width: 230px;" summary="Illustration">
+<tr><td class="figleft1"><img style="width:184px; height:471px" src="images/img881a.jpg" alt="" /></td></tr>
+<tr><td class="caption sc">Fig. 15.</td></tr></table>
+
+<p>The inequality of the electric field in
+the gas when a continuous discharge is
+passing through it is very obvious when
+the pressure of the gas is low. In this
+case the discharge presents a highly
+differentiated appearance of which a
+type is represented in fig. 15. Starting
+from the cathode we have a thin velvety
+luminous glow in contact with the surface;
+this glow is
+often called the &ldquo;first
+cathode layer.&rdquo; Next
+this we have a comparatively
+dark space
+whose thickness increases
+as the pressure
+diminishes; this is
+called the &ldquo;Crookes&rsquo;s dark space,&rdquo; or the
+&ldquo;second cathode layer.&rdquo; Next this we have
+a luminous position called the &ldquo;negative
+glow&rdquo; or the &ldquo;third cathode layer.&rdquo; The
+boundary between the second and third layers
+is often very sharply defined. Next to the
+third layer we have another dark space called
+the &ldquo;Faraday dark space.&rdquo; Next to this and reaching up to the
+anode is another region of luminosity, called the &ldquo;positive
+column,&rdquo; sometimes (as in fig. 15, a) continuous, sometimes (as
+in fig. 15, b) broken up into light or dark patches called &ldquo;striations.&rdquo;
+The dimensions of the Faraday dark space and the positive
+column vary greatly with the current passing through the
+gas and with its pressure; sometimes one or
+other of them is absent. These differences
+in appearances are accompanied by great
+difference in the strength of the electric
+field. The magnitude of the electric force
+at different parts of the discharge is represented
+in fig. 16, where the ordinates represent
+the electric force at different parts of
+the tube, the cathode being on the right.
+We see that the electric force is very large indeed between the
+negative glow and the cathode, much larger than in any other
+part of the tube. It is not constant in this region, but increases
+as we approach the cathode. The force reaches a minimum
+either in the negative glow itself or in the part of the Faraday
+dark space just outside, after which it increases towards the
+positive column. In the case of a uniform positive column the
+electric force along it is constant until we get quite close to the
+anode, when a sudden change, called the &ldquo;anode fall,&rdquo; takes
+place in the potential.</p>
+
+<table class="nobctr" style="clear: both;" summary="Illustration">
+<tr><td class="figcenter"><img style="width:473px; height:306px" src="images/img881b.jpg" alt="" /></td></tr>
+<tr><td class="caption sc">Fig. 16.</td></tr></table>
+
+<p>The difference of potential between the cathode and the
+negative glow is called the &ldquo;cathode potential fall&rdquo; and is
+found to be constant for wide variations in the pressure of the
+gas and the current passing through. It increases, however,
+considerably when the current through the gas exceeds a certain
+critical value, depending among other things on the size of the
+cathode. This cathode fall of potential is shown by experiment
+to be very approximately equal to the minimum potential
+difference. The following table contains a comparison of the
+measurements of the cathode fall of potentials in various gases
+made by Warburg (<i>Wied. Ann.</i>, 1887, 31, p. 545, and 1890, 40,
+p. 1), Capstick (<i>Proc. Roy. Society</i>, 1898, 63, p. 356), and Strutt
+(<i>Phil. Trans.</i>, 1900, 193, p. 377), and the measurements by Strutt
+of the smallest difference of potential which will maintain a
+spark through these gases.</p>
+
+<table class="ws" summary="Contents">
+<tr><td class="tccm allb" rowspan="3">Gas.</td> <td class="tccm allb" colspan="4">Cathode fall in Volts.</td>
+ <td class="tccm allb" rowspan="2">Least potential<br />difference required<br />to maintain a Spark.</td></tr>
+<tr><td class="tccm allb" colspan="3">Platinum Electrodes.</td> <td class="tccm allb">Aluminium<br />Electrodes.</td></tr>
+<tr><td class="tccm allb">Warburg.</td> <td class="tccm allb">Capstick.</td> <td class="tccm allb">Strutt.</td>
+ <td class="tccm allb">Warburg.</td> <td class="tccm allb">Strutt.</td></tr>
+
+<tr><td class="tcl lb rb">Air</td> <td class="tcc rb">340-350</td> <td class="tcc rb">..</td> <td class="tcc rb">..</td> <td class="tcc rb">..</td> <td class="tcc rb">341</td></tr>
+<tr><td class="tcl lb rb">H<span class="su">2</span></td> <td class="tcc rb">about 300</td> <td class="tcc rb">298</td> <td class="tcc rb">..</td> <td class="tcc rb">168</td> <td class="tcc rb">302-308</td></tr>
+<tr><td class="tcl lb rb">O<span class="su">2</span></td> <td class="tcc rb">..</td> <td class="tcc rb">369</td> <td class="tcc rb">..</td> <td class="tcc rb">..</td> <td class="tcc rb">..</td></tr>
+<tr><td class="tcl lb rb">N<span class="su">2</span></td> <td class="tcc rb">230 if free</td> <td class="tcc rb">232</td> <td class="tcc rb">..</td> <td class="tcc rb">207</td> <td class="tcc rb">251</td></tr>
+<tr><td class="tcl lb rb">&nbsp;</td> <td class="tcc rb">from oxygen</td> <td class="tcc rb">&nbsp;</td> <td class="tcc rb">&nbsp;</td> <td class="tcc rb">&nbsp;</td> <td class="tcc rb">&nbsp;</td></tr>
+<tr><td class="tcl lb rb">Hg vapour</td> <td class="tcc rb">340</td> <td class="tcc rb">..</td> <td class="tcc rb">..</td> <td class="tcc rb">..</td> <td class="tcc rb">..</td></tr>
+<tr><td class="tcl lb rb">Helium</td> <td class="tcc rb">..</td> <td class="tcc rb">..</td> <td class="tcc rb">226</td> <td class="tcc rb">..</td> <td class="tcc rb">261-326</td></tr>
+<tr><td class="tcl lb rb">H<span class="su">2</span>O</td> <td class="tcc rb">..</td> <td class="tcc rb">469</td> <td class="tcc rb">..</td> <td class="tcc rb">..</td> <td class="tcc rb">..</td></tr>
+<tr><td class="tcl lb rb bb">NH<span class="su">3</span></td> <td class="tcc rb bb">..</td> <td class="tcc rb bb">582</td> <td class="tcc rb bb">..</td> <td class="tcc rb bb">..</td> <td class="tcc rb bb">..</td></tr>
+</table>
+
+<p>Thus in the cases in which the measurements could be made
+with the greatest accuracy the agreement between the cathode
+fall and the minimum potential difference is very close. The
+cathode fall depends on the material of which the terminals
+are made, as is shown by the following table due to Mey (<i>Verh.
+deutsch. physik. Gesell.</i>, 1903, 5, p. 72).</p>
+
+<table class="ws" summary="Contents">
+<tr><td class="tccm allb" rowspan="2">Gas.</td> <td class="tccm allb" colspan="11">Electrode.</td></tr>
+<tr><td class="tccm allb">Pt</td> <td class="tccm allb">Hg</td> <td class="tccm allb">Ag</td>
+ <td class="tccm allb">Cu</td> <td class="tccm allb">Fe</td> <td class="tccm allb">Zn</td>
+ <td class="tccm allb">Al</td> <td class="tccm allb">Mg</td> <td class="tccm allb">Na</td>
+ <td class="tccm allb">Na-K</td> <td class="tccm allb">K</td></tr>
+
+<tr><td class="tcl lb rb">O<span class="su">2</span></td> <td class="tcc rb">369</td> <td class="tcc rb">..</td> <td class="tcc rb">..</td> <td class="tcc rb">..</td> <td class="tcc rb">..</td> <td class="tcc rb">..</td> <td class="tcc rb">..</td> <td class="tcc rb">..</td> <td class="tcc rb">..</td> <td class="tcc rb">..</td> <td class="tcc rb">..</td></tr>
+<tr><td class="tcl lb rb">H<span class="su">2</span></td> <td class="tcc rb">300</td> <td class="tcc rb">..</td> <td class="tcc rb">295</td> <td class="tcc rb">280</td> <td class="tcc rb">230</td> <td class="tcc rb">213</td> <td class="tcc rb">190</td> <td class="tcc rb">168</td> <td class="tcc rb">185</td> <td class="tcc rb">169</td> <td class="tcc rb">172</td></tr>
+<tr><td class="tcl lb rb">N<span class="su">2</span></td> <td class="tcc rb">232</td> <td class="tcc rb">226</td> <td class="tcc rb">..</td> <td class="tcc rb">..</td> <td class="tcc rb">..</td> <td class="tcc rb">..</td> <td class="tcc rb">..</td> <td class="tcc rb">207</td> <td class="tcc rb">178</td> <td class="tcc rb">125</td> <td class="tcc rb">170</td></tr>
+<tr><td class="tcl lb rb">He</td> <td class="tcc rb">226</td> <td class="tcc rb">..</td> <td class="tcc rb">..</td> <td class="tcc rb">..</td> <td class="tcc rb">..</td> <td class="tcc rb">..</td> <td class="tcc rb">..</td> <td class="tcc rb">..</td> <td class="tcc rb">80</td> <td class="tcc rb">78.5</td> <td class="tcc rb">&ensp;69</td></tr>
+<tr><td class="tcl lb rb bb">Argon</td> <td class="tcc rb bb">167</td> <td class="tcc rb bb">..</td> <td class="tcc rb bb">..</td> <td class="tcc rb bb">..</td> <td class="tcc rb bb">..</td> <td class="tcc rb bb">..</td> <td class="tcc rb bb">100</td> <td class="tcc rb bb">..</td> <td class="tcc rb bb">..</td> <td class="tcc rb bb">..</td> <td class="tcc rb bb">..</td></tr>
+</table>
+
+<p>The dependence of the minimum potential required to produce
+a spark upon the metal of which the cathode is made has not
+been clearly established, some observers being unable to detect
+any difference between the potential required to spark between
+electrodes of aluminium and those of brass, while others thought
+they had detected such a difference. It is only with sparks
+not much longer than the critical spark length that we could
+hope to detect this difference. When the current through the
+gas exceeds a certain critical value depending among other
+things on the size of the cathode, the cathode fall of potential
+increases rapidly and at the same time the thickness of the dark
+<span class="pagenum"><a name="page882" id="page882"></a>882</span>
+spaces diminishes. We may regard the part of the discharge
+between the cathode and the negative glow as a discharge taking
+place under minimum potential difference through a distance
+equal to the critical spark length. An inspection of fig. 16 will
+show that we cannot regard the electric field as constant even
+for this small distance; it thus becomes a matter of interest to
+know what would be the effect on the minimum potential
+difference required to produce a spark if there were sufficient
+ions present to produce variations in the electric field analogous
+to those represented in fig. 16. If the electric force at a distance
+x from the cathode were proportional to &epsilon;<span class="sp">-px</span> we should have a
+state of things much resembling the distribution of electric
+force near the cathode. If we apply to this distribution the
+methods used above for the case when the force was uniform,
+we shall find that the minimum potential is less and the
+critical spark length greater than when the electric force is
+uniform.</p>
+
+<p><i>Potential Difference required to produce a Spark of given Length.</i>&mdash;We
+may regard the region between the cathode and the negative
+glow as a place for the production of corpuscles, these corpuscles
+finding their way from this region through the negative glow.
+The parts of this glow towards the anode we may regard as a
+cathode, from which, as from a hot lime cathode, corpuscles are
+emitted. Let us now consider what will happen to these corpuscles
+shot out from the negative glow with a velocity depending
+on the cathode fall of potential and independent of the pressure.
+These corpuscles will collide with the molecules of the gas, and
+unless there is an external electric field to maintain their velocity
+they will soon come to rest and accumulate in front of the
+negative glow. The electric force exerted by this cloud of
+corpuscles will diminish the strength of the electric field in the
+region between the cathode and the negative glow, and thus
+tend to stop the discharge. To keep up the discharge we must
+have a sufficiently strong electric field between the negative
+glow and the anode to remove the corpuscles from this region as
+fast as they are sent into it from the cathode. If, however,
+there is no production of ions in the region between the negative
+glow and the anode, all the ions in this region will have come
+from near the cathode and will be negatively charged; this
+negative electrification will diminish the electric force on the
+cathode side of it and thus tend to stop the discharge. This
+back electric field could, however, be prevented by a little ionization
+in the region between the anode and glow, for this would
+afford a supply of positive ions, and thus afford an opportunity
+for the gas in this region to have in it as many positive as negative
+ions; in this case it would not give rise to any back electromotive
+force. The ionization which produces these positive
+ions may, if the field is intense, be due to the collisions of corpuscles,
+or it may be due to radiation analogous to ultra-violet,
+or soft Röntgen rays, which have been shown by experiment
+to accompany the discharge. Thus in the most simple conditions
+for discharge we should have sufficient ionization to keep up the
+supply of positive ions, and an electric field strong enough to
+keep the velocity of the negative corpuscle equal to the value
+it has when it emerges from the negative glow. Thus the force
+must be such as to give a constant velocity to the corpuscle,
+and since the force required to move an ion with a given velocity
+is proportional to the pressure, this force will be proportional
+to the pressure of the gas. Let us call this force ap; then if l
+is the distance of the anode from the negative glow the potential
+difference between these points will be alp. The potential
+difference between the negative glow and the cathode is constant
+and equals c; hence if V is the potential difference between
+the anode and cathode, then V = c + alp, a relation which expresses
+the connexion between the potential difference and
+spark length for spark lengths greater than the critical distance.
+It is to be remembered that the result we have obtained applies
+only to such a case as that indicated above, where the electric
+force is constant along the positive column. Experiments
+with the discharge through gases at low pressure show the
+discharge may take other forms. Thus the positive column
+may be striated when the force along it is no longer uniform,
+or the positive column may be absent; the discharge may be
+changed from one of these forms to another by altering the
+current. The relation between the potential and the distance
+between the electrodes varies greatly, as we might expect, with
+the current passing through the gas.</p>
+
+<p>The connexion between the potential difference and the
+spark length has been made the subject of a large number of
+experiments. The first measurements were made by Lord
+Kelvin in 1860 (<i>Collected Papers on Electrostatics and Magnetism</i>,
+p. 247); subsequent experiments have been made by Baille
+(<i>Ann. de chimie et de physique</i>, 5, 25, p. 486), Liebig (<i>Phil. Mag.</i>
+[5], 24, p. 106), Paschen (<i>Wied. Ann.</i> 37, p. 79), Peace (<i>Proc. Roy.
+Soc.</i>, 1892, 52, p. 99), Orgler (<i>Ann. der Phys.</i> 1, p. 159), Strutt
+(<i>Phil. Trans.</i> 193, p. 377), Bouty (<i>Comptes rendus</i>, 131, pp. 469,
+503), Earhart (<i>Phil. Mag.</i> [6], 1, p. 147), Carr (<i>Phil. Trans.</i>, 1903),
+Russell (<i>Phil. Mag.</i> [5], 64, p. 237), Hobbs (<i>Phil. Mag.</i> [6], 10,
+p. 617), Kinsley (<i>Phil. Mag.</i> [6], 9, 692), Ritter (<i>Ann. der Phys.</i>
+14, p. 118). The results of their experiments show that for sparks
+considerably longer than the critical spark length, the relation
+between the potential difference V and the spark length l may
+be expressed when the electrodes are large with great accuracy
+by the linear relation V = c + blp, where p is the pressure and
+c and b are constants depending on the nature of the gas. When
+the sparks are long the term blp is the most important and the
+sparking potential is proportional to the spark length. Though
+there are considerable discrepancies between the results obtained
+by different observers, these indicate that the production of a
+long spark between large electrodes in air at atmospheric pressure
+requires a potential difference of 30,000 volts for each centimetre
+of spark length. In hydrogen only about half this potential
+difference is required, in carbonic acid gas the potential difference
+is about the same as in air, while Ritter&rsquo;s experiments show
+that in helium only about one-tenth of this potential difference
+is required.</p>
+
+<p>In the case when the electric field is not uniform, as for example
+when the discharge takes place between spherical electrodes,
+Russell&rsquo;s experiments show that the discharge takes place as
+soon as the maximum electric force in the field between the
+electrodes reaches a definite value, which he found was for air at
+atmospheric pressure about 38,000 volts per centimetre.</p>
+
+<p><i>Very Short Sparks.</i>&mdash;Some very interesting experiments on the
+potential difference required to produce exceedingly short sparks
+have been made by Earhart, Hobbs and Kinsley; the length of
+these sparks was comparable with the wave length of sodium
+light. With sparks of these lengths it was found that it was
+possible to get a discharge with less than 330 volts, the minimum
+potential difference in air. The results of these observers show
+that there is no diminution in the minimum potential difference
+required to produce discharge until the spark length gets so small
+that the average electric force between the electrodes amounts to
+about one million volts per centimetre. When the force rises to
+this value a discharge takes place even though the potential
+difference is much less than 330 volts; in some of Earhart&rsquo;s
+experiments it was only about 2 volts. This kind of discharge is
+determined not by the condition that the potential difference
+should have a given value, but that the electric force should have
+a given value. Another point in which this discharge differs from
+the ordinary one is that it is influenced entirely by the nature
+of the electrodes and not by the nature or pressure of the gas
+between them, whereas the ordinary discharge is in many cases
+not affected appreciably by changes in the metal of the electrodes,
+but is always affected by changes in the pressure and character
+of the gas between them. Kinsley found that when one of these
+small sparks passed between the electrodes a kind of metallic
+bridge was formed between them, so that they were in metallic
+connexion, and that the distance between them had to be
+considerably increased before the bridge was broken. Almy
+(<i>Phil. Mag.</i>, Sept. 1908), who used very small electrodes, was
+unable to get a discharge with less than the minimum spark
+potential even when the spark length was reduced to one-third of
+the wave length of sodium light. He suggests that the discharges
+obtained with larger electrodes for smaller voltages are
+<span class="pagenum"><a name="page883" id="page883"></a>883</span>
+due to the electrodes being dragged together by the electrostatic
+attraction between them.</p>
+
+<p><i>Constitution of the Electric Spark.</i>&mdash;Schuster and Hemsalech
+(<i>Phil. Trans.</i> 193, p. 189), Hemsalech (<i>Comptes Rendus</i>, 130, p.
+898; 132, p. 917; <i>Jour. de Phys.</i> 3. 9, p. 43, and Schenck, <i>Astrophy.
+Jour.</i> 14, p. 116) have by spectroscopic methods obtained very
+interesting results about the constitution of the spark. The
+method employed by Schuster and Hemsalech was as follows:
+Suppose we photograph the spectrum of a horizontal spark on a
+film which is on the rim of a wheel rotating about a horizontal
+axis with great velocity. If the luminosity travelled with
+infinite speed from one electrode to the other, the image on the
+film would be a horizontal line. If, however, the speed with
+which the luminosity travelled between the electrodes was
+comparable with the speed of the film, the line would be inclined
+to the horizontal, and by measuring the inclinations we could
+find the speed at which the luminosity travelled. In this way
+Schuster and Hemsalech showed that when an oscillating
+discharge passed between metallic terminals in air, the first spark
+passes through the air alone, no lines of the metal appearing in
+its spectrum. This first spark vaporizes some of the metal and
+the subsequent sparks passing mainly through the metallic
+vapour; the appearance of the lines in the film shows that the
+velocity of the luminous part of the vapour was finite. The
+velocity of the vapour of metals of low atomic weight was in
+general greater than that of the vapour of heavier metals.
+Thus the velocity of aluminium vapour was 1890 metres per
+second, that of zinc and cadmium only about 545. Perhaps the
+most interesting point in the investigation was the discovery that
+the velocities corresponding to different lines in the spectrum of
+the same metal were in some cases different. Thus with bismuth
+some of the lines indicated a velocity of 1420 metres per second,
+others a velocity of only 550, while one (&lambda; = 3793) showed
+a still smaller velocity. These results are in accordance with a
+view suggested by other phenomena that many of the lines in a
+spectrum produced by an electrical discharge originate from
+systems formed during the discharge and not from the normal
+atom or molecule. Schuster and Hemsalech found that by
+inserting a coil with large self induction in the primary circuit
+they could obliterate the air lines in the discharge.</p>
+
+<p>Schenck, by observing the appearance presented when an
+alternating current, produced by discharging Leyden jars, was
+examined in a rapidly rotating mirror, found it showed the
+following stages: (1) a thin bright line, followed in some cases at
+intervals of half the period of the discharge by fainter lines; (2)
+bright curved streamers starting from the negative terminal, and
+diminishing rapidly in speed as they receded from the cathode;
+(3) a diffused glow lasting for a much longer period than either
+of the preceding. These constituents gave out quite different
+spectra.</p>
+
+<p>The structure of the discharge is much more easily studied
+when the pressure of the gas is low, as the various parts which
+make up the discharge are more widely separated from each other.
+We have already described the general appearance of the discharge
+through gases at low pressures (see p. 657). There is,
+however, one form of discharge which is so striking and beautiful
+that it deserves more detailed consideration. In this type of
+discharge, known as the striated discharge, the positive column
+is made up of alternate bright and dark patches known as
+<i>striations</i>. Some of these are represented in fig. 17, which is
+taken from a paper by De la Rue and Müller (<i>Phil. Trans.</i>, 1878,
+Pt. 1). This type of discharge only occurs when the current and
+the pressure of the gas are between certain limits. It is most
+beautifully shown when a Wehnelt cathode is used and the
+current is produced by storage cells, as this allows us to use large
+currents and to maintain a steady potential difference between the
+electrodes. The striations are in consequence very bright and
+steady. The facts which have been established about these
+striations are as follows: The distance between the bright parts
+of the striations is greater at low pressures than at high; it
+depends also upon the diameter of the tube, increasing as the
+diameter of the tube increases. If the discharge tube is wide at
+one place and narrow in another the striations will be closer
+together in the narrow parts than in the wide. The distance
+between the striations depends on the current through the tube.
+The relation is not a very simple one, as an increase of current
+sometimes increases while under other circumstances it decreases
+the distance between the striations (see Willows, <i>Proc. Camb.
+Phil. Soc.</i> 10, p. 302). The electric force is not uniform along the
+striated discharge, but is greater in the bright than in the dark
+parts of the striation. An example is shown in fig. 16, due to H.
+A. Wilson, which shows the distribution of electric force at every
+place in a striated discharge. In experiments made by J. J.
+Thomson (<i>Phil. Mag.</i>, Oct. 1909), using a Wehnelt cathode, the
+variations in the electric force were more pronounced than those
+shown in fig. 16. The electric force in this case changed so greatly
+that it actually became negative just on the cathode side of the
+bright part of the striation. Just inside the striation on the anode
+side it rose to a very high value, then continually diminished
+towards the bright side of the next striation when it again
+increased. This distribution of electric force implies that there
+is great excess of negative electricity at the bright head of the
+striation, and a small excess of positive everywhere else. The
+temperature of the gas is higher in the bright than in the dark
+parts of the striations. Wood (<i>Wied. Ann.</i> 49, p. 238), who has
+made a very careful study of the distribution of temperature in
+a discharge tube, finds that in those tubes the temperature varies
+in the same way as the electric force, but that this temperature
+(which it must be remembered is the average temperature of all
+the molecules and not merely of those which are taking part in
+the discharge) is by no means high; in no part of the discharge
+did the temperature in his experiments exceed 100° C.</p>
+
+<table class="nobctr" summary="Illustration">
+<tr><td class="figcenter"><img style="width:517px; height:531px" src="images/img883.jpg" alt="" /></td></tr>
+<tr><td class="caption sc">Fig. 17.</td></tr></table>
+
+<p><i>Theory of the Striations.</i>&mdash;We may regard the heaping up of
+the negative charges at intervals along the discharge as the
+fundamental feature in the striations, and this heaping up may
+be explained as follows. Imagine a corpuscle projected with
+considerable velocity from a place where the electric field is
+strong, such as the neighbourhood of the cathode; as it moves
+towards the anode through the gas it will collide with the molecules,
+ionize them and lose energy and velocity. Thus unless
+the corpuscle is acted on by a field strong enough to supply it
+with the energy it loses by collision, its speed will gradually
+diminish. Further, when its energy falls below a certain value
+it will unite with a molecule and become part of a negative ion,
+instead of a corpuscle; at this stage there will be a sudden and
+<span class="pagenum"><a name="page884" id="page884"></a>884</span>
+very large diminution in its velocity. Let us now follow the
+course of a stream of corpuscles starting from the cathode and
+approaching the anode. If the speed falls off as the stream
+proceeds, the corpuscles in the rear will gain on those in front
+and the density of the stream in the front will be increased.
+If at a certain place the velocity receives a sudden check by the
+corpuscles becoming loaded with a molecule, the density of the
+negative electricity will increase at this place with great rapidity,
+and here there will be a great accumulation of negative electricity,
+as at the bright head on the cathode side of a striation. Now
+this accumulation of negative electricity will produce a large
+electric force on the anode side; this will drive corpuscles
+forward with great velocity and ionize the gas. These corpuscles
+will behave like those shot from the cathode and will accumulate
+again at some distance from their origin, forming the bright
+head of the next striation, when the process will be repeated.
+On this view the bright heads of the striations act like electrodes,
+and the discharge passes from one bright head to the next as by
+a number of stepping stones, and not directly from cathode
+to anode. The luminosity at the head of the striations is due
+to the recombination of the ions. These ions have acquired
+considerable energy from the electric field, and this energy will
+be available for supplying the energy radiated away as light.
+The recombination of ions which do not possess considerable
+amounts of energy does not seem to give rise to luminosity.
+Thus, in an ionized gas not exposed to an electric field, although
+we have recombination between the ions, we need not have
+luminosity. We have at present no exact data as to the amount
+of energy which must be given to an ion to make it luminous
+on recombination; it also certainly varies with the nature of
+the ion; thus even with hot Wehnelt cathodes J. J. Thomson
+has never been able to make the discharge through air luminous
+with a potential less than from 16 to 17 volts. The mercury
+lamps, however, in which the discharge passes through mercury
+vapour are luminous with a potential difference of about 12
+volts. It follows that if the preceding theory be right the
+potential difference between two bright striations must be
+great enough to make the corpuscles ionize by collision and also
+to give enough energy to the ions to make them luminous when
+they recombine. The difference of potential between the bright
+parts of successive striations has been measured by Hohn (<i>Phys.
+Zeit.</i> 9, p. 558); it varies with the pressure and with the gas.
+The smallest value given by Hohn is about 15 volts. In some
+experiments made by J. J. Thomson, when the pressure of the
+gas was very low, the difference of potential between two adjacent
+dark spaces was as low as 3.75 volts.</p>
+
+<p><i>The Arc Discharge.</i>&mdash;The discharges we have hitherto considered
+have been characterized by large potential differences
+and small currents. In the arc discharge we get very large
+currents with comparatively small potential differences. We
+may get the arc discharge by taking a battery of cells large
+enough to give a potential difference of 60 to 80 volts, and
+connecting the cells with two carbon terminals, which are put
+in contact, so that a current of electricity flows round the circuit.
+If the terminals, while the current is on, are drawn apart, a
+bright discharge, which may carry a current of many amperes,
+passes from one to the other. This arc discharge, as it is called, is
+characterized by intense heat and by the brilliant luminosity
+of the terminals. This makes it a powerful source of light.
+The temperature of the positive terminal is much higher than
+that of the negative. According to Violle (<i>Comptes Rendus</i>,
+115, p. 1273) the temperature of the tip of the former is about
+3500° C, and that of the latter 2700° C. The temperature of the
+arc itself he found to be higher than that of either of its terminals.
+As the arc passes, the positive terminal gets hollowed out into
+a crater-like shape, but the negative terminal remains pointed.
+Both terminals lose weight.</p>
+
+<div class="condensed">
+<p>The appearance of the terminals is shown in fig. 18, given by Mrs
+Ayrton (<i>Proc. Inst. Elec. Eng.</i> 28, p. 400); a, b represent the terminals
+when the arc is quiet, and c when it is accompanied by a hissing sound.
+The intrinsic brightness of the positive crater does not increase with
+an increase in the current; an increased current produces an increase
+in the area of the luminous crater, but the amount of light given
+out by each unit of area of luminous surface is unaltered. This
+indicates that the temperature of the crater is constant; it is
+probably that at which carbon volatilizes. W. E. Wilson (<i>Proc.
+Roy. Soc.</i> 58, p. 174; 60, p. 377) has shown that at pressures of
+several atmospheres the intrinsic brightness of the crater is considerably
+diminished.</p>
+
+<table class="nobctr" summary="Illustration">
+<tr><td class="figcenter"><img style="width:443px; height:254px" src="images/img884a.jpg" alt="" /></td></tr>
+<tr><td class="caption sc">Fig. 18.</td></tr></table>
+
+<table class="nobctr" summary="Illustration">
+<tr><td class="figcenter"><img style="width:418px; height:309px" src="images/img884b.jpg" alt="" /></td></tr>
+<tr><td class="caption sc">Fig. 19.</td></tr></table>
+
+<p>The connexion between V, the potential difference between the
+terminals, and l, the length of the arc, is somewhat analogous to
+that which holds for the spark discharge. Fröhlich (<i>Electrotech. Zeit.</i>
+4, p. 150) gives for this connexion the relation V = m + nl, where
+m and n are constants. Mrs Ayrton (<i>The Electric Arc</i>, chap. iv.)
+finds that both m and n depend upon the current passing between
+the terminals, and gives as the relation between V and l,
+V = &alpha; + &beta;/I + (&gamma; + &delta;/I)l,
+where &alpha;, &beta;, &gamma;, &delta; are constants and I the current.
+The relation between current and potential difference was made the
+subject of a series of experiments by Ayrton (<i>Electrician</i>, 1, p. 319;
+xi. p. 418), some of whose results are represented in fig. 19. For a
+quiet arc an increase in current is accompanied by a fall in potential
+difference, while for the hissing arc the potential difference is independent
+of the current. The quantities m and n which occur in
+Fröhlich&rsquo;s equation have been determined by several experimenters.
+For carbon electrodes in air at atmospheric pressure m is about 39
+volts, varying somewhat with the size and purity of the carbons;
+it is diminished by soaking the terminals in salt solution. The
+value of n given by different observers varies considerably, ranging
+from .76 to 2 volts when l is measured in millimetres; it depends
+upon the current, diminishing as the current increases. When
+metallic terminals are used instead of carbons, the value of m
+depends upon the nature of the metal, m in general being larger
+the higher the temperature at which the metal volatilizes. Thus
+v. Lang (<i>Wied. Ann.</i> 31, p. 384) found the following values for m in
+air at atmospheric pressure:&mdash;C = 35; Pt = 27.4; Fe = 25; Ni = 26.18;
+Cu = 23.86; Ag = 15.23; Zn = 19.86; Cd = 10.28. Lecher (<i>Wied.
+Ann.</i> 33, p. 609) gives Pt = 28, Fe = 20, Ag = 8, while Arons (<i>Wied.
+Ann.</i> 31, p. 384) found for Hg the value 12.8; in this case the fall of
+potential along the arc itself was abnormally small. In comparing
+these values it is important to remember that Lecher (loc. cit.) has
+shown that with Fe or Pt terminals the arc discharge is intermittent.
+Arons has shown that this is also the case with Hg terminals, but
+no intermittence has been detected with terminals of C, Ag or Cu.
+The preceding measurements refer to mean potentials, and no
+conclusions as to the actual potential differences at any time can be
+drawn when the discharge is discontinuous, unless we know the law
+of discontinuity. The ease with which an arc is sustained depends
+greatly on the nature of the electrodes; when they are brass, zinc,
+cadmium, or magnesium it is exceedingly difficult to get the arc.</p>
+
+<table class="nobctr" summary="Illustration">
+<tr><td class="figcenter" colspan="2"><img style="width:481px; height:223px" src="images/img885.jpg" alt="" /></td></tr>
+<tr><td class="caption sc">Fig. 20.</td>
+ <td class="caption sc">Fig. 21.</td></tr></table>
+
+<p>The potential difference between the terminals is affected by the
+pressure of the gas. The most extensive series of experiments on
+this point is that made by Duncan, Rowland, and Tod (<i>Electrician</i>,
+<span class="pagenum"><a name="page885" id="page885"></a>885</span>
+31, p. 60), whose results are represented in fig. 20. We see from
+these curves that for very short arcs the potential difference increases
+continuously with the pressure, but for longer ones there is a critical
+pressure at which the potential difference is a minimum, and that
+this critical pressure seems to increase with the length of arc.
+The nature of the gas also affects the potential difference. The
+magnitude of this effect may be gathered from the following values
+given by Arons (<i>Ann. der Phys.</i> 1, p. 700) for the potential difference
+required to produce an arc 1.5 mm. long, carrying a current of 4.5
+amperes, between terminals of different metals in air and pure
+nitrogen.</p>
+
+<table class="ws" summary="Contents">
+<tr><td class="tccm allb">Terminal.</td> <td class="tccm allb">Air.</td> <td class="tccm tb bb lb rb2">Nitrogen.</td> <td class="tccm allb">Terminal.</td> <td class="tccm allb">Air.</td> <td class="tccm allb">Nitrogen.</td></tr>
+
+<tr><td class="tcl lb rb">Ag</td> <td class="tcc rb">21</td> <td class="tcc rb2">?</td> <td class="tcl rb">Pt</td> <td class="tcc rb">36</td> <td class="tcc rb">30</td></tr>
+<tr><td class="tcl lb rb">Zn</td> <td class="tcc rb">23</td> <td class="tcc rb2">21</td> <td class="tcl rb">Al</td> <td class="tcc rb">39</td> <td class="tcc rb">27</td></tr>
+<tr><td class="tcl lb rb">Cd</td> <td class="tcc rb">25</td> <td class="tcc rb2">21</td> <td class="tcl rb">Pb</td> <td class="tcc rb">..</td> <td class="tcc rb">18</td></tr>
+<tr><td class="tcl lb rb">Cu</td> <td class="tcc rb">27</td> <td class="tcc rb2">30</td> <td class="tcl rb">Mg</td> <td class="tcc rb">..</td> <td class="tcc rb">22</td></tr>
+<tr><td class="tcl lb rb bb">Fe</td> <td class="tcc rb bb">29</td> <td class="tcc rb2 bb">20</td> <td class="tcl rb bb">&nbsp;</td> <td class="tcc rb bb">&nbsp;</td> <td class="tcc rb bb">&nbsp;</td></tr>
+</table>
+
+<p class="noind">Thus, with the discharge for an arc of given length and current,
+the nature of the terminals is the most important factor in determining
+the potential difference. The effects produced by the pressure
+and nature of the surrounding gas, although quite appreciable, are
+not of so much importance, while in the spark discharge the nature
+of the terminals is of no importance, everything depending upon
+the nature and pressure of the gas.</p>
+
+<p>The potential gradient in the arc is very far from being uniform.
+With carbon terminals Luggin (<i>Wien. Ber.</i> 98, p. 1192) found that,
+with a current of 15 amperes, there was a fall of potential of 33.7
+close to the anode, and one 8.7 close to the cathode, so that the curve
+representing the distribution of potential between the terminals
+would be somewhat like that shown in fig. 21. We have seen that a
+somewhat analogous distribution of potential holds in the case of
+conduction through flames, though in that case the greatest drop of
+potential is in general at the cathode and not at the anode. The
+difference between the changes of potential at the anode and cathode
+is not so large with Fe and Cu terminals as with carbon ones;
+with mercury terminals, Arons (<i>Wied. Ann.</i> 58, p. 73) found the anode
+fall to be 7.4 volts, the cathode fall 5.4 volts.</p>
+</div>
+
+<p>The case of the arc when the cathode is a pool of mercury and
+the anode a metal wire placed in a vessel from which the air has
+been exhausted is one which has attracted much attention, and
+important investigations on this point have been made by
+Hewitt (<i>Electrician</i>, 52, p. 447), Wills (<i>Electrician</i>, 54, p. 26),
+Stark, Retschinsky and Schnaposnikoff (<i>Ann. der Phys.</i> 18,
+p. 213) and Pollak (<i>Ann. der Phys.</i> 19, p. 217). In this arrangement
+the mercury is vaporized by the heat, and the discharge
+which passes through the mercury vapour gives an exceedingly
+bright light, which has been largely used for lighting factories, &amp;c.
+The arrangement can also be used as a rectifier, for a current
+will only pass through it when the mercury pool is the cathode.
+Thus if such a lamp is connected with an alternating current
+circuit, it lets through the current in one direction and stops
+that in the other, thus furnishing a current which is always in
+one direction.</p>
+
+<p><i>Theory of the Arc Discharge.</i>&mdash;An incandescent body such
+as a piece of carbon even when at a temperature far below that
+of the terminals in an arc, emits corpuscles at a rate corresponding
+to a current of the order of 1 ampere per square centimetre of
+incandescent surface, and as the rate of increase of emission
+with the temperature is very rapid, it is probably at the rate
+of many amperes per square centimetre at the temperature of
+the negative carbon in the arc. If then a piece of carbon were
+maintained at this temperature by some external means, and
+used as a cathode, a current could be sent from it to another
+electrode whether the second electrode were cold or hot. If,
+however, these negatively electrified corpuscles did not produce
+other ions either by collision with the gas through which they
+move or with the anode, the spaces between cathode and anode
+would have a negative charge, which would tend to stop the
+corpuscles leaving the cathode and would require a large potential
+difference between anode and cathode to produce any considerable
+current. If, however, there is ionization either in the gas
+or at the anode, the positive ions will diffuse into the region of
+the discharge until they are sensibly equal in number to the
+negative ions. When this is the case the back electromotive
+force is destroyed and the same potential difference will carry
+a much larger current. The arc discharge may be regarded as
+analogous to the discharge between incandescent terminals,
+the only difference being that in the arc the terminals are maintained
+in the state of incandescence by the current and not by
+external means. On this view the cathode is bombarded by
+positive ions which heat it to such a temperature that negative
+corpuscles sufficient to carry the current are emitted by it.
+These corpuscles bombard the anode and keep it incandescent.
+They ionize also, either directly by collision or indirectly by
+heating the anode, the gas and vapour of the metal of which
+the anode is made, and produce in this way the supply of positive
+ions which keep the cathode hot.</p>
+
+<p><i>Discharge from a Point.</i>&mdash;A very interesting case of electric
+discharge is that between a sharply pointed electrode, such as a
+needle, and a metal surface of considerable area. At atmospheric
+pressures the luminosity is confined to the immediate neighbourhood
+of the point. If the sign of the potential of the point does
+not change, the discharge is carried by ions of one sign&mdash;that of
+the charge on the pointed electrode. The velocity of these ions
+under a given potential gradient has been measured by Chattock
+(<i>Phil. Mag.</i> 32, p. 285), and found to agree with that of the ions
+produced by Röntgen or uranium radiation, while Townsend
+(<i>Phil. Trans.</i> 195, p. 259) has shown that the charge on these
+ions is the same as that on the ions streaming from the point.
+If the pointed electrode be placed at right angles to a metal plane
+serving as the other electrode, the discharge takes place when, for
+a given distance of the point from the plane, the potential
+difference between the electrodes exceeds a definite value
+depending upon the pressure and nature of the gas through which
+the discharge passes; its value also depends upon whether,
+beginning with a small potential difference, we gradually increase
+it until discharge commences, or, beginning with a large potential
+difference, we decrease it until the discharge stops. The value
+found by the latter method is less than that by the former.
+According to Chattock&rsquo;s measurements the potential difference V
+for discharge between the point and the plate is given by the
+linear relation V = a + bl, where l is the distance of the point from
+the plate and a and b are constants. From v. Obermayer&rsquo;s
+(<i>Wien. Ber.</i> 100, 2, p. 127) experiments, in which the distance l
+was greater than in Chattock&rsquo;s, it would seem that the potential
+for larger distances does not increase quite so rapidly with l as
+is indicated by Chattock&rsquo;s relation. The potential required to
+produce this discharge is much less than that required to produce
+a spark of length l between parallel plates; thus from Chattock&rsquo;s
+experiments to produce the point discharge when l = .5 cm. in air
+at atmospheric pressure requires a potential difference of about
+3800 volts when the pointed electrode is positive, while to
+produce a spark at the same distance between plane electrodes
+would require a potential difference of about 15,000 volts.
+Chattock showed that with the same pointed electrode the value
+of the electric intensity at the point was the same whatever the
+distance of the point from the plane. The value of the electric
+intensity depended upon the sharpness of the point. When the
+end of the pointed electrode is a hemisphere of radius a, Chattock
+showed that for the same gas at the same pressure the electric
+intensity &int; when discharge takes place is roughly proportioned to
+a<span class="sp">&minus;0.8</span>. The value of the electric intensity at the pointed electrode
+is much greater than its value at a plane electrode for long
+sparks; but we must remember that at a distance from a
+pointed electrode equal to a small multiple of the radius of
+curvature of its extremity the electric intensity falls very far
+<span class="pagenum"><a name="page886" id="page886"></a>886</span>
+below that required to produce discharge in a uniform field, so
+that the discharge from a pointed electrode ought to be compared
+with a spark whose length is comparable with the radius of
+curvature of the point. For such short sparks the electric
+intensity is very high. The electric intensity required to produce
+the discharge from a gas diminishes as the pressure of the gas
+diminishes, but not nearly so rapidly as the electric intensity for
+long sparks. Here again the discharge from a point is comparable
+with short sparks, which, as we have seen, are much less sensitive
+to pressure changes than longer ones. The minimum potential at
+which the electricity streams from the point does not depend
+upon the material of which the point is made; it varies, however,
+considerably with the nature of the gas. The following are the
+results of some experiments on this point. Those in the first two
+columns are due to Röntgen, those in the third and fourth to
+Precht:&mdash;</p>
+
+<table class="ws" summary="Contents">
+<tr><td class="tccm allb" rowspan="2">Gas.</td> <td class="tccm allb" colspan="2">Discharge Potential. Point +.</td>
+ <td class="tccm allb" colspan="2">Pressure 760.</td></tr>
+<tr><td class="tccm allb">Pressure 205.</td> <td class="tccm allb">Pressure 110.</td> <td class="tccm allb">Point +.</td> <td class="tccm allb">Point &minus;.</td></tr>
+
+<tr><td class="tcl lb rb">&nbsp;</td> <td class="tcc rb">Volts.</td> <td class="tcc rb">Volts.</td> <td class="tcc rb">Volts.</td> <td class="tcc rb">Volts.</td></tr>
+<tr><td class="tcl lb rb">H<span class="su">2</span></td> <td class="tcc rb">1296</td> <td class="tcc rb">1174</td> <td class="tcc rb">2125</td> <td class="tcc rb">1550</td></tr>
+<tr><td class="tcl lb rb">O<span class="su">2</span></td> <td class="tcc rb">2402</td> <td class="tcc rb">1975</td> <td class="tcc rb">2800</td> <td class="tcc rb">2350</td></tr>
+<tr><td class="tcl lb rb">CO</td> <td class="tcc rb">2634</td> <td class="tcc rb">2100</td> <td class="tcc rb">..</td> <td class="tcc rb">..</td></tr>
+<tr><td class="tcl lb rb">CH<span class="su">4</span></td> <td class="tcc rb">2777</td> <td class="tcc rb">2317</td> <td class="tcc rb">..</td> <td class="tcc rb">..</td></tr>
+<tr><td class="tcl lb rb">NO</td> <td class="tcc rb">3188</td> <td class="tcc rb">2543</td> <td class="tcc rb">..</td> <td class="tcc rb">..</td></tr>
+<tr><td class="tcl lb rb">CO<span class="su">2</span></td> <td class="tcc rb">3287</td> <td class="tcc rb">2655</td> <td class="tcc rb">3475</td> <td class="tcc rb">2100</td></tr>
+<tr><td class="tcl lb rb">N<span class="su">2</span></td> <td class="tcc rb">..</td> <td class="tcc rb">..</td> <td class="tcc rb">2600</td> <td class="tcc rb">2000</td></tr>
+<tr><td class="tcl lb rb bb">Air</td> <td class="tcc rb bb">..</td> <td class="tcc rb bb">..</td> <td class="tcc rb bb">2750</td> <td class="tcc rb bb">2050</td></tr>
+</table>
+
+<p class="noind">We see from this table that in the case of the discharge from a
+positively electrified point the greater the molecular weight of the
+gas the greater the potential required for discharge. Röntgen
+concluded from his experiments that the discharging potential
+from a positive point in different gases at the same pressure
+varies inversely as the mean free path of the molecules of the gas.
+In the same gas, however, at different pressures the discharging
+potential does not vary so quickly with the pressure as does the
+mean free path. In Precht&rsquo;s experiments, in which different
+gases were used, the variations in the discharging potential
+are not so great as the variations in the mean free path of the
+gases.</p>
+
+<p>The current of electrified air flowing from the point when the
+electricity is escaping&mdash;the well-known &ldquo;electrical wind&rdquo;&mdash;is
+accompanied by a reaction on the point which tends to drive it
+backwards. This reaction has been measured by Arrhenius
+(<i>Wied. Ann.</i> 63, p. 305), who finds that when positive electricity is
+escaping from a point in air the reaction on the point for a given
+current varies inversely as the pressure of the gas, and for
+different gases (air, hydrogen and carbonic acid) inversely as the
+square root of the molecular weight of the gas. The reaction
+when negative electricity is escaping is much less. The proportion
+between the reactions for positive and negative currents depends
+on the pressure of the gas. Thus for equal positive and negative
+currents in air at a pressure of 70 cm. the reaction for a positive
+point was 1.9 times that of a negative one, at 40 cm. pressure
+2.6 times, at 20 cm. pressure 3.2 times, at 10.3 cm. pressure 7
+times, and at 5.1 cm. pressure 15 times the reaction for the
+negative point. Investigation shows that the reaction should
+be proportional to the quotient of the current by the velocity
+acquired by an ion under unit potential gradient. Now this
+velocity is inversely proportional to the pressure, so that the
+reaction should on this view be directly proportional to the
+pressure. This agrees with Arrhenius&rsquo; results when the point is
+positive. Again, the velocities of an ion in hydrogen, air and
+carbonic acid at the same pressure are approximately inversely
+proportional to the square roots of their molecular weights, so
+that the reaction should be directly proportional to this quantity.
+This also agrees with Arrhenius&rsquo; results for the discharge from a
+positive point. The velocity of the negative ion is greater than
+that of a positive one under the same potential gradient, so that
+the reaction for the negative point should be less than that for a
+positive one, but the excess of the positive reaction over the
+negative is much greater than that of the velocity of the negative
+ion over the velocity of the positive. There is, however, reason to
+believe that a considerable condensation takes place around the
+negative ion as a nucleus after it is formed, so that the velocity of
+the negative ion under a given potential gradient will be greater
+immediately after the ion is formed than when it has existed for
+some time. The measurements which have been made of the
+velocities of the ions relate to those which have been some time in
+existence, but a large part of the reaction will be due to the
+newly-formed ions moving with a greater velocity, and thus
+giving a smaller reaction than that calculated from the observed
+velocity.</p>
+
+<p>With a given potential difference between the point and the
+neighbouring conductor the current issuing from the point is
+greater when the point is negative than when it is positive, except
+in oxygen, when it is less. Warburg (<i>Sitz. Akad. d. Wissensch.
+zu Berlin</i>, 1899, 50, p. 770) has shown that the addition of a
+small quantity of oxygen to nitrogen produces a great diminution
+in the current from a negative point, but has very little effect on
+the discharge from a positive point. Thus the removal of a trace
+of oxygen made a leak from a negative point 50 times what it was
+before. Experiments with hydrogen and helium showed that
+impurities in these gases had a great effect on the current when
+the point was negative, and but little when it was positive. This
+suggests that the impurities, by condensing round the negative
+ions as nuclei, seriously diminish their velocity. If a point is
+charged up to a high and rapidly alternating potential, such as
+can be produced by the electric oscillations started when a Leyden
+jar is discharged, then in hydrogen, nitrogen, ammonia and
+carbonic acid gas a conductor placed in the neighbourhood of the
+point gets a negative charge, while in air and oxygen it gets a
+positive one. There are two considerations which are of importance
+in connexion with this effect. The first is the velocity of
+the ions in the electric field, and the second the ease with which
+the ions can give up their charges to the metal point. The greater
+velocity of the negative ions would, if the potential were rapidly
+alternating, cause an excess of negative ions to be left in the
+surrounding gas. This is the case in hydrogen. If, however, the
+metal had a much greater tendency to unite with negative than
+with positive ions, such as we should expect to be the case in
+oxygen, this would act in the opposite direction, and tend to
+leave an excess of positive ions in the gas.</p>
+
+<p><i>The Characteristic Curve for Discharge through Gases.</i>&mdash;When
+a current of electricity passes through a metallic conductor the
+relation between the current and the potential difference is the
+exceedingly simple one expressed by Ohm&rsquo;s law; the current
+is proportional to the potential difference. When the current
+passes through a gas there is no such simple relation. Thus we
+have already mentioned cases where the current increased as the
+potential increased although not in the same proportion, while
+as we have seen in certain stages of the arc discharge the potential
+difference diminishes as the current increases. Thus the problem
+of finding the current which a given battery will produce when
+part of the circuit consists of a gas discharge is much more
+complicated than when the circuit consists entirely of metallic
+conductors. If, however, we measure the potential difference
+between the electrodes in the gas when different currents are
+sent through it, we can plot a curve, called the &ldquo;characteristic
+curve,&rdquo; whose ordinates are the potential differences between
+the electrodes in the gas and the abscissae the corresponding
+currents. By the aid of this curve we can calculate the current
+produced when a given battery is connected up to the gas by
+leads of known resistance.</p>
+
+<div class="condensed">
+<p>For let E<span class="su">0</span> be the electromotive force of the battery, R the resistance
+of the leads, i the current, the potential difference between
+the terms in the gas will be E<span class="su">0</span> &minus; Ri. Let ABC (fig. 22) be the
+&ldquo;characteristic curve,&rdquo; the ordinates being the potential difference
+between the terminals in the gas, and the abscissae the current.
+Draw the line LM whose equation is E = E<span class="su">0</span> &minus; Ri, then the points
+where this line cuts the characteristic curves will give possible
+values of i and E, the current through the discharge tube and the
+potential difference between the terminals. Some of these points
+may, however, correspond to an unstable position and be impossible
+to realize. The following method gives us a criterion by which we
+can distinguish the stable from the unstable positions. If the current
+<span class="pagenum"><a name="page887" id="page887"></a>887</span>
+is increased by &delta;i, the electromotive force which has to be overcome
+by the battery is R&delta;i + (dE/di)&delta;i. If R + dE/di is positive there will
+be an unbalanced electromotive force round the circuit tending to
+stop the current. Thus the increase in the current will be stopped
+and the condition will be a stable one. If, however, R + dE/di is
+negative there will be an unbalanced
+electromotive force
+tending to increase the current
+still further; thus the current
+will go on increasing and the
+condition will be unstable.
+Thus for stability R + dE/di
+must be positive, a condition
+first given by Kaufmann (<i>Ann.
+der Phys.</i> 11, p. 158). The
+geometrical interpretation of
+this condition is that the
+straight line LM must, at the
+point where it cuts the characteristic
+curve, be steeper
+than the tangent to characteristic
+curve. Thus of the points
+ABC where the line cuts the curve in fig. 22, A and C correspond
+to stable states and B to an unstable one. The state of things
+represented by a point P on the characteristic curve when the slope
+is downward cannot be stable unless there is in the external circuit
+a resistance greater than that represented by the tangent of the
+inclination of the tangent to the curve at P to the horizontal axis.</p>
+
+<table class="nobctr" style="float: left; width: 330px;" summary="Illustration">
+<tr><td class="figleft1"><img style="width:279px; height:227px" src="images/img887a.jpg" alt="" /></td></tr>
+<tr><td class="caption sc">Fig. 22.</td></tr></table>
+
+<p>If we keep the external electromotive force the same and gradually
+increase the resistance in the leads, the line LM will become steeper
+and steeper. C will move to the left so that the current will diminish;
+when the line gets so steep that it touches the curve at C&rsquo;, any
+further increase in the resistance will produce an abrupt change in
+the current; for now the state of things represented by a point near
+A&rsquo; is the only stable state. Thus if the BC part of the curve corresponded
+to a luminous discharge and the A part to a dark discharge,
+we see that if the electromotive force is kept constant there is a
+minimum value of the current for the luminous discharge. If the
+current is reduced below this value, the discharge ceases to be
+luminous, and there is an abrupt diminution in the current.</p>
+</div>
+
+<p style="clear: both;"><i>Cathode Rays.</i>&mdash;When the gas in the discharge tube is at a
+very low pressure some remarkable phenomena occur in the
+neighbourhood of the cathode. These seem to have been first
+observed by Plücker (<i>Pogg. Ann.</i> 107, p. 77; 116, p. 45) who
+noticed on the walls of the glass tube near the cathode a greenish
+phosphorescence, which he regarded as due to rays proceeding
+from the cathode, striking against the sides of the tube, and then
+travelling back to the cathode. He found that the action of a
+magnet on these rays was not the same as the action on the
+part of the discharge near the positive electrode. Hittorf (<i>Pogg.
+Ann.</i> 136, p. 8) showed that the agent producing the phosphorescence
+was intercepted by a solid, whether conductor or insulator,
+placed between the cathode and the sides of the tube. He
+regarded the phosphorescence as caused by a motion starting
+from the cathode and travelling in straight lines through the gas.
+Goldstein (<i>Monat. der Berl. Akad.</i>, 1876, p. 24) confirmed this
+discovery of Hittorf&rsquo;s, and further showed that a distinct,
+though not very sharp, shadow is cast by a small object placed
+near a large plane cathode. This is a proof that the rays producing
+the phosphorescence must be emitted almost normally
+from the cathode, and not, like the rays of light from a luminous
+surface, in all directions, for such rays would not produce a
+perceptible shadow if a small body were placed near the plane.
+Goldstein regarded the phosphorescence as due to waves in the
+ether, for whose propagation the gas was not necessary. Crookes
+(<i>Phil. Trans.</i>, 1879, pt. i. p. 135; pt. ii. pp. 587, 661), who made
+many remarkable researches in this subject, took a different
+view. He regarded the rays as streams of negatively electrified
+particles projected normally from the cathode with great velocity,
+and, when the pressure is sufficiently low, reaching the sides of
+the tube, and by their impact producing phosphorescence and
+heat. The rays on this view are deflected by a magnet, because
+a magnet exerts a force on a charged moving body.</p>
+
+<p>These rays striking against glass make it phosphorescent.
+The colour of the phosphorescence depends on the kind of glass;
+thus the light from soda glass is a yellowish green, and that from
+lead glass blue. Many other bodies phosphoresce when exposed
+to these rays, and in particular the phosphorescence of some
+gems, such as rubies and diamonds, is exceedingly vivid. The
+spectrum of the phosphorescent light is generally continuous,
+but Crookes showed that the phosphorescence of some of the
+rare earths, such as yttrium, gives a spectrum of bright bands,
+and he founded on this fact a spectroscopic method of great
+importance. Goldstein (<i>Wied. Ann.</i> 54, p. 371) discovered
+that the haloid salts of the alkali metals change colour under
+the rays, sodium chloride, for example, becoming violet. The
+coloration is a surface one, and has been traced by E. Wiedemann
+and Schmidt (<i>Wied. Ann.</i> 54, p. 618) to the formation of a subchloride.
+Chlorides of tin, mercury and lead also change colour in
+the same way. E. Wiedemann (<i>Wied. Ann.</i> 56, p. 201) discovered
+another remarkable effect, which he called thermo-luminescence;
+he found that many bodies after being exposed to the cathode
+rays possess for some time the power of becoming luminous
+when their temperature is raised to a point far below that at
+which they become luminous in the normal state. Substances
+belonging to the class called by van &rsquo;t Hoff solid solutions exhibit
+this property of thermo-luminescence to a remarkable extent.
+They are formed when two salts, one greatly in excess of the
+other, are simultaneously precipitated from a solution. A trace
+of MnSO<span class="su">4</span> in CaSO<span class="su">4</span> shows very brilliant thermo-luminescence.
+The impact of cathode rays produces after a time perceptible
+changes in the glass. Crookes (<i>Phil. Trans.</i> pt. ii. 1879, p. 645)
+found that after glass has been phosphorescing for some time
+under the cathode rays it seems to get tired, and the phosphorescence
+is not so
+bright as it was
+initially. Thus, for
+example, when the
+shadow of a Maltese
+cross is thrown
+on the walls of the
+tube as in fig. 23,
+if after the discharge
+has been
+going on for some
+time the cross is
+shaken down or a
+new cathode used
+whose line of fire does not cut the cross, the pattern of the cross
+will still be seen on the glass, but it will now be brighter instead
+of darker than the surrounding portion. The portions shielded
+by the cross, not being tired by being made to phosphoresce
+for a long time, respond more vigorously to the stimulus than
+those portions which have not been protected. Skinner (<i>Proc.
+Camb. Phil. Soc.</i> ix. p. 371) and Thomson found on the glass
+which had been exposed to the rays gelatinous filaments, apparently
+silica, resulting from the reduction of the glass. A reducing
+action was also noticed by Villard (<i>Journ. de phys.</i> 3, viii.
+p. 140) and Wehnelt (<i>Wied. Ann.</i> 67, p. 421). It can be well
+shown by letting the rays fall on a plate of oxidized copper,
+when the part struck by the rays will become bright. The
+rays heat bodies on which they fall, and if they are concentrated
+by using as a cathode a portion of a spherical surface, the heat
+at the centre becomes so great that a piece of platinum wire can
+be melted or a diamond charred. Measurements of the heating
+effects of the rays have been made by Thomson (<i>Phil. Mag.</i>
+[5], 44, p. 293) and Cady (<i>Ann. der Phys.</i> 1, p. 678). Crookes
+(<i>Phil. Trans.</i>, 1879, pt. i. p. 152) showed that a vane mounted
+as in a radiometer is set in rotation by the rays, the direction of
+the rotation being the same as would be produced by a stream
+of particles proceeding from the cathode. The movement is
+not due to the momentum imparted to the vanes by the rays,
+but to the difference in temperature between the sides of the
+vanes, the rays making the side against which they strike
+hotter than the other.</p>
+
+<table class="nobctr" style="float: right; width: 410px;" summary="Illustration">
+<tr><td class="figright1"><img style="width:359px; height:227px" src="images/img887b.jpg" alt="" /></td></tr>
+<tr><td class="caption sc">Fig. 22.</td></tr></table>
+
+<p><i>Effect of a Magnet.</i>&mdash;The rays are deflected by a magnet,
+so that the distribution of phosphorescence over the glass and
+the shape and position of the shadows cast by bodies in the tube
+are altered by the proximity of a magnet. The laws of magnetic
+deflection of these rays have been investigated by Plücker (<i>Pogg.</i>
+<span class="pagenum"><a name="page888" id="page888"></a>888</span>
+<i>Ann.</i> 103, p. 88), Hittorf (<i>Pogg. Ann.</i> 136, p. 213), Crookes (<i>Phil.
+Trans.</i>, 1879, pt. 1, p. 557), and Schuster (<i>Proc. Roy. Soc.</i> 47, p.
+526). The deflection is the same as that of negatively electrified
+particles travelling along the path of the rays. Such particles
+would in a magnetic field be acted on by a force at right angles
+to the direction of motion of the particle and also to the magnetic
+force, the magnitude of the force being proportional to the
+product of the velocity of the particle, the magnetic force, and
+the sine of the angle between these vectors. In this case we have
+seen that if the particle is not acted on by an electrostatic field,
+the path in a uniform magnetic field is a spiral, which, if the
+magnetic force is at right angles to the direction of projection
+of the particle, becomes a circle in the plane at right angles to
+the magnetic force, the radius being mv/He, where m, v, e are
+respectively the mass, velocity and charge on the particle, and
+H is the magnetic force. The smaller the difference of potential
+between the electrodes of the discharge tube the greater the
+deflection produced by a magnetic field of given strength, and as
+the difference of potential rapidly increases with diminution of
+pressure, after a certain pressure has been passed, the higher
+the exhaustion of the tube the less the magnetic deflection of
+the rays. Birkeland (<i>Comptes rendus</i>, 1896, p. 492) has shown
+that when the discharge is from an induction coil the cathode
+rays produced in the tube at any one time are not equally
+deflected by a magnet, but that a narrow patch of phosphorescence
+when deflected by a magnet is split up into several distinct
+patches, giving rise to what Birkeland calls the &ldquo;magnetic
+spectrum.&rdquo; Strutt (<i>Phil. Mag.</i> 48, p. 478) has shown that this
+magnetic spectrum does not occur if the discharge of a large
+number of cells is employed instead of the coil. Thomson (<i>Proc.
+Camb. Phil. Soc.</i> 9, p. 243) has shown that if the potential
+difference between the electrodes is kept the same the magnetic
+deflection is independent of the nature of the gas filling the
+discharge tube; this was tested with gases so different as air,
+hydrogen, carbonic acid and methyl iodide.</p>
+
+<p><i>Charge of Negative Electricity carried by the Rays.</i>&mdash;We have
+seen that the rays are deflected by a magnet, as if they were
+particles charged with negative electricity. Perrin (<i>Comptes
+rendus</i>, 121, p. 1130) showed by direct experiment that a stream
+of negative electricity is associated with the rays. A modification
+made by Thomson of Perrin&rsquo;s experiment is sketched in
+fig. 24 (<i>Phil. Mag.</i> 48, p. 478).</p>
+
+<table class="nobctr" style="float: left; width: 340px;" summary="Illustration">
+<tr><td class="figleft1"><img style="width:291px; height:359px" src="images/img888.jpg" alt="" /></td></tr>
+<tr><td class="caption sc">Fig. 24.</td></tr></table>
+
+<div class="condensed">
+<p>The rays start from the cathode A, and pass through a slit in a
+solid brass rod B fitting tightly into the neck of the tube. This
+rod is connected with earth and used as the anode. The rays after
+passing through the slit travel through the vessel C. D and E are
+two insulated metal cylinders
+insulated from each other,
+and each having a slit cut in
+its face so as to enable the
+rays to pass into the inside of
+the inner cylinder, which is
+connected with an electrometer,
+the outer cylinder
+being connected with the
+earth. The two cylinders are
+placed on the far side of the
+vessel, but out of the direct
+line of fire of the rays. When
+the rays go straight through
+the slit there is only a very
+small negative charge communicated
+to the inner
+cylinder, but when they are
+deflected by a magnet so that
+the phosphorescent patch falls
+on the slit in the outer
+cylinder the inner cylinder
+receives a very large negative
+charge, the increase coinciding
+very sharply with the appearance of the phosphorescent patch on the
+slit. When the patch is so much deflected by the magnet that it
+falls below the slit, the negative charge in the cylinder again disappears.
+This experiment shows that the cathode rays are accompanied
+by a stream of negative electrification. The same apparatus
+can be used to show that the passage of cathode rays through a
+gas makes it a conductor of electricity. For if the induction coil is
+kept running and a stream of the rays kept steadily going into the
+inner cylinder, the potential of the inner cylinder reaches a definite
+negative value below which it does not fall, however long the rays
+may be kept going. The cylinder reaches a steady state in which
+the gain of negative electricity from the cathode rays is equal to the
+loss by leakage through the conducting gas, the conductivity being
+produced by the passage of the rays through it. If the inner cylinder
+is charged up initially with a greater negative charge than corresponds
+to the steady state, on turning the rays on to the cylinder the negative
+charge will decrease and not increase until it reaches the steady
+state. The conductivity produced by the passage of cathode rays
+through a gas diminishes rapidly with the pressure. When rays
+pass through a gas at a low pressure, they are deflected by an electric
+field; when the pressure of the gas is higher the conductivity it
+acquires when the cathode rays pass through it is so large that the
+potential gradient cannot reach a sufficiently high value to produce
+an appreciable deflection.</p>
+</div>
+
+<p>Thus the cathode rays carry a charge of negative electricity;
+the experiment described on page 875 (fig. 13) shows that they
+are deflected by an electric field as if they were negatively
+electrified, and are acted on by a magnetic force in just the way
+this force would act on a negatively electrified body moving
+along the path of the rays. There is therefore every reason for
+believing that they are charges of negative electricity in rapid
+motion. By measuring the deflection produced by magnetic
+and electric fields we can determine the velocity with which
+these particles moved and the ratio of the mass of the particle
+to the charge carried by it.</p>
+
+<p>We may conclude from the experiments that the value of m/e
+for the particles constituting the cathode rays is of the order
+1/1.7&times;10<span class="sp">7</span>, and we have seen that m/e has the same value in
+all the other cases of negative ions in a gas at low pressure for
+which it has been measured&mdash;viz. for the ions produced when
+ultra-violet light falls on a metal plate, or when an incandescent
+carbon filament is surrounded by a gas at a low pressure, and
+for the &beta; particles given out by radio-active bodies. We have
+also seen that the value of the charge on the gaseous ion, in all
+cases in which it has been measured&mdash;viz. the ions produced by
+Röntgen and uranium radiation, by ultra-violet light, and by the
+discharge of electrification from a point&mdash;is the same in magnitude
+as the charge carried by the hydrogen atom in the electrolysis
+of solutions. The mass of the hydrogen alone is, however,
+10<span class="sp">-4</span> times this charge, while the mass of the carriers of negative
+electrification is only 1/1.7&times;10<span class="sp">7</span> times the charge; hence the
+mass of the carriers of the negative electrification is only <span class="spp">1</span>&frasl;<span class="suu">1700</span> of
+the mass of the hydrogen atom. We are thus, by the study of the
+electric discharge, forced to recognize the existence of masses
+very much smaller than the smallest mass hitherto recognized.</p>
+
+<div class="condensed">
+<p>Direct determinations of the velocity of the cathode rays have
+been made by J. J. Thomson (<i>Phil. Mag.</i> 38, p. 358), who measured
+the interval between the appearance of phosphorescence on two
+pieces of glass placed at a known distance apart, and by Maiorana
+(<i>Nuovo Cimento</i>, 4, 6, p. 336) and Battelli and Stefanini (<i>Phys. Zeit.</i>
+1, p. 51), who measured the interval between the arrival of the
+negative charge carried by the rays at two places separated by a
+known distance. The values of the velocity got in this way are much
+smaller than the values got by the indirect methods previously
+described: thus J. J. Thomson at a fairly high pressure found the
+velocity to be 2&times;10<span class="sp">7</span> cm./sec. Maiorana found values ranging
+between 10<span class="sp">7</span> and 6&times;10<span class="sp">7</span> cm./sec, and Battelli and Stefanini values
+ranging from 6&times;10<span class="sp">6</span> to 1.2&times;10<span class="sp">7</span>. In these methods it is very
+difficult to eliminate the effect of the interval which elapses between
+the arrival of the rays and the attainment by the means of detection,
+such as the phosphorescence of the glass or the deflection of the
+electrometer, of sufficient intensity to affect the senses.</p>
+</div>
+
+<table class="nobctr" style="float: left; width: 260px;" summary="Illustration">
+<tr><td class="figleft1"><img style="width:213px; height:89px" src="images/img889a.jpg" alt="" /></td></tr>
+<tr><td class="caption sc">Fig. 25.</td></tr></table>
+
+<p><i>Transmission of Cathode Rays through Solids&mdash;Lenard Rays.</i>&mdash;It
+was for a long time believed that all solids were absolutely
+opaque to these rays, as Crookes and Goldstein had proved that
+very thin glass, and even a film of collodion, cast intensely black
+shadows. Hertz (<i>Wied. Ann.</i> 45, p. 28), however, showed that
+behind a piece of gold-leaf or aluminium foil an appreciable
+amount of phosphorescence occurred on the glass, and that the
+phosphorescence moved when a magnet was brought near. A
+most important advance was next made by Lenard (<i>Wied. Ann.</i>
+51, p. 225), who got the cathode rays to pass from the
+inside of a discharge tube to the air outside. For this purpose he
+used a tube like that shown in fig. 25. The cathode K is an
+aluminium disc 1.2 cm. in diameter fastened to a stiff wire, which
+is surrounded by a glass tube. The anode A is a brass strip partly
+<span class="pagenum"><a name="page889" id="page889"></a>889</span>
+surrounding the cathode. The end of the tube in front of the
+cathode is closed by a strong metal cap, fastened in with marine
+glue, in the middle of which a hole 1.7 mm. in diameter is bored,
+and covered with a piece of very thin aluminium foil about
+.0026 mm. in thickness. The aluminium window is in metallic
+contact with the cap, and this and the anode are connected with
+the earth. The tube is then exhausted until the cathode rays
+strike against the window. Diffuse light spreads from the
+window into the air outside the tube, and can be traced in a dark
+room for a distance of several centimetres. From the window,
+too, proceed rays which, like the cathode rays, can produce
+phosphorescence, for certain bodies phosphoresce when placed
+in the neighbourhood of the window. This effect is conveniently
+observed by the platino-cryanide screens used to detect Röntgen
+radiation. The properties of the rays outside the tube resemble
+in all respects those of cathode rays;
+they are deflected by a magnet and
+by an electric field, they ionize the
+gas through which they pass and make
+it a conductor of electricity, and they
+affect a photographic plate and change
+the colour of the haloid salts of
+the alkali metals. As, however, it is convenient to distinguish
+between cathode rays outside and inside the tube, we shall call
+the former Lenard rays. In air at atmospheric pressure the
+Lenard rays spread out very diffusely. If the aluminium
+window, instead of opening into the air, opens into another tube
+which can be exhausted, it is found that the lower the pressure of
+the gas in this tube the farther the rays travel and the less diffuse
+they are. By filling the tube with different gases Lenard showed
+that the greater the density of the gas the greater is the absorption
+of these rays. Thus they travel farther in hydrogen than in
+any other gas at the same pressure. Lenard showed, too, that if
+he adjusted the pressure so that the density of the gas in this tube
+was the same&mdash;if, for example, the pressure when the tube was
+filled with oxygen was <span class="spp">1</span>&frasl;<span class="suu">16</span> of the pressure when it was filled with
+hydrogen&mdash;the absorption was constant whatever the nature of
+the gas. Becker (<i>Ann. der Phys.</i> 17, p. 381) has shown that this
+law is only approximately true, the absorption by hydrogen
+being abnormally large, and by the inert monatomic gases, such
+as helium and argon, abnormally small. The distance to which
+the Lenard rays penetrate into this tube depends upon the
+pressure in the discharge tube; if the exhaustion in the latter is
+very high, so that there is a large potential difference between
+the cathode and the anode, and therefore a high velocity for the
+cathode rays, the Lenard rays will penetrate farther than when
+the pressure in the discharge tube is higher and the velocity of the
+cathode rays smaller. Lenard showed that the greater the
+penetrating power of his rays the smaller was their magnetic
+deflection, and therefore the greater their velocity; thus the
+greater the velocity of the cathode rays the greater is the velocity
+of the Lenard rays to which they give rise. For very slow
+cathode rays the absorption by different gases departs altogether
+from the density law, so much so that the absorption of these rays
+by hydrogen is greater than that by air (Lenard, <i>Ann. der Phys.</i>
+12, p. 732). Lenard (<i>Wied. Ann.</i> 56, p. 255) studied the passage of
+his rays through solids as well as through gases, and arrived at
+the very interesting result that the absorption of a substance
+depends only upon its density, and not upon its chemical composition
+or physical state; in other words, the amount of
+absorption of the rays when they traverse a given distance
+depends only on the quantity of matter they cut through in the
+distance. McClelland (<i>Proc. Roy. Soc.</i> 61, p. 227) showed that
+the rays carry a charge of negative electricity, and M&rsquo;Lennan
+measured the amount of ionization rays of given intensity
+produced in different gases, finding that if the pressure is adjusted
+so that the density of the different gases is the same the number
+of ions per cubic centimetre is also the same. In this case, as
+Lenard has shown, the absorption is the same, so that with the
+Lenard rays, as with uranium and probably with Röntgen
+rays, equal absorption corresponds to equal ionization. A
+convenient method for producing Lenard rays of great
+intensity has been described by Des Coudres (<i>Wied. Ann.</i>
+62, p. 134).</p>
+
+<p><i>Diffuse Reflection of Cathode Rays.</i>&mdash;When cathode rays fall
+upon a surface, whether of an insulator or a conductor, cathode
+rays start from the surface in all directions. This phenomenon,
+which was discovered by Goldstein (<i>Wied. Ann.</i> 62, p. 134), has
+been investigated by Starke (<i>Wied. Ann.</i> 66, p. 49; <i>Ann. der
+Phys.</i> 111, p. 75), Austin and Starke (<i>Ann. der Phys.</i> 9, p. 271),
+Campbell-Swinton (<i>Proc. Roy. Soc.</i> 64, p. 377), Merritt (<i>Phys.
+Rev.</i> 7, p. 217) and Gehrcke (<i>Ann. der Phys.</i> 8, p. 81); it is often
+regarded as analogous to the diffuse reflection of light from such
+a surface as gypsum, and is spoken of as the diffuse reflection of
+the cathode rays. According to Merritt and Austin and Starke
+the deviation in a magnetic field of these reflected rays is the same
+as that of the incident rays. The experiments, however, were
+confined to rays reflected so that the angle of reflection was
+nearly equal to that of incidence. Gehrcke showed that among
+the reflected rays there were a large number which had a much
+smaller velocity than the incident ones. According to Campbell-Swinton
+the &ldquo;diffuse&rdquo; reflection is accompanied by a certain
+amount of &ldquo;specular&rdquo; reflection. Lenard, who used slower
+cathode rays than Austin and Starke, could not detect in the
+scattered rays any with velocities comparable with that of the
+incident rays; he obtained copious supplies of slow rays whose
+speed did not depend on the angle of incidence of the primary
+rays (<i>Ann. der Phys.</i> 15, p. 485). When the angle of incidence
+is very oblique the surface struck by the rays gets positively
+charged, showing that the secondary rays are more numerous
+than the primary.</p>
+
+<p><i>Repulsion of two Cathode Streams.</i>&mdash;Goldstein discovered that
+if in a tube there are two cathodes connected together, the
+cathodic rays from one cathode are deflected when they pass
+near the other. Experiments bearing on this subject have been
+made by Crookes and Wiedemann and Ebert. The phenomena
+may be described by saying that the repulsion of the rays from
+a cathode A by a cathode B is only appreciable when the rays
+from A pass through the Crookes dark space round B. This is
+what we should expect if we remember that the electric field in
+the dark space is far stronger than in the rest of the discharge,
+and that the gas in the other parts of the tube is rendered a
+conductor by the passage through it of the cathode rays, and
+therefore incapable of transmitting electrostatic repulsion.</p>
+
+<p>Scattering of the Negative Electrodes.&mdash;In addition to the
+cathode rays, portions of metal start normally from the cathode
+and form a metallic deposit on the walls of the tube. The
+amount of this deposit varies very much with the metal. Crookes
+(<i>Proc. Roy. Soc.</i> 50, p. 88) found that the quantities of metal
+torn from electrodes of the same size, in equal times, by the
+same current, are in the order Pd, Au, Ag, Pb, Sn, Pt, Cu, Cd,
+Ni, In, Fe.... In air there is very little deposit from an Al
+cathode, but it is abundant in tubes filled with the monatomic
+gases, mercury vapour, argon or helium. The scattering
+increases as the density of the gas diminishes. The particles
+of metal are at low pressures deflected by a magnet, though not
+nearly to the same extent as the cathode rays. According to
+Grandquist, the loss of weight of the cathode in a given time is
+proportional to the square of the current; it is therefore not,
+like the loss of the cathode in ordinary electrolysis, proportional
+to the quantity of current which passes through it.</p>
+
+<table class="nobctr" style="float: right; width: 330px;" summary="Illustration">
+<tr><td class="figright1"><img style="width:277px; height:138px" src="images/img889b.jpg" alt="" /></td></tr>
+<tr><td class="caption sc">Fig. 26.</td></tr></table>
+
+<p><i>Positive Rays or &ldquo;Canalstrahlen.&rdquo;</i>&mdash;Goldstein (<i>Berl. Sitzungsb.</i>
+39, p. 691) found that with a perforated cathode certain
+rays occurred behind the
+cathode which were not
+appreciably deflected by a
+magnet; these he called
+Canalstrahlen, but we shall,
+for reasons which will appear
+later, call them &ldquo;positive
+rays.&rdquo;</p>
+
+<p>Their appearance is well
+shown in fig. 26, taken from a paper by Wehnelt (<i>Wied. Ann.</i>
+67, p. 421) in which they are represented at B. Goldstein found
+<span class="pagenum"><a name="page890" id="page890"></a>890</span>
+that their colour depends on the gas in which they are
+formed, being gold-colour in air and nitrogen, rose-colour in
+hydrogen, yellowish rose in oxygen, and greenish gray in
+carbonic acid.</p>
+
+<p>The colour of the luminosity due to <span class="correction" title="amended from postive">positive</span> rays is not in
+general the same as that due to anode rays; the difference is
+exceptionally well marked in helium, where the cathode ray
+luminosity is blue while that due to the positive rays is red.
+The luminosity produced when the rays strike against solids
+is also quite distinct. The cathode rays make the body emit
+a continuous spectrum, while the spectrum produced by the
+positive rays often shows bright lines. Thus lithium chloride
+under cathode rays gives out a steely blue light and the spectrum
+is continuous, while under the positive rays the salt gives out a
+brilliant red light and the spectrum shows the red helium line.
+It is remarkable that the lines on the spectra of the alkali metals
+are much more easily produced when the positive rays fall on
+the oxide of the metal than when they fall on the metal itself.
+Thus when the positive rays fall on a pool of the liquid alloy
+of sodium and potassium the specks of oxide on the surface
+shine with a bright yellow light while the untarnished part of
+the surface is quite dark.</p>
+
+<p>W. Wien (<i>Wied. Ann.</i> 65, p. 445) measured the values of
+e/m for the particles forming the positive rays. Other measurements
+have been made by Ewers (<i>Wied. Ann.</i> 69, p. 167) and
+J. J. Thomson (<i>Phil. Mag.</i> 13, p. 561). The differences between
+the values of e/m for the cathode and positive rays are very
+remarkable. For cathode rays whose velocity does not approach
+that of light, e/m is always equal to 1.7&times;10<span class="sp">8</span>, while for the positive
+rays the greatest value of this quantity yet observed is 10<span class="sp">4</span>,
+which is also the value of e/m for the hydrogen ions in the electrolysis
+of dilute solutions. In some experiments made by J. J.
+Thomson (<i>Phil. Mag.</i>, 14, p. 359) it was found that when the
+pressure of the gas was not too low the bright spot produced by
+the impact of a pencil of these rays on a phosphorescent screen
+is deflected by electric and magnetic forces into a continuous
+band extending on both sides of the undeflected position. The
+portion on one side is in general much fainter than that on the
+other. The direction of this deflection shows that it is produced
+by particles charged with negative electricity, while the brighter
+band is due to particles charged with positive electricity. The
+negatively electrified particles which produce the band c.c are
+not corpuscles, for from the electric and magnetic deflections
+we can find the value of e/m. As this proves to be equal to 10<span class="sp">4</span>,
+we see that the mass of the carrier of the negative charge is
+comparable with that of an atom, and so very much greater
+than that of a corpuscle. At very low pressures part of the
+phosphorescence disappears, while the upper portion breaks up
+into two patches (fig. 27). For one of these the maximum value
+of e/m is 10<span class="sp">4</span> and for the other 5&times;10<span class="sp">3</span>. At low pressures the
+appearance of the patches and the values of e/m are the same
+whether the tube is filled originally with air, hydrogen or
+helium. In some of the experiments the tube was exhausted
+until the pressure was too low to allow the discharge to pass.
+A very small quantity of the gas under investigation was then
+admitted into the tube, just sufficient to allow the discharge to
+pass, and the deflection of the phosphorescent patch measured.
+The following gases were admitted into the tube, air, carbonic
+oxide, oxygen, hydrogen, helium, argon and neon, but whatever
+the gas the appearance of the phosphorescence was the same;
+in every case there were two patches, for one of which e/m = 10<span class="sp">4</span>
+and for the other e/m = 5&times;10<span class="sp">3</span>. In helium at higher pressures
+another patch was observed, for which e/m = 2.5&times;10<span class="sp">8</span>. The
+continuous band into which the phosphorescent spot is drawn
+out when the pressure is not exceedingly low, which involves
+the existence of particles for which the mean value of e/m varies
+from zero to 10<span class="sp">4</span>, can be explained as follows. The rays on their
+way to the phosphorescent screen have to pass through gas
+which is ionized by the passage through it of the positive rays;
+this gas will therefore contain free corpuscles. The particles
+which constitute the rays start with a charge of positive electricity.
+Some of these particles in their journey through the
+gas attract a corpuscle whose negative charge neutralizes the
+positive charge on the particle. The particles when in this
+neutral state may be ionized by collision and reacquire a positive
+charge, or by attracting another particle may become negatively
+charged, and this process may be repeated several times on their
+journey to the phosphorescent screen. Thus some of the particles,
+instead of being positively charged for the whole of the time
+they are exposed to the electric and magnetic forces, may be
+for a part of that time without a charge or even have a negative
+charge. The deflection of a particle is proportional to the
+average value of its charge whilst under the influence of the
+deflecting forces. Thus if a particle is without a charge for a
+part of the time, its deflection will be less than that of a particle
+which has retained its positive charge for the whole of its journey,
+while the few particles which have a negative charge for a
+longer time than they have a positive will be deflected in the
+opposite direction to the main portion and will produce the tail
+(fig. 27).</p>
+
+<table class="nobctr" summary="Illustration">
+<tr><td class="figcenter"><img style="width:488px; height:76px" src="images/img890.jpg" alt="" /></td></tr>
+<tr><td class="caption sc">Fig. 27.</td></tr></table>
+
+<p>A similar explanation will apply to the positive rays discovered
+by Villard (<i>Comptes rendus</i>, 143, p. 674) and J. J. Thomson
+(<i>Phil. Mag.</i> 13, p. 359), which travel in the opposite direction to
+the rays we have been considering, <i>i.e.</i> they travel away from the
+cathode and in the direction of the cathode&rsquo;s rays; these rays
+are sometimes called &ldquo;retrograde&rdquo; rays. These as far as has
+been observed have always the same maximum value of e/m,
+<i>i.e.</i> 10<span class="sp">4</span>, and there are a considerable number of negative ones
+always mixed with them. The maximum velocity of both the
+positive and retrograde rays is about 2&times;10<span class="sp">8</span> cm./sec. and varies
+very little with the potential difference between the electrodes
+in the tube in which they are produced (J. J. Thomson, <i>Phil.
+Mag.</i>, Dec. 1909).</p>
+
+<p>The positive rays show, when the pressure is not very low, the
+line spectrum of the gas through which they pass. An exceedingly
+valuable set of observations on this point have been made
+by Stark and his pupils (<i>Physik. Zeit.</i> 6, p. 892; <i>Ann. der
+Phys.</i> 21, pp. 40, 457). Stark has shown that in many gases,
+notably hydrogen, the spectrum shows the Doppler effect, and he
+has been able to calculate in this way the velocity of the positive
+rays.</p>
+
+<p><i>Anode Rays.</i>&mdash;Gehrcke and Reichenhein (<i>Ann. der Phys.</i> 25,
+p. 861) have found that when the anode consists of a <span class="correction" title="amended from mixure">mixture</span> of
+sodium and lithium chloride raised to a high temperature either
+by the discharge itself or by an independent heating circuit, very
+conspicuous rays come from the anode when the pressure of the
+gas in the discharge tube is very low, and a large coil is used to
+produce the discharge. The determination of e/m for these rays
+showed that they are positively charged atoms of sodium or
+lithium, moving with very considerable velocity; in some of
+Gehrcke&rsquo;s experiments the maximum velocity was as great as
+1.8&times;10<span class="sp">7</span> cm./sec. though the average was about 10<span class="sp">7</span> cm./sec.
+These velocities are less than those of the positive rays whose
+maximum velocity is about 2&times;10<span class="sp">8</span> cm./sec.</p>
+<div class="author">(J. J. T.)</div>
+
+<hr class="foot" />
+<div class="note">
+
+<p><a name="Footnote_1" id="Footnote_1" href="#FnAnchor_1"><span class="fn">1</span></a> The values for nickel and bismuth given in the table are much
+higher than later values obtained with pure electrolytic nickel and
+bismuth.</p>
+
+<p><a name="Footnote_2" id="Footnote_2" href="#FnAnchor_2"><span class="fn">2</span></a> The value here given, namely 12.885, for the electric mass-resistivity
+of liquid mercury as determined by Matthiessen is now
+known to be too high by nearly 1%. The value at present accepted
+is 12.789 ohms per metre-gramme at 0° C.</p>
+
+<p><a name="Footnote_3" id="Footnote_3" href="#FnAnchor_3"><span class="fn">3</span></a> The value (1630) here given for hard-drawn copper is about
+¼% higher than the value now adopted, namely, 1626. The difference
+is due to the fact that either Jenkin or Matthiessen did not
+employ precisely the value at present employed for the density of
+hard-drawn and annealed copper in calculating the volume-resistivities
+from the mass-resistivities.</p>
+
+<p><a name="Footnote_4" id="Footnote_4" href="#FnAnchor_4"><span class="fn">4</span></a> Matthiessen&rsquo;s value for nickel is much greater than that obtained
+in more recent researches. (See Matthiessen and Vogt, <i>Phil. Trans.</i>,
+1863, and J. A. Fleming, <i>Proc. Roy. Soc.</i>, December 1899.)</p>
+
+<p><a name="Footnote_5" id="Footnote_5" href="#FnAnchor_5"><span class="fn">5</span></a> Matthiessen&rsquo;s value for mercury is nearly 1% greater than the
+value adopted at present as the mean of the best results, namely
+94,070.</p>
+
+<p><a name="Footnote_6" id="Footnote_6" href="#FnAnchor_6"><span class="fn">6</span></a> The samples of silver, copper and nickel employed for these tests
+were prepared electrolytically by Sir J. W. Swan, and were exceedingly
+pure and soft. The value for volume-resistivity of nickel as
+given in the above table (from experiments by J. A. Fleming, <i>Proc.
+Roy. Soc.</i>, December 1899) is much less (nearly 40%) than the value
+given by Matthiessen&rsquo;s researches.</p>
+
+<p><a name="Footnote_7" id="Footnote_7" href="#FnAnchor_7"><span class="fn">7</span></a> The electrolytic bismuth here used was prepared by Hartmann
+and Braun, and the resistivity taken by J. A. Fleming. The value
+is nearly 20% less than that given by Matthiessen.</p>
+
+<p><a name="Footnote_8" id="Footnote_8" href="#FnAnchor_8"><span class="fn">8</span></a> In 1899 a committee was formed of representatives from eight
+of the leading manufacturers of insulated copper cables with delegates
+from the Post Office and Institution of Electrical Engineers, to
+consider the question of the values to be assigned to the resistivity
+of hard-drawn and annealed copper. The sittings of the committee
+were held in London, the secretary being A. H. Howard. The values
+given in the above paragraphs are in accordance with the decision
+of this committee, and its recommendations have been accepted by
+the General Post Office and the leading manufacturers of insulated
+copper wire and cables.</p>
+
+<p><a name="Footnote_9" id="Footnote_9" href="#FnAnchor_9"><span class="fn">9</span></a> Platinoid is an alloy introduced by Martino, said to be similar
+in composition to German silver, but with a little tungsten added.
+It varies a good deal in composition according to manufacture, and
+the resistivity of different specimens is not identical. Its electric properties
+were first made known by J. T. Bottomley, in a paper read
+at the Royal Society, May 5, 1885.</p>
+
+<p><a name="Footnote_10" id="Footnote_10" href="#FnAnchor_10"><span class="fn">10</span></a> An equivalent gramme molecule is a weight in grammes equal
+numerically to the chemical equivalent of the salt. For instance, one
+equivalent gramme molecule of sodium chloride is a mass of 58.5
+grammes. NaCl = 58.5.</p>
+
+<p><a name="Footnote_11" id="Footnote_11" href="#FnAnchor_11"><span class="fn">11</span></a> F. Kohlrausch and L. Holborn, <i>Das Leitvermögen der Elektrolyte</i>
+(Leipzig, 1898).</p>
+
+<p><a name="Footnote_12" id="Footnote_12" href="#FnAnchor_12"><span class="fn">12</span></a> It should be noticed that the velocities calculated in Kohlrausch&rsquo;s
+theory and observed experimentally are the average velocities, and
+involve both the factors mentioned above; they include the time
+wasted by the ions in combination with each other, and, except at
+great dilution, are less than the velocity with which the ions move
+when free from each other.</p>
+
+</div>
+
+
+<hr class="art" />
+
+
+
+
+
+
+
+
+
+<pre>
+
+
+
+
+
+End of the Project Gutenberg EBook of Encyclopaedia Britannica, 11th
+Edition, Volume 6, Slice 8, by Various
+
+*** END OF THIS PROJECT GUTENBERG EBOOK ENCYC. BRITANNICA, VOL 6, SL 8 ***
+
+***** This file should be named 32063-h.htm or 32063-h.zip *****
+This and all associated files of various formats will be found in:
+ https://www.gutenberg.org/3/2/0/6/32063/
+
+Produced by Marius Masi, Don Kretz, Juliet Sutherland, and
+the Online Distributed Proofreading Team at
+https://www.pgdp.net
+
+
+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
+https://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
+
+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 F3. 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 MERCHANTIBILITY 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 https://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
+https://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 https://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 https://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 including checks, online payments and credit card
+donations. To donate, please visit: https://pglaf.org/donate
+
+
+Section 5. General Information About Project Gutenberg-tm electronic
+works.
+
+Professor Michael S. Hart was 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:
+
+ https://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.
+
+
+</pre>
+
+</body>
+</html>
diff --git a/32063-h/images/img861a.jpg b/32063-h/images/img861a.jpg
new file mode 100644
index 0000000..63445df
--- /dev/null
+++ b/32063-h/images/img861a.jpg
Binary files differ
diff --git a/32063-h/images/img861b.jpg b/32063-h/images/img861b.jpg
new file mode 100644
index 0000000..6fd4332
--- /dev/null
+++ b/32063-h/images/img861b.jpg
Binary files differ
diff --git a/32063-h/images/img862.jpg b/32063-h/images/img862.jpg
new file mode 100644
index 0000000..26e63da
--- /dev/null
+++ b/32063-h/images/img862.jpg
Binary files differ
diff --git a/32063-h/images/img865.jpg b/32063-h/images/img865.jpg
new file mode 100644
index 0000000..018e005
--- /dev/null
+++ b/32063-h/images/img865.jpg
Binary files differ
diff --git a/32063-h/images/img866a.jpg b/32063-h/images/img866a.jpg
new file mode 100644
index 0000000..ddff61e
--- /dev/null
+++ b/32063-h/images/img866a.jpg
Binary files differ
diff --git a/32063-h/images/img866b.jpg b/32063-h/images/img866b.jpg
new file mode 100644
index 0000000..259a142
--- /dev/null
+++ b/32063-h/images/img866b.jpg
Binary files differ
diff --git a/32063-h/images/img868a.jpg b/32063-h/images/img868a.jpg
new file mode 100644
index 0000000..720488c
--- /dev/null
+++ b/32063-h/images/img868a.jpg
Binary files differ
diff --git a/32063-h/images/img868b.jpg b/32063-h/images/img868b.jpg
new file mode 100644
index 0000000..41d85c9
--- /dev/null
+++ b/32063-h/images/img868b.jpg
Binary files differ
diff --git a/32063-h/images/img868c.jpg b/32063-h/images/img868c.jpg
new file mode 100644
index 0000000..d7bb486
--- /dev/null
+++ b/32063-h/images/img868c.jpg
Binary files differ
diff --git a/32063-h/images/img871.jpg b/32063-h/images/img871.jpg
new file mode 100644
index 0000000..cb7926c
--- /dev/null
+++ b/32063-h/images/img871.jpg
Binary files differ
diff --git a/32063-h/images/img872.jpg b/32063-h/images/img872.jpg
new file mode 100644
index 0000000..629b8eb
--- /dev/null
+++ b/32063-h/images/img872.jpg
Binary files differ
diff --git a/32063-h/images/img875.jpg b/32063-h/images/img875.jpg
new file mode 100644
index 0000000..ffefda1
--- /dev/null
+++ b/32063-h/images/img875.jpg
Binary files differ
diff --git a/32063-h/images/img877.jpg b/32063-h/images/img877.jpg
new file mode 100644
index 0000000..5dcafeb
--- /dev/null
+++ b/32063-h/images/img877.jpg
Binary files differ
diff --git a/32063-h/images/img881a.jpg b/32063-h/images/img881a.jpg
new file mode 100644
index 0000000..19ee913
--- /dev/null
+++ b/32063-h/images/img881a.jpg
Binary files differ
diff --git a/32063-h/images/img881b.jpg b/32063-h/images/img881b.jpg
new file mode 100644
index 0000000..aa14664
--- /dev/null
+++ b/32063-h/images/img881b.jpg
Binary files differ
diff --git a/32063-h/images/img883.jpg b/32063-h/images/img883.jpg
new file mode 100644
index 0000000..8599cec
--- /dev/null
+++ b/32063-h/images/img883.jpg
Binary files differ
diff --git a/32063-h/images/img884a.jpg b/32063-h/images/img884a.jpg
new file mode 100644
index 0000000..7fdb392
--- /dev/null
+++ b/32063-h/images/img884a.jpg
Binary files differ
diff --git a/32063-h/images/img884b.jpg b/32063-h/images/img884b.jpg
new file mode 100644
index 0000000..b8935f6
--- /dev/null
+++ b/32063-h/images/img884b.jpg
Binary files differ
diff --git a/32063-h/images/img885.jpg b/32063-h/images/img885.jpg
new file mode 100644
index 0000000..c6c2e12
--- /dev/null
+++ b/32063-h/images/img885.jpg
Binary files differ
diff --git a/32063-h/images/img887a.jpg b/32063-h/images/img887a.jpg
new file mode 100644
index 0000000..db2762a
--- /dev/null
+++ b/32063-h/images/img887a.jpg
Binary files differ
diff --git a/32063-h/images/img887b.jpg b/32063-h/images/img887b.jpg
new file mode 100644
index 0000000..453b955
--- /dev/null
+++ b/32063-h/images/img887b.jpg
Binary files differ
diff --git a/32063-h/images/img888.jpg b/32063-h/images/img888.jpg
new file mode 100644
index 0000000..db3da69
--- /dev/null
+++ b/32063-h/images/img888.jpg
Binary files differ
diff --git a/32063-h/images/img889a.jpg b/32063-h/images/img889a.jpg
new file mode 100644
index 0000000..70bd31b
--- /dev/null
+++ b/32063-h/images/img889a.jpg
Binary files differ
diff --git a/32063-h/images/img889b.jpg b/32063-h/images/img889b.jpg
new file mode 100644
index 0000000..71b6223
--- /dev/null
+++ b/32063-h/images/img889b.jpg
Binary files differ
diff --git a/32063-h/images/img890.jpg b/32063-h/images/img890.jpg
new file mode 100644
index 0000000..3442d5a
--- /dev/null
+++ b/32063-h/images/img890.jpg
Binary files differ
diff --git a/32063.txt b/32063.txt
new file mode 100644
index 0000000..2f85b97
--- /dev/null
+++ b/32063.txt
@@ -0,0 +1,6178 @@
+The Project Gutenberg EBook of Encyclopaedia Britannica, 11th Edition,
+Volume 6, Slice 8, by Various
+
+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
+
+
+Title: Encyclopaedia Britannica, 11th Edition, Volume 6, Slice 8
+ "Conduction, Electric"
+
+Author: Various
+
+Release Date: April 19, 2010 [EBook #32063]
+
+Language: English
+
+Character set encoding: ASCII
+
+*** START OF THIS PROJECT GUTENBERG EBOOK ENCYC. BRITANNICA, VOL 6, SL 8 ***
+
+
+
+
+Produced by Marius Masi, Don Kretz, Juliet Sutherland, and
+the Online Distributed Proofreading Team at
+https://www.pgdp.net
+
+
+
+
+
+
+
+
+
+Transcriber's notes:
+
+(1) Numbers following letters (without space) like C2 were originally
+ printed in subscript. When letters are subscripted, they are
+ preceded by an underscore, like C_n.
+
+(2) Characters following a carat (^) were originally printed in
+ superscript.
+
+(3) Letters topped by Macron are represented as [=x].
+
+(4) [oo] stands for infinity; [int] for integral; [alpha], [beta], etc.
+ for greek letters.
+
+(5) The following typographical errors have been corrected:
+
+ Page 871: "an equation which is very useful, because it enables us,
+ if we know the distribution." 'because' amended from 'becaus'.
+
+ Page 890: "The colour of the luminosity due to positive rays is not
+ in general the same as that due to anode rays." 'positive' amended
+ from 'postive'.
+
+ Page 890: "Anode Rays.--Gehrcke and Reichenhein (Ann. der Phys. 25,
+ p. 861) have found that when the anode consists of a mixture of
+ sodium and lithium chloride." 'mixture' amended from 'mixure'.
+
+
+
+
+ ENCYCLOPAEDIA BRITANNICA
+
+ A DICTIONARY OF ARTS, SCIENCES, LITERATURE
+ AND GENERAL INFORMATION
+
+ ELEVENTH EDITION
+
+
+ VOLUME VI, SLICE VIII
+
+ Conduction, Electric
+
+
+
+
+Article in This Slice:
+
+ CONDUCTION, ELECTRIC
+
+
+
+
+CONDUCTION, ELECTRIC. The electric conductivity of a substance is that
+property in virtue of which all its parts come spontaneously to the same
+electric potential if the substance is kept free from the operation of
+electric force. Accordingly, the reciprocal quality, electric
+resistivity, may be defined as a quality of a substance in virtue of
+which a difference of potential can exist between different portions of
+the body when these are in contact with some constant source of
+electromotive force, in such a manner as to form part of an electric
+circuit.
+
+All material substances possess in some degree, large or small, electric
+conductivity, and may for the sake of convenience be broadly divided
+into five classes in this respect. Between these, however, there is no
+sharply-marked dividing line, and the classification must therefore be
+accepted as a more or less arbitrary one. These divisions are: (1)
+metallic conductors, (2) non-metallic conductors, (3) dielectric
+conductors, (4) electrolytic conductors, (5) gaseous conductors. The
+first class comprises all metallic substances, and those mixtures or
+combinations of metallic substances known as alloys. The second includes
+such non-metallic bodies as carbon, silicon, many of the oxides and
+peroxides of the metals, and probably also some oxides of the
+non-metals, sulphides and selenides. Many of these substances, for
+instance carbon and silicon, are well-known to have the property of
+existing in several allotropic forms, and in some of these conditions,
+so far from being fairly good conductors, they may be almost perfect
+non-conductors. An example of this is seen in the case of carbon in its
+three allotropic conditions--charcoal, graphite and diamond. As charcoal
+it possesses a fairly well-marked but not very high conductivity in
+comparison with metals; as graphite, a conductivity about
+one-four-hundredth of that of iron; but as diamond so little
+conductivity that the substance is included amongst insulators or
+non-conductors. The third class includes those substances which are
+generally called insulators or non-conductors, but which are better
+denominated dielectric conductors; it comprises such solid substances as
+mica, ebonite, shellac, india-rubber, gutta-percha, paraffin, and a
+large number of liquids, chiefly hydrocarbons. These substances differ
+greatly in insulating power, and according as the conductivity is more
+or less marked, they are spoken of as bad or good insulators. Amongst
+the latter many of the liquid gases hold a high position. Thus, liquid
+oxygen and liquid air have been shown by Sir James Dewar to be almost
+perfect non-conductors of electricity.
+
+The behaviour of substances which fall into these three classes is
+discussed below in section I., dealing with metallic conduction.
+
+The fourth class, namely the electrolytic conductors comprises all those
+substances which undergo chemical decomposition when they form part of
+an electric circuit traversed by an electric current. They are discussed
+in section II., dealing with electrolytic conduction.
+
+The fifth and last class of conductors includes the gases. The
+conditions under which this class of substance becomes possessed of
+electric conductivity are considered in section III., on conduction in
+gases.
+
+In connexion with metallic conductors, it is a fact of great interest
+and considerable practical importance, that, although the majority of
+metals when in a finely divided or powdered condition are practically
+non-conductors, a mass of metallic powder or filings may be made to pass
+suddenly into a conductive condition by being exposed to the influence
+of an electric wave. The same is true of the loose contact of two
+metallic conductors. Thus if a steel point, such as a needle, presses
+very lightly against a metallic plate, say of aluminium, it is found
+that this metallic contact, if carefully adjusted, is non-conductive,
+but that if an electric wave is created anywhere in the neighbourhood,
+this non-conducting contact passes into a conductive state. This fact,
+investigated and discovered independently by D. E. Hughes, C. Onesti, E.
+Branly, O. J. Lodge and others, is applied in the construction of the
+"coherer," or sensitive tube employed as a detector or receiver in that
+form of "wireless telegraphy" chiefly developed by Marconi. Further
+references to it are made in the articles ELECTRIC WAVES and TELEGRAPHY:
+_Wireless_.
+
+ _International Ohm._--The practical unit of electrical resistance was
+ legally defined in Great Britain by the authority of the queen in
+ council in 1894, as the "resistance offered to an invariable electric
+ current by a column of mercury at the temperature of melting ice,
+ 14.4521 grammes in mass, of a constant cross-sectional area, and a
+ length 106.3 centimetres." The same unit has been also legalized as a
+ standard in France, Germany and the United States, and is denominated
+ the "International or Standard Ohm." It is intended to represent as
+ nearly as possible a resistance equal to 10 deg. absolute C.G.S. units
+ of electric resistance. Convenient multiples and subdivisions of the
+ ohm are the microhm and the megohm, the former being a millionth part
+ of an ohm, and the latter a million ohms. The resistivity of
+ substances is then numerically expressed by stating the resistance of
+ one cubic centimetre of the substance taken between opposed faces, and
+ expressed in ohms, microhms or megohms, as may be most convenient. The
+ reciprocal of the ohm is called the mho, which is the unit of
+ conductivity, and is defined as the conductivity of a substance whose
+ resistance is one ohm. The absolute unit of conductivity is the
+ conductivity of a substance whose resistivity is one absolute C.G.S.
+ unit, or one-thousandth-millionth part of an ohm. Resistivity is a
+ quality in which material substances differ very widely. The metals
+ and alloys, broadly speaking, are good conductors, and their
+ resistivity is conveniently expressed in microhms per cubic
+ centimetre, or in absolute C.G.S. units. Very small differences in
+ density and in chemical purity make, however, immense differences in
+ electric resistivity; hence the values given by different
+ experimentalists for the resistivity of known metals differ to a
+ considerable extent.
+
+
+I. CONDUCTION IN SOLIDS
+
+It is found convenient to express the resistivity of metals in two
+different ways: (1) We may state the resistivity of one cubic centimetre
+of the material in microhms or absolute units taken between opposed
+faces. This is called the _volume-resistivity_; (2) we may express the
+resistivity by stating the resistance in ohms offered by a wire of the
+material in question of uniform cross-section one metre in length, and
+one gramme in weight. This numerical measure of the resistivity is
+called the _mass-resistivity_. The mass-resistivity of a body is
+connected with its volume-resistivity and the density of the material in
+the following manner:--The mass-resistivity, expressed in microhms per
+metre-gramme, divided by 10 times the density is numerically equal to
+the volume-resistivity per centimetre-cube in absolute C.G.S. units. The
+mass-resistivity per metre-gramme can always be obtained by measuring
+the resistance and the mass of any wire of uniform cross-section of
+which the length is known, and if the density of the substance is then
+measured, the volume-resistivity can be immediately calculated.
+
+ If R is the resistance in ohms of a wire of length l, uniform
+ cross-section s, and density d, then taking [rho] for the
+ volume-resistivity we have 10^9R = [rho]l/s; but lsd = M, where M is
+ the mass of the wire. Hence 10^9R = [rho]dl^2/M. If l = 100 and M = 1,
+ then R = [rho]'= resistivity in ohms per metre-gramme, and 10^9[rho]'
+ = 10,000d[rho], or [rho] = 10^5[rho]'/d, and [rho]' = 10,000MR/l^2.
+
+ The following rules, therefore, are useful in connexion with these
+ measurements. To obtain the mass-resistivity per metre-gramme of a
+ substance in the form of a uniform metallic wire:--Multiply together
+ 10,000 times the mass in grammes and the total resistance in ohms, and
+ then divide by the square of the length in centimetres. Again, to
+ obtain the volume-resistivity in C.G.S. units per centimetre-cube, the
+ rule is to multiply the mass-resistivity in ohms by 100,000 and divide
+ by the density. These rules, of course, apply only to wires of uniform
+ cross-section. In the following Tables I., II. and III. are given the
+ mass and volume resistivity of ordinary metals and certain alloys
+ expressed in terms of the international ohm or the absolute C.G.S.
+ unit of resistance, the values being calculated from the experiments
+ of A. Matthiessen (1831-1870) between 1860 and 1865, and from later
+ results obtained by J. A. Fleming and Sir James Dewar in 1893.
+
+ TABLE I.--_Electric Mass-Resistivity of Various Metals at 0 deg. C.,
+ or Resistance per Metre-gramme in International Ohms at 0 deg. C._
+ (Matthiessen.)
+
+ +---------------------+---------------------+-------------+
+ | | Resistance at 0 deg.| |
+ | | C. in International | Approximate |
+ | Metal. | Ohms of a Wire | Temperature |
+ | | 1 Metre long and | Coefficient |
+ | | Weighing | near 20 |
+ | | 1 Gramme. | deg. C. |
+ +---------------------+---------------------+-------------+
+ |Silver (annealed) | .1523 | 0.00377 |
+ |Silver (hard-drawn) | .1657 | .. |
+ |Copper (annealed) | .1421 | 0.00388 |
+ |Copper (hard-drawn) | .1449 (Matthiessen's Standard) |
+ |Gold (annealed) | .4025 | 0.00365 |
+ |Gold (hard-drawn) | .4094 | .. |
+ |Aluminium (annealed) | .0757 | .. |
+ |Zinc (pressed) | .4013 | .. |
+ |Platinum (annealed) | 1.9337 | .. |
+ |Iron (annealed) | .765 | .. |
+ |Nickel (annealed) | 1.058[1] | .. |
+ |Tin (pressed) | .9618 | 0.00365 |
+ |Lead (pressed) | 2.2268 | 0.00387 |
+ |Antimony (pressed) | 2.3787 | 0.00389 |
+ |Bismuth (pressed) | 12.8554[1] | 0.00354 |
+ |Mercury (liquid) | 12.885[2] | 0.00072 |
+ +---------------------+---------------------+-------------+
+
+ The data commonly used for calculating metallic resistivities were
+ obtained by A. Matthiessen, and his results are set out in the Table
+ II. which is taken from Cantor lectures given by Fleeming Jenkin in
+ 1866 at or about the date when the researches were made. The figures
+ given by Jenkin have, however, been reduced to international ohms and
+ C.G.S. units by multiplying by ([pi]/4) X 0.9866 X 10^5 = 77,485.
+
+ Subsequently numerous determinations of the resistivity of various
+ pure metals were made by Fleming and Dewar, whose results are set out
+ in Table III.
+
+ TABLE II.--_Electric Volume-Resistivity of Various Metals at 0 deg.
+ C., or Resistance per Centimetre-cube in C.G.S. Units at 0 deg. C._
+
+ +---------------------+---------------------+
+ | | Volume-Resistivity. |
+ | Metal. | at 0 deg. C. in |
+ | | C.G.S. Units |
+ +---------------------+---------------------+
+ |Silver (annealed) | 1,502 |
+ |Silver (hard-drawn) | 1,629 |
+ |Copper (annealed) | 1,594 |
+ |Copper (hard-drawn) | 1,630[3] |
+ |Gold (annealed) | 2,052 |
+ |Gold (hard-drawn) | 2,090 |
+ |Aluminium (annealed) | 3,006 |
+ |Zinc (pressed) | 5,621 |
+ |Platinum (annealed) | 9,035 |
+ |Iron (annealed) | 10,568 |
+ |Nickel (annealed) | 12,429[4] |
+ |Tin (pressed) | 13,178 |
+ |Lead (pressed) | 19,580 |
+ |Antimony (pressed) | 35,418 |
+ |Bismuth (pressed) | 130,872 |
+ |Mercury (liquid) | 94,896[5] |
+ +---------------------+---------------------+
+
+ TABLE III.--_Electric Volume-Resistivity of Various Metals at 0 deg.
+ C., or Resistance per Centimetre-cube at 0 deg. C. in C.G.S. Units._
+ (Fleming and Dewar, _Phil. Mag._, September 1893.)
+
+ +--------------------------+------------------+------------------+
+ | | Resistance | Mean Temperature |
+ | Metal. | at 0 deg. C. per | Coefficient |
+ | | Centimetre-cube | between 0 deg. C.|
+ | | in C.G.S. Units. | and 100 deg. C. |
+ +--------------------------+------------------+------------------+
+ |Silver (electrolytic and | | |
+ | well annealed)[6] | 1,468 | 0.00400 |
+ |Copper (electrolytic and | | |
+ | well annealed)[6] | 1,561 | 0.00428 |
+ |Gold (annealed) | 2,197 | 0.00377 |
+ |Aluminium (annealed) | 2,665 | 0.00435 |
+ |Magnesium (pressed) | 4,355 | 0.00381 |
+ |Zinc | 5,751 | 0.00406 |
+ |Nickel (electrolytic)[6] | 6,935 | 0.00618 |
+ |Iron (annealed) | 9,065 | 0.00625 |
+ |Cadmium | 10,023 | 0.00419 |
+ |Palladium | 10,219 | 0.00354 |
+ |Platinum (annealed) | 10,917 | 0.003669 |
+ |Tin (pressed) | 13,048 | 0.00440 |
+ |Thallium (pressed) | 17,633 | 0.00398 |
+ |Lead (pressed) | 20,380 | 0.00411 |
+ |Bismuth (electrolytic)[7] | 110,000 | 0.00433 |
+ +--------------------------+------------------+------------------+
+
+ _Resistivity of Mercury._--The volume-resistivity of pure mercury is a
+ very important electric constant, and since 1880 many of the most
+ competent experimentalists have directed their attention to the
+ determination of its value. The experimental process has usually been
+ to fill a glass tube of known dimensions, having large cup-like
+ extensions at the ends, with pure mercury, and determine the absolute
+ resistance of this column of metal. For the practical details of this
+ method the following references may be consulted:--"The Specific
+ Resistance of Mercury," Lord Rayleigh and Mrs Sidgwick, _Phil.
+ Trans._, 1883, part i. p. 173, and R. T. Glazebrook, _Phil. Mag._,
+ 1885, p. 20; "On the Specific Resistance of Mercury," R. T. Glazebrook
+ and T. C. Fitzpatrick, _Phil. Trans._, 1888, p. 179, or _Proc. Roy.
+ Soc._, 1888, p. 44, or _Electrician_, 1888, 21, p. 538; "Recent
+ Determinations of the Absolute Resistance of Mercury," R. T.
+ Glazebrook, _Electrician_, 1890, 25, pp. 543 and 588. Also see J. V.
+ Jones, "On the Determination of the Specific Resistance of Mercury in
+ Absolute Measure," _Phil. Trans._, 1891, A, p. 2. Table IV. gives the
+ values of the volume-resistivity of mercury as determined by various
+ observers, the constant being expressed (a) in terms of the resistance
+ in ohms of a column of mercury one millimetre in cross-section and 100
+ centimetres in length, taken at 0 deg. C.; and (b) in terms of the
+ length in centimetres of a column of mercury one square millimetre in
+ cross-section taken at 0 deg. C. The result of all the most careful
+ determinations has been to show that the resistivity of pure mercury
+ at 0 deg. C. is about 94,070 C.G.S. electromagnetic units of
+ resistance, and that a column of mercury 106.3 centimetres in length
+ having a cross-sectional area of one square millimetre would have a
+ resistance at 0 deg. C. of one international ohm. These values have
+ accordingly been accepted as the official and recognized values for
+ the specific resistance of mercury, and the definition of the ohm. The
+ table also states the methods which have been adopted by the different
+ observers for obtaining the absolute value of the resistance of a
+ known column of mercury, or of a resistance coil afterwards compared
+ with a known column of mercury. A column of figures is added showing
+ the value in fractions of an international ohm of the British
+ Association Unit (B.A.U.), formerly supposed to represent the true
+ ohm. The real value of the B.A.U. is now taken as .9866 of an
+ international ohm.
+
+ TABLE IV.--_Determinations of the Absolute Value of the
+ Volume-Resistivity of Mercury and the Mercury Equivalent of the
+ Ohm._
+
+ +-----------------+------+----------------------+--------+----------+---------+
+ | | | | |Value of |Value of |
+ | | | |Value of|100 Centi-|Ohm in |
+ | Observer. | Date.| Method. |B.A.U. |metres of |Centi- |
+ | | | |in Ohms.|Mercury |metres of|
+ | | | | |in Ohms. |Mercury. |
+ +-----------------+------+----------------------+--------+----------+---------+
+ |Lord Rayleigh | 1882 | Rotating coil | .98651 | .94133 | 106.31 |
+ |Lord Rayleigh | 1883 | Lorenz method | .98677 | .. | 106.27 |
+ |G. Wiedemann | 1884 | Rotation through 180 | .. | .. | 106.19 |
+ | | | deg. | | | |
+ |E. E. N. Mascart | 1884 | Induced current | .98611 | .94096 | 106.33 |
+ |H. A. Rowland | 1887 | Mean of several | .98644 | .94071 | 106.32 |
+ | | | methods | | | |
+ |F. Kohlrausch | 1887 | Damping of magnets | .98660 | .94061 | 106.32 |
+ |R. T. Glazebrook | 1882 | Induced currents | .98665 | .94074 | 106.29 |
+ | | 1888 | | | | |
+ |Wuilleumeier | 1890 | | .98686 | .94077 | 106.31 |
+ |Duncan and Wilkes| 1890 | Lorenz | .98634 | .94067 | 106.34 |
+ |J. V. Jones | 1891 | Lorenz | .. | .94067 | 106.31 |
+ | | | +--------+----------+---------+
+ | | | Mean value .98653 | | |
+ | | | | | |
+ |Streker | 1885 | An absolute determin-| | .94056 | 106.32 |
+ |Hutchinson | 1888 | ation of resistance| | .94074 | 106.30 |
+ |E. Salvioni | 1890 | was not made. The | | .94054 | 106.33 |
+ |E. Salvioni | .. | value .98656 has | | .94076 | 106.30 |
+ | | | been used | | | |
+ | | | | +----------+---------+
+ | | | Mean value .94076 | 106.31 |
+ | | | | |
+ |H. F. Weber | 1884 | Induced current | Absolute measure- | 105.37 |
+ |H. F. Weber | .. | Rotating coil | ments compared | 106.16 |
+ |A. Roiti | 1884 | Mean effect of | with German silver| 105.89 |
+ | | | induced current | wire coils issued | |
+ |F. Himstedt | 1885 | | by Siemens and | 105.98 |
+ | | | | Streker | |
+ |F. E. Dorn | 1889 | Damping of a magnet | | 106.24 |
+ |Wild | 1883 | Damping of a magnet | | 106.03 |
+ |L. V. Lorenz | 1885 | Lorenz method | | 105.93 |
+ +-----------------+------+----------------------+-------------------+---------+
+
+ For a critical discussion of the methods which have been adopted in
+ the absolute determination of the resistivity of mercury, and the
+ value of the British Association unit of resistance, the reader may be
+ referred to the _British Association Reports_ for 1890 and 1892
+ (_Report of Electrical Standards Committee_), and to the
+ _Electrician_, 25, p. 456, and 29, p. 462. A discussion of the
+ relative value of the results obtained between 1882 and 1890 was given
+ by R. T. Glazebrook in a paper presented to the British Association at
+ Leeds, 1890.
+
+ _Resistivity of Copper._--In connexion with electro-technical work the
+ determination of the conductivity or resistivity values of annealed
+ and hard-drawn copper wire at standard temperatures is a very
+ important matter. Matthiessen devoted considerable attention to this
+ subject between the years 1860 and 1864 (see _Phil. Trans._, 1860, p.
+ 150), and since that time much additional work has been carried out.
+ Matthiessen's value, known as _Matthiessen's Standard_, for the
+ mass-resistivity of pure hard-drawn copper wire, is the resistance of
+ a wire of pure hard-drawn copper one metre long and weighing one
+ gramme, and this is equal to 0.14493 international ohms at 0 deg. C.
+ For many purposes it is more convenient to express temperature in
+ Fahrenheit degrees, and the recommendation of the 1899 committee on
+ copper conductors[8] is as follows:--"Matthiessen's standard for
+ hard-drawn conductivity commercial copper shall be considered to be
+ the resistance of a wire of pure hard-drawn copper one metre long,
+ weighing one gramme which at 60 deg. F. is 0.153858 international
+ ohms." Matthiessen also measured the mass-resistivity of annealed
+ copper, and found that its conductivity is greater than that of
+ hard-drawn copper by about 2.25% to 2.5% As annealed copper may vary
+ considerably in its state of annealing, and is always somewhat
+ hardened by bending and winding, it is found in practice that the
+ resistivity of commercial annealed copper is about 1-1/4% less than
+ that of hard-drawn copper. The standard now accepted for such copper,
+ on the recommendation of the 1899 Committee, is a wire of pure
+ annealed copper one metre long, weighing one gramme, whose resistance
+ at 0 deg. C. is 0.1421 international ohms, or at 60 deg. F., 0.150822
+ international ohms. The specific gravity of copper varies from about
+ 8.89 to 8.95, and the standard value accepted for high conductivity
+ commercial copper is 8.912, corresponding to a weight of 555 lb per
+ cubic foot at 60 deg. F. Hence the volume-resistivity of pure annealed
+ copper at 0 deg. C. is 1.594 microhms per c.c., or 1594 C.G.S. units,
+ and that of pure hard-drawn copper at 0 deg. C. is 1.626 microhms per
+ c.c., or 1626 C.G.S. units. Since Matthiessen's researches, the most
+ careful scientific investigation on the conductivity of copper is that
+ of T. C. Fitzpatrick, carried out in 1890. (_Brit. Assoc. Report_,
+ 1890, Appendix 3, p. 120.) Fitzpatrick confirmed Matthiessen's chief
+ result, and obtained values for the resistivity of hard-drawn copper
+ which, when corrected for temperature variation, are in entire
+ agreement with those of Matthiessen at the same temperature.
+
+The volume resistivity of alloys is, generally speaking, much higher
+than that of pure metals. Table V. shows the volume resistivity at 0
+deg. C. of a number of well-known alloys, with their chemical
+composition.
+
+ TABLE V.--_Volume-Resistivity of Alloys of known Composition at 0 deg.
+ C. in C.G.S. Units per Centimetre-cube. Mean Temperature Coefficients
+ taken at 15 deg. C._ (Fleming and Dewar.)
+
+ +-----------------------+-----------+-------------+-------------------+
+ | |Resistivity| Temperature |Composition in per |
+ | Alloys. | at 0 | Coefficient | cents. |
+ | | deg. C. |at 15 deg. C.| |
+ +-----------------------+-----------+-------------+-------------------+
+ |Platinum-silver | 31,582 | .000243 |Pt 33%, Ag 66% |
+ |Platinum-iridium | 30,896 | .000822 |Pt 80%, Ir 20% |
+ |Platinum-rhodium | 21,142 | .00143 |Pt 90%, Rd 10% |
+ |Gold-silver | 6,280 | .00124 |Au 90%, Ag 10% |
+ |Manganese-steel | 67,148 | .00127 |Mn 12%, Fe 78% |
+ |Nickel-steel | 29,452 | .00201 |Ni 4.35%, remaining|
+ | | | | percentage |
+ | | | | chiefly iron, but|
+ | | | | uncertain |
+ |German silver | 29,982 | .000273 |Cu5Zn3Ni2 |
+ |Platinoid[9] | 41,731 | .00031 | |
+ |Manganin | 46,678 | .0000 |Cu 84%, Mn 12%, |
+ | | | |Ni 4% |
+ |Aluminium-silver | 4,641 | .00238 |Al 94%, Ag 6% |
+ |Aluminium-copper | 2,904 | .00381 |Al 94%, Cu 6% |
+ |Copper-aluminium | 8,847 | .000897 |Cu 97%, Al 3% |
+ |Copper-nickel-aluminium| 14,912 | .000643 |Cu 87%, Ni 6.5%, |
+ | | | | Al 6.5% |
+ |Titanium-aluminium | 3,887 | .00290 | |
+ +-----------------------+-----------+-------------+-------------------+
+
+ Generally speaking, an alloy having high resistivity has poor
+ mechanical qualities, that is to say, its tensile strength and
+ ductility are small. It is possible to form alloys having a
+ resistivity as high as 100 microhms per cubic centimetre; but, on the
+ other hand, the value of an alloy for electro-technical purposes is
+ judged not merely by its resistivity, but also by the degree to which
+ its resistivity varies with temperature, and by its capability of
+ being easily drawn into fine wire of not very small tensile strength.
+ Some pure metals when alloyed with a small proportion of another metal
+ do not suffer much change in resistivity, but in other cases the
+ resultant alloy has a much higher resistivity. Thus an alloy of pure
+ copper with 3% of aluminium has a resistivity about 5-1/2 times that of
+ copper; but if pure aluminium is alloyed with 6% of copper, the
+ resistivity of the product is not more than 20% greater than that of
+ pure aluminium. The presence of a very small proportion of a
+ non-metallic element in a metallic mass, such as oxygen, sulphur or
+ phosphorus, has a very great effect in increasing the resistivity.
+ Certain metallic elements also have the same power; thus platinoid has
+ a resistivity 30% greater than German silver, though it differs from
+ it merely in containing a trace of tungsten.
+
+The resistivity of non-metallic conductors is in all cases higher than
+that of any pure metal. The resistivity of carbon, for instance, in the
+forms of charcoal or carbonized organic material and graphite, varies
+from 600 to 6000 microhms per cubic centimetre, as shown in Table VI.:--
+
+ TABLE VI.--_Electric Volume-Resistivity in Microhms per
+ Centimetre-cube of Various Forms of Carbon at 15 deg. C._
+
+ +-----------------------------------------+---------------+
+ | Substance. | Resistivity. |
+ +-----------------------------------------+---------------+
+ | Arc lamp carbon rod | 8000 |
+ | Jablochkoff candle carbon | 4000 |
+ | Carre carbon | 3400 |
+ | Carbonized bamboo | 6000 |
+ | Carbonized parchmentized thread | 4000 to 5000 |
+ | Ordinary carbon filament from glow-lamp | |
+ | "treated" or flashed | 2400 to 2500 |
+ | Deposited or secondary carbon | 600 to 900 |
+ | Graphite | 400 to 500 |
+ +-----------------------------------------+---------------+
+
+The resistivity of liquids is, generally speaking, much higher than that
+of any metals, metallic alloys or non-metallic conductors. Thus fused
+lead chloride, one of the best conducting liquids, has a resistivity in
+its fused condition of 0.376 ohm per centimetre-cube, or 376,000
+microhms per centimetre-cube, whereas that of metallic alloys only in
+few cases exceeds 100 microhms per centimetre-cube. The resistivity of
+solutions of metallic salts also varies very largely with the proportion
+of the diluent or solvent, and in some instances, as in the aqueous
+solutions of mineral acids; there is a maximum conductivity
+corresponding to a certain dilution. The resistivity of many liquids,
+such as alcohol, ether, benzene and pure water, is so high, in other
+words, their conductivity is so small, that they are practically
+insulators, and the resistivity can only be appropriately expressed in
+megohms per centimetre-cube.
+
+In Table VII. are given the names of a few of these badly-conducting
+liquids, with the values of their volume-resistivity in megohms per
+centimetre-cube:--
+
+ TABLE VII.--_Electric Volume-Resistivity of Various Badly-Conducting
+ Liquids in Megohms per Centimetre-cube._
+
+ +---------------------------+----------------+---------------------+
+ | | Resistivity | |
+ | Substance. | in Megohms | Observer. |
+ | | per c.c. | |
+ +---------------------------+----------------+---------------------+
+ | Ethyl alcohol | 0.5 | Pfeiffer. |
+ | Ethyl ether | 1.175 to 3.760 | W. Kohlrausch. |
+ | Benzene | 4.700 | |
+ | Absolutely pure water | 25.0 at 18 deg.| Value estimated |
+ | approximates probably to | C. | by F. Kohlrausch |
+ | | | and A. Heydweiler. |
+ | All very dilute aqueous | 1.00 at 18 deg.| From results by |
+ | salt solutions having a | C. | F. Kohlrausch |
+ | concentration of about | | and others. |
+ | 0.00001 of an equivalent | | |
+ | gramme molecule[10] per | | |
+ | litre approximate to | | |
+ +---------------------------+----------------+---------------------+
+
+The resistivity of all those substances which are generally called
+dielectrics or insulators is also so high that it can only be
+appropriately expressed in millions of megohms per centimetre-cube, or
+in megohms per quadrant-cube, the quadrant being a cube the side of
+which is 10^9 cms. (see Table VIII.).
+
+Effects of Heat.--Temperature affects the resistivity of these different
+classes of conductors in different ways. In all cases, so far as is yet
+known, the resistivity of a pure metal is increased if its temperature
+is raised, and decreased if the temperature is lowered, so that if it
+could be brought to the absolute zero of temperature (-273 deg. C.) its
+resistivity would be reduced to a very small fraction of its resistance
+at ordinary temperatures. With metallic alloys, however, rise of
+temperature does not always increase resistivity: it sometimes
+diminishes it, so that many alloys are known which have a maximum
+resistivity corresponding to a certain temperature, and at or near this
+point they vary very little in resistance with temperature. Such alloys
+have, therefore, a negative temperature-variation of resistance at and
+above fixed temperatures. Prominent amongst these metallic compounds are
+alloys of iron, manganese, nickel and copper, some of which were
+discovered by Edward Weston, in the United States. One well-known alloy
+of copper, manganese and nickel, now called manganin, which was brought
+to the notice of electricians by the careful investigations made at the
+Berlin Physikalisch-Technische Reichsanstalt, is characterized by having
+a zero temperature coefficient at or about a certain temperature in the
+neighbourhood of 15 deg. C. Hence within a certain range of temperature
+on either side of this critical value the resistivity of manganin is
+hardly affected at all by temperature. Similar alloys can be produced
+from copper and ferro-manganese. An alloy formed of 80% copper and 20%
+manganese in an annealed condition has a nearly zero
+temperature-variation of resistance between 20 deg. C. and 100 deg. C.
+In the case of non-metals the action of temperature is generally to
+diminish the resistivity as temperature rises, though this is not
+universally so. The interesting observation has been recorded by J. W.
+Howell, that "treated" carbon filaments and graphite are substances
+which have a minimum resistance corresponding to a certain temperature
+approaching red heat (_Electrician_, vol. xxxviii. p. 835). At and
+beyond this temperature increased heating appears to increase their
+resistivity; this phenomenon may, however, be accompanied by a molecular
+change and not be a true temperature variation. In the case of
+dielectric conductors and of electrolytes, the action of rising
+temperature is to reduce resistivity. Many of the so-called insulators,
+such as mica, ebonite, indiarubber, and the insulating oils, paraffin,
+&c., decrease in resistivity with great rapidity as the temperature
+rises. With guttapercha a rise in temperature from 0 deg. C. to 24 deg.
+C. is sufficient to reduce the resistivity of one-twentieth part of its
+value at 0 deg. C., and the resistivity of flint glass at 140 deg. C. is
+only one-hundredth of what it is at 60 deg. C.
+
+ TABLE VIII.--_Electric Volume-Resistivity of Dielectrics reckoned in
+ Millions of Megohms (Mega-megohms) per Centimetre-cube, and in Megohms
+ per Quadrant-cube, i.e. a Cube whose Side is 10^9 cms._
+
+ +------------------------+---------------------------+----------+
+ | | Resistivity. | |
+ | +------------+--------------+ Tempera- |
+ | Substance. |Mega-megohms| Megohms per | tura |
+ | | per c.c. |Quadrant-cube.| Cent. |
+ +------------------------+------------+--------------+----------+
+ |Bohemian glass | 61 | .061 | 60 deg. |
+ |Mica | 84 | .084 | 20 deg. |
+ |Gutta-percha | 450 | .45 | 24 deg. |
+ |Flint glass | 1,020 | 1.02 | 60 deg. |
+ |Glover's vulcanized | | | |
+ | indiarubber | 1,630 | 1.63 | 15 deg. |
+ |Siemens' ordinary pure | | | |
+ | vulcanized indiarubber | 2,280 | 2.28 | 15 deg. |
+ |Shellac | 9,000 | 9.0 | 28 deg. |
+ |Indiarubber | 10,900 | 10.9 | 24 deg. |
+ |Siemens' high-insulating| | | |
+ | fibrous material | 11,900 | 11.9 | 15 deg. |
+ |Siemens' special | | | |
+ | high-insulating | | | |
+ | indiarubber. | 16,170 | 16.17 | 15 deg. |
+ |Flint glass | 20,000 | 20.0 | 20 deg. |
+ |Ebonite | 28,000 | 28. | 46 deg. |
+ |Paraffin | 34,000 | 34. | 46 deg. |
+ +------------------------+------------+--------------+----------+
+
+ A definition may here be given of the meaning of the term _Temperature
+ Coefficient_. If, in the first place, we suppose that the resistivity
+ ([rho]t) at any temperature (t) is a simple linear function of the
+ resistivity ([rho]0) at 0 deg. C., then we can write [rho]t = [rho]0(1
+ + [alpha]t), or [alpha] = ([rho]t - [rho]0)/[rho]0t.
+
+ The quantity [alpha] is then called the temperature-coefficient, and
+ its reciprocal is the temperature at which the resistivity would
+ become zero. By an extension of this notion we can call the quantity
+ d[rho]/[rho]dt the temperature coefficient corresponding to any
+ temperature t at which the resistivity is [rho]. In all cases the
+ relation between the resistivity of a substance and the temperature is
+ best set out in the form of a curve called a temperature-resistance
+ curve. If a series of such curves are drawn for various pure metals,
+ temperature being taken as abscissa and resistance as ordinate, and if
+ the temperature range extends from the absolute zero of temperature
+ upwards, then it is found that these temperature-resistance lines are
+ curved lines having their convexity either upwards or downwards. In
+ other words, the second differential coefficient of resistance with
+ respect to temperature is either a positive or negative quantity. An
+ extensive series of observations concerning the form of the
+ resistivity curves for various pure metals over a range of temperature
+ extending from -200 deg. C. to +200 deg. C. was carried out in 1892
+ and 1893 by Fleming and Dewar (_Phil. Mag._ Oct. 1892 and Sept. 1893).
+ The resistance observations were taken with resistance coils
+ constructed with wires of various metals obtained in a state of great
+ chemical purity. The lengths and mean diameters of the wires were
+ carefully measured, and their resistance was then taken at certain
+ known temperatures obtained by immersing the coils in boiling aniline,
+ boiling water, melting ice, melting carbonic acid in ether, and
+ boiling liquid oxygen, the temperatures thus given being +184 deg..5
+ C., +100 deg. C., 0 deg. C., -78 deg..2 C. and -182 deg..5 C. The
+ resistivities of the various metals were then calculated and set out
+ in terms of the temperature. From these data a chart was prepared
+ showing the temperature-resistance curves of these metals throughout a
+ range of 400 degrees. The exact form of these curves through the
+ region of temperature lying between -200 deg. C. and -273 deg. C. is
+ not yet known. As shown on the chart, the curves evidently do not
+ converge to precisely the same point. It is, however, much less
+ probable that the resistance of any metal should vanish at a
+ temperature above the absolute zero than at the absolute zero itself,
+ and the precise path of these curves at their lower ends cannot be
+ delineated until means are found for fixing independently the
+ temperature of some regions in which the resistance of metallic wires
+ can be measured. Sir J. Dewar subsequently showed that for certain
+ pure metals it is clear that the resistance would not vanish at the
+ absolute zero but would be reduced to a finite but small value (see
+ "Electric Resistance Thermometry at the Temperature of Boiling
+ Hydrogen," _Proc. Roy. Soc._ 1904, 73, p. 244).
+
+ The resistivity curves of the magnetic metals are also remarkable for
+ the change of curvature they exhibit at the magnetic critical
+ temperature. Thus J. Hopkinson and D. K. Morris (_Phil. Mag._
+ September 1897, p. 213) observed the remarkable alteration that takes
+ place in the iron resistance temperature curve in the neighbourhood of
+ 780 deg. C. At that temperature the direction of the curvature of the
+ curve changes so that it becomes convex upwards instead of convex
+ downwards, and in addition the value of the temperature coefficient
+ undergoes a great reduction. The mean temperature coefficient of iron
+ in the neighbourhood of 0 deg. C. is 0.0057; at 765 deg. C. it rises
+ to a maximum value 0.0204; but at 1000 deg. C. it falls again to a
+ lower value, 0.00244. A similar rise to a maximum value and subsequent
+ fall are also noted in the case of the specific heat of iron. The
+ changes in the curvature of the resistivity curves are undoubtedly
+ connected with the molecular changes that occur in the magnetic metals
+ at their critical temperatures.
+
+ A fact of considerable interest in connexion with resistivity is the
+ influence exerted by a strong magnetic field in the case of some
+ metals, notably bismuth. It was discovered by A. Righi and confirmed
+ by S. A. Leduc (_Journ. de Phys._ 1886, 5, p. 116, and 1887, 6, p.
+ 189) that if a pure bismuth wire is placed in a magnetic field
+ transversely to the direction of the magnetic field, its resistance is
+ considerably increased. This increase is greatly affected by the
+ temperature of the metal (Dewar and Fleming, _Proc. Roy. Soc._ 1897,
+ 60, p. 427). The temperature coefficient of pure copper is an
+ important constant, and its value as determined by Messrs Clark, Forde
+ and Taylor in terms of Fahrenheit temperature is
+
+ [rho]t = [rho]32 {1 + 0.0023708(t - 32) + 0.0000034548(t - 32)^2}.
+
+_Time Effects._--In the case of dielectric conductors, commonly called
+insulators, such as indiarubber, guttapercha, glass and mica, the
+electric resistivity is not only a function of the temperature but also
+of the time during which the electromotive force employed to measure it
+is imposed. Thus if an indiarubber-covered cable is immersed in water
+and the resistance of the dielectric between the copper conductor and
+the water measured by ascertaining the current which can be caused to
+flow through it by an electromotive force, this current is found to vary
+very rapidly with the time during which the electromotive force is
+applied. Apart from the small initial effect due to the electrostatic
+capacity of the cable, the application of an electromotive force to the
+dielectric produces a current through it which rapidly falls in value,
+as if the electric resistance of the dielectric were increasing. The
+current, however, does not fall continuously but tends to a limiting
+value, and it appears that if the electromotive force is kept applied
+to the cable for a prolonged time, a small and nearly constant current
+will ultimately be found flowing through it. It is customary in
+electro-technical work to consider the resistivity of the dielectric as
+the value it has after the electromotive force has been applied for one
+minute, the standard temperature being 75 deg. F. This, however, is a
+purely conventional proceeding, and the number so obtained does not
+necessarily represent the true or ohmic resistance of the dielectric. If
+the electromotive force is increased, in the case of a large number of
+ordinary dielectrics the apparent resistance at the end of one minute's
+electrification decreases as the electromotive force increases.
+
+_Practical Standards._--The practical measurement of resistivity
+involves many processes and instruments (see WHEATSTONE'S BRIDGE and
+OHMMETER). Broadly speaking, the processes are divided into _Comparison
+Methods_ and _Absolute Methods_. In the former a comparison is effected
+between the resistance of a material in a known form and some standard
+resistance. In the _Absolute Methods_ the resistivity is determined
+without reference to any other substance, but with reference only to the
+fundamental standards of length, mass and time. Immense labour has been
+expended in investigations concerned with the production of a standard
+of resistance and its evaluation in absolute measure. In some cases the
+absolute standard is constructed by filling a carefully-calibrated tube
+of glass with mercury, in order to realize in a material form the
+official definition of the ohm; in this manner most of the principal
+national physical laboratories have been provided with standard mercury
+ohms. (For a full description of the standard mercury ohm of the Berlin
+Physikalisch-Technische Reichsanstalt, see the _Electrician_, xxxvii.
+569.) For practical purposes it is more convenient to employ a standard
+of resistance made of wire.
+
+ Opinion is not yet perfectly settled on the question whether a wire
+ made of any alloy can be considered to be a perfectly unalterable
+ standard of resistance, but experience has shown that a platinum
+ silver alloy (66% silver, 33% platinum), and also the alloy called
+ manganin, seem to possess the qualities of permanence essential for a
+ wire-resistance standard. A comparison made in 1892 and 1894 of all
+ the manganin wire copies of the ohm made at the Reichsanstalt in
+ Berlin, showed that these standards had remained constant for two
+ years to within one or two parts in 100,000. It appears, however, that
+ in order that manganin may remain constant in resistivity when used in
+ the manufacture of a resistance coil, it is necessary that the alloy
+ should be _aged_ by heating it to a temperature of 140 deg. C. for ten
+ hours; and to prevent subsequent changes in resistivity, solders
+ containing zinc must be avoided, and a silver solder containing 75% of
+ silver employed in soldering the manganin wire to its connexions.
+
+The authorities of the Berlin Reichsanstalt have devoted considerable
+attention to the question of the best form for a wire standard of
+electric resistance. In that now adopted the resistance wire is
+carefully insulated and wound on a brass cylinder, being doubled on
+itself to annul inductance as much as possible. In the coil two wires
+are wound on in parallel, one being much finer than the other, and the
+final adjustment of the coil to an exact value is made by shortening the
+finer of the two. A standard of resistance for use in a laboratory now
+generally consists of a wire of manganin or platinum-silver carefully
+insulated and enclosed in a brass case. Thick copper rods are connected
+to the terminals of the wire in the interior of the case, and brought to
+the outside, being carefully insulated at the same time from one another
+and from the case. The coil so constructed can be placed under water or
+paraffin oil, the temperature of which can be exactly observed during
+the process of taking a resistance measurement. Equalization of the
+temperature of the surrounding medium is effected by the employment of a
+stirrer, worked by hand or by a small electric motor. The construction
+of a standard of electrical resistance consisting of mercury in a glass
+tube is an operation requiring considerable precautions, and only to be
+undertaken by those experienced in the matter. Opinions are divided on
+the question whether greater permanence in resistance can be secured by
+mercury-in-glass standards of resistance or by wire standards, but the
+latter are at least more portable and less fragile.
+
+ A full description of the construction of a standard wire-resistance
+ coil on the plan adopted by the Berlin Physikalisch-Technische
+ Reichsanstalt is given in the Report of the British Association
+ Committee on Electrical Standards, presented at the Edinburgh Meeting
+ in 1892. For the design and construction of standards of electric
+ resistances adapted for employment in the comparison and measurement
+ of very low or very high resistances, the reader may be referred to
+ standard treatises on electric measurements.
+
+ BIBLIOGRAPHY.--See also J. A. Fleming, _A Handbook for the Electrical
+ Laboratory and Testing Room_, vol. i. (London, 1901); _Reports of the
+ British Association Committee on Electrical Standards_, edited by
+ Fleeming Jenkin (London, 1873); A. Matthiessen and C. Vogt, "On the
+ Influence of Temperature on the Conducting Power of Alloys," _Phil.
+ Trans._, 1864, 154, p. 167, and _Phil. Mag._, 1865, 29, p. 363; A.
+ Matthiessen and M. Holtzmann, "On the Effect of the Presence of Metals
+ and Metalloids upon the Electric Conducting Power of Pure Copper,"
+ _Phil. Trans._, 1860, 150, p. 85; T. C. Fitzpatrick, "On the Specific
+ Resistance of Copper," _Brit. Assoc. Report_, 1890, p. 120, or
+ _Electrician_, 1890, 25, p. 608; R. Appleyard, _The Conductometer and
+ Electrical Conductivity_; Clark, Forde and Taylor, _Temperature
+ Coefficients of Copper_ (London, 1901). (J. A. F.)
+
+
+II. CONDUCTION IN LIQUIDS
+
+Through liquid metals, such as mercury at ordinary temperatures and
+other metals at temperatures above their melting points, the electric
+current flows as in solid metals without changing the state of the
+conductor, except in so far as heat is developed by the electric
+resistance. But another class of liquid conductors exists, and in them
+the phenomena are quite different. The conductivity of fused salts, and
+of solutions of salts and acids, although less than that of metals, is
+very great compared with the traces of conductivity found in so-called
+non-conductors. In fused salts and conducting solutions the passage of
+the current is always accompanied by definite chemical changes; the
+substance of the conductor or electrolyte is decomposed, and the
+products of the decomposition appear at the electrodes, i.e. the
+metallic plates by means of which the current is led into and out of the
+solution. The chemical phenomena are considered in the article
+ELECTROLYSIS; we are here concerned solely with the mechanism of this
+_electrolytic_ conduction of the current.
+
+To explain the appearance of the products of decomposition at the
+electrodes only, while the intervening solution is unaltered, we suppose
+that, under the action of the electric forces, the opposite parts of the
+electrolyte move in opposite directions through the liquid. These
+opposite parts, named ions by Faraday, must therefore be associated with
+electric charges, and it is the convective movement of the opposite
+streams of ions carrying their charges with them that, on this view,
+constitutes the electric current.
+
+In metallic conduction it is found that the current is proportional to
+the applied electromotive force--a relation known by the name of Ohm's
+law. If we place in a circuit with a small electromotive force an
+electrolytic cell consisting of two platinum electrodes and a solution,
+the initial current soon dies away, and we shall find that a certain
+minimum electromotive force must be applied to the circuit before any
+considerable permanent current passes. The chemical changes which are
+initiated on the surfaces of the electrodes set up a reverse
+electromotive force of polarization, and, until this is overcome, only a
+minute current, probably due to the slow but steady removal of the
+products of decomposition from the electrodes by a process of diffusion,
+will pass through the cell. Thus it is evident that, considering the
+electrolytic cell as a whole, the passage of the current through it
+cannot conform to Ohm's law. But the polarization is due to chemical
+changes, which are confined to the surfaces of the electrodes; and it is
+necessary to inquire whether, if the polarization at the electrodes be
+eliminated, the passage of the current through the bulk of the solution
+itself is proportional to the electromotive force actually applied to
+that solution. Rough experiment shows that the current is proportional
+to the excess of the electromotive force over a constant value, and thus
+verifies the law approximately, the constant electromotive force to be
+overcome being a measure of the polarization. A more satisfactory
+examination of the question was made by F. Kohlrausch in the years 1873
+to 1876. Ohm's law states that the current C is proportional to the
+electromotive force E, or C = kR, where k is a constant called the
+conductivity of the circuit. The equation may also be written as C =
+E/R, where R is a constant, the reciprocal of k, known as the resistance
+of the circuit. The essence of the law is the proportionality between C
+and E, which means that the ratio E/C is a constant. But E/C = R, and
+thus the law may be tested by examining the constancy of the measured
+resistance of a conductor when different currents are passing through
+it. In this way Ohm's law has been confirmed in the case of metallic
+conduction to a very high degree of accuracy. A similar principle was
+applied by Kohlrausch to the case of electrolytes, and he was the first
+to show that an electrolyte possesses a definite resistance which has a
+constant value when measured with different currents and by different
+experimental methods.
+
+_Measurement of the Resistance of Electrolytes._--There are two effects
+of the passage of an electric current which prevent the possibility of
+measuring electrolytic resistance by the ordinary methods with the
+direct currents which are used in the case of metals. The products of
+the chemical decomposition of the electrolyte appear at the electrodes
+and set up the opposing electromotive force of polarization, and unequal
+dilution of the solution may occur in the neighbourhood of the two
+electrodes. The chemical and electrolytic aspects of these phenomena are
+treated in the article ELECTROLYSIS, but from our present point of view
+also it is evident that they are again of fundamental importance. The
+polarization at the surface of the electrodes will set up an opposing
+electromotive force, and the unequal dilution of the solution will turn
+the electrolyte into a concentration cell and produce a subsidiary
+electromotive force either in the same direction as that applied or in
+the reverse according as the anode or the cathode solution becomes the
+more dilute. Both effects thus involve internal electromotive forces,
+and prevent the application of Ohm's law to the electrolytic cell as a
+whole. But the existence of a definite measurable resistance as a
+characteristic property of the system depends on the conformity of the
+system to Ohm's law, and it is therefore necessary to eliminate both
+these effects before attempting to measure the resistance.
+
+The usual and most satisfactory method of measuring the resistance of
+electrolytes consists in eliminating the effects of polarization by the
+use of alternating currents, that is, currents that are reversed in
+direction many times a second.[11] The chemical action produced by the
+first current is thus reversed by the second current in the opposite
+direction, and the polarization caused by the first current on the
+surface of the electrodes is destroyed before it rises to an appreciable
+value. The polarization is also diminished in another way. The
+electromotive force of polarization is due to the deposition of films of
+the products of chemical decomposition on the surface of the electrodes,
+and only reaches its full value when a continuous film is formed. If the
+current be stopped before such a film is completed, the reverse
+electromotive force is less than its full value. A given current flowing
+for a given time deposits a definite amount of substance on the
+electrodes, and therefore the amount per unit area is inversely
+proportional to the area of the electrodes--to the area of contact, that
+is, between the electrode and the liquid. Thus, by increasing the area
+of the electrodes, the polarization due to a given current is decreased.
+Now the area of free surface of a platinum plate can be increased
+enormously by coating the plate with platinum black, which is metallic
+platinum in a spongy state, and with such a plate as electrode the
+effects of polarization are diminished to a very marked extent. The
+coating is effected by passing an electric current first one way and
+then the other between two platinum plates immersed in a 3% solution of
+platinum chloride to which a trace of lead acetate is sometimes added.
+The platinized plates thus obtained are quite satisfactory for the
+investigation of strong solutions. They have the power, however, of
+absorbing a certain amount of salt from the solutions and of giving it
+up again when water or more dilute solution is placed in contact with
+them. The measurement of very dilute solutions is thus made difficult,
+but, if the plates be heated to redness after being platinized, a grey
+surface is obtained which possesses sufficient area for use with dilute
+solutions and yet does not absorb an appreciable quantity of salt.
+
+Any convenient source of alternating current may be used. The currents
+from the secondary circuit of a small induction coil are satisfactory,
+or the currents of an alternating electric light supply may be
+transformed down to an electromotive force of one or two volts. With
+such currents it is necessary to consider the effects of self-induction
+in the circuit and of electrostatic capacity. In balancing the
+resistance of the electrolyte, resistance coils may be used in which
+self-induction and the capacity are reduced to a minimum by winding the
+wire of the coil backwards and forwards in alternate layers.
+
+[Illustration: FIG. 1.]
+
+With these arrangements the usual method of measuring resistance by
+means of Wheatstone's bridge may be adapted to the case of electrolytes.
+With alternating currents, however, it is impossible to use a
+galvanometer in the usual way. The galvanometer was therefore replaced
+by Kohlrausch by a telephone, which gives a sound when an alternating
+current passes through it. The most common plan of the apparatus is
+shown diagrammatically in fig. 1. The electrolytic cell and a resistance
+box form two arms of the bridge, and the sliding contact is moved along
+the metre wire which forms the other two arms till no sound is heard in
+the telephone. The resistance of the electrolyte is to that of the box
+as that of the right-hand end of the wire is to that of the left-hand
+end. A more accurate method of using alternating currents, and one more
+pleasant to use, gets rid of the telephone (_Phil. Trans._, 1900, 194,
+p. 321). The current from one or two voltaic cells is led to an ebonite
+drum turned by a motor or a hand-wheel and cord. On the drum are fixed
+brass strips with wire brushes touching them in such a manner that the
+current from the brushes is reversed several times in each revolution of
+the drum. The wires from the brushes are connected with the Wheatstone's
+bridge. A moving coil galvanometer is used as indicator, its connexions
+being reversed in time with those of the battery by a slightly narrower
+set of brass strips fixed on the other end of the ebonite commutator.
+Thus any residual current through the galvanometer is direct and not
+alternating. The high moment of inertia of the coil makes the period of
+swing slow compared with the period of alternation of the current, and
+the slight periodic disturbances are thus prevented from affecting the
+galvanometer. When the measured resistance is not altered by increasing
+the speed of the commutator or changing the ratio of the arms of the
+bridge, the disturbing effects may be considered to be eliminated.
+
+[Illustration: FIG. 2.]
+
+[Illustration: FIG. 3.]
+
+The form of vessel chosen to contain the electrolyte depends on the
+order of resistance to be measured. For dilute solutions the shape of
+cell shown in fig. 2 will be found convenient, while for more
+concentrated solutions, that indicated in fig. 3 is suitable. The
+absolute resistances of certain solutions have been determined by
+Kohlrausch by comparison with mercury, and, by using one of these
+solutions in any cell, the constant of that cell may be found once for
+all. From the observed resistance of any given solution in the cell the
+resistance of a centimetre cube--the so-called specific resistance--may
+be calculated. The reciprocal of this, or the conductivity, is a more
+generally useful constant; it is conveniently expressed in terms of a
+unit equal to the reciprocal of an ohm. Thus Kohlrausch found that a
+solution of potassium chloride, containing one-tenth of a gram
+equivalent (7.46 grams) per litre, has at 18 deg. C. a specific
+resistance of 89.37 ohms per centimetre cube, or a conductivity of 1.119
+X 10^-2 mhos or 1.119 X 10^-11 C.G.S. units. As the temperature
+variation of conductivity is large, usually about 2% per degree, it is
+necessary to place the resistance cell in a paraffin or water bath, and
+to observe its temperature with some accuracy.
+
+Another way of eliminating the effects of polarization and of dilution
+has been used by W. Stroud and J. B. Henderson (_Phil. Mag._, 1897 [5],
+43, p. 19). Two of the arms of a Wheatstone's bridge are composed of
+narrow tubes filled with the solution, the tubes being of equal diameter
+but of different length. The other two arms are made of coils of wire of
+equal resistance, and metallic resistance is added to the shorter tube
+till the bridge is balanced. Direct currents of somewhat high
+electromotive force are used to work the bridge. Equal currents then
+flow through the two tubes; the effects of polarization and dilution
+must be the same in each, and the resistance added to the shorter tube
+must be equal to the resistance of a column of liquid the length of
+which is equal to the difference in length of the two tubes.
+
+A somewhat different principle was adopted by E. Bouty in 1884. If a
+current be passed through two resistances in series by means of an
+applied electromotive force, the electric potential falls from one end
+of the resistances to the other, and, if we apply Ohm's law to each
+resistance in succession, we see that, since for each of them E = CR,
+and C the current is the same through both, E the electromotive force or
+fall of potential between the ends of each resistance must be
+proportional to the resistance between them. Thus by measuring the
+potential difference between the ends of the two resistances
+successively, we may compare their resistances. If, on the other hand,
+we can measure the potential difference in some known units, and
+similarly measure the current flowing, we can determine the resistance
+of a single electrolyte. The details of the apparatus may vary, but its
+principle is illustrated in the following description. A narrow glass
+tube is fixed horizontally into side openings in two glass vessels, and
+an electric current passed through it by means of platinum electrodes
+and a battery of considerable electromotive force. In this way a steady
+fall of electric potential is set up along the length of the tube. To
+measure the potential difference between the ends of the tube, tapping
+electrodes are constructed, e.g. by placing zinc rods in vessels with
+zinc sulphate solution and connecting these vessels (by means of thin
+siphon tubes also filled with solution) with the vessels at the ends of
+the long tube which contains the electrolyte to be examined. Whatever be
+the contact potential difference between zinc and its solution, it is
+the same at both ends, and thus the potential difference between the
+zinc rods is equal to that between the liquid at the two ends of the
+tube. This potential difference may be measured without passing any
+appreciable current through the tapping electrodes, and thus the
+resistance of the liquid deduced.
+
+_Equivalent Conductivity of Solutions._--As is the case in the other
+properties of solutions, the phenomena are much more simple when the
+concentration is small than when it is great, and a study of dilute
+solutions is therefore the best way of getting an insight into the
+essential principles of the subject. The foundation of our knowledge was
+laid by Kohlrausch when he had developed the method of measuring
+electrolyte resistance described above. He expressed his results in
+terms of "equivalent conductivity," that is, the conductivity (k) of the
+solution divided by the number (m) of gram-equivalents of electrolyte
+per litre. He finds that, as the concentration diminishes, the value of
+k/m approaches a limit, and eventually becomes constant, that is to say,
+at great dilution the conductivity is proportional to the concentration.
+Kohlrausch first prepared very pure water by repeated distillation and
+found that its resistance continually increased as the process of
+purification proceeded. The conductivity of the water, and of the slight
+impurities which must always remain, was subtracted from that of the
+solution made with it, and the result, divided by m, gave the equivalent
+conductivity of the substance dissolved. This procedure appears
+justifiable, for as long as conductivity is proportional to
+concentration it is evident that each part of the dissolved matter
+produces its own independent effect, so that the total conductivity is
+the sum of the conductivities of the parts; when this ceases to hold,
+the concentration of the solution has in general become so great that
+the conductivity of the solvent may be neglected. The general result of
+these experiments can be represented graphically by plotting k/m as
+ordinates and [root 3]m as abscissae, [root 3]m being a number
+proportional to the reciprocal of the average distance between the
+molecules, to which it seems likely that the molecular conductivity may
+be related. The general types of curve for a simple neutral salt like
+potassium or sodium chloride and for a caustic alkali or acid are shown
+in fig. 4. The curve for the neutral salt comes to a limiting value;
+that for the acid attains a maximum at a certain very small
+concentration, and falls again when the dilution is carried farther. It
+has usually been considered that this destruction of conductivity is due
+to chemical action between the acid and the residual impurities in the
+water. At such great dilution these impurities are present in quantities
+comparable with the amount of acid which they convert into a less highly
+conducting neutral salt. In the case of acids, then, the maximum must be
+taken as the limiting value. The decrease in equivalent conductivity at
+great dilution is, however, so constant that this explanation seems
+insufficient. The true cause of the phenomenon may perhaps be connected
+with the fact that the bodies in which it occurs, acids and alkalis,
+contain the ions, hydrogen in the one case, hydroxyl in the other, which
+are present in the solvent, water, and have, perhaps because of this
+relation, velocities higher than those of any other ions. The values of
+the molecular conductivities of all neutral salts are, at great
+dilution, of the same order of magnitude, while those of acids at their
+maxima are about three times as large. The influence of increasing
+concentration is greater in the case of salts containing divalent ions,
+and greatest of all in such cases as solutions of ammonia and acetic
+acid, which are substances of very low conductivity.
+
+[Illustration: FIG. 4.]
+
+_Theory of Moving Ions._--Kohlrausch found that, when the polarization
+at the electrodes was eliminated, the resistance of a solution was
+constant however determined, and thus established Ohm's Law for
+electrolytes. The law was confirmed in the case of strong currents by G.
+F. Fitzgerald and F. T. Trouton (_B.A. Report_, 1886, p. 312). Now,
+Ohm's Law implies that no work is done by the current in overcoming
+reversible electromotive forces such as those of polarization. Thus the
+molecular interchange of ions, which must occur in order that the
+products may be able to work their way through the liquid and appear at
+the electrodes, continues throughout the solution whether a current is
+flowing or not. The influence of the current on the ions is merely
+directive, and, when it flows, streams of electrified ions travel in
+opposite directions, and, if the applied electromotive force is enough
+to overcome the local polarization, give up their charges to the
+electrodes. We may therefore represent the facts by considering the
+process of electrolysis to be a kind of convection. Faraday's classical
+experiments proved that when a current flows through an electrolyte the
+quantity of substance liberated at each electrode is proportional to its
+chemical equivalent weight, and to the total amount of electricity
+passed. Accurate determinations have since shown that the mass of an ion
+deposited by one electromagnetic unit of electricity, i.e. its
+electro-chemical equivalent, is 1.036 X 10^-4 X its chemical equivalent
+weight. Thus the amount of electricity associated with one
+gram-equivalent of any ion is 10^4/1.036 = 9653 units. Each monovalent
+ion must therefore be associated with a certain definite charge, which
+we may take to be a natural unit of electricity; a divalent ion carries
+two such units, and so on. A cation, i.e. an ion giving up its charge at
+the cathode, as the electrode at which the current leaves the solution
+is called, carries a positive charge of electricity; an anion,
+travelling in the opposite direction, carries a negative charge. It will
+now be seen that the quantity of electricity flowing per second, i.e.
+the current through the solution, depends on (1) the number of the ions
+concerned, (2) the charge on each ion, and (3) the velocity with which
+the ions travel past each other. Now, the number of ions is given by the
+concentration of the solution, for even if all the ions are not actively
+engaged in carrying the current at the same instant, they must, on any
+dynamical idea of chemical equilibrium, be all active in turn. The
+charge on each, as we have seen, can be expressed in absolute units, and
+therefore the velocity with which they move past each other can be
+calculated. This was first done by Kohlrausch (_Gottingen Nachrichten_,
+1876, p. 213, and _Das Leitvermogen der Elektrolyte_, Leipzig, 1898)
+about 1879.
+
+ In order to develop Kohlrausch's theory, let us take, as an example,
+ the case of an aqueous solution of potassium chloride, of
+ concentration n gram-equivalents per cubic centimetre. There will then
+ be n gram-equivalents of potassium ions and the same number of
+ chlorine ions in this volume. Let us suppose that on each
+ gram-equivalent of potassium there reside +e units of electricity, and
+ on each gram-equivalent of chlorine ions -e units. If u denotes the
+ average velocity of the potassium ion, the positive charge carried per
+ second across unit area normal to the flow is n e u. Similarly, if v
+ be the average velocity of the chlorine ions, the negative charge
+ carried in the opposite direction is n e v. But positive electricity
+ moving in one direction is equivalent to negative electricity moving
+ in the other, so that, before changes in concentration sensibly
+ supervene, the total current, C, is ne(u + v). Now let us consider the
+ amounts of potassium and chlorine liberated at the electrodes by this
+ current. At the cathode, if the chlorine ions were at rest, the excess
+ of potassium ions would be simply those arriving in one second,
+ namely, nu. But since the chlorine ions move also, a further
+ separation occurs, and nv potassium ions are left without partners.
+ The total number of gram-equivalents liberated is therefore n(u + v).
+ By Faraday's law, the number of grams liberated is equal to the
+ product of the current and the electro-chemical equivalent of the ion;
+ the number of gram-equivalents therefore must be equal to [eta]C,
+ where [eta] denotes the electro-chemical equivalent of hydrogen in
+ C.G.S. units. Thus we get
+
+ n(u + v) = [eta]C = [eta]ne(u + v),
+
+ and it follows that the charge, e, on 1 gram-equivalent of each kind
+ of ion is equal to 1/[eta]. We know that Ohm's Law holds good for
+ electrolytes, so that the current C is also given by k.dP/dx, where k
+ denotes the conductivity of the solution, and dP/dx the potential
+ gradient, i.e. the change in potential per unit length along the lines
+ of current flow. Thus
+
+ n
+ -----(u + v) = kdP/dx;
+ [eta]
+
+ therefore
+
+ k dP
+ u + v = [eta] - --.
+ n dx
+
+ Now [eta] is 1.036 X 10^-4, and the concentration of a solution is
+ usually expressed in terms of the number, m, of gram-equivalents per
+ litre instead of per cubic centimetre. Therefore
+
+ k dP
+ u + v = 1.036 X 10^-1 - -- --.
+ m dx
+
+ When the potential gradient is one volt (10^8 C.G.S. units) per
+ centimetre this becomes
+
+ u + v = 1.036 X 10^-7 X k/m.
+
+ Thus by measuring the value of k/m, which is known as the equivalent
+ conductivity of the solution, we can find u + v, the velocity of the
+ ions relative to each other. For instance, the equivalent conductivity
+ of a solution of potassium chloride containing one-tenth of a
+ gram-equivalent per litre is 1119 X 10^-13 C.G.S. units at 18 deg. C.
+ Therefore
+
+ u + v = 1.036 X 10^7 X 1119 X 10^-13
+ = 1.159 X 10^-3 = 0.001159 cm. per sec.
+
+ In order to obtain the absolute velocities u and v, we must find some
+ other relation between them. Let us resolve u into 1/2(u + v) in one
+ direction, say to the right, and 1/2(u - v) to the left. Similarly v
+ can be resolved into 1/2(v+u) to the left and 1/2(v-u) to the right.
+ On pairing these velocities we have a combined movement of the ions to
+ the right, with a speed of 1/2(u - v) and a drift right and left, past
+ each other, each ion travelling with a speed of 1/2(u + v),
+ constituting the electrolytic separation. If u is greater than v, the
+ combined movement involves a concentration of salt at the cathode, and
+ a corresponding dilution at the anode, and _vice versa_. The rate at
+ which salt is electrolysed, and thus removed from the solution at each
+ electrode, is 1/2(u + v). Thus the total loss of salt at the cathode
+ is 1/2(u + v) - 1/2(u - v) or v, and at the anode, 1/2(v + u) - 1/2(v
+ - u), or u. Therefore, as is explained in the article ELECTROLYSIS, by
+ measuring the dilution of the liquid round the electrodes when a
+ current passed, W. Hittorf (_Pogg. Ann._, 1853-1859, 89, p. 177; 98,
+ p. 1; 103, p. 1; 106, pp. 337 and 513) was able to deduce the ratio of
+ the two velocities, for simple salts when no complex ions are present,
+ and many further experiments have been made on the subject (see _Das
+ Leitvermogen der Elektrolyte_).
+
+ By combining the results thus obtained with the sum of the velocities,
+ as determined from the conductivities, Kohlrausch calculated the
+ absolute velocities of different ions under stated conditions. Thus,
+ in the case of the solution of potassium chloride considered above,
+ Hittorf's experiments show us that the ratio of the velocity of the
+ anion to that of the cation in this solution is .51 : .49. The
+ absolute velocity of the potassium ion under unit potential gradient
+ is therefore 0.000567 cm. per sec., and that of the chlorine ion
+ 0.000592 cm. per sec. Similar calculations can be made for solutions
+ of other concentrations, and of different substances.
+
+Table IX. shows Kohlrausch's values for the ionic velocities of three
+chlorides of alkali metals at 18 deg. C, calculated for a potential
+gradient of 1 volt per cm.; the numbers are in terms of a unit equal to
+10^-6 cm. per sec.:--
+
+ TABLE IX.
+
+ +--------------------------------------------+----------------+
+ | | KCl | NaCl | LiCl |
+ +---------+-----------------+----------------+----------------+
+ | m | u + v u v | u + v u v |u + v u v |
+ +---------+-----------------+----------------+----------------+
+ | 0 | 1350 660 690 | 1140 450 690 | 1050 360 690 |
+ | 0.0001 | 1335 654 681 | 1129 448 681 | 1037 356 681 |
+ | .001 | 1313 643 670 | 1110 440 670 | 1013 343 670 |
+ | .01 | 1263 619 644 | 1059 415 644 | 962 318 644 |
+ | .03 | 1218 597 621 | 1013 390 623 | 917 298 619 |
+ | .1 | 1153 564 589 | 952 360 592 | 853 259 594 |
+ | .3 | 1088 531 557 | 876 324 552 | 774 217 557 |
+ | 1.0 | 1011 491 520 | 765 278 487 | 651 169 482 |
+ | 3.0 | 911 442 469 | 582 206 376 | 463 115 348 |
+ | 5.0 | | 438 153 285 | 334 80 254 |
+ | 10.0 | | | 117 25 92 |
+ +---------+-----------------+----------------+----------------+
+
+These numbers show clearly that there is an increase in ionic velocity
+as the dilution proceeds. Moreover, if we compare the values for the
+chlorine ion obtained from observations on these three different salts,
+we see that as the concentrations diminish the velocity of the chlorine
+ion becomes the same in all of them. A similar relation appears in other
+cases, and, in general, we may say that at great dilution the velocity
+of an ion is independent of the nature of the other ion present. This
+introduces the conception of specific ionic velocities, for which some
+values at 18 deg. C. are given by Kohlrausch in Table X.:--
+
+ Table X.
+
+ +------------------------------------+
+ | K 66 X 10^-5 cms. per sec. |
+ | Na 45 " " |
+ | Li 36 " " |
+ | NH4 66 " " |
+ | H 320 " " |
+ | Ag 57 " " |
+ | Cl 69 " " |
+ | I 69 " " |
+ | NO3 64 " " |
+ | OH 162 " " |
+ | C2H3O2 36 " " |
+ | C3H5O2 33 " " |
+ +------------------------------------+
+
+Having obtained these numbers we can deduce the conductivity of the
+dilute solution of any salt, and the comparison of the calculated with
+the observed values furnished the first confirmation of Kohlrausch's
+theory. Some exceptions, however, are known. Thus acetic acid and
+ammonia give solutions of much lower conductivity than is indicated by
+the sum of the specific ionic velocities of their ions as determined
+from other compounds. An attempt to find in Kohlrausch's theory some
+explanation of this discrepancy shows that it could be due to one of two
+causes. Either the velocities of the ions must be much less in these
+solutions than in others, or else only a fractional part of the number
+of molecules present can be actively concerned in conveying the current.
+We shall return to this point later.
+
+ _Friction on the Ions._--It is interesting to calculate the magnitude
+ of the forces required to drive the ions with a certain velocity. If
+ we have a potential gradient of 1 volt per centimetre the electric
+ force is 10^8 in C.G.S. units. The charge of electricity on 1
+ gram-equivalent of any ion is 1/.0001036 = 9653 units, hence the
+ mechanical force acting on this mass is 9653 X 10^8 dynes. This, let
+ us say, produces a velocity u; then the force required to produce unit
+ velocity is PA = 9.653 X 10^11/u dynes = 9.84 X 10^5/u kilograms-weight.
+ If the ion have an equivalent weight A, the force producing unit
+ velocity when acting on 1 gram is P1 = 9.84 X 10^5/Au kilograms-weight.
+ Thus the aggregate force required to drive 1 gram of potassium ions
+ with a velocity of 1 centimetre per second through a very dilute
+ solution must be equal to the weight of 38 million kilograms.
+
+ Table XI.
+
+ +-------------------------------+
+ | Kilograms-weight. |
+ +-------------------------------+
+ | PA P1 |
+ |K 15 X 10^8 38 X 10^6 |
+ |Na 22 " 95 " |
+ |Li 27 " 390 " |
+ |NH4 15 " 83 " |
+ |H 3.1 " 310 " |
+ |Ag 17 " 16 " |
+ |Cl 14 " 40 " |
+ |I 14 " 11 " |
+ |NO3 15 " 25 " |
+ |OH 5.4 " 32 " |
+ |C2H8O2 27 " 46 " |
+ |C3H5O2 30 " 41 " |
+ +------------- -----------------+
+
+ Since the ions move with uniform velocity, the frictional resistances
+ brought into play must be equal and opposite to the driving forces,
+ and therefore these numbers also represent the ionic friction
+ coefficients in very dilute solutions at 18 deg. C.
+
+_Direct Measurement of Ionic Velocities._--Sir Oliver Lodge was the
+first to directly measure the velocity of an ion (_B.A. Report_, 1886,
+p. 389). In a horizontal glass tube connecting two vessels filled with
+dilute sulphuric acid he placed a solution of sodium chloride in solid
+agar-agar jelly. This solid solution was made alkaline with a trace of
+caustic soda in order to bring out the red colour of a little
+phenol-phthalein added as indicator. An electric current was then passed
+from one vessel to the other. The hydrogen ions from the anode vessel of
+acid were thus carried along the tube, and, as they travelled,
+decolourized the phenol-phthalein. By this method the velocity of the
+hydrogen ion through a jelly solution under a known potential gradient
+was observed to about 0.0026 cm. per sec, a number of the same order as
+that required by Kohlrausch's theory. Direct determinations of the
+velocities of a few other ions have been made by W. C. D. Whetham
+(_Phil. Trans._ vol. 184, A, p. 337; vol. 186, A, p. 507; _Phil. Mag._,
+October 1894). Two solutions having one ion in common, of equivalent
+concentrations, different densities, different colours, and nearly equal
+specific resistances, were placed one over the other in a vertical glass
+tube. In one case, for example, decinormal solutions of potassium
+carbonate and potassium bichromate were used. The colour of the latter
+is due to the presence of the bichromate group, Cr2O7. When a current
+was passed across the junction, the anions CO3 and Cr2O7 travelled in
+the direction opposite to that of the current, and their velocity could
+be determined by measuring the rate at which the colour boundary moved.
+Similar experiments were made with alcoholic solutions of cobalt salts,
+in which the velocities of the ions were found to be much less than in
+water. The behaviour of agar jelly was then investigated, and the
+velocity of an ion through a solid jelly was shown to be very little
+less than in an ordinary liquid solution. The velocities could therefore
+be measured by tracing the change in colour of an indicator or the
+formation of a precipitate. Thus decinormal jelly solutions of barium
+chloride and sodium chloride, the latter containing a trace of sodium
+sulphate, were placed in contact. Under the influence of an
+electromotive force the barium ions moved up the tube, disclosing their
+presence by the trace of insoluble barium sulphate formed. Again, a
+measurement of the velocity of the hydrogen ion, when travelling through
+the solution of an acetate, showed that its velocity was then only about
+the one-fortieth part of that found during its passage through
+chlorides. From this, as from the measurements on alcohol solutions, it
+is clear that where the equivalent conductivities are very low the
+effective velocities of the ions are reduced in the same proportion.
+
+Another series of direct measurements has been made by Orme Masson
+(_Phil. Trans._ vol. 192, A, p. 331). He placed the gelatine solution of
+a salt, potassium chloride, for example, in a horizontal glass tube, and
+found the rate of migration of the potassium and chlorine ions by
+observing the speed at which they were replaced when a coloured anion,
+say, the Cr2O7 from a solution of potassium bichromate, entered the tube
+at one end, and a coloured cation, say, the Cu from copper sulphate, at
+the other. The coloured ions are specifically slower than the colourless
+ions which they follow, and in this case it follows that the coloured
+solution has a higher resistance than the colourless. For the same
+current, therefore, the potential gradient is higher in the coloured
+solution and lower in the colourless one. Thus a coloured ion which gets
+in front of the advancing boundary finds itself acted on by a smaller
+force and falls back into line, while a straggling colourless ion is
+pushed forward again. Hence a sharp boundary is preserved. B. D. Steele
+has shown that with these sharp boundaries the use of coloured ions is
+unnecessary, the junction line being visible owing to the difference in
+the optical refractive indices of two colourless solutions. Once the
+boundary is formed, too, no gelatine is necessary, and the motion can be
+watched through liquid aqueous solutions (see R. B. Denison and B. D.
+Steele, _Phil. Trans._, 1906).
+
+All the direct measurements which have been made on simple binary
+electrolytes agree with Kohlrausch's results within the limits of
+experimental error. His theory, therefore, probably holds good in such
+cases, whatever be the solvent, if the proper values are given to the
+ionic velocities, i.e. the values expressing the velocities with which
+the ions actually move in the solution of the strength taken, and under
+the conditions of the experiment. If we know the specific velocity of
+any one ion, we can deduce, from the conductivity of very dilute
+solutions, the velocity of any other ion with which it may be
+associated, a proceeding which does not involve the difficult task of
+determining the migration constant of the compound. Thus, taking the
+specific ionic velocity of hydrogen as 0.00032 cm. per second, we can
+find, by determining the conductivity of dilute solutions of any acid,
+the specific velocity of the acid radicle involved. Or again, since we
+know the specific velocity of silver, we can find the velocities of a
+series of acid radicles at great dilution by measuring the conductivity
+of their silver salts.
+
+ By such methods W. Ostwald, G. Bredig and other observers have found
+ the specific velocities of many ions both of inorganic and organic
+ compounds, and examined the relation between constitution and ionic
+ velocity. The velocity of elementary ions is found to be a periodic
+ function of the atomic weight, similar elements lying on corresponding
+ portions of a curve drawn to express the relation between these two
+ properties. Such a curve much resembles that giving the relation
+ between atomic weight and viscosity in solution. For complex ions the
+ velocity is largely an additive property; to a continuous additive
+ change in the composition of the ion corresponds a continuous but
+ decreasing change in the velocity. The following table gives Ostwald's
+ results for the formic acid series:--
+
+ Table XII.
+
+ +----------------------+----------+---------------------+
+ | | Velocity.| Difference for CH2. |
+ +----------------------+----------+---------------------+
+ | Formic acid HCO2 | 51.2 | .. |
+ | Acetic " H3C2O2 | 38.3 | -12.9 |
+ | Propionic " H5C3O2 | 34.3 | - 4.0 |
+ | Butyric " H7C4O2 | 30.8 | - 3.5 |
+ | Valeric " H9C5O2 | 28.8 | - 2.0 |
+ | Caprionic " H11C6O2 | 27.4 | - 1.4 |
+ +----------------------+----------+---------------------+
+
+_Nature of Electrolytes._--We have as yet said nothing about the
+fundamental cause of electrolytic activity, nor considered why, for
+example, a solution of potassium chloride is a good conductor, while a
+solution of sugar allows practically no current to pass.
+
+All the preceding account of the subject is, then, independent of any
+view we may take of the nature of electrolytes, and stands on the basis
+of direct experiment. Nevertheless, the facts considered point to a very
+definite conclusion. The specific velocity of an ion is independent of
+the nature of the opposite ion present, and this suggests that the ions
+themselves, while travelling through the liquid, are dissociated from
+each other. Further evidence, pointing in the same direction, is
+furnished by the fact that since the conductivity is proportional to the
+concentration at great dilution, the equivalent-conductivity, and
+therefore the ionic velocity, is independent of it. The importance of
+this relation will be seen by considering the alternative to the
+dissociation hypothesis. If the ions are not permanently free from each
+other their mobility as parts of the dissolved molecules must be secured
+by continual interchanges. The velocity with which they work their way
+through the liquid must then increase as such molecular rearrangements
+become more frequent, and will therefore depend on the number of solute
+molecules, i.e. on the concentration. On this supposition the observed
+constancy of velocity would be impossible. We shall therefore adopt as a
+wording hypothesis the theory, confirmed by other phenomena (see
+ELECTROLYSIS), that an electrolyte consists of dissociated ions.
+
+It will be noticed that neither the evidence in favour of the
+dissociation theory which is here considered, nor that described in the
+article ELECTROLYSIS, requires more than the effective dissociation of
+the ions from each other. They may well be connected in some way with
+solvent molecules, and there are several indications that an ion
+consists of an electrified part of the molecule of the dissolved salt
+with an attendant atmosphere of solvent round it. The conductivity of a
+salt solution depends on two factors--(1) the fraction of the salt
+ionized; (2) the velocity with which the ions, when free from each
+other, move under the electric forces.[12] When a solution is heated,
+both these factors may change. The coefficient of ionization usually,
+though not always, decreases; the specific ionic velocities increase.
+Now the rate of increase with temperature of these ionic velocities is
+very nearly identical with the rate of decrease of the viscosity of the
+liquid. If the curves obtained by observations at ordinary temperatures
+be carried on they indicate a zero of fluidity and a zero of ionic
+velocity about the same point, 38.5 deg. C. below the freezing point of
+water (Kohlrausch, _Sitz. preuss. Akad. Wiss._, 1901, 42, p. 1026). Such
+relations suggest that the frictional resistance to the motion of an ion
+is due to the ordinary viscosity of the liquid, and that the ion is
+analogous to a body of some size urged through a viscous medium rather
+than to a particle of molecular dimensions finding its way through a
+crowd of molecules of similar magnitude. From this point of view W. K.
+Bousfield has calculated the sizes of ions on the assumption that
+Stokes's theory of the motion of a small sphere through a viscous medium
+might be applied (_Zeits. phys. Chem._, 1905, 53, p. 257; _Phil. Trans._
+A, 1906, 206, p. 101). The radius of the potassium or chlorine ion with
+its envelope of water appears to be about 1.2 X 10^-8 centimetres.
+
+ For the bibliography of electrolytic conduction see ELECTROLYSIS. The
+ books which deal more especially with the particular subject of the
+ present article are _Das Leitvermogen der Elektrolyte_, by F.
+ Kohlrausch and L. Holborn (Leipzig, 1898), and _The Theory of Solution
+ and Electrolysis_, by W. C. D. Whetham (Cambridge, 1902).
+ (W. C. D. W.)
+
+
+III. ELECTRIC CONDUCTION THROUGH GASES
+
+A gas such as air when it is under normal conditions conducts
+electricity to a small but only to a very small extent, however small
+the electric force acting on the gas may be. The electrical conductivity
+of gases not exposed to special conditions is so small that it was only
+definitely established in the early years of the 20th century, although
+it had engaged the attention of physicists for more than a hundred
+years. It had been known for a long time that a body charged with
+electricity slowly lost its charge even when insulated with the greatest
+care, and though long ago some physicists believed that part of the leak
+of electricity took place through the air, the general view seems to
+have been that it was due to almost unavoidable defects in the
+insulation or to dust in the air, which after striking the charged body
+was repelled from it and went off with some of the charge. C. A.
+Coulomb, who made some very careful experiments which were published in
+1785 (_Mem. de l'Acad. des Sciences_, 1785, p. 612), came to the
+conclusion that after allowing for the leakage along the threads which
+supported the charged body there was a balance over, which he attributed
+to leakage through the air. His view was that when the molecules of air
+come into contact with a charged body some of the electricity goes on to
+the molecules, which are then repelled from the body carrying their
+charge with them. We shall see later that this explanation is not
+tenable. C. Matteucci (_Ann. chim. phys._, 1850, 28, p. 390) in 1850
+also came to the conclusion that the electricity from a charged body
+passes through the air; he was the first to prove that the rate at
+which electricity escapes is less when the pressure of the gas is low
+than when it is high. He found that the rate was the same whether the
+charged body was surrounded by air, carbonic acid or hydrogen.
+Subsequent investigations have shown that the rate in hydrogen is in
+general much less than in air. Thus in 1872 E. G. Warburg (_Pogg. Ann._,
+1872, 145, p. 578) found that the leak through hydrogen was only about
+one-half of that through air: he confirmed Matteucci's observations on
+the effect of pressure on the rate of leak, and also found that it was
+the same whether the gas was dry or damp. He was inclined to attribute
+the leak to dust in the air, a view which was strengthened by an
+experiment of J. W. Hittorf's (_Wied. Ann._, 1879, 7, p. 595), in which
+a small carefully insulated electroscope, placed in a small vessel
+filled with carefully filtered gas, retained its charge for several
+days; we know now that this was due to the smallness of the vessel and
+not to the absence of dust, as it has been proved that the rate of leak
+in small vessels is less than in large ones.
+
+Great light was thrown on this subject by some experiments on the rates
+of leak from charged bodies in closed vessels made almost simultaneously
+by H. Geitel (_Phys. Zeit._, 1900, 2, p. 116) and C. T. R. Wilson
+(_Proc. Camb. Phil._ Soc., 1900, 11, p. 32). These observers established
+that (1) the rate of escape of electricity in a closed vessel is much
+smaller than in the open, and the larger the vessel the greater is the
+rate of leak; and (2) the rate of leak does not increase in proportion
+to the differences of potential between the charged body and the walls
+of the vessel: the rate soon reaches a limit beyond which it does not
+increase, however much the potential difference may be increased,
+provided, of course, that this is not great enough to cause sparks to
+pass from the charged body. On the assumption that the maximum leak is
+proportional to the volume, Wilson's experiments, which were made in
+vessels less than 1 litre in volume, showed that in dust-free air at
+atmospheric pressure the maximum quantity of electricity which can
+escape in one second from a charged body in a closed volume of V cubic
+centimetres is about 10^-8V electrostatic units. E. Rutherford and S. T.
+Allan (_Phys. Zeit._, 1902, 3, p. 225), working in Montreal, obtained
+results in close agreement with this. Working between pressures of from
+43 to 743 millimetres of mercury, Wilson showed that the maximum rate of
+leak is very approximately proportional to the pressure; it is thus
+exceedingly small when the pressure is low--a result illustrated in a
+striking way by an experiment of Sir W. Crookes (_Proc. Roy. Soc._,
+1879, 28, p. 347) in which a pair of gold leaves retained an electric
+charge for several months in a very high vacuum. Subsequent experiments
+have shown that it is only in very small vessels that the rate of leak
+is proportional to the volume and to the pressure; in large vessels the
+rate of leak per unit volume is considerably smaller than in small ones.
+In small vessels the maximum rate of leak in different gases, is, with
+the exception of hydrogen, approximately proportional to the density of
+the gas. Wilson's results on this point are shown in the following table
+(Proc. Roy. Soc., 1901, 60, p. 277):--
+
+ +---------+------------------------+-----------------+
+ | Gas. | Relative Rate of Leak. | _Rate of Leak._ |
+ | | | Sp. Gr. |
+ +---------+------------------------+-----------------+
+ | Air | 1.00 | 1 |
+ | H2 | .184 | 2.7 |
+ | CO2 | 1.69 | 1.10 |
+ | SO2 | 2.64 | 1.21 |
+ | CH3Cl | 4.7 | 1.09 |
+ | Ni(CO)4 | 5.1 | .867 |
+ +---------+------------------------+-----------------+
+
+The rate of leak of electricity through gas contained in a closed vessel
+depends to some extent on the material of which the walls of the vessel
+are made; thus it is greater, other circumstances being the same, when
+the vessel is made of lead than when it is made of aluminium. It also
+varies, as Campbell and Wood (_Phil. Mag._ [6], 13, p. 265) have shown,
+with the time of the day, having a well-marked minimum at about 3
+o'clock in the morning: it also varies from month to month. Rutherford
+(_Phys. Rev._, 1903, 16, p. 183), Cooke (_Phil. Mag._, 1903 [6], 6, p.
+403) and M'Clennan and Burton (_Phys. Rev._, 1903, 16, p. 184) have
+shown that the leak in a closed vessel can be reduced by about 30% by
+surrounding the vessel with sheets of thick lead, but that the reduction
+is not increased beyond this amount, however thick the lead sheets may
+be. This result indicates that part of the leak is due to a very
+penetrating kind of radiation, which can get through the thin walls of
+the vessel but is stopped by the thick lead. A large part of the leak we
+are describing is due to the presence of radioactive substances such as
+radium and thorium in the earth's crust and in the walls of the vessel,
+and to the gaseous radioactive emanations which diffuse from them into
+the atmosphere. This explains the very interesting effect discovered by
+J. Elster and H. Geitel (_Phys. Zeit._, 1901, 2, p. 560), that the rate
+of leak in caves and cellars when the air is stagnant and only renewed
+slowly is much greater than in the open air. In some cases the
+difference is very marked; thus they found that in the cave called the
+Baumannshohle in the Harz mountains the electricity escaped at seven
+times the rate it did in the air outside. In caves and cellars the
+radioactive emanations from the walls can accumulate and are not blown
+away as in the open air.
+
+The electrical conductivity of gases in the normal state is, as we have
+seen, exceedingly small, so small that the investigation of its
+properties is a matter of considerable difficulty; there are, however,
+many ways by which the electrical conductivity of a gas can be increased
+so greatly that the investigation becomes comparatively easy. Among such
+methods are raising the temperature of the gas above a certain point.
+Gases drawn from the neighbourhood of flames, electric arcs and sparks,
+or glowing pieces of metal or carbon are conductors, as are also gases
+through which Rontgen or cathode rays or rays of positive electricity
+are passing; the rays from the radioactive metals, radium, thorium,
+polonium and actinium, produce the same effect, as does also
+ultra-violet light of exceedingly short wave-length. The gas, after
+being made a conductor of electricity by any of these means, is found to
+possess certain properties; thus it retains its conductivity for some
+little time after the agent which made it a conductor has ceased to act,
+though the conductivity diminishes very rapidly and finally gets too
+small to be appreciable.
+
+[Illustration: FIG. 5.]
+
+This and several other properties of conducting gas may readily be
+proved by the aid of the apparatus represented in fig. 5. V is a testing
+vessel in which an electroscope is placed. Two tubes A and C are fitted
+into the vessel, A being connected with a water pump, while the far end
+of C is in the region where the gas is exposed to the agent which makes
+it a conductor of electricity. Let us suppose that the gas is made
+conducting by Rontgen rays produced by a vacuum tube which is placed in
+a box, covered except for a window at B with lead so as to protect the
+electroscope from the direct action of the rays. If a slow current of
+air is drawn by the water pump through the testing vessel, the charge on
+the electroscope will gradually leak away. The leak, however, ceases
+when the current of air is stopped. This result shows that the gas
+retains its conductivity during the time taken by it to pass from one
+end to the other of the tube C.
+
+The gas loses its conductivity when filtered through a plug of
+glass-wool, or when it is made to bubble through water. This can readily
+be proved by inserting in the tube C a plug of glass-wool or a water
+trap; then if by working the pump a little harder the same current of
+air is produced as before, it will be found that the electroscope will
+now retain its charge, showing that the conductivity can, as it were, be
+filtered out of the gas. The conductivity can also be removed from the
+gas by making the gas traverse a strong electric field. We can show this
+by replacing the tube C by a metal tube with an insulated wire passing
+down the axis of the tube. If there is no potential difference between
+the wire and the tube then the electroscope will leak when a current of
+air is drawn through the vessel, but the leak will stop if a
+considerable difference of potential is maintained between the wire and
+the tube: this shows that a strong electric field removes the
+conductivity from the gas.
+
+The fact that the conductivity of the gas is removed by filtering shows
+that it is due to something mixed with the gas which is removed from it
+by filtration, and since the conductivity is also removed by an electric
+field, the cause of the conductivity must be charged with electricity so
+as to be driven to the sides of the tube by the electric force. Since
+the gas as a whole is not electrified either positively or negatively,
+there must be both negative and positive charges in the gas, the amount
+of electricity of one sign being equal to that of the other. We are thus
+led to the conclusion that the conductivity of the gas is due to
+electrified particles being mixed up with the gas, some of these
+particles having charges of positive electricity, others of negative.
+These electrified particles are called _ions_, and the process by which
+the gas is made a conductor is called the ionization of the gas. We
+shall show later that the charges and masses of the ions can be
+determined, and that the gaseous ions are not identical with those met
+with in the electrolysis of solutions.
+
+[Illustration: FIG. 6.]
+
+One very characteristic property of conduction of electricity through a
+gas is the relation between the current through the gas and the electric
+force which gave rise to it. This relation is not in general that
+expressed by Ohm's law, which always, as far as our present knowledge
+extends, expresses the relation for conduction through metals and
+electrolytes. With gases, on the other hand, it is only when the current
+is very small that Ohm's law is true. If we represent graphically by
+means of a curve the relation between the current passing between two
+parallel metal plates separated by ionized gas and the difference of
+potential between the plates, the curve is of the character shown in
+fig. 6 when the ordinates represent the current and the abscissae the
+difference of potential between the plates. We see that when the
+potential difference is very small, i.e. close to the origin, the curve
+is approximately straight, but that soon the current increases much less
+rapidly than the potential difference, and that a stage is reached when
+no appreciable increase of current is produced when the potential
+difference is increased; when this stage is reached the current is
+constant, and this value of the current is called the "saturation"
+value. When the potential difference approaches the value at which
+sparks would pass through the gas, the current again increases with the
+potential difference; thus the curve representing the relation between
+the current and potential difference over very wide ranges of potential
+difference has the shape shown in fig. 7; curves of this kind have been
+obtained by von Schweidler (_Wien. Ber._, 1899, 108, p. 273), and J. E.
+S. Townsend (_Phil. Mag._, 1901 [6], 1, p. 198). We shall discuss later
+the causes of the rise in the current with large potential differences,
+when we consider ionization by collision.
+
+ The general features of the earlier part of the curve are readily
+ explained on the ionization hypothesis. On this view the Rontgen rays
+ or other ionizing agent acting on the gas between the plates, produces
+ positive and negative ions at a definite rate. Let us suppose that q
+ positive and q negative ions are by this means produced per second
+ between the plates; these under the electric force will tend to move,
+ the positive ones to the negative plate, the negative ones to the
+ positive. Some of these ions will reach the plate, others before
+ reaching the plate will get so near one of the opposite sign that the
+ attraction between them will cause them to unite and form an
+ electrically neutral system; when they do this they end their
+ existence as ions. The current between the plates is proportional to
+ the number of ions which reach the plates per second. Now it is
+ evident that we cannot go on taking more ions out of the gas than are
+ produced; thus we cannot, when the current is steady, have more than q
+ positive ions driven to the negative plate per second, and the same
+ number of negative ions to the positive. If each of the positive ions
+ carries a charge of e units of positive electricity, and if there is
+ an equal and opposite charge on each negative ion, then the maximum
+ amount of electricity which can be given to the plates per second is
+ qe, and this is equal to the saturation current. Thus if we measure
+ the saturation current, we get a direct measure of the ionization, and
+ this does not require us to know the value of any quantity except the
+ constant charge on the ion. If we attempted to deduce the amount of
+ ionization by measurements of the current before it was saturated, we
+ should require to know in addition the velocity with which the ions
+ move under a given electric force, the time that elapses between the
+ liberation of an ion and its combination with one of the opposite
+ sign, and the potential difference between the plates. Thus if we wish
+ to measure the amount of ionization in a gas we should be careful to
+ see that the current is saturated.
+
+ [Illustration: FIG. 7.]
+
+The difference between conduction through gases and through metals is
+shown in a striking way when we use potential differences large enough
+to produce the saturation current. Suppose we have got a potential
+difference between the plates more than sufficient to produce the
+saturation current, and let us increase the distance between the plates.
+If the gas were to act like a metallic conductor this would diminish the
+current, because the greater length would involve a greater resistance
+in the circuit. In the case we are considering the separation of the
+plates will _increase_ the current, because now there is a larger volume
+of gas exposed to the rays; there are therefore more ions produced, and
+as the saturation current is proportional to the number of ions the
+saturation current is increased. If the potential difference between the
+plates were much less than that required to saturate the current, then
+increasing the distance would diminish the current; the gas for such
+potential differences obeys Ohm's law and the behaviour of the gaseous
+resistance is therefore similar to that of a metallic one.
+
+In order to produce the saturation current the electric field must be
+strong enough to drive each ion to the electrode before it has time to
+enter into combination with one of the opposite sign. Thus when the
+plates in the preceding example are far apart, it will take a larger
+potential difference to produce this current than when the plates are
+close together. The potential difference required to saturate the
+current will increase as the square of the distance between the plates,
+for if the ions are to be delivered in a given time to the plates their
+speed must be proportional to the distance between the plates. But the
+speed is proportional to the electric force acting on the ion; hence the
+electric force must be proportional to the distance between the plates,
+and as in a uniform field the potential difference is equal to the
+electric force multiplied by the distance between the plates, the
+potential difference will vary as the square of this distance.
+
+The potential difference required to produce saturation will, other
+circumstances being the same, increase with the amount of ionization,
+for when the number of ions is large and they are crowded together, the
+time which will elapse before a positive one combines with a negative
+will be smaller than when the number of ions is small. The ions have
+therefore to be removed more quickly from the gas when the ionization is
+great than when it is small; thus they must move at a higher speed and
+must therefore be acted upon by a larger force.
+
+
+When the ions are not removed from the gas, they will increase until the
+number of ions of one sign which combine with ions of the opposite sign
+in any time is equal to the number produced by the ionizing agent in
+that time. We can easily calculate the number of free ions at any time
+after the ionizing agent has commenced to act.
+
+ Let q be the number of ions (positive or negative) produced in one
+ cubic centimetre of the gas per second by the ionizing agent, n1, n2,
+ the number of free positive and negative ions respectively per cubic
+ centimetre of the gas. The number of collisions between positive and
+ negative ions per second in one cubic centimetre of the gas is
+ proportional to n1n2. If a certain fraction of the collisions between
+ the positive and negative ions result in the formation of an
+ electrically neutral system, the number of ions which disappear per
+ second on a cubic centimetre will be equal to [alpha]n1 n2, where
+ [alpha] is a quantity which is independent of n1, n2; hence if t is
+ the time since the ionizing agent was applied to the gas, we have
+
+ dn1/dt = q - [alpha]n1 n2, dn2/dt = q - [alpha]n1 n2.
+
+ Thus n1 - n2 is constant, so if the gas is uncharged to begin with, n1
+ will always equal n2. Putting n1 = n2 = n we have
+
+ dn/dt = q - [alpha]n^2 (1),
+
+ the solution of which is, since n = 0 when t = 0,
+
+ k([epsilon]^{2k[alpha]t} - 1)
+ n = ---------------------------- (2)
+ [epsilon]^{2k[alpha]t} + 1
+
+ if k^2 = q/[alpha]. Now the number of ions when the gas has reached a
+ steady state is got by putting t equal to infinity in the preceding
+ equation, and is therefore given by the equation
+
+ n0 = k = [root](q/[alpha]).
+
+ We see from equation (1) that the gas will not approximate to its
+ steady state until 2k[alpha]t is large, that is until t is large
+ compared with 1/2k[alpha] or with 1/2[root](q[alpha]). We may thus
+ take 1/2[root](q[alpha]) as a measure of the time taken by the gas to
+ reach a steady state when exposed to an ionizing agent; as this time
+ varies inversely as [root]q we see that when the ionization is feeble
+ it may take a very considerable time for the gas to reach a steady
+ state. Thus in the case of our atmosphere where the production of ions
+ is only at the rate of about 30 per cubic centimetre per second, and
+ where, as we shall see, [alpha] is about 10^-6, it would take some
+ minutes for the ionization in the air to get into a steady state if
+ the ionizing agent were suddenly applied.
+
+ We may use equation (1) to determine the rate at which the ions
+ disappear when the ionizing agent is removed. Putting q=0 in that
+ equation we get dn/[alpha]t = -[alpha]n^2.
+
+ Hence n = n0/(1 + n0[alpha]t) (3),
+
+ where n0 is the number of ions when t = 0. Thus the number of ions
+ falls to one-half its initial value in the time 1/n0[alpha]. The
+ quantity [alpha] is called the _coefficient of recombination_, and its
+ value for different gases has been determined by Rutherford (_Phil.
+ Mag._ 1897 [5], 44, p. 422), Townsend (_Phil. Trans._, 1900, 193, p.
+ 129), McClung (_Phil. Mag._, 1902 [6], 3, p. 283), Langevin (_Ann.
+ chim. phys._ [7], 28, p. 289), Retschinsky (_Ann. d. Phys._, 1905, 17,
+ p. 518), Hendred (_Phys. Rev._, 1905, 21, p. 314). The values of
+ [alpha]/e, e being the charge on an ion in electrostatic measure as
+ determined by these observers for different gases, is given in the
+ following table:--
+
+ +-----+----------+----------+----------+------------+----------+
+ | | Townsend.| McClung. | Langevin.|Retschinsky.| Hendred. |
+ +-----+----------+----------+----------+------------+----------+
+ | Air | 3420 | 3380 | 3200 | 4140 | 3500 |
+ | O2 | 3380 | | | | |
+ | CO2 | 3500 | 3490 | 3400 | | |
+ | H2 | 3020 | 2940 | | | |
+ +-----+----------+----------+----------+------------+----------+
+
+ The gases in these experiments were carefully dried and free from
+ dust; the apparent value of [alpha] is much increased when dust or
+ small drops of water are present in the gas, for then the ions get
+ caught by the dust particles, the mass of a particle is so great
+ compared with that of an ion that they are practically immovable under
+ the action of the electric field, and so the ions clinging to them
+ escape detection when electrical methods are used. Taking e as 3.5 X
+ 10^-10, we see that [alpha] is about 1.2 X 10^-6, so that the number
+ of recombinations in unit time between n positive and n negative ions
+ in unit volume is 1.2 X 10^-6n^2. The kinetic theory of gases shows
+ that if we have n molecules of air per cubic centimetre, the number of
+ collisions per second is 1.2 X 10^-10n^2 at a temperature of 0 deg. C.
+ Thus we see that the number of recombinations between oppositely
+ charged ions is enormously greater than the number of collisions
+ between the same number of neutral molecules. We shall see that the
+ difference in size between the ion and the molecule is not nearly
+ sufficient to account for the difference between the collisions in the
+ two cases; the difference is due to the force between the oppositely
+ charged ions, which drags ions into collisions which but for this
+ force would have missed each other.
+
+ Several methods have been used to measure [alpha]. In one method air,
+ exposed to some ionizing agent at one end of a long tube, is slowly
+ sucked through the tube and the saturation current measured at
+ different points along the tube. These currents are proportional to
+ the values of n at the place of observation: if we know the distance
+ of this place from the end of the tube when the gas was ionized and
+ the velocity of the stream of gas, we can find t in equation (3), and
+ knowing the value of n we can deduce the value of [alpha] from the
+ equation
+
+ 1/n1 - 1/n2 = [alpha](t1 - t2),
+
+ where n1, n2 are the values of n at the times t1, t2 respectively. In
+ this method the tubes ought to be so wide that the loss of ions by
+ diffusion to the sides of the tube is negligible. There are other
+ methods which involve the knowledge of the speed with which the ions
+ move under the action of known electric forces; we shall defer the
+ consideration of these methods until we have discussed the question of
+ these speeds.
+
+ In measuring the value of [alpha] it should be remembered that the
+ theory of the methods supposes that the ionization is uniform
+ throughout the gas. If the total ionization throughout a gas remains
+ constant, but instead of being uniformly distributed is concentrated
+ in patches, it is evident that the ions will recombine more quickly in
+ the second case than in the first, and that the value of [alpha] will
+ be different in the two cases. This probably explains the large values
+ of [alpha] obtained by Retschinsky, who ionized the gas by the [alpha]
+ rays from radium, a method which produces very patchy ionization.
+
+ _Variation of [alpha] with the Pressure of the Gas._--All observers
+ agree that there is little variation in [alpha] with the pressures for
+ pressures of between 5 and 1 atmospheres; at lower pressures, however,
+ the value of [alpha] seems to diminish with the pressure: thus
+ Langevin (_Ann. chim. phys._, 1903, 28, p. 287) found that at a
+ pressure of 1/5 of an atmosphere the value of [alpha] was about 1/5 of
+ its value at atmospheric pressure.
+
+ _Variation of [alpha] with the Temperature._--Erikson (_Phil. Mag._,
+ Aug. 1909) has shown that the value of [alpha] for air increases as
+ the temperature diminishes, and that at the temperature of liquid air
+ -180 deg. C., it is more than twice as great as at +12 deg. C.
+
+ Since, as we have seen, the recombination is due to the coming
+ together of the positive and negative ions under the influence of the
+ electrical attraction between them, it follows that a large electric
+ force sufficient to overcome this attraction would keep the ions apart
+ and hence diminish the coefficient of recombination. Simple
+ considerations, however, will show that it would require exceedingly
+ strong electric fields to produce an appreciable effect. The value of
+ [alpha] indicates that for two oppositely charged ions to unite they
+ must come within a distance of about 1.5 X 10^-6 centimetres; at this
+ distance the attraction between them is e^2 X 10^12/2.25, and if X is
+ the external electric force, the force tending to pull them apart
+ cannot be greater than Xe; if this is to be comparable with the
+ attraction, X must be comparable with e X 10^12/2.25, or putting e = 4
+ X 10^-10, with 1.8 X 10^2; this is 54,000 volts per centimetre, a
+ force which could not be applied to gas at atmospheric pressure
+ without producing a spark.
+
+ _Diffusion of the Ions._--The ionized gas acts like a mixture of
+ gases, the ions corresponding to two different gases, the non-ionized
+ gas to a third. If the concentration of the ions is not uniform, they
+ will diffuse through the non-ionized gas in such a way as to produce a
+ more uniform distribution. A very valuable series of determinations of
+ the coefficient of diffusion of ions through various gases has been
+ made by Townsend (_Phil. Trans._, 1900, A, 193, p. 129). The method
+ used was to suck the ionized gas through narrow tubes; by measuring
+ the loss of both the positive and negative ions after the gases had
+ passed through a known length of tube, and allowing for the loss by
+ recombination, the loss by diffusion and hence the coefficient of
+ diffusion could be determined. The following tables give the values of
+ the coefficients of diffusion D on the C.G.S. system of units as
+ determined by Townsend:--
+
+ Table I.--_Coefficients of Diffusion (D) in Dry Gases._
+
+ +-----+------------+------------+----------+------------------+
+ |Gas. |D for +ions.|D for -ions.|Mean Value| Ratio of D for |
+ | | | | of D. | - to D for +ions.|
+ +-----+------------+------------+----------+------------------+
+ | Air | .028 | .043 | .0347 | 1.54 |
+ | O2 | .025 | .0396 | .0323 | 1.58 |
+ | CO2 | .023 | .026 | .0245 | 1.13 |
+ | H2 | .123 | .190 | .156 | 1.54 |
+ +-----+------------+------------+----------+------------------+
+
+ Table II.--Coefficients of Diffusion in Moist Gases.
+
+ +-----+------------+------------+----------+------------------+
+ |Gas. |D for +ions.|D for -ions.|Mean Value| Ratio of D for |
+ | | | | of D. | - to D for +ions.|
+ +-----+------------+------------+----------+------------------+
+ | Air | .032 | .037 | .0335 | 1.09 |
+ | O2 | .0288 | .0358 | .0323 | 1.24 |
+ | CO2 | .0245 | .0255 | .025 | 1.04 |
+ | H2 | .128 | .142 | .135 | 1.11 |
+ +-----+------------+------------+----------+------------------+
+
+ It is interesting to compare with these coefficients the values of D
+ when various gases diffuse through each other. D for hydrogen through
+ air is .634, for oxygen through air .177, for the vapour of isobutyl
+ amide through air .042. We thus see that the velocity of diffusion of
+ ions through air is much less than that of the simple gas, but that it
+ is quite comparable with that of the vapours of some complex organic
+ compounds.
+
+ The preceding tables show that the negative ions diffuse more rapidly
+ than the positive, especially in dry gases. The superior mobility of
+ the negative ions was observed first by Zeleny (_Phil. Mag._, 1898
+ [5], 46, p. 120), who showed that the velocity of the negative ions
+ under an electric force is greater than that of the positive. It will
+ be noticed that the difference between the mobility of the negative
+ and the positive ions is much more pronounced in dry gases than in
+ moist. The difference in the rates of diffusion of the positive and
+ negative ions is the reason why ionized gas, in which, to begin with,
+ the positive and negative charges were of equal amounts, sometimes
+ becomes electrified even although the gas is not acted upon by
+ electric forces. Thus, for example, if such gas be blown through
+ narrow tubes, it will be positively electrified when it comes out, for
+ since the negative ions diffuse more rapidly than the positive, the
+ gas in its passage through the tubes will lose by diffusion more
+ negative than positive ions and hence will emerge positively
+ electrified. Zeleny snowed that this effect does not occur when, as in
+ carbonic acid gas, the positive and negative ions diffuse at the same
+ rates. Townsend (loc. cit.) showed that the coefficient of diffusion
+ of the ions is the same whether the ionization is produced by Rontgen
+ rays, radioactive substances, ultra-violet light, or electric sparks.
+ The ions produced by chemical reactions and in flames are much less
+ mobile; thus, for example, Bloch (_Ann. chim. phys._, 1905 [8], 4, p.
+ 25) found that for the ions produced by drawing air over phosphorus
+ the value of [alpha]/e was between 1 and 6 instead of over 3000, the
+ value when the air was ionized by Rontgen rays.
+
+_Velocity of Ions in an Electric Field._--The velocity of ions in an
+electric field, which is of fundamental importance in conduction, is
+very closely related to the coefficient of diffusion. Measurements of
+this velocity for ions produced by Rontgen rays have been made by
+Rutherford (_Phil. Mag._ [5], 44, p. 422), Zeleny (_Phil. Mag._ [5], 46,
+p. 120), Langevin (_Ann. Chim. Phys._, 1903, 28, p. 289), Phillips
+(_Proc. Roy. Soc._ 78, A, p. 167), and Wellisch (_Phil. Trans._, 1909,
+209, p. 249). The ions produced by radioactive substance have been
+investigated by Rutherford (_Phil. Mag._ [5], 47, p. 109) and by Franck
+and Pohl (_Verh. deutsch. phys. Gesell._, 1907, 9, p. 69), and the
+negative ions produced when ultra-violet light falls on a metal plate by
+Rutherford (_Proc. Camb. Phil. Soc._ 9, p. 401). H. A. Wilson (_Phil.
+Trans._ 192, p. 4O9), Marx (_Ann. de Phys._ 11, p. 765), Moreau (_Journ.
+de Phys._ 4, 11, p. 558; _Ann. Chim. Phys._ 7, 30, p. 5) and Gold
+(_Proc. Roy. Soc._ 79, p. 43) have investigated the velocities of ions
+produced by putting various salts into flames; McClelland (_Phil. Mag._
+46, p. 29) the velocity of the ions in gases sucked from the
+neighbourhood of flames and arcs; Townsend (_Proc. Camb. Phil. Soc._ 9,
+p. 345) and Bloch (_loc. cit._) the velocity of ions produced by
+chemical reaction; and Chattock (_Phil. Mag._ [5], 48, p. 401) the
+velocity of the ions produced when electricity escapes from a sharp
+needle point into a gas.
+
+Several methods have been employed to determine these velocities. The
+one most frequently employed is to find the electromotive intensity
+required to force an ion against the stream of gas moving with a known
+velocity parallel to the lines of electric force. Thus, of two
+perforated plane electrodes vertically over each other, suppose the
+lower to be positively, the upper negatively electrified, and suppose
+that the gas is streaming vertically downwards with the velocity V; then
+unless the upward velocity of the positive ion is greater than V, no
+positive electricity will reach the upper plate. If we increase the
+strength of the field between the plates, and hence the upward velocity
+of the positive ion, until the positive ions just begin to reach the
+upper plate, we know that with this strength of field the velocity of
+the positive ion is equal to V. By this method, which has been used by
+Rutherford, Zeleny and H. A. Wilson, the velocity of ions in fields of
+various strengths has been determined.
+
+ The arrangement used by Zeleny is represented in fig. 8. P and Q are
+ square brass plates. They are bored through their centres, and to the
+ openings the tubes R and S are attached, the space between the plates
+ being covered in so as to form a closed box. K is a piece of wire
+ gauze completely covering the opening in Q; T is an insulated piece of
+ wire gauze nearly but not quite filling the opening in the plate P,
+ and connected with one pair of quadrants of an electrometer E. A plug
+ of glass wool G filters out the dust from a stream of gas which enters
+ the vessel by the tube D and leaves it by F; this plug also makes the
+ velocity of the flow of the gas uniform across the section of the
+ tube. The Rontgen rays to ionize the gas were produced by a bulb at
+ O, the bulb and coil being in a lead-covered box, with an aluminium
+ window through which the rays passed. Q is connected with one pole of
+ a battery of cells, P and the other pole of the battery are put to
+ earth. The changes in the potential of T are due to ions giving up
+ their charges to it. With a given velocity of air-blast the potential
+ of T was found not to change unless the difference of potential
+ between P and Q exceeded a critical value. The field corresponding to
+ this critical value thus made the ions move with the known velocity of
+ the blast.
+
+ [Illustration: FIG. 8.]
+
+ [Illustration: FIG. 9.]
+
+ Another method which has been employed by Rutherford and McClelland is
+ based on the action of an electric field in destroying the
+ conductivity of gas streaming through it. Suppose that BAB, DCD (fig.
+ 9) are a system of parallel plates boxed in so that a stream of gas,
+ after flowing between BB, passes between DD without any loss of gas in
+ the interval. Suppose the plates DD are insulated, and connected with
+ one pair of quadrants of an electrometer, by charging up C to a
+ sufficiently high potential we can drive all the positive ions which
+ enter the system DCD against the plates D; this will cause a deflexion
+ of the electrometer, which in one second will be proportional to the
+ number of positive ions which have entered the system in that time. If
+ we charge A up to a high potential, B being put to earth, we shall
+ find that the deflexion of the electrometer connected with DD is less
+ than it was when A and B were at the same potential, because some of
+ the positive ions in their passage through BAB are driven against the
+ plates B. If u is the velocity along the lines of force in the uniform
+ electric field between A and B, and t the time it takes for the gas to
+ pass through BAB, then all the positive ions within a distance ut of
+ the plates B will be driven up against these plates, and thus if the
+ positive ions are equally distributed through the gas, the number of
+ positive ions which emerge from the system when the electric field is
+ on will bear to the number which emerge when the field is off the
+ ratio of 1 - ut/l to unity, where l is the distance between A and B.
+ This ratio is equal to the ratio of the deflexions in one second of
+ the electrometer attached to D, hence the observations of this
+ instrument give 1 - ut/l. If we know the velocity of the gas and the
+ length of the plates A and B, we can determine t, and since l can be
+ easily measured, we can find u, the velocity of the positive ion in a
+ field of given strength. By charging A and C negatively instead of
+ positively we can arrive at the velocity of the negative ion. In
+ practice it is more convenient to use cylindrical tubes with coaxial
+ wires instead of the systems of parallel plates, though in this case
+ the calculation of the velocity of the ions from the observations is a
+ little more complicated, inasmuch as the electric field is not uniform
+ between the tubes.
+
+ [Illustration: FIG. 10.]
+
+ A method which gives very accurate results, though it is only
+ applicable in certain cases, is the one used by Rutherford to measure
+ the velocity of the negative ions produced close to a metal plate by
+ the incidence on the plate of ultra-violet light. The principle of the
+ method is as follows:--AB (fig. 10) is an insulated horizontal plate
+ of well-polished zinc, which can be moved vertically up and down by
+ means of a screw; it is connected with one pair of quadrants of an
+ electrometer, the other pair of quadrants being put to earth. CD is a
+ base-plate with a hole EF in it; this hole is covered with fine wire
+ gauze, through which ultra-violet light passes and falls on the plate
+ AB. The plate CD is connected with an alternating current dynamo,
+ which produces a simply-periodic potential difference between AB and
+ CD, the other pole being put to earth. Suppose that at any instant the
+ plate CD is at a higher potential than AB, then the negative ions from
+ AB will move towards CD, and will continue to do so as long as the
+ potential of CD is higher than that of AB. If, however, the potential
+ difference changes sign before the negative ions reach CD, these ions
+ will go back to AB. Thus AB will not lose any negative charge unless
+ the distance between the plates AB and CD is less than the distance
+ traversed by the negative ion during the time the potential of CD is
+ higher than that of AB. By altering the distance between the plates
+ until CD just begins to lose a negative charge, we find the velocity
+ of the negative ion under unit electromotive intensity. For suppose
+ the difference of potential between AB and CD is equal to a sin pt,
+ then if d is the distance between the plates, the electric intensity
+ is equal to a sin pt/d; if we suppose the velocity of the ion is
+ proportional to the electric intensity, and if u is the velocity for
+ unit electric intensity, the velocity of the negative ion will be ua
+ sin pt/d. Hence if x represent the distance of the ion from AB
+
+ dx ua
+ --- = --- sin pt
+ dT d
+
+ ua
+ x = ----(1 - cos pt), if x = 0 when t = 0.
+ pd
+
+ Thus the greatest distance the ion can get from the plate is equal to
+ 2au/pd, and if the distance between the plates is gradually reduced to
+ this value, the plate AB will begin to lose a negative charge; hence
+ when this happens
+
+ d = 2au/pd, or u = pd^2/2a,
+
+ an equation by means of which we can find u.
+
+ In this form the method is not applicable when ions of both signs are
+ present. Franck and Pohl (_Verh. deutsch. physik. Gesell._ 1907, 9, p.
+ 69) have by a slight modification removed this restriction. The
+ modification consists in confining the ionization to a layer of gas
+ below the gauze EF. If the velocity of the positive ions is to be
+ determined, these ions are forced through the gauze by applying to the
+ ionized gas a small constant electric force acting upwards; if
+ negative ions are required, the constant force is reversed. After
+ passing through the gauze the ions are acted upon by alternating
+ forces as in Rutherford's method.
+
+ Langevin (_Ann. chim. phys._, 1903, 28, p. 289) devised a method of
+ measuring the velocity of the ions which has been extensively used; it
+ has the advantage of not requiring the rate of ionization to remain
+ uniform. The general idea is as follows. Suppose that we expose the
+ gas between two parallel plates A, B to Rontgen rays or some other
+ ionizing agent, then stop the rays and apply a uniform electric field
+ to the region between the plates. If the force on the positive ion is
+ from A to B, the plate B will receive a positive charge of
+ electricity. After the electric force has acted for a time T reverse
+ it. B will now begin to receive negative electricity and will go on
+ doing so until the supply of negative ions is exhausted. Let us
+ consider how the quantity of positive electricity received by B will
+ vary with T. To fix our ideas, suppose the positive ions move more
+ slowly than the negative; let T2 and T1 be respectively the times
+ taken by the positive and negative ions to move under the electric
+ field through a distance equal to AB, the distance between the planes.
+ Then if T is greater than T2 all the ions will have been driven from
+ between the plates before the field is reversed, and therefore the
+ positive charge received by B will not depend upon T. Next let T be
+ less than T2 but greater than T1; then at the time when the field is
+ reversed all the negative ions will have been driven from between the
+ plates, so that the positive charge received by B will not be
+ neutralized by the arrival of fresh ions coming to it after the
+ reversal of the field. The number of positive ions driven against the
+ plate B will be proportional to T. Thus if we measure the value of the
+ positive charge on B for a series of values of T, each value being
+ less than the preceding, we shall find that until T reaches a certain
+ value the charge remains constant, but as soon as we reduce the time
+ below this value the charge diminishes. The value of T when the
+ diminution in the field begins is T2, the time taken for a positive
+ ion to cross from A to B under the electric field; thus from T2 we can
+ calculate the velocity of the positive ion in this field. If we still
+ further diminish T, we shall find that we reach a value when the
+ diminution of the positive charge on B with the time suddenly becomes
+ much more rapid; this change occurs when T falls below T1 the time
+ taken for the negative ions to go from one plate to the other, for now
+ when the field is reversed there are still some negative ions left
+ between the plates, and these will be driven against B and rob it of
+ some of the positive charge it had acquired before the field was
+ reversed. By observing the time when the increase in the rate of
+ diminution of the positive charge with the time suddenly sets in we
+ can determine T1, and hence the velocity of the negative ions.
+
+ The velocity of the ions produced by the discharge of electricity from
+ a fine point was determined by Chattock by an entirely different
+ method. In this case the electric field is so strong and the velocity
+ of the ion so great that the preceding methods are not applicable.
+ Suppose P represents a vertical needle discharging electricity into
+ air, consider the force acting on the ions included between two
+ horizontal planes A, B. If P is the density of the electrification,
+ and Z the vertical component of the electric intensity, F the
+ resultant force on the ions between A and B is vertical and equal to
+ _ _ _
+ / / /
+ | | | Z[rho]dxdydz.
+ _/_/_/
+
+ Let us suppose that the velocity of the ion is proportional to the
+ electric intensity, so that if w is the vertical velocity of the ions,
+ which are supposed all to be of one sign, w = RZ.
+
+ Substituting this value of Z, the vertical force on the ions between A
+ and B is equal to
+ _ _ _
+ 1 / / /
+ - | | | w[rho]dxdydz.
+ R _/_/_/
+
+ But [integral][integral]w[rho]dxdy = [iota], where [iota] is the
+ current streaming from the point. This current, which can be easily
+ measured by putting a galvanometer in series with the discharging
+ point, is independent of z, the vertical distance of a plane between A
+ and B below the charging point. Hence we have
+ _
+ [iota] / [iota]
+ F = ------ | dz = ------.z.
+ R _/ R
+
+ This force must be counterbalanced by the difference of gaseous
+ pressures over the planes A and B; hence if pB and pA denote
+ respectively the pressures over B and A, we have
+
+ [iota]
+ pB - pA = ------ z.
+ R
+
+ Hence by the measurement of these pressures we can determine R, and
+ hence the velocity with which an ion moves under a given electric
+ intensity.
+
+ There are other methods of determining the velocities of the ions, but
+ as these depend on the theory of the conduction of electricity through
+ a gas containing charged ions, we shall consider them in our
+ discussion of that theory.
+
+ By the use of these methods it has been shown that the velocities of
+ the ions in a given gas are the same whether the ionization is
+ produced by Rontgen rays, radioactive substances, ultra-violet light,
+ or by the discharge of electricity from points. When the ionization is
+ produced by chemical action the ions are very much less mobile, moving
+ in the same electric field with a velocity less than one-thousandth
+ part of the velocity of the first kind of ions. On the other hand, as
+ we shall see later, the velocity of the negative ions in flames is
+ enormously greater than that of even the first kind of ion under
+ similar electric fields and at the same pressure. But when these
+ negative ions get into the cold part of the flame, they move
+ sluggishly with velocities of the order of those possessed by the
+ second kind. The results of the various determinations of the
+ velocities of the ions are given in the following table. The
+ velocities are in centimetres per second under an electric force of
+ one volt per centimetre, the pressure of the gas being 1 atmosphere.
+ V+ denotes the velocity of the positive ion, V- that of the negative.
+ V is the mean velocity of the positive and negative ions.
+
+ _Velocities of Ions.--Ions produced by Rontgen Rays._
+
+ +----------------------+------+------+------+-----------------+
+ | Gas. | V+. | V-. | V. | Observer. |
+ +----------------------+------+------+------+-----------------+
+ | Air | .. | .. | 1.6 | Rutherford |
+ | Air (dry) | 1.36 | 1.87 | .. | Zeleny |
+ | " | 1.60 | 1.70 | .. | Langevin |
+ | " | 1.39 | 1.78 | .. | Phillips |
+ | " | 1.54 | 1.78 | .. | Wellisch |
+ | Air (moist) | 1.37 | 1.81 | .. | Zeleny |
+ | Oxygen (dry) | 1.36 | 1.80 | .. | " |
+ | Oxygen (moist) | 1.29 | 1.52 | .. | " |
+ | Carbonic acid (dry) | 0.76 | 0.81 | .. | " |
+ | " | 0.86 | 0.90 | .. | Langevin |
+ | " | 0.81 | 0.85 | .. | Wellisch |
+ | Carbonic acid (moist)| 0.82 | 0.75 | .. | Zeleny |
+ | Hydrogen (dry) | 6.70 | 7.95 | .. | " |
+ | Nitrogen | .. | .. | 1.6 | Rutherford |
+ | Sulphur dioxide | 0.44 | 0.41 | .. | Wellisch |
+ | Hydrochloric acid | .. | .. | 1.27 | Rutherford |
+ | Chlorine | .. | .. | 1.0 | " |
+ | Helium (dry) | 5.09 | 6.31 | .. | Franck and Pohl |
+ | Carbon monoxide | 1.10 | 1.14 | .. | Wellisch |
+ | Nitrous oxide | 0.82 | 0.90 | .. | " |
+ | Ammonia | 0.74 | 0.80 | .. | " |
+ | Aldehyde | 0.31 | 0.30 | .. | " |
+ | Ethyl alcohol | 0.34 | 0.27 | .. | " |
+ | Acetone | 0.31 | 0.29 | .. | " |
+ | Ethyl chloride | 0.33 | 0.31 | .. | " |
+ | Pentane | 0.36 | 0.35 | .. | " |
+ | Methyl acetate | 0.33 | 0.36 | .. | " |
+ | Ethyl formate | 0.30 | 0.31 | .. | " |
+ | Ethyl ether | 0.29 | 0.31 | .. | " |
+ | Ethyl acetate | 0.31 | 0.28 | .. | " |
+ | Methyl bromide | 0.29 | 0.28 | .. | " |
+ | Methyl iodide | 0.21 | 0.22 | .. | " |
+ | Carbon tetrachloride | 0.30 | 0.31 | .. | " |
+ | Ethyl iodide | 0.17 | 0.16 | .. | " |
+ +----------------------+------+------+------+-----------------+
+
+ _Ions produced by Ultra-Violet Light._
+
+ Air 1.4 Rutherford
+ Hydrogen 3.9 Rutherford
+ Carbonic acid 0.78 Rutherford
+
+ _Ions in Gases sucked from Flames._
+
+ Velocities varying from .04 to .23 McClelland
+
+ _Ions in Flames containing Salts._
+
+ Negative ions 12.9 cm./sec. Gold
+ +ions for salts of Li, Na,
+ K, Rb, Cs 62 H. A. Wilson
+ " 200 Marx
+ " 80 Moreau
+
+ _Ions liberated by Chemical Action._
+
+ Velocities of the order of 0.0005 cm./sec. Bloch
+
+ _Ions from Point Discharge._
+
+ +---------------+------+-------+------+----------+
+ | Hydrogen | 5.4 | 7.43 | 6.41 | Chattock |
+ | Carbonic acid | 0.83 | 0.925 | 0.88 | Chattock |
+ | Air | 1.32 | 1.80 | 1.55 | Chattock |
+ | Oxygen | 1.30 | 1.85 | 1.57 | Chattock |
+ +---------------+------+-------+------+----------+
+
+ It will be seen from this table that the greater mobility of the
+ negative ions is very much more marked in the case of the lighter and
+ simpler gases than in that of the heavier and more complicated ones;
+ with the vapours of organic substances there seems but little
+ difference between the mobilities of the positive and negative ions,
+ indeed in one or two cases the positive one seems slightly but very
+ slightly the more mobile of the two. In the case of the simple gases
+ the difference is much greater when the gases are dry than when they
+ are moist. It has been shown by direct experiment that the velocities
+ are directly proportional to the electric force.
+
+ _Variation of Velocities with Pressure._--Until the pressure gets low
+ the velocities of the ions, negative as well as positive, vary
+ inversely as the pressure. Langevin (loc. cit.) was the first to show
+ that at very low pressures the velocity of the negative ions increases
+ more rapidly as the pressure is diminished than this law indicates. If
+ the nature of the ion did not change with the pressure, the kinetic
+ theory of gases indicates that the velocity would vary inversely as
+ the pressure, so that Langevin's results indicate a change in the
+ nature of the negative ion when the pressure is diminished below a
+ certain value. Langevin's results are given in the following table,
+ where p represents the pressure measured in centimetres of mercury, V+
+ and V- the velocities of the positive and negative ions in air under
+ unit electrostatic force, i.e. 300 volts per centimetre:--
+
+ +----------------------+----------------------+
+ | Negative Ions. | Positive Ions. |
+ +-------+------+-------+-------+------+-------+
+ | p. | V-. |pV-/76.| p. | V+. |pV+/76.|
+ +-------+------+-------+-------+------+-------+
+ | 7.5 | 6560 | 647 | 7.5 | 4430 | 437 |
+ | 20.0 | 2204 | 580 | 20.0 | 1634 | 430 |
+ | 41.5 | 994 | 530 | 41.5 | 782 | 427 |
+ | 76.0 | 510 | 510 | 76.0 | 480 | 420 |
+ | 142.0 | 270 | 505 | 142.0 | 225 | 425 |
+ +-------+------+-------+-------+------+-------+
+
+ The increase in the case of pV- indicates that the structure of the
+ negative ion gets simpler as the pressure is reduced. Wallisch in some
+ experiments made at the Cavendish Laboratory found that the diminution
+ in the value of pV- at low pressures is much more marked in some gases
+ than in others, and in some gases he failed to detect it; but it must
+ be remembered that it is difficult to get measurements at pressures of
+ only a few millimetres, as the amount of ionization is so exceedingly
+ small at such pressures that the quantities to be observed are hardly
+ large enough to admit of accurate measurements by the methods
+ available at higher pressures.
+
+ _Effect of Temperature on the Velocity of the Ions._--Phillips (_Proc.
+ Roy. Soc._, 1906, 78, p. 167) investigated, using Langevin's method,
+ the velocities of the + and - ions through air at atmospheric pressure
+ at temperatures ranging from that of boiling liquid air to 411 deg.
+ C.; R1 and R2 are the velocities of the + and - ions respectively when
+ the
+
+ force is a volt per centimetre.
+
+ +-------+-------+---------------------+
+ | R1. | R2. |Temperature Absolute.|
+ +-------+-------+---------------------+
+ | 2.00 | 2.495 | 411 deg. |
+ | 1.95 | 2.40 | 399 deg. |
+ | 1.85 | 2.30 | 383 deg. |
+ | 1.81 | 2.21 | 373 deg. |
+ | 1.67 | 2.125 | 348 deg. |
+ | 1.60 | 2.00 | 333 deg. |
+ | 1.39 | 1.785 | 285 deg. |
+ | 0.945 | 1.23 | 209 deg. |
+ | 0.235 | 0.235 | 94 deg. |
+ +-------+-------+---------------------+
+
+ We see that except in the case of the lowest temperature, that of
+ liquid air, where there is a great drop in the velocity, the
+ velocities of the ions are proportional to the absolute temperature.
+ On the hypothesis of an ion of constant size we should, from the
+ kinetic theory of gases, expect the velocity to be proportional to the
+ square root of the absolute temperature, if the charge on the ion did
+ not affect the number of collisions between the ion and the molecules
+ of the gas through which it is moving. If the collisions were brought
+ about by the electrical attraction between the ions and the molecules,
+ the velocity would be proportional to the absolute temperature. H. A.
+ Wilson (_Phil. Trans._ 192, p. 499), in his experiments on the
+ conduction of flames and hot gases into which salts had been put,
+ found that the velocity of the positive ions in flames at a
+ temperature of 2000 deg. C. containing the salts of the alkali metals
+ was 62 cm./sec. under an electric force of one volt per centimetre,
+ while the velocity of the positive ions in a stream of hot air at 1000
+ deg. C. containing the same salts was only 7 cm./sec. under the same
+ force. The great effect of temperature is also shown in some
+ experiments of McClelland (_Phil. Mag._ [5], 46, p. 29) on the
+ velocities of the ions in gases drawn from Bunsen flames and arcs; he
+ found that these depended upon the distance the gas had travelled from
+ the flame. Thus, the velocity of the ions at a distance of 5.5 cm.
+ from the Bunsen flame when the temperature was 230 deg. C. was .23
+ cm./sec. for a volt per centimetre; at a distance of 10 cm. from the
+ flame when the temperature was 160 deg. C. the velocity was .21
+ cm./sec; while at a distance of 14.5 cm. from the flame when the
+ temperature was 105 deg. C. the velocity was only .04 cm./sec. If the
+ temperature of the gas at this distance from the flame was raised by
+ external means, the velocity of the ions increased.
+
+ We can derive some information as to the constitution of the ions by
+ calculating the velocity with which a molecule of the gas would move
+ in the electric field if it carried the same charge as the ion. From
+ the theory of the diffusion of gases, as developed by Maxwell, we know
+ that if the particles of a gas A are surrounded by a gas B, then, if
+ the partial pressure of A is small, the velocity u with which its
+ particles will move when acted upon by a force Xe is given by the
+ equation
+
+ Xe
+ u = ------- D,
+ (p1/N1)
+
+ where D represents the coefficient of inter-diffusion of A into B, and
+ N1 the number of particles of A per cubic centimetre when the pressure
+ due to A is p1. Let us calculate by this equation the velocity with
+ which a molecule of hydrogen would move through hydrogen if it carried
+ the charge carried by an ion, which we shall prove shortly to be equal
+ to the charge carried by an atom of hydrogen in the electrolysis of
+ solutions. Since p1/N1 is independent of the pressure, it is equal to
+ [Pi]/N, where [Pi] is the atmospheric pressure and N the number of
+ molecules in a cubic centimetre of gas at atmospheric pressure. Now Ne
+ = 1.22 X 10^10, if e is measured in electrostatic units; [Pi] = 10^6
+ and D in this case is the coefficient of diffusion of hydrogen into
+ itself, and is equal to 1.7. Substituting these values we find
+
+ u = 1.97 X 10^4X.
+
+ If the potential gradient is 1 volt per centimetre, X = 1/300.
+ Substituting this value for X, we find u = 66 cm./sec, for the
+ velocity of a hydrogen molecule. We have seen that the velocity of the
+ ion in hydrogen is only about 5 cm./sec, so that the ion moves more
+ slowly than it would if it were a single molecule. One way of
+ explaining this is to suppose that the ion is bigger than the
+ molecule, and is in fact an aggregation of molecules, the charged ion
+ acting as a nucleus around which molecules collect like dust round a
+ charged body. This view is supported by the effect produced by
+ moisture in diminishing the velocity of the negative ion, for, as C.
+ T. R. Wilson (_Phil. Trans._ 193, p. 289) has shown, moisture tends to
+ collect round the ions, and condenses more easily on the negative than
+ on the positive ion. In connexion with the velocities of ions in the
+ gases drawn from flames, we find other instances which suggest that
+ condensation takes place round the ions. An increase in the size of
+ the system is not, however, the only way by which the velocity might
+ fall below that calculated for the hydrogen molecule, for we must
+ remember that the hydrogen molecule, whose coefficient of diffusion is
+ 1.7, is not charged, while the ion is. The forces exerted by the ion
+ on the other molecules of hydrogen are not the same as those which
+ would be exerted by a molecule of hydrogen, and as the coefficient of
+ diffusion depends on the forces between the molecules, the coefficient
+ of diffusion of a charged molecule into hydrogen might be very
+ different from that of an uncharged one.
+
+ Wellisch (_loc. cit._) has shown that the effect of the charge on the
+ ion is sufficient in many cases to explain the small velocity of the
+ ions, even if there were no aggregation.
+
+ _Mixture of Gases._--The ionization of a mixture of gases raises some
+ very interesting questions. If we ionize a mixture of two very
+ different gases, say hydrogen and carbonic acid, and investigate the
+ nature of the ions by measuring their velocities, the question arises,
+ shall we find two kinds of positive and two kinds of negative ions
+ moving with different velocities, as we should do if some of the
+ positive ions were positively charged hydrogen molecules, while others
+ were positively charged molecules of carbonic acid; or shall we find
+ only one velocity for the positive ions and one for the negative? Many
+ experiments have been made on the velocity of ions in mixtures of two
+ gases, but as yet no evidence has been found of the existence of two
+ different kinds of either positive or negative ions in such mixtures,
+ although some of the methods for determining the velocities of the
+ ions, especially Langevin's, ought to give evidence of this effect, if
+ it existed. The experiments seem to show that the positive (and the
+ same is true for the negative) ions in a mixture of gases are all of
+ the same kind. This conclusion is one of considerable importance, as
+ it would not be true if the ions consisted of single molecules of the
+ gas from which they are produced.
+
+ _Recombination._--Several methods enable us to deduce the coefficient
+ of recombination of the ions when we know their velocities. Perhaps
+ the simplest of these consists in determining the relation between the
+ current passing between two parallel plates immersed in ionized gas
+ and the potential difference between the plates. For let q be the
+ amount of ionization, i.e. the number of ions produced per second per
+ unit volume of the gas, A the area of one of the plates, and d the
+ distance between them; then if the ionization is constant through the
+ volume, the number of ions of one sign produced per second in the gas
+ is qAd. Now if i is the current per unit area of the plate, e the
+ charge on an ion, iA/e ions of each sign are driven out of the gas by
+ the current per second. In addition to this source of loss of ions
+ there is the loss due to the recombination; if n is the number of
+ positive or negative ions per unit volume, then the number which
+ recombine per second is [alpha]n^2 per cubic centimetre, and if n is
+ constant through the volume of the gas, as will approximately be the
+ case if the current through the gas is only a small fraction of the
+ saturation current, the number of ions which disappear per second
+ through recombination is [alpha]n^2.Ad. Hence, since when the gas is
+ in a steady state the number of ions produced must be equal to the
+ number which disappear, we have
+
+ qAd = iA/e + [alpha]n^2.Ad,
+ q = i/ed + [alpha]^n2.
+
+ If u1 and u2 are the velocities with which the positive and negative
+ ions move, nu1e and nu2e are respectively the quantities of positive
+ electricity passing in one direction through unit area of the gas per
+ second, and of negative in the opposite direction, hence
+
+ i = nu1e + nu2e.
+
+ If X is the electric force acting on the gas, k1 and k2 the velocities
+ of the positive and negative ions under unit force, u1 = k1X, u2 =
+ k2X; hence
+
+ n = i/(k1 + k2)Xe,
+
+ and we have
+
+ i [alpha]i^2
+ q = -- + -----------------.
+ ed (k1 + k2)^2e^2X^2
+
+ But qed is the saturation current per unit area of the plate; calling
+ this I, we have
+
+ d[alpha]i^2
+ I - i = ---------------
+ e(k1 + k2)^2X^2
+
+ or
+
+ i^2.d[alpha]
+ X^2 = -------------------.
+ e(I - i)(k1 + k2)^2
+
+ Hence if we determine corresponding values of X and i we can deduce
+ the value of [alpha]/e if we also know (k1 + k2). The value of I is
+ easily determined, as it is the current when X is very large. The
+ preceding result only applies when i is small compared with I, as it
+ is only in this case that the values of n and X are uniform throughout
+ the volume of the gas. Another method which answers the same purpose
+ is due to Langevin (_Ann. Chim. Phys._, 1903, 28, p. 289); it is as
+ follows. Let A and B be two parallel planes immersed in a gas, and let
+ a slab of the gas bounded by the planes a, b parallel to A and B be
+ ionized by an instantaneous flash of Rontgen rays. If A and B are at
+ different electric potentials, then all the positive ions produced by
+ the rays will be attracted by the negative plate and all the negative
+ ions by the positive, if the electric field were exceedingly large
+ they would reach these plates before they had time to recombine, so
+ that each plate would receive N0 ions if the flash of Rontgen rays
+ produced N0 positive and N0 negative ions. With weaker fields the
+ number of ions received by the plates will be less as some of them
+ will recombine before they can reach the plates. We can find the
+ number of ions which reach the plates in this case in the following
+ way:--In consequence of the movement of the ions the slab of ionized
+ gas will broaden out and will consist of three portions, one in which
+ there are nothing but positive ions,--this is on the side of the
+ negative plate,--another on the side of the positive plate in which
+ there are nothing but negative ions, and a portion between these in
+ which there are both positive and negative ions; it is in this layer
+ that recombination takes place, and here if n is the number of
+ positive or negative ions at the time t after the flash of Rontgen
+ rays,
+
+ n = n0/(1 + [alpha]n0t).
+
+ With the same notation as before, the breadth of either of the outer
+ layers will in time dt increase by X(k1 + k2)dt, and the number of ions
+ in it by X(k1 + k2)ndt; these ions will reach the plate, the outer
+ layers will receive fresh ions until the middle one disappears, which
+ it will do after a time l/X(k1 + k2), where l is the thickness of the
+ slab ab of ionized gas; hence N, the number of ions reaching either
+ plate, is given by the equation
+ _
+ / l/X(k1+k2) n0X(k1 + k2) X(k1 + k2) / n0[alpha]l \
+ N = | --------------dt = ---------- log( 1 + ---------- ).
+ _/ 0 1 + n0[alpha]t [alpha] \ X(k1 + k2) /
+
+
+ If Q is the charge received by the plate,
+
+ X / Q0[epsilon]\
+ Q = Ne = -------------- log ( 1 + ----------- ),
+ 4[pi][epsilon] \ 4[pi]X /
+
+ where Q0 = n0le is the charge received by the plate when the electric
+ force is large enough to prevent recombination, and [epsilon] =
+ [alpha]4[pi]e(R1 + R2). We can from this result deduce the value of
+ [epsilon] and hence the value of [alpha] when R1+R2 is known.
+
+ _Distribution of Electric Force when a Current is passing through an
+ Ionized Gas._--Let the two plates be at right angles to the axis of x;
+ then we may suppose that between the plates the electric intensity X
+ is everywhere parallel to the axis of x. The velocities of both the
+ positive and negative ions are assumed to be proportional to X. Let
+ k1X, k2X represent these velocities respectively; let n1, n2 be
+ respectively the number of positive and negative ions per unit volume
+ at a point fixed by the co-ordinate x; let q be the number of positive
+ or negative ions produced in unit time per unit volume at this point;
+ and let the number of ions which recombine in unit volume in unit time
+ be [alpha]n1n2; then if e is the charge on the ion, the volume density
+ of the electrification is (n1 - n2)e, hence
+
+ dX
+ -- = 4[pi](n1 - n2)e (1).
+ dx
+
+ If I is the current through unit area of the gas and if we neglect any
+ diffusion except that caused by the electric field,
+
+ n1ek1X + n2ek2X = I (2).
+
+ From equations (1) and (2) we have
+
+ 1 / I k2 dX \
+ n1e = ------- ( - + ----- -- ) (3),
+ k1 + k2 \ X 4[pi] dx /
+
+ 1 / I k1 dX \
+ n2e = ------- ( - - ----- -- ) (4),
+ k1 + k2 \ X 4[pi] dx /
+
+ and from these equations we can, if we know the distribution of
+ electric intensity between the plates, calculate the number of
+ positive and negative ions.
+
+ In a steady state the number of positive and negative ions in unit
+ volume at a given place remains constant, hence neglecting the loss by
+ diffusion, we have
+
+ d
+ --(k1n1X) = q - [alpha]n1n2 (5).
+ dx
+
+ d
+ - --(k2n2X) = q - [alpha]n1n2 (6).
+ dx
+
+ If k1 and K2 are constant, we have from (1), (5) and (6)
+
+ d^2X^2 / 1 1 \
+ ------ = 8[pi]e(q - [alpha]n1n2)( --- + --- ) (7),
+ dx^2 \ k1 k2 /
+
+ an equation which is very useful, because it enables us, if we know
+ the distribution of X^2, to find whether at any point in the gas the
+ ionization is greater or less than the recombination of the ions. We
+ see that q - [alpha]n1n2, which is the excess of ionization over
+ recombination, is proportional to d^2X^2/dx^2. Thus when the ionization
+ exceeds the recombination, i.e. when q - [alpha]n1n2 is positive, the
+ curve for X^2 is convex to the axis of x, while when the recombination
+ exceeds the ionization the curve for X^2 will be concave to the axis of
+ x. Thus, for example, fig. 11 represents the curve for X^2 observed by
+ Graham (_Wied. Ann._ 64, p. 49) in a tube through which a steady
+ current is passing. Interpreting it by equation (7), we infer that
+ ionization was much in excess of recombination at A and B, slightly so
+ along C, while along D the recombination exceeded the ionization.
+ Substituting in equation (7) the values of n1, n2 given in (3), (4),
+ we get
+ _ _
+ d^2X^2 | [alpha] / k^2 dX^2\ / k2 dX^2\ | / 1 1 \
+ ------ = 8[pi]e |q - ----------------- (1 + ----- ---- ) (1 - ----- ---- )| ( --- + --- ) (8).
+ dx^2 |_ e^2X^2(k1 + k2)^2 \ 8[pi] dx / \ 8[pi] dx /_| \k1 k2 /
+
+ [Illustration: Fig. 11.]
+
+ This equation can be solved (see Thomson, _Phil. Mag._ xlvii. P. 253),
+ when q is constant and k1 = k2. From the solution it appears that if
+ X1 be the value of x close to one of the plates, and X0 the value
+ midway between them,
+
+ 1
+ X1/X0 = -------------------
+ [beta]^2 - 2/[beta]
+
+ where [beta] = 8[pi]ek1/[alpha].
+
+
+ Since e = 4 X 10^-10, [alpha] = 2X10^-6, and k1 for air at atmospheric
+ pressure = 450, [beta] is about 2.3 for air at atmospheric pressure
+ and it becomes much greater at lower pressures.
+
+ Thus X1/X0 is always greater than unity, and the value of the ratio
+ increases from unity to infinity as [beta] increases from zero to
+ infinity. As [beta] does not involve either q or I, the ratio of X1 to
+ X0 is independent of the strength of the current and of the intensity
+ of the ionization.
+
+ No general solution of equation (8) has been found when k1 is not
+ equal to k2, but we can get an approximation to the solution when q is
+ constant. The equations (1), (2), (3), (4) are satisfied by the
+ values--
+
+ n1 = n2 = (q / [alpha])^1/2
+
+ k1
+ k1n1Xe = ------- I,
+ k1 + k2
+
+ k2
+ k2n2Xe = ------- I,
+ k1 + k2
+
+
+ /[alpha]\^1/2 I
+ X = ( ------- ) ----------.
+ \ q / e(k1 + k2)
+
+ These solutions cannot, however, hold right up to the surface of the
+ plates, for across each unit of area, at a point P, k1I/(k1+k2)e
+ positive ions pass in unit time, and these must all come from the
+ region between P and the positive plate. If [lambda] is the distance
+ of P from this plate, this region cannot furnish more than q[lambda]
+ positive ions, and only this number if there are no recombinations.
+ Hence the solution cannot hold when q[lambda] is less than k1I/(k1 +
+ k2)e, or where [lambda] is less than k1I/(k1 + k2)qe.
+
+ Similarly the solution cannot hold nearer to the negative plate than
+ the distance k2I/(k1 + k2)qe.
+
+ [Illustration: FIG. 12.]
+
+ The force in these layers will be greater than that in the middle of
+ the gas, and so the loss of ions by recombination will be smaller in
+ comparison with the loss due to the removal of the ions by the
+ current. If we assume that in these layers the loss of ions by
+ recombination can be neglected, we can by the method of the next
+ article find an expression for the value of the electric force at any
+ point in the layer. This, in conjunction with the value
+
+ /[alpha]\^1/2 I
+ X0 = ( ------- ) ----------
+ \ q / e(k1 + k2)
+
+ for the gas outside the layer, will give the value of X at any point
+ between the plates. It follows from this investigation that if X1 and
+ X2 are the values of X at the positive and negative plates
+ respectively, and X0 the value of X outside the layer,
+
+ / k1 I \^1/2 / k2 I \^1/2
+ X1 = X0 ( I + -- --------- ) , X2 = X0 ( I + -- --------- ) ,
+ \ k2 [epsilon]/ \ k1 [epsilon]/
+
+ where [epsilon] = [alpha]/4[pi]e(k1 + k2). Langevin found that for air
+ at a pressure of 152 mm. [epsilon] = 0.01, at 375 mm. [epsilon] =
+ 0.06, and at 760 mm. [epsilon] = 0.27. Thus at fairly low pressures
+ 1/[epsilon] is large, and we have approximately
+
+ /k1\^1/2 I /k2\^1/2 I
+ X1 = X0 ( -- ) ---------------, X2 = X0 ( -- ) ---------------.
+ \k2/ [root][epsilon] \k1/ [root][epsilon]
+
+ Therefore X1/X2 = k1/k2,
+
+ or the force at the positive plate is to that at the negative plate as
+ the velocity of the positive ion is to that of the negative ion. Thus
+ the force at the negative plate is greater than that at the positive.
+ The falls of potential V1, V2 at the two layers when 1/[epsilon] is
+ large can be shown to be given by the equations
+
+ /[epsilon]\^3/2 /k1\^1/2
+ V1 = 8[pi]^2( --------- ) k1 ( -- ) i^2,
+ \q [alpha]/ \k2/
+
+ /[epsilon]\^3/2 /k2\^1/2
+ V2 = 8[pi]^2( --------- ) k2 ( -- ) i^2,
+ \q [alpha]/ \k1/
+
+ hence V1/V2 = k1^2/k2^2,
+
+ so that the potential falls at the electrodes are proportional to the
+ squares of the velocities of the ions. The change in potential across
+ the layers is proportional to the square of the current, while the
+ potential change between the layers is proportional to the current,
+ the total potential difference between the plates is the sum of these
+ changes, hence the relation between V and i will be of the form
+
+ V = Ai + Bi^2.
+
+ Mie (_Ann. der. Phys._, 1904, 13, P. 857) has by the method of
+ successive approximations obtained solutions of equation (8) (i.) when
+ the current is only a small fraction of the saturation current, (ii.)
+ when the current is nearly saturated. The results of his
+ investigations are represented in fig. 12, which represents the
+ distribution of electric force along the path of the current for
+ various values of the current expressed as fractions of the saturation
+ current. It will be seen that until the current amounts to about
+ one-fifth of the maximum current, the type of solution is the one just
+ indicated, i.e. the electric force is constant except in the
+ neighbourhood of the electrodes when it increases rapidly.
+
+ Though we are unable to obtain a general solution of the equation (8),
+ there are some very important special cases in which that equation can
+ be solved without difficulty. We shall consider two of these, the
+ first being that when the current is saturated. In this case there is
+ no loss of ions by recombination, so that using the same notation as
+ before we have
+
+ d
+ --(n1k1X) = q,
+ dx
+
+ d
+ --(n2k2X) = -q.
+ dx
+
+ The solutions of which if q is constant are
+
+ n1k1X = qx,
+
+ n2k2X = I/e - qx = q(l - x),
+
+ if l is the distance between the plates, and x = 0 at the positive
+ electrode. Since
+
+ dX/dx = 4[pi](n1 - n2)e,
+
+ we get
+
+ 1 dX^2 / 1 1 \ l
+ ----- ---- = qx ( -- + -- ) - q --,
+ 8[pi] d^2x \k1 k2 / k2
+
+ or
+
+ X^2 x^2 / 1 1 \ lx
+ ----- = q --- ( -- + -- ) - q -- + C,
+ 8[pi] 2 \k1 k2 / k2
+
+ where C is a quantity to be determined by the condition that
+
+ _
+ / l
+ | Xdx = V,
+ _/0
+
+
+ where V is the given potential difference between the plates. When the
+ force is a minimum dX/dx = 0, hence at this point
+
+ lk1 lk2
+ x = -------, l - x = -------.
+ k1 + k2 k1 + k2
+
+ Hence the ratio of the distances of this point from the positive and
+ negative plates respectively is equal to the ratio of the velocities
+ of the positive and negative ions.
+
+ The other case we shall consider is the very important one in which
+ the velocity of the negative ion is exceedingly large compared with
+ the positive; this is the case in flames where, as Gold (_Proc. Roy.
+ Soc._ 97, p. 43) has shown, the velocity of the negative ion is many
+ thousand times the velocity of the positive; it is also very probably
+ the case in all gases when the pressure is low. We may get the
+ solution of this case either by putting k1/k2 = 0 in equation (8), or
+ independently as follows:--Using the same notation as before, we have
+
+ i = n1k1Xe + n2k2Xe,
+
+ d
+ --(n2k2X) = q - [alpha]n1n2,
+ dx
+
+ dX
+ -- = 4[pi](n1 - n2)e.
+ dx
+
+ In this case practically all the current is carried by the negative
+ ions so that i = n2k2Xe, and therefore q = [alpha]n1n2.
+
+ Thus
+
+ n2 = i/k2Xe, n1 = qk2Xe/[alpha]i.
+
+ Thus
+
+ dX 4[pi]e^2k2qX 4[pi]i
+ -- = ------------ - ------,
+ dx [alpha]i k2X
+
+ or
+
+ dX^2 8[pi]e^2k2qX^2 8[pi]i
+ ---- - -------------- = - ------.
+ dx [alpha]i k2
+
+ The solution of this equation is
+
+ [alpha] i^2
+ X^2 = ------- ------- + C[epsilon]^(8[pi]e^2k2qx/[alpha]i)
+ q k2^2e^2
+
+ Here x is measured from the positive electrode; it is more convenient
+ in this case, however, to measure it from the negative electrode. If x
+ be the distance from the negative electrode at which the electric
+ force is X, we have from equation (7)
+
+ [alpha] i^2
+ X^2 = ------- ------- + C^1[epsilon]^(8[pi]e^2k2qx/[alpha]i)
+ q k2^2e^2
+
+ To find the value of C^1 we see by equation (7) that
+
+ d^2X^2 k1k2 1
+ ------ ------- ------ = q - [alpha]n1n2;
+ dX^2 k1 + k2 8[pi]e
+
+ hence
+ _ _ _
+ | dX^2 k1k2 1 |^x1 / x1
+ | ---- ------- ------ | = | (q - [alpha]n1n2)dx.
+ |_ dX k1 + k2 8[pi]e_| _/0
+
+ The right hand side of this equation is the excess of ionization over
+ recombination in the region extending from the cathode to x1; it must
+ therefore, when things are in a steady state, equal the excess of the
+ number of negative ions which leave this region over those which enter
+ it. The number which leave is i/e and the number which enter is i0/e,
+ if it is the current of negative ions coming from unit area of the
+ cathode, as hot metal cathodes emit large quantities of negative
+ electricity i0 may in some cases be considerable, thus the right hand
+ side of equation is (i - i0)/e. When x1 is large dX^2/dx = 0; hence we
+ have from equation
+
+ [alpha]i(i - i0) k1 + k2
+ C^1 = ---------------- -------,
+ qk1k2e^2 k2
+
+ and since k1 is small compared with k2, we have
+
+ [alpha]i^2 / k2 i - i0 \
+ X^2 = ---------- (1 + -- ------ [epsilon]^{-8[pi]e^2k2.qx/[alpha].i}).
+ qk2^2e^2 \ k1 i /
+
+ From the values which have been found for k2 and [alpha], we know that
+ 8[pi]ek2/[alpha] is a large quantity, hence the second term inside the
+ bracket will be very small when eqx is equal to or greater than i;
+ thus this term will be very small outside a layer of gas next the
+ cathode of such thickness that the number of ions produced on it would
+ be sufficient, if they were all utilized for the purpose, to carry the
+ current; in the case of flames this layer is exceedingly thin unless
+ the current is very large. The value of the electric force in the
+ uniform part of the field is equal to i/k2e.[root]([alpha]/q), while
+ when i0 = 0, the force at the cathode itself bears to the uniform
+ force the ratio of (k1 + k2)^1/2 to k1^1/2. As k1 is many thousand
+ times k2 the force increases with great rapidity as we approach the
+ cathode; this is a very characteristic feature of the passage of
+ electricity through flames and hot gases. Thus in an experiment made
+ by H. A. Wilson with a flame 18 cm. long, the drop of potential within
+ 1 centimetre of the cathode was about five times the drop in the other
+ 17 cm. of the tube. The relation between the current and the potential
+ difference when the velocity of the negative ion is much greater than
+ the positive is very easily obtained. Since the force is uniform and
+ equal to i/k2e.[root]([alpha]/q), until we get close to the cathode
+ the fall of potential in this part of the discharge will be very
+ approximately equal to i/k2e.[root]([alpha]l/q), where l is the
+ distance between the electrodes. Close to the cathode, the electric
+ force when i0 is not nearly equal to i is approximately given by the
+ equation
+
+ i /[alpha]\^1/2
+ X = --------- (---------) [epsilon]^{-4[pi]e^2k2qx/[alpha]i},
+ e(k1k2)^1/2 \ q / ,
+
+ and the fall of potential at the cathode is equal approximately to
+
+ _[oo]
+ /
+ | X dx,
+ _/0
+
+ that is to
+
+ i /[alpha]\^1/2 [alpha]i
+ --------- (---------) -----------.
+ e(k1k2)^1/2 \ q / 4[pi]e^2k2q
+
+ The potential difference between the plates is the sum of the fall of
+ potential in the uniform part of the discharge plus the fall at the
+ cathode, hence
+
+ /[alpha]\^1/2 i / i[alpha]^2 1 \
+ V = (---------) --- ( il + ---------- ------------ ).
+ \ q / ek2 \ 4[pi]e^2q [root](k1k2)/
+
+ The fall of potential at the cathode is proportional to the square of
+ the current, while the fall in the rest of the circuit is directly
+ proportional to the current. In the case of flames or hot gases, the
+ fall of potential at the cathode is much greater than that in the rest
+ of the circuit, so that in such cases the current through the gas
+ varies nearly as the square root of the potential difference. The
+ equation we have just obtained is of the form
+
+ V = Ai + Bi^2,
+
+ and H. A. Wilson has shown that a relation of this form represents the
+ results of his experiments on the conduction of electricity through
+ flames.
+
+ The expression for the fall of potential at the cathode is inversely
+ proportional to q^(3/2), q being the number of ions produced per cubic
+ centimetre per second close to the cathode; thus any increase in the
+ ionization at the cathode will diminish the potential fall at the
+ cathode, and as practically the whole potential difference between the
+ electrodes occurs at the cathode, a diminution in the potential fall
+ there will be much more important than a diminution in the electric
+ force in the uniform part of the discharge, when the force is
+ comparatively insignificant. This consideration explains a very
+ striking phenomenon discovered many years ago by Hittorf, who found
+ that if he put a wire carrying a bead of a volatile salt into the
+ flame, it produced little effect upon the current, unless it were
+ placed close to the cathode where it gave rise to an enormous increase
+ in the current, sometimes increasing the current more than a
+ hundredfold. The introduction of the salt increases very largely the
+ number of ions produced, so that q is much greater for a salted flame
+ than for a plain one. Thus Hittorf's result coincides with the
+ conclusions we have drawn from the theory of this class of conduction.
+
+ The fall of potential at the cathode is proportional to i - i0, where
+ i0 is the stream of negative electricity which comes from the cathode
+ itself, thus as i0 increases the fall of potential at the cathode
+ diminishes and the current sent by a given potential difference
+ through the gas increases. Now all metals give out negative particles
+ when heated, at a rate which increases very rapidly with the
+ temperature, but at the same temperature some metals give out more
+ than others. If the cathode is made of a metal which emits large
+ quantities of negative particles, (i - i0) will for a given value of i
+ be smaller than if the metal only emitted a small number of
+ particles; thus the cathode fall will be smaller for the metal with
+ the greater emissitivity, and the relation between the potential
+ difference and the current will be different in the two cases. These
+ considerations are confirmed by experience, for it has been found that
+ the current between electrodes immersed in a flame depends to a great
+ extent upon the metal of which the electrodes are made. Thus
+ Pettinelli (_Acc. dei Lincei_ [5], v. p. 118) found that, _ceteris
+ paribus_, the current between two carbon electrodes was about 500
+ times that between two iron ones. If one electrode was carbon and the
+ other iron, the current when the carbon was cathode and the iron anode
+ was more than 100 times the current when the electrodes were reversed.
+ The emission of negative particles by some metallic oxides, notably
+ those of calcium and barium, has been shown by Wehnelt (_Ann. der
+ Phys._ 11, p. 425) to be far greater than that of any known metal, and
+ the increase of current produced by coating the cathodes with these
+ oxides is exceedingly large; in some cases investigated by Tufts and
+ Stark (_Physik. Zeits._, 1908, 5, p. 248) the current was increased
+ many thousand times by coating the cathode with lime. No appreciable
+ effect is produced by putting lime on the anode.
+
+ _Conduction when all the Ions are of one Sign._--There are many
+ important cases in which the ions producing the current come from one
+ electrode or from a thin layer of gas close to the electrode, no
+ ionization occurring in the body of the gas or at the other electrode.
+ Among such cases may be mentioned those where one of the electrodes is
+ raised to incandescence while the other is cold, or when the negative
+ electrode is exposed to ultra-violet light. In such cases if the
+ electrode at which the ionization occurs is the positive electrode,
+ all the ions will be positively charged, while if it is the negative
+ electrode the ions will all be charged negatively. The theory of this
+ case is exceedingly simple. Suppose the electrodes are parallel planes
+ at right angles to the axis of x; let X be the electric force at a
+ distance x from the electrode where the ionization occurs, n the
+ number of ions (all of which are of one sign) at this place per cubic
+ centimetre, k the velocity of the ion under unit electric force, e the
+ charge on an ion, and i the current per unit area of the electrode.
+ Then we have dX/dx = 4[pi]ne, and if u is the velocity of the ion neu
+ = i. But u = kX, hence we have kX/4[pi] . dX/dx = i, and since the
+ right hand side of this equation does not depend upon x, we get
+ kX^2/8[pi] = ix + C, where C is a constant to be determined. If l is
+ the distance between the plates, and V the potential difference
+ between them,
+ _ _____ _ _
+ / l 1 /8[pi] | |
+ V = | Xdx = --- / ----- | ( il + C )^3/2 - C^3/2 |.
+ _/0 i \/ k |_ _|
+
+ We shall show that when the current is far below the saturation value,
+ C is very small compared with il, so that the preceding equation
+ becomes
+
+ V^2 = 8[pi]l^3i/k (1).
+
+ To show that for small currents C is small compared with il, consider
+ the case when the ionization is confined to a thin layer, thickness d
+ close to the electrode, in that layer let n0 be the value of n, then
+ we have q = [alpha]n0^2 + i/ed. If X0 be the value of X when x = 0,
+ kX0n0e = i, and,
+
+ kX0^2 i^2 [alpha] i^2
+ C = ----- = ------------ = --------- . -------- (2).
+ 8[pi] n0^2ke.8[pi] 8[pi]ke^2 q + i/ed
+
+ Since [alpha]/8[pi]ke is, as we have seen, less than unity, C will be
+ small compared with il, if i/(eq + i/d) is small compared with l. If
+ I0 is the saturation current, q = I0/ed, so that the former expression
+ = id/(I0 + i), if i is small compared with I0, this expression is
+ small compared with d, and therefore _a fortiori_ compared with l, so
+ that we are justified in this case in using equation (1).
+
+ From equation (2) we see that the current increases as the square of
+ the potential difference. Here an increase in the potential difference
+ produces a much greater percentage increase than in conduction through
+ metals, where the current is proportional to the potential difference.
+ When the ionization is distributed through the gas, we have seen that
+ the current is approximately proportional to the square root of the
+ potential, and so increases more slowly with the potential difference
+ than currents through metals. From equation (1) the current is
+ inversely proportional to the cube of the distance between the
+ electrodes, so that it falls off with great rapidity as this distance
+ is increased. We may note that for a given potential difference the
+ expression for the current does not involve q, the rate of production
+ of the ions at the electrode, in other words, if we vary the
+ ionization the current will not begin to be affected by the strength
+ of the ionization until this falls so low that the current is a
+ considerable fraction of the saturation current. For the same
+ potential difference the current is proportional to k, the velocity
+ under unit electric force of the ion which carries the current. As the
+ velocity of the negative ion is greater than that of the positive, the
+ current when the ionization is confined to the neighbourhood of one of
+ the electrodes will be greater when that electrode is made cathode
+ than when it is anode. Thus the current will appear to pass more
+ easily in one direction than in the opposite.
+
+ Since the ions which carry the current have to travel all the way from
+ one electrode to the other, any obstacle which is impervious to these
+ ions will, if placed between the electrodes, stop the current to the
+ electrode where there is no ionization. A plate of metal will be as
+ effectual as one made of a non-conductor, and thus we get the
+ remarkable result that by interposing a plate of an excellent
+ conductor like copper or silver between the electrode, we can entirely
+ stop the current. This experiment can easily be tried by using a hot
+ plate as the electrode at which the ionization takes place: then if
+ the other electrode is cold the current which passes when the hot
+ plate is cathode can be entirely stopped by interposing a cold metal
+ plate between the electrodes.
+
+_Methods of counting the Number of Ions._--The detection of the ions and
+the estimation of their number in a given volume is much facilitated by
+the property they possess of promoting the condensation of water-drops
+in dust-free air supersaturated with water vapour. If such air contains
+no ions, then it requires about an eightfold supersaturation before any
+water-drops are formed; if, however, ions are present C. T. R. Wilson
+(_Phil. Trans._ 189, p. 265) has shown that a sixfold supersaturation is
+sufficient to cause the water vapour to condense round the ions and to
+fall down as raindrops. The absence of the drops when no ions are
+present is due to the curvature of the drop combined with the surface
+tension causing, as Lord Kelvin showed, the evaporation from a small
+drop to be exceeding rapid, so that even if a drop of water were formed
+the evaporation would be so great in its early stages that it would
+rapidly evaporate and disappear. It has been shown, however (J. J.
+Thomson, _Application of Dynamics to Physics and Chemistry_, p. 164;
+_Conduction of Electricity through Gases_, 2nd ed. p. 179), that if a
+drop of water is charged with electricity the effect of the charge is to
+diminish the evaporation; if the drop is below a certain size the effect
+the charge has in promoting condensation more than counterbalances the
+effect of the surface tension in promoting evaporation. Thus the
+electric charge protects the drop in the most critical period of its
+growth. The effect is easily shown experimentally by taking a bulb
+connected with a piston arranged so as to move with great rapidity. When
+the piston moves so as to increase the volume of the air contained in
+the bulb the air is cooled by expansion, and if it was saturated with
+water vapour before it is supersaturated after the expansion. By
+altering the throw of the piston the amount of supersaturation can be
+adjusted within very wide limits. Let it be adjusted so that the
+expansion produces about a sixfold supersaturation; then if the gas is
+not exposed to any ionizing agents very few drops (and these probably
+due to the small amount of ionization which we have seen is always
+present in gases) are formed. If, however, the bulb is exposed to strong
+Rontgen rays expansion produces a dense cloud which gradually falls down
+and disappears. If the gas in the bulb at the time of its exposure to
+the Rontgen rays is subject to a strong electric field hardly any cloud
+is formed when the gas is suddenly expanded. The electric field removes
+the charged ions from the gas as soon as they are formed so that the
+number of ions present is greatly reduced. This experiment furnishes a
+very direct proof that the drops of water which form the cloud are only
+formed round the ions.
+
+This method gives us an exceedingly delicate test for the presence of
+ions, for there is no difficulty in detecting ten or so raindrops per
+cubic centimetre; we are thus able to detect the presence of this number
+of ions. This result illustrates the enormous difference between the
+delicacy of the methods of detecting ions and those for detecting
+uncharged molecules; we have seen that we can easily detect ten ions per
+cubic centimetre, but there is no known method, spectroscopic or
+chemical, which would enable us to detect a billion (10^12) times this
+number of uncharged molecules. The formation of the water-drops round
+the charged ions gives us a means of counting the number of ions present
+in a cubic centimetre of gas; we cool the gas by sudden expansion until
+the supersaturation produced by the cooling is sufficient to cause a
+cloud to be formed round the ions, and the problem of finding the number
+of ions per cubic centimetre of gas is thus reduced to that of finding
+the number of drops per cubic centimetre in the cloud. Unless the drops
+are very few and far between we cannot do this by direct counting; we
+can, however, arrive at the result in the following way. From the amount
+of expansion of the gas we can calculate the lowering produced in its
+temperature and hence the total quantity of water precipitated. The
+water is precipitated as drops, and if all the drops are the same size
+the number per cubic centimetre will be equal to the volume of water
+deposited per cubic centimetre, divided by the volume of one of the
+drops. Hence we can calculate the number of drops if we know their size,
+and this can be determined by measuring the velocity with which they
+fall under gravity through the air.
+
+ The theory of the fall of a heavy drop of water through a viscous
+ fluid shows that v = (2/9)ga^2/[mu], where a is the radius of the drop,
+ g the acceleration due to gravity, and [mu] the coefficient of
+ viscosity of the gas through which the drop falls. Hence if we know v
+ we can deduce the value of a and hence the volume of each drop and the
+ number of drops.
+
+ _Charge on Ion._--By this method we can determine the number of ions
+ per unit volume of an ionized gas. Knowing this number we can proceed
+ to determine the charge on an ion. To do this let us apply an electric
+ force so as to send a current of electricity through the gas, taking
+ care that the current is only a small fraction of the saturating
+ current. Then if u is the sum of the velocities of the positive and
+ negative ions produced in the electric field applied to the gas, the
+ current through unit area of the gas is neu, where n is the number of
+ positive or negative ions per cubic centimetre, and e the charge on an
+ ion. We can easily measure the current through the gas and thus
+ determine neu; we can determine n by the method just described, and u,
+ the velocity of the ions under the given electric field, is known from
+ the experiments of Zeleny and others. Thus since the product neu, and
+ two of the factors n, u are known, we can determine the other factor
+ e, the charge on the ion. This method was used by J. J. Thomson, and
+ details of the method will be found in _Phil. Mag._ [5], 46, p. 528;
+ [5], 48, p. 547; [6], 5, p. 346. The result of these measurements
+ shows that the charge on the ion is the same whether the ionization is
+ by Rontgen rays or by the influence of ultra-violet light on a metal
+ plate. It is the same whether the gas ionized is hydrogen, air or
+ carbonic acid, and thus is presumably independent of the nature of the
+ gas. The value of e formed by this method was 3.4 X 10^-10
+ electrostatic units.
+
+ H. A. Wilson (_Phil. Mag._ [6], 5, p. 429) used another method. Drops
+ of water, as we have seen, condense more easily on negative than on
+ positive ions. It is possible, therefore, to adjust the expansion so
+ that a cloud is formed on the negative but not on the positive ions.
+ Wilson arranged the experiments so that such a cloud was formed
+ between two horizontal plates which could be maintained at different
+ potentials. The charged drops between the plates were acted upon by a
+ uniform vertical force which affected their rate of fall. Let X be the
+ vertical electric force, e the charge on the drop, v1 the rate of fall
+ of the drop when this force acts, and v the rate of fall due to
+ gravity alone. Then since the rate of fall is proportionate to the
+ force on the drop, if a is the radius of the drop, and [rho] its
+ density, then
+
+ Xe + (4/3)[pi][rho]ga^3 v1
+ ---------------------- = ---,
+ 4/3[pi][rho]ga^3 v
+
+ or Xe = (4/3)[pi][rho]ga^3(v1 - v)/v.
+
+ But v = 2/9ga^2[rho]/[mu],
+
+ so that
+
+ / / [mu]^3 v^(3/2)(v1 - v)
+ Xe = \/ 2.9[pi] - / ------ . ---------------.
+ \/ g[rho] v
+
+ Thus if X, v, v1 are known e can be determined. Wilson by this method
+ found that e was 3.1 X 10^-10 electrostatic units. A few of the ions
+ carried charges 2e or 3e.
+
+ Townsend has used the following method to compare the charge carried
+ by a gaseous ion with that carried by an atom of hydrogen in the
+ electrolysis of solution. We have
+
+ u/D = Ne/[Pi],
+
+ where D is the coefficient of diffusion of the ions through the gas, u
+ the velocity of the ion in the same gas when acted on by unit electric
+ force, N the number of molecules in a cubic centimetre of the gas when
+ the pressure is [Pi] dynes per square centimetre, and e the charge in
+ electrostatic units. This relation is obtained on the hypothesis that
+ N ions in a cubic centimetre produce the same pressure as N uncharged
+ molecules.
+
+ We know the value of D from Townsend's experiments and the values of u
+ from those of Zeleny. We get the following values for Ne X 10^-10:--
+
+ +---------------+---------------------+---------------------+
+ | | Moist Gas. | Moist Gas. |
+ +---------------+----------+----------+----------+----------+
+ | Gas. | Positive | Negative | Positive | Negative |
+ | | Ions. | Ions. | Ions. | Ions. |
+ +---------------+----------+----------+----------+----------+
+ | Air | 1.28 | 1.29 | 1.46 | 1.31 |
+ | Oxygen | 1.34 | 1.27 | 1.63 | 1.36 |
+ | Carbonic acid | 1.01 | .87 | .99 | .93 |
+ | Hydrogen | 1.24 | 1.18 | 1.63 | 1.25 |
+ +---------------+----------+----------+----------+----------+
+ | Mean | 1.22 | 1.15 | 1.43 | 1.21 |
+ +---------------+----------+----------+----------+----------+
+
+
+ Since 1.22 cubic centimetres of hydrogen at the temperature 15 deg. C.
+ and pressure 760 mm. of mercury are liberated by the passage through
+ acidulated water of one electromagnetic unit of electricity or 3 X
+ 10^10 electrostatic units, and since in one cubic centimetre of the
+ gas there are 2.46 N atoms of hydrogen, we have, if E is the charge in
+ electrostatic units, on the atom of hydrogen in the electrolysis of
+ solutions
+
+ 2.46NE = 3 X 10^10,
+
+ or
+
+ NE = 1.22 X 10^10.
+
+ The mean of the values of Ne in the preceding table is 1.24 X 10^10.
+ Hence we may conclude that the charge of electricity carried by a
+ gaseous ion is equal to the charge carried by the hydrogen atom in the
+ electrolysis of solutions. The values of Ne for the different gases
+ differ more than we should have expected from the probable accuracy of
+ the determination of D and the velocity of the ions: Townsend (_Proc.
+ Roy. Soc._ 80, p. 207) has shown that when the ionization is produced
+ by Rontgen rays some of the positive ions carry a double charge and
+ that this accounts for the values of Ne being greater for the positive
+ than for the negative ions. Since we know the value of e, viz. 3.5 X
+ 10^-10, and, also Ne, = 1.24 X 10^10, we find N the number of
+ molecules in a cubic centimetre of gas at standard temperature and
+ pressure to be equal to 3.5 X 10^19. This method of obtaining N is the
+ only one which does not involve any assumption as to the shape of the
+ molecules and the forces acting between them.
+
+ Another method of determining the charge carried by an ion has been
+ employed by Rutherford (_Proc. Roy. Soc._ 81, pp. 141, 162), in which
+ the positively electrified particles emitted by radium are made use
+ of. The method consists of: (1) Counting the number of [alpha]
+ particles emitted by a given quantity of radium in a known time. (2)
+ Measuring the electric charge emitted by this quantity in the same
+ time. To count the number of the [alpha] particles the radium was so
+ arranged that it shot into an ionization chamber a small number of
+ [alpha] particles per minute; the interval between the emission of
+ individual particles was several seconds. When an [alpha] particle
+ passed into the vessel it ionized the gas inside and so greatly
+ increased its conductivity; thus, if the gas were kept exposed to an
+ electric field, the current through the gas would suddenly increase
+ when an [alpha] particle passed into the vessel. Although each [alpha]
+ particle produces about thirty thousand ions, this is hardly large
+ enough to produce the conductivity appreciable without the use of very
+ delicate apparatus; to increase the conductivity Rutherford took
+ advantage of the fact that ions, especially negative ones, when
+ exposed to a strong electric field, produce other ions by collision
+ against the molecules of the gas through which they are moving. By
+ suitably choosing the electric field and the pressure in the
+ ionization chamber, the 30,000 ions produced by each [alpha] particle
+ can be multiplied to such an extent that an appreciable current passes
+ through the ionization chamber on the arrival of each [alpha]
+ particle. An electrometer placed in series with this vessel will show
+ by its deflection when an [alpha] particle enters the chamber, and by
+ counting the number of deflections per minute we can determine the
+ number of [alpha] particles given out by the radium in that time.
+ Another method of counting this number is to let the particles fall on
+ a phosphorescent screen, and count the number of scintillations on the
+ screen in a certain time. Rutherford has shown that these two methods
+ give concordant results.
+
+ The charge of positive electricity given out by the radium was
+ measured by catching the [alpha] particles in a Faraday cylinder
+ placed in a very highly exhausted vessel, and measuring the charge per
+ minute received by this cylinder. In this way Rutherford showed that
+ the charge on the [alpha] particle was 9.4 X 10^-10 electrostatic
+ units. Now e/m for the [alpha] particle = 5 X 10^3, and there is
+ evidence that the [alpha] particle is a charged atom of helium; since
+ the atomic weight of helium is 4 and e/m for hydrogen is 10^4, it
+ follows that the charge on the helium atom is twice that on the
+ hydrogen, so that the charge on the hydrogen atom is 4.7 X 10^-10
+ electrostatic units.
+
+_Calculation of the Mass of the Ions at Low Pressures._--Although at
+ordinary pressures the ion seems to have a very complex structure and to
+be the aggregate of many molecules, yet we have evidence that at very
+low pressures the structure of the ion, and especially of the negative
+one, becomes very much simpler. This evidence is afforded by
+determination of the mass of the atom. We can measure the ratio of the
+mass of an ion to the charge on the ion by observing the deflections
+produced by magnetic and electric forces on a moving ion. If an ion
+carrying a charge e is moving with a velocity v, at a point where the
+magnetic force is H, a mechanical force acts on the ion, whose direction
+is at right angles both to the direction of motion of the ion and to the
+magnetic force, and whose magnitude is evH sin [theta], where [theta] is
+the angle between v and H. Suppose then that we have an ion moving
+through a gas whose pressure is so low that the free path of the ion is
+long compared with the distance through which it moves whilst we are
+experimenting upon it; in this case the motion of the ion will be free,
+and will not be affected by the presence of the gas.
+
+ Since the force is always at right angles to the direction of motion
+ of the ion, the speed of the ion will not be altered by the action of
+ this force; and if the ion is projected with a velocity v in a
+ direction at right angles to the magnetic force, and if the magnetic
+ force is constant in magnitude and direction, the ion will describe a
+ curve in a plane at right angles to the magnetic force. If [rho] is
+ the radius of curvature of this curve, m the mass of the ion,
+ mv^2/[rho] must equal the normal force acting on the ion, i.e. it must
+ be equal to Hev, or [rho] = mv/He. Thus the radius of curvature is
+ constant; the path is therefore a circle, and if we can measure the
+ radius of this circle we know the value of mv/He. In the case of the
+ rapidly moving negative ions projected from the cathode in a highly
+ exhausted tube, which are known as _cathode rays_, the path of the
+ ions can be readily determined since they make many substances
+ luminous when they impinge against them. Thus by putting a screen of
+ such a substance in the path of the rays the shape of the path will be
+ determined. Let us now suppose that the ion is acted upon by a
+ vertical electric force X and is free from magnetic force, if it be
+ projected with a horizontal velocity v, the vertical deflection y
+ after a time t is 1/2 X et^2/m, or if l is the horizontal distance
+ travelled over by the ion in this time we have since l = vt,
+
+ Xe l^2
+ y = 1/2 -- ---.
+ m v^2
+
+ Thus if we measure y and l we can deduce e/mv^2. From the effect of the
+ magnetic force we know e/mv. Combining these results we can find both
+ e/m and v.
+
+ [Illustration: FIG. 13.]
+
+ The method by which this determination is carried out in practice is
+ illustrated in fig. 13. The cathode rays start from the electrode C in
+ a highly exhausted tube, pass through two small holes in the plugs A
+ and B, the holes being in the same horizontal line. Thus a pencil of
+ rays emerging from B is horizontal and produces a bright spot at the
+ far end of the tube. In the course of their journey to the end of the
+ tube they pass between the horizontal plates E and D, by connecting
+ these plates with an electric battery a vertical electric field is
+ produced between E and D and the phosphorescent spot is deflected. By
+ measuring this deflection we determine e/mv^2. The tube is now placed
+ in a uniform magnetic field, the lines of magnetic force being
+ horizontal and at right angles to the plane of the paper. The magnetic
+ force makes the rays describe a circle in the plane of the paper, and
+ by measuring the vertical deflection of the phosphorescent patch at
+ the end of the tube we can determine the radius of this circle, and
+ hence the value of e/mv. From the two observations the value of e/m
+ and v can be calculated.
+
+ Another method of finding e/m for the negative ion which is applicable
+ in many cases to which the preceding one is not suitable, is as
+ follows: Let us suppose that the ion starts from rest and moves in a
+ field where the electric and magnetic forces are both uniform, the
+ electric force X being parallel to the axis of x, and the magnetic
+ force Z parallel to the axis of z; then if x, y, are the co-ordinates
+ of the ion at the time t, the equations of motion of the ion are--
+
+ d^2x dy
+ m ---- = Xe - He --,
+ dt^2 dt
+
+ d^2y dx
+ m ---- = He --.
+ dt^2 dt
+
+ The solution of these equations, if x, y, dx/dt, dy/dt all vanish when
+ t = 0, is
+
+ Xm / / e \ \
+ x = ---- {1 - cos( -- Ht ) }
+ eH^2 \ \ m / /
+
+ Xm /e / e \ \
+ y = ---- {-- Ht - sin( -- Ht ) }.
+ eH^2 \m \ m / /
+
+ These equations show that the path of the ion is a cycloid, the
+ generating circle of which has a diameter equal to 2Xm/eH^2, and rolls
+ on the line x = 0.
+
+ Suppose now that we have a number of ions starting from the plane x =
+ 0, and moving towards the plane x = a. The particles starting from x =
+ 0 describe cycloids, and the greatest distance they can get from the
+ plane is equal to the diameter of the generating circle of the
+ cycloid, i.e. to 2Xm/eH^2. (After reaching this distance they begin to
+ approach the plane.) Hence if a is less than the diameter of the
+ generating circle, all the particles starting from x = 0 will reach
+ the plane x = a, if this is unlimited in extent; while if a is greater
+ than the diameter of the generating circle none of the particles which
+ start from x = 0 will reach the plane x = a. Thus, if x = 0 is a plane
+ illuminated by ultra-violet light, and consequently the seat of a
+ supply of negative ions, and x = a a plane connected with an
+ electrometer, then if a definite electric intensity is established
+ between the planes, i.e. if X be fixed, so that the rate of emission
+ of negative ions from the illuminated plate is given, and if a is less
+ than 2Xm/eH^2, all the ions which start from x = 0 will reach x = a.
+ That is, the rate at which this plane receives an electric charge
+ will be the same whether there is a magnetic field between the plate
+ or not, but if a is greater than 2Xm/eH^2, then no particle which
+ starts from the plate x = 0 will reach the plate x = a, and this plate
+ will receive no charge. Thus the supply of electricity to the plate
+ has been entirely stopped by the magnetic field. Thus, on this theory,
+ if the distance between the plates is less than a certain value, the
+ magnetic force should produce no effect on the rate at which the
+ electrometer plate receives a charge, while if the distance is greater
+ than this value the magnetic force would completely stop the supply of
+ electricity to the plate. The actual phenomena are not so abrupt as
+ this theory indicates. We find that when the plates are very near
+ together the magnetic force produces a very slight effect, and this an
+ increase in the rate of charging of the plate. On increasing the
+ distance we come to a stage where the magnetic force produces a great
+ diminution in the rate of charging. It does not, however, stop it
+ abruptly, there being a considerable range of distance, in which the
+ magnetic force diminishes but does not destroy the current. At still
+ greater distances the current to the plate under the magnetic force is
+ quite inappreciable compared with that when there is no magnetic
+ force. We should get this gradual instead of abrupt decay of the
+ current if some of the particles, instead of all starting from rest,
+ started with a finite velocity; in that case the first particles
+ stopped would be those which started from rest. This would be when a =
+ 2Xm/eH^2. Thus if we measure the value of a when the magnetic force
+ first begins to affect the leak to the electrometer we determine
+ 2Xm/eH^2, and as we can easily measure X and H, we can deduce the value
+ of m/e.
+
+By these methods Thomson determined the value of e/m for the negative
+ions produced when ultra-violet light falls on a metal plate, as well as
+for the negative ions produced by an incandescent carbon filament in an
+atmosphere of hydrogen (_Phil. Mag._ [5], 48, p. 547) as well as for the
+cathode rays. It was found that the value of e/m for the negative ions
+was the same in all these cases, and that it was a constant quantity
+independent of the nature of the gas from which the ions are produced
+and the means used to produce them. It was found, too, that this value
+was more than a thousand times the value of e/M, where e is the charge
+carried by an atom of hydrogen in the electrolysis of solutions, and M
+the mass of an atom of hydrogen. We have seen that this charge is the
+same as that carried by the negative ion in gases; thus since e/m is
+more than a thousand times e/M, it follows that M must be more than a
+thousand times m. Thus the mass of the negative ion is exceedingly small
+compared with the mass of the atom of hydrogen, the smallest mass
+recognized in chemistry. The production of negative ions thus involves
+the splitting up of the atom, as from a collection of atoms something is
+detached whose mass is less than that of a single atom. It is important
+to notice in connexion with this subject that an entirely different line
+of argument, based on the Zeeman effect (see MAGNETO-OPTICS), leads to
+the recognition of negatively electrified particles for which e/m is of
+the same order as that deduced from the consideration of purely
+electrical phenomena. These small negatively electrified particles are
+called corpuscles. The latest determinations of e/m for corpuscles
+available are the following:--
+
+ Observer. e/m.
+
+ Classen (_Ber. deut. phys. Ges._ 6, p. 700) 1.7728 X 10^7
+ Bucherer (_Ann. der Phys._, 28, p. 513) 1.763 X 10^7
+
+It follows from electrical theory that when the corpuscles are moving
+with a velocity comparable with that of light their masses increase
+rapidly with their velocity. This effect has been detected by Kauffmann
+(_Gott. Nach._, Nov. 8, 1901), who used the corpuscles shot out from
+radium, some of which move with velocities only a few per cent less than
+that of light. Other experiments on this point have been made by
+Bucherer (_Ann. der Phys._ 28, p. 513).
+
+_Conductivity Produced by Ultra-Violet Light._--So much use has been
+made in recent times of ultra-violet light for producing ions that it is
+desirable to give some account of the electrical effects produced by
+light. The discovery by Hertz (_Wied. Ann._ 31, p. 983) in 1887, that
+the incidence of ultra-violet light on a spark gap facilitates the
+passage of a spark, led to a series of investigations by Hallwachs,
+Hoor, Righi and Stoletow, on the effect of ultra-violet light on
+electrified bodies. These researches have shown that a freshly cleaned
+metal surface, charged with negative electricity, rapidly loses its
+charge, however small, when exposed to ultra-violet light, and that if
+the surface is insulated and without charge initially, it acquires a
+positive charge under the influence of the light. The magnitude of this
+positive charge may be very much increased by directing a blast of air
+on the plate. This, as Zeleny (_Phil. Mag._ [5], 45, p. 272) showed, has
+the effect of blowing from the neighbourhood of the plate negatively
+electrified gas, which has similar properties to the charged gas
+obtained by the separation of ions from a gas exposed to Rontgen rays or
+uranium radiation. If the metal plate is positively electrified, there
+is no loss of electrification caused by ultra-violet light. This has
+been questioned, but a very careful examination of the question by
+Elster and Geitel (_Wied. Ann._ 57, p. 24) has shown that the apparent
+exceptions are due to the accidental exposure to reflected ultra-violet
+light of metal surfaces in the neighbourhood of the plate negatively
+electrified by induction, so that the apparent loss of charge is due to
+negative electricity coming up to the plate, and not to positive
+electricity going away from it. The ultra-violet light may be obtained
+from an arc-lamp, the effectiveness of which is increased if one of the
+terminals is made of zinc or aluminium, the light from these substances
+being very rich in ultra-violet rays; it may also be got very
+conveniently by sparking with an induction coil between zinc or cadmium
+terminals. Sunlight is not rich in ultra-violet light, and does not
+produce anything like so great an effect as the arc light. Elster and
+Geitel, who have investigated with great success the effects of light on
+electrified bodies, have shown that the more electro-positive metals
+lose negative charges when exposed to ordinary light, and do not need
+the presence of the ultra-violet rays. Thus they found that amalgams of
+sodium or potassium enclosed in a glass vessel lose a negative charge
+when exposed to daylight, though the glass stops the small amount of
+ultra-violet light left in sunlight after its passage through the
+atmosphere. If sodium or potassium be employed, or, what is more
+convenient, the mercury-like liquid obtained by mixing sodium and
+potassium in the proportion of their combining weights, they found that
+negative electricity was discharged by an ordinary petroleum lamp. If
+the still more electro-positive metal rubidium is used, the discharge
+can be produced by the light from a glass rod just heated to redness;
+but there is no discharge till the glass is luminous. Elster and Geitel
+arrange the metals in the following order for the facility with which
+negative electrification is discharged by light: rubidium, potassium,
+alloy of sodium and potassium, sodium, lithium, magnesium, thallium,
+zinc. With copper, platinum, lead, iron, cadmium, carbon and mercury the
+effects with ordinary light are too small to be appreciable. The order
+is the same as that in Volta's electro-chemical series. With
+ultra-violet light the different metals show much smaller differences in
+their power of discharging negative electricity than they do with
+ordinary light. Elster and Geitel found that the ratio of the
+photo-electric effects of two metals exposed to approximately
+monochromatic light depended upon the wave-length of the light,
+different metals showing a maximum sensitiveness in different parts of
+the spectrum. This is shown by the following table for the alkaline
+metals. The numbers in the table are the rates of emission of negative
+electricity under similar circumstances. The rate of emission under the
+light from a petroleum lamp was taken as unity:--
+
+ Blue. Yellow. Orange. Red.
+ Rb .16 .64 .33 .039
+ Na .37 .36 .14 .009
+ K .57 .07 .04 .002
+
+The table shows that the absorption of light by the metal has great
+influence on the photo-electric effect, for while potassium is more
+sensitive in blue light than sodium, the strong absorption of yellow
+light by sodium makes it more than five times more sensitive to this
+light than potassium. Stoletow, at an early period, called attention to
+the connexion between strong absorption and photo-electric effects. He
+showed that water, which does not absorb to any great extent either the
+ultra-violet or visible rays, does not show any photo-electric effect,
+while strongly coloured solutions, and especially solutions of
+fluorescent substances such as methyl green or methyl violet, do so to a
+very considerable extent; indeed, a solution of methyl green is more
+sensitive than zinc. Hallwachs (_Wied. Ann._ 37, p. 666) proved that in
+liquids showing photo-electric effects there is always strong
+absorption; we may, however, have absorption without these effects.
+Phosphorescent substances, such as calcium sulphide show this effect, as
+also do various specimens of fluor-spar. As phosphorescence and
+fluorescence are probably accompanied by a very intense absorption by
+the surface layers, the evidence is strong that to get the
+photo-electric effects we must have strong absorption of some kind of
+light, either visible or ultra-violet.
+
+[Illustration: FIG. 14.]
+
+If a conductor A is placed near a conductor B exposed to ultra-violet
+light, and if B is made the negative electrode and a difference of
+potential established between A and B, a current of electricity will
+flow between the conductors. The relation between the magnitude of the
+current and the difference of potential when A and B are parallel plates
+has been investigated by Stoletow (_Journal de physique_, 1890, 11, p.
+469), von Schweidler (_Wien. Ber._, 1899, 108, p. 273) and Varley
+(_Phil. Trans. A._, 1904, 202, p. 439). The results of some of Varley's
+experiments are represented in the curves shown in fig. 14, in which the
+ordinates are the currents and the abscissae the potentials. It will be
+seen that when the pressure is exceedingly low the current is
+independent of the potential difference and is equal to the negative
+charge carried off in unit time by the corpuscles emitted from the
+surface exposed to the light. At higher pressures the current rises far
+above these values and increases rapidly with the potential difference.
+This is due to the corpuscles emitted by the illuminated surface
+acquiring under the electric field such high velocities that when they
+strike against the molecules of the gas through which they are passing
+they ionize them, producing fresh ions which can carry on additional
+current. The relation between the current and the potential difference
+in this case is in accordance with the results of the theory of
+ionization by collision. The corpuscles emitted from a body under the
+action of ultra-violet light start from the surface with a finite
+velocity. The velocity is not the same for all the corpuscles, nor
+indeed could we expect that it should be: for as Ladenburg has shown
+(_Ann. der Phys._, 1903, 12, p. 558) the seat of their emission is not
+confined to the surface layer of the illuminated metal but extends to a
+layer of finite, though small, thickness. Thus the particles which start
+deep down will have to force their way through a layer of metal before
+they reach the surface, and in doing so will have their velocities
+retarded by an amount depending on the thickness of this layer. The
+variation in the velocity of the corpuscles is shown in the following
+table, due to Lenard (_Ann. der Phys._, 1902, 8, p. 149).
+
+ +------------------------------------+--------+----------+-----------+
+ | | Carbon.| Platinum.| Aluminium.|
+ +------------------------------------+--------+----------+-----------+
+ | Corpuscles emitted with velocities | | | |
+ | between 12 and 8 X 10^7 cm sec. | 0.000 | 0.000 | 0.004 |
+ | between 8 and 4 X 10^7 cm sec. | 0.049 | 0.155 | 0.151 |
+ | between 4 and 0 X 10^7 cm sec. | 0.67 | 0.65 | 0.49 |
+ | | | | |
+ | Corpuscles only emitted with the | | | |
+ | help of an external electric | 0.28 | 0.21 | 0.35 |
+ | field. +--------+----------+-----------|
+ | | 1.00 | 1.00 | 1.00 |
+ +------------------------------------+--------+----------+-----------+
+
+If the illuminated surface is completely surrounded by an envelope of
+the same metal insulated from and completely shielded from the light,
+the emission of the negative corpuscles from the illuminated surface
+would go on until the potential difference V between this surface and
+the envelope became so great that the corpuscles with the greatest
+velocity lost their energy before reaching the envelope, i.e. if m is
+the mass, e the charge on a corpuscle, v the greatest velocity of
+projection, until Ve = 1/2mv^2. The values found for V by different
+observers are not very consistent. Lenard found that V for aluminium was
+about 3 volts and for platinum 2. Millikan and Winchester (_Phil. Mag._,
+July 1907) found for aluminium V = .738. The apparatus used by them was
+so complex that the interpretation of their results is difficult.
+
+An extremely interesting fact discovered by Lenard is that the velocity
+with which the corpuscles are emitted from the metal is independent of
+the intensity of the incident light. The quantity of corpuscles
+increases with the intensity, but the velocity of the individual
+corpuscles does not. It is worthy of notice that in other cases when
+negative corpuscles are emitted from metals, as for example when the
+metals are exposed to cathode rays, Canal-strahlen, or Rontgen rays, the
+velocity of the emitted corpuscles is independent of the intensity of
+the primary radiation which excites them. The velocity is not, however,
+independent of the nature of the primary rays. Thus when light is used
+to produce the emission of corpuscles the velocity, as Ladenburg has
+shown, depends on the wave length of the light, increasing as the wave
+length diminishes. The velocity of corpuscles emitted under the action
+of cathode rays is greater than that of those ejected by light, while
+the incidence of Rontgen rays produces the emission of corpuscles moving
+much more rapidly than those in the cases already mentioned, and the
+harder the primary rays the greater is the velocity of the corpuscles.
+
+The importance of the fact that the velocity and therefore the energy of
+the corpuscles emitted from the metal is independent of the intensity of
+the incident light can hardly be overestimated. It raises the most
+fundamental questions as to the nature of light and the constitution of
+the molecules. What is the source of the energy possessed by these
+corpuscles? Is it the light, or in the stores of internal energy
+possessed by the molecule? Let us follow the consequences of supposing
+that the energy comes from the light. Then, since the energy is
+independent of the intensity of the light, the electric forces which
+liberate the corpuscles must also be independent of that intensity. But
+this cannot be the case if, as is usually assumed in the electromagnetic
+theory, the wave front consists of a uniform distribution of electric
+force without structure, for in this case the magnitude of the electric
+force is proportional to the square root of the intensity. On the
+emission theory of light a difficulty of this kind would not arise, for
+on that theory the energy in a luminiferous particle remains constant as
+the particle pursues its flight through space. Thus any process which a
+single particle is able to effect by virtue of its energy will be done
+just as well a thousand miles away from the source of light as at the
+source itself, though of course in a given space there will not be
+nearly so many particles to do this process far from the source as there
+are close in. Thus, if one of the particles when it struck against a
+piece of metal caused the ejection of a corpuscle with a given velocity,
+the velocity of emission would not depend on the intensity of the light.
+There does not seem any reason for believing that the electromagnetic
+theory is inconsistent with the idea that on this theory, as on the
+emission theory, the energy in the light wave may instead of being
+uniformly distributed through space be concentrated in bundles which
+occupy only a small fraction of the volume traversed by the light, and
+that as the wave travels out the bundles get farther apart, the energy
+in each remaining undiminished. Some such view of the structure of light
+seems to be required to account for the fact that when a plate of metal
+is struck by a wave of ultra-violet light, it would take years before
+the corpuscles emitted from the metal would equal in number the
+molecules on the surface of the metal plate, and yet on the ordinary
+theory of light each one of these is without interruption exposed to the
+action of the light. The fact discovered by E. Ladenburg (_Verh. d.
+deutsch. physik. Ges._ 9, p. 504) that the velocity with which the
+corpuscles are emitted depends on the wave length of the light suggests
+that the energy in each bundle depends upon the wave length and
+increases as the wave length diminishes.
+
+These considerations illustrate the evidence afforded by photo-electric
+effects on the nature of light; these effects may also have a deep
+significance with regard to the structure of matter. The fact that the
+energy of the individual corpuscles is independent of the intensity of
+the light might be explained by the hypothesis that the energy of the
+corpuscles does not come from the light but from the energy stored up in
+the molecules of the metal exposed to the light. We may suppose that
+under the action of the light some of the molecules are thrown into an
+unstable state and explode, ejecting corpuscles; the light in this case
+acts only as a trigger to liberate the energy in the atom, and it is
+this energy and not that of the light which goes into the corpuscles. In
+this way the velocity of the corpuscles would be independent of the
+intensity of the light. But it may be asked, is this view consistent
+with the result obtained by Ladenburg that the velocity of the
+corpuscles depends upon the nature of the light? If light of a definite
+wave length expelled corpuscles with a definite and uniform velocity, it
+would be very improbable that the emission of the corpuscles is due to
+an explosion of the atoms. The experimental facts as far as they are
+known at present do not allow us to say that the connexion between the
+velocity of the corpuscles and the wave length of the light is of this
+definite character, and a connexion such as a gradual increase of
+average velocity as the wave length of the light diminishes, would be
+quite consistent with the view that the corpuscles are ejected by the
+explosion of the atom. For in a complex thing like an atom there may be
+more than one system which becomes unstable when exposed to light. Let
+us suppose that there are two such systems, A and B, of which B ejects
+the corpuscles with the greater velocity. If B is more sensitive to the
+short waves, and A to the long ones, then as the wave length of the
+light diminishes the proportion of the corpuscles which come from B will
+increase, and as these are the faster, the average velocity of the
+corpuscles emitted will also increase. And although the potential
+acquired by a perfectly insulated piece of metal when exposed to
+ultra-violet light would depend only on the velocity of the fastest
+corpuscles and not upon their number, in practice perfect insulation is
+unattainable, and the potential actually acquired is determined by the
+condition that the gain of negative electricity by the metal through
+lack of insulation, is equal to the loss by the emission of negatively
+electrified corpuscles. The potential acquired will fall below that
+corresponding to perfect insulation by an amount depending on the number
+of the faster corpuscles emitted, and the potential will rise if the
+proportion of the rapidly moving corpuscles is increased, even though
+there is no increase in their velocity. It is interesting to compare
+other cases in which corpuscles are emitted with the case of
+ultra-violet light. When a metal or gas is bombarded by cathode rays it
+emits corpuscles and the velocity of these is found to be independent of
+the velocity of the cathode rays which excite them; the velocity is
+greater than for corpuscles emitted under ultra-violet light. Again,
+when bodies are exposed to Rontgen rays they emit corpuscles moving with
+a much greater velocity than those excited by cathode rays, but again
+the velocity does not depend upon the intensity of the rays although it
+does to some extent on their hardness. In the case of cathode and
+Rontgen rays, the velocity with which the corpuscles are emitted seems,
+as far as we know at present, to vary slightly, but only slightly, with
+the nature of the substance on which the rays fall. May not this
+indicate that the first effect of the primary rays is to detach a
+neutral doublet, consisting of a positive and negative charge, this
+doublet being the same from whatever system it is detached? And that the
+doublet is unstable and explodes, expelling the negative charge with a
+high velocity, and the positive one, having a much larger charge, with a
+much smaller velocity, the momentum of the negative charge being equal
+to that of the positive.
+
+Up to now we have been considering the effects produced when light is
+incident on metals. Lenard found (and the result has been confirmed by
+the experiments of J. J. Thomson and Lyman) that certain kinds of
+ultra-violet light ionize a gas when they pass through. The type of
+ultra-violet light which produces this effect is so easily absorbed that
+it is stopped by a layer a few millimetres thick of air at atmospheric
+pressure.
+
+_Ionization by Collision._--When the ionization of the gas is produced
+by external agents such as Rontgen rays or ultra-violet light, the
+electric field produces a current by setting the positive ions moving in
+one direction, and the negative ones in the opposite; it makes use of
+ions already made and does not itself give rise to ionization. In many
+cases, however, such as in electric sparks, there are no external agents
+to produce ionization and the electric field has to produce the ions as
+well as set them in motion. When the ionization is produced by external
+means the smallest electric field is able to produce a current through
+the gas; when, however, these external means are absent no current is
+produced unless the strength of the electric field exceeds a certain
+critical value, which depends not merely upon the nature of the gas but
+also upon the pressure and the dimensions of the vessel in which it is
+contained. The variation of the electric field required to produce
+discharge can be completely explained if we suppose that the ionization
+of the gas is produced by the impact with its molecules of corpuscles,
+and in certain cases of positive ions, which under the influence of the
+electric field have acquired considerable kinetic energy. We have direct
+evidence that rapidly moving corpuscles are able to ionize molecules
+against which they strike, for the cathode rays consist of such
+corpuscles, and these when they pass through a gas produce large amounts
+of ionization. Suppose then that we have in a gas exposed to an electric
+field a few corpuscles. These will be set in motion by the field and
+will acquire an amount of energy in proportion to the product of the
+electric force, their charge, and the distance travelled in the
+direction of the electric field between two collisions with the
+molecules of the gas. If this energy is sufficient to give them the
+ionizing property possessed by cathode rays, then when a corpuscle
+strikes against a molecule it will detach another corpuscle; this under
+the action of the electric field will acquire enough energy to produce
+corpuscles on its own account, and so as the corpuscles move through the
+gas their number will increase in geometrical progression. Thus, though
+there were but few corpuscles to begin with, there may be great
+ionization after these have been driven some distance through the gas by
+the electric field.
+
+ The number of ions produced by collisions can be calculated by the
+ following method. Let the electric force be parallel to the axis of x,
+ and let n be the number of corpuscles per unit volume at a place fixed
+ by the co-ordinate x; then in unit time these corpuscles will make
+ nu/[lambda] collisions with the molecules, if u is the velocity of a
+ corpuscle and [lambda] the mean free path of a corpuscle. When the
+ corpuscles are moving fast enough to produce ions by collision their
+ velocities are very much greater than those they would possess at the
+ same temperature if they were not acted on by electrical force, and so
+ we may regard the velocities as being parallel to the axis of x and
+ determined by the electric force and the mean free path of the
+ corpuscles. We have to consider how many of the nu/[lambda] collisions
+ which take place per second will produce ions. We should expect that
+ the ionization of a molecule would require a certain amount of energy,
+ so that if the energy of the corpuscle fell below this amount no
+ ionization would take place, while if the energy of the corpuscle were
+ exceedingly large, every collision would result in ionization. We
+ shall suppose that a certain fraction of the number of collisions
+ result in ionization and that this fraction is a function of the
+ energy possessed by the corpuscle when it collides against the
+ molecules. This energy is proportional to Xe[lambda] when X is the
+ electric force, e the charge on the corpuscle, and [lambda] the mean
+ free path. If the fraction of collisions which produce ionization is
+ [int](Xe[lambda]), then the number of ions produced per cubic
+ centimetre per second is [int](Xe[lambda])nu/[lambda]. If the
+ collisions follow each other with great rapidity so that a molecule
+ has not had time to recover from one collision before it is struck
+ again, the effect of collisions might be cumulative, so that a
+ succession of collisions might give rise to ionization, though none of
+ the collisions would produce an ion by itself. In this case [int]
+ would involve the frequency of the collisions as well as the energy of
+ the corpuscle; in other words, it might depend on the current through
+ the gas as well as upon the intensity of the electric field. We
+ shall, however, to begin with, assume that the current is so small
+ that this cumulative effect may be neglected.
+
+ Let us now consider the rate of increase, dn/dt, in the number of
+ corpuscles per unit volume. In consequence of the collisions,
+ [int](Xe[lambda])nu/[lambda] corpuscles are produced per second; in
+ consequence of the motion of the corpuscles, the number which leave
+ unit volume per second is greater than those which enter it by
+ (d/dx)(nu); while in a certain number of collisions a corpuscle will
+ stick to the molecule and will thus cease to be a free corpuscle. Let
+ the fraction of the number of collisions in which this occurs be
+ [beta]. Thus the gain in the number of corpuscles is
+ [int](Xe[lambda])nu/[lambda], while the loss is (d/dx)(nu) +
+ [beta](nu)/[lambda]; hence
+
+ dn nu d [beta]nu
+ -- = [int](Xe[lambda]) -------- - --(nu) - --------.
+ dt [lambda] dx [lambda]
+
+ When things are in a steady state dn/dt = 0, and we have
+
+ d 1 / \
+ --(nu) = --------( [int](Xe[lambda]) - [beta] )nu.
+ dx [lambda] \ /
+
+ If the current is so small that the electrical charges in the gas are
+ not able to produce any appreciable variations in the field, X will be
+ constant and we get nu = C[epsilon]^{[alpha]x}, where [alpha] =
+ {[int](Xe[lambda]) - [beta]}/[lambda]. If we take the origin from
+ which we measure x at the cathode, C is the value of nu at the
+ cathode, i.e. it is the number of corpuscles emitted per unit area of
+ the cathode per unit time; this is equal to i0/e if i0 is the quantity
+ of negative electricity coming from unit area of the cathode per
+ second, and e the electric charge carried by a corpuscle. Hence we
+ have nue = i0[epsilon]^{[alpha]x}. If l is the distance between the
+ anode and the cathode, the value of nue, when x = l, is the current
+ passing through unit area of the gas, if we neglect the electricity
+ carried by negatively electrified carriers other than corpuscles.
+ Hence i = i0[epsilon]^{[alpha]l}. Thus the current between the plates
+ increases in geometrical progression with the distance between the
+ plates.
+
+ By measuring the variation of the current as the distance between the
+ plates is increased, Townsend, to whom we owe much of our knowledge on
+ this subject, determined the values of [alpha] for different values of
+ X and for different pressures for air, hydrogen and carbonic acid gas
+ (_Phil. Mag._ [6], 1, p. 198). Since [lambda] varies inversely as the
+ pressure, we see that [alpha] may be written in the form p[phi](X/p)
+ or [alpha]/X = F(X/p). The following are some of the values of [alpha]
+ found by Townsend for air.
+
+ +---------+----------+----------+----------+----------+----------+
+ | X Volts | Pressure | Pressure | Pressure | Pressure | Pressure |
+ | per cm. | .17 mm. | .38 mm. | 1.10 mm. | 2.1 mm. | 4.1 mm. |
+ +---------+----------+----------+----------+----------+----------+
+ | 20 | .24 | | | | |
+ | 40 | .65 | .34 | | | |
+ | 80 | 1.35 | 1.3 | .45 | .13 | |
+ | 120 | 1.8 | 2.0 | 1.1 | .42 | .13 |
+ | 160 | 2.1 | 2.8 | 2.0 | .9 | .28 |
+ | 200 | | 3.4 | 2.8 | 1.6 | .5 |
+ | 240 | 2.45 | 3.8 | 4.0 | 2.35 | .99 |
+ | 320 | 2.7 | 4.5 | 5.5 | 4.0 | 2.1 |
+ | 400 | | 5.0 | 6.8 | 6.0 | 3.6 |
+ | 480 | 3.15 | 5.4 | 8.0 | 7.8 | 5.3 |
+ | 560 | | 5.8 | 9.3 | 9.4 | 7.1 |
+ | 640 | 3.25 | 6.2 | 10.6 | 10.8 | 8.9 |
+ +---------+----------+----------+----------+----------+----------+
+
+ We see from this table that for a given value of X, [alpha] for small
+ pressures increases as the pressure increases; it attains a maximum at
+ a particular pressure, and then diminishes as the pressure increases.
+ The increase in the pressure increases the number of collisions, but
+ diminishes the energy acquired by the corpuscle in the electric field,
+ and thus diminishes the change of any one collision resulting in
+ ionization. If we suppose the field is so strong that at some
+ particular pressure the energy acquired by the corpuscle is well above
+ the value required to ionize at each collision, then it is evident
+ that increasing the number of collisions will increase the amount of
+ ionization, and therefore [alpha], and [alpha] cannot begin to
+ diminish until the pressure has increased to such an extent that the
+ mean free path of a corpuscle is so small that the energy acquired by
+ the corpuscle from the electric field falls below the value when each
+ collision results in ionization.
+
+ The value of p, when X is given, for which [alpha] is a maximum, is
+ proportional to X; this follows at once from the fact that [alpha] is
+ of the form X.F(X/p). The value of X/p for which F(X/p) is a maximum is
+ seen from the preceding table to be about 420, when X is expressed in
+ volts per centimetre and p in millimetres of mercury. The maximum value
+ of F(X/p) is about 1/60. Since the current passing between two planes
+ at a distance l apart is i0[epsilon]^{[alpha]l} or
+ i0[epsilon]^{XlF(X/p)}, and since the force between the plates is
+ supposed to be uniform, Xl is equal to V, the potential between the
+ plates; hence the current between the plates is i0[epsilon]^{VlF(X/p)},
+ and the greatest value it can have is i0[epsilon]^{V/60}. Thus the
+ ratio between the current between the plates when there is ionization
+ and when there is none cannot be greater than [epsilon]^{V/60}, when V
+ is measured in volts. This result is based on Townsend's experiments
+ with very weak currents; we must remember, however, that when the
+ collisions are so frequent that the effects of collisions can
+ accumulate, [alpha] may have much larger values than when the current
+ is small. In some experiments made by J. J. Thomson with intense
+ currents from cathodes covered with hot lime, the increase in the
+ current when the potential difference was 60 volts, instead of being e
+ times the current when there was no ionization, as the preceding theory
+ indicates, was several hundred times that value, thus indicating a
+ great increase in [alpha] with the strength of the current.
+
+ Townsend has shown that we can deduce from the values of [alpha] the
+ mean free path of a corpuscle. For if the ionization is due to the
+ collisions with the corpuscles, then unless one collision detaches
+ more than one corpuscle the maximum number of corpuscles produced will
+ be equal to the number of collisions. When each collision results in
+ the production of a corpuscle, [alpha] = 1/[lambda] and is independent
+ of the strength of the electric field. Hence we see that the value of
+ [alpha], when it is independent of the electric field, is equal to the
+ reciprocal of the free path. Thus from the table we infer that at a
+ pressure of 17 mm. the mean free path is 1/325 cm.; hence at 1 mm. the
+ mean free path of a corpuscle is 1/19 cm. Townsend has shown that this
+ value of the mean free path agrees well with the value 1/21 cm.
+ deduced from the kinetic theory of gases for a corpuscle moving
+ through air. By measuring the values of [alpha] for hydrogen and
+ carbonic acid gas Townsend and Kirby (_Phil. Mag._ [6], 1, p. 630)
+ showed that the mean free paths for corpuscles in these gases are
+ respectively 1/11.5 and 1/29 cm. at a pressure of 1 mm. These results
+ again agree well with the values given by the kinetic theory of gases.
+
+ If the number of positive ions per unit volume is m and v is the
+ velocity, we have nue+mve = i, where i is the current through unit
+ area of the gas. Since nue = i0[epsilon]^nx and i = i0[epsilon]^nl,
+ when l is the distance between the plates, we see that
+
+ nu / mv = [epsilon]^(nx) / ([epsilon]^(nl) - [epsilon]^(nx)),
+
+ n v [epsilon]^(nx)
+ -- = -- . -------------------------------.
+ m u [epsilon]^(ne) - [epsilon]^(nx)
+
+ Since v/u is a very small quantity we see that n will be less than m
+ except when [epsilon]^nl - [epsilon]^nx is small, i.e. except close to
+ the anode. Thus there will be an excess of positive electricity from
+ the cathode almost up to the anode, while close to the anode there
+ will be an excess of negative. This distribution of electricity will
+ make the electric force diminish from the cathode to the place where
+ there is as much positive as negative electricity, where it will have
+ its minimum value, and then increase up to the anode.
+
+ The expression i = i0[epsilon]^[alpha]l applies to the case when there
+ is no source of ionization in the gas other than the collisions; if in
+ addition to this there is a source of uniform ionization producing q
+ ions per cubic centimetre, we can easily show that
+
+ qe
+ i = i0[epsilon]^{[alpha]l} + -------(e^{[alpha]l} - 1).
+ [alpha]
+
+ With regard to the minimum energy which must be possessed by a
+ corpuscle to enable it to produce ions by collision, Townsend (loc.
+ cit.) came to the conclusion that to ionize air the corpuscle must
+ possess an amount of energy equal to that acquired by the fall of its
+ charge through a potential difference of about 2 volts. This is also
+ the value arrived at by H. A. Wilson by entirely different
+ considerations. Stark, however, gives 17 volts as the minimum for
+ ionization. The energy depends upon the nature of the gas; recent
+ experiments by Dawes and Gill and Pedduck (_Phil. Mag._, Aug. 1908)
+ have shown that it is smaller for helium than for air, hydrogen, or
+ carbonic acid gas.
+
+If there is no external source of ionization and no emission of
+corpuscles from the cathode, then it is evident that even if some
+corpuscles happened to be present in the gas when the electric field
+were applied, we could not get a permanent current by the aid of
+collisions made by these corpuscles. For under the electric field, the
+corpuscles would be driven from the cathode to the anode, and in a short
+time all the corpuscles originally present in the gas and those produced
+by them would be driven from the gas against the anode, and if there was
+no source from which fresh corpuscles could be introduced into the gas
+the current would cease. The current, however, could be maintained
+indefinitely if the positive ions in their journey back to the cathode
+also produced ions by collisions, for then we should have a kind of
+regenerative process by which the supply of corpuscles could be
+continually renewed. To maintain the current it is not necessary that
+the ionization resulting from the positive ions should be anything like
+as great as that from the negative, as the investigation given below
+shows a very small amount of ionization by the positive ions will
+suffice to maintain the current. The existence of ionization by
+collision with positive ions has been proved by Townsend. Another method
+by which the current could be and is maintained is by the anode emitting
+corpuscles under the impact of the positive ions driven against it by
+the electric field. J. J. Thomson has shown by direct experiment that
+positively electrified particles when they strike against a metal plate
+cause the metal to emit corpuscles (J. J. Thomson, _Proc. Camb. Phil.
+Soc._ 13, p. 212; Austin, _Phys. Rev._ 22, p. 312). If we assume that
+the number of corpuscles emitted by the plate in one second is
+proportional to the energy in the positive ions which strike the plate
+in that second, we can readily find an expression for the difference of
+potential which will maintain without any external ionization a current
+of electricity through the gas. As this investigation brings into
+prominence many of the most important features of the electric
+discharge, we shall consider it in some detail.
+
+ Let us suppose that the electrodes are parallel plates of metal at
+ right angles to the axis of x, and that at the cathode x = 0 and at the
+ anode x = d, d being thus the distance between the plates. Let us also
+ suppose that the current of electricity flowing between the plates is
+ so small that the electrification between the plates due to the
+ accumulation of ions is not sufficient to disturb appreciably the
+ electric field, which we regard as uniform between the plates, the
+ electric force being equal to V/d, where V is the potential difference
+ between the plates. The number of positive ions produced per second in
+ a layer of gas between the planes x and x+dx is [alpha]nu.dx. Here n is
+ the number of corpuscles per unit volume, [alpha] the coefficient of
+ ionization (for strong electric field [alpha] = 1/[lambda]', where
+ [lambda]' is the mean free path of a corpuscle), and u the velocity of
+ a corpuscle parallel to x. We have seen that nu = i0[epsilon]^[alpha]x,
+ where i0 is the number of corpuscles emitted per second by unit area of
+ the cathode. Thus the number of positive ions produced in the layer is
+ [alpha]i0[epsilon]^[alpha]x dx. If these went straight to the cathode
+ without a collision, each of them would have received an amount of
+ kinetic energy Vex/d when they struck the cathode, and the energy of
+ the group of ions would be Vex/d.[alpha]i0[epsilon]^dx dx. The positive
+ ions will, however, collide with the molecules of the gas through which
+ they are passing, and this will diminish the energy they possess when
+ they reach the cathode.
+
+ The diminution in the energy will increase in geometrical proportion
+ with the length of path travelled by the ion and will thus be
+ proportional to [epsilon]^-[beta]x, [beta] will be proportional to the
+ number of collisions and will thus be proportional to the pressure of
+ the gas. Thus the kinetic energy possessed by the ions when they reach
+ the cathode will be
+
+ [epsilon]^{-[beta]x} . V(ex/d) . [alpha]i0[epsilon]^{[alpha]x} dx,
+
+ and E, the total amount of energy in the positive ions which reach the
+ cathode in unit time, will be given by the equation
+ _
+ /d
+ E = | [epsilon]^{-[beta]x} . V(ex/d) . [alpha]i0[epsilon]^{[alpha]x} dx
+ _/0
+ _
+ Ve[alpha]i0 /d
+ = ----------- | [epsilon]^{-([beta]-[alpha])x}.x.dx
+ d _/0
+
+ Ve[alpha]i0 / 1 / 1 d \ \
+ = ----------- { ------------------ - [epsilon]^{-([beta]-[alpha])d} { ------------------ + ---------------- } } (1).
+ d \([beta]-[alpha])^2 \([beta]-[alpha])^2 ([beta]-[alpha])/ /
+
+
+ If the number of corpuscles emitted by the cathode in unit time is
+ proportional to this energy we have i0 = kE, where k is a constant;
+ hence by equation (1) we have
+
+ ([beta]-[alpha])^2 d
+ V = ------------------ . --,
+ ke[alpha] I
+
+ where
+
+ I = 1 - [epsilon]^{-([beta]-[alpha])d} (1 + d([beta] - [alpha])).
+
+ Since both [beta] and [alpha] are proportional to the pressure, I and
+ ([beta] - [alpha])^2d/[alpha] are both functions of pd, the product of
+ the pressure and the spark length, hence we see that V is expressed by
+ an equation of the form
+
+ 1
+ V = -- [int](pd) (2),
+ ke
+
+ where [int](pd) denotes a function of pd, and neither p nor d enter
+ into the expression for V except in this product. Thus the potential
+ difference required to produce discharge is constant as long as the
+ product of the pressure and spark length remains constant; in other
+ words, the spark potential is constant as long as the mass of the gas
+ between the electrodes is constant. Thus, for example, if we halve the
+ pressure the same potential difference will produce a spark of twice
+ the length. This law, which was discovered by Paschen for fairly long
+ sparks (_Annalen_, 37, p. 79), and has been shown by Carr (_Phil.
+ Trans._, 1903) to hold for short ones, is one of the most important
+ properties of the electric discharge.
+
+ We see from the expression for V that when ([beta] - [alpha])d is very
+ large
+
+ V = ([beta] - [alpha])^2d/ke[alpha].
+
+ Thus V becomes infinite when d is infinite. Again when ([beta] -
+ [alpha])d is very small we find
+
+ V = 1/ke[alpha]d;
+
+ thus V is again infinite when d is nothing. There must therefore be
+ some value of d intermediate between zero and infinity for which V is
+ a minimum. This value is got by finding in the usual way the value of
+ d, which makes the expression for V given in equation (1) a minimum.
+ We find that d must satisfy the equation
+
+ / \
+ 1 = [epsilon]^{-([beta]-[alpha])d} {1 + ([beta] - [alpha])d + ([beta] - [alpha].d)^2}.
+ \ /
+
+ We find by a process of trial and error that ([beta]-[alpha])d = 1.8
+ is approximately a solution of this equation; hence the distance for
+ minimum potential is 1.8/([beta] - [alpha]). Since [beta] and [alpha]
+ are both proportional to the pressure, we see that the critical spark
+ length varies inversely as the pressure. If we substitute this value
+ in the expression for V we find that [=V], the minimum spark
+ potential, is given by
+
+ _ [beta] - [alpha] 2.2
+ V = ---------------- . ---.
+ [alpha] ke
+
+ Since [beta] and [alpha] are each proportional to the pressure, the
+ minimum potential is independent of the pressure of the gas. On this
+ view the minimum potential depends upon the metal of which the cathode
+ is made, since k measures the number of corpuscles emitted per unit
+ time by the cathode when struck by positive ions carrying unit energy,
+ and unless [beta] bears the same ratio to [alpha] for all gases the
+ minimum potential will also vary with the gas. The measurements which
+ have been made of the "cathode fall of potential," which as we shall
+ see is equal to the minimum potential required to produce a spark,
+ show that this quantity varies with the material of which the cathode
+ is made and also with the nature of the gas. Since a metal plate, when
+ bombarded by positive ions, emits corpuscles, the effect we have been
+ considering must play a part in the discharge; it is not, however, the
+ only effect which has to be considered, for as Townsend has shown,
+ positive ions when moving above a certain speed ionize the gas, and
+ cause it to emit corpuscles. It is thus necessary to take into account
+ the ionization of the positive ions.
+
+ Let m be the number of positive ions per unit volume, and w their
+ velocity, the number of collisions which occur in one second in one
+ cubic centimetre of the gas will be proportional to mwp, where p is
+ the pressure of the gas. Let the number of ions which result from
+ these collisions be [gamma]mw; [gamma] will be a function of p and of
+ the strength of the electric field. Let as before n be the number of
+ corpuscles per cubic centimetre, u their velocity, and [alpha]nu the
+ number of ions which result in one second from the collisions between
+ the corpuscles and the gas. The number of ions produced per second per
+ cubic centimetre is equal to [alpha]nu + [gamma]mw; hence when things
+ are in a steady state
+
+ d
+ --(nu) = [alpha]nu + [gamma]mw ,
+ dx
+
+ and
+
+ e(nu + mw) = i,
+
+ where e is the charge on the ion and i the current through the gas.
+ The solution of these equations when the field is uniform between the
+ plates, is
+
+ enu = C[epsilon]^{([alpha]-[gamma])x} - [gamma]i/([alpha] - [gamma]),
+
+ emw = -C[epsilon]^{([alpha]-[gamma])x} + [alpha]i/([alpha] - [gamma]),
+
+ where C is a constant of integration. If there is no emission of
+ positive ions from the anode enu = i, when x = d. Determining C from
+ this condition we find
+
+ i / \
+ enu = ----------------- {[alpha][epsilon]^{([alpha]-[gamma])(x-d)} - [gamma] },
+ [alpha] - [gamma] \ /
+
+ [alpha]i / \
+ emw = ----------------- {1 - [epsilon]^{([alpha]-[gamma])(x-d)} }.
+ [alpha] - [gamma] \ /
+
+ If the cathode did not emit any corpuscles owing to the bombardment by
+ positive ions, the condition that the charge should be maintained is
+ that there should be enough positive ions at the cathode to carry the
+ current i.e. that emw = i; when x = 0, the condition gives
+
+ i / \
+ ----------------- {[alpha][epsilon]^{-([alpha]-[gamma])d} - [gamma] } = 0,
+ [alpha] - [gamma] \ /
+
+ or
+
+ [epsilon]^{[alpha]d}/[alpha] = [epsilon]^{[gamma]d}/[gamma].
+
+ Since [alpha] and [gamma] are both of the form pf(X/p) and X = V/d, we
+ see that V will be a function of pd, in agreement with Paschen's law.
+ If we take into account both the ionization of the gas and the
+ emission of corpuscles by the metal we can easily show that
+
+ _
+ [alpha]-[gamma][epsilon]^{([alpha]-[gamma])d} k[alpha]Ve | 1
+ --------------------------------------------- = ---------- | -------------------------- -
+ [alpha] - [gamma] d |_ ([beta]+[gamma]-[alpha])^2
+ _
+ / 1 d \ |
+ [epsilon]^{-([beta]+[gamma]-[alpha])d} { -------------------------- + ---------------------- } |,
+ \([beta]+[gamma]-[alpha])^2 [beta]+[gamma]-[alpha]/ _|
+
+ where k and [beta] have the same meaning as in the previous
+ investigation. When d is large, [epsilon]^{([alpha]-[gamma])d} is also
+ large; hence in order that the left-hand side of this equation should
+ not be negative [gamma] must be less than [alpha]/[epsilon]^
+ {([alpha]-[gamma])d}; as this diminishes as d increases we see that when
+ the sparks are very long discharge will take place, practically as soon
+ as [gamma] has a finite value, i.e. as soon as the positive ions begin
+ to produce fresh ions by their collisions.
+
+In the preceding investigation we have supposed that the electric field
+between the plates was uniform; if it were not uniform we could get
+discharges produced by very much smaller differences of potential than
+are necessary in a uniform field. For to maintain the discharge it is
+not necessary that the positive ions should act as ionizers all along
+their path; it is sufficient that they should do so in the neighbourhood
+of cathode. Thus if we have a strong field close to the cathode we might
+still get the discharge though the rest of the field were comparatively
+weak. Such a distribution of electric force requires, however, a great
+accumulation of charged ions near the cathode; until these ions
+accumulate the field will be uniform. If the uniform field existing in
+the gas before the discharge begins were strong enough to make the
+corpuscles produce ions by collision, but not strong enough to make the
+positive ions act as ionizers, there would be some accumulation of ions,
+and the amount of this accumulation would depend upon the number of free
+corpuscles originally present in the gas, and upon the strength of the
+electric field. If the accumulation were sufficient to make the field
+near the cathode so strong that the positive ions could produce fresh
+ions either by collision with the cathode or with the gas, the discharge
+would pass though the gas; if not, there will be no continuous
+discharge. As the amount of the accumulation depends on the number of
+corpuscles present in the gas, we can understand how it is that after a
+spark has passed, leaving for a time a supply of corpuscles behind it,
+it is easier to get a discharge to pass through the gas than it was
+before.
+
+[Illustration: Fig. 15.]
+
+The inequality of the electric field in the gas when a continuous
+discharge is passing through it is very obvious when the pressure of the
+gas is low. In this case the discharge presents a highly differentiated
+appearance of which a type is represented in fig. 15. Starting from the
+cathode we have a thin velvety luminous glow in contact with the
+surface; this glow is often called the "first cathode layer." Next this
+we have a comparatively dark space whose thickness increases as the
+pressure diminishes; this is called the "Crookes's dark space," or the
+"second cathode layer." Next this we have a luminous position called the
+"negative glow" or the "third cathode layer." The boundary between the
+second and third layers is often very sharply defined. Next to the third
+layer we have another dark space called the "Faraday dark space." Next
+to this and reaching up to the anode is another region of luminosity,
+called the "positive column," sometimes (as in fig. 15, a) continuous,
+sometimes (as in fig. 15, b) broken up into light or dark patches called
+"striations." The dimensions of the Faraday dark space and the positive
+column vary greatly with the current passing through the gas and with
+its pressure; sometimes one or other of them is absent. These
+differences in appearances are accompanied by great difference in the
+strength of the electric field. The magnitude of the electric force at
+different parts of the discharge is represented in fig. 16, where the
+ordinates represent the electric force at different parts of the tube,
+the cathode being on the right. We see that the electric force is very
+large indeed between the negative glow and the cathode, much larger than
+in any other part of the tube. It is not constant in this region, but
+increases as we approach the cathode. The force reaches a minimum either
+in the negative glow itself or in the part of the Faraday dark space
+just outside, after which it increases towards the positive column. In
+the case of a uniform positive column the electric force along it is
+constant until we get quite close to the anode, when a sudden change,
+called the "anode fall," takes place in the potential.
+
+[Illustration:
+ _Discharge in Hydrogen
+ Pressure 2.25 m.m. Current 0.568.10^-3 ampere_
+FIG. 16.]
+
+The difference of potential between the cathode and the negative glow is
+called the "cathode potential fall" and is found to be constant for wide
+variations in the pressure of the gas and the current passing through.
+It increases, however, considerably when the current through the gas
+exceeds a certain critical value, depending among other things on the
+size of the cathode. This cathode fall of potential is shown by
+experiment to be very approximately equal to the minimum potential
+difference. The following table contains a comparison of the
+measurements of the cathode fall of potentials in various gases made by
+Warburg (_Wied. Ann._, 1887, 31, p. 545, and 1890, 40, p. 1), Capstick
+(_Proc. Roy. Society_, 1898, 63, p. 356), and Strutt (_Phil. Trans._,
+1900, 193, p. 377), and the measurements by Strutt of the smallest
+difference of potential which will maintain a spark through these gases.
+
+ +---------+-----------------------------------------+-----------------+
+ | | Cathode fall in Volts. |Least potential |
+ | Gas. +-----------------------------+-----------+ difference |
+ | | Platinum Electrodes. |Aluminium | required to |
+ | | |Electrodes.|maintain a Spark.|
+ +---------+-----------+---------+-------+-----------+-----------------+
+ | | Warburg. |Capstick.|Strutt.| Warburg. | Strutt. |
+ +---------+-----------+---------+-------+-----------+-----------------+
+ |Air | 340-350 | .. | .. | .. | 341 |
+ |H2 | about 300 | 298 | .. | 168 | 302-308 |
+ |O2 | .. | 369 | .. | .. | .. |
+ |N2 |230 if free| 232 | .. | 207 | 251 |
+ | |from oxygen| | | | |
+ |Hg vapour| 340 | .. | .. | .. | .. |
+ |Helium | .. | .. | 226 | .. | 261-326 |
+ |H2O | .. | 469 | .. | .. | .. |
+ |NH3 | .. | 582 | .. | .. | .. |
+ +---------+-----------+---------+-------+-----------+-----------------+
+
+Thus in the cases in which the measurements could be made with the
+greatest accuracy the agreement between the cathode fall and the minimum
+potential difference is very close. The cathode fall depends on the
+material of which the terminals are made, as is shown by the following
+table due to Mey (_Verh. deutsch. physik. Gesell._, 1903, 5, p. 72).
+
+ +------+---------------------------------------------+
+ | Gas. | Electrode. |
+ +------+---+---+---+---+---+---+---+---+---+-----+---+
+ | | Pt| Hg| Ag| Cu| Fe| Zn| Al| Mg| Na| Na-K| K |
+ +------+---+---+---+---+---+---+---+---+---+-----+---+
+ |O2 |369| ..| ..| ..| ..| ..| ..| ..| ..| .. | ..|
+ |H2 |300| ..|295|280|230|213|190|168|185|169 |172|
+ |N2 |232|226| ..| ..| ..| ..| ..|207|178|125 |170|
+ |He |226| ..| ..| ..| ..| ..| ..| ..| 80| 78.5| 69|
+ |Argon |167| ..| ..| ..| ..| ..|100| ..| ..| .. | ..|
+ +------+---+---+---+---+---+---+---+---+---+-----+---+
+
+The dependence of the minimum potential required to produce a spark upon
+the metal of which the cathode is made has not been clearly established,
+some observers being unable to detect any difference between the
+potential required to spark between electrodes of aluminium and those of
+brass, while others thought they had detected such a difference. It is
+only with sparks not much longer than the critical spark length that we
+could hope to detect this difference. When the current through the gas
+exceeds a certain critical value depending among other things on the
+size of the cathode, the cathode fall of potential increases rapidly and
+at the same time the thickness of the dark spaces diminishes. We may
+regard the part of the discharge between the cathode and the negative
+glow as a discharge taking place under minimum potential difference
+through a distance equal to the critical spark length. An inspection of
+fig. 16 will show that we cannot regard the electric field as constant
+even for this small distance; it thus becomes a matter of interest to
+know what would be the effect on the minimum potential difference
+required to produce a spark if there were sufficient ions present to
+produce variations in the electric field analogous to those represented
+in fig. 16. If the electric force at a distance x from the cathode were
+proportional to [epsilon]^-px we should have a state of things much
+resembling the distribution of electric force near the cathode. If we
+apply to this distribution the methods used above for the case when the
+force was uniform, we shall find that the minimum potential is less and
+the critical spark length greater than when the electric force is
+uniform.
+
+_Potential Difference required to produce a Spark of given Length._--We
+may regard the region between the cathode and the negative glow as a
+place for the production of corpuscles, these corpuscles finding their
+way from this region through the negative glow. The parts of this glow
+towards the anode we may regard as a cathode, from which, as from a hot
+lime cathode, corpuscles are emitted. Let us now consider what will
+happen to these corpuscles shot out from the negative glow with a
+velocity depending on the cathode fall of potential and independent of
+the pressure. These corpuscles will collide with the molecules of the
+gas, and unless there is an external electric field to maintain their
+velocity they will soon come to rest and accumulate in front of the
+negative glow. The electric force exerted by this cloud of corpuscles
+will diminish the strength of the electric field in the region between
+the cathode and the negative glow, and thus tend to stop the discharge.
+To keep up the discharge we must have a sufficiently strong electric
+field between the negative glow and the anode to remove the corpuscles
+from this region as fast as they are sent into it from the cathode. If,
+however, there is no production of ions in the region between the
+negative glow and the anode, all the ions in this region will have come
+from near the cathode and will be negatively charged; this negative
+electrification will diminish the electric force on the cathode side of
+it and thus tend to stop the discharge. This back electric field could,
+however, be prevented by a little ionization in the region between the
+anode and glow, for this would afford a supply of positive ions, and
+thus afford an opportunity for the gas in this region to have in it as
+many positive as negative ions; in this case it would not give rise to
+any back electromotive force. The ionization which produces these
+positive ions may, if the field is intense, be due to the collisions of
+corpuscles, or it may be due to radiation analogous to ultra-violet, or
+soft Rontgen rays, which have been shown by experiment to accompany the
+discharge. Thus in the most simple conditions for discharge we should
+have sufficient ionization to keep up the supply of positive ions, and
+an electric field strong enough to keep the velocity of the negative
+corpuscle equal to the value it has when it emerges from the negative
+glow. Thus the force must be such as to give a constant velocity to the
+corpuscle, and since the force required to move an ion with a given
+velocity is proportional to the pressure, this force will be
+proportional to the pressure of the gas. Let us call this force ap; then
+if l is the distance of the anode from the negative glow the potential
+difference between these points will be alp. The potential difference
+between the negative glow and the cathode is constant and equals c;
+hence if V is the potential difference between the anode and cathode,
+then V = c + alp, a relation which expresses the connexion between the
+potential difference and spark length for spark lengths greater than the
+critical distance. It is to be remembered that the result we have
+obtained applies only to such a case as that indicated above, where the
+electric force is constant along the positive column. Experiments with
+the discharge through gases at low pressure show the discharge may take
+other forms. Thus the positive column may be striated when the force
+along it is no longer uniform, or the positive column may be absent;
+the discharge may be changed from one of these forms to another by
+altering the current. The relation between the potential and the
+distance between the electrodes varies greatly, as we might expect, with
+the current passing through the gas.
+
+The connexion between the potential difference and the spark length has
+been made the subject of a large number of experiments. The first
+measurements were made by Lord Kelvin in 1860 (_Collected Papers on
+Electrostatics and Magnetism_, p. 247); subsequent experiments have been
+made by Baille (_Ann. de chimie et de physique_, 5, 25, p. 486), Liebig
+(_Phil. Mag._ [5], 24, p. 106), Paschen (_Wied. Ann._ 37, p. 79), Peace
+(_Proc. Roy. Soc._, 1892, 52, p. 99), Orgler (_Ann. der Phys._ 1, p.
+159), Strutt (_Phil. Trans._ 193, p. 377), Bouty (_Comptes rendus_, 131,
+pp. 469, 503), Earhart (_Phil. Mag._ [6], 1, p. 147), Carr (_Phil.
+Trans._, 1903), Russell (_Phil. Mag._ [5], 64, p. 237), Hobbs (_Phil.
+Mag._ [6], 10, p. 617), Kinsley (_Phil. Mag._ [6], 9, 692), Ritter
+(_Ann. der Phys._ 14, p. 118). The results of their experiments show
+that for sparks considerably longer than the critical spark length, the
+relation between the potential difference V and the spark length l may
+be expressed when the electrodes are large with great accuracy by the
+linear relation V = c + blp, where p is the pressure and c and b are
+constants depending on the nature of the gas. When the sparks are long
+the term blp is the most important and the sparking potential is
+proportional to the spark length. Though there are considerable
+discrepancies between the results obtained by different observers, these
+indicate that the production of a long spark between large electrodes in
+air at atmospheric pressure requires a potential difference of 30,000
+volts for each centimetre of spark length. In hydrogen only about half
+this potential difference is required, in carbonic acid gas the
+potential difference is about the same as in air, while Ritter's
+experiments show that in helium only about one-tenth of this potential
+difference is required.
+
+In the case when the electric field is not uniform, as for example when
+the discharge takes place between spherical electrodes, Russell's
+experiments show that the discharge takes place as soon as the maximum
+electric force in the field between the electrodes reaches a definite
+value, which he found was for air at atmospheric pressure about 38,000
+volts per centimetre.
+
+_Very Short Sparks._--Some very interesting experiments on the potential
+difference required to produce exceedingly short sparks have been made
+by Earhart, Hobbs and Kinsley; the length of these sparks was comparable
+with the wave length of sodium light. With sparks of these lengths it
+was found that it was possible to get a discharge with less than 330
+volts, the minimum potential difference in air. The results of these
+observers show that there is no diminution in the minimum potential
+difference required to produce discharge until the spark length gets so
+small that the average electric force between the electrodes amounts to
+about one million volts per centimetre. When the force rises to this
+value a discharge takes place even though the potential difference is
+much less than 330 volts; in some of Earhart's experiments it was only
+about 2 volts. This kind of discharge is determined not by the condition
+that the potential difference should have a given value, but that the
+electric force should have a given value. Another point in which this
+discharge differs from the ordinary one is that it is influenced
+entirely by the nature of the electrodes and not by the nature or
+pressure of the gas between them, whereas the ordinary discharge is in
+many cases not affected appreciably by changes in the metal of the
+electrodes, but is always affected by changes in the pressure and
+character of the gas between them. Kinsley found that when one of these
+small sparks passed between the electrodes a kind of metallic bridge was
+formed between them, so that they were in metallic connexion, and that
+the distance between them had to be considerably increased before the
+bridge was broken. Almy (_Phil. Mag._, Sept. 1908), who used very small
+electrodes, was unable to get a discharge with less than the minimum
+spark potential even when the spark length was reduced to one-third of
+the wave length of sodium light. He suggests that the discharges
+obtained with larger electrodes for smaller voltages are due to the
+electrodes being dragged together by the electrostatic attraction
+between them.
+
+_Constitution of the Electric Spark._--Schuster and Hemsalech (_Phil.
+Trans._ 193, p. 189), Hemsalech (_Comptes Rendus_, 130, p. 898; 132, p.
+917; _Jour. de Phys._ 3. 9, p. 43, and Schenck, _Astrophy. Jour._ 14, p.
+116) have by spectroscopic methods obtained very interesting results
+about the constitution of the spark. The method employed by Schuster and
+Hemsalech was as follows: Suppose we photograph the spectrum of a
+horizontal spark on a film which is on the rim of a wheel rotating about
+a horizontal axis with great velocity. If the luminosity travelled with
+infinite speed from one electrode to the other, the image on the film
+would be a horizontal line. If, however, the speed with which the
+luminosity travelled between the electrodes was comparable with the
+speed of the film, the line would be inclined to the horizontal, and by
+measuring the inclinations we could find the speed at which the
+luminosity travelled. In this way Schuster and Hemsalech showed that
+when an oscillating discharge passed between metallic terminals in air,
+the first spark passes through the air alone, no lines of the metal
+appearing in its spectrum. This first spark vaporizes some of the metal
+and the subsequent sparks passing mainly through the metallic vapour;
+the appearance of the lines in the film shows that the velocity of the
+luminous part of the vapour was finite. The velocity of the vapour of
+metals of low atomic weight was in general greater than that of the
+vapour of heavier metals. Thus the velocity of aluminium vapour was 1890
+metres per second, that of zinc and cadmium only about 545. Perhaps the
+most interesting point in the investigation was the discovery that the
+velocities corresponding to different lines in the spectrum of the same
+metal were in some cases different. Thus with bismuth some of the lines
+indicated a velocity of 1420 metres per second, others a velocity of
+only 550, while one ([lambda] = 3793) showed a still smaller velocity.
+These results are in accordance with a view suggested by other phenomena
+that many of the lines in a spectrum produced by an electrical discharge
+originate from systems formed during the discharge and not from the
+normal atom or molecule. Schuster and Hemsalech found that by inserting
+a coil with large self induction in the primary circuit they could
+obliterate the air lines in the discharge.
+
+Schenck, by observing the appearance presented when an alternating
+current, produced by discharging Leyden jars, was examined in a rapidly
+rotating mirror, found it showed the following stages: (1) a thin bright
+line, followed in some cases at intervals of half the period of the
+discharge by fainter lines; (2) bright curved streamers starting from
+the negative terminal, and diminishing rapidly in speed as they receded
+from the cathode; (3) a diffused glow lasting for a much longer period
+than either of the preceding. These constituents gave out quite
+different spectra.
+
+The structure of the discharge is much more easily studied when the
+pressure of the gas is low, as the various parts which make up the
+discharge are more widely separated from each other. We have already
+described the general appearance of the discharge through gases at low
+pressures (see p. 657). There is, however, one form of discharge which
+is so striking and beautiful that it deserves more detailed
+consideration. In this type of discharge, known as the striated
+discharge, the positive column is made up of alternate bright and dark
+patches known as _striations_. Some of these are represented in fig. 17,
+which is taken from a paper by De la Rue and Muller (_Phil. Trans._,
+1878, Pt. 1). This type of discharge only occurs when the current and
+the pressure of the gas are between certain limits. It is most
+beautifully shown when a Wehnelt cathode is used and the current is
+produced by storage cells, as this allows us to use large currents and
+to maintain a steady potential difference between the electrodes. The
+striations are in consequence very bright and steady. The facts which
+have been established about these striations are as follows: The
+distance between the bright parts of the striations is greater at low
+pressures than at high; it depends also upon the diameter of the tube,
+increasing as the diameter of the tube increases. If the discharge tube
+is wide at one place and narrow in another the striations will be
+closer together in the narrow parts than in the wide. The distance
+between the striations depends on the current through the tube. The
+relation is not a very simple one, as an increase of current sometimes
+increases while under other circumstances it decreases the distance
+between the striations (see Willows, _Proc. Camb. Phil. Soc._ 10, p.
+302). The electric force is not uniform along the striated discharge,
+but is greater in the bright than in the dark parts of the striation. An
+example is shown in fig. 16, due to H. A. Wilson, which shows the
+distribution of electric force at every place in a striated discharge.
+In experiments made by J. J. Thomson (_Phil. Mag._, Oct. 1909), using a
+Wehnelt cathode, the variations in the electric force were more
+pronounced than those shown in fig. 16. The electric force in this case
+changed so greatly that it actually became negative just on the cathode
+side of the bright part of the striation. Just inside the striation on
+the anode side it rose to a very high value, then continually diminished
+towards the bright side of the next striation when it again increased.
+This distribution of electric force implies that there is great excess
+of negative electricity at the bright head of the striation, and a small
+excess of positive everywhere else. The temperature of the gas is higher
+in the bright than in the dark parts of the striations. Wood (_Wied.
+Ann._ 49, p. 238), who has made a very careful study of the distribution
+of temperature in a discharge tube, finds that in those tubes the
+temperature varies in the same way as the electric force, but that this
+temperature (which it must be remembered is the average temperature of
+all the molecules and not merely of those which are taking part in the
+discharge) is by no means high; in no part of the discharge did the
+temperature in his experiments exceed 100 deg. C.
+
+[Illustration: FIG. 17.]
+
+_Theory of the Striations._--We may regard the heaping up of the
+negative charges at intervals along the discharge as the fundamental
+feature in the striations, and this heaping up may be explained as
+follows. Imagine a corpuscle projected with considerable velocity from a
+place where the electric field is strong, such as the neighbourhood of
+the cathode; as it moves towards the anode through the gas it will
+collide with the molecules, ionize them and lose energy and velocity.
+Thus unless the corpuscle is acted on by a field strong enough to supply
+it with the energy it loses by collision, its speed will gradually
+diminish. Further, when its energy falls below a certain value it will
+unite with a molecule and become part of a negative ion, instead of a
+corpuscle; at this stage there will be a sudden and very large
+diminution in its velocity. Let us now follow the course of a stream of
+corpuscles starting from the cathode and approaching the anode. If the
+speed falls off as the stream proceeds, the corpuscles in the rear will
+gain on those in front and the density of the stream in the front will
+be increased. If at a certain place the velocity receives a sudden check
+by the corpuscles becoming loaded with a molecule, the density of the
+negative electricity will increase at this place with great rapidity,
+and here there will be a great accumulation of negative electricity, as
+at the bright head on the cathode side of a striation. Now this
+accumulation of negative electricity will produce a large electric force
+on the anode side; this will drive corpuscles forward with great
+velocity and ionize the gas. These corpuscles will behave like those
+shot from the cathode and will accumulate again at some distance from
+their origin, forming the bright head of the next striation, when the
+process will be repeated. On this view the bright heads of the
+striations act like electrodes, and the discharge passes from one bright
+head to the next as by a number of stepping stones, and not directly
+from cathode to anode. The luminosity at the head of the striations is
+due to the recombination of the ions. These ions have acquired
+considerable energy from the electric field, and this energy will be
+available for supplying the energy radiated away as light. The
+recombination of ions which do not possess considerable amounts of
+energy does not seem to give rise to luminosity. Thus, in an ionized gas
+not exposed to an electric field, although we have recombination between
+the ions, we need not have luminosity. We have at present no exact data
+as to the amount of energy which must be given to an ion to make it
+luminous on recombination; it also certainly varies with the nature of
+the ion; thus even with hot Wehnelt cathodes J. J. Thomson has never
+been able to make the discharge through air luminous with a potential
+less than from 16 to 17 volts. The mercury lamps, however, in which the
+discharge passes through mercury vapour are luminous with a potential
+difference of about 12 volts. It follows that if the preceding theory be
+right the potential difference between two bright striations must be
+great enough to make the corpuscles ionize by collision and also to give
+enough energy to the ions to make them luminous when they recombine. The
+difference of potential between the bright parts of successive
+striations has been measured by Hohn (_Phys. Zeit._ 9, p. 558); it
+varies with the pressure and with the gas. The smallest value given by
+Hohn is about 15 volts. In some experiments made by J. J. Thomson, when
+the pressure of the gas was very low, the difference of potential
+between two adjacent dark spaces was as low as 3.75 volts.
+
+_The Arc Discharge._--The discharges we have hitherto considered have
+been characterized by large potential differences and small currents. In
+the arc discharge we get very large currents with comparatively small
+potential differences. We may get the arc discharge by taking a battery
+of cells large enough to give a potential difference of 60 to 80 volts,
+and connecting the cells with two carbon terminals, which are put in
+contact, so that a current of electricity flows round the circuit. If
+the terminals, while the current is on, are drawn apart, a bright
+discharge, which may carry a current of many amperes, passes from one to
+the other. This arc discharge, as it is called, is characterized by
+intense heat and by the brilliant luminosity of the terminals. This
+makes it a powerful source of light. The temperature of the positive
+terminal is much higher than that of the negative. According to Violle
+(_Comptes Rendus_, 115, p. 1273) the temperature of the tip of the
+former is about 3500 deg. C, and that of the latter 2700 deg. C. The
+temperature of the arc itself he found to be higher than that of either
+of its terminals. As the arc passes, the positive terminal gets hollowed
+out into a crater-like shape, but the negative terminal remains pointed.
+Both terminals lose weight.
+
+ The appearance of the terminals is shown in fig. 18, given by Mrs
+ Ayrton (_Proc. Inst. Elec. Eng._ 28, p. 400); a, b represent the
+ terminals when the arc is quiet, and c when it is accompanied by a
+ hissing sound. The intrinsic brightness of the positive crater does
+ not increase with an increase in the current; an increased current
+ produces an increase in the area of the luminous crater, but the
+ amount of light given out by each unit of area of luminous surface is
+ unaltered. This indicates that the temperature of the crater is
+ constant; it is probably that at which carbon volatilizes. W. E.
+ Wilson (_Proc. Roy. Soc._ 58, p. 174; 60, p. 377) has shown that at
+ pressures of several atmospheres the intrinsic brightness of the
+ crater is considerably diminished.
+
+ [Illustration: FIG. 18.]
+
+ [Illustration: FIG. 19.]
+
+ The connexion between V, the potential difference between the
+ terminals, and l, the length of the arc, is somewhat analogous to that
+ which holds for the spark discharge. Frohlich (_Electrotech. Zeit._ 4,
+ p. 150) gives for this connexion the relation V = m + nl, where m and
+ n are constants. Mrs Ayrton (_The Electric Arc_, chap. iv.) finds that
+ both m and n depend upon the current passing between the terminals,
+ and gives as the relation between V and l, V = [alpha] + [beta]/I +
+ ([gamma] + [delta]/I)l, where [alpha], [beta], [gamma], [delta] are
+ constants and I the current. The relation between current and
+ potential difference was made the subject of a series of experiments
+ by Ayrton (_Electrician_, 1, p. 319; xi. p. 418), some of whose
+ results are represented in fig. 19. For a quiet arc an increase in
+ current is accompanied by a fall in potential difference, while for
+ the hissing arc the potential difference is independent of the
+ current. The quantities m and n which occur in Frohlich's equation
+ have been determined by several experimenters. For carbon electrodes
+ in air at atmospheric pressure m is about 39 volts, varying somewhat
+ with the size and purity of the carbons; it is diminished by soaking
+ the terminals in salt solution. The value of n given by different
+ observers varies considerably, ranging from .76 to 2 volts when l is
+ measured in millimetres; it depends upon the current, diminishing as
+ the current increases. When metallic terminals are used instead of
+ carbons, the value of m depends upon the nature of the metal, m in
+ general being larger the higher the temperature at which the metal
+ volatilizes. Thus v. Lang (_Wied. Ann._ 31, p. 384) found the
+ following values for m in air at atmospheric pressure:--C = 35; Pt =
+ 27.4; Fe = 25; Ni = 26.18; Cu = 23.86; Ag = 15.23; Zn = 19.86; Cd =
+ 10.28. Lecher (_Wied. Ann._ 33, p. 609) gives Pt = 28, Fe = 20, Ag =
+ 8, while Arons (_Wied. Ann._ 31, p. 384) found for Hg the value 12.8;
+ in this case the fall of potential along the arc itself was abnormally
+ small. In comparing these values it is important to remember that
+ Lecher (loc. cit.) has shown that with Fe or Pt terminals the arc
+ discharge is intermittent. Arons has shown that this is also the case
+ with Hg terminals, but no intermittence has been detected with
+ terminals of C, Ag or Cu. The preceding measurements refer to mean
+ potentials, and no conclusions as to the actual potential differences
+ at any time can be drawn when the discharge is discontinuous, unless
+ we know the law of discontinuity. The ease with which an arc is
+ sustained depends greatly on the nature of the electrodes; when they
+ are brass, zinc, cadmium, or magnesium it is exceedingly difficult to
+ get the arc.
+
+ [Illustration: FIG. 20.]
+
+ [Illustration: FIG. 21.]
+
+ The potential difference between the terminals is affected by the
+ pressure of the gas. The most extensive series of experiments on this
+ point is that made by Duncan, Rowland, and Tod (_Electrician_, 31, p.
+ 60), whose results are represented in fig. 20. We see from these
+ curves that for very short arcs the potential difference increases
+ continuously with the pressure, but for longer ones there is a
+ critical pressure at which the potential difference is a minimum, and
+ that this critical pressure seems to increase with the length of arc.
+ The nature of the gas also affects the potential difference. The
+ magnitude of this effect may be gathered from the following values
+ given by Arons (_Ann. der Phys._ 1, p. 700) for the potential
+ difference required to produce an arc 1.5 mm. long, carrying a current
+ of 4.5 amperes, between terminals of different metals in air and pure
+ nitrogen.
+
+ +-----------+------+-----------+
+ | Terminal. | Air. | Nitrogen. |
+ +-----------+------+-----------+
+ | Ag | 21 | ? |
+ | Zn | 23 | 21 |
+ | Cd | 25 | 21 |
+ | Cu | 27 | 30 |
+ | Fe | 29 | 20 |
+ | Pt | 36 | 30 |
+ | Al | 39 | 27 |
+ | Pb | .. | 18 |
+ | Mg | .. | 22 |
+ +-----------+------+-----------+
+
+ Thus, with the discharge for an arc of given length and current, the
+ nature of the terminals is the most important factor in determining
+ the potential difference. The effects produced by the pressure and
+ nature of the surrounding gas, although quite appreciable, are not of
+ so much importance, while in the spark discharge the nature of the
+ terminals is of no importance, everything depending upon the nature
+ and pressure of the gas.
+
+ The potential gradient in the arc is very far from being uniform. With
+ carbon terminals Luggin (_Wien. Ber._ 98, p. 1192) found that, with a
+ current of 15 amperes, there was a fall of potential of 33.7 close to
+ the anode, and one 8.7 close to the cathode, so that the curve
+ representing the distribution of potential between the terminals would
+ be somewhat like that shown in fig. 21. We have seen that a somewhat
+ analogous distribution of potential holds in the case of conduction
+ through flames, though in that case the greatest drop of potential is
+ in general at the cathode and not at the anode. The difference between
+ the changes of potential at the anode and cathode is not so large with
+ Fe and Cu terminals as with carbon ones; with mercury terminals, Arons
+ (_Wied. Ann._ 58, p. 73) found the anode fall to be 7.4 volts, the
+ cathode fall 5.4 volts.
+
+The case of the arc when the cathode is a pool of mercury and the anode
+a metal wire placed in a vessel from which the air has been exhausted is
+one which has attracted much attention, and important investigations on
+this point have been made by Hewitt (_Electrician_, 52, p. 447), Wills
+(_Electrician_, 54, p. 26), Stark, Retschinsky and Schnaposnikoff (_Ann.
+der Phys._ 18, p. 213) and Pollak (_Ann. der Phys._ 19, p. 217). In this
+arrangement the mercury is vaporized by the heat, and the discharge
+which passes through the mercury vapour gives an exceedingly bright
+light, which has been largely used for lighting factories, &c. The
+arrangement can also be used as a rectifier, for a current will only
+pass through it when the mercury pool is the cathode. Thus if such a
+lamp is connected with an alternating current circuit, it lets through
+the current in one direction and stops that in the other, thus
+furnishing a current which is always in one direction.
+
+_Theory of the Arc Discharge._--An incandescent body such as a piece of
+carbon even when at a temperature far below that of the terminals in an
+arc, emits corpuscles at a rate corresponding to a current of the order
+of 1 ampere per square centimetre of incandescent surface, and as the
+rate of increase of emission with the temperature is very rapid, it is
+probably at the rate of many amperes per square centimetre at the
+temperature of the negative carbon in the arc. If then a piece of carbon
+were maintained at this temperature by some external means, and used as
+a cathode, a current could be sent from it to another electrode whether
+the second electrode were cold or hot. If, however, these negatively
+electrified corpuscles did not produce other ions either by collision
+with the gas through which they move or with the anode, the spaces
+between cathode and anode would have a negative charge, which would tend
+to stop the corpuscles leaving the cathode and would require a large
+potential difference between anode and cathode to produce any
+considerable current. If, however, there is ionization either in the gas
+or at the anode, the positive ions will diffuse into the region of the
+discharge until they are sensibly equal in number to the negative ions.
+When this is the case the back electromotive force is destroyed and the
+same potential difference will carry a much larger current. The arc
+discharge may be regarded as analogous to the discharge between
+incandescent terminals, the only difference being that in the arc the
+terminals are maintained in the state of incandescence by the current
+and not by external means. On this view the cathode is bombarded by
+positive ions which heat it to such a temperature that negative
+corpuscles sufficient to carry the current are emitted by it. These
+corpuscles bombard the anode and keep it incandescent. They ionize also,
+either directly by collision or indirectly by heating the anode, the gas
+and vapour of the metal of which the anode is made, and produce in this
+way the supply of positive ions which keep the cathode hot.
+
+_Discharge from a Point._--A very interesting case of electric discharge
+is that between a sharply pointed electrode, such as a needle, and a
+metal surface of considerable area. At atmospheric pressures the
+luminosity is confined to the immediate neighbourhood of the point. If
+the sign of the potential of the point does not change, the discharge is
+carried by ions of one sign--that of the charge on the pointed
+electrode. The velocity of these ions under a given potential gradient
+has been measured by Chattock (_Phil. Mag._ 32, p. 285), and found to
+agree with that of the ions produced by Rontgen or uranium radiation,
+while Townsend (_Phil. Trans._ 195, p. 259) has shown that the charge on
+these ions is the same as that on the ions streaming from the point. If
+the pointed electrode be placed at right angles to a metal plane serving
+as the other electrode, the discharge takes place when, for a given
+distance of the point from the plane, the potential difference between
+the electrodes exceeds a definite value depending upon the pressure and
+nature of the gas through which the discharge passes; its value also
+depends upon whether, beginning with a small potential difference, we
+gradually increase it until discharge commences, or, beginning with a
+large potential difference, we decrease it until the discharge stops.
+The value found by the latter method is less than that by the former.
+According to Chattock's measurements the potential difference V for
+discharge between the point and the plate is given by the linear
+relation V = a + bl, where l is the distance of the point from the plate
+and a and b are constants. From v. Obermayer's (_Wien. Ber._ 100, 2, p.
+127) experiments, in which the distance l was greater than in
+Chattock's, it would seem that the potential for larger distances does
+not increase quite so rapidly with l as is indicated by Chattock's
+relation. The potential required to produce this discharge is much less
+than that required to produce a spark of length l between parallel
+plates; thus from Chattock's experiments to produce the point discharge
+when l = .5 cm. in air at atmospheric pressure requires a potential
+difference of about 3800 volts when the pointed electrode is positive,
+while to produce a spark at the same distance between plane electrodes
+would require a potential difference of about 15,000 volts. Chattock
+showed that with the same pointed electrode the value of the electric
+intensity at the point was the same whatever the distance of the point
+from the plane. The value of the electric intensity depended upon the
+sharpness of the point. When the end of the pointed electrode is a
+hemisphere of radius a, Chattock showed that for the same gas at the
+same pressure the electric intensity f when discharge takes place is
+roughly proportioned to a^-0.8. The value of the electric intensity at
+the pointed electrode is much greater than its value at a plane
+electrode for long sparks; but we must remember that at a distance from
+a pointed electrode equal to a small multiple of the radius of curvature
+of its extremity the electric intensity falls very far below that
+required to produce discharge in a uniform field, so that the discharge
+from a pointed electrode ought to be compared with a spark whose length
+is comparable with the radius of curvature of the point. For such short
+sparks the electric intensity is very high. The electric intensity
+required to produce the discharge from a gas diminishes as the pressure
+of the gas diminishes, but not nearly so rapidly as the electric
+intensity for long sparks. Here again the discharge from a point is
+comparable with short sparks, which, as we have seen, are much less
+sensitive to pressure changes than longer ones. The minimum potential at
+which the electricity streams from the point does not depend upon the
+material of which the point is made; it varies, however, considerably
+with the nature of the gas. The following are the results of some
+experiments on this point. Those in the first two columns are due to
+Rontgen, those in the third and fourth to Precht:--
+
+ +------+-----------------------------+--------------------+
+ | |Discharge Potential. Point +.| Pressure 760. |
+ | Gas. +--------------+--------------+----------+---------+
+ | | Pressure 205.| Pressure 110.| Point +. | Point -.|
+ +------+--------------+--------------+----------+---------+
+ | | Volts. | Volts. | Volts. | Volts. |
+ | H2 | 1296 | 1174 | 2125 | 1550 |
+ | O2 | 2402 | 1975 | 2800 | 2350 |
+ | CO | 2634 | 2100 | .. | .. |
+ | CH4 | 2777 | 2317 | .. | .. |
+ | NO | 3188 | 2543 | .. | .. |
+ | CO2 | 3287 | 2655 | 3475 | 2100 |
+ | N2 | .. | .. | 2600 | 2000 |
+ | Air | .. | .. | 2750 | 2050 |
+ +------+--------------+--------------+----------+---------+
+
+We see from this table that in the case of the discharge from a
+positively electrified point the greater the molecular weight of the gas
+the greater the potential required for discharge. Rontgen concluded from
+his experiments that the discharging potential from a positive point in
+different gases at the same pressure varies inversely as the mean free
+path of the molecules of the gas. In the same gas, however, at different
+pressures the discharging potential does not vary so quickly with the
+pressure as does the mean free path. In Precht's experiments, in which
+different gases were used, the variations in the discharging potential
+are not so great as the variations in the mean free path of the gases.
+
+The current of electrified air flowing from the point when the
+electricity is escaping--the well-known "electrical wind"--is
+accompanied by a reaction on the point which tends to drive it
+backwards. This reaction has been measured by Arrhenius (_Wied. Ann._
+63, p. 305), who finds that when positive electricity is escaping from a
+point in air the reaction on the point for a given current varies
+inversely as the pressure of the gas, and for different gases (air,
+hydrogen and carbonic acid) inversely as the square root of the
+molecular weight of the gas. The reaction when negative electricity is
+escaping is much less. The proportion between the reactions for positive
+and negative currents depends on the pressure of the gas. Thus for equal
+positive and negative currents in air at a pressure of 70 cm. the
+reaction for a positive point was 1.9 times that of a negative one, at
+40 cm. pressure 2.6 times, at 20 cm. pressure 3.2 times, at 10.3 cm.
+pressure 7 times, and at 5.1 cm. pressure 15 times the reaction for the
+negative point. Investigation shows that the reaction should be
+proportional to the quotient of the current by the velocity acquired by
+an ion under unit potential gradient. Now this velocity is inversely
+proportional to the pressure, so that the reaction should on this view
+be directly proportional to the pressure. This agrees with Arrhenius'
+results when the point is positive. Again, the velocities of an ion in
+hydrogen, air and carbonic acid at the same pressure are approximately
+inversely proportional to the square roots of their molecular weights,
+so that the reaction should be directly proportional to this quantity.
+This also agrees with Arrhenius' results for the discharge from a
+positive point. The velocity of the negative ion is greater than that of
+a positive one under the same potential gradient, so that the reaction
+for the negative point should be less than that for a positive one, but
+the excess of the positive reaction over the negative is much greater
+than that of the velocity of the negative ion over the velocity of the
+positive. There is, however, reason to believe that a considerable
+condensation takes place around the negative ion as a nucleus after it
+is formed, so that the velocity of the negative ion under a given
+potential gradient will be greater immediately after the ion is formed
+than when it has existed for some time. The measurements which have been
+made of the velocities of the ions relate to those which have been some
+time in existence, but a large part of the reaction will be due to the
+newly-formed ions moving with a greater velocity, and thus giving a
+smaller reaction than that calculated from the observed velocity.
+
+With a given potential difference between the point and the neighbouring
+conductor the current issuing from the point is greater when the point
+is negative than when it is positive, except in oxygen, when it is less.
+Warburg (_Sitz. Akad. d. Wissensch. zu Berlin_, 1899, 50, p. 770) has
+shown that the addition of a small quantity of oxygen to nitrogen
+produces a great diminution in the current from a negative point, but
+has very little effect on the discharge from a positive point. Thus the
+removal of a trace of oxygen made a leak from a negative point 50 times
+what it was before. Experiments with hydrogen and helium showed that
+impurities in these gases had a great effect on the current when the
+point was negative, and but little when it was positive. This suggests
+that the impurities, by condensing round the negative ions as nuclei,
+seriously diminish their velocity. If a point is charged up to a high
+and rapidly alternating potential, such as can be produced by the
+electric oscillations started when a Leyden jar is discharged, then in
+hydrogen, nitrogen, ammonia and carbonic acid gas a conductor placed in
+the neighbourhood of the point gets a negative charge, while in air and
+oxygen it gets a positive one. There are two considerations which are of
+importance in connexion with this effect. The first is the velocity of
+the ions in the electric field, and the second the ease with which the
+ions can give up their charges to the metal point. The greater velocity
+of the negative ions would, if the potential were rapidly alternating,
+cause an excess of negative ions to be left in the surrounding gas. This
+is the case in hydrogen. If, however, the metal had a much greater
+tendency to unite with negative than with positive ions, such as we
+should expect to be the case in oxygen, this would act in the opposite
+direction, and tend to leave an excess of positive ions in the gas.
+
+_The Characteristic Curve for Discharge through Gases._--When a current
+of electricity passes through a metallic conductor the relation between
+the current and the potential difference is the exceedingly simple one
+expressed by Ohm's law; the current is proportional to the potential
+difference. When the current passes through a gas there is no such
+simple relation. Thus we have already mentioned cases where the current
+increased as the potential increased although not in the same
+proportion, while as we have seen in certain stages of the arc discharge
+the potential difference diminishes as the current increases. Thus the
+problem of finding the current which a given battery will produce when
+part of the circuit consists of a gas discharge is much more complicated
+than when the circuit consists entirely of metallic conductors. If,
+however, we measure the potential difference between the electrodes in
+the gas when different currents are sent through it, we can plot a
+curve, called the "characteristic curve," whose ordinates are the
+potential differences between the electrodes in the gas and the
+abscissae the corresponding currents. By the aid of this curve we can
+calculate the current produced when a given battery is connected up to
+the gas by leads of known resistance.
+
+ For let E0 be the electromotive force of the battery, R the resistance
+ of the leads, i the current, the potential difference between the
+ terms in the gas will be E0 - Ri. Let ABC (fig. 22) be the
+ "characteristic curve," the ordinates being the potential difference
+ between the terminals in the gas, and the abscissae the current. Draw
+ the line LM whose equation is E = E0 - Ri, then the points where this
+ line cuts the characteristic curves will give possible values of i and
+ E, the current through the discharge tube and the potential difference
+ between the terminals. Some of these points may, however, correspond
+ to an unstable position and be impossible to realize. The following
+ method gives us a criterion by which we can distinguish the stable
+ from the unstable positions. If the current is increased by [delta]i,
+ the electromotive force which has to be overcome by the battery is
+ R[delta]i + dE/di . [delta]i. If R + dE/di is positive there will be
+ an unbalanced electromotive force round the circuit tending to stop
+ the current. Thus the increase in the current will be stopped and the
+ condition will be a stable one. If, however, R + dE/di is negative
+ there will be an unbalanced electromotive force tending to increase
+ the current still further; thus the current will go on increasing and
+ the condition will be unstable. Thus for stability R + dE/di must be
+ positive, a condition first given by Kaufmann (_Ann. der Phys._ 11, p.
+ 158). The geometrical interpretation of this condition is that the
+ straight line LM must, at the point where it cuts the characteristic
+ curve, be steeper than the tangent to characteristic curve. Thus of
+ the points ABC where the line cuts the curve in fig. 22, A and C
+ correspond to stable states and B to an unstable one. The state of
+ things represented by a point P on the characteristic curve when the
+ slope is downward cannot be stable unless there is in the external
+ circuit a resistance greater than that represented by the tangent of
+ the inclination of the tangent to the curve at P to the horizontal
+ axis.
+
+ [Illustration: FIG. 22.]
+
+ If we keep the external electromotive force the same and gradually
+ increase the resistance in the leads, the line LM will become steeper
+ and steeper. C will move to the left so that the current will
+ diminish; when the line gets so steep that it touches the curve at C',
+ any further increase in the resistance will produce an abrupt change
+ in the current; for now the state of things represented by a point
+ near A' is the only stable state. Thus if the BC part of the curve
+ corresponded to a luminous discharge and the A part to a dark
+ discharge, we see that if the electromotive force is kept constant
+ there is a minimum value of the current for the luminous discharge. If
+ the current is reduced below this value, the discharge ceases to be
+ luminous, and there is an abrupt diminution in the current.
+
+_Cathode Rays._--When the gas in the discharge tube is at a very low
+pressure some remarkable phenomena occur in the neighbourhood of the
+cathode. These seem to have been first observed by Plucker (_Pogg. Ann._
+107, p. 77; 116, p. 45) who noticed on the walls of the glass tube near
+the cathode a greenish phosphorescence, which he regarded as due to rays
+proceeding from the cathode, striking against the sides of the tube, and
+then travelling back to the cathode. He found that the action of a
+magnet on these rays was not the same as the action on the part of the
+discharge near the positive electrode. Hittorf (_Pogg. Ann._ 136, p. 8)
+showed that the agent producing the phosphorescence was intercepted by a
+solid, whether conductor or insulator, placed between the cathode and
+the sides of the tube. He regarded the phosphorescence as caused by a
+motion starting from the cathode and travelling in straight lines
+through the gas. Goldstein (_Monat. der Berl. Akad._, 1876, p. 24)
+confirmed this discovery of Hittorf's, and further showed that a
+distinct, though not very sharp, shadow is cast by a small object placed
+near a large plane cathode. This is a proof that the rays producing the
+phosphorescence must be emitted almost normally from the cathode, and
+not, like the rays of light from a luminous surface, in all directions,
+for such rays would not produce a perceptible shadow if a small body
+were placed near the plane. Goldstein regarded the phosphorescence as
+due to waves in the ether, for whose propagation the gas was not
+necessary. Crookes (_Phil. Trans._, 1879, pt. i. p. 135; pt. ii. pp.
+587, 661), who made many remarkable researches in this subject, took a
+different view. He regarded the rays as streams of negatively
+electrified particles projected normally from the cathode with great
+velocity, and, when the pressure is sufficiently low, reaching the sides
+of the tube, and by their impact producing phosphorescence and heat. The
+rays on this view are deflected by a magnet, because a magnet exerts a
+force on a charged moving body.
+
+These rays striking against glass make it phosphorescent. The colour of
+the phosphorescence depends on the kind of glass; thus the light from soda
+glass is a yellowish green, and that from lead glass blue. Many other
+bodies phosphoresce when exposed to these rays, and in particular the
+phosphorescence of some gems, such as rubies and diamonds, is exceedingly
+vivid. The spectrum of the phosphorescent light is generally continuous,
+but Crookes showed that the phosphorescence of some of the rare earths,
+such as yttrium, gives a spectrum of bright bands, and he founded on this
+fact a spectroscopic method of great importance. Goldstein (_Wied. Ann._
+54, p. 371) discovered that the haloid salts of the alkali metals change
+colour under the rays, sodium chloride, for example, becoming violet. The
+coloration is a surface one, and has been traced by E. Wiedemann and
+Schmidt (_Wied. Ann._ 54, p. 618) to the formation of a subchloride.
+Chlorides of tin, mercury and lead also change colour in the same way. E.
+Wiedemann (_Wied. Ann._ 56, p. 201) discovered another remarkable effect,
+which he called thermo-luminescence; he found that many bodies after being
+exposed to the cathode rays possess for some time the power of becoming
+luminous when their temperature is raised to a point far below that at
+which they become luminous in the normal state. Substances belonging to
+the class called by van 't Hoff solid solutions exhibit this property of
+thermo-luminescence to a remarkable extent. They are formed when two
+salts, one greatly in excess of the other, are simultaneously precipitated
+from a solution. A trace of MnSO4 in CaSO4 shows very brilliant
+thermo-luminescence. The impact of cathode rays produces after a time
+perceptible changes in the glass. Crookes (_Phil. Trans._ pt. ii. 1879, p.
+645) found that after glass has been phosphorescing for some time under
+the cathode rays it seems to get tired, and the phosphorescence is not so
+bright as it was initially. Thus, for example, when the shadow of a
+Maltese cross is thrown on the walls of the tube as in fig. 23, if after
+the discharge has been going on for some time the cross is shaken down or
+a new cathode used whose line of fire does not cut the cross, the pattern
+of the cross will still be seen on the glass, but it will now be brighter
+instead of darker than the surrounding portion. The portions shielded by
+the cross, not being tired by being made to phosphoresce for a long time,
+respond more vigorously to the stimulus than those portions which have not
+been protected. Skinner (_Proc. Camb. Phil. Soc._ ix. p. 371) and Thomson
+found on the glass which had been exposed to the rays gelatinous
+filaments, apparently silica, resulting from the reduction of the glass. A
+reducing action was also noticed by Villard (_Journ. de phys._ 3, viii. p.
+140) and Wehnelt (_Wied. Ann._ 67, p. 421). It can be well shown by
+letting the rays fall on a plate of oxidized copper, when the part struck
+by the rays will become bright. The rays heat bodies on which they fall,
+and if they are concentrated by using as a cathode a portion of a
+spherical surface, the heat at the centre becomes so great that a piece of
+platinum wire can be melted or a diamond charred. Measurements of the
+heating effects of the rays have been made by Thomson (_Phil. Mag._ [5],
+44, p. 293) and Cady (_Ann. der Phys._ 1, p. 678). Crookes (_Phil.
+Trans._, 1879, pt. i. p. 152) showed that a vane mounted as in a
+radiometer is set in rotation by the rays, the direction of the rotation
+being the same as would be produced by a stream of particles proceeding
+from the cathode. The movement is not due to the momentum imparted to the
+vanes by the rays, but to the difference in temperature between the sides
+of the vanes, the rays making the side against which they strike hotter
+than the other.
+
+[Illustration: FIG. 23.]
+
+_Effect of a Magnet._--The rays are deflected by a magnet, so that the
+distribution of phosphorescence over the glass and the shape and
+position of the shadows cast by bodies in the tube are altered by the
+proximity of a magnet. The laws of magnetic deflection of these rays
+have been investigated by Plucker (_Pogg._ _Ann._ 103, p. 88), Hittorf
+(_Pogg. Ann._ 136, p. 213), Crookes (_Phil. Trans._, 1879, pt. 1, p.
+557), and Schuster (_Proc. Roy. Soc._ 47, p. 526). The deflection is the
+same as that of negatively electrified particles travelling along the
+path of the rays. Such particles would in a magnetic field be acted on
+by a force at right angles to the direction of motion of the particle
+and also to the magnetic force, the magnitude of the force being
+proportional to the product of the velocity of the particle, the
+magnetic force, and the sine of the angle between these vectors. In this
+case we have seen that if the particle is not acted on by an
+electrostatic field, the path in a uniform magnetic field is a spiral,
+which, if the magnetic force is at right angles to the direction of
+projection of the particle, becomes a circle in the plane at right
+angles to the magnetic force, the radius being mv/He, where m, v, e are
+respectively the mass, velocity and charge on the particle, and H is the
+magnetic force. The smaller the difference of potential between the
+electrodes of the discharge tube the greater the deflection produced by
+a magnetic field of given strength, and as the difference of potential
+rapidly increases with diminution of pressure, after a certain pressure
+has been passed, the higher the exhaustion of the tube the less the
+magnetic deflection of the rays. Birkeland (_Comptes rendus_, 1896, p.
+492) has shown that when the discharge is from an induction coil the
+cathode rays produced in the tube at any one time are not equally
+deflected by a magnet, but that a narrow patch of phosphorescence when
+deflected by a magnet is split up into several distinct patches, giving
+rise to what Birkeland calls the "magnetic spectrum." Strutt (_Phil.
+Mag._ 48, p. 478) has shown that this magnetic spectrum does not occur
+if the discharge of a large number of cells is employed instead of the
+coil. Thomson (_Proc. Camb. Phil. Soc._ 9, p. 243) has shown that if the
+potential difference between the electrodes is kept the same the
+magnetic deflection is independent of the nature of the gas filling the
+discharge tube; this was tested with gases so different as air,
+hydrogen, carbonic acid and methyl iodide.
+
+_Charge of Negative Electricity carried by the Rays._--We have seen that
+the rays are deflected by a magnet, as if they were particles charged
+with negative electricity. Perrin (_Comptes rendus_, 121, p. 1130)
+showed by direct experiment that a stream of negative electricity is
+associated with the rays. A modification made by Thomson of Perrin's
+experiment is sketched in fig. 24 (_Phil. Mag._ 48, p. 478).
+
+ [Illustration: FIG. 24.]
+
+ The rays start from the cathode A, and pass through a slit in a solid
+ brass rod B fitting tightly into the neck of the tube. This rod is
+ connected with earth and used as the anode. The rays after passing
+ through the slit travel through the vessel C. D and E are two
+ insulated metal cylinders insulated from each other, and each having a
+ slit cut in its face so as to enable the rays to pass into the inside
+ of the inner cylinder, which is connected with an electrometer, the
+ outer cylinder being connected with the earth. The two cylinders are
+ placed on the far side of the vessel, but out of the direct line of
+ fire of the rays. When the rays go straight through the slit there is
+ only a very small negative charge communicated to the inner cylinder,
+ but when they are deflected by a magnet so that the phosphorescent
+ patch falls on the slit in the outer cylinder the inner cylinder
+ receives a very large negative charge, the increase coinciding very
+ sharply with the appearance of the phosphorescent patch on the slit.
+ When the patch is so much deflected by the magnet that it falls below
+ the slit, the negative charge in the cylinder again disappears. This
+ experiment shows that the cathode rays are accompanied by a stream of
+ negative electrification. The same apparatus can be used to show that
+ the passage of cathode rays through a gas makes it a conductor of
+ electricity. For if the induction coil is kept running and a stream of
+ the rays kept steadily going into the inner cylinder, the potential
+ of the inner cylinder reaches a definite negative value below which it
+ does not fall, however long the rays may be kept going. The cylinder
+ reaches a steady state in which the gain of negative electricity from
+ the cathode rays is equal to the loss by leakage through the
+ conducting gas, the conductivity being produced by the passage of the
+ rays through it. If the inner cylinder is charged up initially with a
+ greater negative charge than corresponds to the steady state, on
+ turning the rays on to the cylinder the negative charge will decrease
+ and not increase until it reaches the steady state. The conductivity
+ produced by the passage of cathode rays through a gas diminishes
+ rapidly with the pressure. When rays pass through a gas at a low
+ pressure, they are deflected by an electric field; when the pressure
+ of the gas is higher the conductivity it acquires when the cathode
+ rays pass through it is so large that the potential gradient cannot
+ reach a sufficiently high value to produce an appreciable deflection.
+
+Thus the cathode rays carry a charge of negative electricity; the
+experiment described on page 875 (fig. 13) shows that they are deflected
+by an electric field as if they were negatively electrified, and are
+acted on by a magnetic force in just the way this force would act on a
+negatively electrified body moving along the path of the rays. There is
+therefore every reason for believing that they are charges of negative
+electricity in rapid motion. By measuring the deflection produced by
+magnetic and electric fields we can determine the velocity with which
+these particles moved and the ratio of the mass of the particle to the
+charge carried by it.
+
+We may conclude from the experiments that the value of m/e for the
+particles constituting the cathode rays is of the order 1/1.7 X 10^7,
+and we have seen that m/e has the same value in all the other cases of
+negative ions in a gas at low pressure for which it has been
+measured--viz. for the ions produced when ultra-violet light falls on a
+metal plate, or when an incandescent carbon filament is surrounded by a
+gas at a low pressure, and for the [beta] particles given out by
+radio-active bodies. We have also seen that the value of the charge on
+the gaseous ion, in all cases in which it has been measured--viz. the
+ions produced by Rontgen and uranium radiation, by ultra-violet light,
+and by the discharge of electrification from a point--is the same in
+magnitude as the charge carried by the hydrogen atom in the electrolysis
+of solutions. The mass of the hydrogen alone is, however, 10^-4 times
+this charge, while the mass of the carriers of negative electrification
+is only 1/1.7 X 10^7 times the charge; hence the mass of the carriers of
+the negative electrification is only 1/1700 of the mass of the hydrogen
+atom. We are thus, by the study of the electric discharge, forced to
+recognize the existence of masses very much smaller than the smallest
+mass hitherto recognized.
+
+ Direct determinations of the velocity of the cathode rays have been
+ made by J. J. Thomson (_Phil. Mag._ 38, p. 358), who measured the
+ interval between the appearance of phosphorescence on two pieces of
+ glass placed at a known distance apart, and by Maiorana (_Nuovo
+ Cimento_, 4, 6, p. 336) and Battelli and Stefanini (_Phys. Zeit._ 1,
+ p. 51), who measured the interval between the arrival of the negative
+ charge carried by the rays at two places separated by a known
+ distance. The values of the velocity got in this way are much smaller
+ than the values got by the indirect methods previously described: thus
+ J. J. Thomson at a fairly high pressure found the velocity to be 2 X
+ 10^7 cm./sec. Maiorana found values ranging between 10^7 and 6 X 10^7
+ cm./sec, and Battelli and Stefanini values ranging from 6 X 10^6 to
+ 1.2 X 10^7. In these methods it is very difficult to eliminate the
+ effect of the interval which elapses between the arrival of the rays
+ and the attainment by the means of detection, such as the
+ phosphorescence of the glass or the deflection of the electrometer, of
+ sufficient intensity to affect the senses.
+
+[Illustration: FIG. 25.]
+
+_Transmission of Cathode Rays through Solids--Lenard Rays._--It was for
+a long time believed that all solids were absolutely opaque to these
+rays, as Crookes and Goldstein had proved that very thin glass, and even
+a film of collodion, cast intensely black shadows. Hertz (_Wied. Ann._
+45, p. 28), however, showed that behind a piece of gold-leaf or
+aluminium foil an appreciable amount of phosphorescence occurred on the
+glass, and that the phosphorescence moved when a magnet was brought
+near. A most important advance was next made by Lenard (_Wied. Ann._ 51,
+p. 225), who got the cathode rays to pass from the inside of a discharge
+tube to the air outside. For this purpose he used a tube like that shown
+in fig. 25. The cathode K is an aluminium disc 1.2 cm. in diameter
+fastened to a stiff wire, which is surrounded by a glass tube. The anode
+A is a brass strip partly surrounding the cathode. The end of the tube
+in front of the cathode is closed by a strong metal cap, fastened in
+with marine glue, in the middle of which a hole 1.7 mm. in diameter is
+bored, and covered with a piece of very thin aluminium foil about .0026
+mm. in thickness. The aluminium window is in metallic contact with the
+cap, and this and the anode are connected with the earth. The tube is
+then exhausted until the cathode rays strike against the window. Diffuse
+light spreads from the window into the air outside the tube, and can be
+traced in a dark room for a distance of several centimetres. From the
+window, too, proceed rays which, like the cathode rays, can produce
+phosphorescence, for certain bodies phosphoresce when placed in the
+neighbourhood of the window. This effect is conveniently observed by the
+platino-cryanide screens used to detect Rontgen radiation. The
+properties of the rays outside the tube resemble in all respects those
+of cathode rays; they are deflected by a magnet and by an electric
+field, they ionize the gas through which they pass and make it a
+conductor of electricity, and they affect a photographic plate and
+change the colour of the haloid salts of the alkali metals. As, however,
+it is convenient to distinguish between cathode rays outside and inside
+the tube, we shall call the former Lenard rays. In air at atmospheric
+pressure the Lenard rays spread out very diffusely. If the aluminium
+window, instead of opening into the air, opens into another tube which
+can be exhausted, it is found that the lower the pressure of the gas in
+this tube the farther the rays travel and the less diffuse they are. By
+filling the tube with different gases Lenard showed that the greater the
+density of the gas the greater is the absorption of these rays. Thus
+they travel farther in hydrogen than in any other gas at the same
+pressure. Lenard showed, too, that if he adjusted the pressure so that
+the density of the gas in this tube was the same--if, for example, the
+pressure when the tube was filled with oxygen was 1/16 of the pressure
+when it was filled with hydrogen--the absorption was constant whatever
+the nature of the gas. Becker (_Ann. der Phys._ 17, p. 381) has shown
+that this law is only approximately true, the absorption by hydrogen
+being abnormally large, and by the inert monatomic gases, such as helium
+and argon, abnormally small. The distance to which the Lenard rays
+penetrate into this tube depends upon the pressure in the discharge
+tube; if the exhaustion in the latter is very high, so that there is a
+large potential difference between the cathode and the anode, and
+therefore a high velocity for the cathode rays, the Lenard rays will
+penetrate farther than when the pressure in the discharge tube is higher
+and the velocity of the cathode rays smaller. Lenard showed that the
+greater the penetrating power of his rays the smaller was their magnetic
+deflection, and therefore the greater their velocity; thus the greater
+the velocity of the cathode rays the greater is the velocity of the
+Lenard rays to which they give rise. For very slow cathode rays the
+absorption by different gases departs altogether from the density law,
+so much so that the absorption of these rays by hydrogen is greater than
+that by air (Lenard, _Ann. der Phys._ 12, p. 732). Lenard (_Wied. Ann._
+56, p. 255) studied the passage of his rays through solids as well as
+through gases, and arrived at the very interesting result that the
+absorption of a substance depends only upon its density, and not upon
+its chemical composition or physical state; in other words, the amount
+of absorption of the rays when they traverse a given distance depends
+only on the quantity of matter they cut through in the distance.
+McClelland (_Proc. Roy. Soc._ 61, p. 227) showed that the rays carry a
+charge of negative electricity, and M'Lennan measured the amount of
+ionization rays of given intensity produced in different gases, finding
+that if the pressure is adjusted so that the density of the different
+gases is the same the number of ions per cubic centimetre is also the
+same. In this case, as Lenard has shown, the absorption is the same, so
+that with the Lenard rays, as with uranium and probably with Rontgen
+rays, equal absorption corresponds to equal ionization. A convenient
+method for producing Lenard rays of great intensity has been described
+by Des Coudres (_Wied. Ann._ 62, p. 134).
+
+_Diffuse Reflection of Cathode Rays._--When cathode rays fall upon a
+surface, whether of an insulator or a conductor, cathode rays start from
+the surface in all directions. This phenomenon, which was discovered by
+Goldstein (_Wied. Ann._ 62, p. 134), has been investigated by Starke
+(_Wied. Ann._ 66, p. 49; _Ann. der Phys._ 111, p. 75), Austin and Starke
+(_Ann. der Phys._ 9, p. 271), Campbell-Swinton (_Proc. Roy. Soc._ 64, p.
+377), Merritt (_Phys. Rev._ 7, p. 217) and Gehrcke (_Ann. der Phys._ 8,
+p. 81); it is often regarded as analogous to the diffuse reflection of
+light from such a surface as gypsum, and is spoken of as the diffuse
+reflection of the cathode rays. According to Merritt and Austin and
+Starke the deviation in a magnetic field of these reflected rays is the
+same as that of the incident rays. The experiments, however, were
+confined to rays reflected so that the angle of reflection was nearly
+equal to that of incidence. Gehrcke showed that among the reflected rays
+there were a large number which had a much smaller velocity than the
+incident ones. According to Campbell-Swinton the "diffuse" reflection is
+accompanied by a certain amount of "specular" reflection. Lenard, who
+used slower cathode rays than Austin and Starke, could not detect in the
+scattered rays any with velocities comparable with that of the incident
+rays; he obtained copious supplies of slow rays whose speed did not
+depend on the angle of incidence of the primary rays (_Ann. der Phys._
+15, p. 485). When the angle of incidence is very oblique the surface
+struck by the rays gets positively charged, showing that the secondary
+rays are more numerous than the primary.
+
+_Repulsion of two Cathode Streams._--Goldstein discovered that if in a
+tube there are two cathodes connected together, the cathodic rays from
+one cathode are deflected when they pass near the other. Experiments
+bearing on this subject have been made by Crookes and Wiedemann and
+Ebert. The phenomena may be described by saying that the repulsion of
+the rays from a cathode A by a cathode B is only appreciable when the
+rays from A pass through the Crookes dark space round B. This is what we
+should expect if we remember that the electric field in the dark space
+is far stronger than in the rest of the discharge, and that the gas in
+the other parts of the tube is rendered a conductor by the passage
+through it of the cathode rays, and therefore incapable of transmitting
+electrostatic repulsion.
+
+Scattering of the Negative Electrodes.--In addition to the cathode rays,
+portions of metal start normally from the cathode and form a metallic
+deposit on the walls of the tube. The amount of this deposit varies very
+much with the metal. Crookes (_Proc. Roy. Soc._ 50, p. 88) found that
+the quantities of metal torn from electrodes of the same size, in equal
+times, by the same current, are in the order Pd, Au, Ag, Pb, Sn, Pt, Cu,
+Cd, Ni, In, Fe.... In air there is very little deposit from an Al
+cathode, but it is abundant in tubes filled with the monatomic gases,
+mercury vapour, argon or helium. The scattering increases as the density
+of the gas diminishes. The particles of metal are at low pressures
+deflected by a magnet, though not nearly to the same extent as the
+cathode rays. According to Grandquist, the loss of weight of the cathode
+in a given time is proportional to the square of the current; it is
+therefore not, like the loss of the cathode in ordinary electrolysis,
+proportional to the quantity of current which passes through it.
+
+[Illustration: FIG. 26.]
+
+_Positive Rays or "Canalstrahlen."_--Goldstein (_Berl. Sitzungsb._ 39,
+p. 691) found that with a perforated cathode certain rays occurred
+behind the cathode which were not appreciably deflected by a magnet;
+these he called Canalstrahlen, but we shall, for reasons which will
+appear later, call them "positive rays."
+
+Their appearance is well shown in fig. 26, taken from a paper by Wehnelt
+(_Wied. Ann._ 67, p. 421) in which they are represented at B. Goldstein
+found that their colour depends on the gas in which they are formed,
+being gold-colour in air and nitrogen, rose-colour in hydrogen,
+yellowish rose in oxygen, and greenish gray in carbonic acid.
+
+The colour of the luminosity due to positive rays is not in general the
+same as that due to anode rays; the difference is exceptionally well
+marked in helium, where the cathode ray luminosity is blue while that
+due to the positive rays is red. The luminosity produced when the rays
+strike against solids is also quite distinct. The cathode rays make the
+body emit a continuous spectrum, while the spectrum produced by the
+positive rays often shows bright lines. Thus lithium chloride under
+cathode rays gives out a steely blue light and the spectrum is
+continuous, while under the positive rays the salt gives out a brilliant
+red light and the spectrum shows the red helium line. It is remarkable
+that the lines on the spectra of the alkali metals are much more easily
+produced when the positive rays fall on the oxide of the metal than when
+they fall on the metal itself. Thus when the positive rays fall on a
+pool of the liquid alloy of sodium and potassium the specks of oxide on
+the surface shine with a bright yellow light while the untarnished part
+of the surface is quite dark.
+
+W. Wien (_Wied. Ann._ 65, p. 445) measured the values of e/m for the
+particles forming the positive rays. Other measurements have been made
+by Ewers (_Wied. Ann._ 69, p. 167) and J. J. Thomson (_Phil. Mag._ 13,
+p. 561). The differences between the values of e/m for the cathode and
+positive rays are very remarkable. For cathode rays whose velocity does
+not approach that of light, e/m is always equal to 1.7 X 10^8, while for
+the positive rays the greatest value of this quantity yet observed is
+10^4, which is also the value of e/m for the hydrogen ions in the
+electrolysis of dilute solutions. In some experiments made by J. J.
+Thomson (_Phil. Mag._, 14, p. 359) it was found that when the pressure
+of the gas was not too low the bright spot produced by the impact of a
+pencil of these rays on a phosphorescent screen is deflected by electric
+and magnetic forces into a continuous band extending on both sides of
+the undeflected position. The portion on one side is in general much
+fainter than that on the other. The direction of this deflection shows
+that it is produced by particles charged with negative electricity,
+while the brighter band is due to particles charged with positive
+electricity. The negatively electrified particles which produce the band
+c.c are not corpuscles, for from the electric and magnetic deflections
+we can find the value of e/m. As this proves to be equal to 10^4, we see
+that the mass of the carrier of the negative charge is comparable with
+that of an atom, and so very much greater than that of a corpuscle. At
+very low pressures part of the phosphorescence disappears, while the
+upper portion breaks up into two patches (fig. 27). For one of these the
+maximum value of e/m is 10^4 and for the other 5 X 10^3. At low pressures
+the appearance of the patches and the values of e/m are the same whether
+the tube is filled originally with air, hydrogen or helium. In some of
+the experiments the tube was exhausted until the pressure was too low to
+allow the discharge to pass. A very small quantity of the gas under
+investigation was then admitted into the tube, just sufficient to allow
+the discharge to pass, and the deflection of the phosphorescent patch
+measured. The following gases were admitted into the tube, air, carbonic
+oxide, oxygen, hydrogen, helium, argon and neon, but whatever the gas
+the appearance of the phosphorescence was the same; in every case there
+were two patches, for one of which e/m = 10^4 and for the other e/m =
+5 X 10^3. In helium at higher pressures another patch was observed, for
+which e/m = 2.5 X 10^8. The continuous band into which the
+phosphorescent spot is drawn out when the pressure is not exceedingly
+low, which involves the existence of particles for which the mean value
+of e/m varies from zero to 10^4, can be explained as follows. The rays
+on their way to the phosphorescent screen have to pass through gas which
+is ionized by the passage through it of the positive rays; this gas will
+therefore contain free corpuscles. The particles which constitute the
+rays start with a charge of positive electricity. Some of these
+particles in their journey through the gas attract a corpuscle whose
+negative charge neutralizes the positive charge on the particle. The
+particles when in this neutral state may be ionized by collision and
+reacquire a positive charge, or by attracting another particle may
+become negatively charged, and this process may be repeated several
+times on their journey to the phosphorescent screen. Thus some of the
+particles, instead of being positively charged for the whole of the time
+they are exposed to the electric and magnetic forces, may be for a part
+of that time without a charge or even have a negative charge. The
+deflection of a particle is proportional to the average value of its
+charge whilst under the influence of the deflecting forces. Thus if a
+particle is without a charge for a part of the time, its deflection will
+be less than that of a particle which has retained its positive charge
+for the whole of its journey, while the few particles which have a
+negative charge for a longer time than they have a positive will be
+deflected in the opposite direction to the main portion and will produce
+the tail (fig. 27).
+
+[Illustration: Fig. 27.]
+
+A similar explanation will apply to the positive rays discovered by
+Villard (_Comptes rendus_, 143, p. 674) and J. J. Thomson (_Phil. Mag._
+13, p. 359), which travel in the opposite direction to the rays we have
+been considering, i.e. they travel away from the cathode and in the
+direction of the cathode's rays; these rays are sometimes called
+"retrograde" rays. These as far as has been observed have always the
+same maximum value of e/m, i.e. 10^4, and there are a considerable
+number of negative ones always mixed with them. The maximum velocity of
+both the positive and retrograde rays is about 2 X 10^8 cm./sec. and
+varies very little with the potential difference between the electrodes
+in the tube in which they are produced (J. J. Thomson, _Phil. Mag._,
+Dec. 1909).
+
+The positive rays show, when the pressure is not very low, the line
+spectrum of the gas through which they pass. An exceedingly valuable set
+of observations on this point have been made by Stark and his pupils
+(_Physik. Zeit._ 6, p. 892; _Ann. der Phys._ 21, pp. 40, 457). Stark has
+shown that in many gases, notably hydrogen, the spectrum shows the
+Doppler effect, and he has been able to calculate in this way the
+velocity of the positive rays.
+
+_Anode Rays._--Gehrcke and Reichenhein (_Ann. der Phys._ 25, p. 861)
+have found that when the anode consists of a mixture of sodium and
+lithium chloride raised to a high temperature either by the discharge
+itself or by an independent heating circuit, very conspicuous rays come
+from the anode when the pressure of the gas in the discharge tube is
+very low, and a large coil is used to produce the discharge. The
+determination of e/m for these rays showed that they are positively
+charged atoms of sodium or lithium, moving with very considerable
+velocity; in some of Gehrcke's experiments the maximum velocity was as
+great as 1.8 X 10^7 cm./sec. though the average was about 10^7 cm./sec.
+These velocities are less than those of the positive rays whose maximum
+velocity is about 2 X 10^8 cm./sec. (J. J. T.)
+
+
+FOOTNOTES:
+
+ [1] The values for nickel and bismuth given in the table are much
+ higher than later values obtained with pure electrolytic nickel and
+ bismuth.
+
+ [2] The value here given, namely 12.885, for the electric
+ mass-resistivity of liquid mercury as determined by Matthiessen is
+ now known to be too high by nearly 1%. The value at present accepted
+ is 12.789 ohms per metre-gramme at 0 deg. C.
+
+ [3] The value (1630) here given for hard-drawn copper is about 1/4%
+ higher than the value now adopted, namely, 1626. The difference is
+ due to the fact that either Jenkin or Matthiessen did not employ
+ precisely the value at present employed for the density of hard-drawn
+ and annealed copper in calculating the volume-resistivities from the
+ mass-resistivities.
+
+ [4] Matthiessen's value for nickel is much greater than that obtained
+ in more recent researches. (See Matthiessen and Vogt, _Phil. Trans._,
+ 1863, and J. A. Fleming, _Proc. Roy. Soc._, December 1899.)
+
+ [5] Matthiessen's value for mercury is nearly 1% greater than the
+ value adopted at present as the mean of the best results, namely
+ 94,070.
+
+ [6] The samples of silver, copper and nickel employed for these tests
+ were prepared electrolytically by Sir J. W. Swan, and were
+ exceedingly pure and soft. The value for volume-resistivity of nickel
+ as given in the above table (from experiments by J. A. Fleming,
+ _Proc. Roy. Soc._, December 1899) is much less (nearly 40%) than the
+ value given by Matthiessen's researches.
+
+ [7] The electrolytic bismuth here used was prepared by Hartmann and
+ Braun, and the resistivity taken by J. A. Fleming. The value is
+ nearly 20% less than that given by Matthiessen.
+
+ [8] In 1899 a committee was formed of representatives from eight of
+ the leading manufacturers of insulated copper cables with delegates
+ from the Post Office and Institution of Electrical Engineers, to
+ consider the question of the values to be assigned to the resistivity
+ of hard-drawn and annealed copper. The sittings of the committee were
+ held in London, the secretary being A. H. Howard. The values given in
+ the above paragraphs are in accordance with the decision of this
+ committee, and its recommendations have been accepted by the General
+ Post Office and the leading manufacturers of insulated copper wire
+ and cables.
+
+ [9] Platinoid is an alloy introduced by Martino, said to be similar
+ in composition to German silver, but with a little tungsten added. It
+ varies a good deal in composition according to manufacture, and the
+ resistivity of different specimens is not identical. Its electric
+ properties were first made known by J. T. Bottomley, in a paper read
+ at the Royal Society, May 5, 1885.
+
+ [10] An equivalent gramme molecule is a weight in grammes equal
+ numerically to the chemical equivalent of the salt. For instance, one
+ equivalent gramme molecule of sodium chloride is a mass of 58.5
+ grammes. NaCl = 58.5.
+
+ [11] F. Kohlrausch and L. Holborn, _Das Leitvermogen der Elektrolyte_
+ (Leipzig, 1898).
+
+ [12] It should be noticed that the velocities calculated in
+ Kohlrausch's theory and observed experimentally are the average
+ velocities, and involve both the factors mentioned above; they
+ include the time wasted by the ions in combination with each other,
+ and, except at great dilution, are less than the velocity with which
+ the ions move when free from each other.
+
+
+
+
+
+End of the Project Gutenberg EBook of Encyclopaedia Britannica, 11th
+Edition, Volume 6, Slice 8, by Various
+
+*** END OF THIS PROJECT GUTENBERG EBOOK ENCYC. BRITANNICA, VOL 6, SL 8 ***
+
+***** This file should be named 32063.txt or 32063.zip *****
+This and all associated files of various formats will be found in:
+ https://www.gutenberg.org/3/2/0/6/32063/
+
+Produced by Marius Masi, Don Kretz, Juliet Sutherland, and
+the Online Distributed Proofreading Team at
+https://www.pgdp.net
+
+
+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
+https://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
+
+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 F3. 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 MERCHANTIBILITY 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 https://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
+https://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 https://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 https://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 including checks, online payments and credit card
+donations. To donate, please visit: https://pglaf.org/donate
+
+
+Section 5. General Information About Project Gutenberg-tm electronic
+works.
+
+Professor Michael S. Hart was 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:
+
+ https://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.
diff --git a/32063.zip b/32063.zip
new file mode 100644
index 0000000..84ff3f1
--- /dev/null
+++ b/32063.zip
Binary files differ
diff --git a/LICENSE.txt b/LICENSE.txt
new file mode 100644
index 0000000..6312041
--- /dev/null
+++ b/LICENSE.txt
@@ -0,0 +1,11 @@
+This eBook, including all associated images, markup, improvements,
+metadata, and any other content or labor, has been confirmed to be
+in the PUBLIC DOMAIN IN THE UNITED STATES.
+
+Procedures for determining public domain status are described in
+the "Copyright How-To" at https://www.gutenberg.org.
+
+No investigation has been made concerning possible copyrights in
+jurisdictions other than the United States. Anyone seeking to utilize
+this eBook outside of the United States should confirm copyright
+status under the laws that apply to them.
diff --git a/README.md b/README.md
new file mode 100644
index 0000000..72364c4
--- /dev/null
+++ b/README.md
@@ -0,0 +1,2 @@
+Project Gutenberg (https://www.gutenberg.org) public repository for
+eBook #32063 (https://www.gutenberg.org/ebooks/32063)