summaryrefslogtreecommitdiff
diff options
context:
space:
mode:
-rw-r--r--.gitattributes3
-rw-r--r--38722-pdf.pdfbin0 -> 3217612 bytes
-rw-r--r--38722-pdf.zipbin0 -> 2753837 bytes
-rw-r--r--38722-t.zipbin0 -> 1956822 bytes
-rw-r--r--38722-t/38722-t.tex15606
-rw-r--r--38722-t/images/fig1.pngbin0 -> 28665 bytes
-rw-r--r--38722-t/images/fig10.pngbin0 -> 11499 bytes
-rw-r--r--38722-t/images/fig11.pngbin0 -> 18678 bytes
-rw-r--r--38722-t/images/fig12.pngbin0 -> 6453 bytes
-rw-r--r--38722-t/images/fig13.pngbin0 -> 6270 bytes
-rw-r--r--38722-t/images/fig13b.pdfbin0 -> 11563 bytes
-rw-r--r--38722-t/images/fig14.pngbin0 -> 50119 bytes
-rw-r--r--38722-t/images/fig15.pngbin0 -> 2734 bytes
-rw-r--r--38722-t/images/fig16.pngbin0 -> 25135 bytes
-rw-r--r--38722-t/images/fig17.jpgbin0 -> 88099 bytes
-rw-r--r--38722-t/images/fig18.jpgbin0 -> 90830 bytes
-rw-r--r--38722-t/images/fig19.jpgbin0 -> 145066 bytes
-rw-r--r--38722-t/images/fig2.pngbin0 -> 17960 bytes
-rw-r--r--38722-t/images/fig20.pngbin0 -> 8027 bytes
-rw-r--r--38722-t/images/fig21.pngbin0 -> 8101 bytes
-rw-r--r--38722-t/images/fig22.pngbin0 -> 15248 bytes
-rw-r--r--38722-t/images/fig23.jpgbin0 -> 94898 bytes
-rw-r--r--38722-t/images/fig24.pngbin0 -> 7758 bytes
-rw-r--r--38722-t/images/fig25.pngbin0 -> 9637 bytes
-rw-r--r--38722-t/images/fig26.pngbin0 -> 5367 bytes
-rw-r--r--38722-t/images/fig27.pngbin0 -> 4542 bytes
-rw-r--r--38722-t/images/fig28.pngbin0 -> 7291 bytes
-rw-r--r--38722-t/images/fig29.pngbin0 -> 72485 bytes
-rw-r--r--38722-t/images/fig3.jpgbin0 -> 61974 bytes
-rw-r--r--38722-t/images/fig30.pngbin0 -> 5708 bytes
-rw-r--r--38722-t/images/fig31.pngbin0 -> 3851 bytes
-rw-r--r--38722-t/images/fig32.jpgbin0 -> 131548 bytes
-rw-r--r--38722-t/images/fig33.jpgbin0 -> 92070 bytes
-rw-r--r--38722-t/images/fig34.pngbin0 -> 35839 bytes
-rw-r--r--38722-t/images/fig35.pngbin0 -> 17978 bytes
-rw-r--r--38722-t/images/fig36.pngbin0 -> 7658 bytes
-rw-r--r--38722-t/images/fig37.pngbin0 -> 10983 bytes
-rw-r--r--38722-t/images/fig38.pngbin0 -> 13902 bytes
-rw-r--r--38722-t/images/fig39.pngbin0 -> 11348 bytes
-rw-r--r--38722-t/images/fig4.jpgbin0 -> 160817 bytes
-rw-r--r--38722-t/images/fig40.jpgbin0 -> 167372 bytes
-rw-r--r--38722-t/images/fig41.jpgbin0 -> 95790 bytes
-rw-r--r--38722-t/images/fig42.pngbin0 -> 10195 bytes
-rw-r--r--38722-t/images/fig43.pngbin0 -> 5232 bytes
-rw-r--r--38722-t/images/fig5.jpgbin0 -> 144036 bytes
-rw-r--r--38722-t/images/fig6.pngbin0 -> 5532 bytes
-rw-r--r--38722-t/images/fig7.pngbin0 -> 19893 bytes
-rw-r--r--38722-t/images/fig8.pngbin0 -> 10965 bytes
-rw-r--r--38722-t/images/riverside.pngbin0 -> 15848 bytes
-rw-r--r--38722-t/old/38722-t.tex15600
-rw-r--r--38722-t/old/38722-t.zipbin0 -> 1956947 bytes
-rw-r--r--LICENSE.txt11
-rw-r--r--README.md2
53 files changed, 31222 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/38722-pdf.pdf b/38722-pdf.pdf
new file mode 100644
index 0000000..c083292
--- /dev/null
+++ b/38722-pdf.pdf
Binary files differ
diff --git a/38722-pdf.zip b/38722-pdf.zip
new file mode 100644
index 0000000..2bb29af
--- /dev/null
+++ b/38722-pdf.zip
Binary files differ
diff --git a/38722-t.zip b/38722-t.zip
new file mode 100644
index 0000000..48b8d40
--- /dev/null
+++ b/38722-t.zip
Binary files differ
diff --git a/38722-t/38722-t.tex b/38722-t/38722-t.tex
new file mode 100644
index 0000000..0149a5e
--- /dev/null
+++ b/38722-t/38722-t.tex
@@ -0,0 +1,15606 @@
+% %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% %
+% %
+% The Project Gutenberg EBook of The Tides and Kindred Phenomena in the %
+% Solar System, by Sir George Howard Darwin %
+% %
+% 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.net %
+% %
+% %
+% Title: The Tides and Kindred Phenomena in the Solar System %
+% The Substance of Lectures Delivered in 1897 at the Lowell %
+% Institute, Boston, Massachusetts %
+% %
+% Author: Sir George Howard Darwin %
+% %
+% Release Date: January 31, 2012 [EBook #38722] %
+% Most recently updated: June 11, 2021 %
+% %
+% Language: English %
+% %
+% Character set encoding: UTF-8 %
+% %
+% *** START OF THIS PROJECT GUTENBERG EBOOK TIDES AND KINDRED PHENOMENA ***
+% %
+% %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% %
+
+\def\ebook{38722}
+%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
+%% %%
+%% Packages and substitutions: %%
+%% %%
+%% book: Required. %%
+%% inputenc: Standard DP encoding. Required. %%
+%% %%
+%% ifthen: Logical conditionals. Required. %%
+%% %%
+%% amsmath: AMS mathematics enhancements. Required. %%
+%% amssymb: Additional mathematical symbols. Required. %%
+%% %%
+%% alltt: Fixed-width font environment. Required. %%
+%% %%
+%% multicol: Multicolumn environment for index. Required. %%
+%% makeidx: Indexing. Required. %%
+%% %%
+%% footmisc: Extended footnote capabilities. Required. %%
+%% %%
+%% indentfirst: Indent first word of each sectional unit. Required. %%
+%% icomma: Make the comma a decimal separator in math. Required. %%
+%% %%
+%% calc: Length calculations. Required. %%
+%% %%
+%% array: Enhanced arrays. Required. %%
+%% %%
+%% yfonts: Gothic font on title page. Optional. %%
+%% %%
+%% fancyhdr: Enhanced running headers and footers. Required. %%
+%% %%
+%% graphicx: Standard interface for graphics inclusion. Required. %%
+%% caption: Caption customization. Required. %%
+%% %%
+%% geometry: Enhanced page layout package. Required. %%
+%% hyperref: Hypertext embellishments for pdf output. Required. %%
+%% %%
+%% %%
+%% Producer's Comments: %%
+%% %%
+%% Changes are noted in this file in multiple ways. %%
+%% 1. \DPnote{} for in-line `placeholder' notes. %%
+%% 2. \DPtypo{}{} for typographical corrections, showing original %%
+%% and replacement text side-by-side. %%
+%% 3. \DPchg{}{} for stylistic changes made for consistency. %%
+%% 4. [** TN: Note]s for lengthier or stylistic comments. %%
+%% %%
+%% %%
+%% Compilation Flags: %%
+%% %%
+%% The following behavior may be controlled by boolean flags. %%
+%% %%
+%% ForPrinting (false by default): %%
+%% Compile a print-optimized PDF file. Set to false for screen- %%
+%% optimized file (pages cropped, one-sided, blue hyperlinks). %%
+%% %%
+%% %%
+%% PDF pages: 422 (if ForPrinting set to false) %%
+%% PDF page size: 4.5 x 6in (non-standard) %%
+%% %%
+%% Images: 1 pdf diagram, 32 png images, 11 jpg images. %%
+%% %%
+%% Summary of log file: %%
+%% * Two underfull hboxes. %%
+%% %%
+%% %%
+%% Compile History: %%
+%% %%
+%% January, 2012: adhere (Andrew D. Hwang) %%
+%% texlive2007, GNU/Linux %%
+%% %%
+%% Command block: %%
+%% %%
+%% pdflatex x2 %%
+%% makeindex %%
+%% pdflatex x2 %%
+%% %%
+%% %%
+%% January 2012: pglatex. %%
+%% Compile this project with: %%
+%% pdflatex 38722-t.tex ..... TWO times %%
+%% makeindex 38722-t.idx %%
+%% pdflatex 38722-t.tex ..... TWO times %%
+%% %%
+%% pdfTeX, Version 3.1415926-1.40.10 (TeX Live 2009/Debian) %%
+%% %%
+%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
+\listfiles
+\documentclass[12pt]{book}[2005/09/16]
+
+%%%%%%%%%%%%%%%%%%%%%%%%%%%%% PACKAGES %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
+\usepackage[utf8]{inputenc}[2006/05/05]
+
+\usepackage{ifthen}[2001/05/26] %% Logical conditionals
+
+\usepackage{amsmath}[2000/07/18] %% Displayed equations
+\usepackage{amssymb}[2002/01/22] %% and additional symbols
+
+\usepackage{alltt}[1997/06/16] %% boilerplate, credits, license
+
+ %% extended footnote capabilities
+\usepackage[perpage]{footmisc}[2005/03/17]
+
+\usepackage{multicol}[2006/05/18]
+\usepackage{makeidx}[2000/03/29]
+
+\usepackage{indentfirst}[1995/11/23]
+\usepackage{icomma}[2002/03/10]
+
+\usepackage{calc}[2005/08/06]
+
+\usepackage{array}
+
+\IfFileExists{yfonts.sty}{%
+ \usepackage{yfonts}[2003/01/08]%
+}{%
+ \providecommand{\textgoth}[1]{\textbf{#1}}%
+}
+
+% for running heads
+\usepackage{fancyhdr}
+
+\usepackage{graphicx}[1999/02/16]%% For diagrams
+\usepackage[labelformat=empty,justification=centerlast,textfont={sc,footnotesize}]{caption}[2007/01/07]
+
+%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
+%%%% Interlude: Set up PRINTING (default) or SCREEN VIEWING %%%%
+%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
+
+% ForPrinting=true (default) false
+% Asymmetric margins Symmetric margins
+% Black hyperlinks Blue hyperlinks
+% Start Preface, ToC, etc. recto No blank verso pages
+%
+\newboolean{ForPrinting}
+%% UNCOMMENT the next line for a PRINT-OPTIMIZED VERSION of the text %%
+%\setboolean{ForPrinting}{true}
+
+%% Initialize values to ForPrinting=false
+\newcommand{\ChapterSpace}{}
+\newcommand{\Margins}{hmarginratio=1:1} % Symmetric margins
+\newcommand{\HLinkColor}{blue} % Hyperlink color
+\newcommand{\PDFPageLayout}{SinglePage}
+\newcommand{\TransNote}{Transcriber's Note}
+\newcommand{\TransNoteCommon}{%
+ Minor typographical corrections and presentational changes have been
+ made without comment. All changes are detailed in the \LaTeX\ source
+ file, which may be downloaded from
+ \begin{center}
+ \texttt{www.gutenberg.org/ebooks/\ebook}.
+ \end{center}
+ \bigskip
+}
+\newcommand{\TransNoteText}{%
+ \TransNoteCommon
+
+ This PDF file is optimized for screen viewing, but may easily be
+ recompiled for printing. Please consult the preamble of the \LaTeX\
+ source file for instructions.
+}
+%% Re-set if ForPrinting=true
+\ifthenelse{\boolean{ForPrinting}}{%
+ \renewcommand{\ChapterSpace}{\vspace*{1in}}
+ \renewcommand{\Margins}{hmarginratio=2:3} % Asymmetric margins
+ \renewcommand{\HLinkColor}{black} % Hyperlink color
+ \renewcommand{\PDFPageLayout}{TwoPageRight}
+ \renewcommand{\TransNoteText}{%
+ \TransNoteCommon
+
+ This PDF file is optimized for printing, but may easily be
+ recompiled for screen viewing. Please consult the preamble of the
+ \LaTeX\ source file for instructions.
+ }
+}{% If ForPrinting=false, don't skip to recto
+ \renewcommand{\cleardoublepage}{\clearpage}
+ \raggedbottom
+}
+%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
+%%%% End of PRINTING/SCREEN VIEWING code; back to packages %%%%
+%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
+
+\ifthenelse{\boolean{ForPrinting}}{%
+ \setlength{\paperwidth}{8.5in}%
+ \setlength{\paperheight}{11in}%
+ \usepackage[body={4.25in,7.5in},\Margins]{geometry}[2002/07/08]
+}{%
+ \setlength{\paperwidth}{4.5in}%
+ \setlength{\paperheight}{6in}%
+ \usepackage[body={4.25in,5in},\Margins,includeheadfoot]{geometry}[2002/07/08]
+}
+
+\providecommand{\ebook}{00000} % Overridden during white-washing
+\usepackage[pdftex,
+ hyperfootnotes=false,
+ pdftitle={The Project Gutenberg eBook \#\ebook: The tides and kindred phenomena in the solar system},
+ pdfauthor={George Howard Darwin},
+ pdfkeywords={Andrew D. Hwang, Bryan Ness,
+ Project Gutenberg Online Distributed Proofreading Team,
+ Internet Archive/American Libraries},
+ pdfstartview=Fit, % default value
+ pdfstartpage=1, % default value
+ pdfpagemode=UseNone, % default value
+ bookmarks=true, % default value
+ linktocpage=false, % default value
+ pdfpagelayout=\PDFPageLayout,
+ pdfdisplaydoctitle,
+ pdfpagelabels=true,
+ bookmarksopen=true,
+ bookmarksopenlevel=1,
+ colorlinks=true,
+ linkcolor=\HLinkColor]{hyperref}[2007/02/07]
+
+%%%% Fixed-width environment to format PG boilerplate %%%%
+\newenvironment{PGtext}{%
+\begin{alltt}
+\fontsize{8.1}{9.2}\ttfamily\selectfont}%
+{\end{alltt}}
+
+%%%% Global style parameters %%%%
+\makeindex
+
+% Running heads
+\renewcommand{\headrulewidth}{0pt}
+\setlength{\headheight}{15pt}
+
+% Float placement
+\renewcommand{\floatpagefraction}{0.75}
+\renewcommand{\topfraction}{0.95}
+\renewcommand{\bottomfraction}{0.95}
+
+% Misc.
+\widowpenalty=3000
+\setlength{\arraycolsep}{2pt}
+
+% Loosen horizontal spacing
+\setlength{\emergencystretch}{1.5em}
+\newcommand{\Loosen}{\spaceskip 0.375em plus 0.75em minus 0.25em}
+
+% "Scratch pad" for length calculations
+\newlength{\TmpLen}
+
+%% Parametrized vertical space %%
+\newcommand{\Strut}[1][12pt]{\rule{0pt}{#1}}
+
+%%%% Corrections and in-line transcriber's notes %%%%
+% In-line notes
+\newcommand{\DPnote}[1]{}
+% Errors
+\newcommand{\DPtypo}[2]{#2}
+% Stylistic changes
+\newcommand{\DPchg}[2]{#2}
+
+%%%% Running heads %%%%
+\newcommand{\FlushRunningHeads}{%
+ \clearpage
+ \pagestyle{fancy}
+ \fancyhf{}
+ \cleardoublepage
+ \thispagestyle{empty}
+
+ \ifthenelse{\boolean{ForPrinting}}
+ {\fancyhead[RO,LE]{\thepage}}
+ {\fancyhead[R]{\thepage}}
+}
+
+
+\newcommand{\SetCenterHeads}[2][]{%
+ \ifthenelse{\equal{#1}{}}{%
+ \fancyhead[C]{{\small \MakeUppercase{#2}}}%
+ }{%
+ \fancyhead[CE]{{\small \MakeUppercase{#1}}}%
+ \fancyhead[CO]{{\small \MakeUppercase{#2}}}%
+ }%
+}
+
+\newcommand{\BookMark}[2]{%
+ \phantomsection%
+ \pdfbookmark[#1]{#2}{#2}%
+}
+
+%%%% Major document divisions %%%%
+\newcommand{\FrontMatter}{%
+ \cleardoublepage
+ \frontmatter
+ \BookMark{-1}{Front Matter}
+}
+\newcommand{\PGBoilerPlate}{%
+ \pagenumbering{Alph}
+ \pagestyle{empty}
+ \BookMark{0}{PG Boilerplate}
+}
+\newcommand{\MainMatter}{%
+ \FlushRunningHeads
+ \mainmatter
+ \BookMark{-1}{Main Matter}
+}
+\newcommand{\BackMatter}{%
+ \FlushRunningHeads
+ \backmatter
+ \BookMark{-1}{Back Matter}
+}
+\newcommand{\PGLicense}{%
+ \FlushRunningHeads
+ \pagenumbering{Roman}
+ \BookMark{0}{PG License}
+ \SetCenterHeads{License}
+}
+
+\newcommand{\TranscribersNote}[1]{%
+ \begin{minipage}{0.85\textwidth}
+ \small
+ \BookMark{0}{Transcriber's Note}
+ \subsection*{\centering\normalfont\scshape\normalsize\TransNote}
+ #1
+ \end{minipage}
+}
+
+%%%% Table of Contents %%%%
+\newcommand{\Contents}{%
+ \FlushRunningHeads
+ \SetCenterHeads{Contents}
+ \BookMark{0}{Contents}
+ \thispagestyle{empty}
+ \Section{Contents}
+}
+
+\newcommand{\Illustrations}{%
+ \FlushRunningHeads
+ \SetCenterHeads{List of Illustrations}
+ \BookMark{0}{List of Illustrations}
+ \thispagestyle{empty}
+ \Section{List of Illustrations}
+}
+
+% Chapter entries
+\newcommand{\ToCChap}[2]{%
+ \subsection*{\centering\normalfont\small CHAPTER #1}
+ \subsubsection*{\centering\normalfont\footnotesize #2}
+}
+
+\newcommand{\ToCFont}{\small\raggedright}
+
+% Section(s) entries
+\newcommand{\ToCSect}[2]{%
+ \settowidth{\TmpLen}{ #2}%
+ \par\noindent\strut\parbox[b]{\textwidth-\TmpLen}{\ToCFont\hangindent2em%
+ #1~\dotfill}\makebox[\TmpLen][r]{#2}%
+}
+
+% Macro discards 3rd argument (original page number)
+\newcommand{\LoFFig}[3]{%
+%[** TN: Figures 8, 9 bundled together, Fig 13b needs ad hoc handling]
+ \smallskip\noindent\strut\ifthenelse{\equal{#1}{8, 9}}{% Fig 8--9
+ \settowidth{\TmpLen}{ \PgNo[fig]{8}}%
+ \parbox[b]{\textwidth-\TmpLen}{\ToCFont\hangindent3em%
+ \PadTxt[r]{99}{8}, 9.~\textsc{#2} \dotfill}\makebox[\TmpLen][r]{\PgNo[fig]{8}}%
+ }{% Fig 13b
+ \ifthenelse{\equal{#1}{}}{%
+ \settowidth{\TmpLen}{ \PgNo[fig]{13b}}%
+ \parbox[b]{\textwidth-\TmpLen}{\ToCFont\hangindent3em%
+ \PadTxt[r]{99.}{} \textsc{#2} \dotfill}\makebox[\TmpLen][r]{\PgNo[fig]{13b}}%
+ }{% Generic handling
+ \settowidth{\TmpLen}{ \PgNo[fig]{#1}}%
+ \parbox[b]{\textwidth-\TmpLen}{\ToCFont\hangindent3em%
+ \PadTxt[r]{99}{#1}.~\textsc{#2} \dotfill}\makebox[\TmpLen][r]{\PgNo[fig]{#1}}%
+ }%
+ }%
+}
+
+% Page numbers
+% ** Approximate by default (points to original page separators)
+\newcommand{\PgNo}[2][page]{\pageref{#1:#2}}
+
+\newcommand{\PgNos}[3][--]{%
+ \ifthenelse{\equal{\pageref{page:#2}}{\pageref{page:#3}}}{%
+ \PgNo{#2}%
+ }{%
+ \ifthenelse{\equal{#1}{--}}{%
+ \PgNo{#2}#1\PgNo{#3}%
+ }{%
+ \PgNo{#2}\PadTxt[l]{--}{#1}\PgNo{#3}%
+ }%
+ }%
+}
+
+%% Index formatting
+\makeindex
+\makeatletter
+\renewcommand{\@idxitem}{\par\hangindent 30\p@\global\let\idxbrk\nobreak}
+\renewcommand\subitem{\idxbrk\@idxitem \hspace*{12\p@}\let\idxbrk\relax}
+\renewcommand{\indexspace}{\par\penalty-3000 \vskip 10pt plus5pt minus3pt\relax}
+
+\renewenvironment{theindex}{%
+ \setlength\columnseprule{0.5pt}\setlength\columnsep{18pt}%
+ \FlushRunningHeads
+ \SetCenterHeads{Index}
+ \BookMark{0}{Index}
+ \begin{multicols}{2}[\Section{Index}\small]% ** N.B. font size
+ \setlength\parindent{0pt}\setlength\parskip{0pt plus 0.3pt}%
+ \thispagestyle{empty}\let\item\@idxitem\raggedright%
+ }{%
+ \end{multicols}\FlushRunningHeads
+}
+\makeatother
+
+%%%% Document Sectioning %%%%
+\newcommand{\Section}[1]{\section*{\centering\normalfont\MakeUppercase{#1}}}
+
+\newcommand{\Preface}{%
+ \FlushRunningHeads
+ \SetCenterHeads{Preface}
+ \Section{Preface}}
+
+% \Chapter[Running head]{Number}{Heading title}
+\newcommand{\Chapter}[3][]{%
+ \FlushRunningHeads
+ \phantomsection
+ \label{chapter:#2}
+ \thispagestyle{plain}
+ \ifthenelse{\equal{#1}{}}{%
+ \SetCenterHeads[Chapter #2]{#3}%
+ \BookMark{0}{Ch. #2: #3}%
+ }{%
+ \SetCenterHeads[Chapter #2]{#1}%
+ \BookMark{0}{Ch. #2: #1}%
+ }
+ \ChapterSpace
+ \ifthenelse{\equal{#2}{I}}{%
+ \section*{\centering\normalfont\Large THE TIDES}
+ }{}
+ \ifthenelse{\equal{#2}{III}\or\equal{#2}{XVI}}{
+ \section*{\centering\normalfont CHAPTER #2\footnotemark}
+ }{%
+ \section*{\centering\normalfont CHAPTER #2}
+ }
+ \subsection*{\centering\normalfont\small\MakeUppercase{#3}}
+}
+
+\newcommand{\Signature}[2]{%
+ \null\hfill#1\hspace*{\parindent}\\
+ \hspace*{\parindent}#2%
+}
+
+%%%% Misc. textual macros %%%%
+\newcommand{\First}[1]{\textsc{#1}}
+\newcommand{\Title}[1]{\textit{#1}}
+
+\newcommand{\eg}{e.\;g.}
+\newcommand{\ie}{i.\;e.}
+\newcommand{\NB}{N.\;B.}
+
+\renewcommand{\bar}[1]{%
+ \ifthenelse{\equal{#1}{i}}{%
+ \ensuremath{\overline{\PadTxt{ }{\i}}}%
+ }{%
+ \ensuremath{\overline{\PadTxt{r}{#1}}}%
+ }%
+}
+
+\newcommand{\ieme}{\textsuperscript{e}}
+\newcommand{\iere}{\textsuperscript{re}}
+
+% For small caps A.M. and P.M. Usage: 11:30 o'clock, \PM,
+% If followed by a space, use a backslash, e.g. By 6:00~\AM\ the sun will rise.
+\newcommand{\SCAbbrev}[3]{%
+ \ifthenelse{\equal{#3}{.}}%
+ {\textsc{\MakeLowercase{#1.#2}.}}%
+ {\textsc{\MakeLowercase{#1.#2}.}\@#3}%
+}
+
+\newcommand{\AM}[1]{\SCAbbrev{A}{M}{#1}}
+\newcommand{\PM}[1]{\SCAbbrev{P}{M}{#1}}
+
+\newcommand{\AD}[1]{\SCAbbrev{A}{D}{#1}}
+
+%%%% Illustrations %%%%
+% Inclusion wrapper: \Graphics[ext]{width}{filename}
+\newcommand{\Graphic}[3][png]{\includegraphics[width=#2]{./images/#3.#1}}
+
+%\Figure[width]{Figure number}{Caption}{ext}
+\newcommand{\Figure}[4][1.0]{%
+ \begin{figure}[hbtp!]
+ \centering
+ \Graphic[#4]{#1\textwidth}{fig#2}
+ \ifthenelse{\equal{#2}{8}}{%
+ \caption{Fig.~8, 9.---#3}
+ \label{fig:8}\label{fig:9}
+ }{%
+ \ifthenelse{\not\equal{#3}{}}{%
+ \caption{Fig.~#2.---#3}%
+ }{%
+ \caption{Fig.~#2.}%
+ }%
+ \label{fig:#2}
+ }
+ \end{figure}\ignorespaces%
+}
+
+\newcommand{\TallFig}[4][0.95]{%
+ \begin{figure}[hbtp!]
+ \centering
+ \includegraphics[height=#1\textheight]{./images/fig#2.#4}%
+ \caption{Fig.~#2.---#3}%
+ \label{fig:#2}
+ \end{figure}\ignorespaces%
+}
+
+\newenvironment{Authorities}[1][Authorities]{%
+\subsection*{\centering\normalfont\small\scshape #1.}
+\small}{\normalsize}
+
+\newcommand{\Journal}{\medskip}
+\newcommand{\Paper}[1]{\hspace*{\parindent}\hangindent3\parindent#1\par}
+
+\newenvironment{Remark}{\medskip\footnotesize}{\medskip\normalsize}
+
+%%%% Cross-referencing %%%%
+
+%% Anchors
+\newcommand{\PageSep}[1]{\PageLabel{#1}\ignorespaces}
+\newcommand{\PageLabel}[2][page]{\phantomsection\label{#1:#2}}
+
+%% Links
+% Used only once; \PageRef{p.}{329}
+\newcommand{\PageRef}[2]{\hyperref[pg:#2]{#1~\pageref*{pg:#2}}}
+
+\newcommand{\figref}[1]{\hyperref[fig:#1]{#1}}
+\newcommand{\fig}[1]{\hyperref[fig:#1]{fig.~#1}}
+\newcommand{\Fig}[1]{\hyperref[fig:#1]{Fig.~#1}}
+
+% Chapter/appendix reference
+\newcommand{\Ref}[2]{%
+ \ifthenelse{\not\equal{#1}{}}{%
+ \hyperref[chapter:#2]{#1~#2}%
+ }{%
+ \hyperref[chapter:#2]{#2}%
+ }%
+}
+
+%%%% Typographical conveniences %%%%
+\newcommand{\TB}{\bigskip}
+\newcommand{\Seg}[3][]{\ifthenelse{\not\equal{#1}{}}{#1\;}{}#2\;#3}
+
+\newcommand{\Z}{\phantom{0}}
+
+\newcommand{\TEntry}[2][1.25in]{%
+ \parbox[c]{#1}{\centering\footnotesize #2}%
+}
+
+\newcommand{\ColHead}[2][1.25in]{%
+ \TEntry[#1]{\medskip#2\medskip}%
+}
+
+\newcommand{\Dash}{\text{---}}
+\newcommand{\mm}{^{\text{m}}}
+
+% \PadTxt[#1]{#2}{#3} sets #3 in a box of width #2, aligned at #1 (default [c])
+\newcommand{\PadTxt}[3][c]{%
+ \settowidth{\TmpLen}{\text{#2}}%
+ \makebox[\TmpLen][#1]{#3}%
+}
+
+\newcommand{\Ditto}[1]{\PadTxt{#1}{``}}%''
+
+%%%% Misc. mathematical setup %%%%
+\SetSymbolFont{operators}{normal}{T1}{cmsc}{m}{n}
+%% Upright small-cap letters in math mode
+\DeclareMathSymbol{A}{\mathalpha}{operators}{`a}
+\DeclareMathSymbol{B}{\mathalpha}{operators}{`b}
+\DeclareMathSymbol{C}{\mathalpha}{operators}{`c}
+\DeclareMathSymbol{D}{\mathalpha}{operators}{`d}
+\DeclareMathSymbol{E}{\mathalpha}{operators}{`e}
+\DeclareMathSymbol{F}{\mathalpha}{operators}{`f}
+\DeclareMathSymbol{G}{\mathalpha}{operators}{`g}
+\DeclareMathSymbol{I}{\mathalpha}{operators}{`i}
+\DeclareMathSymbol{M}{\mathalpha}{operators}{`m}
+\DeclareMathSymbol{P}{\mathalpha}{operators}{`p}
+\DeclareMathSymbol{V}{\mathalpha}{operators}{`v}
+
+% Handle degree symbols and centered dots as Latin-1 characters
+\DeclareUnicodeCharacter{00A3}{\pounds}
+\DeclareUnicodeCharacter{00B0}{{}^\circ}
+\DeclareUnicodeCharacter{00B1}{\pm}
+\DeclareUnicodeCharacter{00B7}{\cdot}
+\DeclareUnicodeCharacter{00D7}{\times}
+\DeclareUnicodeCharacter{00F7}{\div}
+
+%%%%%%%%%%%%%%%%%%%%%%%% START OF DOCUMENT %%%%%%%%%%%%%%%%%%%%%%%%%%
+\begin{document}
+\FrontMatter
+%%%% PG BOILERPLATE %%%%
+\PGBoilerPlate
+\begin{center}
+\begin{minipage}{\textwidth}
+\small
+\begin{PGtext}
+The Project Gutenberg EBook of The Tides and Kindred Phenomena in the
+Solar System, by Sir George Howard Darwin
+
+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.net
+
+
+Title: The Tides and Kindred Phenomena in the Solar System
+ The Substance of Lectures Delivered in 1897 at the Lowell
+ Institute, Boston, Massachusetts
+
+Author: Sir George Howard Darwin
+
+Release Date: January 31, 2012 [EBook #38722]
+Most recently updated: June 11, 2021
+
+Language: English
+
+Character set encoding: UTF-8
+
+*** START OF THIS PROJECT GUTENBERG EBOOK TIDES AND KINDRED PHENOMENA ***
+\end{PGtext}
+\end{minipage}
+\end{center}
+\clearpage
+
+%%%% Credits and transcriber's note %%%%
+\begin{center}
+\begin{minipage}{\textwidth}
+\begin{PGtext}
+Produced by Andrew D. Hwang, Bryan Ness and the Online
+Distributed Proofreading Team at http://www.pgdp.net (This
+file was produced from images generously made available
+by The Internet Archive/Canadian Libraries)
+\end{PGtext}
+\end{minipage}
+\vfill
+\TranscribersNote{\TransNoteText}
+\end{center}
+%%%%%%%%%%%%%%%%%%%%%%%%%%% FRONT MATTER %%%%%%%%%%%%%%%%%%%%%%%%%%
+\cleardoublepage
+\pagenumbering{roman}
+\PageSep{iii}
+\begin{center}
+\textbf{\Huge THE TIDES}
+\vfil
+\textbf{\large AND KINDRED PHENOMENA IN THE \\[4pt]
+SOLAR SYSTEM}
+\vfil
+{\footnotesize THE SUBSTANCE OF LECTURES DELIVERED \\[4pt]
+IN 1897 AT THE LOWELL INSTITUTE, \\[4pt]
+BOSTON, MASSACHUSETTS}
+\vfil
+BY
+\vfil
+\textbf{\Large GEORGE HOWARD DARWIN}
+\vfil
+\scriptsize PLUMIAN PROFESSOR AND FELLOW OF TRINITY COLLEGE IN THE \\[4pt]
+UNIVERSITY OF CAMBRIDGE
+\vfil\vfil
+\Graphic[png]{1.25in}{riverside}
+\vfil\vfil\vfil
+\footnotesize BOSTON AND NEW YORK \\
+\normalsize HOUGHTON, MIFFLIN AND COMPANY \\
+\textgoth{The Riverside Press, Cambridge} \\
+1899
+\end{center}
+\PageSep{iv}
+\clearpage
+\null\vfill
+\begin{center}
+\footnotesize
+COPYRIGHT, 1898, BY GEORGE HOWARD DARWIN \\
+ALL RIGHTS RESERVED
+\end{center}
+\vfill
+\PageSep{v}
+
+
+\Preface
+
+\First{In} 1897 I delivered a course of lectures on
+the Tides at the Lowell Institute in Boston,
+Massachusetts, and this book contains the substance
+of what I then said. The personal form
+of address appropriate to a lecture is, I think,
+apt to be rather tiresome in a book, and I have
+therefore taken pains to eliminate all traces of
+the lecture from what I have written.
+
+A mathematical argument is, after all, only
+organized common sense, and it is well that men
+of science should not always expound their work
+to the few behind a veil of technical language,
+but should from time to time explain to a larger
+public the reasoning which lies behind their
+mathematical notation. To a man unversed in
+popular exposition it needs a great effort to shell
+away the apparatus of investigation and the
+technical mode of speech from the thing behind
+it, and I owe a debt of gratitude to Mr.~Lowell,
+trustee of the Institute, for having afforded me
+the occasion for making that effort.
+\PageSep{vi}
+
+It is not unlikely that the first remark of
+many who see my title will be that so small a
+subject as the Tides cannot demand a whole volume;
+but, in fact, the subject branches out in
+so many directions that the difficulty has been to
+attain to the requisite compression of my matter.
+Many popular works on astronomy devote a few
+pages to the Tides, but, as far as I know, none
+of these books contain explanations of the practical
+methods of observing and predicting the
+Tides, or give any details as to the degree of
+success attained by tidal predictions. If these
+matters are of interest, I invite my readers not
+to confine their reading to this preface. The
+later chapters of this book are devoted to the
+consideration of several branches of speculative
+Astronomy, with which the theory of the Tides
+has an intimate relationship. The problems involved
+in the origin and history of the solar
+and of other celestial systems have little bearing
+upon our life on the earth, yet these questions
+can hardly fail to be of interest to all those
+whose minds are in any degree permeated by
+the scientific spirit.
+
+I think that there are many who would like to
+understand the Tides, and will make the attempt
+to do so provided the exposition be sufficiently
+\PageSep{vii}
+simple and clear; it is to such readers I address
+this volume. It is for them to say how far I
+have succeeded in rendering these intricate subjects
+interesting and intelligible, but if I have
+failed it has not been for lack of pains.
+
+The figures and diagrams have, for the most
+part, been made by Mr.~Edwin Wilson of Cambridge,
+but I have to acknowledge the courtesy
+of the proprietors of \Title{Harper's}, the \Title{Century},
+and the \Title{Atlantic Monthly} magazines, in supplying
+me with some important illustrations.
+
+A considerable portion of \Ref{Chapter}{III}.\ on the
+``Bore'' is to appear as an article in the \Title{Century
+Magazine} for October, 1898, and the reproductions
+of Captain Moore's photographs of the
+``Bore'' in the Tsien-Tang-Kiang have been
+prepared for that article. The \Title{Century} has also
+kindly furnished the block of Dr.~Isaac Roberts's
+remarkable photograph of the great nebula in
+the constellation of Andromeda; it originally
+appeared in an article on Meteorites in the number
+for October,~1890. The greater portion of
+the text and the whole of the illustrations of
+\Ref{Chapter}{XX}.\ were originally published in \Title{Harper's
+Magazine} for June,~1889. Lastly, portions
+of Chapters \Ref{}{XV}.~and~\Ref{}{XVI}.\ appeared in
+the \Title{Atlantic Monthly} for April, 1898, published
+\PageSep{viii}
+by Messrs.\ Houghton, Mifflin~\&~Co., who also
+make themselves responsible for the publication
+of the American edition of this book.
+
+In conclusion, I wish to take this opportunity
+of thanking my American audience for the cordiality
+of their reception, and my many friends
+across the Atlantic for their abundant hospitality
+and kindness.
+
+\Signature{G. H. DARWIN.}
+{\textsc{Cambridge}, \textit{August}, 1898.}
+\PageSep{ix}
+
+
+\Contents
+
+\ToCChap{I}
+{TIDES AND METHODS OF OBSERVATION}
+
+\ToCSect{Definition of tide}{\PgNos{1}{3}}
+\ToCSect{Oceanic tides}{\PgNos[,]{4}{5}}
+\ToCSect{Methods of observation}{\PgNo{6}}
+\ToCSect{Tide-gauge}{\PgNos{7}{12}}
+\ToCSect{Tide-curve}{\PgNo{12}}
+\ToCSect{Site for tide-gauge}{\PgNo{13}}
+\ToCSect{Irregularities in tide-curve}{\PgNos[,]{14}{15}}
+\ToCSect{Authorities}{\PgNo{16}}
+
+
+\ToCChap{II}
+{SEICHES IN LAKES}
+
+\ToCSect{Meaning of seiche}{\PgNo{17}}
+\ToCSect{Uses of scientific apparatus}{\PgNo{18}}
+\ToCSect{Forel's plemyrameter}{\PgNos[,]{19}{20}}
+\ToCSect{Records of the level of the lake}{\PgNo{20}}
+\ToCSect{Interpretation of record}{\PgNos{21}{23}}
+\ToCSect{Limnimeter}{\PgNo{24}}
+\ToCSect{Mode of oscillation in seiches}{\PgNos{25}{28}}
+\ToCSect{Wave motion in deep and in shallow water}{\PgNos{29}{32}}
+\ToCSect{Composition of waves}{\PgNos{32}{36}}
+\ToCSect{Periods of seiches}{\PgNos[,]{37}{38}}
+\ToCSect{Causes of seiches}{\PgNos[,]{39}{40}}
+\ToCSect{Vibrations due to wind and to steamers}{\PgNos{41}{47}}
+\ToCSect{Aerial waves and their action on lakes and on the sea}{\PgNos{48}{53}}
+\ToCSect{Authorities}{\PgNos[,]{53}{54}}
+\PageSep{x}
+
+
+\ToCChap{III}
+{TIDES IN RIVERS---TIDE MILLS}
+
+\ToCSect{Definition of ebb and flow}{\PgNo{56}}
+\ToCSect{Tidal currents in rivers}{\PgNos{56}{58}}
+\ToCSect{Progressive change of wave in shallow water}{\PgNos[,]{58}{59}}
+\ToCSect{The bore}{\PgNo{59}}
+\ToCSect{Captain Moore's survey of the Tsien-Tang-Kiang}{\PgNos{60}{64}}
+\ToCSect{Diagram of water-levels during the flow}{\PgNos[,]{64}{65}}
+\ToCSect{Chinese superstition}{\PgNos{68}{71}}
+\ToCSect{Pictures of the bore}{\PgNo{69}}
+\ToCSect{Other cases of bores}{\PgNo{71}}
+\ToCSect{Causes of the bore}{\PgNo{72}}
+\ToCSect{Tidal energy}{\PgNos[,]{73}{74}}
+\ToCSect{Tide mills}{\PgNos[,]{74}{75}}
+\ToCSect{Authorities}{\PgNo{75}}
+
+
+\ToCChap{IV}
+{HISTORICAL SKETCH}
+
+\ToCSect{Theories of the Chinese}{\PgNos[,]{76}{77}}
+\ToCSect{Theories of the Arabs}{\PgNos{77}{79}}
+\ToCSect{Theories of the Norsemen}{\PgNos[,]{79}{80}}
+\ToCSect{Writings of Posidonius and Strabo}{\PgNos{80}{84}}
+\ToCSect{Seleucus the Babylonian on the diurnal inequality}{\PgNos[,]{84}{85}}
+\ToCSect{Galileo and Kepler}{\PgNo{85}}
+\ToCSect{Newton and his successors}{\PgNos{86}{88}}
+\ToCSect{Empirical method of tidal prediction}{\PgNos{88}{90}}
+\ToCSect{Authorities}{\PgNo{90}}
+
+
+\ToCChap{V}
+{TIDE-GENERATING FORCE}
+
+\ToCSect{Inertia and centrifugal force}{\PgNos{91}{93}}
+\ToCSect{Orbital motion of earth and moon}{\PgNos{93}{95}}
+\ToCSect{Tide-generating force}{\PgNos{96}{100}}
+\ToCSect{Law of its dependence on the moon's distance}{\PgNos{101}{103}}
+\ToCSect{Earth's rotation}{\PgNos[,]{103}{104}}
+\PageSep{xi}
+\ToCSect{Second explanation of tide-generating force}{\PgNos[,]{104}{105}}
+\ToCSect{Horizontal tide-generating force}{\PgNos[,]{105}{106}}
+\ToCSect{Successive changes of force in the course of a day}{\PgNos[,]{107}{108}}
+\ToCSect{Authorities}{\PgNo{108}}
+
+
+\ToCChap{VI}
+{DEFLECTION OF THE VERTICAL}
+
+\ToCSect{Deflection of a pendulum by horizontal tidal force}{\PgNos{109}{111}}
+\ToCSect{Path pursued by a pendulum under tidal force}{\PgNos{111}{113}}
+\ToCSect{Object of measuring the deflection of a pendulum}{\PgNos{113}{115}}
+\ToCSect{Attempt to measure deflection by bifilar pendulum}{\PgNos{115}{125}}
+\ToCSect{Microsisms}{\PgNos{125}{127}}
+\ToCSect{The microphone as a seismological instrument}{\PgNos{127}{130}}
+\ToCSect{Paschwitz's work with the horizontal pendulum}{\PgNos[,]{131}{132}}
+\ToCSect{Supposed measurement of the lunar deflection of gravity}{\PgNo{132}}
+\ToCSect{Authorities}{\PgNo{133}}
+
+
+\ToCChap{VII}
+{THE ELASTIC DISTORTION OF THE EARTH'S SURFACE BY
+VARYING LOADS}
+
+\ToCSect{Distortion of an elastic surface by superincumbent load}{\PgNos{134}{137}}
+\ToCSect{Application of the theory to the earth}{\PgNos[,]{137}{138}}
+\ToCSect{Effects of tidal load}{\PgNos{138}{140}}
+\ToCSect{Probable deflections at various distances from the coast}{\PgNos{140}{143}}
+\ToCSect{Deflections observed by M.~d'Abbadie and by Dr.~Paschwitz}{\PgNos{143}{145}}
+\ToCSect{Effects of atmospheric pressure on the earth's surface}{\PgNos{145}{147}}
+\ToCSect{Authorities}{\PgNo{148}}
+
+
+\ToCChap{VIII}
+{EQUILIBRIUM THEORY OF TIDES}
+
+\ToCSect{Explanation of the figure of equilibrium}{\PgNos{149}{151}}
+\ToCSect{Map of equilibrium tide}{\PgNos{151}{153}}
+\ToCSect{Tides according to the equilibrium theory}{\PgNos{153}{156}}
+\ToCSect{Solar tidal force compared with lunar}{\PgNos{156}{158}}
+\PageSep{xii}
+\ToCSect{Composition of lunar and solar tides}{\PgNos[,]{158}{159}}
+\ToCSect{Points of disagreement between theory and fact}{\PgNos{159}{162}}
+\ToCSect{Authorities}{\PgNo{162}}
+
+
+\ToCChap{IX}
+{DYNAMICAL THEORY OF THE TIDE WAVE}
+
+\ToCSect{Free and forced waves in an equatorial canal}{\PgNos{163}{165}}
+\ToCSect{Critical depth of canal}{\PgNos{165}{167}}
+\ToCSect{General principle as to free and forced oscillations}{\PgNos{167}{174}}
+\ToCSect{Inverted and direct oscillation}{\PgNos[,]{172}{173}}
+\ToCSect{Canal in latitude~$60°$}{\PgNos[,]{174}{175}}
+\ToCSect{Tides where the planet is partitioned into canals}{\PgNo{175}}
+\ToCSect{Removal of partitions; vortical motion of the water}{\PgNos[,]{176}{177}}
+\ToCSect{Critical latitude where the rise and fall vanish}{\PgNos[,]{177}{178}}
+\ToCSect{Diurnal inequality}{\PgNos{178}{180}}
+\ToCSect{Authorities}{\PgNo{181}}
+
+
+\ToCChap{X}
+{TIDES IN LAKES---COTIDAL CHART}
+
+\ToCSect{The tide in a lake}{\PgNos{182}{185}}
+\ToCSect{The Mediterranean Sea}{\PgNos[,]{185}{186}}
+\ToCSect{Derived tide of the Atlantic}{\PgNos{186}{188}}
+\ToCSect{Cotidal chart}{\PgNos{188}{192}}
+\ToCSect{Authorities}{\PgNo{192}}
+
+
+\ToCChap{XI}
+{HARMONIC ANALYSIS OF THE TIDE}
+
+\ToCSect{Tide in actual oceans due to single equatorial satellite}{\PgNos{193}{196}}
+\ToCSect{Substitution of ideal satellites for the moon}{\PgNos{197}{199}}
+\ToCSect{Partial tide due to each ideal satellite}{\PgNos{199}{201}}
+\ToCSect{Three groups of partial tides}{\PgNo{201}}
+\ToCSect{Semidiurnal group}{\PgNos{201}{204}}
+\ToCSect{Diurnal group}{\PgNos{204}{206}}
+\ToCSect{Meteorological tides}{\PgNos[,]{206}{207}}
+\ToCSect{Shallow water tides}{\PgNos{207}{210}}
+\ToCSect{Authorities}{\PgNo{210}}
+\PageSep{xiii}
+
+
+\ToCChap{XII}
+{REDUCTION OF TIDAL OBSERVATIONS}
+
+\ToCSect{Method of singling out a single partial tide}{\PgNos{211}{214}}
+\ToCSect{Variety of plans adopted}{\PgNos{214}{217}}
+\ToCSect{Tidal abacus}{\PgNos{217}{220}}
+\ToCSect{Authorities}{\PgNo{220}}
+
+
+\ToCChap{XIII}
+{TIDE TABLES}
+
+\ToCSect{Definition of special and general tables}{\PgNo{221}}
+\ToCSect{Reference to moon's transit}{\PgNos[,]{222}{223}}
+\ToCSect{Examples at Portsmouth and at Aden}{\PgNos{223}{228}}
+\ToCSect{General inadequacy of tidal information}{\PgNos[,]{229}{230}}
+\ToCSect{Method of calculating tide tables}{\PgNos{230}{233}}
+\ToCSect{Tide-predicting machine}{\PgNos{233}{241}}
+\ToCSect{Authorities}{\PgNo{241}}
+
+
+\ToCChap{XIV}
+{THE DEGREE OF ACCURACY OF TIDAL PREDICTION}
+
+\ToCSect{Effects of wind and barometric pressure}{\PgNos[,]{242}{243}}
+\ToCSect{Errors at Portsmouth}{\PgNos[,]{243}{244}}
+\ToCSect{Errors at Aden}{\PgNos{245}{249}}
+\ToCSect{Authorities}{\PgNo{250}}
+
+
+\ToCChap{XV}
+{CHANDLER'S NUTATION---THE RIGIDITY OF THE EARTH}
+
+\ToCSect{Nutation of the earth and variation of latitude}{\PgNos{251}{254}}
+\ToCSect{Elasticity of the earth}{\PgNos[,]{254}{255}}
+\ToCSect{Tide due to variation of latitude}{\PgNos[,]{255}{256}}
+\ToCSect{Rigidity of the earth}{\PgNos{256}{260}}
+\ToCSect{Transmission of earthquake shocks}{\PgNos[,]{261}{262}}
+\ToCSect{Authorities}{\PgNos[,]{262}{263}}
+\PageSep{xiv}
+
+
+\ToCChap{XVI}
+{TIDAL FRICTION}
+
+\ToCSect{Friction retards the tide}{\PgNos{264}{267}}
+\ToCSect{Retardation of planetary rotation}{\PgNos{267}{269}}
+\ToCSect{Reaction on the satellite}{\PgNos{269}{272}}
+\ToCSect{Ancient eclipses of the sun}{\PgNos[,]{272}{273}}
+\ToCSect{Law of variation of tidal friction with moon's distance}{\PgNos{273}{275}}
+\ToCSect{Transformations of the month and of the day}{\PgNos{275}{280}}
+\ToCSect{Initial and final conditions of motion}{\PgNos[,]{280}{281}}
+\ToCSect{Genesis of the moon}{\PgNos{281}{285}}
+\ToCSect{Minimum time requisite for the evolution}{\PgNos[,]{285}{286}}
+\ToCSect{Rotation of the moon}{\PgNos[,]{286}{287}}
+\ToCSect{The month ultimately to be shorter than the day}{\PgNos{287}{289}}
+
+
+\ToCChap{XVII}
+{TIDAL FRICTION (\textit{continued})}
+
+\ToCSect{Discovery of the Martian satellites}{\PgNos{290}{298}}
+\ToCSect{Rotation of Mercury, of Venus, and of the Jovian satellites}{\PgNos[,]{298}{299}}
+\ToCSect{Adaptation of the earth's figure to changed rotation}{\PgNos{299}{302}}
+\ToCSect{Ellipticity of the internal strata of the earth}{\PgNos{302}{304}}
+\ToCSect{Geological evidence}{\PgNos{304}{306}}
+\ToCSect{Distortion of a plastic planet and trend of continents}{\PgNos{306}{308}}
+\ToCSect{Obliquity of the ecliptic}{\PgNos{308}{312}}
+\ToCSect{Eccentricity of lunar orbit}{\PgNos[,]{312}{313}}
+\ToCSect{Eccentricity of the orbits of double stars}{\PgNo{313}}
+\ToCSect{Plane of the lunar orbit}{\PgNos[,]{313}{314}}
+\ToCSect{Short summary}{\PgNos[,]{314}{315}}
+\ToCSect{Authorities}{\PgNo{315}}
+
+
+\ToCChap{XVIII}
+{THE FIGURES OF EQUILIBRIUM OF A ROTATING MASS OF
+LIQUID}
+
+\ToCSect{Plateau's experiment}{\PgNos{316}{319}}
+\ToCSect{Stability of a celestial sphere of liquid}{\PgNos{319}{321}}
+\ToCSect{The two ellipsoids of Maclaurin and that of Jacobi}{\PgNos{321}{323}}
+\PageSep{xv}
+\ToCSect{Transitions with change of rotation}{\PgNos[,]{323}{324}}
+\ToCSect{Coalescence of Jacobi's with Maclaurin's ellipsoid}{\PgNos{324}{326}}
+\ToCSect{Poincaré's law of stability and coalescence}{\PgNos[,]{326}{327}}
+\ToCSect{Poincaré's pear-shaped figure}{\PgNos[,]{327}{328}}
+\ToCSect{Hour-glass figure of equilibrium}{\PgNos{328}{332}}
+\ToCSect{Figures of planets}{\PgNos[,]{332}{333}}
+\ToCSect{Authorities}{\PgNo{333}}
+
+
+\ToCChap{XIX}
+{THE EVOLUTION OF CELESTIAL SYSTEMS}
+
+\ToCSect{The Nebular Hypothesis}{\PgNos{334}{338}}
+\ToCSect{Nebula in Andromeda}{\PgNos[,]{338}{339}}
+\ToCSect{Distribution of satellites in the solar system}{\PgNos{339}{341}}
+\ToCSect{Genesis of celestial bodies by fission}{\PgNo{342}}
+\ToCSect{Dr.~See's speculations as to systems of double stars}{\PgNos{342}{344}}
+\ToCSect{Diversity of celestial bodies}{\PgNos{344}{346}}
+\ToCSect{Authorities}{\PgNo{346}}
+
+
+\ToCChap{XX}
+{SATURN'S RINGS}
+
+\ToCSect{Description}{\PgNos[,]{347}{348}}
+\ToCSect{Discovery of Saturn's rings}{\PgNos{348}{352}}
+\ToCSect{Diagram of the rings}{\PgNos{353}{356}}
+\ToCSect{Roche's investigation}{\PgNos[,]{356}{357}}
+\ToCSect{Roche's limit}{\PgNos{358}{360}}
+\ToCSect{The limit for the several planets}{\PgNos[,]{360}{361}}
+\ToCSect{Meteoric constitution of Saturn's rings}{\PgNos[,]{361}{362}}
+\ToCSect{Maxwell's investigations}{\PgNos{362}{367}}
+\ToCSect{Spectroscopic examination of the rings}{\PgNos{367}{369}}
+\ToCSect{Authorities}{\PgNo{369}}
+\PageSep{xvi}
+%[Blank Page]
+\PageSep{xvii}
+
+
+% [** TN: Plates (Figs. 23, 33, and 40) listed separately in the original;
+% omitted separate "FULL-PAGE" and "IN TEXT" headings, folded plate
+% references into main illustration list.]
+% LIST OF ILLUSTRATIONS
+\Illustrations
+
+\LoFFig{1}{Well for Tide-Gauge}{7}
+\LoFFig{2}{Pipe of Tide-Gauge}{9}
+\LoFFig{3}{Indian Tide-Gauge}{10}
+\LoFFig{4}{Légé's Tide-Gauge}{11}
+\LoFFig{5}{Bombay Tide-Curve from Noon, April 22, to Noon, April 30, 1884}{14}
+\LoFFig{6}{Sites for a Tide-Gauge}{15}
+\LoFFig{7}{Plemyrameter}{20}
+\LoFFig{8, 9}{Records of Seiches at Évian}{23}
+\LoFFig{10}{Map of Lake of Geneva}{26}
+\LoFFig{11}{Wave in Deep Water}{30}
+\LoFFig{12}{Wave in Shallow Water}{31}
+\LoFFig{13}{Simple Wave}{33}
+\LoFFig{}{Composition of Two Equal and Opposite Waves}{34}
+\LoFFig{14}{Vibrations due to Steamers}{45}
+\LoFFig{15}{Progressive Change of a Wave in Shallow Water}{59}
+\LoFFig{16}{Chart of the Estuary of the Tsien-Tang-Kiang}{61}
+\LoFFig{17}{Bore-Shelter on the Tsien-Tang-Kiang}{64}
+\LoFFig{18}{Diagram of the Flow of the Tide on the Tsien-Tang-Kiang}{66}
+\LoFFig{19}{Pictures of the Bore on the Tsien-Tang-Kiang}{69}
+\LoFFig{20}{Earth and Moon}{93}
+\PageSep{xviii}
+\LoFFig{21}{Revolution of a Body without Rotation}{98}
+\LoFFig{22}{Tide-generating Force}{100}
+\LoFFig{23}{Horizontal Tide-generating Force}{106}
+\LoFFig{24}{Deflection of a Pendulum; the Moon and Observer on the Equator}{111}
+\LoFFig{25}{Deflection of a Pendulum; the Moon in N. Declination~$15°$,
+the Observer in N. Latitude~$30°$}{112}
+\LoFFig{26}{Bifilar Pendulum}{115}
+\LoFFig{27}{Form of Dimple in an Elastic Surface}{135}
+\LoFFig{28}{Distortion of Land and Sea-Bed by Tidal Load}{139}
+\LoFFig{29}{Chart of Equilibrium Tides}{152}
+\LoFFig{30}{Forced Oscillations of a Pendulum}{171}
+\LoFFig{31}{The Tide in a Lake}{184}
+\LoFFig{32}{Chart of Cotidal Lines}{190}
+\LoFFig{33}{Tidal Abacus}{218}
+\LoFFig{34}{Curves of Intervals and Heights at Portsmouth and at Aden}{227}
+\LoFFig{35}{Diagram of Tide-predicting Instrument}{235}
+\LoFFig{36}{Frictionally Retarded Tide}{266}
+\LoFFig{37}{Maclaurin's and Jacobi's Ellipsoids of Equilibrium}{323}
+\LoFFig{38}{Figures of Equilibrium}{325}
+\LoFFig{39}{Hour-glass Figure of Equilibrium}{329}
+\LoFFig{40}{Nebula in Andromeda}{339}
+\LoFFig{41}{The Planet Saturn}{349}
+\LoFFig{42}{Diagram of Saturn and his Rings}{353}
+\LoFFig{43}{Roche's Figure of a Satellite when elongated to the utmost}{357}
+
+\MainMatter
+\PageSep{1}
+\index{Barometric pressure|see{Atmospheric pressure}}%
+\index{Bifilar|see{Pendulum}}%
+\index{D'Abbadie|see{Abbadie}}%
+\index{Horizontal tide-generating force|see{also Pendulum}}%
+\index{Rebeur|see{Paschwitz}}%
+\index{Thomson, Sir W.|see{Kelvin}}%
+\index{Tidal problem|see{Laplace, Harmonic Analysis, etc.}}%
+\index{Tide|see{also other headings; \eg\ for tide-generating force \textit{see} Force}}%
+\index{Vertical|see{Deflection}}%
+% [** TN: Text printed by \Chapter macro]
+% THE TIDES
+
+
+\Chapter{I}
+{Tides and Methods of Observation}
+
+\First{The} great wave caused by an earthquake is
+\index{Tide!definition|(}%
+often described in the newspapers as a tidal
+wave, and the same name is not unfrequently
+applied to such a short series of enormous waves
+as is occasionally encountered by a ship in the
+open sea. We must of course use our language
+in the manner which is most convenient, but as
+in this connection the adjective ``tidal'' implies
+simply greatness and uncommonness, the use of
+the term in such a sense cannot be regarded as
+appropriate.
+
+The word ``tidal'' should, I think, only be
+used when we are referring to regular and persistent
+alternations of rise and fall of sea-level.
+Even in this case the term may perhaps be used
+in too wide a sense, for in many places there is a
+regular alternation of the wind, which blows in-shore
+during the day and out during the night
+with approximate regularity, and such breezes
+\PageSep{2}
+alternately raise and depress the sea-level, and
+thus produce a sort of tide. Then in the Tropics
+there is a regularly alternating, though small,
+periodicity in the pressure of the atmosphere,
+which is betrayed by an oscillation in the height
+of the barometer. Now the ocean will respond
+to the atmospheric pressure, so that the sea-level
+will fall with a rising barometer, and rise
+with a falling barometer. Thus a regularly
+periodic rise and fall of the sea-level must result
+from this cause also. Again, the melting of the
+snows in great mountain ranges, and the annual
+variability in rainfall and evaporation, produce
+approximately periodic changes of level in the
+estuaries of rivers, and although the period of
+these changes is very long, extending as they do
+over the whole year, yet from their periodicity
+they partake of the tidal character.
+
+These changes of water level are not, however,
+tides in the proper sense of the term, and a true
+tide can only be adequately defined by reference
+to the causes which produce it. A tide, in fact,
+means a rising and falling of the water of the
+ocean caused by the attractions of the sun and
+moon.
+
+Although true tides are due to astronomical
+causes, yet the effects of regularly periodic winds,
+variation of atmospheric pressure, and rainfall
+are so closely interlaced with the true tide that
+in actual observation of the sea it is necessary to
+\PageSep{3}
+consider them both together. It is accordingly
+practically convenient to speak of any regular
+alternation of sea-level, due to the wind and to
+the other influences to which I have referred, as
+a Meteorological Tide. The addition of the adjective
+``meteorological'' justifies the use of the
+term ``tide'' in this connection.
+
+We live at the bottom of an immense sea of
+air, and if the attractions of the sun and moon
+affect the ocean, they must also affect the air.
+This effect will be shown by a regular rise and
+fall in the height of the barometer. Although
+such an effect is undoubtedly very small, yet it
+is measurable. The daily heating of the air by
+the sun, and its cooling at night, produce marked
+alternations in the atmospheric pressure, and this
+effect may by analogy be called an atmospheric
+meteorological tide.
+
+The attractions of the moon and sun must certainly
+act not only on the sea, but also on the
+solid earth; and, since the earth is not perfectly
+rigid or stiff, they must produce an alternating
+change in its shape. Even if the earth is now
+so stiff that the changes in its shape escape
+detection through their minuteness, yet such
+changes of shape must exist. There is much
+evidence to show that in the early stages of their
+histories the planets consisted largely or entirely
+of molten rock, which must have yielded to tidal
+influences. I shall, then, extend the term ``tide''
+\index{Tide!definition|)}%
+\PageSep{4}
+so as to include such alternating deformations of
+a solid and elastic, or of a molten and plastic,
+globe. These corporeal tides will be found to
+lead us on to some far-reaching astronomical
+speculations. The tide, in the sense which I
+\index{Tide!general description|(}%
+have attributed to the term, covers a wide field
+of inquiry, and forms the subject of the present
+volume.
+
+I now turn to the simplest and best known
+form of tidal phenomena. When we are at the
+seashore, or on an estuary, we see that the water
+rises and falls nearly twice a day. To be more
+exact, the average interval from one high water
+to the next is twelve hours twenty-five minutes,
+and so high water falls later, according to the
+clock, by twice twenty-five minutes, or by fifty
+minutes, on each successive day. Thus if high
+water falls to-day at noon, it will occur to-morrow
+at ten minutes to one. Before proceeding, it
+may be well to remark that I use high water and
+low water as technical terms. In common parlance
+the level of water may be called high or
+low, according as whether it is higher or lower
+than usual. But when the level varies periodically,
+there are certain moments when it is highest
+and lowest, and these will be referred to as
+the times of high and low water, or of high and
+of low tide. In the same way I shall speak of
+the heights at high and low water, as denoting
+the water-level at the moments in question.
+\PageSep{5}
+
+The most elementary observations would show
+that the time of high water has an intimate relationship
+to the moon's position. The moon, in
+fact, passes the meridian on the average fifty
+minutes later on each succeeding day, so that if
+high water occurs so many hours after the moon
+is due south on any day, it will occur on any
+other day about the same number of hours after
+the moon was south. This rule is far from being
+exact, for it would be found that the interval
+from the moon's passage to high water differs
+considerably according to the age of the moon.
+I shall not, however, attempt to explain at present
+how this rough rule as to the time of high
+water must be qualified, so as to convert it into
+an accurate statement.
+
+But it is not only the hour of high water which
+changes from day to day, for the height to which
+the water rises varies so conspicuously that the
+fact could not escape the notice of even the most
+casual observer. It would have been necessary
+to consult a clock to discover the law by which
+the hour of high water changes from day to day;
+but at the seashore it would be impossible to
+avoid noticing that some rocks or shoals which
+are continuously covered by the sea at one part
+of a fortnight are laid bare at others. It is, in
+fact, about full and new moon that the range
+from low to high water is greatest, and at the
+moon's first or third quarter that the range is
+\PageSep{6}
+least. The greater tides are called ``springs,''
+and the smaller ``neaps.''
+
+The currents produced in the sea by tides are
+\index{Gauge, tide!description of|(}%
+often very complicated where the open sea is
+broken by islands and headlands, and the knowledge
+of tidal currents at each place is only to be
+gained by the practical experience of the pilot.
+Indeed, in the language of sailors, the word
+``tide'' is not unfrequently used as meaning
+tidal current, without reference to rise and fall.
+These currents are often of great violence, and
+vary from hour to hour as the water rises and
+falls, so that the pilot requires to know how the
+water stands in-shore in order to avail himself of
+his practical knowledge of how the currents will
+make in each place. A tide table is then of
+\index{Tide!general description|)}%
+much use, even at places where the access to a
+harbor is not obstructed by a bar or shoal. It
+is, of course, still more important for ships to
+have a correct forecast of the tides where the
+entrance to the harbor is shallow.
+
+I have now sketched in rough outline some
+of the peculiarities of the tides, and it will have
+become clear that the subject is a complicated
+one, not to be unraveled without regular observation.
+\index{Observation!methods of tidal|(}%
+I shall, therefore, explain how tides are
+observed scientifically, and how the facts are collected
+upon which the scientific treatment of the
+tides is based.
+
+The rise and fall of the sea may, of course, be
+\PageSep{7}
+roughly estimated by observing the height of the
+water on posts or at jetties, which jut out into
+moderately deep water. But as the sea is continually
+disturbed by waves, observations of this
+kind are not susceptible of accuracy, and for
+\Figure{1}{Well for Tide-Gauge}{png}
+scientific purposes more elaborate apparatus is
+required. The exact height of the water can
+only be observed in a place to which the sea has
+a moderately free access, but where the channel
+is so narrow as to prevent the waves from sensibly
+disturbing the level of the water. This result
+\PageSep{8}
+is obtained in a considerable variety of ways,
+but one of them may be described as typical of
+all.
+
+A well (\fig{1}) about two feet in diameter is
+dug to a depth of several feet below the lowest
+tide and in the neighborhood of deep water. The
+well is lined with iron, and a two-inch iron pipe
+runs into the well very near its bottom, and passes
+down the shore to the low-water line. Here it is
+joined to a flexible pipe running out into deep
+water, and ending with a large rose pierced with
+many holes, like that of a watering can. The
+rose (\fig{2}) is anchored to the bottom of the
+sea, and is suspended by means of a buoy, so as
+to be clear of the bottom. The tidal water can
+thus enter pretty freely into the well, but the
+passage is so narrow that the wave motion is not
+transmitted into the well. Inside the well there
+floats a water-tight copper cylinder, weighted at
+the bottom so that it floats upright, and counter-poised
+so that it only just keeps its top clear of
+the water. To the top of the float there is fastened
+a copper tape or wire, which runs up to
+the top of the well and there passes round a
+wheel. Thus as the water rises and falls this
+wheel turns backwards and forwards.
+
+It is hardly necessary to describe in detail the
+simple mechanism by which the turning of this
+wheel causes a pencil to move backwards and forwards
+in a straight line. The mechanism is,
+\PageSep{9}
+however, such that the pencil moves horizontally
+backwards and forwards by exactly the same
+amount as the water rises or falls in the well; or,
+\Figure{2}{Pipe of Tide-Gauge}{png}
+if the rise and fall of the tide is considerable,
+the pencil only moves by half as much, or one
+third, or even one tenth as much as the water.
+At each place a scale of reduction is so chosen
+as to bring the range of motion of the pencil
+within convenient limits. We thus have a pencil
+which will draw the rise and fall of the tide
+on the desired scale.
+
+It remains to show how the times of the rise
+and fall are indicated. The end of the pencil
+touches a sheet of paper which is wrapped round
+a drum about five feet long and twenty-four
+\PageSep{10}
+\index{Curve, tide!irregularities in|(}%
+inches in circumference. If the drum were kept
+still the pencil would simply draw a straight line
+to and fro along the length of the drum as the
+water rises and falls. But the drum is kept
+turning by clockwork, so that it makes exactly
+one revolution in twenty-four hours. Since the
+drum is twenty-four inches round, each inch of
+circumference corresponds to one hour. If the
+water were at rest the pencil would simply draw
+a circle round the paper, and the beginning and
+ending of the line would join, whilst if the drum
+remained still and the water moved, the pencil
+\Figure{3}{Indian Tide-Gauge}{jpg}
+would draw a straight line along the length of
+the cylinder; but when both drum and water
+are in motion, the pencil draws a curve on the
+cylinder from which the height of water may be
+read off at any time in each day and night. At
+the end of twenty-four hours the pencil has returned
+to the same part of the paper from which
+\PageSep{11}
+\Figure[0.7]{4}{Légé's Tide-Gauge}{jpg}
+\index{Gauge, tide!description of|)}%
+\PageSep{12}
+it started, and it might be thought that there
+would be risk of confusion between the tides of
+to-day and those of yesterday. But since to-day
+the tides happen about three quarters of an hour
+later than yesterday, it is found that the lines
+keep clear of one another, and, in fact, it is
+usual to allow the drum to run for a fortnight
+before changing the paper, and when the old
+sheet is unwrapped from the drum, there is
+written on it a tidal record for a fortnight.
+
+The instrument which I have described is
+called a ``tide-gauge,'' and the paper a ``tide-curve.''
+As I have already said, tide-gauges
+may differ in many details, but this description
+will serve as typical of all. Another form of
+tide-gauge is shown in \fig{4}; here a continuous
+sheet of paper is placed over the drum, so that
+there is no crossing of the curves, as in the first
+example. Yet another form, designed by Lord
+Kelvin, is shown on p.~170 of vol.~iii.\ of his
+``Popular Lectures.''
+
+The actual record for a week is exhibited in
+\index{Curve, tide!at Bombay}%
+\fig{5}, on a reduced scale. This tide-curve was
+drawn at Bombay by a tide-gauge of the pattern
+first described. When the paper was wrapped
+on the drum, the right edge was joined to the
+left, and now that it is unwrapped the curve
+must be followed out of the paper on the left
+and into it again on the right. The figure
+shows that spring tide occurred on April~26,
+\PageSep{13}
+1884; the preceding neap tide was on the~18th,
+and is not shown. It may be noticed that the
+law of the tide is conspicuously different from
+that which holds good on the coast of England,
+for the two successive high or low waters which
+occur on any day have very different heights.
+Thus, for example, on April~26 low water occurred
+at 5.50~\PM, and the water fell to $5$~ft.\
+$2$~in., whereas the next low water, occurring at
+5.45~\AM\ of the~27th, fell to $1$~ft.\ $3$~in., the
+heights being in both cases measured from a
+certain datum. When we come to consider the
+theory of the tides the nature of this irregularity
+will be examined.
+
+The position near the seashore to be chosen
+for the erection of the tide-gauge is a matter of
+much importance. The choice of a site is generally
+limited by nature, for it should be near
+the open sea, should be sheltered from heavy
+weather, and deep water must be close at hand
+even at low tide.
+
+In the sketch map shown in \fig{6} a site such
+as~$A$ is a good one when the prevailing wind
+blows in the direction of the arrow. A position
+such as~$B$, although well sheltered from heavy
+seas, is not so good, because it is found that
+tide-curves drawn at~$B$ would be much zigzagged.
+These zigzags appear in the Bombay curves,
+although at Bombay they are usually very
+smooth ones.
+\PageSep{14}
+
+These irregularities in the tide-curve are not
+due to tides, and as the object of the observation
+\index{Observation!methods of tidal|)}%
+is to determine the nature of the tides it is
+\Figure{5}{Bombay Tide-Curve from Noon, April~22,
+to Noon, April~30, 1884}{jpg}
+desirable to choose a site for the gauge where
+\index{Gauge, tide!site for}%
+the zigzags shall not be troublesome; but it is
+not always easy to foresee the places which will
+furnish smooth tide-curves.
+
+Most of us have probably at some time or
+other made a scratch on the sand by the seashore,
+\PageSep{15}
+and watched the water rise over it. We
+generally make our mark on the sand at the
+furthest point, where the wash of a rather large
+wave has brought up the water. For perhaps
+five or ten minutes no wave brings the water up
+as far as the mark, and one begins to think that
+it was really an extraordinarily large wave which
+was marked, although it did not seem so at the
+time. Then a wave brings up the water far over
+the mark, and immediately all the waves submerge
+it. This little observation simply points
+\Figure{6}{Sites for a Tide-Gauge}{png}
+to the fact that the tide is apt to rise by jerks,
+and it is this irregularity of rise and fall which
+marks the notches in the tide-curves to which I
+have drawn attention.
+
+Now in scientific matters it is well to follow
+up the clues afforded by such apparently insignificant
+facts as this. An interesting light is
+indeed thrown on the origin of these notches on
+tide-curves by an investigation, not very directly
+\PageSep{16}
+connected with tidal observation, on which I shall
+make a digression in the following chapter.
+\index{Curve, tide!irregularities in|)}%
+
+\begin{Authorities}
+Baird's \Title{Manual for Tidal Observations} (Taylor \& Francis,
+\index{Baird, \Title{Manual for Tidal Observation}}%
+1886). Price 7\textit{s.}~6\textit{d.} Figs.\ \figref{1},~\figref{2},~\figref{3},~\figref{6} are reproduced from this
+work.
+
+The second form of tide-gauge shown in \fig{4} is made by
+Messrs.\ Légé, and is reproduced from a woodcut kindly provided
+by them.
+
+Sir William Thomson's (Lord Kelvin's) \Title{Popular Lectures and
+Addresses}, vol.~iii. (Macmillan,~1891).
+\end{Authorities}
+\PageSep{17}
+
+
+\Chapter{II}
+{Seiches in Lakes}
+
+\First{It} has been known for nearly three centuries
+\index{Lakes!seiches in|(}%
+\index{Seiches!definition}%
+that the water of the Lake of Geneva is apt to
+\index{Geneva!seiches in lake|(}%
+rise and fall by a few inches, sometimes irregularly
+and sometimes with more or less regularity;
+and the same sort of oscillation has been observed
+in other Swiss lakes. These quasi-tides,
+called seiches, were until recently supposed only
+to occur in stormy weather, but it is now known
+that small seiches are of almost daily occurrence.\footnote
+ {The word ``seiche'' is a purely local one. It has been alleged
+ to be derived from ``sèche,'' but I can see no reason for
+ associating dryness with the phenomenon.}
+
+Observations were made by Vaucher in the
+\index{Vaucher, record of a great seiche at Geneva}%
+last century on the oscillations of the Lake of
+Geneva, and he gave an account of a celebrated
+seiche in the year~1600, when the water oscillated
+through three or four feet; but hardly any
+systematic observation had been undertaken when
+Professor Forel, of Lausanne, attacked the subject,
+\index{Forel!on seiches|(}%
+and it is his very interesting observations
+which I propose to describe.
+
+Doctor Forel is not a mathematician, but is
+\PageSep{18}
+rather a naturalist of the old school, who notes any
+interesting fact and then proceeds carefully to
+investigate its origin. His papers have a special
+charm in that he allows one to see all the workings
+of his mind, and tells of each difficulty as it
+arose and how he met it. To those who like to
+read of such work, almost in the form of a narrative,
+I can strongly recommend these papers,
+which afford an admirable example of research
+thoroughly carried out with simple appliances.
+
+People are nowadays too apt to think that
+science can only be carried to perfection with
+elaborate appliances, and yet it is the fact that
+many of the finest experiments have been made
+with cardboard, cork, and sealing-wax. The principal
+reason for elaborate appliances in the laboratories
+of universities is that a teacher could not
+deal with a large number of students if he had
+to show each of them how to make and set up
+his apparatus, and a student would not be able
+to go through a large field of study if he had to
+spend days in preparation. Great laboratories
+have, indeed, a rather serious defect, in that they
+tend to make all but the very best students helpless,
+and thus to dwarf their powers of resource
+and inventiveness. The mass of scientific work
+is undoubtedly enormously increased by these
+institutions, but the number of really great investigators
+seems to remain almost unaffected by
+them. But I must not convey the impression
+\PageSep{19}
+that, in my opinion, great laboratories are not
+useful. It is obvious, indeed, that without them
+science could not be taught to large numbers of
+students, and, besides, there are many investigations
+in which every possible refinement of apparatus
+is necessary. But I do say that the
+number of great investigators is but little increased
+by laboratories, and that those who are
+interested in science, but yet have not access to
+laboratories, should not give up their study in
+despair.
+
+Doctor Forel's object was, in the first instance,
+to note the variations of the level of the lake, after
+obliterating the small ripple of the waves on the
+surface. The instrument used in his earlier investigations
+was both simple and delicate. Its principle
+was founded on casual observation at the port
+of Morges, where there happens to be a breakwater,
+pierced by a large ingress for ships and a small
+one for rowing boats. He accidentally noticed
+that at the small passage there was always a current
+setting either inwards or outwards, and it
+occurred to him that such a current would form
+a very sensitive index of the rise and fall of the
+water in the lake. He therefore devised an instrument,
+illustrated in \fig{7}, and called by him
+a plemyrameter, for noting currents of even the
+\index{Plemyrameter, observation of seiches with|(}%
+most sluggish character. Near the shore he
+made a small tank, and he connected it with the
+lake by means of an india-rubber siphon pipe of
+\PageSep{20}
+small bore. Where the pipe crossed the edge
+of the tank he inserted a horizontal glass tube
+of seven millimetres diameter, and in that tube
+he put a float of cork, weighted with lead so that
+it should be of the same density as water. At
+the ends of the glass tube there were stops, so
+that the float could not pass out of it. When
+the lake was higher than the tank, the water ran
+\Figure{7}{Plemyrameter}{png}
+through the siphon pipe from lake to tank, and
+the float remained jammed in the glass tube
+against the stop on the side towards the tank;
+and when the lake fell lower than the tank, the
+float traveled slowly to the other end and remained
+there. The siphon pipe being small, the
+only sign of the waves in the lake was that the
+float moved with slight jerks, instead of uniformly.
+Another consequence of the smallness
+of the tube was that the amount of water which
+could be delivered into the tank or drawn out of
+it in one or two hours was so small that it might
+\PageSep{21}
+practically be neglected, so that the water level
+in the tank might be considered as invariable.
+
+This apparatus enabled Forel to note the rise
+and fall of the water, and he did not at first
+attempt to measure the height of rise and fall,
+as it was the periodicity in which he was principally
+interested.
+
+In order to understand the record of observations,
+\index{Seiches!records of}%
+it must be remembered that when the
+float is towards the lake, the water in the tank
+stands at the higher level, and when the float
+is towards the tank the lake is the higher. In
+the diagrams, of which \fig{8} is an example, the
+straight line is divided into a scale of hours and
+minutes. The zigzag line gives the record, and
+the lower portions represent that the water of
+the lake was below the tank, and the upper line
+that it was above the tank. The fact that the
+float only moved slowly across from end to end
+of the glass tube, is indicated by the slope of
+the lines, which join the lower and upper portions
+of the zigzags. Then on reading \fig{8} we
+see that from $2$~hrs.\ $1$~min.\ to $2$~hrs.\ $4$~mins.\ the
+water was high and the float was jammed against
+the tank end of the tube, because there was a
+current from the lake to the tank. The float
+then slowly left the tank end and traveled
+across, so that at $2$~hrs.\ $5$~mins.\ the water was
+low in the lake. It continued, save for transient
+changes of level, to be low until $2$~hrs.\ $30$~mins.,
+\PageSep{22}
+when it rose again. Further explanation seems
+unnecessary, as it should now be easy to read
+this diagram, and that shown in~\fig{9}.
+\index{Plemyrameter, observation of seiches with|)}%
+
+The sharp pinnacles indicate alternations of
+level so transient that the float had not time to
+travel across from one end of the glass tube
+to the other, before the current was reversed.
+These pinnacles may be disregarded for the
+present, since we are only considering seiches of
+considerable period.
+
+These two diagrams are samples of hundreds
+which were obtained at various points on the
+shores of Geneva, and of other lakes in Switzerland.
+In order to render intelligible the method
+by which Forel analyzed and interpreted these
+records, I must consider \fig{8} more closely. In
+this case it will be noticed that the record shows
+a long high water separated from a long low
+water by two pinnacles with flat tops. These
+pieces at the ends have an interesting significance.
+When the water of the lake is simply
+oscillating with a period of about an hour we
+have a trace of the form shown in~\fig{9}. But
+when there exists concurrently with this another
+oscillation, of much smaller range and of short
+period, the form of the trace will be changed.
+When the water is high in consequence of the
+large and slow oscillation, the level of the lake
+cannot be reduced below that of the tank by
+the small short oscillation, and the water merely
+\PageSep{23}
+stands a little higher or a little lower, but always
+remains above the level of the tank, so that the
+trace continues on the higher level. But when,
+in course of the changes of the large oscillation,
+the water has sunk to near the mean level of the
+%[** TN: Figures 8 and 9 combined; cross-refs handled by preamble code]
+\Figure{8}{Records of Seiches at Évian}{png}
+lake, the short oscillation will become manifest,
+and so it is only at the ends of the long flat
+pieces that we shall find evidence of the quick
+oscillation.
+
+Thus, in these two figures there was in one
+case only one sort of wave, and in the other
+there were two simultaneous waves. These
+records are amongst the simplest of those obtained
+by Forel, and yet even here the oscillations
+of the water were sufficiently complicated.
+It needed, indeed, the careful analysis of many
+records to disentangle the several waves and to
+determine their periods.
+
+After having studied seiches with a plemyrameter
+for some time, Forel used another form of
+\PageSep{24}
+apparatus, by which he could observe the amplitude
+of the waves as well as their period. His
+apparatus was, in fact, a very delicate tide-gauge,
+which he called a limnimeter. The only
+\index{Limnimeter, a form of tide gauge}%
+difference between this instrument and the one
+already described as a tide-gauge is that the
+drum turned much more rapidly, so that five
+feet of paper passed over the drum in twenty-four
+hours, and that the paper was comparatively
+narrow, the range of the oscillation being
+small. The curve was usually drawn on the full
+scale, but it could be quickly reduced to half
+scale when large seiches were under observation.
+
+It would be impossible in a book of this kind
+\index{Lakes!mode of rocking in seiches}%
+to follow Forel in the long analysis by which he
+interpreted his curves. He speaks thus of the
+complication of simultaneous waves: ``All these
+oscillations are embroidered one on the other
+and interlace their changes of level. There is
+here matter to disturb the calmest mind. I
+must have a very stout faith in the truth of my
+hypothesis to persist in maintaining that, in the
+midst of all these waves which cross and mingle,
+there is, nevertheless, a recognizable rhythm.
+This is, however, what I shall try to prove.''\footnote
+ {\Title{Deuxième Étude}, p.~544.}
+The hypothesis to which he here refers, and
+triumphantly proves, is that seiches consist of a
+rocking of the whole water of the lake about
+fixed lines, just as by tilting a trough the water
+\PageSep{25}
+in it may be set swinging, so that the level at
+the middle remains unchanged, while at the two
+ends the water rises and falls alternately.
+
+In another paper he remarks: ``If you will
+\index{Lakes!mode of rocking in seiches}%
+follow and study with me these movements you
+will find a great charm in the investigation.
+When I see the water rising and falling on the
+shore at the end of my garden I have not before
+me a simple wave which disturbs the water of
+the bay of Morges, but I am observing the manifestation
+of a far more important phenomenon.
+It is the whole water of the lake which is rocking.
+It is a gigantic impulse which moves the
+whole liquid mass of Leman throughout its
+length, breadth, and depth\dots. It is probable
+that the same thing would be observed in far
+larger basins of water, and I feel bound to
+recognize in the phenomenon of seiches the
+\index{Seiches!longitudinal and transverse|(}%
+grandest oscillatory movement which man can
+study on the face of our globe.''\footnote
+ {\Title{Les Seiches, Vagues d'Oscillation}, p.~11.}
+
+It will now be well to consider the map of
+Geneva in~\fig{10}. Although the lake somewhat
+resembles the arc of a circle, the curvature of
+its shores will make so little difference in the
+nature of the swinging of the water that we
+may, in the first instance, consider it as practically
+straight.
+
+Forel's analysis of seiches led him to conclude
+that the oscillations were of two kinds, the longitudinal
+\PageSep{26}
+and the transverse. In the longitudinal
+seiche the water rocks about a line drawn across
+the lake nearly through Morges, and the water
+at the east end of the lake rises when that at
+the west falls, and vice versa. The line about
+which the water rocks is called a node, so that in
+this case there is one node at the middle of the
+lake. This sort of seiche is therefore called a
+uninodal longitudinal seiche. The period of the
+\Figure{10}{Map of Lake of Geneva}{png}
+oscillation is the time between two successive
+high waters at any place, and it was found to be
+seventy-three minutes, but the range of rise and
+fall was very variable. There are also longitudinal
+seiches in which there are two nodes,
+dividing the lake into three parts, of which the
+central one is twice as long as the extreme parts;
+such an oscillation is called a binodal longitudinal
+seiche. In this mode the water at the middle
+\PageSep{27}
+of the lake is high when that at the two ends
+is low, and vice versa; the period is thirty-five
+minutes.
+
+Other seiches of various periods were observed,
+\index{Seiches!periods of}%
+some of which were no doubt multinodal. Thus
+in a trinodal seiche, the nodes divide the lake
+into four parts, of which the two central ones
+are each twice as long as the extreme parts. If
+there are any number of nodes, their positions
+are such that the central portion of the lake is
+divided into equal lengths, and the terminal
+parts are each of half the length of the central
+part or parts. This condition is necessary in
+order that the ends of the lake may fall at places
+where there is no horizontal current. In all such
+modes of oscillation the places where the horizontal
+current is evanescent are called loops, and
+these are always halfway between the nodes,
+where there is no rise and fall.
+
+A trinodal seiche should have a period of
+about twenty-four minutes, and a quadrinodal
+seiche should oscillate in about eighteen minutes.
+The periods of these quicker seiches would, no
+doubt, be affected by the irregularity in the form
+and depth of the lake, and it is worthy of notice
+that Forel observed at Morges seiches with
+periods of about twenty minutes and thirty minutes,
+which he conjectured to be multinodal.
+
+The second group of seiches were transverse,
+\index{Seiches!longitudinal and transverse|)}%
+being observable at Morges and Évian. It was
+\PageSep{28}
+clear that these oscillations, of which the period
+was about ten minutes, were transversal, because
+at the moment when the water was highest at
+Morges it was lowest at Évian, and vice versa.
+As in the case of the longitudinal seiches, the
+principal oscillation of this class was uninodal,
+but the node was, of course, now longitudinal to
+the lake. The irregularity in the width and
+depth of the lake must lead to great diversity of
+period in the transverse seiches appropriate to
+the various parts of the lake. The transverse
+seiches at one part of the lake must also be
+transmitted elsewhere, and must confuse the
+seiches appropriate to other parts. Accordingly
+there is abundant reason to expect oscillations of
+such complexity as to elude complete explanation.
+
+The great difficulty of applying deductive
+reasoning to the oscillations of a sheet of water
+of irregular outline and depth led Forel to construct
+a model of the lake. By studying the
+waves in his model he was able to recognize
+many of the oscillations occurring in the real
+lake, and so obtained an experimental confirmation
+of his theories, although the periods of
+oscillation in the model of course differed enormously
+from those observed in actuality.
+
+The theory of seiches cannot be considered as
+demonstrated, unless we can show that the water
+of such a basin as that of Geneva is capable of
+\index{Geneva!seiches in lake|)}%
+\index{Geneva!model of lake}%
+\PageSep{29}
+swinging at the rates observed. I must, therefore,
+now explain how it may be proved that the
+periods of the actual oscillations agree with the
+facts of the case.
+
+As a preliminary let us consider the nature of
+wave motion. There are two very distinct cases
+of the undulatory motion of water, which nevertheless
+graduate into one another. The distinction
+lies in the depth of the water compared with
+the length of the wave, measured from crest to
+crest, in the direction of wave propagation. The
+wave-length may be used as a measuring rod,
+and if the depth of the water is a small fraction
+of the wave-length, it must be considered shallow,
+but if its depth is a multiple of the wave-length,
+it will be deep. The two extremes of
+course graduate into one another.
+
+In a wave in deep water the motion dies out
+pretty rapidly as we go below the surface, so that
+when we have gone down half a wave-length
+below the surface, the motion is very small. In
+shallow water, on the other hand, the motion extends
+quite to the bottom, and in water which is
+neither deep nor shallow, the condition of affairs
+is intermediate. The two figures, \figref{11}~and~\figref{12},
+show the nature of the movement in the two
+classes of waves. In both cases the dotted lines
+\index{Waves!in deep and shallow water}%
+show the position of the water when at rest, and
+the full lines show the shapes assumed by the
+rectangular blocks marked out by the dotted
+\PageSep{30}
+lines, when wave motion is disturbing the water.
+It will be observed that in the deep water, as
+shown in \fig{11}, the rectangular blocks change
+their shape, rise and fall, and move to and fro.
+Taking the topmost row of rectangles, each block
+of water passes successively in time through all
+the forms and positions shown by the top row
+of quasi-parallelograms. So also the successive
+changes of the second row of blocks are indicated
+by the second strip, and the third and the fourth
+indicate the same. The changes in the bottom
+\Figure[0.7]{11}{Wave in Deep Water}{png}
+row are relatively very small both as to shape
+and as to displacement, so that it did not seem
+worth while to extend the figure to a greater
+depth.
+
+Turning now to the wave in shallow water in~\fig{12},
+we see that each of the blocks is simply
+displaced sideways and gets thinner or more
+\PageSep{31}
+squat as the wave passes along. Now, I say that
+we may roughly classify the water as being deep
+with respect to wave motion when its depth is
+more than half a wave-length, and as being shallow
+when it is less. Thus the same water may
+be shallow for long waves and deep for short
+\index{Waves!speed of}%
+ones. For example, the sea is very shallow for
+\Figure[0.7]{12}{Wave in Shallow Water}{png}
+the great wave of the oceanic tide, but it is very
+deep even for the largest waves of other kinds.
+Deepness and shallowness are thus merely relative
+to wave-length.
+
+The rate at which a wave moves can be exactly
+calculated from mathematical formulæ,
+from which it appears that in the deep sea a
+wave $63$~metres in length travels at $36$~kilometres
+per hour, or, in British measure, a wave of
+$68$~yards in length travels $22\frac{1}{2}$~miles an hour.
+Then, the rule for other waves is that the speed
+varies as the square root of the wave-length, so
+that a wave $16$~metres long---that is, one quarter
+of $63$~metres---travels at $18$~kilometres an
+hour, which is half of $36$~kilometres an hour.
+Or if its length were $7$~metres, or one ninth as
+\PageSep{32}
+long, it would travel at $12$~kilometres an hour,
+or one third as quick.
+
+Although the speed of waves in deep water
+depends on wave-length, yet in shallow water the
+speed is identical for waves of all lengths, and
+depends only on the depth of the water. In
+water $10$~metres deep, the calculated velocity of
+a wave is $36$~kilometres an hour; or if the water
+were $2\frac{1}{2}$~metres deep (quarter of $10$~metres), it
+would travel $18$~kilometres (half of $36$~kilometres)
+an hour; the law of variation being that
+the speed of the wave varies as the square root
+of the depth. For water that is neither deep nor
+shallow, the rate of wave propagation depends
+both on depth and on wave-length, according to
+a law which is somewhat complicated.
+
+In the case of seiches, the waves are very long
+compared with the depth, so that the water is to
+be considered as shallow; and here we know
+that the speed of propagation of the wave depends
+only on depth. The average depth of the
+Lake of Geneva may be taken as about $150$~metres,
+and it follows that the speed of a long wave
+in the lake is about $120$~kilometres an hour.
+
+In order to apply this conclusion to the study
+of seiches, we have to consider what is meant by
+the composition of two waves. If I take the
+series of numbers
+\[
+\text{\&c.}\quad
+100\quad 71\quad 0\quad -71\quad -100\quad -71\quad 0\quad 71\quad 100\quad
+\text{\&c.}
+\]
+and plot out, at equal distances, a figure of
+\PageSep{33}
+\index{Waves!composition of|(}%
+heights proportional to these numbers, setting
+off the positive numbers above and the negative
+numbers below a horizontal line, I get the simple
+wave line shown in~\fig{13}. Now, if this
+wave is traveling to the right, the same series of
+numbers will represent the wave at a later time,
+when they are all displaced towards the right, as
+in the dotted line.
+
+Now turn to the following schedule of numbers,
+and consider those which are written in the
+top row of each successive group of three rows.
+The columns represent equidistant spaces, and
+the rows equidistant times. The first set of
+numbers, $-100$,~$-71$,~$0$,~\&c., are those which
+were plotted out as a wave in~\fig{13}; in the top
+\Figure[0.7]{13}{Simple Wave}{png}
+row of the second group they are the same, but
+moved one space to the right, so that they represent
+the movement of the wave to the right in
+one interval of time. In the top row of each
+successive group the numbers are the same, but
+always displaced one more space to the right;
+they thus represent the successive positions of a
+\PageSep{34}
+\begin{figure}[hp!]
+\centering
+\ifthenelse{\boolean{ForPrinting}}{%
+ \makebox[0pt][c]{\includegraphics[height=0.9\textheight]{./images/fig13b.pdf}}
+}{
+ \includegraphics[width=0.75\textwidth]{./images/fig13b.pdf}
+}
+\caption{Composition of Two Equal and Opposite Waves}
+\label{fig:13b}
+\iffalse
+-100 -71 0 71 100 71 0 -71 -100
+
+-100 -71 0 71 100 71 0 -71 -100
+[*--]
+-200 -142 0 142 200 142 0 -142 -200
+[*--]
+ -71 -100 -71 0 71 100 71 0 -71
+
+ -71 0 71 100 71 0 -71 -100 -71
+[*--]
+-142 -100 0 100 142 100 0 -100 -142
+[*--]
+ 0 -71 -100 -71 0 71 100 71 0
+
+ 0 71 100 71 0 -71 -100 -71 0
+[*--]
+ 0 0 0 0 0 0 0 0 0
+[*--]
+ 71 0 -71 -100 -71 0 71 100 71
+
+ 71 100 71 0 -71 -100 -71 0 71
+[*--]
+ 142 100 0 -100 -142 -100 0 100 142
+[*--]
+ 100 71 0 -71 -100 -71 0 71 100
+
+ 100 71 0 -71 -100 -71 0 71 100
+[*--]
+ 200 142 0 -142 -200 -142 0 142 200
+[*--]
+ 71 100 71 0 -71 -100 -71 0 71
+
+ 71 0 -71 -100 -71 0 71 100 71
+[*--]
+ 142 100 0 -100 -142 -100 0 100 142
+[*--]
+ 0 71 100 71 0 -71 -100 -71 0
+
+ 0 -71 -100 -71 0 71 100 71 0
+[*--]
+ 0 0 0 0 0 0 0 0 0
+[*--]
+ -71 0 71 100 71 0 -71 -100 -71
+
+ -71 -100 -71 0 71 100 71 0 -71
+[*--]
+-142 -100 0 100 142 100 0 -100 -142
+[*--]
+-100 \DPtypo{71}{-71} 0 71 100 71 0 -71 -100
+
+-100 -71 0 71 100 71 0 -71 -100
+[*--]
+-200 -142 0 142 200 142 0 -142 -200
+[*--]
+\fi
+\end{figure}
+\PageSep{35}
+wave moving to the right. The table ends in
+the same way as it begins, so that in eight of
+these intervals of time the wave has advanced
+through a space equal to its own length.
+
+If we were to invert these upper figures, so
+that the numbers on the right are exchanged
+with those on the left, we should have a series of
+numbers representing a wave traveling to the
+left. Such numbers are shown in the second
+row in each group.
+
+When these two waves coëxist, the numbers
+must be compounded together by addition, and
+then the result is the series of numbers written
+in the third rows. These numbers represent the
+resultant of a wave traveling to the right, and of
+an equal wave traveling simultaneously to the
+left.
+
+It may be well to repeat that the first row of
+each group represents a wave moving to the
+right, the second row represents a wave moving
+to the left, and the third represents the resultant
+of the two. Now let us consider the nature
+of this resultant motion; the third and the
+seventh columns of figures are always zero, and
+therefore at these two places the water neither
+rises nor falls,---they are, in fact, nodes. If the
+schedule were extended indefinitely both ways,
+exactly halfway between any pairs of nodes
+there would be a loop, or line across which there
+is no horizontal motion. In the schedule, as it
+\PageSep{36}
+stands, the first, fifth, and ninth columns are
+loops.
+
+At the extreme right and at the extreme left
+the resultant numbers are the same, and represent
+a rise of the water from $-200$ to~$+200$,
+and a subsequent fall to~$-200$ again. If these
+nine columns represent the length of the lake,
+the motion is that which was described as binodal,
+for there are two nodes dividing the lake
+into three parts, there is a loop at each end, and
+when the water is high in the middle it is low at
+the ends, and vice versa. It follows that two
+equal waves, each as long as the lake, traveling
+in opposite directions, when compounded together
+give the motion which is described as the
+binodal longitudinal seiche.
+
+Now let us suppose that only five columns of
+the table represent the length of the lake. The
+resultant numbers, which again terminate at
+each end with a loop, are:---
+\begin{align*}
+-200&& -142&& 0&& 142&& 200& \displaybreak[0] \\
+-142&& -100&& 0&& 100&& 142& \displaybreak[0] \\
+ 0&& 0&& 0&& 0&& 0& \displaybreak[0] \\
+ 142&& 100&& 0&& -100&& -142& \displaybreak[0] \\
+ 200&& 142&& 0&& -142&& -200& \displaybreak[0] \\
+ 142&& 100&& 0&& -100&& -142& \displaybreak[0] \\
+ 0&& 0&& 0&& 0&& 0& \displaybreak[0] \\
+-142&& -100&& 0&& 100&& 142& \displaybreak[0] \\
+-200&& -142&& 0&& 142&& 200&
+\end{align*}
+
+Since the middle column consists of zero
+throughout, the water neither rises nor falls
+\PageSep{37}
+there, and there is a node at the middle. Again,
+since the numbers at one end are just the same
+as those at the other, but reversed as to positive
+and negative, when the water is high at one end
+it is low at the other. The motion is, in fact, a
+simple rocking about the central line, and is that
+described as the uninodal longitudinal seiche.
+
+The motion is here again the resultant of two
+equal waves moving in opposite directions, and
+the period of the oscillation is equal to the time
+which either simple wave takes to travel through
+its own length. But the length of the wave is
+now twice that of the lake. Hence it follows
+that the period of the rocking motion is the
+time occupied by a wave in traveling twice the
+length of the lake. We have already seen that
+in shallow water the rate at which a wave moves
+is independent of its length and depends only
+on the depth of the water, and that in water of
+the same depth as the Lake of Geneva the wave
+travels $120$~kilometres an hour. The Lake of
+Geneva is $70$~kilometres long, so that the two
+waves, whose composition produces a simple rocking
+\index{Waves!composition of|)}%
+of the water, must each of them have a
+length of $140$~kilometres. Hence it follows that
+the period of a simple rocking motion, with one
+node in the middle of the Lake of Geneva, will
+be almost exactly $\frac{140}{120}$~of an hour, or $70$~minutes.
+Forel, in fact, found the period to be $73$~minutes.
+He expresses this result by saying that
+\PageSep{38}
+a uninodal longitudinal seiche in the Lake of
+Geneva has a period of $73$~minutes. His observations
+also showed him that the period of a
+binodal seiche was $35$~minutes. It follows from
+the previous discussion that when there are two
+nodes the period of the oscillation should be
+half as long as when there is one node. Hence,
+we should expect that the period would be
+about $36$~or $37$~minutes, and the discrepancy
+between these two results may be due to the
+fact that the formula by which we calculate the
+period of a binodal seiche would require some
+correction, because the depth of the lake is not
+so very small compared with the length of these
+shorter waves.
+
+It is proper to remark that the agreement
+between the theoretical and observed periods is
+suspiciously exact. The lake differs much in
+depth in different parts, and it is not quite certain
+what is the proper method of computing
+the average depth for the determination of the
+period of a seiche. It is pretty clear, in fact,
+that the extreme closeness of the agreement is
+accidentally due to the assumption of a round
+number of metres as the average depth of the
+lake. The concordance between theory and observation
+must not, however, be depreciated too
+much, for it is certain that the facts of the case
+agree well with what is known of the depth of
+the lake.
+\index{Forel!on seiches|)}%
+\PageSep{39}
+
+The height of the waves called ``seiches'' is
+\index{Seiches!causes of}%
+\index{Storms a cause of seiches}%
+very various. I have mentioned an historical
+seiche which had a range of as much as four
+feet, and Forel was able by his delicate instruments
+still to detect them when they were only a
+millimetre or a twenty-fifth of an inch in height.
+It is obvious, therefore, that whatever be the
+cause of seiches, that cause must vary widely
+in intensity. According to Forel, seiches arise
+from several causes. It is clear that anything
+which heaps up the water at one end of the
+lake, and then ceases to act, must tend to produce
+an oscillation of the whole. Now, a rise
+of water level at one end or at one side of the
+lake may be produced in various ways. Some,
+and perhaps many, seiches are due to the tilting
+of the whole lake bed by minute earthquakes.
+\index{Earthquakes!a cause of seiches}%
+Modern investigations seem to show that this is
+a more fertile cause than Forel was disposed to
+allow, and it would therefore be interesting to
+see the investigation of seiches repeated with the
+aid of delicate instruments for the study of
+earthquakes, some of which will be described in
+\Ref{Chapter}{VI}. I suspect that seiches would be
+observed at times when the surface of the earth
+is much disturbed.
+
+The wind is doubtless another cause of seiches.
+\index{Wind!a cause of seiches}%
+When it blows along the lake for many hours in
+one direction, it produces a superficial current,
+and heaps up the water at the end towards
+\PageSep{40}
+\index{Atmospheric pressure!cause of seiches}%
+which it is blowing. If such a wind ceases
+somewhat suddenly, a seiche will certainly be
+started, and will continue for hours until it dies
+out from the effects of the friction of the water
+on the lake bottom. Again, the height of the
+barometer will often differ slightly at different
+parts of the lake, and the water will respond, just
+as does the mercury, to variations of atmospheric
+pressure. About a foot of rise of water should
+correspond to an inch of difference in the height
+of barometer. The barometric pressure cannot
+be quite uniform all over the Lake of Geneva,
+and although the differences must always be
+exceedingly small, yet it is impossible to doubt
+that this cause, combined probably with wind,
+will produce many seiches. I shall return later
+\index{Seiches!causes of}%
+to the consideration of an interesting speculation
+as to the effects of barometric pressure on
+the oscillation of lakes and of the sea. Lastly,
+Forel was of opinion that sudden squalls or local
+storms were the most frequent causes of seiches.
+\index{Storms a cause of seiches}%
+I think that he much overestimated the efficiency
+of this cause, because his theory of the path of
+the wind in sudden and local storms is one that
+would hardly be acceptable to most meteorologists.
+
+Although, then, it is possible to indicate causes
+competent to produce seiches, yet we cannot as
+yet point out the particular cause for any individual
+seiche. The complication of causes is so
+\PageSep{41}
+\index{Lakes!vibrations|(}%
+\index{Vibration of lakes|(}%
+great that this degree of uncertainty will probably
+never be entirely removed.
+
+But I have not yet referred to the point which
+justifies this long digression on seiches in a book
+on the tides. The subject was introduced by
+the irregularities in the line traced by the tide-gauge
+at Bombay, which indicated that there
+are oscillations of the water with periods ranging
+from two minutes to a quarter of an hour or
+somewhat longer. Now these zigzags are not
+found in the sea alone, for Forel observed on
+the lake oscillations of short period, which resembled
+seiches in all but the fact of their more
+rapid alternations. Some of these waves are
+perhaps multinodal seiches, but it seems that
+they are usually too local to be true seiches
+affecting the whole body of the lake at one time.
+Forel calls these shorter oscillations ``vibrations,''
+thus distinguishing them from proper seiches.
+A complete theory of the so-called vibrations
+has not yet been formulated, although, as I shall
+show below, a theory is now under trial which
+serves to explain, at least in part, the origin of
+vibrations.
+
+Forel observed with his limnimeter or tide-gauge
+that when there is much wind, especially
+\index{Wind!vibrations of lakes due to}%
+from certain quarters, vibrations arise which are
+quite distinct from the ordinary visible wave
+motion. The period of the visible waves on the
+\PageSep{42}
+Lake of Geneva is from $4$~to $5$~seconds,\footnote
+ {I observed when it was blowing half a gale on Ullswater, in
+ Cumberland, that the waves had a period of about a second.}
+whereas
+vibrations have periods ranging from $45$~seconds
+to $4$~minutes. Thus there is a clear line separating
+waves from vibrations. Forel was unable
+to determine what proportion of the area of the
+lake is disturbed by vibrations at any one time,
+and although their velocity was not directly observed,
+there can be no doubt that these waves
+are propagated at a rate which corresponds to
+their length and to the depth of the water. I
+have little doubt but that the inequalities which
+produce notches in a tide-curve have the same
+origin as vibrations on lakes.
+
+It is difficult to understand how a wind, whose
+\index{Wind!vibrations of lakes due to}%
+only visible effect is short waves, can be responsible
+for raising waves of a length as great as a
+thousand yards or a mile, and yet we are driven
+to believe that this is the case. But Forel also
+found that steamers produce vibrations exactly
+like those due to wind. The resemblance was
+indeed so exact that vibrations due to wind
+could only be studied at night, when it was
+known that no steamers were traveling on the
+lake, and, further, the vibrations due to steamers
+could only be studied when there was no wind.
+
+His observations on the steamer vibrations are
+amongst the most curious of all his results.
+When a boat arrives at the pier at Morges, the
+\PageSep{43}
+water rises slowly by about $5$~to $8$~millimetres,
+and then falls in about $20$~to $30$~seconds. The
+amount and the rapidity of the rise and fall
+vary with the tonnage of the boat and with the
+rate of her approach. After the boat has passed,
+the trace of the limnimeter shows irregularities
+with sharp points, the variations of height ranging
+from about two to five millimetres, with a
+period of about two minutes. These vibrations
+continue to be visible during two to three hours
+after the boat has passed. As these boats travel
+at a speed of $20$~kilometres an hour, the vibrations
+persist for a long time after any renewal
+of them by the boat has ceased. These vibrations
+are called by Forel ``the subsequent steamer
+vibrations.''
+
+That the agitation of the water should continue
+for more than two hours is very remarkable,
+and shows the delicacy of the method of
+observation. But it seems yet more strange
+that, when a boat is approaching Morges, the
+vibrations should be visible during $25$~minutes
+before she reaches the pier. These he calls
+``antecedent steamer vibrations.'' They are
+more rapid than the subsequent ones, having a
+period of a minute to a minute and a quarter.
+Their height is sometimes two millimetres (a
+twelfth of an inch), but they are easily detected
+when less than one millimetre in height. It
+appears that these antecedent vibrations are first
+\PageSep{44}
+noticeable when the steamer rounds the mole of
+Ouchy, when she is still at a distance of $10$~kilometres.
+As far as one can judge from the speed
+at which waves are transmitted in the Lake of
+Geneva, the antecedent vibrations, which are
+noticed $25$~minutes before the arrival of the
+boat, must have been generated when she was at
+a distance of $12$~kilometres from Morges. \Fig{14}
+gives an admirable tracing of these steamer
+vibrations.\footnote
+ {From \Title{Les Seiches, Vagues d'Oscillation fixe des Lacs}, 1876.}
+
+In this figure the line~$\Seg{a}{a'}$ was traced between
+two and three o'clock in the morning, and shows
+scarcely any sign of perturbation. Between
+three and eight o'clock in the morning no observations
+were taken, but the record begins again
+at eight o'clock. The portion marked~$\Seg{b}{b'}$ shows
+weak vibrations, probably due to steamers passing
+along the coast of Savoy. The antecedent
+vibrations, produced by a steamer approaching
+Morges, began about the time of its departure
+from Ouchy, and are shown at~$\Seg{c}{c'}$. The point~$d$
+shows the arrival of this boat at Morges, and
+$d'$~shows the effect of another boat coming from
+Geneva. The portion marked~$\Seg[e]{e}{e}$ shows the
+subsequent steamer vibrations, which were very
+clear during more than two hours after the boats
+had passed.
+
+Dr.~Forel was aware that similar vibrations occur
+in the sea, for he says: ``What are these
+\index{Sea!vibrations of}%
+\PageSep{45}
+oscillations with periods of
+$5$,~$10$, $20$, or $100$~minutes,
+which are sometimes irregular?
+Are they analogous to
+our seiches? Not if we define
+seiches as uninodal oscillations,
+for it is clear that
+if, in a closed basin of $70$~kilometres
+in length, uninodal
+seiches have a period of
+$73$~minutes, in the far greater
+basin of the Mediterranean,
+or of the ocean, a uninodal
+wave of oscillation must have
+a much longer period. They
+resemble much more closely
+what I have called vibrations,
+and, provisionally, I
+shall call them by the name
+of `vibrations of the sea.' I
+\index{Sea!vibrations of}%
+venture to invite men of science
+who live on the seacoast
+to follow this study.
+It presents a fine subject for
+research, either in the interpretation
+of the phenomenon
+or in the establishment of
+the relations between these
+movements and meteorological
+conditions.''\footnote
+ {\Title{Seiches et Vibrations des Lacs et de la Mer}, 1879, p.~5.}
+%[** TN: Figure wrapped in the original]
+\TallFig{14}{Vibrations due to Steamers}{png}
+\PageSep{46}
+
+These vibrations are obviously due to the wind
+or to steamers, but it is a matter of no little surprise
+that such insignificant causes should produce
+even very small waves of half a mile to a
+mile in length.
+
+The manner in which this is brought about is
+undoubtedly obscure, yet it is possible to obtain
+some sort of insight into the way in which these
+long waves arise. When a stone falls into calm
+water waves of all sorts of lengths are instantaneously
+generated, and the same is true of
+any other isolated disturbance. Out of all these
+waves the very long ones and the very short
+ones are very small in height. Theoretically,
+waves of infinitely great and of infinitely small
+lengths, yet in both cases of infinitely small
+heights, are generated at the instant of the impulse,
+but the waves of enormous length and
+those of very small length are of no practical
+importance, and we need only consider the moderate
+waves. For the shorter of these the water
+is virtually deep, and so they will each travel
+outwards at a pace dependent on length, the
+longer ones outstripping the shorter ones. But
+for the longer waves the water will be shallow,
+and they will all travel together. Thus the general
+effect at a distance is the arrival of a long
+wave first, followed by an agitated rippling.
+The point which we have to note is that an isolated
+disturbance will generate long waves and
+\PageSep{47}
+that they will run ahead of the small ones. It
+is important also to observe that the friction of
+the water annuls the oscillation in the shorter
+waves more rapidly than it does that of the
+longer ones, and therefore the long waves are
+more persistent. Now we may look at the disturbance
+due to a steamer or to the wind as consisting
+of a succession of isolated disturbances,
+each of which will create long waves outstripping
+the shorter ones. These considerations afford a
+sort of explanation of what is observed, but I do
+not understand how it is that the separation of
+the long from the short waves is so complete, nor
+what governs the length of the waves, nor have
+I made any attempt to evaluate the greater rapidity
+of decrease of short waves than long ones.\footnote
+ {See, however, S.~S. Hough, \Title{Proc.\ Lond.\ Math.\ Soc.}, xxviii.\
+\index{Hough, S. S.!frictional extinction of waves}%
+ p.~276.}
+It must then be left to future investigators to
+elucidate these points.
+
+\TB
+
+The subject of seiches and vibrations clearly
+affords an interesting field for further research.
+The seiches of Lake George in New South Wales
+have been observed by Mr.~Russell, the government
+\index{Russell, observation of seiches in New South Wales}%
+astronomer at Sydney; but until last year
+they do not seem to have been much studied on
+any lakes outside of Switzerland. The great
+lakes of North America are no doubt agitated by
+seiches on a much larger scale than those on the
+\PageSep{48}
+comparatively small basin of Geneva. This idea
+appears to have struck Mr.~Napier Denison of
+\index{Denison, F. Napier, vibrations and seiches on lakes|(}%
+Toronto, and he has been so fortunate as to enlist
+the interest of Mr.~Bell Dawson, the chief of
+\index{Dawson coöperates in investigation of seiches}%
+the Canadian Tidal Survey, and of Mr.~Stupart,
+\index{Stupart coöperates in investigation of seiches}%
+the director of the Meteorological Department.
+Mr.~Denison's attention has been, in the first instance,
+principally directed towards those notches
+in tide-curves which have afforded the occasion
+for the present discussion of this subject. He
+has made an interesting suggestion as to the
+origin of these oscillations, which I will now
+explain.
+
+The wind generally consists of a rather shallow
+\index{Waves!in atmosphere|(}%
+current, so that when it is calm at the earth's
+surface there is often a strong wind at the top
+of a neighboring mountain; or the wind aloft
+may blow from a different quarter from that below.
+If we ascend a mountain or go up in a
+balloon, the temperature of the air falls on the
+average by a certain definite number of degrees
+per thousand feet. But the normal rate of fall
+of temperature is generally interrupted on passing
+into an upper current, which blows from a
+different direction. This abrupt change of temperature
+corresponds with a sudden change of
+density, so that the upper layer of air must be
+regarded as a fluid of different density from that
+of the lower air, over which it slides.
+
+Now Helmholtz has pointed out that one layer
+\index{Atmospheric waves, Helmholtz on|(}%
+\index{Helmholtz!on atmospheric waves|(}%
+\PageSep{49}
+of fluid cannot slide over another, without generating
+waves at the surface of separation. We
+are familiar with this fact in the case of sea-waves
+generated by wind. A mackerel sky
+\index{Mackerel sky, evidence of air-waves}%
+proves also the applicability to currents of air of
+Helmholtz's observation. In this case the moisture
+of the air is condensed into clouds at the
+crests of the air waves, and reabsorbed in the
+hollows, so that the clouds are arranged in a visible
+ripple-mark. A mackerel sky is not seen in
+stormy weather, for it affords proof of the existence
+of an upper layer of air sliding with only
+moderate velocity over a lower layer. The distance
+from crest to crest must be considerable
+as measured in yards, yet we must regard the
+mackerel sky as a mere ripple formed by a slow
+relative velocity of the two layers. If this is so,
+it becomes of interest to consider what wave-lengths
+may be expected to arise when the upper
+current is moving over the lower with a speed of
+perhaps a hundred miles an hour. The problem
+is not directly soluble, for even in the case of
+sea-waves it is impossible to predict the wave-lengths.
+We do know, however, that the duration
+of the wind and the size of the basin are
+material circumstances, and that in gales in the
+open ocean the waves attain a very definite magnitude.
+
+Although the problem involved is not a soluble
+one, yet Helmholtz has used the analogy of
+\PageSep{50}
+oceanic waves for an approximate determination
+of the sizes of the atmospheric ones. His
+method is a very fertile one in many complex
+physical investigations, where an exact solution
+is not attainable. The method may be best illustrated
+by one or two simple cases.
+
+It is easy for the mathematician to prove that
+the period of a swing of a simple pendulum must
+vary as the square root of its length. The proof
+does not depend on the complete solution of the
+problem, so that even if it were insoluble he
+would still be sure of the correctness of his conclusion.
+If, then, a given pendulum is observed
+to swing in a certain period, it is certain that a
+similar pendulum of four times the length will
+take twice as long to perform its oscillation. In
+the same way, the engine power required for a
+ship is determinable from experiments on the
+resistance suffered by a small model when towed
+through the water. The correct conclusion is
+discovered in this case, although it is altogether
+impossible to discover the resistance of a ship
+by \textit{à~priori} reasoning.
+
+The wave motion at the surface separating
+two fluids of different densities presents another
+problem of the same kind, and if the result is
+known in one case, it can be confidently predicted
+in another. Now oceanic waves generated
+\index{Waves!in atmosphere|)}%
+by wind afford the known case, and Helmholtz
+has thence determined by analogy the
+\PageSep{51}
+lengths of the atmospheric waves which must
+exist aloft. By making plausible suppositions
+as to the densities of the two layers of air and
+as to their relative velocity, he has shown that
+sea-waves of ten yards in length will correspond
+with air-waves of as much as twenty miles. A
+wave of this length would cover the whole sky,
+and might have a period of half an hour. It is
+clear then that mackerel sky will disappear in
+stormy weather, because we are too near to the
+crests and furrows to observe the orderly arrangement
+of the clouds.
+
+Although the waves are too long to be seen as
+such, yet the unsteadiness of the barometer in a
+gale of wind affords evidence of the correctness
+of this theory. In fact, when the crest of denser
+air is over the place of observation the barometer
+rises, and it falls as the hollow passes. The
+waves in the continuous trace of the barometer
+have some tendency to regularity, and have
+periods of from ten minutes to half an hour.
+The analogy seems to be pretty close with the
+confused and turbulent sea often seen in a gale
+of wind in the open ocean.\footnote
+ {A gust of wind will cause the barometer to vary, without a
+ corresponding change in the density of the air. It is not therefore
+ safe to interpret the oscillations of the barometer as being
+ due entirely to true changes of pressure. If, however, the intermittent
+ squalls in a gale are connected with the waves aloft,
+ the waviness of the barometric trace would still afford signals
+ of the passage of crests and hollows above.}
+\index{Atmospheric waves, Helmholtz on|)}%
+\index{Helmholtz!on atmospheric waves|)}%
+\PageSep{52}
+
+Mr.~Denison's application of this theory consists
+in supposing that the vibrations of the sea
+and of lakes are the response of the water to
+variations in the atmospheric pressure. The sea,
+being squeezed down by the greater pressure,
+should fall as the barometer rises, and conversely
+should rise as the barometer falls. He is engaged
+in a systematic comparison of the simultaneous
+excursions of the water and of the barometer
+on Lake Huron. Thus far the evidence
+seems decidedly favorable to the theory. He
+concludes that when the water is least disturbed,
+so also is the barometric trace; and that when
+the undulations of the lake become large and
+rapid, the atmospheric waves recorded by the
+barometer have the same character. There is
+also a considerable degree of correspondence
+between the periods of the two oscillations. The
+smaller undulations of the water correspond with
+the shorter air-waves, and are magnified as they
+run into narrower and shallower places, so as to
+make conspicuous ``vibrations.''
+
+It is interesting to note that the vibrations of
+the water have a tendency to appear before those
+in the barometer, so that they seem to give a
+warning of approaching change of weather. It
+is thus not impossible that we here have the
+foreshadowing of a new form of meteorological
+instrument, which may be of service in the forecasting
+of the weather.
+\PageSep{53}
+
+I must, however, emphasize that these conclusions
+are preliminary and tentative, and that
+much observation will be needed before they can
+be established as definite truths. Whatever
+may be the outcome, the investigation appears
+promising, and it is certainly already interesting.
+\index{Denison, F. Napier, vibrations and seiches on lakes|)}%
+\index{Lakes!vibrations|)}%
+\index{Vibration of lakes|)}%
+
+\begin{Authorities}
+Papers by Dr.~Forel on Seiches.
+\index{Forel!list of papers}%
+
+\Journal ``Bibliothèque Universelle, Archives des Sciences physiques
+et naturelles,'' Geneva:---
+
+\Paper{\Title{Formule des Seiches}, 1876.}
+
+\Paper{\Title{Limnimètre Enregistreur}, 1876.}
+
+\Paper{\Title{Essai monographique}, 1877.}
+
+\Paper{\Title{Causes des Seiches}, \DPchg{Sept.~15}{15~Sept.}, 1878.}
+
+\Paper{\Title{Limnographe}, 15~Déc., 1878.}
+
+\Paper{\Title{Seiche du \emph{20}~Février, \emph{1879}}, 15~Avril, 1879.}
+
+\Paper{\Title{Seiches dicrotes}, 15~Jan., 1880.}
+
+\Paper{\Title{Formules des Seiches}, 15~Sept., 1885.}
+
+\Journal ``Bulletin de la Soc.\ Vaudoise des Sciences naturelles:''---
+
+\Paper{\Title{Première Étude}, 1873.}
+
+\Paper{\Title{Deuxième Étude}, 1875.}
+
+\Paper{\Title{Limnimétrie du Lac Léman}. I\iere~Série. Bull.~xiv.\ 1877.
+II\ieme~Série. Bull.~xv. III\ieme~Série. Bull.~xv. 1879.}
+
+\Journal ``Actes de la Soc.\ helv.\ Andermatt:''---
+
+\Paper{\Title{Les Seiches, Vagues d'Oscillation}, 1875.}
+
+\Journal ``Association Française pour l'avancement,'' etc.:---
+
+\Paper{\Title{Seiches et Vibrations}, Congrès de Montpelier, 1879.}
+
+\Journal ``Annales de Chimie et de Physique:''---
+
+\Paper{\Title{Les Seiches, Vagues d'Oscillation}, 1876.}
+
+\Paper{\Title{Un Limnimètre Enregistreur}, 1876.}
+
+\TB
+
+Helmholtz, Sitzungsberichte der Preuss.\ Akad.\ der Wissenschaft,
+July~25, 1889; transl.\ by Abbe in \Title{Smithsonian Reports}.
+\PageSep{54}
+
+F.~Napier Denison:---
+
+\Paper{\Title{Secondary Undulations~\dots\ found in Tide-Gauges.} ``Proc.\
+Canadian Institute,'' Jan.~16, 1897.}
+
+\Paper{\Title{The Great Lakes as a Sensitive Barometer.} ``Proc.\ Canadian
+Institute,'' Feb.~6, 1897.}
+\index{Lakes!seiches in|)}%
+
+\Paper{Same title, but different paper, ``Canadian Engineer,'' Oct.\
+and Nov., 1897.}
+\end{Authorities}
+\index{Forel!list of papers}%
+\PageSep{55}
+
+
+%[** TN: Footnote mark handled by \Chapter logic]
+\Chapter{III}
+{Tides in Rivers---Tide Mills}
+
+\footnotetext{The account of the bore in this chapter appeared as an
+ article in the \Title{Century Magazine} for August,~1898. The illustrations
+ then used are now reproduced, through the courtesy of
+ the proprietors.}
+
+\First{Since} most important towns are situated on
+\index{Rivers!tide wave in|(}%
+rivers or on estuaries, a large proportion of our
+tidal observations relates to such sites. I shall
+therefore now consider the curious, and at times
+very striking phenomena which attend the rise
+and fall of the tide in rivers.
+
+The sea resembles a large pond in which the
+water rises and falls with the oceanic tide, and a
+river is a canal which leads into it. The rhythmical
+rise and fall of the sea generate waves
+which would travel up the river, whatever were
+the cause of the oscillation of the sea. Accordingly,
+a tide wave in a river owes its origin
+directly to the tide in the sea, which is itself
+produced by the tidal attractions of the sun and
+moon.
+
+We have seen in \Ref{Chapter}{II}.\ that long waves
+progress in shallow water at a speed which depends
+only on the depth of the water, and that
+\PageSep{56}
+waves are to be considered as long when their
+length is at least twice the depth of the water.
+Now the tide wave in a river is many hundreds
+of times as long as the depth, and it must therefore
+progress at a speed dependent only on the
+depth. That speed is very slow compared with
+the motion of the great tide wave in the open
+ocean.
+
+The terms ``ebb'' and ``flow'' are applied to
+\index{Ebb and flow defined}%
+\index{Flow and ebb defined}%
+tidal currents. The current ebbs when the
+water is receding from the land seaward, and
+flows when it is approaching the shore. On the
+open seacoast the water ebbs as the water-level
+falls, and it flows as the water rises. Thus at
+high and low tide the water is neither flowing
+landward nor ebbing seaward, and we say that
+it is slack or dead. In this case ebb and flow
+are simultaneous with rise and fall, and it is not
+uncommon to hear the two terms used synonymously;
+but we shall see that this usage is incorrect.
+
+I begin by considering the tidal currents in a
+river of uniform depth, so sluggish in its own
+proper current that it may be considered as a
+stagnant canal, and the only currents to be considered
+are tidal currents. At any point on the
+river bank there is a certain mean height of
+water, such that the water rises as much above
+that level at high water as it falls below it at
+low water. The law of tidal current is, then,
+\index{Currents, tidal, in rivers}%
+\PageSep{57}
+very simple. Whenever the water stands above
+the mean level the current is up-stream and progresses
+along with the tide wave; and whenever
+it stands below mean level the current is down-stream
+and progresses in the direction contrary
+to the tide wave. Since the current is up-stream
+when the water is higher than the mean, and
+down-stream when it is lower, it is obvious that
+when it stands exactly at mean level the current
+is neither up nor down, and the water is slack
+or dead. Also, at the moment of high water
+the current is most rapid up-stream, and at low
+water it is most rapid down-stream. Hence the
+tidal current ``flows'' for a long time after high
+water has passed and when the water-level is
+falling, and ``ebbs'' for a long time after low
+water and when the water-level is rising.
+
+The law of tidal currents in a uniform canal
+communicating with the sea is thus very different
+from that which holds on an open seacoast,
+where slack water occurs at high and at low
+water, instead of at mean water. But rivers
+gradually broaden and become deeper as they
+approach the coast, and therefore the tidal currents
+in actual estuaries must be intermediate
+between the two cases of the open seacoast and
+the uniform canal.
+
+A river has also to deliver a large quantity of
+water into the sea in the course of a single tidal
+oscillation, and its own proper current is superposed
+\PageSep{58}
+on the tidal currents. Hence in actual
+rivers the resultant current continues to flow up
+stream after high water is reached, with falling
+water-level, but ceases flowing before mean water-level
+is reached, and the resultant current ebbs
+down-stream after low water, and continues to
+ebb with the rising tide until mean water is
+reached, and usually for some time afterward.
+The downward stream, in fact, lasts longer than
+the upward one. The moments at which the
+currents change will differ in each river according
+to the depth, the rise and fall of the tide at
+the mouth, and the amount of water delivered
+by the river. An obvious consequence of this
+is that in rivers the tide rises quicker than it
+falls, so that a shorter time elapses between low
+water and high water than between high water
+and low water.
+
+The tide wave in a river has another peculiarity
+of which I have not yet spoken. The complete
+theory of waves would be too technical for a book
+of this sort, and I must ask the reader to accept
+as a fact that a wave cannot progress along a
+river without changing its shape. The change
+is such that the front slope of the wave gradually
+gets steeper, and the rear slope becomes more
+gradual. This is illustrated in~\fig{15}, which
+shows the progress of a train of waves in shallow
+water as calculated theoretically. If the
+steepening of the advancing slope of a wave
+\PageSep{59}
+were carried to an extreme, the wave would present
+the form of a wall of water; but the mere
+advance of a wave into shallow water would by
+itself never suffice to produce so great a change
+of form without the concurrence of the natural
+\Figure{15}{Progressive Change of a Train
+of Waves in Shallow Water}{png}
+stream of the river. The downward current in
+the river has, in fact, a very important influence
+in heading the sea-water back, and this coöperates
+with the natural change in the shape of a
+wave as it runs into shallow water, so as to exaggerate
+the steepness of the advancing slope of
+the wave.
+
+There are in the estuaries of many rivers
+\index{Rivers!tide wave in|)}%
+broad flats of mud or sand which are nearly dry
+at low water, and in such situations the tide not
+unfrequently rises with such great rapidity that
+the wave assumes the form of a wall of water.
+This sort of tide wave is called a ``bore,'' and in
+\index{Bore!definition}%
+French \textit{mascaret}. Notwithstanding the striking
+nature of the phenomenon, very little has been
+published on the subject, and I know of only one
+series of systematic observations of the bore.
+As the account to which I refer is contained in
+the official publications of the English Admiralty,
+it has probably come under the notice of only a
+\PageSep{60}
+small circle of readers. But the experiences of
+the men engaged in making these observations
+were so striking that an account of them should
+prove of interest to the general public. I have,
+moreover, through the kindness of Admiral Sir
+William Wharton and of Captain Moore, the
+\index{Moore, Captain!survey of Tsien-Tang-Kiang|(}%
+advantage of supplementing verbal description
+by photographs.
+
+The estuary on which the observations were
+made is that of the Tsien-Tang-Kiang, a considerable
+\index{Tsien-Tang-Kiang, the bore in|(}%
+river which flows into the China Sea about
+sixty miles south of the great Yang-Tse-Kiang.
+At most places the bore occurs only intermittently,
+but in this case it travels up the river at
+every tide. The bore may be observed within
+seventy miles of Shanghai, and within an easy
+walk of the great city of Hangchow; and yet
+\index{Hangchow, the bore at|(}%
+nothing more than a mere mention of it is to be
+found in any previous publication.
+
+In 1888 Captain Moore, R.~N., in command
+of Her Majesty's surveying ship Rambler,
+thought that it was desirable to make a thorough
+survey of the river and estuary. He returned
+to the same station in~1892; and the account
+which I give of his survey is derived from reports
+drawn up after his two visits. The annexed
+sketch-map shows the estuary of the
+Tsien-Tang, and the few places to which I shall
+have occasion to refer are marked thereon.
+
+On the morning of September~19, 1888, the
+\PageSep{61}
+Rambler was moored near an island, named
+after the ship, to the southwest of Chapu Bay;
+and on the~20th the two steam cutters Pandora
+and Gulnare, towing the sailing cutter
+\Figure{16}{Chart of the Estuary of the Tsien-Tang-Kiang}{png}
+Brunswick, left the ship with instruments for
+observing and a week's provisions.
+
+Captain Moore had no reason to suspect that
+the tidal currents would prove dangerous out
+in the estuary, and he proposed to go up the
+estuary about thirty miles to Haining, and then
+follow the next succeeding bore up-stream to
+Hangchow. Running up-stream with the flood,
+all went well until about~11.30, when they were
+about fifteen miles southwest by west of Kanpu.
+The leading boat, the Pandora, here grounded,
+and anchored quickly, but swung round violently
+as far as the keel would let her. The other
+boats, being unable to stop, came up rapidly;
+and the Gulnare, casting off the Brunswick,
+\PageSep{62}
+struck the Pandora, and then drove on to and
+over the bank, and anchored. The boats soon
+floated in the rising flood, and although the engines
+of the steam cutters were kept going
+full speed, all three boats dragged their anchors
+in an eleven-knot stream. When the flood
+slackened, the three boats pursued their course
+to the mouth of the river, where they arrived
+about 4~\PM. The ebb was, however, so violent
+that they were unable to anchor near one another.
+Their positions were chosen by the advice of
+some junkmen, who told Captain Moore, very
+erroneously as it turned out, that they would be
+safe from the night bore.
+
+The night was calm, and at~11.29 the murmur
+of the bore was heard to the eastward; it could
+be seen at~11.55, and passed with a roar at~12.20,
+well over toward the opposite bank, as predicted
+by the Chinese. The danger was now supposed
+to be past; but at~1~\AM\ a current of extreme
+violence caught the Pandora, and she had
+much difficulty to avoid shipwreck. In the
+morning it was found that her rudder-post and
+propeller-guard were broken, and the Brunswick
+and Gulnare were nowhere to be seen.
+They had, in fact, been in considerable danger,
+and had dragged their anchors three miles up
+the river. At 12.20~\AM\ they had been struck
+by a violent rush of water in a succession of big
+ripples. In a few moments they were afloat in
+\PageSep{63}
+an eight-knot current; in ten minutes the water
+rose nine feet, and the boats began to drag their
+anchors, although the engines of the Gulnare
+were kept going full speed. After the boats had
+dragged for three miles, the rush subsided, and
+when the anchor was hove up the pea and the
+greater part of the chain were as bright as polished
+silver.
+
+This account shows that all the boats were in
+imminent danger, and that great skill was needed
+to save them. After this experience and warning,
+the survey was continued almost entirely
+from the shore.
+
+The junks which navigate the river are well
+aware of the dangers to which the English boats
+were exposed, and they have an ingenious method
+of avoiding them. At various places on the
+bank of the river there are shelter platforms, of
+which I show an illustration in~\fig{17}. Immediately
+after the passing of the bore the
+\index{Bore!bore-shelter}%
+junks run up-stream with the after-rush and
+make for one of these shelters, where they allow
+themselves to be left stranded on the raised
+platform shown in the picture. At the end of
+this platform there is a sort of round tower
+jutting out into the stream. The object of this
+is to deflect the main wave of the bore so as to
+protect the junks from danger. After the passage
+of the bore, the water rises on the platform
+very rapidly, but the junks are just able to float
+\PageSep{64}
+in safety. Captain Moore gives a graphic account
+of the spectacle afforded by the junks as
+they go up-stream, and describes how on one
+occasion he saw no less than thirty junks swept
+\Figure{17}{Bore-Shelter on the Tsien-Tang-Kiang}{jpg}
+up in the after-rush, at a rate of ten knots, past
+the town of Haining toward Hangchow, with all
+sail set but with their bows in every direction.
+
+Measurements of the water-level were made
+in the course of the survey, and the results, in
+the form of a diagram, \fig{18}, exhibit the nature
+of the bore with admirable clearness. The
+observations of water-level were taken simultaneously
+at three places, viz., Volcano Island
+in the estuary, Rambler Island near the mouth
+of the river, and Haining, twenty-six miles up
+the river. In the figure, the distance between
+\PageSep{65}
+the lines marked Rambler and Volcano represents
+fifty-one miles, and that between Rambler
+and Haining twenty-six miles. The vertical
+scales show the height of water, measured in
+feet, above and below the mean level of the
+water at these three points. The lines joining
+these vertical scales, marked with the hours of
+the clock, show the height of the water simultaneously.
+The hour of~8.30 is indicated by
+the lowest line it shows that the water was
+one foot below mean level at Volcano Island,
+twelve feet below at Rambler Island, and eight
+feet below at Haining. Thus the water sloped
+down from Haining to Rambler, and from Volcano
+to Rambler; the water was running up the
+estuary toward Rambler Island, and down the
+estuary to the same point. At 9~and at~9.30
+there was no great change, but the water had
+risen two or three feet at Volcano Island and at
+Rambler Island. By ten~o'clock the water was
+rising rapidly at Rambler Island, so that there
+was a nearly uniform slope up the river from
+Volcano Island to Haining. The rise at Rambler
+Island then continued to be very rapid,
+while the water at Haining remained almost
+stationary. This state of affairs went on until
+midnight, by which time the water had risen
+twenty-one feet at Rambler Island, and about
+six feet at Volcano Island, but had not yet risen
+at all at Haining. No doubt through the whole
+\PageSep{66}
+of this time the water was running down the
+river from Haining towards its mouth. It is
+clear that this was a state of strain which could
+not continue long, for there was over twenty
+feet of difference of level between Rambler
+Island, outside, and Haining, in the river. Almost
+exactly at midnight the strain broke down
+and the bore started somewhere between Rambler
+\index{Bore!diagram of rise in Tsien-Tang}%
+Island and Kanpu, and rushed up the river
+in a wall of water twelve feet high. This result
+is indicated in the figure by the presence of two
+lines marked ``midnight.'' After the bore had
+\Figure{18}{Diagram of the Flow of the Tide on
+the Tsien-Tang-Kiang}{jpg}
+passed there was an after-rush that carried the
+water up eight feet more. It was on this that
+the junks were swept up the stream, as already
+described. At~1.30 the after-rush was over,
+\PageSep{67}
+but the water was still somewhat higher at
+Rambler Island than at Haining, and a gentle
+current continued to set up-stream. The water
+then began to fall at Rambler Island, while it
+continued to rise at Haining up to three o'clock.
+At this point the ebb of the tide sets in. I do
+not reproduce the figure which exhibits the fall
+of the water in the ebbing tide, for it may suffice
+to say that there is no bore down-stream,
+\index{Bore!pictures}%
+although there is at one time a very violent
+current.
+
+In 1892 Captain Moore succeeded, with considerable
+\index{Moore, Captain!illustrations of bore}%
+difficulty, in obtaining photographs of
+the bore as it passed Haining. They tell more
+of the violence of the wave than could be conveyed
+by any amount of description. The photographs,
+reproduced in~\fig{19}, do not, however,
+show that the broken water in the rear of the
+crest is often disturbed by a secondary roller, or
+miniature wave, which leaps up, from time to
+time, as if struck by some unseen force, and disappears
+in a cloud of spray. These breakers
+were sometimes twenty to thirty feet above the
+level of the river in front of the bore.
+
+The upper of these pictures is from a photograph,
+taken at a height of twenty-seven feet
+above the river, as the bore passed Haining on
+October~10, 1892. The height of this bore was
+eleven feet. The lower pictures, also taken at
+Haining, represent the passage of the bore on
+\PageSep{68}
+October~9, 1892. The first of these photographs
+was taken at 1.29~\PM, and the second
+represents the view only one minute later.
+
+The Chinese regard the bore with superstitious
+\index{Bore!Chinese superstition|(}%
+\index{Chinese!superstition as to bore|(}%
+reverence, and their explanation, which I quote
+from Captain Moore's report, is as follows:
+``Many hundred years ago there was a certain
+general who had obtained many victories over
+the enemies of the Emperor, and who, being
+constantly successful and deservedly popular
+among his countrymen, excited the jealousy of
+his sovereign, who had for some time observed
+with secret wrath his growing influence. The
+Emperor accordingly caused him to be assassinated
+and thrown into the Tsien-Tang-Kiang,
+where his spirit conceived the idea of revenging
+itself by bringing the tide in from the ocean in
+such force as to overwhelm the city of Hangchow,
+then the magnificent capital of the empire.
+As my interpreter, who has been for some years
+in America, put it, `his sowl felt a sort of ugly-like
+arter the many battles he had got for the
+Emperor.' The spirit so far succeeded as to
+flood a large portion of the country, when the
+Emperor, becoming alarmed at the distress and
+loss of property occasioned, endeavored to enter
+into a sort of compact with it by burning paper
+and offering food upon the sea-wall. This, however,
+did not have the desired effect, as the high
+tide came in as before; and it was at last determined
+\PageSep{69}
+\ifthenelse{\boolean{ForPrinting}}{%
+ \TallFig[0.7]{19}{Pictures of the Bore on the Tsien-Tang-Kiang}{jpg}
+}{%
+ \Figure[0.75]{19}{Pictures of the Bore on the Tsien-Tang-Kiang}{jpg}
+}
+\index{Wharton, Sir W. J., illustration of bore}%
+\PageSep{70}
+to erect a pagoda at the spot where the
+worst breach in the embankment had been made.
+Hence the origin of the Bhota Pagoda. A
+pagoda induces the good \textit{fungshui}, or spirit.
+After it was built the flood tide, though it still
+continued to come in the shape of a bore, did
+not flood the country as before.''
+
+We ``foreign devils'' may take the liberty of
+suspecting that the repairs to the embankment
+had also some share in this beneficial result.
+
+This story is remarkable in that it refers to
+the reign of an Emperor whose historical existence
+is undoubted. It thus differs from many
+of the mythical stories which have been invented
+by primitive peoples to explain great natural
+phenomena. There is good reason to suppose,
+in fact, that this bore had no existence some centuries
+ago; for Marco Polo, in the thirteenth
+\index{Marco Polo, resident of Hangchow}%
+century, stayed about a year and a half at
+Hangchow, and gives so faithful and minute
+\index{Hangchow, the bore at|)}%
+an account of that great town that it is almost
+impossible to believe that he would have omitted
+to notice a fact so striking. But the Emperor
+referred to in the Chinese legend reigned some
+centuries before the days of Marco Polo, so that
+we have reason to believe that the bore is intermittent.
+\index{Bore!Chinese superstition|)}%
+I have also learned from Captain
+Moore himself that at the time of the great
+\index{Moore, Captain!survey of Tsien-Tang-Kiang|)}%
+Taiping rebellion, the suppression of which was
+principally due to ``Chinese'' Gordon, the intensity
+\index{Chinese!superstition as to bore|)}%
+\index{Tsien-Tang-Kiang, the bore in|)}%
+\PageSep{71}
+of the bore was far less than it is to-day.
+\index{Bore!rivers where found}%
+This shows that the bore is liable to great variability,
+according as the silting of the estuary
+changes.
+
+The people at Haining still continue to pay
+religious reverence to the bore, and on one of
+the days when Captain Moore was making observations
+some five or six thousand people assembled
+on the river-wall to propitiate the god of
+the waters by throwing in offerings. This was
+the occasion of one of the highest bores at spring
+tide, and the rebound of the bore from the sea-wall,
+and the sudden heaping up of the waters
+as the flood conformed to the narrow mouth of
+the river, here barely a mile in width at low
+water, was a magnificent spectacle. A series of
+breakers were formed on the back of the advancing
+flood, which for over five minutes were not
+less than twenty-five feet above the level of the
+river in front of the bore. On this occasion
+Captain Moore made a rough estimate that a
+million and three quarters of tons of water passed
+the point of observation in one minute.
+
+The bore of which I have given an account is
+perhaps the largest known; but relatively small
+ones are to be observed on the Severn and Wye
+\index{Severn, bore in the}%
+\index{Wye, bore in the}%
+in England, on the Seine in France, on the Petitcodiac
+\index{Petitcodiac, bore in the}%
+\index{Seine, bore in the}%
+in Canada, on the Hugli in India, and
+\index{Hugli, bore on the}%
+doubtless in many other places. In general,
+however, it is only at spring tides and with certain
+\PageSep{72}
+winds that the phenomenon is at all striking.
+In September,~1897, I was on the banks of the
+Severn at spring tide; but there was no proper
+bore, and only a succession of waves up-stream,
+\index{Bore!causes}%
+and a rapid rise of water-level.
+
+I have shown, at the beginning of this chapter,
+that the heading back of the sea water by
+the natural current of a river, and the progressive
+change of shape of a wave in shallow water combine
+to produce a rapid rise of the tide in rivers.
+But the explanation of the bore, as resulting
+from these causes, is incomplete, because it leaves
+their relative importance indeterminate, and
+serves rather to explain a rapid rise than an absolutely
+sudden one. I think that it would be
+impossible, from the mere inspection of an estuary,
+to say whether there would be a bore there;
+we could only say that the situation looked
+promising or the reverse.
+
+The capriciousness of the appearance of the
+bore proves in fact that it depends on a very nice
+balance between conflicting forces, and the irregularity
+in the depth and form of an estuary renders
+the exact calculation of the form of the
+rising tide an impossibility. It would be easy
+to imitate the bore experimentally on a small
+scale; but, as in many other physical problems,
+we must rest satisfied with a general comprehension
+of the causes which produce the observed
+result.
+\PageSep{73}
+
+The manner in which the Chinese avail themselves
+of the after-rush for ascending the river
+affords an illustration of the utilization by mankind
+of tidal energy. In going up-stream, a
+\index{Energy, tidal, utilization of}%
+barge, say of one hundred tons, may rise some
+twenty or thirty feet. There has, then, been
+done upon that barge a work of from two to
+three thousand foot-tons. Whence does this
+energy come? Now, I say that it comes from
+the rotation of the earth; for we are making the
+tide do the work for us, and thus resisting the
+tidal movement. But resistance to the tide has
+the effect of diminishing the rate at which the
+earth is spinning round. Hence it is the earth's
+rotation which carries the barge up the river, and
+we are retarding the earth's rotation and making
+the day infinitesimally longer by using the tide
+in this way. This resistance is of an analogous
+character to that due to tidal friction, the consideration
+of which I must defer to a future
+chapter, as my present object is to consider the
+uses which may be made of tidal energy.
+
+It has been supposed by many that when the
+coal supply of the world has been exhausted we
+shall fall back on the tides to do our work. But
+a little consideration will show that although this
+source of energy is boundless, there are other far
+more accessible funds on which to draw.
+
+I saw some years ago a suggestion that the
+rise and fall of old hulks on the tide would afford
+\PageSep{74}
+\index{Energy, tidal, utilization of}%
+serviceable power. If we picture to ourselves the
+immense weight of a large ship, we may be deluded
+for a moment into agreement with this
+project, but numerical calculation soon shows its
+futility. The tide takes about six hours to rise
+from low water to high water, and the same
+period to fall again. Let us suppose that the
+water rises ten feet, and that a hulk of $10,000$
+tons displacement is floating on it; then it is
+easy to show that only twenty horse-power will
+be developed by its rise and fall. We should
+then require ten such hulks to develop as much
+work as would be given by a steam engine of
+very moderate size, and the expense of the installation
+would be far better bestowed on water-wheels
+in rivers or on wind-mills. I am glad to
+\index{Mills worked by the tide}%
+say that the projector of this scheme gave it up
+when its relative insignificance was pointed out
+to him. It is the only instance of which I ever
+heard where an inventor was deterred by the impracticability
+of his plan.
+
+We may, then, fairly conclude that, with existing
+mechanical appliances, the attempt to utilize
+the tide on an open coast is futile. But
+where a large area of tidal water can be easily
+trapped at high water, its fall may be made to
+work mill-wheels or turbines with advantage.
+The expense of building long jetties to catch the
+water is prohibitive, and therefore tide mills are
+only practicable where there exists an easily
+\PageSep{75}
+adaptable configuration of shoals in an estuary.
+There are, no doubt, many such mills in the
+\index{Mills worked by the tide}%
+world, but the only one which I happen to have
+seen is at Bembridge, in the Isle of Wight. At
+this place embankments formed on the natural
+shoals are furnished with lock-gates, and inclose
+many acres of tidal water. The gates open automatically
+with the rising tide, and the incipient
+outward current at the turn of the tide closes
+the gates again, so that the water is trapped.
+The water then works a mill wheel of moderate
+size. When we reflect on the intermittence of
+work from low water to high water and the great
+inequality of work with springs and neaps, it
+may be doubted whether this mill is worth the
+expense of retaining the embankments and lock-gates.
+
+We see then that, notwithstanding the boundless
+energy of the tide, rivers and wind and fuel
+are likely for all time to be incomparably more
+important for the use of mankind.
+
+\begin{Authorities}
+On waves in rivers see Airy's article on \Title{Tides and Waves} in
+\index{Airy, Sir G. B.!tides in rivers}%
+\index{Rivers!Airy on tide in}%
+the ``Encyclopædia Metropolitana.'' Some of his results will
+also be found in the article \Title{Tides} in the ``Encyclopædia Britannica.''
+
+Commander Moore, R.~N., \Title{Report on the Bore of the Tsien-Tang-Kiang.}
+Sold by Potter, Poultry, London,~1888.
+
+\Title{Further Report},~\&c., by the same author and publisher,~1893.
+\end{Authorities}
+\PageSep{76}
+
+
+\Chapter{IV}
+{Historical Sketch}
+
+\First{I Cannot} claim to have made extensive investigations
+\index{History!of tidal theories|(}%
+as to the ideas of mankind at different
+periods on the subject of the tides, but I propose
+in the present chapter to tell what I have
+been able to discover.
+
+No doubt many mythologies contain stories
+explanatory of the obvious connection between
+the moon and the tide. But explanations, professing
+at least to be scientific, would have been
+brought forward at periods much later than
+those when the mythological stories originated,
+and I shall only speak of the former.
+
+I have to thank my colleagues at Cambridge
+for the translations from the Chinese, Arabic,
+\index{Chinese!theories of tide}%
+Icelandic, and classical literatures of such passages
+as they were able to discover.
+
+I learn from Professor Giles that Chinese
+\index{Giles on Chinese theories of the tide}%
+writers have suggested two causes for the tides:
+first, that water is the blood of the earth, and
+that the tides are the beating of its pulse; and
+secondly, that the tides are caused by the earth
+breathing. Ko~Hung, a writer of the fourth
+century of our era, gives a somewhat obscure
+\PageSep{77}
+\index{Chinese!theories of tide}%
+\index{Giles on Chinese theories of the tide}%
+explanation of spring and neap tides. He says
+that every month the sky moves eastward and
+then westward, and hence the tides are greater
+and smaller alternately. Summer tides are said
+to be higher than winter tides, because in summer
+the sun is in the south and the sky is $15,000$~li
+($5,000$~miles) further off, and therefore in
+summer the female or negative principle in nature
+is weak, and the male or positive principle
+strong.
+
+In China the diurnal inequality is such that
+in summer the tide rises higher in the daytime
+than in the night, whilst the converse is true
+in winter. I suggest that this fact affords the
+justification for the statement that the summer
+tides are great.
+
+\TB
+
+Mr.~E.~G. Browne has translated for me the
+\index{Arabian theories of tide|(}%
+\index{Browne, E. G., Arabian theories of tide|(}%
+following passage from the ``Wonders of Creation''
+of Zakariyy\bar{a} ibn Muhammad ibn Mahm\bar{u}d
+al Qazv\bar{i}n\bar{i}, who died in \AD~1283.\footnote
+ {Wüstenfeld's edition, pp.~103,~104.}
+
+``Section treating of certain wonderful conditions
+of the sea.
+
+``Know that at different periods of the four
+seasons, and on the first and last days of the
+months, and at certain hours of the night and
+day, the seas have certain conditions as to the
+rising of their waters and the flow and agitation
+thereof.
+\PageSep{78}
+
+``As to the rising of the waters, it is supposed
+that when the sun acts on them it rarefies them,
+and they expand and seek a space ampler than
+that wherein they were before, and the one part
+repels the other in the five directions eastwards,
+westwards, southwards, northwards, and upwards,
+and there arise at the same time various winds
+on the shores of the sea. This is what is said
+as to the cause of the rising of the waters.
+
+``As for the flow of certain seas at the time
+of the rising of the moon, it is supposed that at
+the bottom of such seas there are solid rocks
+and hard stones, and that when the moon rises
+over the surface of such a sea, its penetrating
+rays reach these rocks and stones which are at
+the bottom, and are then reflected back thence;
+and the waters are heated and rarefied and seek
+an ampler space and roll in waves towards the
+seashore~\dots\ and so it continues as long as
+the moon shines in mid-heaven. But when she
+begins to decline, the boiling of the waters
+ceases, and the particles cool and become dense
+and return to their state of rest, and the currents
+run according to their wont. This goes
+on until the moon reaches the western horizon,
+when the flow begins again, as it did when the
+moon was in the eastern horizon. And this
+flow continues until the moon is at the middle
+of the sky below the horizon, when it ceases.
+Then when the moon comes upward, the flow
+\PageSep{79}
+begins again until she reaches the eastern horizon.
+This is the account of the flow and ebb
+of the sea.
+
+``The agitation of the sea resembles the agitation
+of the humours in men's bodies, for verily
+as thou seest in the case of a sanguine or bilious
+man,~\&c., the humours stirring in his body, and
+then subsiding little by little; so likewise the
+sea has matters which rise from time to time as
+they gain strength, whereby it is thrown into
+violent commotion which subsides little by little.
+And this the Prophet (on whom be the blessings
+of God and his peace) hath expressed in a poetical
+manner, when he says: `Verily the Angel,
+who is set over the seas, places his foot in the
+sea and thence comes the flow; then he raises it
+and thence comes the ebb.'\,''
+\index{Arabian theories of tide|)}%
+\index{Browne, E. G., Arabian theories of tide|)}%
+
+\TB
+
+Mr.~Magnússon has kindly searched the old
+\index{Icelandic theory of tides}%
+\index{Magnússon on Icelandic theories of tides}%
+Icelandic literature for references to the tides.
+In the Rimbegla he finds this passage:---
+
+``Beda the priest says that the tides follow
+the moon, and that they ebb through her blowing
+on them, but wax in consequence of her
+movement.''
+
+And again:---
+
+``(At new moon) the moon stands in the way
+of the sun and prevents him from drying up the
+sea; she also drops down her own moisture.
+For both these reasons, at every new moon, the
+\PageSep{80}
+\index{Icelandic theory of tides}%
+\index{Magnússon on Icelandic theories of tides}%
+ocean swells and makes those tides which we call
+spring tides. But when the moon gets past the
+sun, he throws down some of his heat upon
+the sea, and diminishes thereby the fluidity of
+the water. In this way the tides of the sea
+are diminished.''
+
+In another passage the author writes:---
+
+``But when the moon is opposite to the sun,
+the sun heats the ocean greatly, and as nothing
+impedes that warmth, the ocean boils and the
+sea flood is more impetuous than before---just
+as one may see water rise in a kettle when it
+boils violently. This we call spring tide.''
+
+There seems to be a considerable inconsistency
+in explaining one spring tide by the interception
+of the sun's heat by the moon, and the next one
+by the excess of that heat.
+
+But it is not necessary to search ancient literature
+for grotesque theories of the tides. In
+1722 E.~Barlow, gentleman, in ``An Exact Survey
+of the Tide,''\footnote
+ {``The Second Edition, with Curious Maps.'' (London: John
+ Hooke, 1722.)}
+attributes it to the pressure
+of the moon on the atmosphere. And theories
+not less absurd have been promulgated during
+the last twenty years.
+
+\TB
+
+The Greeks and Romans, living on the shores
+of the Mediterranean, had not much occasion to
+learn about the tide, and the passages in classical
+\PageSep{81}
+\index{Greek!theory and description of tides|(}%
+\index{Roman description of tides|(}%
+literature which treat of this matter are but
+few. But where the subject is touched on we
+see clearly their great intellectual superiority over
+those other peoples, whose ideas have just been
+quoted.
+
+The only author who treats of the tide in any
+\index{Strabo on tides|(}%
+detail is Posidonius, and we have to rely for our
+\index{Posidonius on tides|(}%
+knowledge of his work entirely on quotations
+from him by Strabo.\footnote
+ {My attention was drawn to Strabo by a passage in Sir W.
+ Thomson's (Lord Kelvin's) Popular Lectures, \Title{The Tides}, vol.~ii.
+ I have to thank Mr.~Duff for the translations which follow from
+ Strabo and Posidonius. The work consulted was Bake's \Title{Posidonius}
+ (Leiden,~1810), but Mr.~Duff tells me that the text is very
+ corrupt in some places, and he has therefore also consulted a
+ more recent text.}
+
+Posidonius says that Aristotle attributed the
+\index{Aristotle on tides}%
+flow and ebb of the sea at Cadiz to the mountainous
+formation of the coast, but he very justly
+pronounces this to be nonsense, particularly as
+the coast of Spain is flat and sandy. He himself
+attributes the tides to the moon's influence, and
+the accuracy of his observations is proved by the
+following interesting passage from Strabo:\footnotemark---
+\footnotetext{Teubner's \Title{Strabo},~i.\ p.~236.}
+
+``Posidonius says that the movement of the
+ocean observes a regular series like a heavenly
+body, there being a daily, monthly, and yearly
+movement according to the influence of the
+moon. For when the moon is above the (eastern)
+horizon by the distance of one sign of the
+zodiac (\ie~$30°$) the sea begins to flow, and encroaches
+\PageSep{82}
+visibly on the land until the moon
+reaches the meridian. When she has passed the
+meridian, the sea in turn ebbs gradually, until
+the moon is above the western horizon by the
+distance of one sign of the zodiac. The sea then
+remains motionless while the moon is actually
+setting, and still more so (\textit{sic}) so long as the
+moon is moving beneath the earth as far as a
+sign of the zodiac beneath the horizon. Then
+the sea again advances until the moon has
+reached the meridian below the earth; and retreats
+while the moon is moving towards the east,
+until she is the distance of a sign of the zodiac
+below the horizon; it remains at rest until the
+moon is the same distance above the horizon, and
+then begins to flow again. Such is the daily
+movement of the tides, according to Posidonius.
+
+``As to their monthly movement, he says that
+the ebbs are greatest at the conjunctions [of
+the sun and moon], and then grow less until the
+time of half moon, and increase again until the
+time of full moon, and grow less again until
+the moon has waned to half. Then the increase
+of the tide follows until the conjunction. But
+the increases last longer and come quicker [this
+phrase is very obscure].
+
+``The yearly movements of the tides he says
+he learned from the people of Cadiz. They told
+him that the ebb and flow alike were greatest at
+the summer solstice. He guesses for himself
+\PageSep{83}
+that the tides grow less from the solstice to the
+equinox, and then increase between the equinox
+and the winter solstice, and then grow less until
+the spring equinox, and greater until the summer
+solstice.''
+
+This is an excellent account of the tides at
+Cadiz, but I doubt whether there is any foundation
+\index{Polibius on tides at Cadiz}%
+for that part which was derived from hearsay.
+Lord Kelvin remarks, however, that it is interesting
+to note that inequalities extending over
+the year should have been recognized.
+
+Strabo also says that there was a spring near
+Cadiz in which the water rose and fell, and that
+this was believed by the inhabitants, and by
+Polybius, to be due to the influence of the ocean
+tide, but Posidonius was not of this opinion.
+Strabo says:---
+
+``Posidonius denies this explanation. He says
+there are two wells in the precinct of Hercules at
+Cadiz, and a third in the city. Of the two former
+the smaller runs dry while people are drawing
+water from it, and when they stop drawing water
+it fills again; the larger continues to supply
+water all day, but, like all other wells, it falls
+during the day but is replenished at night, when
+the drawing of water has ceased. But since the
+ebb tide often coincides with the replenishing of
+the well, therefore, says Posidonius, the idle story
+of the tidal influence has been believed by the
+inhabitants.''
+\PageSep{84}
+
+Since the wells follow the sun, whilst the tide
+follows the moon, the criticism of Posidonius is
+a very just one. But Strabo blames him for
+distrusting the Cadizians in a simple matter of
+everyday experience, whilst accepting their evidence
+as to an annual inequality in the tides.
+
+There is another very interesting passage in
+Strabo, the meaning of which was obviously unknown
+to the Dutch commentator Bake---and
+indeed must necessarily have been unintelligible
+to him at the time when he wrote, on account of
+the then prevailing ignorance of tidal phenomena
+in remoter parts of the world. Strabo
+writes:---
+
+``Anyhow Posidonius says that Seleucus of
+\index{Diurnal inequality!observed by Seleucus}%
+\index{Posidonius on tides|)}%
+\index{Seleucus, observation of tides of Indian Ocean}%
+the Red Sea [also called the Babylonian] declares
+that there is a certain irregularity and regularity
+in these phenomena [the tides], according
+to the different positions [of the moon] in the
+zodiac. While the moon is in the equinoctial
+signs, the phenomena are regular; but while she
+is in the signs of the solstices, there is irregularity
+both in the height and speed of the tides,
+and in the other signs there is regularity or the
+reverse in proportion to their nearness to the solstices
+or to the equinoxes.''
+
+Now let us consider the meaning of this.
+When the moon is in the equinoxes she is on
+the equator, and when she is in the solstices she
+is at her maximum distances to the north or
+\PageSep{85}
+south of the equator---or, as astronomers say, in
+her greatest north or south declination. Hence
+Seleucus means that, when the moon is on the
+\index{Seleucus, observation of tides of Indian Ocean}%
+equator, the tides follow one another, with two
+equal high and low waters a day; but when she
+is distant from the equator, the regular sequence
+is interrupted. In other words, the diurnal
+inequality (which I shall explain in a later chapter)
+vanishes when the moon is on the equator,
+and is at its maximum when the declination is
+greatest. This is quite correct, and since the
+diurnal inequality is almost evanescent in the
+\index{Diurnal inequality!observed by Seleucus}%
+Atlantic, whilst it is very great in the Indian
+Ocean, especially about Aden, it is clear that
+Seleucus had watched the sea there, just as we
+should expect him to do from his place of origin.
+
+\TB
+
+Many centuries elapsed after the classical
+period before any scientific thought was bestowed
+on the tides. Kepler recognized the
+\index{Kepler!ideas concerning tides}%
+tendency of the water on the earth to move
+towards the sun and the moon, but he was unable
+to submit his theory to calculation. Galileo
+\index{Galileo!blames Kepler for his tidal theory}%
+expresses his regret that so acute a man as
+Kepler should have produced a theory, which
+appeared to him to reintroduce the occult qualities
+of the ancient philosophers. His own explanation
+referred the phenomenon to the rotation
+of the earth, and he considered that it afforded
+a principal proof of the Copernican system.
+\index{Greek!theory and description of tides|)}%
+\index{Roman description of tides|)}%
+\index{Strabo on tides|)}%
+\PageSep{86}
+\index{Kepler!ideas concerning tides}%
+
+The theory of tide-generating force which will
+be set forth in \Ref{Chapter}{V}.\ is due to Newton,
+\index{Newton!founder of tidal theory}%
+who expounded it in his ``Principia'' in~1687.
+His theory affords the firm basis on which all
+subsequent work has been laid.
+
+In 1738 the Academy of Sciences of Paris
+offered the theory of the tides as the subject for
+a prize. The authors of four essays received
+prizes, viz., Daniel Bernoulli, Euler, Maclaurin,
+\index{Bernoulli, Daniel, essay on tides}%
+\index{Euler, essay on tides}%
+\index{Maclaurin!essay on tides}%
+and Cavalleri. The first three adopted, not only
+\index{Cavalleri, essay on tides}%
+the theory of gravitation, but also Newton's
+theory to its fullest extent. A considerable
+portion of Bernoulli's work is incorporated in
+the account of the theory of the tides which I
+shall give later. The essays of Euler and Maclaurin
+contained remarkable advances in mathematical
+knowledge, but did not add greatly to
+the theory of the tides. The Jesuit priest
+Cavalleri adopted the theory of vortices to explain
+the tides, and it is not worth while to
+follow him in his erroneous and obsolete speculations.
+
+Nothing of importance was added to our
+knowledge until the great French mathematician
+Laplace took up the subject in~1774. It was he
+\index{Laplace!theory of tides|(}%
+who for the first time fully recognized the difficulty
+of the problem, and showed that the earth's
+rotation is an essential feature in the conditions.
+The actual treatment of the tidal problem is in
+effect due to Laplace, although the mode of
+\PageSep{87}
+presentment of the theory has come to differ
+considerably from his.
+
+Subsequently to Laplace, the most important
+workers in this field have been Sir John Lubbock
+\index{Lubbock, Sir J., senior, on tides}%
+senior, Whewell, Airy, and Lord Kelvin.
+\index{Kelvin, Lord!initiates harmonic analysis}%
+\index{Whewell!on tides}%
+\index{Whewell!empirical construction of tide tables|(}%
+The work of Lubbock and Whewell is chiefly
+remarkable for the coördination and analysis of
+enormous masses of data at various ports, and
+the construction of trustworthy tide tables.
+Airy contributed an important review of the
+whole tidal theory. He also studied profoundly
+the theory of waves in canals, and considered
+the effects of frictional resistances on the progress
+of tidal and other waves.
+
+Lord Kelvin initiated a new and powerful
+method of considering tidal oscillations. His
+method possesses a close analogy with that already
+used in discussing the irregularities in the
+motions of the moon and planets. His merit
+consists in the clear conception that the plan of
+procedure which has been so successful in the
+one case would be applicable to the other. The
+difference between the laws of the moon's motion
+and those of tidal oscillations is, however,
+so great that there is scarcely any superficial
+resemblance between the two methods. This
+so-called ``harmonic analysis'' of the tides is
+\index{Harmonic analysis!initiated by Lord Kelvin}%
+daily growing in favor in the eyes of men of
+science, and is likely to supersede all the older
+methods. I shall explain it in a future chapter.
+\PageSep{88}
+
+Amongst all the grand work which has been
+bestowed on this difficult subject, Newton stands
+out first, and next to him we must rank Laplace.
+However original any future contribution to the
+science of tides may be, it would seem as though
+it must perforce be based on the work of these
+two. The exposition which I shall give hereafter
+of the theory of oceanic tides is based on
+the work of Newton, Bernoulli, Laplace, and
+\index{Bernoulli, Daniel, essay on tides}%
+\index{Laplace!theory of tides|)}%
+Kelvin, in proportions of which it would be
+difficult to assign the relative importance.
+
+\TB
+
+The connection between the moon and the
+tide is so obvious that long before the formulation
+of a satisfactory theory fairly accurate predictions
+of the tides were made and published.
+On this head Whewell\footnote
+ {\Title{History of the Inductive Sciences}, 1837, vol.~ii.\ p.~248 \textit{et~seq.}}
+\index{History!of tidal theories|)}%
+has the following interesting
+passage:---
+
+``The course which analogy would have recommended
+for the cultivation of our knowledge of
+tides would have been to ascertain by an analysis
+of long series of observations, the effects of
+changes in the time of transit, parallax, and
+declination of the moon, and thus to obtain the
+laws of phenomena; and then to proceed to
+investigate the laws of causation.
+
+``Though this was not the course followed by
+mathematical theorists, it was really pursued by
+those who practically calculated tide tables; and
+\PageSep{89}
+the application of knowledge to the useful purposes
+of life, being thus separated from the
+promotion of the theory, was naturally treated
+as a gainful property, and preserved by secrecy.
+\dots~Liverpool, London, and other places, had
+their tide tables, constructed by undivulged
+methods, which methods, in some instances at
+least, were handed down from father to son for
+several generations as a family possession; and
+the publication of new tables accompanied by a
+statement of the mode of calculation was resented
+as an infringement of the rights of property.
+
+``The mode in which these secret methods
+were invented was that which we have pointed
+out,---the analysis of a considerable series of
+observations. Probably the best example of this
+was afforded by the Liverpool tide tables. These
+were deduced by a clergyman named Holden,
+from observations made at that port by a harbor
+master of the name of Hutchinson, who was
+led, by a love of such pursuits, to observe the
+tides for above twenty years, day and night.
+Holden's tables, founded on four years of these
+observations, were remarkably accurate.
+
+``At length men of science began to perceive
+that such calculations were part of their business;
+and that they were called upon, as the
+guardians of the established theory of the universe,
+to compare it in the greatest possible
+\PageSep{90}
+detail with the facts. Mr.~Lubbock was the
+first mathematician who undertook the extensive
+labors which such a conviction suggested. Finding
+that regular tide observations had been made
+at the London docks from~1795, he took nineteen
+years of these (purposely selecting the
+length of the cycle of the motions of the lunar
+orbit), and caused them (in~1831) to be analyzed
+by Mr.~Dessiou, an expert calculator. He thus
+obtained tables for the effect of the moon's
+declination, parallax, and hour of transit, on the
+tides; and was enabled to produce tide tables
+founded upon the data thus obtained. Some
+mistakes in these as first published (mistakes unimportant
+as to the theoretical value of the work)
+served to show the jealousy of the practical tide
+table calculators, by the acrimony with which the
+oversights were dwelt upon; but in a very few
+years the tables thus produced by an open and scientific
+process were more exact than those which
+resulted from any of the secrets; and thus practice
+was brought into its proper subordination to
+theory.''
+\index{Whewell!empirical construction of tide tables|)}%
+
+\begin{Authorities}
+The history from Galileo to Laplace is to be found in the
+\Title{Mécanique Céleste} of Laplace, book~xiii, chapter~i.
+
+The other authorities are quoted in the text or in footnotes.
+\end{Authorities}
+\PageSep{91}
+
+
+\Chapter{V}
+{Tide-generating Force}
+
+\First{It} would need mathematical reasoning to fully
+\index{Centripetal and centrifugal forces|(}%
+explain how the attractions of the sun and moon
+give rise to tide-generating forces. But as this
+\index{Forces!centripetal and centrifugal|(}%
+book is not intended for the mathematician, I
+must endeavor to dispense with technical language.
+
+A body in motion will move in a straight line,
+unless it is deflected from its straight path by
+some external force, and the resistance to the
+deflection is said to be due to inertia. The motion
+of the body then is equivalent in its effect
+to a force which opposes the deflection due to
+the external force, and in many cases it is permissible
+to abstract our attention from the motion
+of the system and to regard it as at rest, if
+at the same time we introduce the proper ideal
+forces, due to inertia, so that they shall balance
+the action of the real external forces.
+
+If I tie a string to a stone and whirl it round,
+the string is thrown into a state of tension. The
+natural tendency of the stone, at each instant, is
+to move onward in a straight line, but it is continuously
+deflected from its straight path by the
+\PageSep{92}
+tension of the string. In this case the ideal
+force, due to inertia, whereby the stone resists
+its continuous deflection, is called centrifugal
+force. This force is in reality only a substitute
+for the motion, but if we withdraw our attention
+from the motion, it may be regarded as a reality.
+
+The centrifugal force is transmitted to my
+hand through the string, and I thus experience
+an outward or centrifugal tendency. But the
+stone itself is continually pulled inward by the
+string, and the force is called centripetal. When
+a string is under tension, as in this experiment,
+it is subject to equal and opposite forces, so that
+the tension implies the existence of a pair of
+forces, one towards and the other away from the
+centre of rotation. The force is to be regarded
+as away from the centre when we consider the
+sensation of the whirler, and as towards the centre
+when we consider the thing whirled. A similar
+double view occurs in commerce, where a
+transaction which stands on the credit side in the
+books of one merchant appears on the debit side
+in the books of the other.
+
+This simple experiment exemplifies the mechanism
+by which the moon is kept revolving round
+the earth. There is not of course any visible
+connection between the two bodies, but an invisible
+bond is provided by the attraction of gravity,
+which replaces the string which unites the
+stone to the hand. The moon, then, whirls
+\PageSep{93}
+\index{Forces!tide-generating|(}%
+\index{Orbit!of moon and earth|(}%
+round the earth at just such a rate and at just
+\index{Earth and moon!diagram}%
+such a distance, that her resistance to circular
+motion, called centrifugal force, is counterbalanced
+by the centripetal tendency of gravity. If
+\index{Centripetal and centrifugal forces|)}%
+she were nearer to us the attraction of gravity
+would be greater, and she would have to go
+round the earth faster, so as to make enough
+centrifugal force to counterbalance the greater
+\Figure[0.8]{20}{Earth and Moon}{png}
+gravity. The converse would be true, and the
+moon would go round slower, if she were further
+from us.
+
+The moon and the earth go round the sun in
+companionship once in a year, but this annual
+motion does not affect the interaction between
+them, and we may put aside the orbital motion
+of the earth, and suppose the moon and earth to
+\index{Moon and earth!diagram}%
+be the only pair of bodies in existence. When
+the principle involved in a purely lunar tide is
+grasped, the action of the sun in producing a
+\index{Forces!centripetal and centrifugal|)}%
+\PageSep{94}
+solar tide will become obvious. But the analogy
+of the string and stone is imperfect in one
+respect where the distinction is important; the
+moon, in fact, does not revolve exactly about
+the earth, but about the centre of gravity of
+the earth and moon. The earth is eighty times
+as heavy as the moon, and so this centre of gravity
+is not very far from the earth's centre. The
+upper part of \fig{20} is intended to represent a
+planet and its satellite; the lower part shows
+the earth and the moon in their true proportions.
+The upper figure is more convenient for
+our present argument, and the planet and satellite
+may be described as the earth and the moon,
+notwithstanding the exaggeration of their relative
+proportions. The point~$G$ is the centre of
+gravity of the two, and the axis about which
+they revolve passes through~$G$. This point is
+sufficiently near to the centre of the earth to
+permit us, for many purposes, to speak of the
+moon as revolving round the earth. But in the
+present case we must be more accurate and must
+regard the moon and earth as revolving round~$G$,
+their centre of gravity. The moon and earth
+are on opposite sides of this point, and describe
+circles round it. The distance of the moon's
+centre from~$G$ is $237,000$~miles, whilst that of
+the earth's centre is only $\DPchg{3000}{3,000}$~miles in the opposite
+direction. The $\DPchg{3000}{3,000}$~and $237,000$~miles
+together make up the $240,000$~miles which separate
+the centres of the two bodies.
+\PageSep{95}
+
+A system may now be devised so as to resemble
+the earth and moon more closely than that
+of the string and stone with which I began. If
+a large stone and a small one are attached to one
+another by a light and stiff rod, the system can
+be balanced horizontally about a point in the rod
+called the centre of gravity~$G$. The two weights
+may then be set whirling about a pivot at~$G$, so
+that the rod shall always be horizontal. In consequence
+of the rotation the rod is brought into
+a state of stress, just as was the string in the
+first example, and the centripetal stress in the
+rod exactly counterbalances the centrifugal force.
+The big and the little stones now correspond to
+the earth and the moon, and the stress in the rod
+plays the same part as the invisible bond of
+gravity between the earth and the moon. Fixing
+our attention on the smaller stone or moon
+at the end of the longer arm of the rod, we see
+that the total centrifugal force acting on the
+moon, as it revolves round the centre of gravity,
+is equal and opposite to the attraction of the
+earth on the moon. On considering the short
+arm of the rod between the pivot and the big
+stone, we see also that the centrifugal force acting
+on the earth is equal and opposite to the
+attraction of the moon on it. In this experiment
+as well as in the former one, we consider
+the total of centrifugal force and of attraction,
+but every particle of both the celestial bodies is
+\index{Orbit!of moon and earth|)}%
+\PageSep{96}
+\index{Davis, method of presenting tide-generating force}%
+acted on by these forces, and accordingly a
+closer analysis is necessary.
+
+It will now simplify matters if we make a supposition
+which departs from actuality, introducing
+the true conditions at a later stage in the
+argument.
+
+The earth's centre describes a circle about the
+centre of gravity~$G$, with a radius of $\DPchg{3000}{3,000}$~miles,
+and the period of the revolution is of course one
+month. Now whilst this motion of revolution
+of the earth's centre continues, let it be supposed
+that the diurnal rotation is annulled. As this
+is a mode of revolution which differs from that
+of a wheel, it is well to explain exactly what is
+meant by the annulment of the diurnal rotation.
+This is illustrated in~\fig{21}, which shows the
+successive positions assumed by an arrow in revolution
+without rotation. The shaft of the arrow
+always remains parallel to the same direction in
+space, and therefore it does not rotate, although
+the whole arrow revolves. It is obvious that every
+particle of the arrow describes a circle of the
+same radius, but that the circles described by
+them are not concentric. The circles described
+by the point and by the base of the arrow are
+shown in the figure, and their centres are separated
+by a distance equal to the length of the
+arrow. Now the centrifugal force on a revolving
+particle acts along the radius of the circle described,
+and in this case the radii of the circles
+\PageSep{97}
+described by any two particles in the arrow are
+always parallel. The parallelism of the centrifugal
+forces at the two ends of the arrow is
+indicated in the figure. Then again, the centrifugal
+force must everywhere be equal as well
+as parallel, because its intensity depends both on
+the radius and on the speed of revolution, and
+these are the same for every part. It follows
+that if a body revolves without rotation, every
+part of it is subject to equal and parallel centrifugal
+forces. The same must therefore be
+true of the earth when deprived of diurnal rotation.
+Accordingly every particle of the idealized
+non-rotating earth is continuously subject to
+equal and parallel centrifugal forces, in consequence
+of the revolution of the earth's centre
+in its monthly orbit with a radius of $\DPchg{3000}{3,000}$~miles.\footnote
+ {I owe the suggestion of this method of presenting the origin
+ of tide-generating force to Professor Davis of Harvard
+\index{Davis, method of presenting tide-generating force}%
+ University.}
+
+We have seen that the total of centrifugal
+force acting on the whole earth must be just
+such as to balance the total of the centripetal
+forces due to the moon's attraction. If, then,
+the attractional forces, acting on every particle
+of the earth, were also equal and parallel, there
+would be a perfect balance throughout. We
+shall see, however, that although there is a perfect
+balance on the whole, there is not that uniformity
+\PageSep{98}
+which would render the balance perfect
+at every particle.
+
+As far as concerns the totality of the attraction
+the analogy is complete between the larger
+stone, revolving at the end of the shorter arm
+of the rod, and the earth revolving in its small
+\Figure[0.7]{21}{Revolution of a Body without Rotation}{png}
+orbit round~$G$. But a difference arises when we
+compare the distribution of the tension of the
+rod with that of the lunar attraction; for the
+rod only pulls at the stone at the point where it
+is attached to it, whereas the moon attracts every
+particle of the earth. She does not, however,
+attract every particle with equal force, for she
+pulls the nearer parts more strongly than the
+further, as is obvious from the nature of the law
+of gravitation. The earth's centre is distant
+sixty times its radius from the moon, so that the
+nearest and furthest parts are distant fifty-nine
+\PageSep{99}
+and sixty-one radii respectively. Hence the attractions
+at the nearest and furthest parts differ
+only a little from the average, namely, that at
+the centre; but it is just these small differences
+which are important in this matter.
+
+Since on the whole the attractions and the centrifugal
+forces are equal and opposite, and since
+the centrifugal forces acting on the non-rotating
+earth are equal and parallel at every part, and
+since the attraction at the earth's centre is the
+average attraction, it follows that where the attraction
+is stronger than the average it overbalances
+the centrifugal force, and where it is weaker
+it is overbalanced thereby.
+
+The result of the contest between the two sets
+of forces is illustrated in~\fig{22}. The circle
+represents a section of the earth, and the moon
+is a long way off in the direction~$M$.
+
+Since the moon revolves round the earth,
+whilst the earth is still deprived of rotation, the
+figure only shows the state of affairs at a definite
+instant of time. The face which the earth exhibits
+to the moon is always changing, and the
+moon returns to the same side of the earth only
+at the end of the month. Hence the section of
+the earth shown in this figure always passes
+through the moon, while it is continually shifting
+with respect to the solid earth. The arrows in
+the figure show by their directions and lengths
+the magnitudes and directions of the overbalance
+\PageSep{100}
+in the contest between centrifugal and centripetal
+tendencies. The point~$V$ in the figure is
+the middle of the hemisphere, which at the moment
+portrayed faces full towards the moon. It
+\Figure[0.7]{22}{Tide-Generating Force}{png}
+is the middle of the round disk which the man in
+the moon looks at. The middle of the face invisible
+to the man in the moon is at~$I$. The
+point of the earth which is only fifty-nine earth's
+radii from the moon is at~$V$. Here attraction
+overbalances centrifugal force, and this is indicated
+by an arrow pointing towards the moon.
+The point distant sixty-one earth's radii from
+the moon is at~$I$, and attraction is here overbalanced,
+as indicated by the arrow pointing away
+from the moon.
+
+I shall have to refer hereafter to the intensities
+\PageSep{101}
+of these forces, and will therefore here pause
+to make some numerical calculations.
+
+The moon is distant from the earth's centre
+sixty times the earth's radius, and the attraction
+of gravity varies inversely as the square of the
+distance. Hence we may take $\frac{1}{60^{2}}$ or $\frac{1}{3,600}$ as a
+measure of the intensity of the moon's attraction
+at the earth's centre. The particle which occupies
+the centre of the earth is also that particle
+which is at the average distance of all the particles
+constituting the earth's mass. Hence $\frac{1}{60^{2}}$ or
+$\frac{1}{3,600}$ may be taken as a measure of the average
+attraction of the moon on every particle of the
+earth.
+
+Now the point~$V$ is only distant fifty-nine
+earth's radii from the moon, and therefore, on
+the same scale, the moon attraction is measured
+by $\frac{1}{59^{2}}$ or~$\frac{1}{3,481}$.
+
+The attraction therefore at~$V$ exceeds the average
+by $\frac{1}{59^{2}} - \frac{1}{60^{2}}$, or $\frac{1}{3,481} - \frac{1}{3,600}$. It will be well to
+express these results in decimals; now $\frac{1}{3,481}$ is
+$.000,287,27$, and $\frac{1}{3,600}$ is $.000,277,78$, so that the
+difference is~$.000,009,49$. It is important to
+notice that $\frac{2}{60^{3}}$ or $\frac{2}{216,000}$ is equal to~$.000,009,26$;
+so that the difference is nearly equal to~$\frac{2}{60^{3}}$.
+
+{\Loosen Again, the point~$I$ is distant sixty-one earth's
+radii from the moon, and the moon's attraction
+there is to be measured by $\frac{1}{61^{2}}$ or $\frac{1}{3,721}$. The attraction
+at~$I$ therefore falls below the average by
+$\frac{1}{60^{2}} - \frac{1}{61^{2}}$, or $\frac{1}{3,600} - \frac{1}{3,721}$; that is, by~$.000,277,78 - .000,268,75$,
+\PageSep{102}
+which is equal to~$.000,009,03$.
+This again does not differ much from~$\frac{2}{60^{3}}$.}
+
+These calculations show that the excess of the
+actual attraction at~$V$ above the average attraction
+is nearly equal to the excess of the average
+above the actual attraction at~$I$. These two
+excesses only differ from one another by $5$~per
+cent.\ of either, and they are both approximately
+equal to~$\frac{2}{60^{3}}$ on the adopted scale of measurement.
+
+The use of any particular scale of measurement
+is not material to this argument, and we
+should always find that the two excesses are
+nearly equal to one another. And further, if
+the moon were distant from the earth by any
+other number of earth's radii, we should find
+that the two excesses are each nearly equal to $2$
+divided by the cube of that number.\footnote
+ {\Loosen This argument is very easily stated in algebraic notation.
+ If $x$~be the number of earth's radii at which the moon is
+ placed, the points $V$~and~$I$ are respectively distant $x - 1$ and
+ $x + 1$ radii. Now $(x - 1)^{2}$~is nearly equal to~$x^{2} - 2x$ or to
+ $x^{2}(1 - \frac{2}{x})$, and therefore $\frac{1}{(x - 1)^{2}}$~is nearly equal to~$\frac{1}{x^{2}(1 - \frac{2}{x})}$, which is
+ nearly equal to~$\frac{1}{x^{2}}(1 + \frac{2}{x})$. Hence $\frac{1}{(x - 1)^{2}} - \frac{1}{x^{2}}$~is nearly equal to~$\frac{2}{x^{3}}$.
+ By a similar argument $(x + 1)^{2}$~is nearly equal to~$x^{2}(1 + \frac{2}{x})$,
+ and $\frac{1}{(x + 1)^{2}}$~is nearly equal to~$\frac{1}{x^{2}}(1 - \frac{2}{x})$; so that $\frac{1}{x^{2}} - \frac{1}{(x + 1)^{2}}$~is nearly
+ equal to~$\frac{2}{x^{3}}$.}
+
+We conclude then that the two overbalances
+at $V$~and~$I$, which will be called tide-generating
+forces, are nearly equal to one another, and vary
+\PageSep{103}
+inversely as the cube of the distance of the moon
+from the earth.
+
+The fact of the approximate equality of the
+overbalance or excess on the two sides of the
+earth is noted in the figure by two arrows at $V$
+and $I$ of equal lengths. The argument would
+be a little more complicated, if I were to attempt
+to follow the mathematician in his examination
+of the whole surface of the earth, and to trace
+from point to point how the balance between
+the opposing forces turns. The reader must
+accept the results of such an analysis as shown
+in \fig{22} by the directions and lengths of the
+arrows.
+
+We have already seen that the forces at $V$ and~$I$,
+the middles of the faces of the earth which
+are visible and invisible to the man in the moon,
+are directed away from the earth's centre. The
+edges of the earth's disk as seen from the moon
+are at $D$ and~$D$, and here the arrows point inwards
+to the earth's centre and are half as long
+as those at $V$ and~$I$. At intermediate points,
+they are intermediate both in size and direction.
+
+The only point in which the system considered
+differs from actuality is that the earth has
+been deprived of rotation. But this restriction
+may be removed, for, when the earth rotates
+once in $24$~hours, no difference is made in the
+forces which I have been trying to explain,
+\PageSep{104}
+although of course the force of gravity and the
+shape of the planet are affected by the rotation.
+This figure is called a diagram of tide-generating
+forces, because the tides of the ocean are due to
+the action of this system of forces.
+
+The explanation of tide-generating force is
+the very kernel of our subject, and, at the risk
+of being tedious, I shall look at it from a slightly
+different point of view. If every particle of the
+earth and of the ocean were acted on by equal
+and parallel forces, the whole system would
+move together and the ocean would not be displaced
+relatively to the earth; we should say
+that the ocean was at rest. If the forces were
+not quite equal and not quite parallel, there
+would be a slight residual effect tending to make
+the ocean move relatively to the solid earth. In
+other words, any defect from equality and parallelism
+in the forces would cause the ocean to
+move on the earth's surface.
+
+The forces which constitute the departure
+from equality and parallelism are called ``tide-generating
+forces,'' and it is this system which
+is indicated by the arrows in~\fig{22}. Tide-generating
+force is, in fact, that force which,
+superposed on the average force, makes the actual
+force. The average direction of the forces
+which act on the earth, as due to the moon's
+attraction, is along the line joining the earth's
+centre to the moon's centre, and its average
+\PageSep{105}
+intensity is equal to the force at the earth's
+centre.
+
+Now at~$V$ the actual force is straight towards~$M$,
+in the same direction as the average, but of
+greater intensity. Hence we find an arrow
+directed towards~$M$, the moon. At~$I$, the actual
+force is again in the same direction as, but of
+less intensity than, the average, and the arrow is
+directed away from~$M$, the moon. At~$D$, the
+actual force is almost exactly of the same intensity
+as the average, but it is not parallel thereto,
+and we must insert an inward force as shown by
+the arrow, so that when this is compounded with
+the average force we may get a total force in
+the right direction.
+
+Now let us consider how these forces tend to
+affect an ocean lying on the surface of the earth.
+The moon is directly over the head of an inhabitant
+of the earth, that is to say in his zenith,
+when he is at~$V$; she is right under his feet in
+the nadir when he is at~$I$; and she is in the
+observer's horizon, either rising or setting, when
+he is anywhere on the circle~$D$. When the
+inhabitant is at~$V$ or at~$I$ he finds that the tide-generating
+force is towards the zenith; when he
+is anywhere on the circle~$D$ he finds it towards
+the nadir. At other places he finds it directed
+towards or away from some point in the sky,
+except along two circles halfway between $V$ and~$D$,
+or between $I$ and~$D$, where the tide-generating
+\PageSep{106}
+force is level along the earth's surface, and may
+be called horizontal.
+
+A vertical force cannot make things move
+sideways, and so the sea will not be moved horizontally
+by it. The vertical part of the tide-generating
+force is not sufficiently great to
+overcome gravity, but will have the effect of
+making the water appear lighter or heavier. It
+will not, however, be effective in moving the
+water, since the water must remain in contact
+with the earth. We want, then, to omit the
+vertical part of the force and leave behind only
+the horizontal part, by which I mean a force
+which, to an observer on the earth's surface, is
+not directed either upwards or downwards, but
+along the level to any point of the compass.
+
+If there be a force acting at any point of the
+earth's surface, and directed upwards or downwards
+away from or towards some point in the
+sky other than the zenith, it may be decomposed
+into two forces, one vertically upwards or downwards,
+and another along the horizontal surface.
+Now as concerns the making of the tides,
+no attention need be paid to that part which
+is directed straight up or down, and the only
+important part is that along the surface,---the
+horizontal portion.
+
+Taking then the diagram of tide-generating
+forces in~\fig{22}, and obliterating the upward
+and downward portions of the force, we are left
+%% Plate 1
+\Figure[0.9]{23}{Horizontal Tide-Generating Force}{jpg}
+%% Facing page
+%[Blank Page]
+\PageSep{107}
+\index{Horizontal tide-generating force}%
+with a system of forces which may be represented
+by the arrows in the perspective picture of horizontal
+tide-generating force shown in~\fig{23}.
+
+If we imagine an observer to wander over the
+earth, $V$~is the place at which the moon is vertically
+over his head, and the circle~$D$, shown by
+the boundary of the shadow, passes through all
+the places at which the moon is in the horizon,
+just rising or setting. Then there is no horizontal
+force where the moon is over his head or under
+his feet, or where the moon is in his horizon
+either rising or setting, but everywhere else there
+is a force directed along the surface of the earth
+in the direction of the point at which the moon
+is straight overhead or underfoot.
+
+Now suppose $P$ to be the north pole of the
+earth, and that the circle $A_{1}$,~$A_{2}$, $A_{3}$, $A_{4}$,~$A_{5}$ is a
+parallel of latitude---say the latitude of London.
+Then if we watch our observer from external
+space, he first puts in an appearance on the picture
+at~$A_{1}$, and is gradually carried along to~$A_{2}$
+by the earth's rotation, and so onwards. Just before
+he comes to~$A_{2}$, the moon is due south of him,
+and the tide-generating force is also south, but
+not very large. It then increases, so that nearly
+three hours later, when he has arrived at~$A_{3}$, it
+is considerably greater. It then wanes, and
+when he is at~$A_{4}$ the moon is setting and the
+force is nil. After the moon has set, the force
+is directed towards the moon's antipodes, and it
+\PageSep{108}
+is greatest about three hours after moonset, and
+vanishes when the moon, still being invisible, is
+on the meridian.
+
+It must be obvious from this discussion that
+the lunar horizontal tide-generating force will
+differ, both as to direction and magnitude, according
+to the position of the observer on the
+earth and of the moon in the heavens, and that
+it can only be adequately stated by means of
+mathematical formulæ. I shall in the following
+chapter consider the general nature of the
+changes which the forces undergo at any point
+\index{Forces!tide-generating|)}%
+on the earth's surface.
+
+But before passing on to that matter it should
+be remarked that if the earth and sun had been
+the only pair of bodies in existence the whole of
+the argument would have applied equally well.
+Hence it follows that there is also a solar tide-generating
+force, which in actuality coëxists
+with the lunar force. I shall hereafter show
+how the relative importance of these two influences
+is to be determined.
+
+\begin{Authorities}
+Any mathematical work on the theory of the tides; for example,
+Thomson and Tait's \Title{Natural Philosophy}, Lamb's \Title{Hydrodynamics},
+Bassett's \Title{Hydrodynamics}, article \Title{Tides}, ``Encycl.\ Britan.,''
+Laplace's \Title{Mécanique Céleste},~\&c.
+\end{Authorities}
+\PageSep{109}
+
+
+\Chapter{VI}
+{Deflection of the Vertical}
+
+\First{The} intensity of tide-generating force is to be
+\index{Deflection of the vertical|(}%
+estimated by comparison with some standard, and
+it is natural to take as that standard the force of
+gravity at the earth's surface. Gravity acts in a
+vertical direction, whilst that portion of the tidal
+force which is actually efficient in disturbing the
+ocean is horizontal. Now the comparison between
+a small horizontal force and gravity is
+easily effected by means of a pendulum. For if
+the horizontal force acts on a suspended weight,
+the pendulum so formed will be deflected from
+the vertical, and the amount of deflection will
+measure the force in comparison with gravity.
+A sufficiently sensitive spirit level would similarly
+show the effect of a horizontal force by the
+displacement of the bubble. When dealing with
+tidal forces the displacements of either the pendulum
+\index{Forces!numerical estimate|(}%
+\index{Forces!deflection of vertical by|(}%
+or the level must be exceedingly minute,
+but, if measurable, they will show themselves as
+a change in the apparent direction of gravity.
+Accordingly a disturbance of this kind is often
+described as a deflection of the vertical.
+
+The maximum horizontal force due to the
+\PageSep{110}
+moon may be shown by a calculation, which involves
+the mass and distance of the moon, to
+have an intensity of $\frac{1}{11,660,000}$ of gravity.\footnote
+ {It does not occur to me that there is any very elementary
+ method of computing the maximum horizontal tidal force, but it
+ is easy to calculate the vertical force at the points $V$~or~$I$ in~\fig{22}.
+
+ {\Loosen The moon weighs $\frac{1}{80}$~of the earth, and has a radius $\frac{1}{4}$~as large.
+ Hence lunar gravity on the moon's surface is~$\frac{1}{80} × 4^{2}$, or $\frac{1}{5}$~of
+ terrestrial gravity at the earth's surface. The earth's radius is
+ $4,000$~miles and the moon's distance from the earth's centre
+ $240,000$~miles. Hence her distance from the nearer side of the
+ earth is $236,000$~miles. Therefore lunar gravity at the earth's
+ centre is $\frac{1}{5} × \frac{1}{240^{2}}$~of terrestrial gravity, and lunar gravity at the
+ point~$V$ is $\frac{1}{5} × \frac{1}{236^{2}}$~of the same. Therefore the tidal force at~$V$
+ is $\frac{1}{5} × \frac{1}{236^{2}} - \frac{1}{5} × \frac{1}{240^{2}}$~of terrestrial gravity. On multiplying the
+ squares of~$236$ and of~$240$ by~$5$, we find that this difference is
+ $\frac{1}{278,480} - \frac{1}{288,000}$. If these fractions are reduced to decimals
+ and the subtraction is performed, we find that the force at~$V$
+ is $.000,000,118,44$~of terrestrial gravity. When this decimal is
+ written as a fraction, we find the result to be $\frac{1}{8,450,000}$~of
+ gravity.}
+
+ Now it is the fact, although I do not see how to prove it in an
+ equally elementary manner, that the maximum horizontal tide-generating
+ force has an intensity equal to $\frac{3}{4}$~of the vertical force
+ at $V$~or~$I$. To find $\frac{3}{4}$~of the above fraction we must augment the
+ denominator by one third part. Hence the maximum horizontal
+ force is $\frac{1}{11,260,000}$~of gravity. This number does not agree exactly
+ with that given in the text; the discrepancy is due to the
+ fact that round numbers have been used to express the sizes and
+ distance apart of the earth and the moon, and their relative
+ masses.}
+Such a
+force must deflect the bob of a pendulum by the
+same fraction of the length of the cord by which
+it is suspended. If therefore the string were $10$~metres
+or $33$~feet in length, the maximum deflection
+of the weight would be $\frac{1}{11,660,000}$~of $10$~metres,
+\PageSep{111}
+or $\frac{1}{1,166}$~of a millimetre. In English measure this
+is $\frac{1}{29,000}$~of an inch. But the tidal force is reversed
+in direction about every six hours, so that the
+pendulum will depart from its mean direction by
+\index{Pendulum!curves traced by, under tidal force}%
+as much in the opposite direction. Hence the
+\Figure[0.8]{24}{Deflection of a Pendulum; the Moon and
+Observer on the Equator}{png}
+excursion to and fro of the pendulum under the
+lunar influence will be $\frac{1}{14,500}$~of an inch. With a
+pendulum one metre, or $3$~ft.\ $3$~in.\ in length,
+the range of motion of the pendulum bob is
+$\frac{1}{145,000}$~of an inch. For any pendulum of manageable
+length this displacement is so small, that it
+seems hopeless to attempt to measure it by direct
+observation. Nevertheless the mass and distance
+of the moon and the intensity of gravity being
+known with a considerable degree of accuracy, it
+is easy to calculate the deflection of the vertical
+at any time.
+\index{Forces!numerical estimate|)}%
+
+The curves which are traced out by a pendulum
+present an infinite variety of forms, corresponding
+\PageSep{112}
+to various positions of the observer on
+the earth and of the moon in the heavens. Two
+illustrations of these curves must suffice. \Fig{24}
+shows the case when the moon is on the celestial
+equator and the observer on the terrestrial
+equator. The path is here a simple ellipse,
+which is traversed twice over in the lunar day by
+the pendulum. The hours of the lunar day at
+\index{Pendulum!curves traced by, under tidal force}%
+which the bob occupies successive positions are
+marked on the curve.
+
+If the larger ellipse be taken to show the displacement
+of a pendulum when the sun and
+\Figure[0.8]{25}{Deflection of a Pendulum; the Moon in N.
+Declination~$15°$, the Observer in N. Latitude~$30°$}{png}
+moon coöperate at spring tide, the smaller one
+will show its path at the time of neap tide.
+
+In \fig{25} the observer is supposed to be in
+latitude~$30°$, whilst the moon stands $15°$~N. of
+the equator; in this figure no account is taken
+\PageSep{113}
+of the sun's force. Here also the hours are
+marked at the successive positions of the pendulum,
+which traverses this more complex curve
+only once in the lunar day. These curves are
+somewhat idealized, for they are drawn on the
+hypothesis that the moon does not shift her
+position in the heavens. If this fact were taken
+into account, we should find that the curve
+would not end exactly where it began, and that
+the character of the curve would change slowly
+from day to day.
+
+But even after the application of a correction
+for the gradual shift of the moon in the heavens,
+the curves would still be far simpler than in actuality,
+because the sun's influence has been left
+out of account. It has been remarked in the
+last chapter that the sun produces a tide-generating
+force, and it must therefore produce a
+deflection of the vertical. Although the solar
+deflection is considerably less than the lunar, yet
+it would serve to complicate the curve to a great
+degree, and it must be obvious then that when
+the full conditions of actuality are introduced
+the path of the pendulum will be so complicated,
+that mathematical formulæ are necessary for
+complete representation.
+
+Although the direct observation of the tidal
+deflection of the vertical would be impossible
+even by aid of a powerful microscope, yet several
+attempts have been made by more or less
+\PageSep{114}
+indirect methods. I have just pointed out that
+the path of a pendulum, although drawn on an
+ultra-microscopic scale, can be computed with a
+high degree of accuracy. It may then occur to
+the reader that it is foolish to take a great deal
+of trouble to measure a displacement which is
+scarcely measurable, and which is already known
+with fair accuracy. To this it might be answered
+that it would be interesting to watch the direct
+gravitational effects of the moon on the earth's
+surface. But such an interest does not afford
+the principal grounds for thinking that this
+attempted measurement is worth making. If the
+solid earth were to yield to the lunar attraction
+with the freedom of a perfect fluid, its surface
+would always be perpendicular to the direction
+of gravity at each instant of time. Accordingly
+a pendulum would then always hang perpendicularly
+to the average surface of the earth, and so
+there would be no displacement of the pendulum
+with reference to the earth's surface. If, then,
+the solid earth yields partially to the lunar attraction,
+the displacements of a pendulum must be
+of smaller extent relatively to the earth than if
+the solid earth were absolutely rigid. I must
+therefore correct my statement as to our knowledge
+of the path pursued by a pendulum, and
+say that it is known if the earth is perfectly
+unyielding. The accurate observation of the
+movement of a pendulum under the influence of
+\PageSep{115}
+the moon, and the comparison of the observed
+oscillation, with that computed on the supposition
+that the earth is perfectly stiff, would afford
+the means of determining to
+what extent the solid earth is
+yielding to tidal forces. Such
+a result would be very interesting
+as giving a measure of the
+stiffness of the earth as a whole.
+
+I must pass over the various
+\index{Cambridge, experiments with bifilar pendulum at|(}%
+\index{Darwin, G. H.!bifilar pendulum|(}%
+\index{Darwin, Horace, bifilar pendulum|(}%
+\index{Deflection of the vertical!experiments to measure|(}%
+\index{Pendulum!bifilar|(}%
+earlier attempts to measure the
+lunar attraction, and will only
+explain the plan, although it
+was abortive, used in~1879 by
+my brother Horace and myself.
+
+Our object was to measure
+the ultra-microscopic displacements
+of a pendulum with reference
+to the ground on which it
+stood. The principle of the apparatus
+used for this purpose is
+due to Lord Kelvin; it is very
+simple, although the practical
+application of it was not easy.
+
+%[** TN: Figure wrapped in the original]
+\Figure[0.2]{26}{Bifilar Pendulum}{png}
+
+\Fig{26} shows diagrammatically, and not drawn
+to scale, a pendulum~$\Seg{A}{B}$ hanging by two wires.
+At the foot of the pendulum there is a support~$C$
+attached to the stand of the pendulum; $D$~is a
+small mirror suspended by two silk fibres, one
+being attached to the bottom of the pendulum~$B$
+\PageSep{116}
+and the other to the support~$C$. When the
+two fibres are brought very close together, any
+movement of the pendulum perpendicular to the
+plane of the mirror causes the mirror to turn
+through a considerable angle. The two silk
+fibres diverge from one another, but if two vertical
+lines passing through the two points of suspension
+are $\frac{1}{1,000}$~of an inch apart, then when the
+pendulum moves one of these points through a
+millionth of an inch, whilst the other attached to~$C$
+remains at rest, the mirror will turn through
+an angle of more than three minutes of arc.
+A lamp is placed opposite to the mirror, and
+the image of the lamp formed by reflection in
+the mirror is observed. A slight rotation of the
+mirror corresponds to an almost infinitesimal
+motion of the pendulum, and even excessively
+small movements of the mirror are easily detected
+by means of the reflected image of the light.
+
+In our earlier experiments the pendulum was
+hung on a solid stone gallows; and yet, when
+the apparatus was made fairly sensitive, the image
+of the light danced and wandered incessantly.
+Indeed, the instability was so great that
+the reflected image wandered all across the room.
+We found subsequently that this instability was
+due both to changes of temperature in the stone
+gallows, and to currents in the air surrounding
+the pendulum.
+
+To tell of all the difficulties encountered
+\PageSep{117}
+might be as tedious as the difficulties themselves,
+so I shall merely describe the apparatus in its
+ultimate form. The pendulum was suspended,
+as shown in~\fig{26}, by two wires; the two wires
+being in an east and west plane, the pendulum
+could only swing north and south. It was hung
+inside a copper tube, just so wide that the solid
+copper cylinder, forming the pendulum bob, did
+not touch the sides of the tube. A spike projected
+from the base of the pendulum bob
+through a hole in the bottom of the tube. The
+mirror was hung in a little box, with a plate-glass
+front, which was fastened to the bottom of the
+copper tube. The only communication between
+the tube and the mirror-box was by the hole
+through which the spike of the pendulum projected,
+but the tube and mirror-box together
+formed a water-tight vessel, which was filled with
+a mixture of spirits of wine and boiled water.
+The object of the fluid was to steady the
+mirror and the pendulum, while allowing its
+slower movements to take place. The water was
+boiled to get rid of air in it, and the spirits of
+wine was added to increase the resistance of the
+fluid, for it is a remarkable fact that a mixture
+of spirits and water has considerably more viscosity
+or stickiness than either pure spirits or
+pure water.
+
+The copper tube, with the pendulum and mirror-box,
+was supported on three legs resting on
+\PageSep{118}
+a block of stone weighing a ton, and this stood
+on the native gravel in a north room in the laboratory
+at Cambridge. The whole instrument
+was immersed in a water-jacket, which was furnished
+with a window near the bottom, so that
+the little mirror could be seen from outside. A
+water ditch also surrounded the stone pedestal,
+and the water jacketing of the whole instrument
+made the changes of temperature very slow.
+
+A gas jet, only turned up at the moment of
+observation, furnished the light to be observed
+by reflection in the little mirror. The gas
+burner could be made to travel to and fro along
+a scale in front of the instrument. In the preliminary
+description I have spoken of the motion
+of the image of a fixed light, but it clearly
+amounts to the same thing if we measure the
+motion of the light, keeping the point of observation
+fixed. In our instrument the image of
+the movable gas jet was observed by a fixed telescope
+placed outside of the room. A bright
+light was unfortunately necessary, because there
+was a very great loss of light in the passages to
+and fro through two pieces of plate glass and a
+considerable thickness of water.
+
+Arrangements were made by which, without
+entering the room, the gas jet could be turned
+up and down, and could be made to move to and
+fro in the room in an east and west direction,
+until its image was observed in the telescope.
+\PageSep{119}
+There were also adjustments by which the two
+silk fibres from which the mirror hung could be
+brought closer together or further apart, thus
+making the instrument more or less sensitive.
+There was also an arrangement by which the image
+of the light could be brought into the field
+of view, when it had wandered away beyond the
+limits allowed for by the traverse of the gas jet.
+
+When the instrument was in adjustment, an
+observation consisted of moving the gas jet until
+its image was in the centre of the field of
+view of the telescope; a reading of the scale, by
+another telescope, determined the position of the
+gas jet to within about a twentieth of an inch.
+
+The whole of these arrangements were arrived
+at only after laborious trials, but all the precautions
+were shown by experience to be necessary,
+and were possibly even insufficient to guard the
+instrument from the effects of changes of temperature.
+I shall not explain the manner in which
+we were able to translate the displacements of
+the gas jet into displacements of the pendulum.
+It was not very satisfactory, and only gave approximate
+results. A subsequent form of an
+instrument of this kind, designed by my brother,
+has been much improved in this respect. It was
+he also who designed all the mechanical appliances
+in the experiment of which I am speaking.
+
+It may be well to reiterate that the pendulum
+was only free to move north and south, and that
+\PageSep{120}
+our object was to find how much it swung. The
+east and west motion of a pendulum is equally
+interesting, but as we could not observe both
+displacements at the same time, we confined our
+attention in the first instance to the northerly
+and southerly movements.
+
+When properly adjusted the apparatus was so
+sensitive that, if the bob of the pendulum moved
+through $\frac{1}{40,000}$~of a millimetre, that is, a millionth
+part of an inch, we could certainly detect the
+movement, for it corresponded to a twentieth
+of an inch in our scale of position of the gas
+jet. When the pendulum bob moved through
+this amount, the wires of the pendulum turned
+through one two-hundredth of a second of arc;
+this is the angle subtended by one inch at $770$~miles
+distance. I do not say that we could actually
+measure with this degree of refinement, but
+we could detect a change of that amount. In
+view of the instability of the pendulum, which
+still continued to some extent, it may be hard to
+gain credence for the statement that such a small
+deflection was a reality, so I will explain how we
+were sure of our correctness.
+
+In setting up the apparatus, work had to be
+conducted inside the room, and some preliminary
+observations of the reflected image of a stationary
+gas jet were made without the use of the telescope.
+The scale on which the reflected spot
+of light fell was laid on the ground at about
+\PageSep{121}
+seven feet from the instrument; in order to
+watch it I knelt on the pavement behind the
+scale, and leant over it. I was one day watching
+on the scale the spot of light which revealed
+the motion of the pendulum, and, being tired
+with kneeling, supported part of my weight on
+my hands a few inches in front of the scale.
+The place where my hands rested was on the
+bare earth, from which a paving stone had been
+removed. I was surprised to find quite a large
+change in the reading. It seemed at first incredible
+that my change of position was the cause,
+but after several trials I found that light pressure
+with one hand was quite sufficient to produce
+an effect. It must be remembered that this was
+not simply a small pressure delivered on the bare
+earth at, say, seven feet distance, but it was the
+difference of effect produced by the same pressure
+at seven feet and six feet; for, of course,
+the change only consisted in the distribution of
+the weight of a small portion of my body.
+
+It is not very easy to catch the telescopic image
+of a spot of light reflected from a mirror of
+the size of a shilling. Accordingly, in setting
+up our apparatus, we availed ourselves of this result,
+for we found that the readiest way of bringing
+the reflected image into the telescopic field
+of view was for one of us to move slowly about
+the room, until the image of the light was
+brought, by the warping of the soil due to his
+\PageSep{122}
+weight, into the field of view of the telescope.
+He then placed a heavy weight on the floor
+where he had been standing; this of course
+drove the image out of the field of view, but
+after he had left the room the image of the flame
+was found to be in the field.
+
+We ultimately found, even when no special
+pains had been taken to render the instrument
+sensitive, that if one of us was in the room, and
+stood at about sixteen feet south of the instrument
+with his feet about a foot apart, and slowly
+shifted his weight from one foot to the other, a
+distinct change was produced in the image of the
+gas flame, and of course in the position of the
+little mirror, from which the image was derived
+by reflection. It may be well to consider for
+a moment the meaning of this result. If one
+presses with a finger on a flat slab of jelly, a sort
+of dimple is produced, and if a pin were sticking
+upright in the jelly near the dimple, it would tilt
+slightly towards the finger. Now this is like
+what we were observing, for the jelly represents
+the soil, and the tilt of the pin corresponds to
+that of the pendulum. But the scale of the displacement
+is very different, for our pendulum
+stood on a block of stone weighing nearly a ton,
+which rested on the native gravel at two feet below
+the level of the floor, and the slabs of the
+floor were removed from all round the pendulum.
+The dimple produced by a weight of $140$~lbs.\ on
+\PageSep{123}
+\index{Distortion of soil!by weight}%
+\index{Elastic distortion!of soil by weight}%
+the stone paved floor must have been pretty
+small, and the slope of the sides of that dimple
+\index{Dimple!in soil, due to weight}%
+at sixteen feet must have been excessively slight;
+but we were here virtually observing the change
+of slope at the instrument, when the centre of
+the dimple was moved from a distance of fifteen
+feet to sixteen feet.
+
+It might perhaps be thought that all observation
+would be rendered impossible by the street
+traffic and by the ordinary work of the laboratory.
+But such disturbances only make tremors
+of very short period, and the spirits and water
+damped out quick oscillations so thoroughly, that
+no difference could be detected in the behavior
+of the pendulum during the day and during the
+night. Indeed, we found that a man could stand
+close to the instrument and hit the tub and pedestal
+smart blows with a stick, without producing
+any sensible effect. But it was not quite easy to
+try this experiment, because there was a considerable
+disturbance on our first entering the room;
+and when this had subsided small movements of
+the body produced a sensible deflection, by slight
+changes in the distribution of the experimenter's
+weight.
+
+It is clear that we had here an instrument of
+amply sufficient delicacy to observe the lunar
+tide-generating force, and yet we completely
+failed to do so. The pendulum was, in fact,
+always vacillating and changing its position by
+\PageSep{124}
+many times the amount of the lunar effect which
+we sought to measure.
+
+An example will explain how this was: A series
+of frequent readings were taken from July
+21st to~25th, 1881, with the pendulum arranged
+to swing north and south. We found that there
+was a distinct diurnal period, with a maximum at
+noon, when the pendulum bob stood furthest
+northward. The path of the pendulum was interrupted
+by many minor zigzags, and it would
+sometimes reverse its motion for an hour together.
+But the diurnal oscillation was superposed on a
+gradual drift of the pendulum, for the mean
+diurnal position traveled slowly southward. Indeed,
+in these four days the image disappeared
+from the scale three times over, and was brought
+back into the field of view three times by the
+appliance for that purpose. On the night between
+the 24th and~25th the pendulum took an
+abrupt turn northward, and the scale reading
+was found, on the morning of the~25th, nearly
+at the opposite end of the scale from that towards
+which it had been creeping for four days
+previously.
+
+Notwithstanding all our precautions the pendulum
+was never at rest, and the image of the
+flame was always trembling and dancing, or waving
+slowly to and fro. In fact, every reading of
+our scale had to be taken as the mean of the
+excursions to right and left. Sometimes for two
+\PageSep{125}
+or three days together the dance of the image
+would be very pronounced, and during other
+days it would be remarkably quiescent.
+
+The origin of these tremors and slower movements
+\index{Earthquakes!microsisms and earth tremors|(}%
+\index{Italian investigations in seismology|(}%
+\index{Tremors, earth}%
+is still to some extent uncertain. Quite
+recent investigations by Professor Milne seem to
+\index{Milne on seismology}%
+show that part of them are produced by currents
+in the fluid surrounding the pendulum, that
+others are due to changes in the soil of a very
+local character, and others again to changes
+affecting a considerable tract of soil. But when
+all possible allowance is made for these perturbations,
+it remains certain that a large proportion
+of these mysterious movements are due to minute
+earthquakes.
+
+Some part of the displacements of our pendulum
+was undoubtedly due to the action of the
+moon, but it was so small a fraction of the whole,
+that we were completely foiled in our endeavor
+to measure it.\footnote
+ {Since the date of our experiment the bifilar pendulum has
+ been perfected by my brother, and it is now giving continuous
+ photographic records at several observatories. It is now made
+ to be far less sensitive than in our original experiment, and no
+ attempt is made to detect the direct effect of the moon.}
+\index{Cambridge, experiments with bifilar pendulum at|)}%
+\index{Darwin, G. H.!bifilar pendulum|)}%
+\index{Darwin, Horace, bifilar pendulum|)}%
+\index{Deflection of the vertical!experiments to measure|)}%
+\index{Pendulum!bifilar|)}%
+
+The minute earthquakes of which I have
+\index{Microsisms, minute earthquakes|(}%
+spoken are called by Italian observers microsisms,
+and this name has been very generally
+adopted. The literature on the subject of seismology
+is now very extensive, and it would be
+out of place to attempt to summarize here the
+\PageSep{126}
+conclusions which have been drawn from observation.
+I may, however, permit myself to add a
+few words to indicate the general lines of the research,
+which is being carried on in many parts
+of the world.
+
+Italy is a volcanic country, and the Italians
+have been the pioneers in seismology. Their
+observations have been made by means of pendulums
+of various lengths, and with instruments
+of other forms, adapted for detecting vertical
+movements of the soil. The conclusions at
+which Father Bertelli arrived twenty years ago
+\index{Bertelli on Italian seismology}%
+may be summarized as follows:---
+
+The oscillation of the pendulum is generally
+\index{Pendulum!as seismological instrument}%
+parallel to valleys or chains of mountains in the
+neighborhood. The oscillations are independent
+of local tremors, velocity and direction of wind,
+rain, change of temperature, and atmospheric
+electricity.
+
+Pendulums of different lengths betray the
+movements of the soil in different manners, according
+to the agreement or disagreement of
+their natural periods of oscillation with the period
+of the terrestrial vibrations.
+
+The disturbances are not strictly simultaneous
+in the different towns of Italy, but succeed one
+another at short intervals.
+
+After earthquakes the ``tromometric'' or microseismic\DPnote{** [sic]}
+\index{Tromometer, a seismological instrument}%
+movements are especially apt to be in
+a vertical direction. They are always so when
+\PageSep{127}
+\index{Bertelli on Italian seismology}%
+the earthquake is local, but the vertical movements
+are sometimes absent when the shock
+occurs elsewhere. Sometimes there is no movement
+at all, even when the shock occurs quite
+close at hand.
+
+The positions of the sun and moon appear to
+have some influence on the movements of the
+pendulum, but the disturbances are especially
+\index{Pendulum!as seismological instrument}%
+frequent when the barometer is low.
+
+The curves of ``the monthly means of the
+tromometric movement'' exhibit the same forms
+in the various towns of Italy, even those which
+are distant from one another.
+
+The maximum of disturbance occurs near the
+winter solstice and the minimum near the summer
+solstice.
+
+At Florence a period of earthquakes is presaged
+\index{Earthquakes!microsisms and earth tremors|)}%
+\index{Tromometer, a seismological instrument}%
+by the magnitude and frequency of oscillatory
+movements in a vertical direction. These
+movements are observable at intervals and during
+several hours after each shock.
+
+Some very curious observations on microsisms
+\index{Microsisms, minute earthquakes|)}%
+have also been made in Italy with the microphone,
+by which very slight movements of the
+soil are rendered audible.
+
+Cavaliere de Rossi, of Rome, has established a
+``geodynamic'' observatory in a cave $700$~metres
+above the sea at Rocca di Papa, on the external
+slope of an extinct volcano.
+
+At this place, remote from all carriages and
+\PageSep{128}
+\index{Rossi on Italian seismology|(}%
+roads, he placed his microphone at a depth of $20$~metres
+\index{Microphone as a seismological instrument|(}%
+below the ground. It was protected
+against insects by woolen wrappings. Carpet
+was spread on the floor of the cave to deaden
+the noise from particles of stone which might
+possibly fall. Having established his microphone,
+he waited till night, and then heard noises which
+he says revealed ``natural telluric phenomena.''
+The sounds which he heard he describes as
+``roarings, explosions occurring isolated or in
+volleys, and metallic or bell-like sounds'' (\textit{fremiti,
+scopii isolati o di moschetteria, e suoni-metallici
+o di campana}). They all occurred
+mixed indiscriminately, and rose to maxima at
+irregular intervals. By artificial means he was
+able to cause noises which he calls ``rumbling (?)
+or crackling'' (\textit{rullo o crepito}). The roaring
+(\textit{fremito}) was the only noise which he could reproduce
+artificially, and then only for a moment.
+It was done by rubbing together the conducting
+wires, ``in the same manner as the rocks must
+rub against one another when there is an earthquake.''
+
+A mine having been exploded in a quarry at
+some distance, the tremors in the earth were
+audible in the microphone for some seconds
+subsequently.
+
+There was some degree of coincidence between
+the agitation of the pendulum-seismograph and
+the noises heard with the microphone.
+\PageSep{129}
+
+At a time when Vesuvius became active,
+Rocca di Papa was agitated by microsisms, and
+the shocks were found to be accompanied by the
+very same microphonic noises as before. The
+noises sometimes became ``intolerably loud;''
+especially on one occasion in the middle of the
+night, half an hour before a sensible earthquake.
+The agitation of the microphone corresponded
+exactly with the activity of Vesuvius.
+
+Rossi then transported his microphone to
+Palmieri's Vesuvian observatory, and worked in
+conjunction with him. He there found that
+each class of shock had its corresponding noise.
+The sussultorial shocks, in which I conceive the
+movement of the ground is vertically up and
+down, gave the volleys of musketry (\textit{i~colpi di
+moschetteria}), and the undulatory shocks gave
+the roarings (\textit{i~fremiti}). The two classes of
+noises were sometimes mixed up together.
+
+Rossi makes the following remarks: ``On
+Vesuvius I was put in the way of discovering
+that the simple fall and rise in the ticking which
+occurs with the microphone [\textit{battito del orologio
+unito al microfono}] (a phenomenon observed
+by all, and remaining inexplicable to all) is a
+consequence of the vibration of the ground.''
+This passage alone might perhaps lead one to
+suppose that clockwork was included in the circuit;
+but that this was not the case, and that
+``ticking'' is merely a mode of representing a
+\PageSep{130}
+\index{Paschwitz, von Rebeur!on horizontal pendulum|(}%
+\index{Pendulum!horizontal|(}%
+natural noise is proved by the fact that he subsequently
+says that he considers the ticking to
+be ``a telluric phenomenon.''
+
+Rossi then took the microphone to the Solfatara
+\index{Rossi on Italian seismology|)}%
+of Pozzuoli, and here, although no sensible
+tremors were felt, the noises were so loud as
+to be heard simultaneously by all the people in
+the room. The ticking was quite masked by
+other natural noises. The noises at the Solfatara
+were imitated by placing the microphone
+on the lid of a vessel of boiling water. Other
+seismic noises were then imitated by placing the
+microphone on a marble slab, and scratching
+\index{Microphone as a seismological instrument|)}%
+and tapping the under surface of it.
+
+The observations on Vesuvius led him to the
+conclusion that the earthquake oscillations have
+sometimes fixed ``nodes,'' for there were places
+on the mountain where no effects were observed.
+There were also places where the movement was
+intensified, and hence it may be concluded that
+the centre of disturbance may sometimes be very
+distant, even when the observed agitation is
+considerable.
+
+At the present time perhaps the most distinguished
+investigator in seismology is Professor
+Milne, formerly of the Imperial College of Engineering
+\index{Milne on seismology}%
+at Tokyo. His residence in Japan gave
+\index{Japan, frequency of earthquakes}%
+him peculiar opportunities of studying earthquakes,
+for there is, in that country, at least one
+earthquake per diem of sufficient intensity to
+\index{Italian investigations in seismology|)}%
+\PageSep{131}
+\index{Japan, frequency of earthquakes}%
+affect a seismometer. The instrument of which
+he now makes most use is called a horizontal
+pendulum. The principle involved in it is old,
+but it was first rendered practicable by von
+Rebeur-Paschwitz, whose early death deprived
+the world of a skillful and enthusiastic investigator.
+
+The work of Paschwitz touches more closely
+on our present subject than that of Milne, because
+he made a gallant attempt to measure the
+moon's tide-generating force, and almost persuaded
+himself that he had done so.
+
+The horizontal pendulum is like a door in its
+mode of suspension. If a doorpost be absolutely
+vertical, the door will clearly rest in any
+position, but if the post be even infinitesimally
+tilted the door naturally rests in one definite
+position. A very small shift of the doorpost is
+betrayed by a considerable change in the position
+of the door. In the pendulum the door is
+replaced by a horizontal boom, and the hinges
+by steel points resting in agate cups, but the
+principle is the same.
+
+The movement of the boom is detected and
+registered photographically by the image of a
+light reflected from certain mirrors. Paschwitz
+made systematic observations with his pendulum
+at Wilhelmshaven, Potsdam, Strassburg, and
+Orotava. He almost convinced himself at one
+time that he could detect, amidst the wanderings
+\PageSep{132}
+of the curves of record, a periodicity corresponding
+to the direct effect of the moon's action.
+But a more searching analysis of his results left
+the matter in doubt. Since his death the observations
+at Strassburg have been continued by
+M.~Ehlert. His results show an excellent consistency
+\index{Ehlert, observation with horizontal pendulum}%
+with those of Paschwitz, and are therefore
+\index{Paschwitz, von Rebeur!on horizontal pendulum|)}%
+confirmatory of the earlier opinion of the
+latter. I am myself disposed to think that the
+detection of the lunar attraction is a reality, but
+the effect is so minute that it cannot yet be
+relied on to furnish a trustworthy measurement
+of the amount of the yielding of the solid earth
+to tidal forces.
+
+It might be supposed that doubt could hardly
+arise as to whether or not the direct effect of
+the moon's attraction had been detected. But
+I shall show in the next chapter that at many
+places the tidal forces must exercise in an indirect
+manner an effect on the motion of a pendulum
+\index{Pendulum!horizontal|)}%
+much greater than the direct effect.
+
+It was the consideration of this indirect effect,
+and of other concomitants, which led us to
+abandon our attempted measurement, and to
+conclude that all endeavors in that direction
+were doomed to remain for ever fruitless. I can
+but hope that a falsification of our forecast by
+M.~Ehlert and by others may be confirmed.
+\PageSep{133}
+
+\begin{Authorities}
+G.~H. Darwin and Horace Darwin, ``Reports to the British
+Association for the Advancement of Science:''---
+
+\Title{Measurement of the Lunar Disturbance of Gravity.} York
+meeting, 1881, pp.~93--126.
+
+\Title{Second Report on the same}, with appendix. Southampton
+meeting, 1882, pp.~95--119.
+
+E.~von Rebeur-Paschwitz, \Title{Das Horizontalpendel}.
+
+``Nova Acta Leop.\ Carol.\ Akad.,'' 1892, vol.~lx.\ no.~1, p.~213;
+also ``Brit.\ Assoc.\ Reports,''~1893.
+
+E.~von Rebeur-Paschwitz, \Title{Ueber Horizontalpendel-Beobach\-tungen
+in Wilhelmshaven, Potsdam und Puerto Orotava auf Tenerifa}.
+
+``Astron.\ Nachrichten,'' vol.~cxxx.\ pp.~194--215.
+
+R.~Ehlert, \Title{Horizontalpendel-Beobachtungen}.
+
+``Beiträge zur Geophysik,'' vol.~iii.\ Part~I., 1896.
+
+C.~Davison, \Title{History of the Horizontal and Bifilar Pendulums}.
+\index{Davison, history of bifilar and horizontal pendulums}%
+
+``Appendix to Brit.\ Assoc.\ Report on Earth Tremors.'' Ipswich
+meeting, 1895, pp.~184--192.
+
+``British Association Reports of Committees.''
+
+\Title{On Earth Tremors}, 1891--95 (the first being purely formal).
+
+\Title{On Seismological Investigation}, 1896.
+
+The literature on Seismology is very extensive, and would
+\index{Seismology}%
+need a considerable index; the reader may refer to \Title{Earthquakes}
+and to \Title{Seismology} by John Milne. Both works form volumes in
+the International Scientific Series, published by Kegan Paul,
+Trench, Trübner \&~Co.
+\end{Authorities}
+\index{Deflection of the vertical|)}%
+\index{Forces!deflection of vertical by|)}%
+\PageSep{134}
+
+
+\Chapter[Distortion of the Earth's Surface]{VII}
+{The Elastic Distortion of the Earth's Surface
+by Varying Loads}
+
+\First{When} the tide rises and falls on the seacoast,
+\index{Darwin, G. H.!distortion of earth's surface by varying loads|(}%
+\index{Deflection of the vertical!due to tide|(}%
+\index{Distortion of soil!by varying loads|(}%
+\index{Elastic distortion!of earth by varying loads|(}%
+many millions of tons of water are brought alternately
+nearer and further from the land. Accordingly
+a pendulum suspended within a hundred
+miles or so of a seacoast should respond to the
+attraction of the sea water, swinging towards the
+sea at high water, and away from it at low water.
+Since the rise and fall has a lunar periodicity the
+pendulum should swing in the same period, even
+if the direct attraction of the moon did not affect
+it. But, as I shall now show, the problem is
+further confused by another effect of the varying
+tidal load.
+
+We saw in \Ref{Chapter}{VI}.\ how a weight resting
+on the floor in the neighborhood of our pendulum
+produced a dimple by which the massive
+stone pedestal of our instrument was tilted over.
+Now as low tide changes to high tide the position
+of an enormous mass of water is varied with
+respect to the land. Accordingly the whole
+coast line must rock to and fro with the varying
+tide. We must now consider the nature of the
+\PageSep{135}
+distortion of the soil produced in this way. The
+mathematical investigation of the form of the
+dimple in a horizontal slab of jelly or other elastic
+\index{Dimple!form of, in elastic slab}%
+material, due to pressure at a single point,
+shows that the slope at any place varies inversely
+as the square of the distance from the centre.
+That is to say, if starting from any point we
+proceed to half our original distance, we shall
+find four times as great a slope, and at one third
+\Figure[0.7]{27}{Form of Dimple in an Elastic Surface}{png}
+of the original distance the slope will be augmented
+ninefold.
+
+The theoretical form of dimple produced by
+pressure at a single mathematical point is shown
+in~\fig{27}. The slope is exaggerated so as to
+render it visible, and since the figure is drawn on
+the supposition that the pressure is delivered at
+a mathematical point, the centre of the dimple
+is infinitely deep. If the pressure be delivered
+by a blunt point, the slope at a little distance
+\PageSep{136}
+will be as shown, but the centre will not be infinitely
+deep. If therefore we pay no attention to
+the very centre, this figure serves to illustrate
+the state of the case. When the dimple is produced
+by the pressure of a weight, that weight,
+being endowed with gravitation, attracts any
+other body with a force varying inversely as the
+square of the distance. It follows, therefore,
+that the slope of the dimple is everywhere exactly
+\index{Slope of soil!due to elastic distortion}%
+proportional to the gravitational attraction
+\index{Attraction!of weight resting on elastic slab proportional to slope}%
+of the weight. Since this is true of a single
+weight, it is true of a group of weights, each
+producing its own dimple by pressure and its own
+attraction, strictly proportional to one another.
+Thus the whole surface is deformed by the superposition
+of dimples, and the total attraction is
+the sum of all the partial attractions.
+
+Let us then imagine a very thick horizontal
+slab of glass supporting any weights at any parts
+of its surface. The originally flat surface of the
+slab will be distorted into shallow valleys and
+low hills, and it is clear that the direct attraction
+of the weights will everywhere be exactly proportional
+to the slopes of the hillsides; also the
+direction of the greatest slope at each place must
+agree with the direction of the attraction. The
+direct attraction of the weights will deflect a
+pendulum from the vertical, and the deflection
+must be exactly proportional to the slope produced
+by the pressure of the weights. It may
+\PageSep{137}
+be proved that if the slab is made of a very stiff
+glass the angular deflection of the pendulum
+under the influence of attraction will be one fifth
+\index{Attraction!of weight resting on elastic slab proportional to slope}%
+of the slope of the hillside; if the glass were
+of the most yielding kind, the fraction would be
+one eighth. The fraction depends on the degree
+of elasticity of the material, and the stiffer it is
+the larger the fraction.
+
+The observation of a pendulum consists in
+noting its change of position with reference to
+the surface of the soil; hence the slope of the
+soil, and the direct attraction of the weight
+which causes that slope, will be absolutely fused
+together, and will be indistinguishable from one
+another.
+
+Now, this conclusion may be applied to the
+tidal load, and we learn that, if rocks are of the
+same degree of stiffness as glass of medium
+quality, the direct attraction of the tidal load
+produces one sixth of the apparent deflection of
+a pendulum produced by the tilting of the soil.
+
+If any one shall observe a pendulum, within
+say a hundred miles of the seacoast, and shall
+detect a lunar periodicity in its motion, he can
+only conclude that what he observes is partly
+due to the depression and tilting of the soil,
+partly to attraction of the sea water, and partly
+to the direct attraction of the moon. Calculation
+indicates that, with the known average elasticity
+of rock, the tilting of the soil is likely to
+\PageSep{138}
+\index{Elastic distortion!calculation and illustration|(}%
+\index{Slope of soil!calculation and illustration of|(}%
+be far greater than the other two effects combined.
+Hence, if the direct attraction of the
+moon is ever to be measured, it will first be
+necessary to estimate and to allow for other important
+oscillations with lunar periodicity. The
+difficulty thus introduced into this problem is so
+serious that it has not yet been successfully met.
+It may perhaps some day be possible to distinguish
+the direct effects of the moon's tidal attraction
+from the indirect effects, but I am not
+very hopeful of success in this respect. It was
+pointed out in \Ref{Chapter}{VI}.\ that there is some
+reason to think that a lunar periodicity in the
+swing of a pendulum has been already detected,
+and if this opinion is correct, the larger part of
+the deflection was probably due to these indirect
+effects.
+
+The calculation of the actual tilting of the
+coast line by the rising tide would be excessively
+complex even if accurate estimates were obtainable
+of the elasticity of the rock and of the tidal
+load. It is, however, possible to formulate a
+soluble problem of ideal simplicity, which will
+afford us some idea of the magnitude of the
+results occurring in nature.
+
+In the first place, we may safely suppose the
+earth to be flat, because the effect of the tidal
+load is quite superficial, and the curvature of the
+earth is not likely to make much difference in
+the result. In the second place, it greatly simplifies
+\PageSep{139}
+the calculation to suppose the ocean to
+consist of an indefinite number of broad canals,
+separated from one another by broad strips of
+land of equal breadth. Lastly, we shall suppose
+that each strip of sea rocks about its middle line,
+so that the water oscillates as in a seiche of the
+Lake of Geneva; thus, when it is high water
+on the right-hand coast of a strip of sea, it is
+low water on the left-hand coast, and vice versa.
+We have then to determine the change of shape
+of the ocean-bed and of the land, as the tide
+rises and falls. The problem as thus stated is
+\Figure{28}{Distortion of Land and Sea-bed by Tidal Load}{png}
+vastly simpler than in actuality, yet it will suffice
+to give interesting indications of what must
+occur in nature.
+
+The figure~\figref{28} shows the calculated result, the
+slopes being of course enormously exaggerated.
+The straight line represents the level surface of
+land and sea before the tidal oscillation begins,
+the shaded part being the land and the dotted
+part the sea. Then the curved line shows the
+form of the land and of the sea-bed, when it is
+low water at the right of the strip of land and high
+\PageSep{140}
+water at the left. The figure would be reversed
+when the high water interchanges position
+with the low water. Thus both land and sea
+rock about their middle lines, but the figure
+shows that the strip of land remains nearly flat
+although not horizontal, whilst the sea-bed becomes
+somewhat curved.
+
+It will be noticed that there is a sharp nick at
+the coast line. This arises from the fact that
+deep water was assumed to extend quite up to
+the shore line; if, however, the sea were given
+a shelving shore, as in nature, the sharp nick
+would disappear, although the form of the distorted
+rocks would remain practically unchanged
+elsewhere.
+
+Thus far the results have been of a general
+character, and we have made no assumptions as
+to the degree of stiffness of the rock, or as to
+the breadths of the oceans and continents. Let
+us make hypotheses which are more or less
+plausible. At many places on the seashore the
+tide ranges through twenty or thirty feet, but
+these great tides only represent the augmentation
+of the tide-wave as it runs into shallow
+water, and it would not be fair to suppose our
+tide to be nearly so great. In order to be moderate,
+I will suppose the tide to have a range of
+$160$~centimetres, or, in round numbers, about $5$~feet.
+Then, at the high-water side of the sea,
+the water is raised by $80$~centimetres, and at the
+\index{Elastic distortion!calculation and illustration|)}%
+\index{Slope of soil!calculation and illustration of|)}%
+\PageSep{141}
+low-water side it is depressed by the same
+amount. The breadth of the Atlantic is about
+$4,000$~or $5,000$~miles. I take then, the breadth of
+the oceans and of the continents as $3,900$~miles,
+or $6,280$~kilometres. Lastly, as rocks are usually
+stiffer than glass, I take the rock bed to
+be twice as stiff as the most yielding glass, and
+quarter as stiff again as the stiffest glass; this
+assumption as to the elasticity of rock makes the
+attraction at any place one quarter of the slope.
+For a medium glass we found the fraction to be
+about one sixth. These are all the data required
+for determining the slope.
+
+It is of course necessary to have a unit of
+measurement for the slope of the surface. Now
+a second of arc is the name for the angular
+magnitude of an inch seen at $3\frac{1}{4}$~miles, and accordingly
+a hundredth of a second of arc, usually
+written~$0''.01$, is the angular magnitude of
+an inch seen at $325$~miles; the angles will then
+be measured in hundredths of seconds.
+
+Before the tide rises, the land and sea-bed
+are supposed to be perfectly flat and horizontal.
+Then at high tides the slopes on the land are as
+follows:---
+\[
+\begin{array}{r@{\ }lc}
+\multicolumn{2}{c}{\ColHead{Distance from high-water mark}} &
+\multicolumn{1}{c}{\ColHead[2in]{Slope of the land measured in hundredths of seconds of arc}} \\
+ 10 &\text{metres} & 10 \\
+100 &\text{metres} & \Z8 \\
+ 1 &\text{kilometre} & \Z6 \\
+ 10 &\text{kilometres} & \Z4 \\
+ 20 &\text{kilometres} & \Z3\rlap{$\frac{1}{2}$} \\
+100 &\text{kilometres} & \Z2
+\end{array}
+\]
+\PageSep{142}
+The slope is here expressed in hundredths of a
+second of arc, so that at $100$~kilometres from the
+coast, where the slope is~$2$, the change of plane
+amounts to the angle subtended by one inch at
+$162$~miles.
+
+When high water changes to low water, the
+slopes are just reversed, hence the range of
+change of slope is represented by the doubles of
+these angles. If the change of slope is observed
+by some form of pendulum, allowance must be
+made for the direct attraction of the sea, and it
+appears that with the supposed degree of stiffness
+of rock these angles of slope must be augmented
+in the proportion of $5$~to~$4$. Thus, we
+double the angles to allow of change from high
+to low water, and augment the numbers as $5$ is
+to~$4$, to allow for the direct attraction of the sea.
+Finally we find results which may be arranged
+in the following tabular form:---
+\[
+\begin{array}{r@{\ }lc}
+\multicolumn{2}{c}{\ColHead{Distance from high-water mark}} &
+\multicolumn{1}{c}{\ColHead[1.75in]{Apparent range of deflection of the vertical}} \\
+ 10 &\text{metres} & 0''.25\Z \\
+100 &\text{metres} & 0''.20\Z \\
+ 1 &\text{kilometre} & 0''.15\Z \\
+ 10 &\text{kilometres} & 0''.10\Z \\
+ 20 &\text{kilometres} & 0''.084 \\
+100 &\text{kilometres} & 0''.050
+\end{array}
+\]
+At the centre of the continent, $1,950$~miles from
+the coast, the range will be~$0''.012$.
+
+If all the assumed data be varied, the ranges
+of the slopes are easily calculable, but these
+\PageSep{143}
+results may be taken as fairly representative, although
+perhaps somewhat underestimated. Lord
+Kelvin has made an entirely independent estimate
+\index{Kelvin, Lord!calculation of tidal attraction}%
+of the probable deflection of a pendulum
+by the direct attraction of the sea at high tide.
+\index{Attraction!of tide calculated}%
+He supposes the tide to have a range of $10$~feet
+from low water to high water, and he then estimates
+the attraction of a slab of water $10$~feet
+thick, $50$~miles broad perpendicular to the coast,
+and $100$~miles long parallel to the coast, on a
+plummet $100$~yards from low-water mark and
+opposite the middle of the $100$~miles. This
+would, he thinks, very roughly represent the
+state of things at St.~Alban's Head, in England.
+He finds the attraction such as to deflect the
+plumb-line, as high water changes to low water,
+by a twentieth of a second of arc. The general
+law as to the proportionality of slope to
+attraction shows that, with our supposed degree
+of stiffness of rock, the apparent deflection of a
+plumb-line, due to the depression of the coast
+and the attraction of the sea as high water
+changes to low water, will then be a quarter of a
+second of arc. Postulating a smaller tide, but
+spread over a wider area, I found the result
+would be a fifth of a second; thus the two results
+present a satisfactory agreement.
+
+This speculative investigation receives confirmation
+from observation. The late M.~d'Abbadie
+\index{Abbadie, tidal deflection of vertical}%
+established an observatory at his château of
+\index{Deflection of the vertical!due to tide|)}%
+\PageSep{144}
+\index{Abbadie, tidal deflection of vertical}%
+Abbadia, close to the Spanish frontier and within
+a quarter of a mile of the Bay of Biscay. Here
+he constructed a special form of instrument for
+detecting small changes in the direction of gravity.
+Without going into details, it may suffice
+to state that he compared a fixed mark with its
+image formed by reflection from a pool of mercury.
+He took $359$~special observations at the
+times of high and low tide in order to see, as he
+says, whether the water exercised an attraction
+on the pool of mercury, for it had not occurred
+to him that the larger effect would probably
+arise from the bending of the rock. He found
+that in $243$~cases the pool of mercury was tilted
+towards the sea at high water or away from it at
+low water; in $59$~cases there was no apparent
+effect, and in the remaining $57$~cases the action
+was inverted. The observations were repeated
+later by his assistant in the case of $71$~successive
+high waters\footnote
+ {Presumably the observation at one high water was defective.}
+and $73$~low waters, and he also
+found that in about two thirds of the observations
+the sea seemed to exercise its expected
+influence. We may, I think, feel confident that
+on the occasions where no effect or a reversal
+was perceived, it was annulled or reversed by a
+warping of the soil, such as is observed with
+seismometers.
+
+Dr.~von Rebeur-Paschwitz also noted deflections
+\index{Paschwitz, von Rebeur!tidal deflection of vertical at Wilhelmshaven}%
+due to the tide at Wilhelmshaven in Germany.
+\PageSep{145}
+\index{Atmospheric pressure!distortion of soil by}%
+\index{Elastic distortion!by atmospheric pressure|(}%
+\index{Pressure of atmosphere, elastic distortion of soil by}%
+The deflection was indeed of unexpected
+magnitude at this place, and this may probably
+be due to the peaty nature of the soil, which
+renders it far more yielding than if the observatory
+were built on rock.
+
+This investigation has another interesting application,
+for the solid earth has to bear another
+varying load besides that of the tide. The
+atmosphere rests on the earth and exercises a
+variable pressure, as shown by the varying
+height of the barometer. The variation of
+pressure is much more considerable than one
+would be inclined to suspect off-hand. The
+height of the barometer ranges through nearly
+two inches, or say five centimetres; this means
+that each square yard of soil supports a weight
+greater by $1,260$~lbs.\ when the barometer is very
+high, than when it is very low. If we picture
+to ourselves a field loaded with half a ton to
+each square yard, we may realize how enormous
+is the difference of pressure in the two cases.
+
+In order to obtain some estimate of the effects
+of the changing pressure, I will assume, as before,
+that the rocks are a quarter as stiff again
+as the stiffest glass. On a thick slab of this
+material let us imagine a train of parallel waves
+of air, such that at the crests of the waves the
+barometer is $5$~centimetres higher than at the
+hollow. Our knowledge of the march of barometric
+gradients on the earth's surface makes it
+\PageSep{146}
+plausible to assume that it is $1,500$~miles from
+the line of highest to that of lowest pressure.
+Calculation then shows that the slab is distorted
+into parallel ridges and valleys, and that the
+tops of the ridges are $9$~centimetres, or $3\frac{1}{2}$~inches,
+higher than the hollows. Although the actual
+distribution of barometric pressures is not of this
+simple character, yet this calculation shows, with
+a high degree of probability, that when the
+barometer is very high we are at least $3$~inches
+nearer the earth's centre than when it is very
+low.
+
+The consideration of the effects of atmospheric
+\index{Atmospheric pressure!distortion of soil by}%
+\index{Level of sea affected by atmospheric pressure}%
+\index{Pressure of atmosphere, elastic distortion of soil by}%
+pressure leads also to other curious conclusions.
+I have remarked before that the sea must respond
+\index{Sea!level affected by atmospheric pressure}%
+to barometric pressure, being depressed
+by high and elevated by low pressure. Since a
+column of water $68$~centimetres ($2$~ft.\ $3$~in.)\ in
+height weighs the same as a column, with the
+same cross section, of mercury, and $5$~centimetres
+in height, the sea should be depressed by $68$~centimetres
+under the very high barometer as compared
+with the very low barometer. But the
+height of the water can only be determined with
+reference to the land, and we have seen that the
+land must be depressed by $9$~centimetres. Hence
+the sea would be apparently depressed by only $59$~centimetres.
+
+It is probable that, in reality, the larger barometric
+inequalities do not linger quite long
+\PageSep{147}
+enough over particular areas to permit the sea to
+attain everywhere its due slope, and therefore the
+full difference of water level can only be attained
+occasionally. On the other hand the elastic compression
+\index{Elastic distortion!by atmospheric pressure|)}%
+of the ground must take place without
+sensible delay. Thus it seems probable that this
+compression must exercise a very sensible effect
+in modifying the apparent depression or elevation
+of the sea under high and low barometer.
+
+If delicate observations are made with some
+form of pendulum, the air waves and the consequent
+distortions of the soil should have a sensible
+effect on the instrument. In the ideal case
+which I have described above, it appears that
+the maximum apparent deflection of the plumb-line
+would be $\frac{1}{90}$~of a second of arc; this would
+be augmented to $\frac{1}{70}$~of a second by the addition
+of the true deflection, produced by the attraction
+of the air. Lastly, since the slope and attraction
+would be absolutely reversed when the air wave
+assumed a different position with respect to the
+observer, it is clear that the range of apparent
+oscillation of the pendulum might amount to
+$\frac{1}{35}$~of a second of arc.
+
+This oscillation is actually greater than that
+due to the direct tidal force of the moon acting
+on a pendulum suspended on an ideally unyielding
+earth. Accordingly we have yet another
+reason why the direct measurement of the tidal
+force presents a problem of the extremest difficulty.
+\PageSep{148}
+
+\begin{Authorities}
+G.~H. Darwin, \Title{Appendix to the Second Report on Lunar Disturbance
+of Gravity}. ``Brit.\ Assoc.\ Reports.'' Southampton,~1882.
+
+Reprint of the same in the ``Philosophical Magazine.''
+
+d'Abbadie, \Title{Recherches sur la verticale}. ``Ann.\ de~la Soc.\ Scient.\
+de~Bruxelles,'' 1881.
+
+von Rebeur-Paschwitz, \Title{Das Horizontalpendel}. ``Nova Acta
+K.~Leop.\ Car.\ Akad.,'' Band~60, No.~1, 1892.
+\end{Authorities}
+\index{Darwin, G. H.!distortion of earth's surface by varying loads|)}%
+\index{Distortion of soil!by varying loads|)}%
+\index{Elastic distortion!of earth by varying loads|)}%
+\PageSep{149}
+
+
+\Chapter{VIII}
+{Equilibrium Theory of Tides}
+
+\First{It} is clearly necessary to proceed step by step
+\index{Equilibrium theory of tides|(}%
+towards the actual conditions of the tidal problem,
+and I shall begin by supposing that the
+oceans cover the whole earth, leaving no dry
+land. It has been shown in \Ref{Chapter}{V}.\ that the
+tidal force is the resultant of opposing centrifugal
+and centripetal forces. The motion of the
+system is therefore one of its most essential features.
+We may however imagine a supernatural
+being, who carries the moon round the earth and
+makes the earth rotate at the actual relative
+speeds, but with indefinite slowness as regards
+absolute time. This supernatural being is further
+to have the power of maintaining the tidal forces
+at exactly their present intensities, and with their
+actual relationship as regards the positions of
+the moon and earth. Everything, in fact, is to
+remain as in reality, except time, which is to be
+indefinitely protracted. The question to be considered
+is as to the manner in which the tidal
+forces will cause the ocean to move on the slowly
+revolving earth.
+
+It appears from \fig{23} that the horizontal
+\PageSep{150}
+tidal force acts at right angles to the circle, where
+the moon is in the horizon, just rising or just
+setting, towards those two points, $V$~and~$I$, where
+the moon is overhead in the zenith, or underfoot
+in the nadir. The force will clearly generate
+currents in the water away from the circle of
+moonrise and moonset, and towards $V$~and~$I$.
+The currents will continue to flow until the water
+level is just so much raised above the primitive
+surface at $V$~and~$I$, and depressed along the circle,
+that the tendency to flow downhill towards
+the circle is equal to the tendency to flow uphill
+under the action of the tide-generating force.
+When the currents have ceased to flow, the figure
+of the ocean has become elongated, or egg-shaped
+with the two ends alike, and the longer
+axis of the egg is pointed at the moon. When
+this condition is attained the system is at rest or
+in equilibrium, and the technical name for the
+egg-like form is a ``prolate ellipsoid of revolution''---``prolate''
+because it is elongated, and
+``of revolution'' because it is symmetrical with
+respect to the line pointing at the moon. Accordingly
+the mathematician says that the figure
+of equilibrium under tide-generating force is a
+prolate ellipsoid of revolution, with the major
+axis directed to the moon.
+
+It has been supposed that the earth rotates and
+that the moon revolves, but with such extreme
+slowness that the ocean currents have time
+\PageSep{151}
+\index{Figure of equilibrium!of ocean under tidal forces|(}%
+\index{Forces!figure of equilibrium under tidal|(}%
+enough to bring the surface to its form of equilibrium,
+\index{Equilibrium theory of tides!chart and law of tide|(}%
+at each moment of time. If the time be
+sufficiently protracted, this is a possible condition
+of affairs. It is true that with the earth spinning
+at its actual rate, and with the moon revolving
+as in nature, the form of equilibrium can
+never be attained by the ocean; nevertheless it
+is very important to master the equilibrium
+theory.
+
+\Fig{29} represents the world in two hemispheres,
+as in an ordinary atlas, with parallels
+of latitude drawn at $15°$~apart. At the moment
+represented, the moon is supposed to be in the
+zenith at $15°$~of north latitude, in the middle of
+the right-hand hemisphere. The diametrically
+opposite point is of course at $15°$~of south latitude,
+in the middle of the other hemisphere.
+These are the two points $V$~and~$I$ of figs.\ \figref{22}~and~\figref{23},
+towards which the water is drawn, so that the
+vertices of the ellipsoid are at these two spots.
+A scale of measurement must be adopted for
+estimating the elevation of the water above, and
+its depression below the original undisturbed surface
+of the globe. It will be convenient to measure
+the elevation at these two spots by the
+number~$2$. A series of circles are drawn round
+these points, but one of them is, of necessity,
+presented as partly in one hemisphere and partly
+in the other. In the map they are not quite concentric
+with the two spots, but on the actual
+\PageSep{152}
+\TallFig{29}{Chart of Equilibrium Tide}{png}
+\PageSep{153}
+\index{Semidiurnal tide!in equilibrium theory|(}%
+globe they would be so. These circles show
+where, on the adopted scale of measurement, the
+elevation of height is successively $1\frac{1}{2}$,~$1$,~$\frac{1}{2}$. The
+fourth circle, marked in chain dot, shows where
+there is no elevation or depression above the original
+surface. The next succeeding and dotted
+circle shows where there is a depression of~$\frac{1}{2}$, and
+the last dotted line is the circle of lowest water
+where the depression is~$1$; it is the circle~$\Seg{D}{D}$ of~\fig{22},
+and the circle of the shadow in~\fig{23}.
+
+The elevation above the original spherical surface
+at the vertices or highest points is just twice
+as great as the greatest depression. But the
+greatest elevation only occurs at two points,
+whereas the greatest depression is found all along
+a circle round the globe. The horizontal tide-generating
+force is everywhere at right angles to
+these circles, and the present figure is in effect a
+reproduction, in the form of a map, of the perspective
+picture in~\fig{23}.
+
+Now as the earth turns from west to east, let
+us imagine a man standing on an island in the
+otherwise boundless sea, and let us consider what
+he will observe. Although the earth is supposed
+to be revolving very slowly, we may still call the
+twenty-fourth part of the time of its rotation an
+hour. The man will be carried by the earth's
+rotation along some one of the parallels of latitude.
+If, for example, his post of observation is
+in latitude $30°$~N., he will pass along the second
+\index{Equilibrium theory of tides!chart and law of tide|)}%
+\index{Figure of equilibrium!of ocean under tidal forces|)}%
+\index{Forces!figure of equilibrium under tidal|)}%
+\PageSep{154}
+parallel to the north of the equator. This parallel
+cuts several of the circles which indicate the
+elevation and depression of the water, and therefore
+he will during his progress pass places where
+the water is shallower and deeper alternately, and
+he would say that the water was rising and falling
+rhythmically. Let us watch his progress
+across the two hemispheres, starting from the
+extreme left. Shortly after coming into view he
+is on the dotted circle of lowest water, and he
+says it is low tide. As he proceeds the water
+rises, slowly at first and more rapidly later, until
+he is in the middle of the hemisphere; he arrives
+there six hours later than when we first began to
+watch him. It will have taken him about $5\frac{1}{2}$~hours
+to pass from low water to high water. At
+low water he was depressed by~$1$ below the original
+level, and at high water he is raised by~$\frac{1}{2}$
+above that level, so that the range from low
+water to high water is represented by~$1\frac{1}{2}$. After
+the passage across the middle of the hemisphere,
+the water level falls, and after about $5\frac{1}{2}$~hours
+more the water is again lowest, and the depression
+is measured by~$1$ on the adopted scale.
+Soon after this he passes out of this hemisphere
+into the other one, and the water rises again
+until he is in the middle of that hemisphere.
+But this time he passes much nearer to the vertex
+of highest water than was the case in the other
+hemisphere, so that the water now rises to a
+\PageSep{155}
+height represented by about~$1\frac{4}{5}$. In this half of
+his daily course the range of tide is from $1$~below
+to $1\frac{4}{5}$~above, and is therefore~$2\frac{4}{5}$, whereas before
+it was only~$1\frac{1}{2}$. The fact that the range of two
+successive tides is not the same is of great importance
+in tidal theory; it is called the diurnal
+inequality of the tide.
+
+It will have been noticed that in the left hemisphere
+the range of fall below the original spherical
+surface is greater than the range of rise
+above it; whereas in the right hemisphere the
+rise is greater than the fall. Mean water mark
+is such that the tide falls on the average as much
+below it as it rises above it, but in this case the
+rise and fall have been measured from the originally
+undisturbed surface. In fact the mean
+level of the water, in the course of a day, is not
+identical with the originally undisturbed surface,
+although the two levels do not differ much from
+one another.
+
+The reader may trace an imaginary observer
+in his daily progress along any other parallel of
+latitude, and will find a similar series of oscillations
+in the ocean; each latitude will, however,
+present its own peculiarities. Then again the
+moon moves in the heavens. In \fig{29} she has
+been supposed to be $15°$~north of the equator,
+but she might have been yet further northward,
+or on the equator, or to the south of it. Her
+extreme range is in fact $28°$~north or south of
+\PageSep{156}
+\index{Forces!those of sun and moon compared|(}%
+\index{Lunar!tide-generating force compared with solar|(}%
+\index{Moon and earth!tide-generating force compared with sun's|(}%
+\index{Solar!tide-generating force compared with lunar|(}%
+the equator. To represent each such case a new
+map would be required, which would, however,
+only differ from this one by the amount of displacement
+of the central spots from the equator.
+
+It is obvious that the two hemispheres in \fig{29}
+are exactly alike, save that they are inverted
+with respect to north and south; the right hemisphere
+is in fact the same as the left upside down.
+It is this inversion which causes the two successive
+tides to be unlike one another, or, in other
+words, gives rise to the diurnal inequality. But
+\index{Diurnal inequality!according to equilibrium theory}%
+there is one case where inversion makes no difference;
+this is when the central spot is on the
+equator in the left hemisphere, for its inversion
+then makes the right hemisphere an exact reproduction
+of the left one. In this case therefore
+the two successive tides are exactly alike, and
+there is no diurnal inequality. Hence the diurnal
+inequality vanishes when the moon is on the
+equator.
+
+Our figure exhibits another important point,
+for it shows that the tide has the greater range
+in that hemisphere where the observer passes
+nearest to one of the two central spots. That is
+to say, the higher tide occurs in that half of the
+daily circuit in which the moon passes nearest to
+the zenith or to the nadir of the observer.
+
+Thus far I have supposed the moon to exist
+alone, but the sun also acts on the ocean according
+\index{Sun!tide-generating force of, compared with that of moon|(}%
+to similar laws, although with less intensity.
+\index{Semidiurnal tide!in equilibrium theory|)}%
+\PageSep{157}
+We must now consider how the relative strengths
+of the actions of the two bodies are to be determined.
+It was indicated in \Ref{Chapter}{V}.\ that
+tide-generating force varies inversely as the cube
+of the distance from the earth of the tide-generating
+body. The force of gravity varies inversely
+as the square of the distance, so that, as
+we change the distance of the attracting body,
+tidal force varies with much greater rapidity than
+does the direct gravitational attraction. Thus if
+the moon stood at half her present distance from
+the earth, her tide-generating force would be $8$~times
+as great, whereas her direct attraction would
+only be multiplied $4$~times. It is also obvious
+that if the moon were twice as heavy as in reality,
+her tide-generating force would be doubled;
+and if she were half as heavy it would be halved.
+Hence we conclude that tide-generating force
+varies directly as the mass of the tide-generating
+body, and inversely as the cube of the distance.
+
+The application of this law enables us to compare
+the sun's tidal force with that of the moon.
+The sun is $25,500,000$~times as heavy as the
+moon, so that, on the score of mass, the solar
+tidal force should be $25\frac{1}{2}$~million times greater
+than that of the moon. But the sun is $389$~times
+as distant as the moon. And since the
+cube of~$389$ is about $59$~millions, the solar tidal
+force should be $59$~million times weaker than
+that of the moon, on the score of distance.
+\PageSep{158}
+
+We have, then, a force which is $25\frac{1}{2}$~million
+times stronger on account of the sun's greater
+weight, and $59$~million times weaker on account
+of his greater distance; it follows that the sun's
+tide-generating force is $25\frac{1}{2}$-$59$ths, or a little
+less than half of that of the moon.
+
+We conclude then that if the sun acted alone
+on the water, the degree of elongation or distortion
+of the ocean, when in equilibrium, would
+be a little less than half of that due to the moon
+alone. When both bodies act together, the distortion
+of the surface due to the sun is superposed
+on that due to the moon, and a terrestrial
+observer perceives only the total or sum of the
+two effects.
+
+When the sun and moon are together on the
+same side of the earth, or when they are diametrically
+opposite, the two distortions conspire
+together, and the total tide will be half as great
+again as that due to the moon alone, because
+the solar tide is added to the lunar tide. And
+when the sun and moon are at right angles to
+\index{Sun!tide-generating force of, compared with that of moon|)}%
+one another, the two distortions are at right
+angles, and the low water of the solar tide conspires
+with the high water of the lunar tide.
+The composite tide has then a range only half as
+great as that due to the moon alone, because the
+solar tide, which has a range of about half that
+\index{Solar!tide-generating force compared with lunar|)}%
+of the lunar tide, is deducted from the lunar
+\index{Lunar!tide-generating force compared with solar|)}%
+tide. Since one and a half is three times a half,
+\index{Forces!those of sun and moon compared|)}%
+\index{Moon and earth!tide-generating force compared with sun's|)}%
+\PageSep{159}
+\index{Spring and neap tides!in equilibrium theory}%
+it follows that when the moon and sun act together
+the range of tide is three times as great
+as when they act adversely. The two bodies
+are together at change of moon and opposite at
+full moon. In both of these positions their
+actions conspire; hence at the change and the
+full of moon the tides are at their largest, and
+are called spring tides. When the two bodies
+are at right angles to one another, it is half
+moon, either waxing or waning, the tides have
+their smallest range, and are called neap tides.
+\index{Neap and spring tides!in equilibrium theory}%
+
+The observed facts agree pretty closely with
+this theory in several respects, for spring tide
+occurs about the full and change of moon, neap
+tide occurs at the half moon, and the range at
+springs is usually about three times as great as
+that at neaps. Moreover, the diurnal inequality
+conforms to the theory in vanishing when the
+moon is on the equator, and rising to a maximum
+when the moon is furthest north or south. The
+amount of the diurnal inequality does not, however,
+agree with theory, and in many places the
+tide which should be the greater is actually the
+less.
+
+The theory which I have sketched is called
+the Equilibrium Theory of the Tides, because
+it supposes that at each moment the ocean is
+in that position of rest or equilibrium which it
+would attain if indefinite time were allowed.
+The general agreement with the real phenomena
+\PageSep{160}
+\index{Equilibrium theory of tides!defects of}%
+proves the theory to have much truth about it,
+but a detailed comparison with actuality shows
+that it is terribly at fault. The lunar and solar
+tidal ellipsoids were found to have their long
+axes pointing straight towards the tide-generating
+bodies, and, therefore, at the time when the
+moon and sun pull together, it ought to be high
+water just when they are due south. In other
+words, at full and change of moon, it ought
+to be high water exactly at noon and at midnight.
+\index{High water!under moon in equilibrium theory}%
+Now observation at spring tides shows
+that at most places this is utterly contradictory
+to fact.
+
+It is a matter of rough observation that the
+tides follow the moon's course, so that high
+water always occurs about the same number of
+hours after the moon is due south. This rule
+has no pretension to accuracy, but it is better
+than no rule at all. Now at change and full of
+the moon, the moon crosses the meridian at the
+same hour of the clock as the sun, for at change
+of moon they are together, and at full moon
+they are twelve hours apart. Hence the hour
+of the clock at which high water occurs at
+change and full of moon is in effect a statement
+of the number of hours which elapse after the
+moon's passage of the meridian up to high
+water. This clock time affords a rough rule for
+the time of high water at any other phase of the
+moon; if, for example, it is high water at eight
+\PageSep{161}
+o'clock at full and change, approximately eight
+hours will always elapse after the moon's passage
+until high water occurs. Mariners call the clock
+time of high water at change and full of moon
+\index{Establishment of port!definition}%
+\index{Establishment of port!zero in equilibrium theory}%
+``the establishment of the port,'' because it
+establishes a rough rule of the tide at all other
+times.
+
+According to the equilibrium theory, high
+water falls at noon and midnight at full and
+change of moon, or in the language of the mariner
+the establishment of all ports should be
+zero. But observation shows that the establishment
+at actual ports has all sorts of values, and
+that in the Pacific Ocean (where the tidal forces
+have free scope) it is at least much nearer to six
+hours than to zero. High water cannot be more
+than six hours before or after noon or midnight
+on the day of full or change of moon, because if
+it occurs more than six hours after one noon, it
+is less than six hours before the following midnight;
+hence the establishment of any port
+cannot possibly be more than six hours before or
+after. Accordingly, the equilibrium theory is
+nearly as much wrong as possible, in respect to
+the time of high water. In fact, in many places
+it is nearly low water at the time that the equilibrium
+theory predicts high water.
+
+It would seem then as if the tidal action of
+the moon was actually to repel the water instead
+of attracting it, and we are driven to ask whether
+\PageSep{162}
+\index{Establishment of port!definition}%
+this result can possibly be consistent with the
+theory of universal gravitation.
+
+The existence of continental barriers across
+the oceans must obviously exercise great influence
+on the tides, but this fact can hardly be
+responsible for a reversal of the previsions of the
+equilibrium theory. It was Newton who showed
+that a depression of the ocean under the moon
+is entirely consistent with the theory of gravitation.
+In the following chapter I shall explain
+Newton's theory, and show how it explains the
+discrepancy which we have found between the
+equilibrium theory and actuality.
+
+\begin{Authorities}
+An exposition of the equilibrium theory will be found in any
+mathematical work on the subject, or in the article \Title{Tides} in the
+``Encyclopædia Britannica.''
+\end{Authorities}
+\index{Equilibrium theory of tides|)}%
+\PageSep{163}
+
+
+\Chapter{IX}
+{Dynamical Theory of the Tide Wave}
+
+\First{The} most serious difficulties in the complete
+\index{Dynamical theory of tide-wave|(}%
+tidal problem do not arise in a certain special
+case which was considered by Newton. His supposition
+was that the sea is confined to a canal
+\index{Canal!critical depth|(}%
+circling the equator, and that the moon and sun
+move exactly in the equator.
+
+An earthquake or any other gigantic impulse
+may be supposed to generate a great wave in this
+equatorial canal. The rate of progress of such
+a wave is dependent on the depth of the canal
+only, according to the laws sketched in \Ref{Chapter}{II}.,
+and the earth's rotation and the moon's attraction
+make no sensible difference in its speed
+of transmission. If, for example, the canal were
+$5$~kilometres ($3$~miles) in depth, such a great
+wave would travel $796$~kilometres ($500$~miles)
+per~hour. If the canal were shallower the speed
+would be less than this; if deeper, greater.
+Now there is one special depth which will be
+found to have a peculiar importance in the theory
+of the tide, namely, where the canal is $13\frac{3}{4}$~miles
+deep. In this case the wave travels $1,042$~miles
+an hour, so that it would complete the
+\PageSep{164}
+$25,000$~miles round the earth in exactly $24$~hours.
+It is important to note that if the depth of the
+equatorial canal be less than $13\frac{3}{4}$~miles, a wave
+takes more than a day to complete the circuit of
+the earth, and if the depth be greater the circuit
+is performed in less than a day.
+
+The great wave, produced by an earthquake or
+other impulse, is called a ``free wave,'' because
+\index{Free wave, explanation and contrast with forced}%
+\index{Waves!forced and free}%
+when once produced it travels free from the action
+of external forces, and would persist forever,
+were it not for the friction to which water is
+necessarily subject. But the leading characteristic
+of the tide wave is that it is generated and
+kept in action by continuous forces, which act
+on the fluid throughout all time. Such a wave
+is called a ``forced wave,'' because it is due to
+\index{Forced wave, explanation and contrast with free wave}%
+the continuous action of external forces. The
+rate at which the tide wave moves is moreover
+dependent only on the rate at which the tidal
+forces travel over the earth, and not in any degree
+on the depth of the canal. It is true that
+the depth of the canal exercises an influence on
+the height of the wave generated by the tidal
+forces, but the wave itself must always complete
+the circuit of the earth in a day, because the
+earth turns round in that period.
+
+We must now contrast the progress of any
+long ``free wave'' in the equatorial canal with
+that of the ``forced'' tide wave. I may premise
+that it will here be slightly more convenient to
+\PageSep{165}
+consider the solar instead of the lunar tide. The
+lunar wave is due to a stronger tide-generating
+force, and since the earth takes $24$~hours $50$~minutes
+to turn round with respect to the moon, that
+is the time which the lunar tide wave takes to
+complete the circuit of the earth; but these differences
+are not material to the present argument.
+The earth turns with respect to the sun
+in exactly one day, or as we may more conveniently
+say, the sun completes the circuit of the
+earth in that time. Therefore the solar tidal
+influence travels over the surface of the earth
+at the rate of $1,042$~miles an hour. Now this is
+exactly the pace at which a ``free wave'' travels
+in a canal of a depth of $13\frac{3}{4}$~miles; accordingly
+\index{Canal!theory of tide wave in|(}%
+in such a canal any long free wave just keeps
+pace with the sun.
+
+We have seen in \Ref{Chapter}{V}.\ that the solar
+tide-generating force \emph{tends} to make a wave crest,
+at those points of the earth's circumference where
+it is noon and midnight. At each moment of
+time the sun is generating a new wave, and after
+it is generated that wave travels onwards as a
+free wave. If therefore the canal has a depth
+\index{Canal!critical depth|)}%
+of $13\frac{3}{4}$~miles, each new wave, generated at each
+moment of time, keeps pace with the sun, and
+the summation of them all must build up two
+enormous wave crests at opposite sides of the
+earth.
+
+If the velocity of a free wave were absolutely
+\PageSep{166}
+the same whatever were its height, the crests of
+the two tide waves would become infinite in
+height. As a fact the rate of progress of a wave
+is somewhat influenced by its height, and therefore,
+when the waves get very big, they will
+cease to keep pace exactly with the sun, and
+the cause for continuous exaggeration of their
+heights will cease to exist. We may, however,
+express this conclusion by saying that, when the
+canal is $13\frac{3}{4}$~miles deep, the height of the tide
+wave becomes mathematically infinite. This does
+not mean that mathematicians assert that the
+wave would really become infinite, but only that
+the simple method of treatment which supposes
+the wave velocity to depend only on the depth
+of water becomes inadequate. If the ocean was
+really confined to an equatorial canal, of this exact
+depth, the tides would be of very great
+height, and the theory would be even more complex
+than it is. It is, however, hardly necessary
+to consider this special case in further detail.
+
+We conclude then that for the depth of $13\frac{3}{4}$~miles,
+the wave becomes infinite in height, in
+the qualified sense of infinity which I have described.
+We may feel sure that the existence
+of the quasi-infinite tide betokens that the behavior
+of the water in a canal shallower than
+$13\frac{3}{4}$~miles differs widely from that in a deeper
+one. It is therefore necessary to examine into
+the essential point in which the two cases differ
+\PageSep{167}
+from one another. In the shallower canal a free
+wave covers less than $25,000$~miles a day, and
+thus any wave generated by the sun would tend
+to be left behind by him. On the other hand,
+in the deeper canal a free wave would outstrip
+the sun, and a wave generated by the sun tends
+to run on in advance of him. But these are
+only tendencies, for in both the shallower and
+the deeper canal the actual tide wave exactly
+keeps pace with the sun.
+
+It would be troublesome to find out what
+would happen if we had the water in the canal
+at rest, and were suddenly to start the sun to
+work at it; and it is fortunately not necessary
+to attempt to do so. It is, however, certain that
+for a long time the motion would be confused,
+but that the friction of the water would finally
+produce order out of chaos, and that ultimately
+there would be a pair of antipodal tide crests
+traveling at the same pace as the sun. Our
+task, then, is to discover what that final state of
+motion may be, without endeavoring to unravel
+the preliminary chaos.
+
+Let us take a concrete case, and suppose our
+canal to be $3$~miles deep, in which we have seen
+that a free wave will travel $500$~miles an hour.
+Suppose, then, we start a long free wave in the
+equatorial canal of $3$~miles deep, with two crests
+\index{Canal!theory of tide wave in|)}%
+$12,500$~miles apart, and therefore antipodal to
+one another. The period of a wave is the time
+\PageSep{168}
+between the passage of two successive crests
+past any fixed point. In this case the crests
+are antipodal to one another, and therefore the
+wave length is $12,500$~miles, and the wave
+travels $500$~miles an hour, so that the period of a
+free wave is $25$~hours. But the tide wave keeps
+pace with the sun, traveling $1,042$~miles an
+hour, and there are two antipodal crests, $12,500$~miles
+apart; hence, the time between the passage
+of successive tide crests is $12$~hours.
+
+In this case a free wave would have a period
+of $25$~hours, and the tide wave, resulting from
+the action of solar tide-generating force, has a
+period of $12$~hours. The contrast then lies between
+the free wave, with a period of $25$~hours,
+and the forced wave, with a period of $12$~hours.
+
+For any other depth of ocean the free wave
+will have another period depending on the depth,
+but the period of the forced wave is always $12$~hours,
+because it depends on the sun. If the
+ocean be shallower than $3$~miles, the free period
+will be greater than $25$~hours, and, if deeper,
+less than $25$~hours. But if the ocean be deepened
+to $13\frac{3}{4}$~miles, the free wave travels at the
+same pace as the forced wave, and therefore the
+two periods are coincident. For depths greater
+than $13\frac{3}{4}$~miles, the period of the free wave is
+less than that of the forced wave; and the
+converse is true for canals less than $13\frac{3}{4}$~miles in
+depth.
+\PageSep{169}
+\index{Forced oscillation!principle of}%
+\index{Free oscillation contrasted with forced}%
+\index{Principle of forced oscillations}%
+
+Now let us generalize this conception; we
+have a system which if disturbed and left to
+itself will oscillate in a certain period, called the
+free period. Periodic disturbing forces act on
+this system and the period of the disturbance
+is independent of the oscillating system itself.
+The period of the disturbing forces is called the
+forced period. How will such a system swing,
+when disturbed with this forced periodicity?
+
+A weight tied to the end of a string affords
+an example of a very simple system capable of
+oscillation, and the period of its free swing depends
+on the length of the string only. I will
+suppose the string to be $3$~feet, $3$~inches, or one
+metre in length, so that the period of the swing
+from right to left, or from left to right is one
+second.\footnote
+ {A pendulum of one metre in length is commonly called a
+ seconds-pendulum, although its complete period is two seconds.}
+If, holding the string, I move my
+hand horizontally to and fro through a small
+distance with a regular periodicity, I set the
+pendulum a-swinging. The period of the movement
+of my hand is the forced period, and the
+free period is two seconds, being the time occupied
+by a metre-long pendulum in moving from
+right to left and back again to right. If I time
+the to and fro motion of my hand so that its
+period from right to left, or from left to right,
+is exactly one second, the excursions of the pendulum
+bob grow greater and greater without limit,
+\PageSep{170}
+\index{Forced oscillation!principle of}%
+\index{Free oscillation contrasted with forced}%
+\index{Principle of forced oscillations}%
+because the successive impulses are stored up in
+the pendulum, which swings further and further
+with each successive impulse. This case is
+exactly analogous with the quasi-infinite tides
+which would arise in a canal $13\frac{3}{4}$~miles deep, and
+here also this case is critical, separating two
+modes of oscillation of the pendulum of different
+characters.
+
+Now when the hand occupies more than one
+second in moving from right to left, the forced
+period is greater than the free period of the
+pendulum; and when the system is swinging
+steadily, it will be observed that the excursion
+of the hand agrees in direction with the excursion
+of the pendulum, so that when the hand is
+furthest to the right so is also the pendulum,
+and vice versa. When the period of the force
+is greater than the free period of the system, at
+the time when the force tends to make the pendulum
+move to the right, it is furthest to the
+right. The excursions of the pendulum agree
+in direction with that of the hand.
+
+Next, when the hand occupies less than one
+second to move from right to left or from left
+to right, the forced period is less than the free
+period, and it will be found that when the hand
+is furthest to the right the pendulum is furthest
+to the left. The excursions of the pendulum
+are opposite in direction from those of the hand.
+These two cases are illustrated by~\fig{30}, which
+\PageSep{171}
+\index{High water!position in shallow and deep canals in dynamical theory}%
+will, perhaps, render my meaning more obvious.
+We may sum up this discussion by saying that
+in the case of a slowly varying disturbing force,
+the oscillation and the force are consentaneous,
+but that with a quickly oscillating force, the
+oscillation is exactly inverted with respect to the
+force.
+
+Now, this simple case illustrates a general
+dynamical principle, namely, that if a system
+\Figure{30}{Forced Oscillations of a Pendulum}{png}
+capable of oscillating with a certain period is
+acted on by a periodic force, when the period of
+the force is greater than the natural free period
+of the system, the oscillations of the system
+agree with the oscillations of the force; but if
+the period of the force is less than the natural
+free period of the system the oscillations are
+inverted with reference to the force.
+
+This principle may be applied to the case of
+the tides in the canal. When the canal is more
+than $13\frac{3}{4}$~miles deep, the period of the sun's disturbing
+force is $12$~hours and is greater than the
+\PageSep{172}
+\index{Newton!theory of tide in equatorial canal}%
+natural free period of the oscillation, because a
+free wave would go more than half round the
+earth in $12$~hours. We conclude, then, that when
+the tide-generating forces are trying to make it
+high water, it will be high water. It has been
+\index{High water!position in shallow and deep canals in dynamical theory}%
+shown that these forces are tending to make high
+water immediately under the sun and at its antipodes,
+and there accordingly will the high water
+be. In this case the tide is said to be direct.
+
+But when the canal is less than $13\frac{3}{4}$~miles
+deep, the sun's disturbing force has, as before, a
+period of $12$~hours, but the period of the free
+wave is more than $12$~hours, because a free wave
+would take more than $12$~hours to get half round
+the earth. Thus the general principle shows
+that where the forces are trying to make high
+water, there will be low water, and vice versa.
+Here, then, there will be low water under the sun
+and at its antipodes, and such a tide is said to
+be inverted, because the oscillation is the exact
+inversion of what would be naturally expected.
+
+All the oceans on the earth are very much
+shallower than fourteen miles, and so, at least
+near the equator, the tides ought to be inverted.
+The conclusion of the equilibrium theory will
+therefore be the exact opposite of the truth, near
+the equator.
+
+This argument as to the solar tide requires
+but little alteration to make it applicable to the
+lunar tide. In fact the only material difference
+\PageSep{173}
+\index{Waves!of tide in equatorial canal}%
+in the conditions is that the period of the lunar
+tide is $12$~hours $25$~minutes, instead of $12$~hours,
+and so the critical depth of an equatorial canal,
+\index{Equatorial canal, tide wave in}%
+which would allow the lunar tide to become
+quasi-infinite, is a little less than that for the
+solar tide. This depth for the lunar tide is in
+fact nearly $13$~miles.\footnote
+ {It is worthy of remark that if the canal had a depth of between
+ $13\frac{3}{4}$ and $13$~miles, the solar tides would be inverted, and
+ the lunar tides would be direct. We should then, at the equator,
+ have spring tide at half moon, when our actual neaps occur;
+ and neap tide at full and change, when our actual springs occur.
+ The tides would also be of enormous height, because the depth
+ is nearly such as to make both tides quasi-infinite. If the depth
+ of the canal were very nearly $13\frac{3}{4}$~miles the solar tide might be
+ greater than the lunar. But these exceptional cases have only
+ a theoretical interest.}
+
+This discussion should have made it clear that
+any tidal theory, worthy of the name, must take
+account of motion, and it explains why the prediction
+of the equilibrium theory is so wide from
+the truth. Notwithstanding, however, this condemnation
+of the equilibrium theory, it is of the
+utmost service in the discussion of the tides,
+because by far the most convenient and complete
+way of specifying the forces which act on the
+ocean at each instant is to determine the figure
+which the ocean would assume, if the forces had
+abundant time to act.
+
+\TB
+
+When the sea is confined to an equatorial
+canal, the tidal problem is much simpler than
+\PageSep{174}
+when the ocean covers the whole planet, and
+this is much simpler than when the sea is interrupted
+by continents. Then again, we have
+thus far supposed the sun and moon to be always
+exactly over the equator, whereas they actually
+range a long way both to the north and to the
+south of the equator; and so here also the true
+problem is more complicated than the one under
+consideration. Let us next consider a case, still
+far simpler than actuality, and suppose that
+whilst the moon or sun still always move over
+the equator, the ocean is confined to several
+canals which run round the globe, following parallels
+of latitude.
+\index{Latitude!tidal wave in canal in high|(}%
+
+The circumference of a canal in latitude~$60°$
+\index{Canal!canal in high latitude|(}%
+is only $12,500$~miles, instead of~$25,000$. If a
+free wave were generated in such a canal with
+two crests at opposite sides of the globe, the distance
+from crest to crest would be $6,250$~miles.
+Now if an equatorial canal and one in latitude~$60°$
+have equal depths, a free wave will travel at
+the same rate along each; and if in each canal
+there be a wave with two antipodal crests, the
+time occupied by the wave in latitude~$60°$ in
+traveling through a space equal to its length will
+be only half of the similar period for the equatorial
+waves. The period of a free wave in latitude~$60°$
+\index{Waves!in canal in high latitude|(}%
+is therefore half what it is at the
+equator, for a pair of canals of equal depths.
+But there is only one sun, and it takes $12$~hours
+\PageSep{175}
+to go half round the planet, and therefore for
+both canals the forced tide wave has a period of
+$12$~hours. If, for example, both canals were
+$8$~miles deep, in the equatorial canal the
+\index{Canal!tides in ocean partitioned into canals}%
+period of the free wave would be greater than
+$12$~hours, whilst in the canal at $60°$~of latitude
+it would be less than $12$~hours. It follows then
+from the general principle as to forced and free
+oscillations, that whilst the tide in the equatorial
+canal would be inverted, that in latitude~$60°$
+would be direct. Therefore, whilst it would be
+low water under the moon at the equator, it
+would be high water under the moon in latitude~$60°$.
+Somewhere, between latitude~$60°$ and the
+equator, there must be a place at which the free
+period in a canal $8$~miles deep is the same as
+the forced period, and in a canal at this latitude
+the tide would be infinite in height, in the modified
+sense explained earlier. It follows therefore
+that there is for any given depth of canal, less
+than $14$~miles, a critical latitude, at which the
+tide tends to become infinite in height.
+
+We conclude, that if the whole planet were
+divided up into canals each partitioned off from
+its neighbor, and if the canals were shallower
+than $14$~miles, we should have inverted tides in
+the equatorial region, and direct tides in the
+polar regions, and, in one of the canals in some
+middle latitude, very great tides the nature of
+which cannot be specified exactly.
+\PageSep{176}
+
+The supposed partitions between neighboring
+canals have introduced a limitation which must
+be removed, if we are to approach actuality, but
+I am unable by general reasoning to do more
+than indicate what will be the effect of the removal
+of the partitions. It is clear that when
+the sea swells up to form the high water, the
+water comes not only from the east and the west
+of the place of high water, but also from the
+north and south. The earth, as it rotates, carries
+with it the ocean; the equatorial water is
+carried over a space of $25,000$~miles in $24$~hours,
+whereas the water in latitude~$60°$ is carried over
+only $12,500$~miles in the same time. When, in
+the northern hemisphere, water moves from north
+to south it passes from a place where the surface
+of the earth is moving slower, to where it is
+moving quicker. Then, as the water goes to the
+south, it carries with it only the velocity adapted
+to the northern latitude, and so it gets left behind
+by the earth. Since the earth spins from
+west to east, a southerly current acquires a westward
+trend. Conversely, when water is carried
+northward of its proper latitude, it leaves the
+\index{Latitude!tidal wave in canal in high|)}%
+earth behind and is carried eastward. Hence
+the water cannot oscillate northward and southward,
+without at the same time oscillating eastward
+and westward. Since in an ocean not
+partitioned into canals, the water must necessarily
+move not only east and west but also north
+\index{Canal!canal in high latitude|)}%
+\index{Waves!in canal in high latitude|)}%
+\PageSep{177}
+\index{Earth and moon!rotation of, effects on tides}%
+\index{Rotation!of earth involved in tidal problem}%
+\index{Vortical motion in oceanic tides}%
+and south, it follows that tidal movements in the
+ocean must result in eddies or vortices. The
+\index{Eddies, tidal oscillation involves}%
+eddying motion of the water must exist everywhere,
+but it would be impossible, without mathematical
+reasoning, to explain how all the eddies
+fit into one another in time and place. It must
+suffice for the present discussion for the reader
+to know that the full mathematical treatment of
+the problem shows this general conclusion to be
+correct.
+
+The very difficult mathematical problem of
+the tides of an ocean covering the globe to a
+uniform depth was first successfully attacked by
+Laplace. He showed that whilst the tides of a
+\index{Laplace!theory of tides|(}%
+shallow ocean are inverted at the equator, as
+proved by Newton, that they are direct towards
+the pole. We have just arrived at the same
+conclusion by considering the tide wave in a
+canal in latitude~$60°$. But our reasoning indicated
+that somewhere in between higher latitudes
+and the equator, the tide would be of an undefined
+character, with an enormous range of rise
+and fall. The complete solution of the problem
+shows, however, that this indication of the
+canal theory is wrong, and that the tidal variation
+of level absolutely vanishes in some latitude
+intermediate between the equator and the pole.
+The conclusion of the mathematician is that
+there is a certain circle of latitude, whose position
+depends on the depth of the sea, where
+there is neither rise nor fall of tide.
+\PageSep{178}
+\index{Vortical motion in oceanic tides}%
+
+At this circle the water flows northward and
+southward, and to and fro between east and
+west, but in such a way as never to raise or depress
+the level of the sea. It is not true to say
+that there is no tide at this circle, for there are
+tidal currents without rise and fall. When the
+ocean was supposed to be cut into canals, we
+thereby obliterated the northerly and southerly
+currents, and it is exactly these currents which
+prevent the tides becoming very great, as we
+were then led to suppose they would be.
+
+It may seem strange that, whereas the first
+rough solution of the problem indicates an oscillation
+of infinite magnitude at a certain parallel
+of latitude, the more accurate treatment of the
+case should show that there is no oscillation of
+level at all. Yet to the mathematician such a
+result is not a cause of surprise. But whether
+strange or not, it should be clear that if at the
+equator it is low water under the moon, and if
+near the pole it is high water under the moon,
+there must in some intermediate latitude be a
+place where the water is neither high nor low,
+that is to say, where there is neither rise nor fall.\footnote
+ {The mathematician knows that a quantity may change sign,
+ either by passing through infinity or through zero. Where a
+ change from positive to negative undoubtedly takes place, and
+ where a passage through infinity can have no physical meaning,
+ the change must take place by passage through zero.}
+
+\TB
+
+Now let us take one more step towards actuality,
+and suppose the earth's equator to be
+\PageSep{179}
+oblique to the orbits of the moon and sun, so
+that they may sometimes stand to the north and
+sometimes to the south of the equator. We
+have seen that in this case the equilibrium theory
+indicates that the two successive tides on any
+one day have unequal ranges. The mathematical
+solution of the problem shows that this conclusion
+is correct. It appears also that if the
+ocean is deeper at the poles than at the equator,
+that tide is the greater which is asserted to be
+so by the equilibrium theory. If, however, the
+ocean is shallower at the poles than at the equator,
+it is found that the high water which the
+equilibrium theory would make the larger is actually
+the smaller and vice versa.
+
+If the ocean is of the same depth everywhere,
+we have a case intermediate between the two,
+where it is shallower at the poles, and where it is
+deeper at the poles. Now in one of these cases
+it appears that the higher high water occurs
+where in the other we find the lower high water
+to occur; and so, when the depth is uniform,
+the higher high water and the lower high water
+must attain the same heights. We thus arrive
+at the remarkable conclusion that, in an ocean
+of uniform depth, the diurnal inequality of the
+\index{Diurnal inequality!in Laplace's solution}%
+tide is evanescent. There are, however, diurnal
+inequalities in the tidal currents, which are so
+adjusted as not to produce a rise or fall. This
+result was first arrived at by the great mathematician
+Laplace.
+\PageSep{180}
+
+According to the equilibrium theory, when the
+moon stands some distance north of the equator,
+the inequality between the successive tides on
+the coasts of Europe should be very great, but
+the difference is actually so small as to escape
+ordinary observation. In the days of Laplace,
+the knowledge of the tides in other parts of the
+world was very imperfect, and it was naturally
+thought that the European tides were fairly
+representative of the whole world. When, then,
+it was discovered that there would be no diurnal
+inequality in an ocean of uniform depth covering
+the whole globe, it was thought that a fair explanation
+had been found for the absence of
+that inequality in Europe. But since the days
+of Laplace much has been learnt about the tides
+\index{Laplace!theory of tides|)}%
+in the Pacific and Indian oceans, and we now
+know that a large diurnal inequality is almost
+\index{Diurnal inequality!in Atlantic, Pacific, and Indian Oceans}%
+universal, so that the tides of the North Atlantic
+are exceptional in their simplicity. In fact, the
+evanescence of the diurnal inequality is not much
+closer to the truth than the large inequality
+predicted by the equilibrium theory; and both
+theories must be abandoned as satisfactory explanations
+of the true condition of affairs. But
+notwithstanding their deficiencies both these
+theories are of importance in teaching us how
+the tides are to be predicted. In the next chapter
+I shall show how a further approximation to
+the truth is attainable.
+\PageSep{181}
+
+\begin{Authorities}
+The canal theory in its elementary form is treated in many
+works on Hydrodynamics, and in \Title{Tides}, ``Encyclopædia Britannica.''
+
+An elaborate treatment of the subject is contained in Airy's
+\index{Airy, Sir G. B.!attack on Laplace}%
+\Title{Tides and Waves}, ``Encyclopædia Metropolitana.'' Airy there
+attacks Laplace for his treatment of the wider tidal problem,
+but his strictures are now universally regarded as unsound.
+
+Laplace's theory is contained in the \Title{Mécanique Céleste}, but it
+is better studied in more recent works.
+
+A full presentment of this theory is contained in Professor
+Horace Lamb's \Title{Hydrodynamics}, Camb.\ Univ.\ Press, 1895, chapter~viii.
+\index{Lamb, H., presentation of Laplace's theory}%
+
+Important papers, extending Laplace's work, by Mr.~S.~S.
+Hough, are contained in the \Title{Philosophical Transactions of the
+\index{Hough, S. S.!dynamical solution of tidal problem}%
+Royal Society}, A.~1897, pp.~201--258, and A.~1898, pp.~139--185.
+\end{Authorities}
+\index{Dynamical theory of tide-wave|)}%
+\PageSep{182}
+
+
+\Chapter{X}
+{Tides in Lakes---Cotidal Chart}
+
+\First{If} the conditions of the tidal problem are to
+\index{Lakes!tides in|(}%
+agree with reality, an ocean must be considered
+which is interrupted by continental barriers of
+land. The case of a sea or lake entirely surrounded
+by land affords the simplest and most
+complete limitation to the continuity of the
+water. I shall therefore begin by considering
+the tides in a lake.
+
+The oscillations of a pendulum under the tidal
+attraction of the moon were considered in \Ref{Chapter}{VI}.,
+and we there saw that the pendulum
+would swing to and fro, although the scale of
+displacement would be too minute for actual
+observation. Now a pendulum always hangs
+perpendicularly to the surface of water, and
+must therefore be regarded as a sort of level.
+As it sways to and fro under the changing action
+of the tidal force, so also must the surface
+of water. If the water in question is a lake, the
+rocking of the level of the lake is a true tide.
+
+A lake of say a hundred miles in length is
+very small compared with the size of the earth,
+and its waters must respond almost instantaneously
+\PageSep{183}
+to the changes in the tidal force. Such
+a lake is not large enough to introduce, to a
+perceptible extent, those complications which
+make the complete theory of oceanic tides so
+difficult. The equilibrium theory is here actually
+true, because the currents due to the changes in
+the tidal force have not many yards to run before
+equilibrium is established, and the lake may
+be regarded as a level which responds almost
+instantaneously to the tidal deflections of gravity.
+The open ocean is a great level also, but sufficient
+time is not allowed it to respond to the changes
+in the direction of gravity, before that direction
+has itself changed.
+
+It was stated in \Ref{Chapter}{V}.\ that the maximum
+horizontal force due to the moon has an intensity
+equal to $\frac{1}{11,664,000}$~part of gravity, and that
+therefore a pendulum $10$~metres long is deflected
+through $\frac{1}{11,664,000}$~of $10$~metres, or through $\frac{1}{1,166}$~of
+a millimetre. Now suppose our lake, $200$~kilometres
+in length, runs east and west, and that
+our pendulum is hung up at the middle of the
+lake, $100$~kilometres from either end. In \fig{31}
+let $\Seg{C}{D}$ represent the level of the lake as
+undisturbed, and $\Seg{A}{B}$~an exaggerated pendulum.
+When the tide-generating force displaces the
+pendulum to~$\Seg{A}{B'}$, the surface of the lake must
+assume the position~$\Seg{C'}{D'}$. Now $\Seg{A}{B}$~being $10$~metres,
+$\Seg{B}{B'}$~may range as far as $\frac{1}{1,166}$~of a millimetre;
+and it is obvious that $\Seg{C}{C'}$~must bear the
+\PageSep{184}
+same relation to~$\Seg{C}{B}$ that $\Seg{B}{B'}$ does to~$\Seg{A}{B}$.
+Hence $\Seg{C}{C'}$ at its greatest may be $\frac{1}{11,664,000}$~of half
+the length of the lake. The lake is supposed
+to be twice $100$~kilometres in length, and $100$~kilometres
+is $10$~million centimetres; thus $\Seg{C}{C'}$~is
+$\frac{1}{1.1664}$~centimetre, or $\frac{9}{10}$~of a centimetre. When
+the pendulum is deflected in the other direction
+the lake rocks the other way, and $C'$~is just as
+much above~$C$ as it was below it before. It
+follows from this that the lunar tide at the ends
+of a lake, $200$~kilometres or $120$~miles in length,
+has a range of $1\frac{3}{4}$~centimetres or $\frac{2}{3}$~of an inch.
+The solar tidal force is a little less than half as
+strong as that due to the moon, and when the
+two forces conspire together at the times of
+spring tide, we should find a tide with a range
+of $2\frac{1}{2}$~centimetres.
+\Figure{31}{The Tide in a Lake}{png}
+
+If the same rule were to apply to a lake $2,000$~kilometres
+or $1,200$~miles in length, the range
+of lunar tide would be about $17$~centimetres or
+$7$~inches, and the addition of solar tides would
+bring the range up to $25$~centimetres or $10$~inches.
+\PageSep{185}
+I dare say that, for a lake of such a
+size, this rule would not be very largely in error.
+But as we make the lake longer, the currents
+set up by the tidal forces have not sufficient time
+to produce their full effects before the intensity
+and direction of the tidal forces change. Besides
+this, if the lake were broad from north to south,
+the earth's rotation would have an appreciable
+effect, so that the water which flows from the
+north to the south would be deflected westward,
+and that which flows from south to north would
+tend to flow eastward. The curvature of the
+earth's surface must also begin to affect the
+motion. For these reasons, such a simple rule
+would then no longer suffice for calculating the
+tide.
+
+Mathematicians have not yet succeeded in
+solving the tidal problem for a lake of large
+dimensions, and so it is impossible to describe
+the mode of oscillation. It may, however, be asserted
+that the shape, dimensions, and depth of
+the lake, and the latitudes of its boundaries will
+affect the result. The tides on the northern and
+southern shores will be different, and there will
+be nodal lines, along which there will be no rise
+and fall of the water.
+\index{Lakes!tides in|)}%
+
+The Straits of Gibraltar are so narrow, that
+\index{Mediterranean Sea, tides in}%
+the amount of water which can flow through
+them in the six hours which elapse between
+high and low water in the Atlantic is inconsiderable.
+\PageSep{186}
+\index{Waves!propagated northward in Atlantic|(}%
+Hence the Mediterranean Sea is virtually
+\index{Mediterranean Sea, tides in}%
+a closed lake. The tides of this sea are
+much complicated by the constriction formed
+by the Sicilian and Tunisian promontories. Its
+tides probably more nearly resemble those of two
+lakes than of a single sheet of water. The tides
+of the Mediterranean are, in most places, so inconspicuous
+that it is usually, but incorrectly,
+described as a tideless sea. Every visitor to
+Venice must, however, have seen, or may we say
+smelt, the tides, which at springs have a range of
+some four feet. The considerable range of tide
+at Venice appears to indicate that the Adriatic
+\index{Adriatic, tide in}%
+acts as a resonator for the tidal oscillation, in the
+same way that a hollow vessel, tuned to a particular
+note, picks out and resonates loudly when
+that note is sounded.
+
+We see, then, that whilst the tides of a small
+lake are calculable by the equilibrium theory,
+those of a large one, such as the Mediterranean,
+remain intractable. It is clear, then, that the
+tides of the ocean must present a problem yet
+more complex than those of a large lake.
+
+In the Pacific and Southern oceans the tidal
+\index{Pacific Ocean, tide in, affects Atlantic}%
+forces have almost uninterrupted sway, but the promontories
+of Africa and of South America must
+profoundly affect the progress of the tide wave
+from east to west. The Atlantic Ocean forms a
+\index{Atlantic, tide in|(}%
+great bay in this vaster tract of water. If this
+inlet were closed by a barrier from the Cape of
+\PageSep{187}
+\index{Pacific Ocean, tide in, affects Atlantic}%
+Good Hope to Cape Horn, it would form a lake
+large enough for the generation of much larger
+tides than those of the Mediterranean Sea, although
+probably much smaller than those which
+we actually observe on our coasts. Let us now
+suppose that the tides proper to the Atlantic are
+non-existent, and let us remove the barrier between
+the two capes. Then the great tide wave
+sweeps across the Southern ocean from east to
+west, and, on reaching the tract between Africa
+and South America, generates a wave which
+travels northward up the Atlantic inlet. This
+secondary wave travels ``freely,'' at a rate dependent
+only on the depth of the ocean. The
+energy of the wave motion is concentrated, where
+the channel narrows between North Africa and
+Brazil, and the height of the wave must be augmented
+in that region. Then the energy is
+weakened by spreading, where the sea broadens
+again, and it is again reconcentrated by the projection
+of the North American coast line towards
+Europe. Hence, even in this case, ideally simplified
+as it is by the omission of the direct action
+of the moon and sun, the range of tide would
+differ at every portion of the coasts on each side
+of the Atlantic.
+
+The time of high water at any place must also
+depend on the varying depth of the ocean, for it
+is governed by the time occupied by the ``free
+wave'' in traveling from the southern region to
+\PageSep{188}
+\index{Cotidal chart}%
+the north. But in the south, between the two
+capes of Africa and South America, the tidal
+oscillation is constrained to keep regular time
+with the moon, and so it will keep the same
+rhythm at every place to the northward, at whatever
+variable pace the wave may move. The
+time of high water will of course differ at every
+point, being later as we go northward. The
+wave may indeed occupy so long on its journey,
+that one high water may have only just arrived
+at the northern coast of Africa, when another is
+rounding the Cape of Good Hope.
+
+Under the true conditions of the case, this
+``free'' wave, generated in and propagated from
+the southern ocean, is fused with the true
+``forced'' tide wave generated in the Atlantic itself.
+\index{Atlantic, tide in|)}%
+It may be conjectured that on the coast of
+Europe the latter is of less importance than the
+\index{Europe, tides on coasts of}%
+former. It is interesting to reflect that our tides
+to-day depend even more on what occurred yesterday
+or the day before in the Southern Pacific
+and Indian oceans, than on the direct action of
+the moon to-day. But the relative importance
+of the two causes must remain a matter of conjecture,
+for the problem is one of insoluble complexity.
+
+Some sixty years ago Whewell, and after him
+\index{Whewell!on cotidal charts}%
+Airy, drew charts illustrative of what has just
+\index{Airy, Sir G. B.!cotidal chart}%
+been described. A map showing the march
+of the tide wave is reproduced from Airy's
+\index{Waves!propagated northward in Atlantic|)}%
+\PageSep{189}
+\index{Establishment of port!shown in cotidal chart}%
+``Tides and Waves,'' in~\fig{32}. It claims to
+show, from the observed times of high water at
+the various parts of the earth, how the tide wave
+travels over the oceans. Whewell and Airy were
+\index{Whewell!on cotidal charts}%
+well aware that their map could only be regarded
+as the roughest approximation to reality. Much
+has been learnt since their days, and the then
+incomplete state of knowledge hardly permitted
+them to fully realize how very rough was their
+approximation to the truth. No more recent attempt
+has been made to construct such a map,
+and we must rest satisfied with this one. Even
+if its lines may in places depart pretty widely
+from the truth, it presents features of much interest.
+I do not reproduce the Pacific Ocean,
+because it is left almost blank, from deficiency
+of data. Thus, in that part of the world where
+the tides are most normal, and where the knowledge
+of them would possess the greatest scientific
+interest, we are compelled to admit an almost
+total ignorance.
+
+The lines on the map, \fig{32}, give the Greenwich
+times of high water at full and change of
+moon. They thus purport to represent the successive
+positions of the crest of the tide wave.
+For example, at noon and midnight (XII~o'clock),
+at full and change of moon, the crest
+of the tide wave runs from North Australia to
+Sumatra, thence to Ceylon, whence it bends back
+to the Island of Bourbon, and, passing some hundreds
+\PageSep{190}
+\TallFig[0.875]{32}{Chart of Cotidal Lines}{jpg}
+\PageSep{191}
+\index{Cotidal chart!for diurnal tide hitherto undetermined}%
+of miles south of the Cape of Good Hope,
+trends away towards the Antarctic Ocean. At
+the same moment the previous tide crest has
+traveled up the Atlantic, and is found running
+across from Newfoundland to the Canary
+Islands. A yet earlier crest has reached the
+north of Norway. At this moment it is low
+water from Brazil to the Gold Coast, and again
+at Great Britain.
+
+The successive lines then exhibit the progress
+of the wave from hour to hour, and we see how
+the wave is propagated into the Atlantic. The
+crowding together of lines in places is the graphical
+representation of the retardation of the
+wave, as it runs into shallower water.
+
+But even if this chart were perfectly trustworthy,
+it would only tell us of the progress of
+the ordinary semidiurnal wave, which produces
+high water twice a day. We have, however, seen
+reason to believe that two successive tides should
+not rise to equal heights, and this figure does
+not even profess to give any suggestion as to
+how this inequality is propagated. In other
+words, it is impossible to say whether two successive
+tides of unequal heights tend to become
+more or less unequal, as they run into any of
+the great oceanic inlets. Thus the map affords
+no indication of the law of the propagation of
+the diurnal inequality.
+\index{Diurnal inequality!not shown in cotidal chart}%
+
+This sketch of the difficulties in the solution
+\PageSep{192}
+\index{Cotidal chart!for diurnal tide hitherto undetermined}%
+of the full tidal problem might well lead to despair
+of the possibility of tidal prediction on our
+coasts. I shall, however, show in the next chapter
+how such prediction is possible.
+
+\begin{Authorities}
+For cotidal charts see Whewell, \Title{Phil.\ Trans.\ Roy.\ Soc.}\ 1833,
+or Airy's \Title{Tides and Waves}, ``Encyclopædia Metropolitana.''
+\index{Airy, Sir G. B.!\Title{Tides and Waves}}%
+\end{Authorities}
+\PageSep{193}
+
+
+\Chapter{XI}
+{Harmonic Analysis of the Tide}
+
+\First{It} is not probable that it will ever be possible
+\index{Analysis, harmonic, of tide|(}%
+\index{Harmonic analysis!account of|(}%
+\index{Moon and earth!tide due to ideal, moving in equator}%
+to determine the nature of the oceanic oscillation
+as a whole with any accuracy. It is true that
+we have already some knowledge of the general
+march of the tide wave, and we shall doubtless
+learn more in the future, but this can never suffice
+for accurate prediction of the tide at any
+place.
+
+Although the equilibrium theory is totally
+false as regards its prediction of the time of passage
+and of the height of the tide wave, yet it
+furnishes the stepping-stone leading towards the
+truth, because it is in effect a compendious statement
+of the infinite variety of the tidal force in
+time and place.
+
+I will begin my explanation of the practical
+method of tidal prediction by obliterating the
+sun, and by supposing that the moon revolves in
+an equatorial circle round the earth. In this
+case the equilibrium theory indicates that each
+tide exactly resembles its predecessors and its
+successors for all time, and that the successive
+and simultaneous passages of the moon and of
+\PageSep{194}
+\index{Moon and earth!tide due to ideal, moving in equator}%
+the wave crests across any place follow one
+another at intervals of $12$~hours $25$~minutes. It
+would always be exactly high water under or
+opposite to the moon, and the height of high
+water would be exactly determinate. In actual
+oceans, even although only subject to the action
+of such a single satellite, the motion of the water
+would be so complex that it would be impossible
+to predict the exact height or time of high or
+of low water. But since the tidal forces operate
+in a stereotyped fashion day after day, there will
+be none of that variability which actually occurs
+on the real earth under the actions of the real
+sun and moon, and we may positively assert that
+whatever the water does to-day it will do to-morrow.
+Thus, if at a given place it is high water
+at a definite number of hours after the equatorial
+moon has crossed the meridian to-day, it will be
+so to-morrow at the same number of hours after
+the moon's passage, and the water will rise and
+fall every day to the same height above and below
+the mean sea level. If then we wanted to
+know how the tide would rise and fall in a given
+harbor, we need only watch the motion of the
+sea at that place, for however the water may
+move elsewhere its motion will always produce
+the same result at the port of observation.
+Thus, apart from the effects of wind, we should
+only have to note the tide on any one day
+to be able to predict it for all time. For by a
+\PageSep{195}
+\index{Satellites!tide due to single equatorial}%
+single day of observation it would be easy to
+note how many hours after the moon's passage
+high water occurs, and how many feet it rises
+and falls with reference to some fixed mark on
+the shore. The delay after the moon's passage
+and the amount of rise and fall would differ geographically,
+but at each place there would be two
+definite numbers giving the height of the tide
+and the interval after the moon's passage until
+high water. These two numbers are called the
+tidal constants for the port; they would virtually
+\index{Constants, tidal, explained}%
+contain tidal predictions for all time.
+
+Now if the moon were obliterated, leaving the
+sun alone, and if he also always moved over
+the equator, a similar rule would hold good,
+but exactly $12$~hours would elapse from one
+high water to the next, instead of $12$~hours $25$~minutes
+as in the case of the moon's isolated
+action. Thus two other tidal constants, expressive
+of height and interval, would virtually contain
+tidal prediction for the solar tide for all
+time.
+
+Theory here gives us some power of foreseeing
+the relative importance of the purely lunar
+and of the purely solar tide. The two waves
+due to the sun alone or to the moon alone have
+the same character, but the solar waves follow
+one another a little quicker than the lunar waves,
+and the sun's force is a little less than half the
+moon's force. The close similarity between the
+\PageSep{196}
+\index{Satellites!tide due to single equatorial}%
+actions of the sun and moon makes it safe to conclude
+that the delay of the isolated solar wave
+after the passage of the sun would not differ
+much from the delay of the isolated lunar wave
+after the passage of the moon, and that the
+height of the solar wave would be about half of
+that of the lunar wave. But theory can only be
+trusted far enough to predict a rough proportionality
+of the heights of the two tide waves to their
+respective generating forces, and the approximate
+equality of the intervals of retardation; but the
+height and retardation of the solar wave could
+not be accurately foretold from observation of
+the lunar wave.
+
+When the sun and moon coëxist, but still
+move in equatorial circles, the two waves, which
+we have considered separately, are combined.
+The four tidal constants, two for the moon and
+two for the sun, would contain the prediction of
+the height of water for all time, for it is easy at
+any future moment of time to discover the two
+intervals of time since the moon and since the
+sun have crossed the meridian of the place of
+observation; we should then calculate the height
+of the water above some mark on the shore on
+the supposition that the moon exists alone, and,
+again, on the supposition that the sun exists
+alone, and adding the two results together, should
+obtain the required height of the water at the
+moment in question.
+\PageSep{197}
+
+But the real moon and sun do not move in
+equatorial circles, but in planes which are oblique
+to the earth's equator, and they are therefore
+sometimes to the north and sometimes to the
+south of the equator; they are also sometimes
+nearer and sometimes further from the earth on
+account of the eccentricity of the orbits in which
+they move. Now the mathematician treats this
+complication in the following way: he first considers
+the moon alone and replaces it by a number
+of satellites of various masses, which move
+in various planes. It is a matter of indifference
+that such a system of satellites could not maintain
+the orbits assigned to them if they were allowed
+to go free, but a mysterious being may be
+postulated who compels the satellites to move in
+the assigned orbits. One, and this is the largest
+of these ideal satellites, has nearly the same mass
+as the real moon and moves in a circle over the
+equator; it is in fact the simple isolated moon
+whose action I first considered. Another small
+satellite stands still amongst the stars; others
+move in such orbits that they are always vertically
+overhead in latitude~$45°$; others repel instead
+of attracting; and others move backwards
+amongst the stars. Now all these satellites are
+so arranged as to their masses and their orbits,
+that the sum of their tidal forces is exactly the
+same as those due to the real moon moving in
+her actual orbit.
+\PageSep{198}
+\index{Interval from moon's transit to high water!in case of ideal satellite}%
+
+So far the problem seems to be complicated
+rather than simplified, for we have to consider a
+dozen moons instead of one. The simplification,
+however, arises from the fact that each satellite
+either moves uniformly in an orbit parallel to the
+equator, or else stands still amongst the stars.
+It follows that each of the ideal satellites creates
+a tide in the ocean which is of a simple character,
+and repeats itself day after day in the same
+way as the tide due to an isolated equatorial
+moon. If all but one of these ideal satellites
+were obliterated the observation of the tide for
+a single day would enable us to predict the tide
+for all time; because it would only be necessary
+to note the time of high water after the ideal
+satellite had crossed the meridian, and the height
+\index{Height of tide!due to ideal satellite}%
+of the high water, and then these two data would
+virtually contain a tidal prediction for that tide
+at the place of observation for all future time.
+The interval and height are together a pair of
+``tidal constants'' for the particular satellite in
+question, and refer only to the particular place
+at which the observation is made.
+
+In actuality all the ideal satellites coëxist, and
+the determination of the pair of tidal constants
+appropriate to any one of them has to be made
+by a complex method of analysis, of which I shall
+say more hereafter. For the present it will suffice
+to know that if we could at will annul all
+the ideal satellites except one, and observe its
+\PageSep{199}
+\index{Moon and earth!ideal satellites replacing actual}%
+\index{Partial tides in harmonic method}%
+tide even for a single day, its pair of constants
+could be easily determined. It would then only
+be necessary to choose in succession all the satellites
+\index{Satellites!ideal replacing sun and moon in harmonic analysis}%
+as subjects of observation, and the materials
+for a lunar tide table for all time would be obtained.
+
+The motion of the sun round the earth is analogous
+to that of the moon, and so the sun has
+also to be replaced by a similar series of ideal
+suns, and the partial tide due to each of them
+has to be found. Finally at any harbor some
+twenty pairs of numbers, corresponding to twenty
+ideal moons and suns, give the materials for tidal
+prediction for all time. Theoretically an infinite
+number of ideal bodies is necessary for an absolutely
+perfect representation of the tides, but
+after we have taken some twenty of them, the
+remainder are found to be excessively small in
+mass, and therefore the tides raised by them are
+so minute that they may be safely omitted. This
+method of separating the tide wave into a number
+of partial constituents is called ``harmonic
+analysis.'' It was first suggested, and put into
+practice as a practical treatment of the tidal
+problem, by Sir William Thomson, now Lord
+Kelvin, and it is in extensive use.
+\index{Kelvin, Lord!initiates harmonic analysis}%
+
+In this method the aggregate tide wave is considered
+as the sum of a number of simple waves
+following one another at exactly equal intervals
+of time, and always presenting a constant rise
+\PageSep{200}
+\index{Moon and earth!ideal satellites replacing actual}%
+\index{Prediction of tide!due to ideal satellite}%
+and fall at the place of observation. When the
+time of high water and the height of any one of
+these constituent waves is known on any one
+day, we can predict, with certainty, the height
+of the water, as due to it alone, at any future
+time however distant. The period of time which
+elapses between the passage of one crest and of
+the next is absolutely exact, for it is derived from
+a study of the motions of the moon or sun, and
+is determined to within a thousandth of a second.
+The instant at which any one of the satellites
+\index{Satellites!ideal replacing sun and moon in harmonic analysis}%
+passes the meridian of the place is also
+known with absolute accuracy, but the interval
+after the passage of the satellite up to the high
+water of any one of these constituent waves, and
+the height to which the water will rise are only
+derivable from observation at each port.
+
+Since there are about twenty coëxistent waves
+of sensible magnitude, a long series of observations
+is requisite for disentangling any particular
+wave from among the rest. The series must
+also be so long that the disturbing influence of
+the wind, both on height and time, may be eliminated
+by the taking of averages. It may be
+well to reiterate that each harbor has to be considered
+by itself, and that a separate set of tidal
+constants has to be found for each place. If it
+is only required to predict the tides with moderate
+accuracy some eight partial waves suffice, but
+if high accuracy is to be attained, we have to
+\PageSep{201}
+consider a number of the smaller ones, bringing
+the total up to $20$ or~$25$.
+
+When the observed tidal motions of the sea
+have been analyzed into partial tide waves, they
+are found to fall naturally into three groups,
+which correspond with the dissections of the sun
+\index{Sun!ideal, replacing real sun in harmonic analysis}%
+and moon into the ideal satellites. In the first
+and most important group the crests follow one
+another at intervals of somewhere about $12$~hours;
+these are called the semidiurnal tides.
+In the second group, the waves of which are in
+most places of somewhat less height than those
+of the semidiurnal group, the crests follow one
+another at intervals of somewhere about $24$~hours,
+and they are called diurnal. The tides
+of the third group have a very slow periodicity,
+for their periods are a fortnight, a month, half
+a year, and a year; they are commonly of very
+small height, and have scarcely any practical
+importance; I shall therefore make no further
+reference to them.
+
+Let us now consider the semidiurnal group.
+The most important of these is called ``the principal
+lunar semidiurnal tide.'' It is the tide
+\index{Lunar!tide, principal}%
+\index{Semidiurnal tide!in harmonic method|(}%
+raised by an ideal satellite, which moves in a circle
+round the earth's equator. I began my explanation
+of this method by a somewhat detailed
+consideration of this wave. In this case, the
+wave crests follow one another at intervals of
+$12$~hours $25$~minutes $14\frac{1}{6}$~seconds. The average
+\PageSep{202}
+interval of time between the successive visible
+transits of the moon over the meridian of the
+place of observation is $24$~hours $50$~minutes $28\frac{1}{3}$~seconds;
+and as the invisible transit corresponds
+to a tide as well as the visible one, the interval
+between the successive high waters is the time
+between the successive transits, of which only
+each alternate one is visible.
+
+The tide next in importance is ``the principal
+solar semidiurnal tide.'' This tide bears the
+\index{Solar!principal tide}%
+same relationship to the real sun that the principal
+lunar semidiurnal tide bears to the real moon.
+The crests follow one another at intervals of
+exactly $12$~hours, which is the time from noon to
+midnight and of midnight to noon. The height
+of this partial wave is, at most places, a little less
+than half of that of the principal lunar tide.
+
+The interval between successive lunar tides is
+$25\frac{1}{4}$~minutes longer than that between successive
+solar tides, and as there are two tides a day, the
+lunar tide falls behind the solar tide by $50\frac{1}{2}$~minutes
+a day. If we imagine the two tides to start
+together with simultaneous high waters, then in
+about $7$~days the lunar tide will have fallen about
+$6$~hours behind the solar tide, because $7$~times
+$50\frac{1}{2}$~minutes is $5$~hours $54$~minutes. The period
+from high water to low water of the principal
+solar semidiurnal tide is $6$~hours, being half the
+time between successive high waters. Accordingly,
+when the lunar tide has fallen $6$~hours
+\PageSep{203}
+\index{Spring and neap tides!represented by principal lunar and solar tides}%
+behind the solar tide, the low water of the solar
+tide falls in with the high water of the lunar
+tide. It may facilitate the comprehension of
+this matter to take a numerical example; suppose
+then that the lunar tide rises $4$~feet above
+and falls by the same amount below the mean
+level of the sea, and that the solar tide rises and
+falls $2$~feet above and below the same level;
+then if the two partial waves be started with their
+high waters simultaneous, the joint wave will at
+first rise and fall by $6$~feet. But after $7$~days it
+is low solar tide when it is high lunar tide, and
+so the solar tide is subtracted from the lunar
+tide, and the compound wave has a height of
+$4$~feet less $2$~feet, that is to say, of $2$~feet.
+After nearly another $7$~days, or more exactly
+after $14\frac{1}{2}$~days from the start, the lunar tide has
+lost another $6$~hours, so that it has fallen back
+$12$~hours in all, and the two high waters agree
+together again, and the joint wave has again a
+rise and fall of $6$~feet. When the two high
+waters conspire it is called spring tide, and when
+the low water of the solar tide conspires with the
+high water of the lunar tide, it is called neap
+tide. It thus appears that the principal lunar
+and principal solar semidiurnal tides together
+represent the most prominent feature of the tidal
+oscillation.
+
+The next in importance of the semidiurnal
+waves is called the ``lunar elliptic tide,'' and here
+\PageSep{204}
+\index{Neap and spring tides!represented by principal lunar and solar tides}%
+the crests follow one another at intervals of $12$~hours
+$39$~minutes $30$~seconds. Now the interval
+between the successive principal lunar tides was
+\index{Lunar!elliptic tide}%
+$12$~hours $25$~minutes $14$~seconds; hence, this
+new tide falls behind the principal lunar tide by
+$14\frac{1}{2}$~minutes in each half day. If this tide starts
+so that its high water agrees with that of the
+principal lunar tide, then after $13\frac{3}{4}$~days from
+the start, its hollow falls in with the crest of the
+former, and in $27\frac{1}{2}$~days from the start the two
+crests agree again.
+
+The moon moves round the earth in an ellipse,
+and if to-day it is nearest to the earth, in $13\frac{3}{4}$~days
+it will be furthest, and in $27\frac{1}{2}$~days it will
+be nearest again. The moon must clearly exercise
+a stronger tidal force and create higher
+tides when she is near than when she is far;
+hence every $27\frac{1}{2}$~days the tides must be larger,
+and halfway between they must be smaller.
+But the tide under consideration conspires with
+the principal lunar tide every $27\frac{1}{2}$~days, and,
+accordingly, the joint wave is larger every $27\frac{1}{2}$~days
+and smaller in between. Thus this lunar
+elliptic tide represents the principal effect of the
+\index{Elliptic tide, lunar}%
+elliptic motion of the moon round the earth.
+There are other semidiurnal waves besides the
+three which I have mentioned, but it would
+hardly be in place to consider them further
+here.
+\index{Semidiurnal tide!in harmonic method|)}%
+
+Now turning to the waves of the second kind,
+\PageSep{205}
+\index{Diurnal inequality!in harmonic method}%
+which are diurnal in character, we find three, all
+of great importance. In one of them the high
+waters succeed one another at intervals of $25$~hours
+$49$~minutes $9\frac{1}{2}$~seconds, and of the second
+and third, one has a period of about $4$~minutes
+less than $24$~hours and the other of about $4$~minutes
+greater than the $24$~hours. It would
+hardly be possible to show by general reasoning
+how these three waves arise from the attraction
+of three ideal satellites, and how these satellites
+together are a substitute for the actions of the
+true moon and sun. It must, however, be obvious
+that the oscillation resulting from three coëxistent
+waves will be very complicated.
+
+All the semidiurnal tides result from waves of
+essentially similar character, although some follow
+one another a little more rapidly than others,
+and some are higher and some are lower. An
+accurate cotidal map, illustrating the progress of
+any one of these semidiurnal waves over the
+ocean, would certainly tell all that we care to
+know about the progress of all the other waves
+of the group.
+
+Again, all the diurnal tides arise from waves
+of the same character, but they are quite diverse
+in origin from the semidiurnal waves, and have
+only one high water a day instead of two. A
+complete knowledge of the behavior of semidiurnal
+waves would afford but little insight into
+the behavior of the diurnal waves. At some
+\PageSep{206}
+time in the future the endeavor ought to be
+made to draw a diurnal cotidal chart distinct
+from the semidiurnal one, but our knowledge is
+not yet sufficiently advanced to make the construction
+of such a chart feasible.
+
+\TB
+
+All the waves of which I have spoken thus
+far are generated by the attractions of the sun
+and moon and are therefore called astronomical
+tides, but the sea level is also affected by other
+oscillations arising from other causes.
+
+Most of the places, at which a knowledge of
+the tides is practically important, are situated in
+estuaries and in rivers. Now rain is more prevalent
+\index{Rivers!annual meteorological tide in}%
+at one season than at another, and mountain
+snow melts in summer; hence rivers and
+estuaries are subject to seasonal variability of
+level. In many estuaries this kind of inequality
+may amount to one or two feet, and such a considerable
+change cannot be disregarded in tidal
+prediction. It is represented by inequalities with
+periods of a year and of half a year, which are
+called the annual and semiannual meteorological
+\index{Annual and semi-annual tides}%
+\index{Meteorological!tides}%
+tides.
+
+Then again, at many places, especially in the
+Tropics, there is a regular alternation of day and
+night breezes, the effect of which is to heap up
+% [** TN: "inland", "off-shore" on line breaks in the original; sole instances]
+the water in-shore as long as the wind blows in-land,
+\index{Wind!a cause of meteorological tides}%
+and to lower it when the wind blows off-shore.
+Hence there results a diurnal inequality
+\PageSep{207}
+\index{Estuary, annual meteorological tide in}%
+of sea-level, which is taken into account in tidal
+prediction by means of a ``solar diurnal meteorological
+\index{Meteorological!tides}%
+tide.'' Although these inequalities depend
+entirely on meteorological influences and
+have no astronomical counterpart, yet it is necessary
+to take them into account in tidal prediction.
+
+\TB
+
+But besides their direct astronomical action,
+the sun and moon exercise an influence on the
+sea in a way of which I have not yet spoken.
+We have seen how waves gradually change their
+shape as they progress in a shallow river, so that
+the crests become sharper and the hollows flatter,
+while the advancing slope becomes steeper and
+the receding one less steep. An extreme exaggeration
+of this sort of change of shape was
+found in the bore. Now it is an absolute rule,
+in the harmonic analysis of the tide, that the
+partial waves shall be of the simplest character,
+and shall have a certain standard law of slope
+on each side of their crests. If then any wave
+ceases to present this standard simple form, it is
+necessary to conceive of it as compound, and to
+build it up out of several simple waves. By the
+composition of a simple wave with other simple
+waves of a half, a third, a quarter of the wave
+length, a resultant wave can be built up which
+shall assume any desired form. For a given
+compound wave, there is no alternative of choice,
+\PageSep{208}
+for it can only be built up in one way. The
+analogy with musical notes is here complete, for
+a musical note of any quality is built up from
+a fundamental, together with its octave and
+twelfth, which are called overtones. So also the
+distorted tide wave in a river is regarded as consisting
+of simple fundamental tide, with over-tides
+of half and third length. The periods of
+these over-tides are also one half and one third
+of that of the fundamental wave.
+
+Out in the open ocean, the principal lunar
+semidiurnal tide is a simple wave, but when it
+runs into shallow water at the coast line, and
+still more so in an estuary, it changes its shape.
+\index{Estuary, annual meteorological tide in}%
+The low water lasts longer than the high water,
+and the time which elapses from low water to
+high water is usually shorter than that from
+high water to low water. The wave is in fact
+no longer simple, and this is taken into account
+by considering it to consist of a fundamental
+lunar semidiurnal wave with a period of $12$~hours
+$50$~minutes, of the first over-tide or octave
+with a period of $6$~hours $25$~minutes, of the second
+over-tide or twelfth with a period of $4$~hours
+$17$~minutes, and of the third over-tide or
+double octave with a period of $3$~hours $13$~minutes.
+In estuaries, the first over-tide of the
+lunar semidiurnal tide is often of great importance,
+and even the second is considerable; the
+third is usually very small, and the fourth and
+\PageSep{209}
+higher over-tides are imperceptible. In the same
+way over-tides must be introduced, to represent
+the change of form of the principal solar semidiurnal
+tide. But it is not usually found necessary
+to consider them in the cases of the less
+important partial tides. The octave, the twelfth,
+and the upper octave may be legitimately described
+as tides, because they are due to the
+attractions of the moon and of the sun, although
+they arise indirectly through the distorting influence
+of the shallowness of the water.
+
+\TB
+
+I have said above that about twenty different
+simple waves afford a good representation of the
+tides at any port. Out of these twenty waves,
+some represent the seasonal change of level in
+the water due to unequal rainfall and evaporation
+at different times of the year, and others
+represent the change of shape of the wave due
+to shallowing of the water. Deducting these
+quasi-tides, we are left with about twelve to
+represent the true astronomical tide. It is not
+possible to give an exact estimate of the number
+of partial tides necessary to insure a good representation
+of the aggregate tide wave, because
+the characteristics of the motion are so different
+at various places that partial waves, important
+at one place, are insignificant at others. For
+example, at an oceanic island the tides may be
+more accurately represented as the sum of a
+\PageSep{210}
+dozen simple waves than by two dozen in a tidal
+river.
+
+The method of analyzing a tide into its constituent
+parts, of which I have now given an
+account, is not the only method by which the
+tides may be treated, but as it is the most recent
+and the best way, I shall not consider the older
+methods in detail. The nature of the procedure
+adopted formerly will, however, be indicated in
+\Ref{Chapter}{XIII}.
+
+\begin{Authorities}[Authority]
+G.~H. Darwin, \Title{Harmonic Analysis of Tidal Observations}:
+\index{Analysis, harmonic, of tide|)}%
+\index{Darwin, G. H.!harmonic analysis}%
+\index{Harmonic analysis!account of|)}%
+``Report to British Association.'' Southport,~1883.
+
+An outline of the method is also contained in \Title{Tides}, ``Encyclopædia
+Britannica.''
+\end{Authorities}
+\PageSep{211}
+
+
+\Chapter{XII}
+{Reduction of Tidal Observations}
+
+\First{I have} now to explain the process by which
+\index{Reduction of tidal observations|(}%
+the several partial tides may be disentangled
+from one another.
+
+The tide gauge furnishes a complete tidal record,
+so that measurement of the tide curve gives
+the height of the water at every instant of time
+during the whole period of observation. The
+\index{Observation!reduction of tidal|(}%
+record may be supposed to begin at noon of a
+given day, say of the first of January. The
+longitude of the port of observation is of course
+known, and the Nautical Almanack gives the
+positions of the sun and moon on the day and
+at the hour in question, with perfect accuracy.
+The real moon has now to be replaced by a
+series of ideal satellites, and the rules for the
+substitution are absolutely precise. Accordingly,
+the position in the heavens of each of
+the ideal satellites is known at the moment of
+time at which the observations begin. The
+same is true of the ideal suns which replace the
+actual sun.
+
+I shall now refer to only a single one of the
+ideal moons or suns, for, \textit{mutatis mutandis},
+\PageSep{212}
+what is true of one is true of all. It is easy to
+calculate at what hour of the clock, measured in
+the time of the place of observation, the satellite
+in question will be due south. If the ideal
+satellite under consideration were the one which
+generates the principal lunar semidiurnal tide, it
+would be due south very nearly when the real
+moon is south, and the ideal sun which generates
+the principal solar tide is south exactly at noon.
+But there is no such obvious celestial phenomenon
+associated with the transit of any other of
+the satellites, although it is easy to calculate the
+time of the southing of each of them. We have
+then to discover how many hours elapse after
+the passage of the particular satellite up to the
+high water of its tide wave. The height of
+the wave crest above, and the depression of the
+wave hollow below the mean water mark must
+also be determined. When this problem has
+been solved for all the ideal satellites and suns,
+the tides are said to be reduced, and the reduction
+furnishes the materials for a tide table for
+the place of observation.
+
+The difficulty of finding the time of passage
+and the height of the wave due to any one of
+the satellites arises from the fact that all the
+waves really coëxist, and are not separately
+manifest. The nature of the disentanglement
+may be most easily explained from a special
+case, say for example that of the principal lunar
+\PageSep{213}
+semidiurnal tide, of which the crests follow one
+another at intervals of $12$~hours $25$~minutes $14\frac{1}{6}$~seconds.
+
+Since the waves follow one another at intervals
+of approximately, but not exactly, a half-day, it
+is convenient to manipulate the time scale so as
+\index{Time!lunar}%
+to make them exactly semidiurnal. Accordingly
+we describe $24$~hours $50$~minutes $28\frac{1}{3}$~seconds as
+a lunar day, so that there are exactly two waves
+\index{Lunar!time}%
+following one another in the lunar day.
+
+The tide curve furnishes the height of the
+\index{Curve, tide!partitioned into lunar time}%
+water at every moment of time, but the time
+having been registered by the clock of the tide
+gauge is partitioned into ordinary days and
+hours. It may, however, be partitioned at intervals
+of $24$~hours $50$~minutes $28\frac{1}{3}$~seconds, and
+into the twenty-fourth parts of that period, and
+it will then be divided into lunar days and hours.
+On each lunar day the tide for which we are
+searching presents itself in the same way, so
+that it is always high and low water at the same
+hour of the lunar clock, with exactly two high
+waters and two low waters in the lunar day.
+
+Now the other simple tides are governed by
+other scales of time, so that in a long succession
+of days their high waters and low waters occur
+at every hour of the lunar clock. If then we
+find the average curve of rise and fall of the
+water, when the time is divided into lunar days
+and hours, and if we use for the average a long
+\PageSep{214}
+succession of days, all the other tide waves will
+disappear, and we shall be left with only the
+lunar semidiurnal tide, purified from all the
+others which really coëxist with it.
+
+The numerical process of averaging thus leads
+to the obliteration of all but one of the ideal
+satellites, and this is the foundation of the
+method of harmonic analysis. The average
+lunar tide curve may be looked on as the outcome
+of a single day of observation, when all
+but the selected satellite have been obliterated.
+The height of the average wave, and the interval
+after lunar noon up to high water, are the
+two tidal constants for the lunar semidiurnal
+tide, and they enable us to foretell that tide for
+all future time.
+
+If the tide curve were partitioned into other
+days and hours of appropriate lengths, it would
+be possible by a similar process of averaging to
+single out another of the constituent tide waves,
+and to determine its two tidal constants, which
+contain the elements of prediction with respect
+to it. By continued repetition of operations of
+this kind, all the constituents of practical importance
+can be determined, and recorded numerically
+by means of their pairs of tidal constants.
+
+The possibility of the disentanglement has
+now been demonstrated, but the work of carrying
+out these numerical operations would be
+\PageSep{215}
+\index{Schedule for reducing tidal observations}%
+fearfully laborious. The tide curve would have
+to be partitioned into about a dozen kinds of
+days of various lengths, and the process would
+entail measurements at each of the $24$~hours of
+each sort of day throughout the whole series.
+There are about nine thousand hours in a year,
+and it would need about a hundred thousand
+measurements of the curve to evaluate twelve
+different partial tides; each set of measured
+heights would then have to be treated separately
+to find the several sorts of averages. Work of
+this kind has usually to be done by paid computers,
+and the magnitude of the operation
+would make it financially prohibitive. It is,
+however, fortunately possible to devise abridged
+methods, which bring the work within manageable
+limits.
+
+In order to minimize the number of measurements,
+the tide curve is only measured at each
+of the $24$~exact hours of ordinary time, the
+height at noon being numbered $0$~hr., and that
+at midnight $12$~hrs., and so on up to $24$~hrs.
+After obtaining a set of $24$~measurements for
+each day, the original tide curve is of no further
+use. The number of measurements involved is
+still large, but not prohibitive. It would be
+somewhat too technical, in a book of this kind,
+to explain in detail how the measured heights of
+the water at the exact hours of ordinary time
+may be made to give, with fair approximation,
+\PageSep{216}
+\index{Schedule for reducing tidal observations}%
+the heights at the exact hours of other time
+scales. It may, however, be well to explain that
+this approximate method is based on the fact,
+that each exact hour of any one of the special
+time scales must of necessity fall within half an
+hour of one of the exact hours of ordinary time.
+The height of the water at the nearest ordinary
+hour is then accepted as giving the height at the
+exact hour of the special time. The results, as
+computed in this way, are subjected to a certain
+small correction, which renders the convention
+accurate enough for all practical purposes.
+
+A schedule, serviceable for all time and for
+all places, is prepared which shows the hour of
+ordinary time lying nearest to each successive
+hour of any one of the special times. The successive
+$24$~hourly heights, as measured on the tide
+curve, are entered in this schedule, and when
+the entry is completed the heights are found to
+be arranged in columns, which follow the special
+time scale with a sufficiently good approximation
+to accuracy. A different form of schedule is
+required for each partial tide, and the entry of
+the numbers therein is still enormously laborious,
+although far less so than the re-partitions and
+re-measurements of the tide curve would be.
+
+The operation of sorting the numbers into
+schedules has been carried out in various ways.
+In the work of the Indian Survey, the numbers
+\index{Indian Survey!method of reducing tidal observations}%
+have been re-copied over and over again. In
+\PageSep{217}
+\index{Abacus for reducing tidal observations|(}%
+\index{Darwin, G. H.!tidal abacus|(}%
+\index{Indian Survey!method of reducing tidal observations}%
+the office of the United States Coast Survey use
+\index{United States Coast Survey!method of reducing tidal observations}%
+is made of certain card templates pierced with
+holes. These templates are laid upon the tabulation
+of the measurements of the tide curve,
+and the numbers themselves are visible through
+the holes. On the surface of the template lines
+are drawn from hole to hole, and these lines
+indicate the same grouping of the numbers as
+that given by the Indian schedules. Dr.~Börgen,
+\index{Borgen@Börgen, method of reducing tidal observations}%
+of the Imperial German Marine Observatory
+\index{German method of reducing tidal observations}%
+at Wilhelmshaven, has used sheets of tracing
+paper to attain the same end. The Indian procedure
+is unnecessarily laborious, and the American
+and German plans appear to have some
+disadvantage in the fact that the numbers to be
+added together lie diagonally across the page.
+I am assured by some professional computers
+that diagonal addition is easy to perform correctly;
+nevertheless this appeared to me to be
+so serious a drawback, that I devised another
+plan by which the numbers should be brought
+into vertical columns, without the necessity of
+re-copying them. In my plan each day is treated
+as a unit and is shifted appropriately. It might
+be thought that the results of the grouping
+would be considerably less accurate than in the
+former methods, but in fact there is found to be
+no appreciable loss of accuracy.
+
+I have $74$~narrow writing-tablets of xylonite,
+divided by lines into $24$~compartments; the
+\PageSep{218}
+tablets are furnished with spikes on the under
+side, so that they can be fixed temporarily in any
+position on an ordinary drawing-board. The
+compartments on each strip are provided for the
+entry of the $24$~tidal measurements appertaining
+to each day. Each strip is stamped at its end
+with a number specifying the number of the day
+to which it is appropriated.
+
+The arrangement of these little tablets, so that
+the numbers written on them may fall into columns,
+is indicated by a sheet of paper marked
+with a sort of staircase, which shows where each
+tablet is to be set down, with its spikes piercing
+the guide sheet. When the strips are in place,
+as shown in~\fig{33}, the numbers fall into $48$~columns,
+numbered $0$,~$1$,~\dots~$23$, $0$,~$1$,~\dots~$23$
+twice over. The guide sheet shown in the figure~\figref{33}
+is the one appropriate for the lunar semidiurnal
+tide for the fourth set of $74$~days of
+a year of observation. The upper half of the
+tablets are in position, but the lower ones are
+left unmounted, so as the better to show the
+staircase of marks.
+
+Then I say that the average of all the $74$~numbers
+standing under the two~$0$'s combined
+will give the average height of water at $0$~hr.\
+of lunar time, and the average of the numbers
+under~$1$, that at $1$~hr.\ of lunar time, and so forth.
+Thus, after the strips are pegged out, the computer
+has only to add the numbers in columns in
+%% Plate 2
+\TallFig{33}{Tidal Abacus}{jpg}
+%% Facing page
+%[Blank Page]
+\PageSep{219}
+order to find the averages. There are other
+sheets of paper marked for such other rearrangements
+of the strips that each new setting gives
+one of the required results; thus a single writing
+of the numbers serves for the whole computation.
+It is usual to treat a whole year of
+observations at one time, but the board being
+adapted for taking only $74$~successive days, five
+series of writings are required for $370$~days,
+which is just over a year. The number~$74$ was
+chosen for simultaneous treatment, because $74$~days
+is almost exactly five semilunations, and
+accordingly there will always be five spring tides
+on record at once.
+
+In order to guard the computer against the
+use of the wrong paper with any set of strips,
+the guide sheets for the first set of $74$~days are
+red; for the second they are yellow; for the
+third green; for the fourth blue; for the fifth
+violet, the colors being those of the rainbow.
+
+The preparation of these papers entailed a
+great deal of calculation in the first instance, but
+the tidal computer has merely to peg out the
+tablets in their right places, verifying that the
+numbers stamped on the ends of the strips agree
+with the numbers on the paper. The addition
+of the long columns of figures is certainly laborious,
+but it is a necessary incident of every
+method of reducing tidal observations.
+
+The result of all the methods is that for each
+\PageSep{220}
+partial tide we have a set of $24$~numbers, which
+represent the oscillations of the sea due to the
+isolated action of one of the ideal satellites, during
+the period embraced between two successive
+passages of that satellite to the south of the
+place of observation. The examination of each
+partial tide wave gives its height, and the interval
+of time which elapses after its satellite has
+passed the meridian until it is high water for
+that particular tide. The height and interval
+are the tidal constants for that particular tide, at
+the port of observation.
+\index{Observation!reduction of tidal|)}%
+
+The results of this ``reduction of the observations''
+are contained in some fifteen or twenty
+pairs of tidal constants, and these numbers contain
+a complete record of the behavior of the sea
+at the place in question.
+
+\begin{Authorities}
+G.~H. Darwin, \Title{Harmonic Analysis, \&c.}: ``Report to British Association,''
+1883.
+
+G.~H. Darwin, \Title{On an apparatus for facilitating the reduction of
+\index{Darwin, G. H.!tidal abacus|)}%
+tidal observations}: ``Proceedings of the Royal Society,'' vol.~lii.\
+1892.
+\end{Authorities}
+\index{Abacus for reducing tidal observations|)}%
+\index{Reduction of tidal observations|)}%
+\PageSep{221}
+
+
+\Chapter{XIII}
+{Tide Tables}
+
+\First{A tide table} professes to tell, at a given
+\index{Tables, tide|(}%
+place and on a given day, the time of high and
+low water, together with the height of the rise
+and the depth of the fall of the water, with
+reference to some standard mark on the shore.
+A perfect tide table would tell the height of the
+water at every moment of the day, but such a
+table would be so bulky that it is usual to predict
+only the high and low waters.
+
+There are two kinds of tide table, namely,
+those which give the heights and times of high
+and low water for each successive day of each
+year, and those which predict the high and low
+water only by reference to some conspicuous
+celestial phenomenon. Both sorts of tide table
+refer only to the particular harbor for which they
+are prepared.
+
+The first kind contains definite forecasts for
+each day, and may be called a special tide table.
+Such a table is expensive to calculate, and must
+be published a full year beforehand. Special
+tide tables are published by all civilized countries
+for their most important harbors. I believe that
+\PageSep{222}
+\index{Indian Survey!tide tables}%
+\index{United States Coast Survey!tide tables of}%
+the most extensive publications are those of the
+Indian Government for the Indian Ocean, and
+of the United States Government for the coasts
+of North America. The Indian tables contain
+\index{America, North, tide tables for}%
+predictions for about thirty-seven ports.
+
+The second kind of table, where the tide is
+given by reference to a celestial phenomenon,
+may be described as a general one. It is here
+necessary to refer to the Nautical Almanack for
+the time of occurrence of the celestial phenomenon,
+and a little simple calculation must then be
+made to obtain the prediction. The phenomenon
+to which the tide is usually referred is the passage
+of the moon across the meridian of the place of
+observation, and the table states that high and
+low water will occur so many hours after the
+moon's passage, and that the water will stand at
+such and such a height.
+
+The moon, at her change, is close to the sun
+and crosses the meridian at noon; she would
+then be visible but for the sun's brightness, and
+if she did not turn her dark side towards us.
+She again crosses the meridian invisibly at midnight.
+At full moon she is on the meridian,
+visibly at midnight, and invisibly at noon. At
+waxing half moon she is visibly on the meridian
+at six at night, and at waning half moon at six
+in the morning. The hour of the clock at which
+the moon passes the meridian is therefore in effect
+a statement of her phase. Accordingly the
+\PageSep{223}
+relative position of the sun and moon is directly
+involved in a statement of the tide as corresponding
+to a definite hour of the moon's passage. A
+table founded on the time of the moon's passage
+must therefore involve the principal lunar and
+solar semidiurnal tides.
+
+At places where successive tides differ but little
+from one another, a simple table of this kind
+suffices for rough predictions. The curves marked
+Portsmouth in \fig{34} show graphically the interval
+after the moon's passage, and the height
+of high water at that port, for all the hours of
+the moon's passage. We have seen in \Ref{Chapter}{X}.\
+that the tide in the North Atlantic is principally
+due to a wave propagated from the Southern
+Ocean. Since this wave takes a considerable
+time to travel from the Cape of Good Hope to
+England, the tide here depends, in great measure,
+on that generated in the south at a considerable
+time earlier. It has therefore been found better
+to refer the high water to a transit of the moon
+which occurred before the immediately preceding
+one. The reader will observe that it is noted on
+the upper figure that $28$~hours have been subtracted
+from the Portsmouth intervals; that is
+to say, the intervals on the vertical scale marked
+$6$,~$7$,~$8$ hours are, for Portsmouth, to be interpreted
+as meaning $34$,~$35$,~$36$ hours. These are
+the hours which elapse after any transit of the
+moon up to high water. The horizontal scale is
+\PageSep{224}
+\index{Moon and earth!tidal prediction by reference to transit|(}%
+one of the times of moon's transit and of phases
+of the moon; the vertical scale in the lower figure
+is one of feet, and it shows the height to
+which the water will rise measured from a certain
+mark ashore. These Portsmouth curves do not
+extend beyond 12~o'clock of moon's transit; this
+is because there is hardly any diurnal inequality,
+\index{Diurnal inequality!complicates prediction}%
+and it is not necessary to differentiate the hours
+as either diurnal or nocturnal, the statement being
+equally true of either day or night. Thus
+if the Portsmouth curves had been extended onward
+from $12$~hours to $24$~hours of the clock time
+of the moon's passage, the second halves of the
+curves would have been merely the duplicates of
+the first halves.\footnote
+ {Before the introduction of the harmonic analysis of the tides
+ described in preceding chapters, tidal observations were ``reduced''
+ by the construction of such figures as these, directly from
+ the tidal observations. Every high water was tabulated as appertaining
+ to a particular phase of the moon, both as to its height
+ and as to the interval between the moon's transit and the occurrence
+ of high water. The average of a long series of observations
+ may be represented in the form of curves by such figures
+ as these.}
+
+But the time of the moon's passage leaves her
+angular distance from the equator and her linear
+distance from the earth indeterminate; and since
+the variability of both of these has its influence
+on the tide, corrections are needed which add
+something to or subtract something from the
+tabular values of the interval and height, as dependent
+solely on the time of the moon's passage.
+\PageSep{225}
+\index{Diurnal inequality!complicates prediction}%
+\index{Interval from moon's transit to high water!at Portsmouth and at Aden}%
+The sun also moves in a plane which is oblique
+to the equator, and so similar allowances must be
+made for the distance of the sun from the equator,
+and for the variability in his distance from the
+earth. In order to attain accuracy with a tide
+table of this sort, eight or ten corrections are
+needed, and the use of the table becomes complicated.
+
+It is, however, possible by increasing the number
+of such figures or tables to introduce into
+them many of the corrections referred to; and
+the use of a general tide table then becomes comparatively
+simple. The sun occupies a definite
+position with reference to the equator, and stands
+at a definite distance from the earth on each day
+of the year; also the moon's path amongst the
+stars does not differ very much from the sun's.
+Accordingly a tide table which states the interval
+after the moon's passage to high or low water
+and the height of the water on a given day of
+\index{Height of tide!at Portsmouth and at Aden}%
+the year will directly involve the principal inequalities
+in the tides. As the sun moves slowly
+amongst the stars, a table applicable to a given
+day of the year is nearly correct for a short time
+before and after that date. If, then, a tide table,
+stating the time and height of the water by reference
+to the moon's passage, be computed for
+say every ten days of the year, it will be very
+nearly correct for five days before and for five
+days after the date for which it is calculated.
+\PageSep{226}
+
+The curves marked Aden, March and June, in
+\index{Prediction of tide!example at Aden|(}%
+\fig{34}, show the intervals and heights of tide,
+on the 15th of those months at that port, for all
+the hours of the moon's passage. The curves are
+to be read in the same way as those for Portsmouth,
+but it is here necessary to distinguish the
+hours of the day from those of the night, and
+accordingly the clock times of moon's transit are
+numbered from $0$~hr.\ at noon up to $24$~hrs.\ at
+the next noon. The curves for March differ so
+much from those for June, that the corrections
+would be very large, if the tides were treated at
+Aden by a single pair of average curves as at
+Portsmouth.
+
+The law of the tides, as here shown graphically,
+may also be stated numerically, and the
+use of such a table is easy. The process will be
+best explained by an example, which happens to
+be retrospective instead of prophetic. It will involve
+that part of the complete table (or series of
+curves) for Aden which applies to the 15th of
+March of any year. Let it be required then to
+find the time and height of high water on March~17,
+1889. The Nautical Almanack for that year
+shows that on that day the moon passed the meridian
+of Aden at eleven minutes past noon of
+Aden time, or in astronomical language at $0$~hr.\
+$11$~mins. Now the table, or the figure of intervals,
+shows that if the moon had passed at $0$~hr.,
+or exactly at noon, the interval would have been
+\PageSep{227}
+$8$~hrs.\ $9$~mins., and that if she had passed at $0$~hr.\
+$20$~mins., or 12.20~\PM\ of the day, the interval
+would have been $7$~hrs.\ $59$~mins. But on
+March~17th the moon actually crossed at $0$~hr.\
+\Figure{34}{Curves of Intervals and Heights at Portsmouth
+and at Aden}{png}
+$11$~mins., very nearly halfway between noon and
+$20$~mins.\ past noon. Hence the interval was
+halfway between $8$~hrs.\ $9$~mins.\ and $7$~hrs.\ $59$~mins.,
+so that it was $8$~hrs.\ $4$~mins. Accordingly
+it was high water $8$~hrs.\ $4$~mins.\ after the moon
+\PageSep{228}
+crossed the meridian. But the moon crossed at
+$0$~hr.\ $11$~mins., therefore the high water occurred
+at 8.15~\PM.
+
+Again the table of heights, or the figure, shows
+that on March~15th, if the moon crossed at $0$~hr.\
+$0$~min.\ the high water would be $6.86$~ft.\ above
+a certain mark ashore, and if she crossed at $0$~hr.\
+$20$~mins.\ the height would be $6.92$~ft. But on
+March~17th the moon crossed halfway between
+$0$~hr.\ $0$~min.\ and $0$~hr.\ $20$~mins., and therefore
+the height was halfway between $6.86$~ft.\ and
+$6.92$~ft., that is to say, it was $6.89$~ft., or $6$~ft.\
+$11$~in. We therefore conclude that on March~17,
+1889, the sea at high water rose to $6$~ft.\
+$11$~in., at 8.15~\PM. I have no information as
+to the actual height and time of high water on
+that day, but from the known accuracy of other
+predictions at Aden we may be sure that this
+agrees pretty nearly with actuality. The predictions
+derived from this table are markedly improved
+when a correction, either additive or subtractive,
+is applied, to allow for the elliptic motion
+of the moon round the earth. On this particular
+occasion the moon stood rather nearer the earth
+than the average, and therefore the correction to
+the height is additive; the correction to the time
+also happens to be additive, although it could
+not be foreseen by general reasoning that this
+would be the case. The corrections for March~17,
+1889, are found to add about $2$~mins.\ to the
+\PageSep{229}
+time, bringing it to 8.17~\PM, and nearly two
+inches to the height, bringing it to $7$~ft.\ $1$~in.
+
+This sort of elaborate general tide table has
+been, as yet, but little used. It is expensive to
+calculate, in the first instance, and it would occupy
+two or three pages of a book. The expense
+is, however, incurred once for all, and the table
+is available for all time, provided that the tidal
+observations on which it is based have been good.
+A sea captain arriving off his port of destination
+would not take five minutes to calculate the two
+or three tides he might require to know, and the
+information would often be of the greatest value
+to him.
+
+As things stand at present, a ship sailing to
+most Chinese, Pacific, or Australian ports is only
+furnished with a statement, often subject to considerable
+error, that the high water will occur at
+so many hours after the moon's passage and will
+rise so many feet. The average rise at springs and
+neaps is generally stated, but the law of the variability
+according to the phases of the moon is wanting.
+But this is not the most serious defect in the
+information, for it is frequently noted that the
+tide is ``affected by diurnal inequality,'' and this
+note is really a warning to the navigator that he
+cannot foretell the time of high water within two
+or three hours of time, or the height within several
+feet.
+
+Tables of the kind I have described would
+\PageSep{230}
+\index{Prediction of tide!method of computing|(}%
+banish this extreme vagueness, but they are more
+likely to be of service at ports of second-rate importance
+than at the great centres of trade, because
+at the latter it is worth while to compute
+full special tide tables for each year.
+\index{Tables, tide!method of calculating|(}%
+
+It is unnecessary to comment on the use of
+tables containing predictions for definite days,
+since it merely entails reference to a book, as to
+a railway time table. Such special tables are undoubtedly
+the most convenient, but the number
+of ports which can ever be deemed worthy of the
+great expense incidental to their preparation
+must always be very limited.
+
+\TB
+
+We must now consider the manner in which
+tide tables are calculated. It is supposed that
+careful observations have been made, and that
+the tidal constants, which state the laws governing
+the several partial tides, have been accurately
+determined by harmonic analysis. The analysis
+of tidal observations consists in the dissection of
+the aggregate tide wave into its constituent partial
+waves, and prediction involves the recomposition
+or synthesis of those waves. In the synthetic
+\index{Synthesis of partial tides for prediction|(}%
+process care must be taken that the partial
+waves shall be recompounded in their proper
+relative positions, which are determined by the
+places of the moon and sun at the moment of
+time chosen for the commencement of prediction.
+
+The synthesis of partial waves may be best
+\index{Moon and earth!tidal prediction by reference to transit|)}%
+\index{Prediction of tide!example at Aden|)}%
+\PageSep{231}
+arranged in two stages. It has been shown in
+\Ref{Chapter}{XI}.\ that the partial waves fall naturally
+into three groups, of which the third is practically
+insignificant. The first and second are the
+semidiurnal and diurnal groups. The first process
+is to unite each group into a single wave.
+
+We will first consider the semidiurnal group.
+Let us now, for the moment, banish the tides
+from our minds, and imagine that there are two
+trains of waves traveling simultaneously along a
+straight canal. If either train existed by itself
+every wave would be exactly like all its brethren,
+both in height, length, and period. Now suppose
+that the lengths and periods of the waves
+of the two coëxistent trains do not differ much
+from one another, although their heights may
+differ widely. Then the resultant must be a single
+train of waves of lengths and periods intermediate
+between those of the constituent waves,
+but in one part of the canal the waves will be
+high, where the two sets of crests fall in the
+same place, whilst in another they will be low,
+where the hollow of the smaller constituent wave
+falls in with the crest of the larger. If only one
+part of the canal were visible to us, a train of
+waves would pass before us, whose heights would
+gradually vary, whilst their periods would change
+but little.
+
+In the same way two of the semidiurnal tide
+waves, when united by the addition of their separate
+\PageSep{232}
+displacements from the mean level, form a
+single wave of variable height, with a period still
+semidiurnal, although slightly variable. But
+there is nothing in this process which limits the
+synthesis to two waves, and we may add a third
+and a fourth, finally obtaining a single semidiurnal
+wave, the height of which varies according
+to a very complex law.
+
+A similar synthesis is then applied to the second
+group of waves, so that we have a single
+variable wave of approximately diurnal period.
+The final step consists in the union of the single
+semidiurnal wave with the single diurnal one into
+a resultant wave. When the diurnal wave is
+large, the resultant is found to undergo very
+great variability both in period and height. The
+principal variations in the relative positions of
+the partial tide waves are determined by the
+phases of the moon and by the time of year, and
+there is, corresponding to each arrangement of
+the partial waves, a definite form for the single
+resultant wave. The task of forming a general
+tide table therefore consists in the determination
+of all the possible periods and heights of the resultant
+wave and the tabulation of the heights
+and intervals after the moon's passage of its high
+and low waters.
+
+I supposed formerly that the captain would
+himself calculate the tide he required from the
+general tide table, but such calculation may be
+\PageSep{233}
+\index{Machine, tide-predicting}%
+\index{Predicting machine for tides|(}%
+done beforehand for every day of a specified
+year, and the result will be a special tide table.
+There are about $\DPchg{1400}{1,400}$ high and low waters in
+a year, so that the task is very laborious, and
+has to be repeated each year.
+
+\TB
+
+It is, however, possible to compute a special
+tide table by a different and far less laborious
+method. In this plan an ingenious mechanical
+device replaces the labor of the computer. The
+first suggestion for instrumental prediction of
+tides was made, I think, by Sir William Thomson,
+now Lord Kelvin, in~1872. Mr.~Edward
+\index{Kelvin, Lord!tide predicting machine}%
+Roberts bore an important part in the practical
+\index{Roberts, E., the tide-predicting machine}%
+realization of such a machine, and a tide predicter
+was constructed by Messrs.\ Légé for the
+\index{Lege@Légé, constructor of tide-predicting machine}%
+Indian Government under his supervision. This
+is, as yet, the only complete instrument in existence.
+But others are said to be now in course
+of construction for the Government of the
+United States and for that of France. The
+Indian machine cost so much and works so well,
+that it is a pity it should not be used to the full
+extent of its capacity. The Indian Government
+has, of course, the first claim on it, but the use
+of it is allowed to others on the payment of a
+small fee. I believe that, pending the construction
+of their own machine, the French authorities
+are obtaining the curves for certain tidal
+predictions from the instrument in London.
+\index{Prediction of tide!method of computing|)}%
+\index{Synthesis of partial tides for prediction|)}%
+\PageSep{234}
+
+Although the principle involved in the tide
+predicter is simple, yet the practical realization
+of it is so complex that a picture of the whole
+machine would convey no idea of how it works.
+I shall therefore only illustrate it diagrammatically,
+in~\fig{35}, without any pretension to scale
+or proportion. The reader must at first imagine
+that there are only two pulleys, namely, $A$~and~$B$,
+so that the cord passes from the fixed end~$F$
+under~$A$ and over~$B$, and so onward to the pencil.
+The pulley~$B$ is fixed, and the pulley~$A$ can slide
+vertically up and down in a slot, which is not
+shown in the diagram. If $A$~moves vertically
+through any distance, the pencil must clearly
+move through double that distance, so that
+when $A$~is highest the pencil is lowest, and vice
+versa.
+
+The pencil touches a uniformly revolving
+drum, covered with paper; thus if the pulley~$A$
+executes a simple vertical oscillation, the pencil
+draws a simple wave on the drum. Now the
+pulley is mounted on an inverted T-shaped
+frame, and a pin, fixed in a crank~$C$, engages in
+the slit in the horizontal arm of the T-piece.
+When the crank~$C$ revolves, the pulley~$A$ executes
+a simple vertical oscillation with a range depending
+on the throw of the crank.\footnote
+ {I now notice that the throw of the crank~$C$ is too small to
+ have allowed the pencil to draw so large a wave as that shown
+ on the drum. But as this is a mere diagram, I have not thought
+ it worth while to redraw the whole.}
+The position
+\PageSep{235}
+of the pin is susceptible of adjustment on the
+crank, so that its throw and the range of oscillation
+of the pulley can be set to any required
+\Figure[0.7]{35}{Diagram of Tide-predicting Instrument}{png}
+length---of course within definite limits determined
+by the size of the apparatus.
+
+The drum is connected to the crank~$C$ by a
+train of wheels, so that as the crank rotates the
+drum also turns at some definitely proportional
+rate. If, for example, the crank revolves twice
+for one turn of the drum, the pencil will draw a
+simple wave, with exactly two crests in one circumference
+of the drum. If one revolution of
+the drum represents a day, the graphical time
+scale is $24$~hours to the circumference of the
+\PageSep{236}
+drum. If the throw of the crank be one inch,
+the pulley will oscillate with a total range of two
+inches, and the pencil with a total range of four
+inches. Then taking two inches lengthwise on
+the drum to represent a foot of water, the curve
+drawn by the pencil might be taken to represent
+the principal solar semidiurnal tide, rising one
+foot above and falling one foot below the mean
+sea level.
+
+I will now show how the machine is to be
+adjusted so as to give predictions. We will
+suppose that it is known that, at noon of the
+first day for which prediction is required, the
+solar tide will stand at $1$~ft.\ $9$~in.\ above mean
+sea level and that the water will be rising. Then,
+the semi-range of this tide being one foot, the
+pin is adjusted in the crank at one inch from
+the centre, so as to make the pencil rock through
+a total range of $4$~inches, representing $2$~feet.
+The drum is now turned so as to bring the noon-line
+of its circumference under the pencil, and
+the crank is turned so that the pencil shall be
+$3\frac{1}{2}$~inches (representing $1$~ft.\ $9$~in.\ of water)
+below the middle of the drum, and so that when
+the machine starts, the pencil will begin to descend.
+The curve being drawn upside-down,
+the pencil is set below the middle line because
+the water is to be above mean level, and it must
+begin to descend because the water is to ascend.
+The train of wheels connecting the crank and
+\PageSep{237}
+drum is then thrown into gear, and the machine
+is started; it will then draw the solar tide curve,
+on the scale of $2$~inches to the foot, for all
+time.
+
+If the train of wheels connecting the crank to
+the drum were to make the drum revolve once
+whilst the crank revolves $1.93227$~times, the
+curve would represent a lunar semidiurnal tide.
+The reason of this is that $1.93227$~is the ratio
+of $24$~hours to $12$~h.\ $25$~m.\ $14$~s., that is to say,
+of a day to a lunar half day. We suppose the
+circumference of the drum still to represent an
+ordinary day of $24$~hours, and therefore the
+curve drawn by the pencil will have lunar semidiurnal
+periodicity. In order that these curves
+may give predictions of the future march of that
+tide, the throw of the crank must be set to give the
+correct range and its angular position must give
+the proper height at the moment of time chosen
+for beginning. When these adjustments are
+made the curve will represent that tide for all
+time.
+
+We have now shown that, by means of appropriate
+trains of wheels, the machine can be made
+to predict either the solar or the lunar tide; but
+we have to explain the arrangement for combining
+them. If, still supposing there to be
+only the two pulleys $A$,~$B$, the end~$F$ of the cord
+were moved up or down, its motion would be
+transmitted to the pencil, whether the crank~$C$
+\PageSep{238}
+and pulley~$A$ were in motion, or at rest; but if
+they were in motion, the pencil would add the
+motion of the end of the cord to that of the
+pulley. If then there be added another fixed
+pulley~$B'$, and another movable pulley~$A'$, driven
+by a crank and T-piece (not shown in the diagram),
+the pencil will add together the movements
+of the two pulleys $A$~and~$A'$. There must
+now be two trains of wheels, one connecting $A$
+with the drum and the other for~$A'$. If a single
+revolution of the drum causes the crank~$C$ to
+turn twice, whilst it makes the crank of~$A'$ rotate
+$1.93227$~times, the curve drawn will represent
+the union of the principal solar and lunar semidiurnal
+tides. The trains of wheels requisite for
+transmitting motion from the drum to the two
+cranks in the proper proportions are complicated,
+but it is obviously only a matter of calculation
+to determine the numbers of the teeth in the
+several wheels in the trains. It is true that rigorous
+accuracy is not attainable, but the mechanism
+is made so nearly exact that the error in the
+sum of the two tides would be barely sensible
+even after $\DPchg{3000}{3,000}$~revolutions of the drum. It is
+of course necessary to set the two cranks with
+their proper throws and at their proper angles
+so as to draw a curve which shall, from the noon
+of a given day, correspond to the tide at a given
+place.
+
+It must now be clear that we may add as
+\PageSep{239}
+many more movable pulleys as we like. When
+the motion of each pulley is governed by an
+appropriate train of wheels, the movement of
+the pencil, in as far as it is determined by that
+pulley, corresponds to the tide due to one of our
+ideal satellites. The resultant curve drawn on
+the drum is then the synthesis of all the partial
+tides, and corresponds with the motion of the
+sea.
+
+The instrument of the Indian Government
+unites twenty-four partial tides. In order to
+trace a tide curve, the throws of all the cranks
+are set so as to correspond with the known
+heights of the partial tides, and each crank is set
+at the proper angle to correspond with the moment
+of time chosen for the beginning of the tide
+table. It is not very difficult to set the cranks
+and pins correctly, although close attention is of
+course necessary. The apparatus is then driven
+by the fall of a weight, and the paper is fed
+automatically on to the drum and coiled off on
+to a second drum, with the tide curve drawn on
+it. It is only necessary to see that the paper
+runs on and off smoothly, and to write the date
+from time to time on the paper as it passes, in
+order to save future trouble in the identification
+of the days. It takes about four hours to run
+off the tides for a year.
+
+The Indian Government sends home annually
+the latest revision of the tidal constants for
+\PageSep{240}
+thirty-seven ports in the Indian Ocean. Mr.~Roberts
+sets the machine for each port, so as to
+correspond with noon of a future 1st~of January,
+and then lets it run off a complete tide
+curve for a whole year. The curve is subsequently
+measured for the time and height of
+each high and low water, and the printed tables
+are sold at the moderate price of four rupees.
+The publication is made sufficiently long beforehand
+to render the tables available for future
+voyages. These tide tables are certainly amongst
+the most admirable in the world.
+
+\TB
+
+It is characteristic of England that the machine
+is not, as I believe, used for any of the
+home ports, and only for a few of the colonies.
+The neglect of the English authorities is not,
+however, so unreasonable as it might appear to
+be. The tides at English ports are remarkably
+simple, because the diurnal inequality is practically
+absent. The applicability of the older
+methods of prediction, by means of such curves
+as that for Portsmouth in~\fig{34}, is accordingly
+easy, and the various corrections are well determined.
+The arithmetical processes are therefore
+not very complicated, and ordinary computers
+are capable of preparing the tables with but
+little skilled supervision. Still it is to be regretted
+that this beautiful instrument should not
+be more used for the home and colonial ports.
+\PageSep{241}
+
+The excellent tide tables of the Government
+of the United States have hitherto been prepared
+by the aid of a machine of quite a different
+character, the invention of the late Professor
+Ferrel. This apparatus virtually carries out
+\index{Ferrel, tide-predicting instrument}%
+\index{Predicting machine for tides!Ferrel's}%
+that process of compounding all the waves together
+into a single one, which I have described
+as being done by a computer for the formation
+of a general tide table. It only registers, however,
+the time and height of the maxima and
+minima---the high and low waters. I do not
+think it necessary to describe its principle in
+detail, because it will shortly be superseded by a
+machine like, but not identical with, that of the
+Indian Government.
+
+\begin{Authorities}
+G.~H. Darwin, \Title{On Tidal Prediction}. ``Philosophical Transactions
+of the Royal Society,'' A.~1891, pp.~159--229.
+
+In the example of the use of a general tide table at Aden,
+given in this chapter, the datum from which the height is measured
+is $0.37$~ft.\ higher than that used in the Indian Tide Tables;
+\index{Tables, tide|)}%
+\index{Tables, tide!method of calculating|)}%
+accordingly $4\frac{1}{2}$~inches must be added to the height, in order to
+bring it into accordance with the official table.
+
+Sir William Thomson, \Title{Tidal Instruments}, and the subsequent
+discussion. ``Institute of Civil Engineers,'' vol.~lxv.
+
+William Ferrel, \Title{Description of a Maxima and Minima Tide-predicting
+\index{Machine, tide-predicting}%
+Machine}. ``United States Coast Survey,'' 1883.
+\end{Authorities}
+\index{Predicting machine for tides|)}%
+\PageSep{242}
+
+
+\Chapter[Accuracy of Tidal Prediction]{XIV}
+{The Degree of Accuracy of Tidal Prediction}
+
+\First{The} success of tidal predictions varies much
+\index{Atmospheric pressure!influence on tidal prediction}%
+\index{Prediction of tide!errors in|(}%
+according to the place of observation. They are
+not unfrequently considerably in error in our
+latitude, and throughout those regions called by
+sailors ``the roaring forties.'' The utmost that
+can be expected of a tide table is that it shall
+be correct in calm weather and with a steady
+barometer. But such conditions are practically
+non-existent, and in the North Atlantic the great
+variability in the meteorological elements renders
+tidal prediction somewhat uncertain.
+
+The sea generally stands higher when the
+barometer is low, and lower when the barometer
+is high, an inch of mercury corresponding to
+rather more than a foot of water. The pressure
+of the air on the sea in fact depresses it in those
+places where the barometer is high, and allows it
+to rise where the opposite condition prevails.
+
+Then again a landward wind usually raises the
+\index{Wind!perturbation of, in tidal prediction}%
+sea level, and in estuaries the rise is sometimes
+very great. There is a known instance when the
+Thames at London was raised by five feet in a
+strong gale. Even on the open coast the effect
+\PageSep{243}
+\index{Atmospheric pressure!influence on tidal prediction}%
+of wind is sometimes great. A disastrous example
+\index{Wind!perturbation of, in tidal prediction}%
+of this was afforded on the east coast of England
+in the autumn of~1897, when the conjunction
+of a gale with springtide caused the sea to
+do an enormous amount of damage, by breaking
+embankments and flooding low-lying land.
+
+But sometimes the wind has no apparent effect,
+and we must then suppose that it had been blowing
+previously elsewhere in such a way as to depress
+the water at the point at which we watch it.
+The gale might then only restore the water to its
+normal level, and the two effects might mask one
+another. The length of time during which the
+wind has lasted is clearly an important factor,
+because the currents generated by the wind must
+be more effective in raising or depressing the sea
+level the longer they have lasted.
+
+It does not then seem possible to formulate
+any certain system of allowance for barometric
+pressure and wind. There are, at each harbor,
+certain rules of probability, the application of
+which will generally lead to improvement in the
+prediction; but occasionally such empirical corrections
+will be found to augment the error.
+
+But notwithstanding these perturbations, good
+tide tables are usually of surprising accuracy
+even in northern latitudes; this may be seen
+from the following table showing the results of
+comparisons between prediction and actuality at
+Portsmouth. The importance of the errors in
+\index{Errors in tidal prediction|(}%
+\PageSep{244}
+height depends of course on the range of tide;
+it is therefore well to note that the average ranges
+of tide at springs and neaps are $13$~ft.\ $9$~in.\ and
+$7$~ft.\ $9$~in.\ respectively.
+\begin{table}[hbtp!]
+\caption{Table of Errors in the Prediction of High Water at
+Portsmouth in the Months of January, May, And
+\index{Portsmouth, table of errors in tidal predictions}%
+September, 1897.}
+\[
+\begin{array}{|c|c||c|c|}
+\hline
+\multicolumn{2}{|c||}{\ColHead{Time}} &
+\multicolumn{2}{c|}{\ColHead{Height}} \\
+\hline
+\ColHead[1.2in]{Magnitude of error} &
+\ColHead[0.5in]{Number of cases} &
+\ColHead[1.2in]{Magnitude of error} &
+\ColHead[0.5in]{Number of cases} \\
+\hline
+&& \TEntry{Inches} & \\
+\Z0\mm\text{ to }\Z5\mm& 69 & \Z0\text{ to }\Z6 & 89 \\
+\Z6\mm\text{ to }10\mm & 50 & \Z7\text{ to }12 & 58 \\
+ 11\mm\text{ to }15\mm & 25 & 13\text{ to }18 & 24 \\
+ 16\mm\text{ to }20\mm & 10 & 19\text{ to }24 & \Z6 \\
+ 21\mm\text{ to }25\mm & 11 & \Dash & \Dash \\
+ 26\mm\text{ to }30\mm & \Z7 & \Dash & \Dash \\
+ 31\mm\text{ to }35\mm & \Z4 & \Dash & \Dash \\
+ 52\mm & \Z1 & \Dash & \Dash \\
+\hline
+\Strut\Dash & \llap{$1$}77&\Dash & \llap{$1$}77 \\
+\hline
+\end{array}
+\]
+\end{table}
+\begin{table}[hbtp!]
+\caption{Errors in Height for the Year 1892,
+Excepting Part of July}
+\[
+\begin{array}{|c|c|}
+\hline
+\ColHead{Magnitude of error} & \ColHead{Number of cases} \\
+\hline
+\TEntry{Inches} & \\
+\Z0\text{ to }\Z6& 381 \\
+\Z7\text{ to }12 & 228 \\
+ 13\text{ to }18 & \Z52 \\
+ 19\text{ to }24 & \Z\Z8 \\
+ 31 & \Z\Z1 \\
+\hline
+\Strut\Dash & 670 \\
+\hline
+\end{array}
+\]
+\end{table}
+\PageSep{245}
+
+\begin{Remark}
+\NB---The comparison seems to indicate that these predictions
+might be much improved, because the predicted height is
+nearly always above the observed height, and because the diurnal
+inequality has not been taken into account sufficiently, if at
+all.
+\end{Remark}
+
+In tropical regions the weather is very uniform,
+and in many places the ``meteorological
+tides'' produced by the regularly periodic variations
+of wind and barometric pressure are taken
+into account in tidal predictions.
+
+The apparent irregularity of the tides at Aden
+is so great, that an officer of the Royal Engineers
+has told me that, when he was stationed there
+many years ago, it was commonly believed that
+the strange inequalities of water level were due
+to the wind at distant places. We now know
+that the tide at Aden is in fact marvelously
+regular, although the rule according to which it
+proceeds is very complex. In almost every month
+in the year there are a few successive days when
+there is only one high water and one low water
+in the $24$~hours; and the water often remains
+almost stagnant for three or four hours at a
+time. This apparent irregularity is due to the
+diurnal inequality, which is very great at Aden,
+whereas on the coasts of Europe it is insignificant.
+
+I happen to have a comparison with actuality
+of a few predictions of high water at Aden,
+where the maximum range of the tide is about
+$8$~ft.\ $6$~in. They embrace the periods from March~10
+\index{Errors in tidal prediction|)}%
+\PageSep{246}
+\index{Aden, errors of tidal prediction at}%
+\index{Tables, tide!amount of error in}%
+to April~9, and again from November~12 to
+December~12, 1884. In these two periods there
+were $118$~high waters, but through an accident
+to the tide gauge one high water was not registered.
+On one occasion, when the regular semidiurnal
+sequence of the tide would lead us to
+expect high water, there occurred one of those
+periods of stagnation to which I have referred.
+Thus we are left with $116$~cases of comparison
+between the predicted and actual high waters.
+
+The results are exhibited in the following
+table:---
+\[
+\begin{array}{|c|c||c|c|}
+\hline
+\multicolumn{2}{|c||}{\ColHead{Time}} &
+\multicolumn{2}{c|}{\ColHead{Height}} \\
+\hline
+\ColHead{Magnitude of errors} &
+\ColHead[0.5in]{Number of high waters} &
+\ColHead{Magnitude of errors} &
+\ColHead[0.5in]{Number of high waters} \\
+\hline
+&& \TEntry{Inches} & \\
+\Z0\mm\PadTxt{ and }{to}\Z5\mm & 35 & 0 & 15 \\
+\Z5\mm\PadTxt{ and }{to}10\mm & 32 & 1 & 48 \\
+ 10\mm\PadTxt{ and }{to}15\mm & 19 & 2 & 28 \\
+ 15\mm\PadTxt{ and }{to}20\mm & 19 & 3 & 14 \\
+ 20\mm\PadTxt{ and }{to}25\mm & \Z5 & 4 & 11 \\
+ 26\mm\text{ and }28\mm & \Z2 &
+\multicolumn{1}{c|}{\TEntry{No high water}} & \Z1 \\
+ 33\mm\text{ and }36\mm & \Z2 & \Dash & \Dash \\
+ 56\mm\text{ and }57\mm & \Z2 & \Dash & \Dash \\
+\multicolumn{1}{|c|}{\TEntry{No high water}} & \Z1 & \Dash & \Dash \\
+\hline
+\Strut & 117 && 117 \\
+\hline
+\end{array}
+\]
+
+It would be natural to think that when the
+prediction is erroneous by as much as $57$~minutes,
+it is a very bad one; but I shall show that
+\PageSep{247}
+\index{Tables, tide!amount of error in}%
+this would be to do injustice to the table. On
+several of the occasions comprised in this list
+the water was very nearly stagnant. Now if the
+water only rises about a foot from low to high
+water in the course of four or five hours, it is
+almost impossible to say with accuracy when it
+was highest, and two observers might differ in
+their estimate by half an hour or even by an
+hour.
+
+In the table of comparison there are $11$~cases
+in which the error of time is equal to or greater
+than twenty minutes, and I have examined these
+cases in order to see whether the water was then
+nearly stagnant. A measure of the degree of
+stagnation is afforded by the amount of the rise
+from low water to high water, or of the fall from
+high water to low water. The following table
+gives a classification of the errors of time according
+to the rise or fall:---
+\begin{table}[hp!]
+\caption{Analysis of Errors in Time.}
+\centering
+\begin{tabular}{|c|c|}
+\hline
+\ColHead[1.5in]{Ranges from low water to high water} &
+\ColHead{Errors of time} \\
+\hline
+\Strut
+Nil & \Dash \\
+$6$ in.\ to $8$ in. & $22$, $26$, $28$, $56$, $57$~minutes \\
+$13$ in. & $36$~minutes \\
+$17$ in. & $22$ \Ditto{minutes} \\
+$19$ in. & $33$ \Ditto{minutes} \\
+$2$ ft.\ $10$ in. & $22$ \Ditto{minutes} \\
+$3$ ft.\ $\Z9$ in. & $23$ \Ditto{minutes} \\
+$3$ ft.\ $11$ in. & $20$ \Ditto{minutes} \\
+\hline
+\end{tabular}
+\end{table}
+\PageSep{248}
+
+There are then only three cases when the rise
+of water was considerable, and in the greatest of
+them it was only $3$~ft.\ $11$~in.
+
+If we deduct all the tides in which the range
+between low and high water was equal to or less
+than $19$~inches, we are left with $108$~predictions,
+and in these cases the greatest error in time is
+$23$~mins. In $86$~cases the error is equal to or less
+than a quarter of an hour. This leaves $22$~cases
+where the error was greater than $15$~mins.\ made
+up as follows: $18$~cases with error greater than
+$15$~mins.\ and less than $20$~mins.\ and $3$~cases with
+errors of $20$~mins., $22$~mins., $23$~mins. Thus in
+$106$ out of~$108$ predictions the error of time was
+equal to or less than $20$~minutes.
+
+Two independent measurements of a tide
+curve, for the determination of the time of high
+water, lead to results which frequently differ by
+five minutes, and sometimes by ten minutes. It
+may therefore be claimed that these predictions
+have a very high order of accuracy as regards
+time.
+
+Turning now to the heights, out of $116$~predictions
+the error in the predicted height was
+equal to or less than $2$~inches in $91$~cases, it
+amounted to $3$~inches in $14$~cases, and in the
+remaining $11$~cases it was $4$~inches. It thus appears
+that, as regards the height of the tide also,
+the predictions are of great accuracy. This
+short series of comparisons affords a not unduly
+\PageSep{249}
+favorable example of the remarkable success attainable,
+where tidal observation and prediction
+have been thoroughly carried out at a place
+subject to only slight meteorological disturbance.
+
+If our theory of tides were incorrect, so that
+we imagined that there was a partial tide wave
+of a certain period, whereas in fact such a wave
+has no true counterpart in physical causation,
+the reduction of a year of tidal observation would
+undoubtedly assign some definite small height,
+and some definite retardation of the high water
+after the passage of the corresponding, but
+erroneous, satellite. But when a second series
+of observations is reduced, the two tidal constants
+would show no relationship to their previous
+evaluations. If then reductions carried
+out year after year assign, as they do, fairly
+consistent values to the tidal constants, we may
+feel confident that true physical causation is involved,
+even when the heights of some of the
+constituent tide waves do not exceed an inch
+or two.
+
+Prediction must inevitably fail, unless we have
+lighted on the true causes of the phenomena;
+success is therefore a guarantee of the truth of
+the theory. When we consider that the incessant
+variability of the tidal forces, the complex
+outlines of our coasts, the depth of the sea and
+the earth's rotation are all involved, we should
+\PageSep{250}
+regard good tidal prediction as one of the
+greatest triumphs of the theory of universal
+gravitation.
+
+\begin{Authorities}
+The Portsmouth comparisons were given to the author by the
+Hydrographer of the Admiralty, Admiral Sir W.~J. Wharton.
+
+G.~H. Darwin, \Title{On Tidal Prediction}. ``Philosophical Transactions
+of the Royal Society,'' A.~1891.
+\end{Authorities}
+\index{Prediction of tide!errors in|)}%
+\PageSep{251}
+
+
+\Chapter[Rigidity of the Earth]{XV}
+{Chandler's Nutation---The Rigidity of the
+Earth}
+
+\First{In} the present chapter I have to explain the
+\index{Nutation!Chandler's|(}%
+\index{Variation of latitude|(}%
+origin of a tide of an entirely different character
+from any of those considered hitherto. It may
+fairly be described as a true tide, although it is
+not due to the attraction of either the sun or
+the moon.
+
+We have all spun a top, and have seen it, as
+boys say, go to sleep. At first it nods a little,
+but gradually it settles down to perfect steadiness.
+Now the earth may be likened to a top,
+and it also may either have a nutational or nodding
+motion, or it may spin steadily; it is only
+by observation that we can decide whether it is
+nodding or sound asleep.
+
+The equator must now be defined as a plane
+through the earth's centre at right angles to the
+axis of rotation, and not as a plane fixed with
+reference to the solid earth. The latitude of
+\index{Latitude!periodic variations of|(}%
+any place is the angle\footnote
+ {This angle is technically called the geocentric latitude; the
+ distinction between true and geocentric latitude is immaterial in
+ the present discussion.}
+between the equator and
+\PageSep{252}
+a line drawn from the centre of the earth to the
+place of observation. Now when the earth
+nutates, the axis of rotation shifts, and its
+extremity describes a small circle round the spot
+which is usually described as the pole. The
+equator, being perpendicular to the axis of rotation,
+of course shifts also, and therefore the
+latitude of a place fixed on the solid earth varies.
+During the whole course of the nutation, the
+earth's axis of rotation is always directed towards
+the same point in the heavens, and therefore the
+angle between the celestial pole and the vertical
+or plumb-line at the place of observation must
+oscillate about some mean value; the period of
+the oscillation is that of the earth's nutation.
+This movement is called a ``free'' nutation,
+because it is independent of the action of external
+forces.
+
+There are, besides, other nutations resulting
+from the attractions of the moon and sun on the
+protuberant matter at the equator, and from the
+same cause there is a slow shift in space of the
+earth's axis, called the precession. These movements
+are said to be ``forced,'' because they are
+due to external forces. The measurements of
+the forced nutations and of the precession afford
+the means of determining the period of the free
+nutation, if it should exist. It has thus been
+concluded that if there is any variation in the
+latitude, it should be periodic in $305$~days; but
+\PageSep{253}
+only observation can decide whether there is
+such a variation of latitude or not.
+
+Until recently astronomers were so convinced
+of the sufficiency of this reasoning, that, when
+they made systematic examination of the latitudes
+of many observatories, they always searched
+for an inequality with a period of $305$~days.
+Some thought that they had detected it, but
+when the observations extended over long periods,
+it always seemed to vanish, as though what
+they had observed were due to the inevitable
+errors of observation. At length it occurred
+to Mr.~Chandler to examine the observations
+\index{Chandler, free nutation of earth, and variation of latitude|(}%
+of latitude without any prepossession as to the
+period of the inequality. By the treatment of
+enormous masses of observation, he came to the
+conclusion that there is really such an inequality,
+but that the period is $427$~days instead of $305$~days.
+He also found other inequalities in the
+motion of the axis of rotation, of somewhat
+obscure origin, and of which I have no occasion
+to say more.\footnote
+ {They are perhaps due to the unequal melting of polar
+ ice and unequal rainfall in successive years. These irregular
+ variations in the latitude are such that some astronomers are
+ still skeptical as to the reality of Chandler's nutation, and think
+ that it will perhaps be found to lose its regularly rhythmical
+ character in the future.}
+
+The question then arises as to how the theory
+can be so amended as to justify the extension of
+the period of nutation. It was, I believe, Newcomb,
+\PageSep{254}
+\index{Newcomb, S., theoretical explanation of Chandler's nutation}%
+of the United States Naval Observatory,
+who first suggested that the explanation is to be
+sought in the fact that the axis of rotation is an
+axis of centrifugal repulsion, and that when it
+shifts, the distribution of centrifugal force is
+changed with reference to the solid earth, so
+that the earth is put into a state of stress, to
+which it must yield like any other elastic body.
+The strain or yielding consequent on this stress
+must be such as to produce a slight variability
+in the position of the equatorial protuberance
+with reference to places fixed on the earth.
+Now the period of $305$~days was computed on
+the hypothesis that the position of the equatorial
+protuberance is absolutely invariable, but
+periodic variations of the earth's figure would
+operate so as to lengthen the period of the free
+nutation, to an extent dependent on the average
+elasticity of the whole earth.
+\index{Elasticity of earth}%
+
+Mr.~Chandler's investigation demanded the
+utmost patience and skill in marshaling large
+masses of the most refined astronomical observations.
+His conclusions are not only of the
+greatest importance to astronomy, but they also
+give an indication of the amount by which the
+solid earth is capable of yielding to external
+forces. It would seem that the average stiffness
+of the whole earth must be such that it yields a
+little less than if it were made of steel.\footnote
+ {Mr.~S. S. Hough, p.~338 of the paper referred to in the list
+\index{Hough, S. S.!rigidity of earth}%
+ of authorities at the end of the chapter.}
+But
+\PageSep{255}
+\index{Elasticity of earth}%
+the amount by which the surface yields remains
+unknown, because we are unable to say what
+proportion of the aggregate change is superficial
+and what is deep-seated. It is, however, certain
+that the movements are excessively small, because
+the circle described by the extremity of
+the earth's axis of rotation, about the point on
+the earth which we call the pole, has a radius of
+only fifteen feet.
+
+It is easily intelligible that as the axis of
+rotation shifts in the earth, the oceans will tend
+to swash about, and that a sort of tide will be
+generated. If the displacement of the axis were
+considerable, whole continents would be drowned
+by a gigantic wave, but the movement is so
+small that the swaying of the ocean is very
+feeble. Two investigators have endeavored to
+detect an oceanic tide with a period of $427$~days;
+they are Dr.~Bakhuyzen of Leyden and
+\index{Bakhuyzen on tide due to variation of latitude}%
+Mr.~Christie of the United States Coast Survey.
+\index{Christie, A. S., tide due to variation of latitude}%
+The former considered observations of sea-level
+on the coasts of Holland, the latter those on the
+coasts of the United States; and they both conclude
+that the sea-level undergoes a minute
+variability with a period of about $430$~days. A
+similar investigation is now being prosecuted by
+the Tidal Survey of India, and as the Indian
+tidal observations are amongst the best in the
+world, we may hope for the detection of this
+minute tide in the Indian Ocean also.
+\PageSep{256}
+\index{Earth and moon!rigidity of|(}%
+\index{Rigidity of earth|(}%
+
+The inequality in water level is so slight and
+extends over so long a period that its measurement
+cannot yet be accepted as certain. The
+mean level of the sea is subject to slight irregular
+variations, which are probably due to unequal
+rainfall and unequal melting of polar ice in
+successive years. But whatever be the origin of
+these irregularities they exceed in magnitude the
+one to be measured. The arithmetical processes,
+employed to eliminate the ordinary tides and the
+irregular variability, will always leave behind
+some residual quantities, and therefore the examination
+of a tidal record will always apparently
+yield an inequality of any arbitrary period whatever.
+It is only when several independent determinations
+yield fairly consistent values of the
+magnitude of the rise and fall and of the moment
+of high water, that we can feel confidence
+in the result. Now although the reductions of
+Bakhuyzen and Christie are fairly consistent
+\index{Bakhuyzen on tide due to variation of latitude}%
+\index{Christie, A. S., tide due to variation of latitude}%
+with one another, and with the time and height
+suggested by Chandler's nutation, yet it is by no
+means impossible that accident may have led to
+this agreement. The whole calculation must
+therefore be repeated for several places and at
+several times, before confidence can be attained
+in the detection of this latitudinal tide.
+
+\TB
+
+The prolongation of the period of Chandler's
+nutation from $305$ to $427$~days seems to indicate
+\index{Latitude!periodic variations of|)}%
+\index{Nutation!Chandler's|)}%
+\index{Variation of latitude|)}%
+\PageSep{257}
+that our planet yields to external forces, and we
+naturally desire to learn more on so interesting
+a subject. Up to fifty years ago it was generally
+held that the earth was a globe of molten
+matter covered by a thin crust. The ejection of
+lava from volcanoes and the great increase of
+temperature in mines seemed to present evidence
+in favor of this belief. But the geologists and
+physicists of that time seemed not to have perceived
+that the inference might be false, if great
+pressure is capable of imparting rigidity to matter
+at a very high temperature, because the interior
+of the earth might then be solid although
+very hot. Now it has been proved experimentally
+that rock expands in melting, and a physical
+corollary from this is that when rock is under
+great pressure a higher temperature is needed to
+melt it than when the pressure is removed. The
+pressure inside the earth much exceeds any that
+can be produced in the laboratory, and it is uncertain
+up to what degree of increase of pressure
+the law of the rise of the temperature of
+melting would hold good; but there can be no
+doubt that, in so far as experiments in the laboratory
+can be deemed applicable to the conditions
+prevailing in the interior of the earth, they
+tend to show that the matter there is not improbably
+solid.
+
+But Lord Kelvin reinforces this argument
+\index{Kelvin, Lord!rigidity of earth|(}%
+from another point of view. Rock in the solid
+\index{Chandler, free nutation of earth, and variation of latitude|)}%
+\PageSep{258}
+condition is undoubtedly heavier than when it is
+molten. Now the solidified crust on the surface
+of a molten planet must have been fractured
+many times during the history of the planet,
+and the fragments would sink through the liquid,
+and thus build up a solid nucleus. It will
+be observed that this argument does not repose
+on the rise in the melting temperature of rock
+through pressure, although it is undoubtedly
+reinforced thereby.
+
+Hopkins was, I think, the first to adduce arguments
+\index{Hopkins on rigidity of earth}%
+of weight in favor of the earth's solidity.
+He examined the laws of the precession and
+nutation of a rigid shell inclosing liquid, and
+found that the motion of such a system would
+differ to a marked degree from that of the earth.
+From this he concluded that the interior of the
+earth was not liquid.
+
+Lord Kelvin has pointed out that although
+Hopkins's investigation is by no means complete,
+yet as he was the first to show that the
+motion of the earth as a whole affords indications
+of the condition of the interior, an important
+share in the discovery of the solidity of the
+earth should be assigned to him. Lord Kelvin
+then resumed Hopkins's work, and showed that
+if the liquid interior of the planet were inclosed
+in an unyielding crust, a very slight departure
+from perfect sphericity in the shell would render
+the motion of the system almost identical with
+\PageSep{259}
+\index{Height of tide!reduced by elastic yielding of earth}%
+that of a globe solid from centre to surface,
+although this would not be the case with the
+more rapid nutations. A yet more important
+deficiency in Hopkins's investigation is that he
+\index{Hopkins on rigidity of earth}%
+did not consider that, unless the crust were more
+rigid than the stiffest steel, it would yield to the
+surging of the imprisoned liquid as freely as
+india-rubber; and, besides, that if the crust
+yielded freely, the precession and nutations of
+the whole mass would hardly be distinguishable
+from those of a solid globe. Hopkins's argument,
+as thus amended by Lord Kelvin, leads
+to one of two alternatives: either the globe is
+solid throughout, or else the crust yields with
+nearly the same freedom to external forces as
+though it were liquid.
+
+We have now to show that the latter hypothesis
+is negatived by other considerations. The
+oceanic tides, as we perceive them, consist in a
+motion of the water relatively to the land. Now
+if the solid earth were to yield to the tidal forces
+with the same freedom as the superjacent sea,
+the cause for the relative movement of the sea
+would disappear. And if the solid yielded to
+some extent, the apparent oceanic tide would be
+proportionately diminished. The very existence
+of tides in the sea, therefore, proves at least that
+the land does not yield with perfect freedom.
+
+Lord Kelvin has shown that the oceanic tides,
+on a globe of the same rigidity as that of glass,
+\PageSep{260}
+would only have an apparent range of two fifths
+of those on a perfectly rigid globe; whilst, if
+the rigidity was equal to that of steel, the fraction
+of diminution would be two thirds. I have
+myself extended his argument to the hypothesis
+that the earth may be composed of a viscous
+material, which yields slowly under the application
+of continuous forces, and also to the hypothesis
+of a material which shares the properties
+of viscosity and rigidity, and have been led to
+\index{Rigidity of earth|)}%
+analogous conclusions.
+
+The difficulty of the problem of oceanic tides
+is so great that we cannot say how high the tides
+would be if the earth were absolutely rigid, but
+Lord Kelvin is of opinion that they certainly
+\index{Kelvin, Lord!rigidity of earth|)}%
+would not be twice as great as they are, and he
+concludes that the earth possesses a greater average
+stiffness than that of glass, although perhaps
+not greater than that of steel. It is proper to
+add that the validity of this argument depends
+principally on the observed height of an inequality
+of sea level with a period of a fortnight. This
+is one of the partial tides of the third kind, which
+I described in \Ref{Chapter}{XI}.\ as practically unimportant,
+and did not discuss in detail. The value
+of this inequality in the present argument is due
+to the fact that it is possible to form a much
+closer estimate of its magnitude on a rigid earth
+\index{Earth and moon!rigidity of|)}%
+than in the case of the semidiurnal and diurnal
+tides.
+\PageSep{261}
+\index{Darwin, G. H.!rigidity of earth}%
+\index{Earthquakes!shock perceptible at great distance}%
+
+It may ultimately be possible to derive further
+indications concerning the physical condition of
+the inside of the earth from the science of seismology.
+The tremor of an earthquake has frequently
+been observed instrumentally at an enormous
+distance from its origin; as, for example,
+when the shock of a Japanese earthquake is
+perceived in England.
+
+The vibrations which are transmitted through
+the earth are of two kinds. The first sort of wave
+is one in which the matter through which it passes
+is alternately compressed and dilated; it may be
+described as a wave of compression. In the
+second sort the shape of each minute portion of
+the solid is distorted, but the volume remains
+unchanged, and it may be called a wave of distortion.
+These two vibrations travel at different
+speeds, and the compressional wave outpaces
+the distortional one. Now the first sign of a
+distant earthquake is that the instrumental record
+shows a succession of minute tremors.
+These are supposed to be due to waves of compression,
+and they are succeeded by a much
+more strongly marked disturbance, which, however,
+lasts only a short time. This second phase
+in the instrumental record is supposed to be due
+to the wave of distortion.
+
+If the natures of these two disturbances are
+correctly ascribed to their respective sources, it
+is certain that the matter through which the vibration
+\PageSep{262}
+\index{Darwin, G. H.!rigidity of earth}%
+has passed was solid. For, although a
+compressional wave might be transmitted without
+much loss of intensity, from a solid to a
+liquid and back again to a solid, as would have
+to be the case if the interior of the earth is molten,
+yet this cannot be true of the distortional
+wave. It has been supposed that vibrations due
+to earthquakes pass in a straight line through
+the earth; if then this could be proved, we
+should know with certainty that the earth is
+solid, at least far down towards its centre.
+
+Although there are still some---principally
+amongst the geologists---who believe in the existence
+of liquid matter immediately under the
+solid crust of the earth,\footnote
+ {See the Rev.~Osmond Fisher's \Title{Physics of the Earth's Crust}.}
+\index{Fisher, Osmond, on molten interior of earth}%
+yet the arguments which
+I have sketched appear to most men of science
+conclusive against such belief.
+
+\begin{Authorities}
+Mr.~S.~C. Chandler's investigations are published in the ``Astronomical
+Journal,'' vol.~11 and following volumes. A summary
+is contained in ``Science,'' May~3, 1895.
+
+R.~S. Woodward, \Title{Mechanical Interpretation of the Variations of
+\index{Woodward on variation of latitude}%
+Latitude}, ``Ast.\ Journ.'' vol.~15, May,~1895.
+
+Simon Newcomb, \Title{On the Dynamics of the Earth's Rotation},
+``Monthly Notices of the R.~Astron.\ Soc.,'' vol.~52 (1892),
+p.~336.
+
+S.~S. Hough, \Title{The Rotation of an Elastic Spheroid}, ``Philosoph.\
+\index{Hough, S. S.!Chandler's nutation}%
+Trans.\ of the Royal Society,'' A.~1896, p.~319. He indicates a
+slight oversight on the part of Newcomb.
+
+H.~G. van~de Sande Bakhuyzen, \Title{Ueber die Aenderung der Polhoehe},
+``Astron.\ Nachrichten,'' No.~3261.
+\PageSep{263}
+
+A.~S. Christie, \Title{The Latitude-variation Tide}, ``Phil.\ Soc.\ of
+Washington, Bulletin,'' vol.~12 (1895), p.~103.
+
+Lord Kelvin, in Thomson and Tait's ``Natural Philosophy,''
+on the Rigidity of the Earth; and ``Popular Lectures,'' vol.~3.
+
+G.~H. Darwin, \Title{Bodily Tides of Viscous and Semi-elastic Spheroids,~\&c.},
+``Philosoph.\ Trans.\ of the Royal Society,'' Part.~I.
+1879.
+\end{Authorities}
+\PageSep{264}
+
+
+%[** TN: Footnote mark handled by \Chapter logic]
+\Chapter{XVI}
+{Tidal Friction}
+\footnotetext{A considerable portion of this and of the succeeding chapter
+ appeared as an article in \Title{The Atlantic Monthly} for April,~1898.}
+
+\First{The} fact that the earth, the moon, and the
+\index{Friction of tides|(}%
+planets are all nearly spherical proves that in
+early times they were molten and plastic, and
+assumed their present round shape under the
+influence of gravitation. When the material of
+which any planet is formed was semi-liquid
+through heat, its satellites, or at any rate the
+sun, must have produced tidal oscillations in the
+molten rock, just as the sun and moon now produce
+the tides in our oceans.
+
+Molten rock and molten iron are rather sticky
+or viscous substances, and any movement which
+agitates them must be subject to much friction.
+Even water, which is a very good lubricant, is
+not entirely free from friction, and so our present
+oceanic tides must be influenced by fluid
+friction, although to a far less extent than the
+molten solid just referred to. Now, all moving
+systems which are subject to friction gradually
+come to rest. A train will run a long way when
+the steam is turned off, but it stops at last, and
+\PageSep{265}
+a fly-wheel will continue to spin for only a limited
+time. This general law renders it certain that
+the friction of the tide, whether it consists in the
+swaying of molten lava or of an ocean, must be
+retarding the rotation of the planet, or at any
+rate retarding the motion of the system in some
+way.
+
+It is the friction upon its bearings which brings
+a fly-wheel to rest; but as the earth has no bearings,
+it is not easy to see how the friction of the
+tidal wave, whether corporeal or oceanic, can
+tend to stop its rate of rotation. The result
+must clearly be brought about, in some way, by
+the interaction between the moon and the earth.
+Action and reaction must be equal and opposite,
+and if we are correct in supposing that the friction
+of the tides is retarding the earth's rotation,
+there must be a reaction upon the moon which
+must tend to hurry her onwards. To give a
+homely illustration of the effects of reaction, I
+may recall to mind how a man riding a high
+bicycle, on applying the brake too suddenly, was
+thrown over the handles. The desired action
+was to stop the front wheel, but this could not
+be done without the reaction on the rider, which
+sometimes led to unpleasant consequences.
+
+The general conclusion as to the action and
+reaction due to tidal friction is of so vague a
+character that it is desirable to consider in detail
+how they operate.
+\PageSep{266}
+
+The circle in \fig{36} is supposed to represent
+the undisturbed shape of the planet, which rotates
+in the direction of the curved arrow. A portion
+of the orbit of the satellite is indicated by part
+\Figure{36}{Frictionally retarded Tide}{png}
+of a circle, and the direction of its motion is
+shown by an arrow. I will first suppose that the
+water lying on the planet, or the molten rock of
+which it is formed, is a perfect lubricant devoid
+of friction, and that at the moment represented
+in the figure the satellite is at~$M'$. The fluid will
+then be distorted by the tidal force until it assumes
+the egg-like shape marked by the ellipse,
+projecting on both sides beyond the circle. It
+will, however, be well to observe that if this figure
+represents an ocean, it must be a very deep
+one, far deeper than those which actually exist
+on the earth; for we have seen that it is only in
+deep oceans that the high water stands underneath
+and opposite to the moon; whereas in
+shallow water it is low water where we should
+\PageSep{267}
+naturally expect high water. Accepting the hypothesis
+that the high tide is opposite to the
+moon, and supposing that the liquid is devoid of
+friction, the long axis of the egg is always directed
+straight towards the satellite~$M'$, and the
+liquid maintains a continuous rhythmical movement,
+so that as the planet rotates and the satellite
+revolves, it always maintains the same shape
+and attitude towards the satellite.
+
+But when, as in reality, the liquid is subject to
+friction, it gets belated in its rhythmical rise and
+fall, and the protuberance is carried onward by
+the rotation of the planet beyond its proper
+place. In order to make the same figure serve
+for this condition, I set the satellite backward to~$M$;
+for this amounts to just the same thing, and
+is less confusing than redrawing the protuberance
+in its more advanced position. The planet
+then constantly maintains this shape and attitude
+with regard to the satellite, and the interaction
+between the two will be the same as though the
+planet were solid, but continually altering its
+shape.
+
+We have now to examine what effects must
+follow from the attraction of the satellite on an
+egg-shaped planet, when the two constantly
+maintain the same attitude relatively to each
+other. It will make the matter somewhat easier
+of comprehension if we replace the tidal protuberances
+by two particles of equal masses, one at~$P$,
+\PageSep{268}
+\index{Earth and moon!rotation retarded by tidal friction}%
+\index{Retardation of earth's rotation}%
+and the other at~$P'$. If the masses of these
+particles be properly chosen, so as to represent
+the amount of matter in the protuberances, the
+proposed change will make no material difference
+in the action.
+
+The gravitational attraction of the satellite is
+greater on bodies which are near than on those
+which are far, and accordingly it attracts the
+particle~$P$ more strongly than the particle~$P'$. It
+is obvious from the figure that the attraction on~$P$
+must tend to stop the planet's rotation, whilst
+\index{Rotation!retarded by tidal friction}%
+that on~$P'$ must tend to accelerate it. If a man
+pushes equally on the two pedals of a bicycle,
+the crank has no tendency to turn, and besides
+there are dead points in the revolution where
+pushing and pulling have no effect. So also in
+the astronomical problem, if the two attractions
+were exactly equal, or if the protuberances were
+at a dead point, there would be no resultant effect
+on the rotation of the planet. But it is
+obvious that here the retarding pull is stronger
+than the accelerating one, and that the set of the
+protuberances is such that we have passed the
+dead point. It follows from this that the primary
+effect of fluid friction is to throw the tidal
+protuberance forward, and the secondary effect
+is to retard the planet's rotation.
+
+It has been already remarked that this figure is
+drawn so as to apply only to the case of corporeal
+tides or to those of a very deep ocean. If
+\PageSep{269}
+\index{Moon and earth!retardation of motion by tidal friction}%
+the ocean were shallow and frictionless, it would
+be low water under and opposite to the satellite.
+If then the effect of friction were still to throw
+the protuberances forward, the rotation of the
+planet would be accelerated instead of retarded.
+But in fact the effect of fluid friction in a shallow
+ocean is to throw the protuberances backward,
+and a similar figure, drawn to illustrate such a
+displacement of the tide, would at once make it
+clear that here also tidal friction will lead to the
+retardation of the planet's rotation. Henceforth
+then I shall confine myself to the case illustrated
+by~\fig{36}.
+
+Action and reaction are equal and opposite,
+and if the satellite pulls at the protuberances,
+they pull in return on the satellite. The figure
+shows that the attraction of the protuberance~$P$
+tends in some measure to hurry the satellite onward
+in its orbit, whilst that of~$P'$ tends to retard
+it. But the attraction of~$P$ is stronger than that
+of~$P'$, and therefore the resultant of the two is a
+force tending to carry the satellite forward in the
+direction of the arrow.
+
+If a stone be whirled at the end of an elastic
+string, a retarding force, such as the friction of
+the air, will cause the string to shorten, and an
+accelerating force will make it lengthen. In the
+same way the satellite, being as it were tied to
+the planet by the attraction of gravitation, when
+subjected to an onward force, recedes from the
+\PageSep{270}
+\index{Moon and earth!retardation of motion by tidal friction}%
+planet, and moves in a spiral curve at ever increasing
+distances. The time occupied by the
+satellite in making a circuit round the planet is
+prolonged, and this lengthening of the periodic
+time is not merely due to the lengthening of the
+arc described by it, but also to an actual retardation
+of its velocity. It appears paradoxical that
+the effect of an accelerating force should be a
+retardation, but a consideration of the mode in
+which the force operates will remove the paradox.
+The effect of the tangential accelerating
+force on the satellite is to make it describe an
+increasing spiral curve. Now if the reader will
+draw an exaggerated figure to illustrate part of
+such a spiral orbit, he will perceive that the central
+force, acting directly towards the planet,
+must operate in some measure to retard the velocity
+of the satellite. The central force is very
+great compared with the tangential force due to
+the tidal friction, and therefore a very small
+fraction of the central force may be greater than
+the tangential force. Although in a very slowly
+increasing spiral the fraction of the central force
+productive of retardation is very small, yet it is
+found to be greater than the tangential accelerating
+force, and thus the resultant effect is a
+retardation of the satellite's velocity.
+
+The converse case where a retarding force results
+in increase of velocity will perhaps be more
+intelligible, as being more familiar. A meteorite,
+\PageSep{271}
+rushing through the earth's atmosphere, moves
+faster and faster, because it gains more speed
+from the attraction of gravity than it loses by the
+friction of the air.
+
+Now let us apply these ideas to the case of the
+earth and the moon. A man standing on the
+planet, as it rotates, is carried past places where
+the fluid is deeper and shallower alternately; at
+the deep places he says that it is high tide, and
+at the shallow places that it is low tide. In \fig{36}
+it is high tide when the observer is carried
+past~$P$. Now it was pointed out that when there
+is no fluid friction we must put the moon at~$M'$,
+but when there is friction she must be at~$M$.
+Accordingly, if there is no friction it is high tide
+when the moon is over the observer's head, but
+when there is friction the moon has passed his
+zenith before he reaches high tide. Hence he
+would remark that fluid friction retards the time
+of high tide.
+
+A day is the name for the time in which the
+earth rotates once, and a month for the time in
+which the moon revolves once. Then since tidal
+friction retards the earth's rotation and the
+moon's revolution, we may state that both the
+day and the month are being lengthened, and
+that these results follow from the retardation of
+the time of high tide.
+
+It must also be noted that the spiral in which
+the moon moves is an increasing one, so that her
+\PageSep{272}
+\index{Assyrian records of eclipses}%
+distance from the earth also increases. These
+are absolutely certain and inevitable results of
+the mechanical interaction of the two bodies.
+
+At the present time the rates of increase of
+the day and month are excessively small, so that
+it has not been found possible to determine them
+with any approach to accuracy. It may be well
+to notice in passing that if the rate of either increase
+of element were determinable, that of the
+other would be deducible by calculation.
+
+The extreme slowness of the changes within
+historical times is established by the early records
+in Greek and Assyrian history of eclipses of the
+\index{Eclipses, ancient, and earth's rotation}%
+\index{Greek!records of ancient eclipses}%
+sun, which occurred on certain days and in certain
+places. Notwithstanding the changes in the
+calendar, it is possible to identify the day according
+to our modern reckoning, and the identification
+of the place presents no difficulty.
+Astronomy affords the means of calculating the
+exact time and place of the occurrence of an
+eclipse even three thousand years ago, on the
+supposition that the earth spun at the same rate
+then as now, and that the complex laws governing
+the moon's motion are unchanged.
+
+The particular eclipse referred to in history is
+known, but any considerable change in the
+earth's rotation and in the moon's position would
+have shifted the position of visibility on the
+earth from the situation to which modern computation
+would assign it. Most astronomical
+\PageSep{273}
+observations would be worthless if the exact time
+of the occurrence were uncertain, but in the
+case of eclipses the place of observation affords
+\index{Eclipses, ancient, and earth's rotation}%
+just that element of precision which is otherwise
+wanting. As, then, the situations of the ancient
+eclipses agree fairly well with modern computations,
+we are sure that there has been no great
+change within the last three thousand years,
+either in the earth's rotation or in the moon's
+motion. There is, however, a small outstanding
+discrepancy which indicates that there has been
+some change. But the exact amount of change
+involves elements of uncertainty, because our
+knowledge of the laws of the moon's motion is
+not yet quite accurate enough for the absolutely
+perfect calculation of eclipses which occurred
+many centuries ago. In this way, it is known
+that within historical times the retardation of the
+earth's rotation and the recession of the moon
+have been at any rate very slow.
+
+It does not, however, follow from this that
+the changes have always been equally slow; indeed,
+it may be shown that the efficiency of tidal
+friction increases with great rapidity as we bring
+the tide-generating satellite nearer to the planet.
+
+It has been shown in \Ref{Chapter}{V}.\ that the intensity
+of tide-generating force varies as the inverse
+cube of the distance between the moon and
+the earth, so that if the moon's distance were
+reduced successively to $\frac{1}{2}$,~$\frac{1}{3}$,~$\frac{1}{4}$, of its original distance,
+\PageSep{274}
+the force and the tide generated by it
+would be multiplied $8$,~$27$,~$64$ times. But the
+efficiency of tidal friction increases far more rapidly
+than this, because not only is the tide itself
+augmented, but also the attraction of the moon.
+In order to see how these two factors will coöperate,
+let us begin by supposing that the
+height of the tide remains unaffected by the approach
+or retrogression of the moon. Then the
+same line of argument, which led to the conclusion
+that tide-generating force varies inversely as
+the cube of the distance, shows that the action
+of the moon on protuberances of definite magnitude
+must also vary inversely as the cube of the
+distance. But the height of the tide is not in
+fact a fixed quantity, but varies inversely as the
+cube of the distance, so that when account is
+taken both of the augmentation of the tide and
+of the increased attraction of the moon, it follows
+that the tidal retardation of the earth's rotation
+must vary as the inverse sixth power of
+the distance. Now since the sixth power of~$2$ is~$64$,
+the lunar tidal friction, with the moon at
+half her present distance, would be $64$~times as
+efficient as at present. Similarly, if her distance
+were diminished to a third and a quarter of what
+it is, the tidal friction would act with $729$ and
+$4,096$~times its present strength. Thus, although
+the action may be insensibly slow now, it must
+have gone on with much greater rapidity when
+the moon was nearer to us.
+\PageSep{275}
+
+There are many problems in which it would
+be very difficult to follow the changes according
+to the times of their occurrence, but where it is
+possible to banish time from consideration, and
+to trace the changes themselves, in due order,
+without reference to time. In the sphere of
+common life, we know the succession of stations
+which a train must pass between London and
+Edinburgh, although we may have no time-table.
+This is the case with our astronomical
+problem; for although we have no time-table,
+yet the sequence of the changes in the system
+can be traced accurately.
+
+Let us then banish time, and look forward to
+the ultimate outcome of the tidal interaction of
+the moon and earth. The day and the month
+\index{Day, change in length of, under tidal friction}%
+\index{Month, change in, under tidal friction|(}%
+are lengthening at relative rates which are calculable,
+although the absolute rates in time are
+unknown. It will suffice for a general comprehension
+of the problem to know that the present
+rate of increase of the day is much more rapid
+than that of the month, and that this will hold
+good in the future. Thus, the number of rotations
+of the earth in the interval comprised in
+one revolution of the moon diminishes; or, in
+other words, the number of days in the month
+diminishes, although the month itself is longer
+than at present. For example, when the day
+shall be equal in length to two of our actual
+days, the month may be as long as thirty-seven
+\PageSep{276}
+of our days, and then the earth will spin round
+only about eighteen times in the month.
+
+This gradual change in the day and month
+\index{Day, change in length of, under tidal friction}%
+proceeds continuously until the duration of a
+rotation of the earth is prolonged to fifty-five of
+our present days. At the same time the month,
+or the time of revolution of the moon round the
+earth, will also occupy fifty-five of our days.
+Since the month here means the period of the
+return of the moon to the same place among the
+stars, and since the day is to be estimated in
+the same way, the moon must then always face
+the same part of the earth's surface, and the
+two bodies must move as though they were
+united by a bar. The outcome of the lunar
+tidal friction will therefore be that the moon
+and the earth go round as though locked together,
+in a period of fifty-five of our present
+days, with the day and the month identical in
+length.
+
+Now looking backward in time, we find the
+day and the month shortening, but the day
+changing more rapidly than the month. The
+earth was therefore able to complete more revolutions
+in the month, although that month was
+itself shorter than it is now. We get back in
+fact to a time when there were $29$~rotations of
+the earth in a month instead of~$27\frac{1}{3}$, as at present.
+This epoch is a sort of crisis in the history
+of the moon and the earth, for it may be proved
+\PageSep{277}
+that there never could have been more than $29$~days
+in the month. Earlier than this epoch, the
+days were fewer than~$29$, and later fewer also.
+Although measured in years, this epoch in the
+earth's history must be very remote, yet when we
+contemplate the whole series of changes it must
+be considered as a comparatively recent event.
+In a sense, indeed, we may be said to have passed
+recently through the middle stage of our history.
+
+Now, pursuing the series of changes further
+back than the epoch when there was the maximum
+number of days in the month, we find the
+earth still rotating faster and faster, and the
+moon drawing nearer and nearer to the earth,
+and revolving in shorter and shorter periods.
+But a change has now supervened, so that the
+rate at which the month is shortening is more
+rapid than the rate of change in the day. Consequently,
+the moon now gains, as it were, on
+the earth, which cannot get round so frequently
+in the month as it did before. In other words,
+the number of days in the month declines from
+the maximum of~$29$, and is finally reduced to
+one. When there is only one day in the month,
+\index{Month, change in, under tidal friction|)}%
+the earth and the moon go round at the same
+rate, so that the moon always looks at the same
+side of the earth, and so far as concerns the
+motion they might be fastened together by a
+rigid bar.
+
+This is the same conclusion at which we arrived
+\PageSep{278}
+with respect to the remote future. But
+the two cases differ widely; for whereas in the
+future the period of the common rotation will
+be $55$~of our present days, in the past we find
+the two bodies going round each other in between
+three and five of our present hours. A
+satellite revolving round the earth in so short a
+period must almost touch the earth's surface.
+The system is therefore traced until the moon
+nearly touches the earth, and the two go round
+each other like a single solid body in about three
+to five hours.
+
+The series of changes has been traced forward
+and backward from the present time, but it will
+make the whole process more intelligible, and
+the opportunity will be afforded for certain further
+considerations, if I sketch the history again
+\index{History!of earth and moon|(}%
+in the form of a continuous narrative.
+
+Let us imagine a planet attended by a satellite
+which revolves so as nearly to touch its surface,
+and continuously to face the same side of the
+planet's surface. If now, for some reason, the
+satellite's month comes to differ very slightly
+from the planet's day, the satellite will no longer
+continuously face the same side of the planet,
+but will pass over every part of the planet's
+equator in turn. This is the condition necessary
+for the generation of tidal oscillations in the
+planet, and as the molten lava, of which we
+suppose it to be formed, is a sticky or viscous
+\PageSep{279}
+fluid, the tidal oscillations must be subject to
+friction. Tidal friction will then begin to do its
+work, but the result will be very different according
+as the satellite revolves a little faster or
+a little slower than the planet. If it revolves a
+little faster, so that the month is shorter than
+the day, we have a condition not contemplated
+in~\fig{36}; it is easy to see, however, that as
+the satellite is always leaving the planet behind
+it, the apex of the trial protuberance must be
+directed to a point behind the satellite in its
+orbit. In this case the rotation of the planet
+must be \DPtypo{acclerated}{accelerated} by the tidal friction, and the
+satellite will be drawn inward towards the planet,
+into which it must ultimately fall. In the application
+of this theory to the earth and moon, it
+is obvious that the very existence of the moon
+negatives the hypothesis that the initial month
+was even infinitesimally shorter than the day.
+We must then suppose that the moon revolved
+a little more slowly than the earth rotated. In
+this case the tidal friction would retard the
+earth's rotation, and force the moon to recede
+from the earth, and so perform her orbit more
+slowly. Accordingly, the primitive day and the
+primitive month lengthen, but the month increases
+much more rapidly than the day, so that
+the number of days in a month increases. This
+proceeds until that number reaches a maximum,
+which in the case of our planet is about~$29$.
+\PageSep{280}
+\index{Instability!nature of dynamical, and initial of moon's motion|(}%
+\index{Stability!nature of dynamical}%
+
+After the epoch of the maximum number of
+days in the month, the rate of change in the
+length of the day becomes less rapid than that
+in the length of the month; and although both
+periods increase, the number of days in the
+month begins to diminish. The series of
+changes then proceeds until the two periods
+come again to an identity, when we have the
+earth and the moon as they were at the beginning,
+revolving in the same period, with the
+moon always facing the same side of the earth.
+But in her final condition the moon will be a
+long way off the earth instead of being quite
+close to it.
+
+Although the initial and final states resemble
+each other, yet they differ in one respect which
+is of much importance, for in the initial condition
+the motion is unstable, whilst finally it is
+stable. The meaning of this is, that if the
+moon were even infinitesimally disturbed from
+the initial mode of motion, she would necessarily
+either fall into the planet, or recede therefrom,
+and it would be impossible for her to continue
+to move in that neighborhood. She is unstable
+in the same sense in which an egg when balanced
+on its point is unstable; the smallest mote
+of dust will upset it, and practically it cannot
+stay in that position. But the final condition
+resembles the case of the egg lying on its side,
+which only rocks a little when we disturb it.
+\PageSep{281}
+\index{Stability!nature of dynamical}%
+So if the moon were slightly disturbed from her
+final condition, she would continue to describe
+very nearly the same path round the earth, and
+would not assume some entirely new form of
+orbit.
+
+It is by methods of rigorous argument that
+the moon is traced back to the initial unstable
+condition when she revolved close to the earth.
+But the argument here breaks down, and calculation
+is incompetent to tell us what occurred
+before, and how she attained that unstable mode
+of motion. If we were to find a pendulum
+swinging in a room, where we knew that it had
+been undisturbed for a long time, we might, by
+observing its velocity and allowing for the resistance
+of the air, conclude that at some previous
+moment it had just been upside down, but
+calculation could never tell us how it had
+reached that position. We should of course
+feel confident that some one had started it.
+Now a similar hiatus must occur in the history
+of the moon, but it is not so easy to supply the
+missing episode. It is indeed only possible to
+speculate as to the preceding history.
+
+But there is some basis for our speculation;
+for I say that if a planet, such as the earth,
+made each rotation in three hours, it would very
+nearly fly to pieces. The attraction of gravity
+would be barely strong enough to hold it together,
+just as the cohesive strength of iron is
+\PageSep{282}
+\index{Forced oscillation!due to solar tide, possibly related to birth of moon|(}%
+\index{Moon and earth!origin of}%
+insufficient to hold a fly-wheel together if it is
+spun too fast. There is, of course, an important
+distinction between the case of the ruptured
+fly-wheel and the supposed break-up of the
+earth; for when a fly-wheel breaks, the pieces
+are hurled apart as soon as the force of cohesion
+fails, whereas when a planet breaks up through
+too rapid rotation, gravity must continue to
+hold the pieces together after they have ceased
+to form parts of a single body.
+
+Hence we have grounds for conjecturing that
+the moon is composed of fragments of the primitive
+planet which we now call the earth, which
+detached themselves when the planet spun very
+swiftly, and afterwards became consolidated. It
+surpasses the power of mathematical calculation
+to trace the details of the process of this rupture
+and subsequent consolidation, but we can hardly
+doubt that the system would pass through a
+period of turbulence, before order was reëstablished
+in the formation of a satellite.
+
+I have said above that rapid rotation was probably
+the cause of the birth of the moon, but it
+may perhaps not have been brought about by
+this cause alone. There are certain considerations
+which make it difficult to ascertain the
+initial common period of revolution of the moon
+and the earth with accuracy; it may lie between
+three and five hours. Now I think that such
+a speed might not quite suffice to cause the
+\index{Instability!nature of dynamical, and initial of moon's motion|)}%
+\PageSep{283}
+\index{Moon and earth!origin of}%
+primitive planet to break up. In \Ref{Chapter}{XVIII}.\
+we shall consider in greater detail the conditions
+under which a rotating mass of liquid would
+rupture, but for the present it may suffice to say
+that, where the rotating body is heterogeneous in
+density, like the earth, the exact determination
+of the limiting speed of rotation is not possible.
+Is there, then, any other cause which might coöperate
+with rapid rotation in producing rupture?
+I think there is such a cause, and, although
+we are here dealing with guesswork, I
+will hazard the suggestion.
+
+The primitive planet, before the birth of the
+moon, was rotating rapidly with reference to the
+sun, and it must therefore have been agitated by
+solar tides. In \Ref{Chapter}{IX}.\ it was pointed out
+that there is a general dynamical law which enables
+us to foresee the magnitude of the oscillations
+of a system under the action of external
+forces. That law depended on the natural or
+free period of the oscillation of the system when
+disturbed and left to itself, free from the intervention
+of external forces. We saw that the
+more nearly the periodic forces were timed to
+agree with the free period, the greater was the
+amplitude of the oscillations of the system. Now
+it is easy to calculate the natural or free period
+of the oscillation of a homogeneous liquid globe
+of the same density as the earth, namely, five
+and a half times as heavy as water; the period
+\PageSep{284}
+\index{Sun!possible influence of, in assisting birth of moon}%
+is found to be $1$~hour $34$~minutes. The heterogeneity
+of the earth introduces a complication of
+which we cannot take account, but it seems likely
+that the period would be from $1\frac{1}{2}$ to $2$~hours.
+The period of the solar semidiurnal tide is half a
+\index{Solar!possible effect of tide in assisting birth of moon}%
+day, and if the day were from $3$ to $4$ of our present
+hours the forced period of the tide would
+be in close agreement with the free period of
+oscillation.
+
+May we not then conjecture that as the rotation
+of the primitive earth was gradually reduced
+by solar tidal friction, the period of the solar tide
+was brought into closer and closer agreement
+with the free period, and that consequently the
+solar tide increased more and more in height?
+In this case the oscillation might at length become
+so violent that, in coöperation with the
+rapid rotation, it shook the planet to pieces, and
+that huge fragments were detached which ultimately
+became our moon.
+
+There is nothing to tell us whether this theory
+affords the true explanation of the birth of the
+moon, and I say that it is only a wild speculation,
+incapable of verification.
+
+But the truth or falsity of this speculation
+does not militate against the acceptance of the
+general theory of tidal friction, which, standing
+on the firm basis of mechanical necessity, throws
+much light on the history of the earth and the
+moon, and correlates the lengths of our present
+day and month.
+\index{Forced oscillation!due to solar tide, possibly related to birth of moon|)}%
+\PageSep{285}
+\index{Sun!possible influence of, in assisting birth of moon}%
+
+I have said above that the sequence of events
+has been stated without reference to the scale of
+time. It is, however, of the utmost importance
+\index{Time!requisite for evolution of moon}%
+to gain some idea of the time requisite for all the
+changes in the system. If millions of millions
+of years were necessary, the theory would have
+to be rejected, because it is known from other
+lines of argument that there is not an unlimited
+bank of time on which to draw. The uncertainty
+as to the duration of the solar system is
+\index{Solar!possible effect of tide in assisting birth of moon}%
+wide, yet we are sure that it has not existed for
+an almost infinite past.
+
+Now, although the actual time scale is indeterminate,
+it is possible to find the minimum time
+adequate for the transformation of the moon's
+orbit from its supposed initial condition to its
+present shape. It may be proved, in fact, that
+if tidal friction always operated under the conditions
+most favorable for producing rapid change,
+the sequence of events from the beginning until
+to-day would have occupied a period of between
+$50$ and $60$~millions of years. The actual period,
+of course, must have been much greater. Various
+lines of argument as to the age of the solar
+system have led to results which differ widely
+among themselves, yet I cannot think that the
+applicability of the theory is negatived by the
+magnitude of the period demanded. It may be
+that science will have to reject the theory in its
+full extent, but it seems unlikely that the ultimate
+\PageSep{286}
+\index{Moon and earth!rotation annulled by tidal friction and present libration}%
+verdict will be adverse to the preponderating
+influence of the tide in the evolution of our
+planet.
+
+\TB
+
+If this history be true of the earth and moon,
+\index{History!of earth and moon|)}%
+it should throw light on many peculiarities of the
+solar system. In the first place, a corresponding
+series of changes must have taken place in the
+moon herself. Once on a time the moon must
+have been molten, and the great extinct volcanoes
+revealed by the telescope are evidences of
+her primitive heat. The molten mass must have
+been semi-fluid, and the earth must have raised
+in it enormous tides of molten lava. Doubtless
+the moon once rotated rapidly on her axis, and
+the frictional resistance to her tides must have
+impeded her rotation. This cause must have
+\index{Rotation!of moon annulled by tidal friction}%
+added to the moon's recession from the earth,
+but as the moon's mass is only an eightieth part
+of that of the earth, the effect on the moon's
+orbit must have been small. The only point to
+which we need now pay attention is that the
+rate of her rotation was reduced. She rotated
+then more and more slowly until the tide solidified,
+and thenceforward and to the present day
+she has shown the same face to the earth. Kant
+\index{Kant!rotation of moon}%
+and Laplace in the last century, and Helmholtz
+\index{Helmholtz!on rotation of the moon}%
+\index{Laplace!on rotation of moon}%
+in recent times, have adduced this as the explanation
+of the fact that the moon always shows
+us the same face. Our theory, then, receives a
+\PageSep{287}
+striking confirmation from the moon; for, having
+ceased to rotate relatively to us, she has actually
+advanced to that condition which may be
+foreseen as the fate of the earth.
+
+The earth tide in the moon is now solidified
+so that the moon's equator is not quite circular,
+and the longer axis is directed towards the earth.
+Laplace has considered the action of the earth
+\index{Laplace!on rotation of moon}%
+on this solidified tide, and has shown that the
+moon must rock a little as she moves round the
+earth. In consequence of this rocking motion or
+libration of the moon, and also of the fact that
+her orbit is elliptic, we are able to see just a little
+more than half of the moon's surface.
+
+\TB
+
+Thus far I have referred in only one passage
+to the influence of solar tides, but these are of
+considerable importance, being large enough to
+cause the conspicuous phenomena of spring and
+neap tides. Now, whilst the moon is retarding
+the earth's rotation, the sun is doing so also.
+But these solar tides react only on the earth's
+motion round the sun, leaving the moon's motion
+round the earth unaffected. It might perhaps
+be expected that parallel changes in the
+earth's orbit would have proceeded step by step,
+and that the earth might be traced to an origin
+close to the sun. The earth's mass is less than $\frac{1}{300,000}$~part
+of the sun's, and the reactive effect on the
+earth's orbit round the sun is altogether negligible.
+\PageSep{288}
+It is improbable, in fact, that the year is,
+from this cause at any rate, longer by more than
+a few seconds than it was at the very birth of
+the solar system.
+
+Although the solar tides cannot have had any
+perceptible influence upon the earth's movement
+in its orbit, they will have affected the rotation
+of the earth to a considerable extent. Let us
+imagine ourselves transported to the indefinite
+future, when the moon's orbital period and the
+earth's diurnal period shall both be prolonged to
+$55$~of our present days. The lunar tide in the
+earth will then be unchanging, just as the earth
+tide in the moon is now fixed; but the earth will
+be rotating with reference to the sun, and, if
+there are still oceans on the earth, her rotation
+will be subject to retardation in consequence of
+the solar tidal friction. The day will then become
+longer than the month, whilst the moon
+will at first continue to revolve round the earth
+in $55$~days. Lunar tides will now be again generated,
+but as the motion of the earth will be
+very slow relatively to the moon, the oscillations
+will also be very slow, and subject to little friction.
+But that friction will act in opposition to
+the solar tides, and the earth's rotation will to
+some slight extent be assisted by the moon.
+The moon herself will slowly approach the earth,
+moving with a shorter period, and must ultimately
+fall back into the earth. We know that
+\PageSep{289}
+there are neither oceans nor atmosphere on the
+moon, but if there were such, the moon would
+have been subject to solar tidal friction, and
+would now be rotating slower than she revolves.
+
+%[** TN: Not hyperlinking chapter reference]
+\begin{Authorities}
+See the end of Chapter~XVII.
+\end{Authorities}
+\PageSep{290}
+
+
+\Chapter[Tidal Friction]{XVII}
+{Tidal Friction (Continued)}
+
+\First{It} has been shown in the last chapter that the
+prolongation of the day and of the month under
+the influence of tidal friction takes place in such
+a manner that the month will ultimately become
+longer than the day. Until recent times no case
+had been observed in the solar system in which
+a satellite revolved more rapidly than its planet
+rotated, and this might have been plausibly adduced
+as a reason for rejecting the actual efficiency
+of solar tidal friction in the process of
+celestial evolution. At length however, in~1877,
+Professor Asaph Hall discovered in the system
+\index{Hall, Asaph, discovery of Martian satellites|(}%
+of the planet Mars a case of the kind of motion
+\index{Mars!discovery of satellites|(}%
+which we foresee as the future fate of the moon
+and earth, for he found that the planet was attended
+by two satellites, the nearer of which has
+\index{Satellites!discovery of those of Mars|(}%
+a month shorter than the planet's day. He gives
+an interesting account of what had been conjectured,
+partly in jest and partly in earnest, as to
+the existence of satellites attending that planet.
+This foreshadowing of future discoveries is so
+curious that I quote the following passage from
+Professor Hall's paper. He writes:---
+\PageSep{291}
+
+``Since the discovery of the satellites of Mars,
+the remarkable statements of Dean Swift and
+Voltaire concerning the satellites of this planet,
+and the arguments of Dr.~Thomas Dick and
+others for the existence of such bodies, have attracted
+so much attention, that a brief account
+of the writings on this subject may be interesting.
+
+``The following letter of Kepler was written
+\index{Kepler!argument respecting Martian satellites}%
+to one of his friends soon after the discovery by
+Galileo in~1610 of the four satellites of Jupiter,
+\index{Galileo!discovery of Jupiter's satellites}%
+and when doubts had been expressed as to the
+reality of this discovery. The news of the discovery
+was communicated to him by his friend
+Wachenfels; and Kepler says:---
+
+``\,`Such a fit of wonder seized me at a report
+which seemed to be so very absurd, and I was
+thrown into such agitation at seeing an old dispute
+between us decided in this way, that between
+his joy, my coloring, and the laughter of
+both, confounded as we were by such a novelty,
+we were hardly capable, he of speaking, or I of
+listening. On our parting, I immediately began
+to think how there could be any addition to the
+number of the planets without overturning my
+``Cosmographic Mystery,'' according to which
+Euclid's five regular solids do not allow more
+than six planets round the sun\dots. I am so
+far from disbelieving the existence of the four
+circumjovial planets, that I long for a telescope,
+to anticipate you, if possible, in discovering \emph{two}
+\PageSep{292}
+\index{Kepler!argument respecting Martian satellites}%
+round Mars, as the proportion seems to require,
+\emph{six} or \emph{eight} round Saturn, and perhaps \emph{one} each
+round Mercury and Venus.'
+
+``Dean Swift's statement concerning the satellites
+\index{Swift, satire on mathematicians|(}%
+of Mars is in his famous satire, `The
+Travels of Mr.~Lemuel Gulliver.' After describing
+\index{Gulliver@\Title{Gulliver's Travels}, satire on mathematics|(}%
+his arrival in Laputa, and the devotion
+of the Laputians to mathematics and music,
+Gulliver says:---
+
+``\,`The knowledge I had in mathematics gave
+me great assistance in acquiring their phraseology,
+which depended much upon that science,
+and music; and in the latter I was not unskilled.
+Their ideas were perpetually conversant in lines
+and figures. If they would, for example, praise
+the beauty of a woman, or of any other animal,
+they describe it by rhombs, circles, parallelograms,
+ellipses, and other geometrical terms, or
+by words of art drawn from music, needless here
+to repeat\dots. And although they are dexterous
+enough upon a piece of paper, in the management
+of the rule, the pencil, and the divider,
+yet in the common actions and the behavior of
+life, I have not seen a more clumsy, awkward,
+and unhandy people, nor so slow and perplexed
+in their conceptions upon all subjects, except
+those of mathematics and music. They are very
+bad reasoners, and vehemently given to opposition,
+unless when they happen to be of the right
+opinion, which is seldom their case\dots. These
+\PageSep{293}
+people are under continual disquietudes, never
+enjoying a minute's peace of mind; and their
+disturbances proceed from causes which very
+little affect the rest of mortals. Their apprehensions
+arise from several changes they dread
+in the celestial bodies. For instance, that the
+earth, by the continual approaches of the sun
+towards it, must, in the course of time, be absorbed,
+or swallowed up. That the face of the
+sun will, by degrees, be encrusted with its own
+effluvia, and give no more light to the world.
+That the earth very narrowly escaped a brush
+from the tail of the last comet, which would
+have infallibly reduced it to ashes; and that the
+next, which they have calculated for one-and-thirty
+years hence, will probably destroy us.
+For if, in its perihelion, it should approach
+within a certain degree of the sun (as by their
+calculations they have reason to dread,) it will
+receive a degree of heat ten thousand times
+more intense than that of red-hot glowing iron;
+and, in its absence from the sun, carry a blazing
+tail ten hundred thousand and fourteen miles
+long; through which, if the earth should pass
+at the distance of one hundred thousand miles
+from the nucleus, or main body of the comet, it
+must, in its passage, be set on fire, and reduced
+to ashes. That the sun, daily spending its rays,
+without any nutriment to supply them, will at
+last be wholly consumed and annihilated; which
+\PageSep{294}
+must be attended with the destruction of this
+earth, and of all the planets that receive their
+light from it.
+
+``\,`They are so perpetually alarmed with the
+apprehension of these, and the like impending
+dangers, that they can neither sleep quietly in
+their beds, nor have any relish for the common
+pleasures and amusements of life. When they
+meet an acquaintance in the morning, the first
+question is about the sun's health, how he looked
+at his setting and rising, and what hopes they had
+to avoid the stroke of the approaching comet\dots.
+They spend the greatest part of their lives
+in observing the celestial bodies, which they do
+by the assistance of glasses, far excelling ours in
+goodness. For although their largest telescopes
+do not exceed three feet, they magnify much
+more than those of a hundred with us, and show
+the stars with greater clearness. This advantage
+has enabled them to extend their discoveries
+much further than our astronomers in Europe;
+for they have made a catalogue of ten thousand
+fixed stars, whereas the largest of ours do not
+contain above one-third of that number\dots.
+They have likewise discovered two lesser stars,
+or satellites, which revolve about Mars; whereof
+the innermost is distant from the centre of the
+primary planet exactly three of his diameters,
+and the outermost, five; the former revolves in
+the space of ten hours, and the latter in twenty-one
+\PageSep{295}
+and a half; so that the squares of their
+periodical times are very near in the same proportion
+with the cubes of their distance from
+the centre of Mars; which evidently shows them
+to be governed by the same law of gravitation
+that influences the other heavenly bodies.'
+
+``The reference which Voltaire makes to the
+\index{Voltaire, satire on mathematicians, and Martian satellites}%
+moons of Mars is in his `Micromegas, Histoire
+Philosophique.' Micromegas was an inhabitant
+of Sirius, who, having written a book which a
+suspicious old man thought smelt of heresy, left
+Sirius and visited our solar system. Voltaire
+says:---
+
+``\,`Mais revenons à nos voyageurs. En sortant
+de Jupiter, ils traversèrent un espace d'environ
+cent millions de lieues, et ils côtoyèrent
+la planète de Mars, qui, comme on sait, est cinq
+fois plus petite que noire petit globe; ils virent
+deux lunes qui servent à cette planète, et qui ont
+échappé aux regards de nos astronomes. Je sais
+bien que le père \emph{Castel} écrira, et même plaisamment,
+\index{Castel, Father, ridiculed by Voltaire}%
+contre l'existence de ces deux lunes; mais
+je m'en rapporte à ceux qui raisonnent par analogie.
+Ces bons philosophes-là savent combien il
+serait difficile que Mars, qui est si loin du soleil,
+se passât à moins de deux lunes.'
+
+``The argument by analogy for the existence
+of a satellite of Mars was revived by writers like
+Dr.~Thomas Dick, Dr.~Lardner, and others. In
+\index{Dick, argument as to Martian satellites}%
+\index{Lardner, possibility of Martian satellites}%
+addition to what may be called the analogies of
+\index{Gulliver@\Title{Gulliver's Travels}, satire on mathematics|)}%
+\index{Swift, satire on mathematicians|)}%
+\PageSep{296}
+astronomy, these writers appear to rest on the
+idea that a beneficent Creator would not place
+a planet so far from the sun as Mars without
+giving it a satellite. This kind of argument has
+passed into some of our handbooks of astronomy,
+and is stated as follows by Mr.~Chambers
+\index{Chambers on possible existence of Martian satellites}%
+in his excellent book on `Descriptive Astronomy,'
+2d~edition, p.~89, published in~1867:---
+
+``\,`As far as we know, Mars possesses no satellite,
+though analogy does not forbid, but rather,
+on the contrary, infers the existence of one; and
+its never having been seen, in this case at least,
+proves nothing. The second satellite of Jupiter
+is only $\frac{1}{43}$~of the diameter of the primary, and
+a satellite $\frac{1}{43}$~of the diameter of Mars would
+be less than $100$~miles in diameter, and therefore
+of a size barely within the reach of our largest
+telescopes, allowing nothing for its possibly close
+proximity to the planet. The fact that one of
+the satellites of Saturn was only discovered a
+few years ago renders the discovery of a satellite
+of Mars by no means so great an improbability
+as might be imagined.'
+
+``Swift seems to have had a hearty contempt
+for mathematicians and astronomers, which he
+has expressed in his description of the inhabitants
+of Laputa. Voltaire shared this contempt,
+\index{Voltaire, satire on mathematicians, and Martian satellites}%
+and delighted in making fun of the philosophers
+whom Frederick the Great collected at Berlin.
+The `père Castel' may have been le~père Louis
+\index{Castel, Father, ridiculed by Voltaire}%
+\PageSep{297}
+Castel, who published books on physics and
+mathematics at Paris in 1743 and~1758. The
+probable origin of these speculations about the
+moons of Mars was, I think, Kepler's analogies.
+Astronomers failing to verify these, an opportunity
+was afforded to satirists like Swift and
+Voltaire to ridicule such arguments.''\footnote
+ {\Title{Observations and Orbits of the Satellites of Mars}, by Asaph
+ Hall. Washington, Government Printing Office, 1878.}
+
+As I have already said, these prognostications
+were at length verified by Professor Asaph Hall
+in the discovery of two satellites, which he named
+Phobos and Deimos---Fear and Panic, the dogs
+\index{Deimos, a satellite of Mars}%
+\index{Phobos, a satellite of Mars}%
+of war. The period of Deimos is about $30$~hours,
+and that of Phobos somewhat less than $8$~hours,
+whilst the Martian day is of nearly the same
+length as our own. The month of the inner
+minute satellite is thus less than a third of the
+planet's day; it rises to the Martians in the west,
+and passes through all its phases in a few hours;
+sometimes it must even rise twice in a single
+Martian night. As we here find an illustration
+of the condition foreseen for the earth and moon,
+it seems legitimate to suppose that solar tidal
+friction has retarded the planet's rotation until it
+has become slower than the revolution of one of
+the satellites. It would seem as if the ultimate
+fate of Phobos will be absorption in the planet.
+
+Several of the satellites of Jupiter and of Saturn
+present faint inequalities of coloring, and
+\PageSep{298}
+\index{Jupiter!satellites constantly face planet}%
+\index{Saturn!satellites always face the planet}%
+telescopic examination has led astronomers to believe
+that they always present the same face to
+their planets. The theory of tidal friction would
+\index{Planets!rotation of some, annulled by tidal friction}%
+certainly lead us to expect that these enormous
+planets should work out the same result for their
+relatively small satellites that the earth has produced
+\index{Satellites!discovery of those of Mars|)}%
+\index{Satellites!rotation of those of Jupiter and Saturn annulled}%
+in the moon.
+
+The proximity of the planets Mercury and
+\index{Mercury, rotation of}%
+Venus to the sun should obviously render solar
+\index{Venus, rotation of}%
+tidal friction far more effective than with us.
+The determination of the periods of rotation of
+\index{Rotation!of Mercury, Venus, and satellites of Jupiter and Saturn annulled by tidal friction}%
+these planets thus becomes a matter of much interest.
+But the markings on their disks are so
+obscure that the rates of their rotations have remained
+under discussion for many years. Until
+recently the prevailing opinion was that in both
+cases the day was of nearly the same length as
+ours; but a few years ago Schiaparelli of Milan,
+\index{Schiaparelli on rotation of Venus and Mercury}%
+an observer endowed with extraordinary acuteness
+of vision, announced as the result of his observations
+that both Mercury and Venus rotate
+only once in their respective years, and that
+each of them constantly presents the same face
+to the sun. These conclusions have recently been
+confirmed by Mr.~Percival Lowell from observations
+\index{Lowell, P., on rotations of Venus and Mercury}%
+made in Arizona. Although on reading
+the papers of these astronomers it is not easy
+to see how they can be mistaken, yet it should
+be noted that others have failed to detect the
+markings on the planet's disks, although they
+\index{Hall, Asaph, discovery of Martian satellites|)}%
+\index{Mars!discovery of satellites|)}%
+\PageSep{299}
+\index{Lowell, P., on rotations of Venus and Mercury}%
+apparently enjoyed equal advantages for observation.\footnote
+ {Dr.~See, a member of the staff of the Flagstaff Observatory,
+ Arizona, tells me that he has occasionally looked at these planets
+ through the telescope, although he took no part in the systematic
+ observation. In his opinion it would be impossible for any one
+ at Flagstaff to doubt the reality of the markings. There are,
+ however, many astronomers of eminence who suspend their
+ judgment, and await confirmation by other observers at other
+ stations.}
+
+If, as I am disposed to do, we accept these observations
+as sound, we find that evidence favorable
+to the theory of tidal friction is furnished
+by the planets Mercury and Venus, and by the
+\index{Mercury, rotation of}%
+\index{Venus, rotation of}%
+satellites of the earth, Jupiter and Saturn, whilst
+\index{Earth and moon!figure of}%
+\index{Earth and moon!adjustment of figure to suit change of rotation|(}%
+the Martian system is yet more striking as an
+instance of an advanced stage in evolution.
+
+\TB
+
+It is well known that the figure of the earth
+is flattened by the diurnal rotation, so that the
+polar axis is shorter than any equatorial diameter.
+At the present time the excess of the equatorial
+radius over the polar radius is $\frac{1}{290}$~part of
+either of them. Now in tracing the history of
+the earth and moon, we found that the earth's
+rotation had been retarded, so that the day is
+now longer than it was. If then the solid earth
+has always been absolutely unyielding, and if an
+ocean formerly covered the planet to a uniform
+depth, the sea must have gradually retreated
+towards the poles, leaving the dry land exposed
+at the equator. If on the other hand the solid
+\PageSep{300}
+\index{Geological evidence of earth's plasticity}%
+\index{Plasticity of earth under change of rotation|(}%
+earth had formerly its present shape, there must
+then have been polar continents and a deep equatorial
+sea.
+
+But any considerable change in the speed of
+the earth's rotation would, through the action of
+gravity, bring enormous forces to bear on the
+solid earth. These forces are such as would, if
+they acted on a plastic material, tend to restore
+the planet's figure to the form appropriate to its
+changed rotation. It has been shown experimentally
+by M.~Tresca and others that even very
+\index{Tresca on flow of solids}%
+rigid and elastic substances lose their rigidity
+and their elasticity, and become plastic under the
+action of sufficiently great forces. It appears to
+me, therefore, legitimate to hold to the belief in
+the temporary rigidity of the earth's mass, as explained
+in \Ref{Chapter}{XV}., whilst contending that
+under a change of rotational velocity the earth
+may have become plastic, and so have maintained
+a figure adapted to its speed. Geological observation
+shows that rocks have been freely twisted
+and bent near the earth's surface, and it is impossible
+to doubt that under altered rotation the
+deeper portions of the earth would have been
+subjected to very great stress. I conjecture that
+the internal layers might adapt themselves by
+continuous flow, whilst the superficial portion
+might yield impulsively. Earthquakes are probably
+due to unequal shrinkage of the planet in
+cooling, and each shock would tend to bring the
+\PageSep{301}
+strata into their position of rest; thus the earth's
+surface would avail itself of the opportunity afforded
+by earthquakes of acquiring its proper
+shape. The deposit in the sea of sediment, derived
+from the denudation of continents, affords
+another means of adjustment of the figure of the
+planet. I believe then that the earth has always
+maintained a shape nearly appropriate to its rotation.
+The existence of the continents proves
+that the adjustment has not been perfect, and we
+shall see reason to believe that there has been
+also a similar absence of complete adjustment in
+the interior.
+
+But the opinion here maintained is not shared
+by the most eminent of living authorities, Lord
+Kelvin; for he holds that the fact that the average
+\index{Kelvin, Lord!denies adjustment of earth's figure to changed rotation}%
+figure of the earth corresponds with the
+actual length of the day proves that the planet
+was consolidated at a time when the rotation was
+but little more rapid than it is now. The difference
+between us is, however, only one of degree,
+for he considers that the power of adjustment is
+slight, whilst I hold that it would be sufficient
+to bring about a considerable change of shape
+within the period comprised in geological history.
+
+If the adjustment of the planet's figure were
+perfect, the continents would sink below the
+ocean, which would then be of uniform depth.
+But there is no superficial sign, other than the
+dry land, of absence of adaptation to the present
+\PageSep{302}
+\index{Moon and earth!inequality in motion indicates internal density of earth}%
+rotation---unless indeed the deep polar sea discovered
+by Nansen be such. Yet, as I have
+hinted above, some tokens still exist in the earth
+\index{Earth and moon!internal density}%
+of the shorter day of the past. The detection of
+this evidence depends however on arguments of
+so technical a character that I cannot hope in
+such a work as this to do more than indicate the
+nature of the proof.
+
+The earth is denser towards the centre than
+outside, and the layers of equal density are concentric.
+\index{Density!of earth, law of internal}%
+If then the materials were perfectly
+plastic throughout, not only the surface, but
+also each of these layers would be flattened to a
+definite extent, which depends on the rate of rotation
+and on the law governing the internal
+density of the earth. Although the rate at
+which the earth gets denser is unknown, yet it is
+possible to assign limits to the density at various
+depths. Thus it can be proved that at any internal
+point the density must lie between two
+values which depend on the position of the point
+in question. So also, the degree of flattening at
+any internal point is found to lie between two
+extreme limits, provided that all the internal layers
+are arranged as they would be if the whole
+mass were plastic.
+
+Now variations in the law of internal density
+and in the internal flattening would betray themselves
+to our observation in several ways. In
+the first place, gravity on the earth's surface
+\index{Earth and moon!adjustment of figure to suit change of rotation|)}%
+\index{Gravity, variation according to latitude}%
+\index{Plasticity of earth under change of rotation|)}%
+\PageSep{303}
+\index{Meteorological!conditions dependent on earth's rotation}%
+\index{Moon and earth!inequality in motion indicates internal density of earth}%
+\index{Nutation!value of, indicates internal density of earth}%
+would be changed. The force of gravity at the
+\index{Gravity, variation according to latitude}%
+poles is greater than at the equator, and the law
+of its variation according to latitude is known.
+In the second place the amount of the flattening
+of the earth's surface would be altered, and the
+present figure of the earth is known with considerable
+exactness. Thirdly the figure and law of
+density of the earth govern a certain irregularity
+or inequality in the moon's motion, which has
+been carefully evaluated by astronomers. Lastly
+the precessional and nutational motion of the
+earth is determined by the same causes, and these
+motions also are accurately known. These four
+facts of observation---gravity, the ellipticity of
+\index{Ellipticity of earth's strata in excess for present rotation}%
+the earth, the lunar inequality, and the precessional
+and nutational motion of the earth---are
+so intimately intertwined that one of them cannot
+be touched without affecting the others.
+
+Now Édouard Roche, a French mathematician,
+\index{Roche, E.!ellipticity of internal strata of earth}%
+has shown that if the earth is perfectly plastic,
+so that each layer is exactly of the proper shape
+for the existing rotation, it is not possible to adjust
+the unknown law of internal density so as
+\index{Precession, value of, indicates internal density of earth}%
+to make the values of all these elements accord
+with observation. If the density be assumed
+such as to fit one of the data, it will produce a
+disagreement with observation in others. If,
+however, the hypothesis be abandoned that the
+internal strata all have the proper shapes, and if
+it be granted that they are a little more flattened
+\PageSep{304}
+\index{Ellipticity of earth's strata in excess for present rotation}%
+\index{Geological evidence of earth's plasticity!as to retardation of earth's rotation|(}%
+than is due to the present rate of rotation, the
+data are harmonized together; and this is just
+what would be expected according to the theory
+of tidal friction. But it would not be right to
+attach great weight to this argument, for the
+absence of harmony is so minute that it might
+be plausibly explained by errors in the numerical
+data of observation. I notice, however, that the
+most competent judges of this intricate subject
+are disposed to regard the discrepancy as a
+reality.
+
+\DPchg{}{\TB}
+
+We have seen in the preceding chapter that
+the length of day has changed but little within
+historical times. But the period comprised in
+written history is almost as nothing compared
+with the whole geological history of the earth.
+We ought then to consider whether geology furnishes
+any evidence bearing on the theory of
+tidal friction. The meteorological conditions on
+the earth are dependent to a considerable extent
+on the diurnal rotation of the planet, and therefore
+those conditions must have differed in the
+past. Our storms are of the nature of aerial eddies,
+and they derive their rotation from that of
+the earth. Accordingly storms were probably
+more intense when the earth spun more rapidly.
+The trunks of trees should be stronger than they
+are now to withstand more violent storms. But
+I cannot learn that there is any direct geological
+evidence on this head, for deciduous trees with
+\PageSep{305}
+\index{Ripple mark in sand preserved in geological strata}%
+stiff trunks seem to have been a modern product
+of geological time, whilst the earlier trees more
+nearly resembled bamboos, which yield to the
+wind instead of standing up to it. It seems possible
+that trees and plants would not be exterminated,
+even if they suffered far more wreckage
+than they do now. If trees with stiff trunks
+could only withstand the struggle for existence
+when storms became moderate in intensity, their
+absence from earlier geological formations would
+be directly due to the greater rapidity of the
+earth's rotation in those times.
+
+According to our theory the tides on the seacoast
+must certainly have had a much wider
+range, and river floods must probably have been
+more severe. The question then arises whether
+these agencies should have produced sedimentary
+deposits of coarser grain than at present. Although
+I am no geologist, I venture to express a
+doubt whether it is possible to tell, within very
+wide limits, the speed of the current or the range
+of the tide that has brought down and distributed
+any sedimentary deposit. I doubt whether any
+geologist would assert that floods might not have
+been twice or thrice as frequent, or that the tide
+might not have had a very much greater range
+than at present.
+
+In some geological strata ripple-marks have
+been preserved which exactly resemble modern
+ones. This has, I believe, been adduced as an
+\PageSep{306}
+argument against the existence of tides of great
+range. Ripples are, however, never produced
+by a violent scour of water, but only by gentle
+currents or by moderate waves. The turn of
+the tide must be gentle to whatever height it
+rises, and so the formation of ripple-mark should
+have no relationship to the range of tide.
+
+It appears then that whilst geology affords no
+direct confirmation of the theory, yet it does not
+present any evidence inconsistent with it. Increased
+activity in the factors of change is important
+to geologists, since it renders intelligible
+a diminution in the time occupied by the history
+of the earth; and thus brings the views of the
+\index{Earth and moon!probably once molten}%
+geologist and of the physicist into better harmony.
+
+Although in this discussion I have maintained
+the possibility that a considerable portion of the
+changes due to tidal friction may have occurred
+within geological history, yet it seems to me
+probable that the greater part must be referred
+back to pre-geological times, when the planet
+was partially or entirely molten.
+
+\TB
+
+The action of the moon and sun on a plastic
+and viscous planet would have an effect of which
+some remains may perhaps still be traceable.
+The relative positions of the moon and of the
+frictionally retarded tide were illustrated in the
+last chapter by~\fig{36}. That figure shows that
+\index{Geological evidence of earth's plasticity!as to retardation of earth's rotation|)}%
+\PageSep{307}
+the earth's rotation is retarded by forces acting
+\index{Earth and moon!distortion under primeval tidal friction}%
+on the tidal protuberances in a direction adverse
+to the planet's rotation. As the plastic substance,
+of which we now suppose the planet to
+be formed, rises and falls rhythmically with the
+tide, the protuberant portions are continually
+subject to this retarding force. Meanwhile the
+internal portions are urged onward by the
+inertia due to their velocity. Accordingly there
+must be a slow motion of the more superficial
+portions with reference to the interior. From
+the same causes, under present conditions, the
+whole ocean must have a slow westerly drift, although
+it has not been detected by observation.
+
+Returning however to our plastic planet, the
+equatorial portion is subjected to greater force
+than the polar regions, and if meridians were
+painted on its surface, as on a map, they would
+gradually become distorted. In the equatorial
+belt the original meridional lines would still run
+north and south, but in the northern hemisphere
+they would trend towards the northeast, and in
+the southern hemisphere towards the southeast.
+This distortion of the surface would cause the
+surface to wrinkle, and the wrinkles should be
+warped in the directions just ascribed to the
+meridional lines. If the material yielded very
+easily I imagine that the wrinkles would be
+small, but if it were so stiff as only to yield with
+difficulty they might be large.
+\PageSep{308}
+
+There can be no doubt as to the correctness
+\index{History!of earth and moon|(}%
+of this conclusion as to a stiff yet viscous planet,
+but the application of these ideas to the earth is
+hazardous and highly speculative. We do, however,
+observe that the continents, in fact, run
+\index{Continents, trend of, possibly due to primeval tidal friction}%
+roughly north and south. It may appear fanciful
+to note, also, that the northeastern coast of
+America, the northern coast of China, and the
+southern extremity of South America have the
+proper theoretical trends. But the northwestern
+coast of America follows a line directly adverse
+to the theory, and the other features of the globe
+are by no means sufficiently regular to inspire
+much confidence in the justice of the conjecture.\footnote
+ {See, also, W. Prinz, \Title{Torsion apparente des planètes}, ``Annuaire
+ de l'Obs.~R. de~Bruxelles,'' 1891.}
+
+\TB
+
+We must now revert to the astronomical aspects
+of our problem. It is natural to inquire
+whether the theory of tidal friction is competent
+to explain any peculiarities of the motion of the
+moon and earth other than those already considered.
+It has been supposed thus far that the
+moon moves over the earth's equator in a circular
+orbit, and that the equator coincides with the
+plane in which the earth moves in its orbit. But
+the moon actually moves in a plane different
+from that in which the earth revolves round the
+sun, her orbit is not circular but elliptic, and the
+\PageSep{309}
+earth's equator is oblique to the orbit. We must
+consider, then, how tidal friction will affect these
+three factors.
+
+Let us begin by considering the obliquity of
+the equator to the ecliptic, which produces the
+seasonal changes of winter and summer. The
+problem involved in the disturbance of the motion
+of a rotating body by any external force is
+too complex for treatment by general reasoning,
+and I shall not attempt to explain in detail the
+interaction of the moon and earth in this respect.
+
+The attractions of the moon and sun on the
+equatorial protuberance of the earth causes the
+earth's axis to move slowly and continuously
+with reference to the fixed stars. At present,
+the axis points to the pole-star, but $13,000$~years
+hence the present pole-star will be $47°$~distant
+from the pole, and in another $13,000$~years it
+will again be the pole-star. Throughout this
+precessional movement the obliquity of the equator
+to the ecliptic remains constant, so that winter
+and summer remain as at present. There is
+also, superposed on the precession, the nutational
+or nodding motion of the pole to which I referred
+in \Ref{Chapter}{XV}. In the absence of tidal
+friction the attractions of the moon and sun on
+the tidal protuberance would slightly augment
+the precession due to the solid equatorial protuberance,
+and would add certain very minute
+nutations of the earth's axis; the amount of
+\PageSep{310}
+these tidal effects, is, however, quite insignificant.
+But under the influence of tidal friction,
+the matter assumes a different aspect, for the
+earth's axis will not return at the end of each
+nutation to exactly the same position it would
+have had in the absence of friction, and there is
+a minute residual effect which always tends in
+the same direction. A motion of the pole may
+be insignificant when it is perfectly periodic, but
+it becomes important in a very long period of
+time when the path described is not absolutely
+reëntrant. Now this is the case with regard to
+the motion of the earth's axis under the influence
+of frictionally retarded tides, for it is found
+to be subject to a gradual drift in one direction.
+
+In tracing the history of the earth and moon
+backwards in time we found the day and month
+growing shorter, but at such relative speeds that
+the number of days in the month diminished until
+the day and month became equal. This conclusion
+remains correct when the earth is oblique
+to its orbit, but the effect on the obliquity is
+\index{Ecliptic, obliquity of, due to tidal friction|(}%
+\index{Obliquity of ecliptic, effects of tidal friction on|(}%
+found to depend in a remarkable manner upon
+the number of days in the month. At present
+and for a long time in the past the obliquity
+is increasing, so that it was smaller long ago.
+But on going back to the time when the day
+was six and the month twelve of our present
+hours we find that the tendency for the obliquity
+to increase vanishes. In other words, if
+\PageSep{311}
+there are more than two days in a month the
+obliquity will increase, if less than two it will
+diminish.
+
+Whatever may be the number of days in the
+month, the rate of increase or diminution of
+obliquity varies as the obliquity which exists at
+the moment under consideration. If, then, a
+planet be spinning about an axis absolutely perpendicular
+to the plane of its satellite's orbit, the
+obliquity remains invariable. But if we impart
+infinitesimal obliquity to a planet whose day is
+less than half a month, that infinitesimal obliquity
+will increase; whilst, if the day is more
+than half a month, the infinitesimal obliquity
+will diminish. Accordingly, the motion of a
+planet spinning upright is stable, if there are
+less than two days in a month, and unstable if
+there are more than two.
+
+It is not legitimate to ascribe the whole of
+the present obliquity of~$23\frac{1}{2}°$ to the influence of
+tidal friction, because it appears that when there
+were only two days in the month, the obliquity
+was still as much as~$11°$. It is, moreover, impossible
+to explain the considerable obliquity of the
+other planets to their orbits by this cause. It
+must, therefore, be granted that there was some
+unknown cause which started the planets in rotation
+about axes oblique to their orbits. It remains,
+however, certain that a planet, rotating primitively
+without obliquity, would gradually become
+\PageSep{312}
+inclined to its orbit, although probably not to so
+great an extent as we find in the case of the
+earth.
+
+The next subject to be considered is the fact
+that the moon's orbit is not circular but eccentric.
+Here, again, it is found that if the tides
+were not subject to friction, there would be no
+sensible effect on the shape of the moon's path,
+but tidal friction produces a reaction on the
+moon tending to change the degree of eccentricity.
+In this case, it is possible to indicate by
+general reasoning the manner in which this reaction
+operates. We have seen that tidal reaction
+tends to increase the moon's distance from the
+earth. Now, when the moon is nearest, in perigee,
+the reaction is stronger than when she is
+furthest, in apogee. The effect of the forces in
+perigee is such that the moon's distance at the
+next succeeding apogee is greater than it was at
+the next preceding apogee; so, also, the effect
+of the forces in apogee is an increase in the perigeal
+distance. But the perigeal effect is stronger
+than the apogeal, and, therefore, the apogeal distances
+increase more rapidly than the perigeal
+ones. It follows, therefore, that, whilst the orbit
+as a whole expands, it becomes at the same time
+more eccentric.
+\index{Ecliptic, obliquity of, due to tidal friction|)}%
+\index{Obliquity of ecliptic, effects of tidal friction on|)}%
+
+The lunar orbit is then becoming more eccentric,
+and numerical calculation shows that in
+very early times it must have been nearly circular.
+\PageSep{313}
+\index{Eccentricity of orbit!due to tidal friction}%
+\index{Moon and earth!eccentricity of orbit increased by tidal friction}%
+\index{Saint@St.\ Vénant on flow of solids}%
+\index{See, T. J. J.!eccentricity of orbits of double stars}%
+But mathematical analysis indicates that in
+this case, as with the obliquity, the rate of
+increase depends in a remarkable manner upon
+the number of days in the month. I find in
+fact that if eighteen days are less than eleven
+months the eccentricity will increase, but in the
+converse case it will diminish; in other words
+the critical stage at which the eccentricity is
+stationary is when $1\frac{7}{11}$~days is equal to the
+month. It follows from this that the circular
+orbit of the satellite is dynamically stable or
+\index{Orbit!of double stars, very eccentric}%
+unstable according as $1\frac{7}{11}$~days is less or greater
+than the month.
+
+The effect of tidal friction on the eccentricity
+has been made the basis of extensive astronomical
+speculations by Dr.~See. I shall revert to
+this subject in \Ref{Chapter}{XIX}., and will here
+merely remark that systems of double stars are
+\index{Stars!double, eccentricity of orbits}%
+found to revolve about one another in orbits of
+great eccentricity, and that Dr.~See supposes
+that the eccentricity has arisen from the tidal
+action of each star on the other.
+
+The last effect of tidal friction to which I
+have to refer is that on the plane of the moon's
+orbit. The lunar orbit is inclined to that of the
+earth round the sun at an angle of~$5°$, and the
+problem to be solved is as to the nature of the
+effect of tidal friction on that inclination. The
+nature of the relation of the moon's orbit to the
+ecliptic is however so complex that it appears
+\index{History!of earth and moon|)}%
+\PageSep{314}
+\index{Eccentricity of orbit!due to tidal friction}%
+\index{Moon and earth!eccentricity of orbit increased by tidal friction}%
+hopeless to explain the effects of tidal action
+without the use of mathematical language, and
+I must frankly give up the attempt. I may,
+however, state that when the moon was near the
+earth she must have moved nearly in the plane
+of the earth's equator, but that the motion gradually
+changed so that she has ultimately come to
+move nearly in the plane of the ecliptic. These
+two extreme cases are easily intelligible, but the
+transition from one case to the other is very
+complicated. It may suffice for this general
+account of the subject to know that the effects
+of tidal friction are quite consistent with the
+present condition of the moon's motion, and
+with the rest of the history which has been
+traced.
+
+This discussion of the effects of tidal friction
+may be summed up thus:---
+
+If a planet consisted partly or wholly of molten
+lava or of other fluid, and rotated rapidly about
+an axis perpendicular to the plane of its orbit,
+and if that planet was attended by a single satellite,
+revolving with its month a little longer than
+the planet's day, then a system would necessarily
+be developed which would have a strong resemblance
+to that of the earth and moon.
+
+A theory reposing on \textit{veræ causæ} which brings
+into quantitative correlation the lengths of the
+present day and month, the obliquity of the
+ecliptic, the eccentricity and the inclination of
+\PageSep{315}
+the lunar orbit, should have strong claims to
+acceptance.
+
+\begin{Authorities}
+G.~H. Darwin. A series of papers in the ``Phil.\ Trans.\ Roy.\
+\index{Darwin, G. H.!papers on tidal friction}%
+Soc.'' pt.~i.\ 1879, pt.~ii.\ 1879, pt.~ii.\ 1880, pt.~ii.\ 1881, pt.~i.\ 1882,
+and abstracts (containing general reasoning) in the corresponding
+Proceedings; also ``Proc.\ Roy.\ Soc.''\ vol.~29, 1879, p.~168 (in
+part republished in Thomson and Tait's \Title{Natural Philosophy}),
+and vol.~30, 1880, p.~255.
+
+Lord Kelvin, \Title{On Geological Time}, ``Popular Lectures and
+\index{Kelvin, Lord!on geological time}%
+Addresses,'' vol.~iii. Macmillan, 1894.
+
+Roche. The investigations of Roche and of others are given
+in Tisserand's \Title{Mécanique Céleste}, vol.~ii. Gauthier-Villars, 1891.
+\index{Tisserand, Roche's investigations as to earth's figure}%
+
+Tresca and St.~Vénant, \Title{Sur l'écoulement des Corps Solides},
+``Mémoires des Savants Étrangers,'' Académie des Sciences de
+Paris, vols.\ 18~and~20.
+
+Schiaparelli, \Title{Considerazioni sul moto rotatorio del pianeta
+\index{Schiaparelli on rotation of Venus and Mercury}%
+Venere}. Five notes in the ``Rendiconti del R.~Istituto Lombardo,''
+vol.~23, and \Title{Sulla rotazione di Mercurio}, ``Ast.\ Nach.,''
+No.~2944. An abstract is given in ``Report of Council of R.
+Ast.\ Soc.,'' Feb.~1891.
+
+Lowell, Mercury, ``Ast.\ Nach.,'' No.~3417. \Title{Mercury and Determination
+\index{Lowell, P., on rotations of Venus and Mercury}%
+of Rotation Period~\dots\ of Venus}, ``Monthly Notices
+R. Ast.\ Soc.,'' vol.~57, 1897, p.~148. \Title{Further proof},~\&c., \textit{ibid}.\
+p.~402.
+
+Douglass, \Title{Jupiter's third Satellite}, ``Ast.\ Nach.,'' No.~3432.
+\index{Douglass, rotation of Jupiter's satellites}%
+\Title{Rotation des IV~Jupitersmondes}, ``Ast.\ Nach.,'' No.~3427, confirming
+Engelmann, \Title{Ueber~\dots\ Jupiterstrabanten}, Leipzig, 1871.
+
+Barnard, \Title{The third and fourth Satellites of Jupiter}, ``Ast.\
+\index{Barnard, rotation of Jupiter's satellites}%
+Nach.,'' No.~3453.
+\end{Authorities}
+\index{Friction of tides|)}%
+\PageSep{316}
+
+
+\Chapter[Figures of Equilibrium]{XVIII}
+{The Figures of Equilibrium of a Rotating
+Mass of Liquid}
+
+\First{The} theory of the tides involves the determination
+\index{Equilibrium, figures of, of rotating liquid|(}%
+\index{Figure of equilibrium!of rotating liquid|(}%
+\index{Rotating liquid, figures of equilibrium|(}%
+of the form assumed by the ocean under
+the attraction of a distant body, and it now
+remains to discuss the figure which a rotating
+mass of liquid may assume when it is removed
+from all external influences. The forces which
+act upon the liquid are the mutual gravitation
+of its particles, and the centrifugal force due to
+its rotation. If the mass be of the appropriate
+shape, these two opposing forces will balance
+one another, and the shape will be permanent.
+The problem in hand is, then, to determine
+what shapes of this kind are possible.
+
+In 1842 a distinguished Belgian physicist, M.~Plateau,\footnote
+ {He is justly celebrated not only for his discoveries, but also
+ for his splendid perseverance in continuing his researches after
+ he had become totally blind.}
+devised an experiment which affords
+\index{Capillarity of liquids, and Plateau's experiment|(}%
+\index{Plateau, experiment on figure of rotating globule|(}%
+a beautiful illustration of the present subject.
+The experiment needs very nice adjustment in
+several respects, but I refer the reader to
+Plateau's paper for an account of the necessary
+\PageSep{317}
+\index{Surface tension of liquids}%
+precautions. Alcohol and water may be so
+mixed as to have the same density as olive oil.
+If the adjustment of density is sufficiently exact,
+a mass of oil will float in the mixture, in the
+form of a spherical globule, without any tendency
+to rise or fall. The oil is thus virtually
+relieved from the effect of gravity. A straight
+wire, carrying a small circular disk at right
+angles to itself, is then introduced from the top
+of the vessel. When the disk reaches the
+globule, the oil automatically congregates itself
+round the disk in a spherical form, symmetrical
+with the wire.
+
+The disk is then rotated slowly and uniformly,
+and carries with it the oil, but leaves the surrounding
+mixture at rest. The globule is then
+seen to become flattened like an orange, and as
+the rotation quickens it dimples at the centre,
+and finally detaches itself from the disk in the
+form of a perfect ring. This latter form is only
+transient; for the oil usually closes in again
+round the disk, or sometimes, with slightly different
+manipulation, the ring may break into
+drops which revolve round the centre, rotating
+round their axes as they go.
+
+The force which holds a drop of water, or
+this globule of oil, together is called ``surface
+tension'' or ``capillarity.'' It is due to a certain
+molecular attraction, quite distinct from
+that of gravitation, and it produces the same
+\PageSep{318}
+effect as if the surface of the liquid were enclosed
+in an elastic skin. There is of course no
+actual skin, and yet when the liquid is stirred
+the superficial particles attract their temporary
+neighbors so as to restore the superficial elasticity,
+continuously and immediately. The intensity
+of surface tension depends on the nature
+\index{Surface tension of liquids}%
+of the material with which the liquid is in contact;
+thus there is a definite degree of tension
+in the skin of olive oil in contact with spirits
+and water.
+
+A globule at rest necessarily assumes the form
+of a sphere under the action of surface tension,
+but when it rotates it is distorted by centrifugal
+force. The polar regions become less curved,
+and the equatorial region becomes more curved,
+until the excess of the retaining power at the
+equator over that at the poles is sufficient to
+restrain the centrifugal force. Accordingly the
+struggle between surface tension and centrifugal
+force results in the assumption by the globule
+of an orange-like shape, or, with greater speed
+of rotation, of the other figures of equilibrium.
+\index{Capillarity of liquids, and Plateau's experiment|)}%
+
+In very nearly the same way a large mass of
+gravitating and rotating liquid will naturally
+assume certain definite forms. The simplest
+case of the kind is when the fluid is at rest in
+space, without any rotation. Then mutual gravitation
+is the only force which acts on the system.
+The water will obviously crowd together
+\PageSep{319}
+into the smallest possible space, so that every
+particle may get as near to the centre as its
+neighbors will let it. I suppose the water to be
+incompressible, so that the central portion, although
+pressed by that which lies outside of it,
+does not become more dense; and so the water
+does not weigh more per cubic foot near the
+centre than towards the outside. Since there
+is no upwards and downwards, or right and
+left about the system, it must be symmetrical in
+every direction; and the only figure which possesses
+this quality of universal symmetry is the
+sphere. A sphere is then said to be a figure of
+equilibrium of a mass of fluid at rest.
+
+If such a sphere of water were to be slightly
+deformed, and then released, it would oscillate
+to and fro, but would always maintain a nearly
+spherical shape. The speed of the oscillation
+depends on the nature of the deformation impressed
+upon it. If the water were flattened to
+the shape of an orange and released, it would
+spring back towards the spherical form, but
+would overshoot the mark, and pass on to a
+lemon shape, as much elongated as the orange
+was flattened. It would then return to the
+orange shape, and so on backwards and forwards,
+passing through the spherical form at
+each oscillation. This is the simplest kind of
+oscillation which the system can undergo, but
+there is an infinite number of other modes of
+\index{Plateau, experiment on figure of rotating globule|)}%
+\PageSep{320}
+any degree of complexity. The mathematician
+can easily prove that a liquid globe, of the same
+density as the earth, would take an hour and a
+half to pass from the orange shape to the lemon
+shape, and back to the orange shape. At present,
+the exact period of the oscillation is not
+the important point, but it is to be noted that if
+the body be set oscillating in any way whatever,
+it will continue to oscillate and will always remain
+nearly spherical. We say then that the
+sphere is a stable form of equilibrium of a mass
+of fluid. The distinction between stability and
+instability has been already illustrated in \Ref{Chapter}{XVI}.\
+by the cases of an egg lying on its
+side and balanced on its end, and there is a
+similar distinction between stable and unstable
+modes of motion.
+
+Let us now suppose the mass of water to rotate
+slowly, all in one piece as if it were solid.
+We may by analogy with the earth describe the
+axis of rotation as polar, and the central plane,
+at right angles to the axis, as equatorial. The
+equatorial region tends to move outwards in consequence
+of the centrifugal force of the rotation,
+and this tendency is resisted by gravitation which
+tends to draw the water together towards the
+centre. As the rotation is supposed to be very
+slow, centrifugal force is weak, and its effects are
+small thus the globe is very slightly flattened at
+the poles, like an orange or like the earth itself.
+\PageSep{321}
+Such a body resembles the sphere in its behavior
+when disturbed; it will oscillate, and its average
+figure in the course of its swing is the orange
+shape. It is therefore stable.
+
+But it has been discovered that the liquid may
+also assume two other alternative forms. One
+of these is extremely flattened and resembles a
+flat cheese with rounded edges. As the disk of
+liquid is very wide, the centrifugal force at the
+equator is very great, although the rotation is
+very slow. In the case of the orange-shaped figure,
+the slower the rotation the less is the equatorial
+centrifugal force, because it diminishes
+both with diminution of radius and fall of speed.
+But in the cheese shape the equatorial centrifugal
+force gains more by the increase of equatorial
+radius than it loses by diminution of rotation.
+Therefore the slower the rotation the broader the
+disk, and, if the rotation were infinitely slow, the
+liquid would be an infinitely thin, flat, circular
+disk.
+
+The cheese-like form differs in an important
+respect from the orange-like form. If it were
+slightly disturbed, it would break up, probably
+into a number of detached pieces. The nature
+of the break-up would depend on the disturbance
+from which it started, but it is impossible to trace
+the details of the rupture in any case. We say
+then that the cheese shape is an unstable figure
+of equilibrium of a rotating mass of liquid.
+\PageSep{322}
+\index{Stability!of figures of equilibrium}%
+
+The third form is strikingly different from
+either of the preceding ones. We must now imagine
+the liquid to be shaped like a long cigar,
+and to be rotating about a central axis perpendicular
+to its length. Here again the ends of
+the cigar are so distant from the axis of rotation
+that the centrifugal force is great, and with infinitely
+slow rotation the figure becomes infinitely
+long and thin. Now this form resembles the
+cheese in being unstable. It is remarkable that
+these three forms are independent of the scale on
+which they are constructed, for they are perfectly
+similar whether they contain a few pounds of
+water or millions of tons.\footnote
+ {It is supposed that they are more than a fraction of an inch
+ across, otherwise surface tension would be called into play.}
+If the period of rotation
+and the density of the liquid are given,
+the shapes are absolutely determinable.
+
+The first of the three figures resembles the
+earth and may be called the planetary figure, and
+\index{Planetary figure of equilibrium of rotating liquid}%
+I may continue to refer to the other two as the
+cheese shape and the cigar shape. The planetary
+and cheese shape are sometimes called the spheroids
+of Maclaurin, after their discoverer, and
+\index{Maclaurin!figure of equilibrium of rotating liquid|(}%
+the cigar shape is generally named after Jacobi,
+\index{Jacobi, figure of equilibrium of rotating liquid|(}%
+the great German mathematician. For slow rotations
+the planetary form is stable, and the
+cheese and cigar are unstable. There are probably
+other possible forms of equilibrium, such as
+a ring, or several rings, or two detached masses
+\PageSep{323}
+revolving about one another like a planet and
+satellite, but for the present I only consider these
+three forms.
+
+Now imagine three equal masses of liquid, infinitely
+distant from one another, and each rotating
+\Figure[0.85]{37}{}{png}
+at the same slow speed, and let one of them
+have the planetary shape, the second the cheese
+shape, and the third the cigar shape. When the
+rotations are simultaneously and equally augmented,
+we find the planetary form becoming
+flatter, the cheese form shrinking in diameter
+and thickening, and the cigar form shortening
+and becoming fatter. There is as yet no change
+in the stability, the first remaining stable and
+\index{Stability!of figures of equilibrium}%
+the second and third unstable. The three figures
+are illustrated in~\fig{37}, but the cigar shape
+is hardly recognizable by that name, since it has
+already become quite short and its girth is
+considerable.
+\PageSep{324}
+
+Now it has been proved that as the cigar shape
+shortens, its tendency to break up becomes less
+marked, or in other words its degree of instability
+diminishes. At a certain stage, not as yet
+exactly determined, but which probably occurs
+when the cigar is about twice as long as broad,
+the instability disappears and the cigar form just
+becomes stable. I shall have to return to the
+consideration of this phase later. The condition
+of the three figures is now as follows: The planetary
+form of Maclaurin has become much flattened,
+but is still stable; the cigar form of Jacobi
+has become short and thick, and is just stable;
+and the cheese form of Maclaurin is still unstable,
+\index{Maclaurin!figure of equilibrium of rotating liquid|)}%
+but its diameter has shrunk so much that the
+figure might be better described as a very flat
+orange.
+
+On further augmenting the rotation the form
+of Jacobi still shrinks in length and increases in
+girth, until its length becomes equal to its
+greater breadth. Throughout the transformation
+the axis of rotation has always remained the
+shortest of the three, so that when the length
+becomes equal to the shorter equatorial diameter,
+the shape is not spherical, but resembles that of
+a much flattened orange. In fact, at this stage
+Jacobi's figure of equilibrium has degenerated to
+\index{Jacobi, figure of equilibrium of rotating liquid|)}%
+identity with the planetary shape. One of the
+upper ovals in \fig{38} represents the section of
+the form in which the planetary figure and the
+\PageSep{325}
+cigar figure coalesce, the former by continuous
+flattening, the latter by continuous shortening.
+The other upper figure represents the form to
+which the cheese-like figure of Maclaurin has
+\Figure[0.85]{38}{}{png}
+\index{Poincaré!figure of rotating liquid}%
+been reduced; it will be observed that it presents
+some resemblance to the coalescent form.
+
+When the rotation is further augmented, there
+is no longer the possibility of an elongated Jacobian
+figure, and there remain only the two
+spheroids of Maclaurin. But an important change
+has now supervened, for both these are now unstable,
+and indeed no stable form consisting of a
+single mass of liquid has yet been discovered.
+
+Still quickening the rotation, the two remaining
+forms, both unstable, grow in resemblance to
+one another, until at length they become identical
+in shape. This limiting form of Maclaurin's
+spheroids is shown in the lower part of~\fig{38}.
+If the liquid were water, it must rotate in $2$~hours
+\PageSep{326}
+$25$~minutes to attain this figure, but it would be
+unstable.
+
+A figure for yet more rapid rotation has not
+been determined, but it seems probable that
+dimples would be formed on the axis, that the
+dimples would deepen until they met, and that
+the shape would then be annular. The actual
+existence of such figures in Plateau's experiment
+is confirmatory of this conjecture.
+
+We must now revert to the consideration of
+the cigar-shaped figure of Jacobi, at the stage
+when it has just become stable. The whole of
+this argument depends on the fact that any figure
+of equilibrium is a member of a continuous
+series of figures of the same class, which gradually
+transforms itself as the rotation varies. Now
+M.~Poincaré has proved that, when we follow a
+\index{Poincaré!law of interchange of stability}%
+given series of figures and find a change from instability
+to stability, we are, as it were, served with
+a notice that there exists another series of figures
+coalescent with the first at that stage. We have already
+seen an example of this law, for the planetary
+figure of Maclaurin changed from stability
+to instability at the moment of its coalescence
+with the figure of Jacobi. Now I said that when
+the cigar form of Jacobi was very long it was
+unstable, but that when its length had shrunk to
+about twice its breadth it became stable; hence
+we have notice that at the moment of change
+another series of forms was coalescent with the
+\PageSep{327}
+cigar. It follows also from Poincaré's investigation
+\index{Poincaré!law of interchange of stability}%
+\index{Poincaré!figure of rotating liquid}%
+that the other series of forms must have
+been stable before the coalescence.
+
+Let us imagine then a mass of liquid in the
+form of Jacobi's cigar-shaped body rotating at
+the speed which just admits of stability, and let
+us pursue the series of changes backwards by
+making it rotate a little slower. We know that
+this retardation of rotation lengthens Jacobi's
+figure, and induces instability, but Poincaré has
+not only proved the existence and stability of the
+other series, but has shown that the shape is
+something like a pear.
+
+Poincaré's figure is represented approximately
+in~\fig{38}, but the mathematical difficulty of the
+problem has been too great to admit of an absolutely
+exact drawing. The further development
+of the pear shape is unknown, when the rotation
+slackens still more. There can, however, be
+hardly any doubt that the pear becomes more
+constricted in the waist, and begins to resemble
+an hour-glass; that the neck of the hour-glass
+becomes thinner, and that ultimately the body
+separates into two parts. It is of course likewise
+unknown up to what stage in these changes
+Poincaré's figure retains its stability.
+
+I have myself attacked this problem from an
+entirely different point of view, and my conclusions
+throw an interesting light on the subject,
+although they are very imperfect in comparison
+\PageSep{328}
+\index{Darwin, G. H.!hour-glass figure of rotating liquid|(}%
+with Poincaré's masterly work. To understand
+this new point of view, we must consider a new
+series of figures, namely that of a liquid planet
+attended by a liquid satellite. The two bodies
+are supposed to move in a circle round one another,
+and each is also to revolve on its axis at
+such a speed as always to exhibit the same face
+to its neighbor. Such a system, although divided
+into two parts, may be described as a figure of
+equilibrium. If the earth were to turn round
+once in twenty-seven days, it would always show
+to the moon the same side, and the moon actually
+does present the same side to us. In this
+case the earth and the moon would form such a
+system as that I am describing. Both the planet
+and the satellite are slightly flattened by their
+rotations, and each of them exercises a tidal influence
+on the other, whereby they are elongated
+towards the other.
+
+The system then consists of a liquid planet
+and liquid satellite revolving round one another,
+so as always to exhibit the same face to one another,
+and each tidally distorting the other. It
+is certain that if the two bodies are sufficiently
+far apart the system is a stable one, for if any
+slight disturbance be given, the whole system will
+not break up. But little is known as yet as to
+the limiting proximity of the planet and satellite,
+which will insure stability.
+
+Now if the rotations and revolutions of the
+\PageSep{329}
+bodies be accelerated, the two masses must be
+brought nearer together in order that the greater
+attraction may counterbalance the centrifugal
+force. But as the two are brought nearer the
+tide-generating force increases in intensity with
+great rapidity, and accordingly the tidal elongation
+of the two bodies is much augmented.
+
+A time will at length come when the ends of
+the two bodies will just touch, and we then have
+a form shaped like an hour-glass with a very
+\Figure[0.7]{39}{Hour-glass Figure of Equilibrium}{png}
+\PageLabel[pg]{329}% [** TN: Used by reference on p. 356 of the original]
+thin neck. The form is clearly Poincaré's figure,
+at an advanced stage of its evolution.
+
+The figure~\figref{39} shows the form of one possible
+\PageSep{330}
+figure of this class; it arises from the coalescence
+of two equal masses of liquid, and the
+shape shown was determined by calculation.
+But there are any number of different sorts of
+hour-glass shapes, according to the relative sizes
+of the planet and satellite which coalesce; and
+in order to form a continuous series with Poincaré's
+pear, it would be necessary to start with
+a planet and satellite of some definitely proportionate
+sizes. Unfortunately I do not know
+what the proportion may be. There are, however,
+certain indications which may ultimately
+lead to a complete knowledge of the series of
+figures from Jacobi's cigar shape down to the
+planet and satellite. It may be shown---and I
+shall have in \Ref{Chapter}{XX}.\ to consider the point
+more in detail---that if our liquid satellite had
+only, say, a thousandth of the mass of the planet,
+and if the two bodies were brought nearer one
+another, at a certain calculable distance the tidal
+action of the big planet on the very small satellite
+would become so intense that it would tear
+it to pieces. Accordingly the contact and coalescence
+of a very small satellite with a large
+planet is impossible. It is, however, certain that
+a large enough satellite---say of half the mass
+of the planet---could be brought up to contact
+with the planet, without the tidal action of the
+planet on the satellite becoming too intense to
+admit of the existence of the latter. There
+\PageSep{331}
+must then be some mass of the satellite, which
+will just allow the two to touch at the same
+moment that the tidal action of the larger on
+the smaller body is on the point of disrupting
+it. Now I suspect, although I do not know,
+that the series of figures which we should find in
+this case is in fact Poincaré's series. This discussion
+shows that the subject still affords an
+interesting field for future mathematicians.
+
+These investigations as to the form of rotating
+masses of liquid are of a very abstract character,
+and seem at first sight remote from practical
+conclusions, yet they have some very interesting
+applications.
+
+The planetary body of Maclaurin is flattened
+at the poles like the actual planets, and the
+degree of its flattening is exactly appropriate to
+the rapidity of its rotation. Although the planets
+are, at least in large part, composed of solid
+matter, yet that matter is now, or was once,
+sufficiently plastic to permit it to yield to the
+enormous forces called into play by rotation and
+gravitation. Hence it follows that the theory
+of Maclaurin's figure is the foundation of that
+of the figures of planets, and of the variation of
+gravity at the various parts of their surfaces.
+In the liquid considered hitherto, every particle
+attracted every other particle, the fluid was
+equally dense throughout, and the figure assumed
+was the resultant of the battle between
+\PageSep{332}
+\index{Figure of planets and their density}%
+\index{Gravity, variation according to latitude}%
+\index{Saturn!law of density and figure}%
+the centrifugal force and gravitation. At every
+part of the liquid the resultant attraction was
+directed nearly, but not quite, towards the
+centre of the shape. But if the attraction had
+everywhere been directed exactly to the centre,
+the degree of flattening would have been
+diminished. We may see that this must be so,
+because if the rotation were annulled, the mass
+would be exactly spherical, and if the rotation
+were not annulled, yet the forces would be such
+as to make the fluid pack closer, and so assume
+a more nearly spherical form than when the
+forces were not absolutely directed to the centre.
+It may be shown in fact that the flattening is
+$2\frac{1}{2}$~times greater in the case of Maclaurin's
+body than it is when the seat of gravitation is
+exactly central.
+
+In the case of actual planets the denser matter
+\index{Planets!figures and internal densities}%
+must lie in the centre and the less dense outside.
+If the central matter were enormously
+denser than superficial rock, the attraction would
+be directed towards the centre. There are then
+two extreme cases in which the degree of flattening
+can be determined,---one in which the density
+\index{Density!of planets determinable from their figures}%
+of the planet is the same all through, giving
+Maclaurin's figure; the other when the density
+is enormously greater at the centre. The flattening
+in the former is $2\frac{1}{2}$~times as great as in
+the latter. The actual condition of a real planet
+must lie between these two extremes. The
+\index{Darwin, G. H.!hour-glass figure of rotating liquid|)}%
+\PageSep{333}
+\index{Figure of planets and their density}%
+knowledge of the rate of rotation of a planet
+and of the degree of its flattening furnishes us
+with some insight into the law of its internal
+density. If it is very much less flat than Maclaurin's
+\index{Density!of planets determinable from their figures}%
+figure, we conclude that it is very dense
+in its central portion. In this way it is known
+with certainty that the central portions of the
+planets Jupiter and Saturn are much denser,
+\index{Jupiter!figure and law of internal density}%
+\index{Planets!figures and internal densities}%
+compared with their superficial portions, than is
+the case with the earth.
+
+I do not propose to pursue this subject into
+the consideration of the law of the variation of
+gravity on the surface of a planet; but enough
+has been said to show that these abstract investigations
+have most important practical applications.
+
+\begin{Authorities}
+Plateau, ``Mémoires de l'Académie Royale de~Belgique,''
+vol.~xvi. 1843.
+
+Thomson and Tait's \Title{Natural Philosophy} or other works on
+hydrodynamics give an account of figures of equilibrium.
+
+Poincaré, \Title{Sur l'équilibre d'une masse fluide animée d'un mouvement
+de rotation}, ``Acta Mathematica,'' vol.~7, 1885.
+
+An easier and different presentation of the subject is contained
+in an inaugural dissertation by Schwarzschild (Annals of Munich
+\index{Schwarzschild!exposition of Poincaré's theory}%
+Observatory, vol.~iii. 1896). He considers that Poincaré's
+proof of the stability of his figure is not absolutely conclusive.
+
+G.~H. Darwin, \Title{Figures of Equilibrium of Rotating Masses of
+\index{Darwin, G. H.!Jacobi's ellipsoid}%
+Fluid}, ``Transactions of Royal Society,'' vol.~178, 1887.
+
+G.~H. Darwin, \Title{Jacobi's Figure of Equilibrium},~\&c., ``Proceedings
+\index{Equilibrium, figures of, of rotating liquid|)}%
+\index{Figure of equilibrium!of rotating liquid|)}%
+Roy.\ Soc.,'' vol.~41, 1886, p.~319.
+
+S.~Krüger, \Title{Ellipsoidale Evenwichtsvormen},~\&c., Leeuwen, Leiden,
+\index{Krüger, figures of equilibrium of liquid}%
+1896; \Title{Sur l'ellipsoïde de Jacobi}, ``Nieuw Archief voor Wiskunde,''
+2d~series, 3d~part, 1898. The author shows that G.~H.
+Darwin had been forestalled in much of his work on Jacobi's
+figure, and he corrects certain mistakes.
+\end{Authorities}
+\index{Rotating liquid, figures of equilibrium|)}%
+\PageSep{334}
+
+
+\Chapter{XIX}
+{The Evolution of Celestial Systems}
+
+\First{Men} will always aspire to peer into the remote
+\index{Evolution of celestial systems|(}%
+\index{Solar!system, nebular hypothesis as to origin of|(}%
+past to the utmost of their power, and the fact
+that their success or failure cannot appreciably
+influence their life on the earth will never deter
+them from such endeavors. From this point
+of view the investigations explained in the last
+chapter acquire much interest, since they form
+the basis of the theories of cosmogony which
+seem most probable by the light of our present
+knowledge.
+
+We have seen that an annular figure of equilibrium
+\index{Kant!nebular hypothesis|(}%
+\index{Nebular hypothesis|(}%
+actually exists in Plateau's experiment,
+and it is almost certainly a possible form amongst
+celestial bodies. Plateau's ring has however
+only a transient existence, and tends to break up
+into globules, spinning on their axes and revolving
+round the centre. In this result we saw a
+close analogy with the origin of the planets, and
+regarded his experiment as confirmatory of the
+Nebular Hypothesis, of which I shall now give a
+short account.\footnote
+ {My knowledge of the history of the Nebular Hypothesis is
+ entirely derived from an interesting paper by Mr.~G.~F. Becker,
+\index{Becker, G. F., on Nebular Hypothesis}%
+ on ``Kant as a Natural Philosopher,'' \Title{American Journal of Science},
+ vol.~v. Feb.~1898.}
+\PageSep{335}
+
+The first germs of this theory are to be found
+in Descartes' ``Principles of Philosophy,'' published
+\index{Descartes, vortical theory of cosmogony}%
+in~1644. According to him the sun and
+planets were represented by eddies or vortices in
+a primitive chaos of matter, which afterwards
+formed the centres for the accretion of matter.
+As the theory of universal gravitation was propounded
+for the first time half a century later
+than the date of Descartes' book, it does not
+seem worth while to follow his speculations
+further. Swedenborg formulated another vortical
+cosmogony in~1734, and Thomas Wright of
+\index{Wright, Thomas, on a theory of cosmogony}%
+Durham published in 1750 a book of preternatural
+dullness on the same subject. It might not
+have been worth while to mention Wright, but
+that Kant acknowledges his obligation to him.
+
+The Nebular Hypothesis has been commonly
+associated with the name of Laplace, and he undoubtedly
+\index{Laplace!nebular hypothesis|(}%
+avoided certain errors into which his
+precursors had fallen. I shall therefore explain
+Laplace's theory, and afterwards show how he
+was, in most respects, really forestalled by the
+great German philosopher Kant.
+
+Laplace supposed that the matter now forming
+the solar system once existed in the form of a
+lens-shaped nebula of highly rarefied gas, that it
+rotated slowly about an axis perpendicular to the
+present orbits of the planets, and that the nebula
+extended beyond the present orbit of the furthest
+planet. The gas was at first expanded by heat,
+\PageSep{336}
+and as the surface cooled the central portion
+condensed and its temperature rose. The speed
+of rotation increased in consequence of the contraction,
+according to a well known law of mechanics
+called ``the conservation of moment of
+momentum;''\footnote
+ {Kant fell into error through ignorance of the generality of
+ this law, for he imagined that rotation could be generated from
+ rest.}
+the edges of the lenticular mass
+of gas then ceased to be continuous with the
+more central portion, and a ring of matter was
+detached, in much the same way as in Plateau's
+experiment. Further cooling led to further contraction
+and consequently to increased rotation,
+until a second ring was shed, and so on successively.
+The rings then ruptured and aggregated
+themselves into planets whilst the central nucleus
+formed the sun.
+
+Virtually the same theory had been propounded
+by Kant many years previously, but I am not
+aware that there is any reason to suppose that
+Laplace had ever read Kant's works. In a paper,
+to which I have referred above, Mr.~G.~F.
+Becker makes the following excellent summary
+\index{Becker, G. F., on Nebular Hypothesis|(}%
+of the relative merits of Kant and Laplace; he
+writes:---
+
+``Kant seems to have anticipated Laplace almost
+completely in the more essential portions
+of the nebular hypothesis. The great Frenchman
+was a child when Kant's theory was issued,
+\PageSep{337}
+and the `Système du Monde,' which closes with
+the nebular hypothesis, did not appear until
+1796. Laplace, like Kant, infers unity of origin
+for the members of the solar system from the
+similarity of their movements, the small obliquity
+and small eccentricity of the orbits of either
+planets or satellites.\footnote
+ {``The retrograde satellites of Uranus were discovered by
+ Herschel in~1787, but Laplace in his hypothesis does not refer to
+ them.''}
+Only a fluid extending
+throughout the solar system could have produced
+such a result. He is led to conclude that the
+atmosphere of the sun, in virtue of excessive
+heat, originally extended beyond the solar system
+and gradually shrank to its present limits. This
+nebula was endowed with moment of momentum
+which Kant tried to develop by collisions. Planets
+formed from zones of vapor, which on breaking
+agglomerated\dots. The main points of
+comparison between Kant and Laplace seem to
+be these. Kant begins with a cold, stationary
+nebula which, however, becomes hot by compression
+and at its first regenesis would be in a state
+of rotation. It is with a hot, rotating nebula
+that Laplace starts, without any attempt to account
+for the heat. Kant supposes annular
+zones of freely revolving nebulous matter to
+gather together by attraction during condensation
+of the nebula. Laplace supposes rings left
+\index{Laplace!nebular hypothesis|)}%
+behind by the cooling of the nebula to agglomerate
+\PageSep{338}
+in the same way as Kant had done. While
+both appeal to the rings of Saturn as an example
+of the hypothesis, neither explains satisfactorily
+why the planetary rings are not as stable
+as those of Saturn. Both assert that the positive
+rotation of the planets is a necessary
+consequence of agglomeration, but neither is
+sufficiently explicit. The genesis of satellites is
+for each of them a repetition on a small scale of
+the formation of the system\dots. While Laplace
+assigns no cause for the heat which he ascribes
+to his nebula, Lord Kelvin goes further
+back and supposes a cold nebula consisting of
+separate atoms or of meteoric stones, initially
+possessed of a resultant moment of momentum
+equal or superior to that of the solar system.
+Collision at the centre will reduce them to a
+vapor which then expanding far beyond Neptune's
+orbit will give a nebula such as Laplace
+postulates.\footnote
+ {\Title{Popular Lectures}, vol.~i.\ p.~421.}
+Thus Kelvin goes back to the same
+initial condition as Kant, excepting that Kant
+endeavored (of course vainly) to develop a moment
+of momentum for his system from collisions.''\footnote
+ {Becker, \Title{Amer.\ Journ.\ Science}, vol.~v. 1898, pp.~107,~108.}
+\index{Becker, G. F., on Nebular Hypothesis|)}%
+
+There is good reason for believing that the
+Nebular Hypothesis presents a true statement in
+outline of the origin of the solar system, and of
+the planetary subsystems, because photographs
+%% Plate 3
+\TallFig[0.85]{40}{Nebula in Andromeda}{jpg}
+%% Facing page
+%[Blank Page]
+\PageSep{339}
+of nebulæ have been taken recently in which we
+can almost see the process in action. \Fig{40} is
+a reproduction of a remarkable photograph by
+Dr.~Isaac Roberts of the great nebula in the constellation
+\index{Roberts, I., photograph of nebula in Andromeda}%
+of Andromeda. In it we may see the
+\index{Andromeda, nebula in}%
+\index{Nebula in Andromeda}%
+lenticular nebula with its central condensation,
+the annulation of the outer portions, and even
+the condensations in the rings which will doubtless
+at some time form planets. This system is
+built on a colossal scale, compared with which
+our solar system is utterly insignificant. Other
+\index{Solar!system, nebular hypothesis as to origin of|)}%
+\index{Solar!system, distribution of satellites in|(}%
+nebulæ show the same thing, and although they
+are less striking we derive from them good
+grounds for accepting this theory of evolution
+as substantially true.
+
+\TB
+
+I explained in \Ref{Chapter}{XVI}.\ how the theory
+of tidal friction showed that the moon took her
+origin very near to the present surface of the
+earth. But it was also pointed out that the same
+theory cannot be invoked to explain an origin
+for the planets at a point close to the sun. They
+must in fact have always moved at nearly their
+present distances. In the same way the dimensions
+of the orbits of the satellites of Mars, Jupiter,
+\index{Satellites!distribution of, in solar system|(}%
+Saturn, and Neptune cannot have been
+largely augmented, whatever other effects tidal
+friction may have had. We must therefore still
+rely on the Nebular Hypothesis for the explanation
+\index{Nebular hypothesis|)}%
+of the main features of the system as a
+whole.
+\index{Kant!nebular hypothesis|)}%
+\PageSep{340}
+
+It may, at first sight, appear illogical to maintain
+that an action, predominant in its influence
+on our satellite, should have been insignificant
+in regulating the orbits of all the other bodies
+of the system. But this is not so, for whilst the
+earth is only $80$~times as heavy as the moon, Saturn
+weighs about $4,600$~times as much as its
+satellite Titan, which is by far the largest satellite
+in the solar system; and all the other satellites
+are almost infinitesimal in comparison with their
+primaries. Since, then, the relationship of the
+moon to the earth is unique, it may be fairly contended
+that a factor of evolution, which has been
+predominant in our own history, has been relatively
+insignificant elsewhere.
+
+There is indeed a reason explanatory of this
+singularity in the moon and earth; it lies in the
+fact that the earth is nearer to the sun than any
+other planet attended by a satellite. To explain
+the bearing of this fact on the origin of satellites
+and on their sizes, I must now show how tidal
+friction has probably operated as a perturbing
+influence in the sequence of events, which would
+be normal according to the Nebular Hypothesis.
+
+We have seen that rings should be shed from
+the central nucleus, when the contraction of the
+nebula has induced a certain degree of augmentation
+of rotation. Now if the rotation were
+retarded by some external cause, the genesis of
+a ring would be retarded, or might be entirely
+prevented.
+\PageSep{341}
+
+The friction of the solar tides in a planetary
+nebula furnishes such an external cause, and accordingly
+the rotation of a planetary nebula near
+to the sun might be so much retarded that a ring
+would never be detached from it, and no satellite
+would be generated. From this point of view
+it is noteworthy that Mercury and Venus have
+no satellites; that Mars has two, Jupiter five,
+and that all the exterior planets have several
+satellites. I suggest then that the solar tidal
+friction of the terrestrial nebula was sufficient to
+retard the birth of a satellite, but not to prevent
+it, and that the planetary mass had contracted
+to nearly the present dimensions of the earth
+and had partially condensed into the solid and
+liquid forms, before the rotation had augmented
+sufficiently to permit the birth of a satellite.
+When satellites arise under conditions which are
+widely different, it is reasonable to suppose that
+their masses will also differ much. Hence we can
+understand how it has come about that the relationship
+between the moon and the earth is so
+unlike that between other satellites and their
+planets. In \Ref{Chapter}{XVII}.\ I showed that there
+are reasons for believing that solar tidal friction
+\index{Solar!system, distribution of satellites in|)}%
+has really been an efficient cause of change, and
+this makes it legitimate to invoke its aid in explaining
+the birth and distribution of satellites.
+\index{Satellites!distribution of, in solar system|)}%
+
+\TB
+
+In speaking of the origin of the moon I have
+\PageSep{342}
+\index{Eccentricity of orbit!theory of, in case of double stars}%
+\index{See, T. J. J.!theory of evolution of double stars|(}%
+been careful not to imply that the matter of
+which she is formed was necessarily first arranged
+in the form of a ring. Indeed, the genesis of
+the hour-glass figure of equilibrium from Jacobi's
+form and its fission into two parts indicate the
+possibility of an entirely different sequence of
+events. It may perhaps be conjectured that the
+moon was detached from the primitive earth in
+this way, possibly with the help of tidal oscillations
+due to the solar action. Even if this suggestion
+is only a guess, it is interesting to make
+such speculations, when they have some basis of
+reason.
+
+In recent years astronomers have been trying,
+principally by aid of the spectroscope, to determine
+the orbits of pairs of double stars around
+\index{Stars!theory of evolution|(}%
+one another. It has been observed that, in the
+majority of these systems, the masses of the two
+component stars do not differ from one another
+extremely; and Dr.~See, who has specially devoted
+himself to this research, has drawn attention
+to the great contrast between these systems
+and that of the sun, attended by a retinue of
+infinitesimal planets. He maintains, with justice,
+that the paths of evolution pursued in the two
+cases have probably also been strikingly different.
+
+It is hardly credible that two stars should
+have gained their present companionship by an
+accidental approach from infinite space. They
+cannot always have moved as they do now, and
+\PageSep{343}
+so we are driven to reflect on the changes which
+might supervene in such a system under the
+action of known forces.
+
+The only efficient interaction between a pair
+of celestial bodies, which is known hitherto, is
+a tidal one, and the friction of the oscillations
+introduces a cause of change in the system.
+Tidal friction tends to increase the eccentricity
+of the orbit in which two bodies revolve about
+one another, and its efficiency is much increased
+when the pair are not very unequal in mass and
+when each is perturbed by the tides due to the
+other. The fact that the orbits of the majority
+of the known pairs are very eccentric affords a
+reason for accepting the tidal explanation. The
+only adverse reason, that I know of, is that the
+eccentricities are frequently so great that we
+may perhaps be putting too severe a strain on
+the supposed cause.
+
+But the principal effect of tidal friction is the
+repulsion of the two bodies from one another,
+so that when their history is traced backwards
+we ultimately find them close together. If then
+this cause has been as potent as Dr.~See believes
+it to have been, the two components of a binary
+system must once have been close together.
+From this stage it is but a step to picture to
+ourselves the rupture of a nebula, in the form
+of an hour-glass, into two detached masses.
+
+The theory embraces all the facts of the case,
+\PageSep{344}
+and as such is worthy of at least a provisional
+acceptance. But we must not disguise from
+ourselves that out of the thousands, and perhaps
+millions of double stars which may be visible
+from the earth, we only as yet know the orbits
+and masses of a dozen.
+
+Many years ago Sir John Herschel drew a
+\index{Herschel, observations of twin nebulæ}%
+number of twin nebulæ as they appear through
+a powerful telescope. The drawings probably
+possess the highest degree of accuracy attainable
+by this method of delineation, and the shapes
+present evidence confirmatory of the theory of
+the fission of nebulæ adopted by Dr.~See. But
+since Herschel's time it has been discovered that
+many details, to which our eyes must remain forever
+blind, are revealed by celestial photography.
+The photographic film is, in fact, sensitive to
+those ``actinic'' rays which we may call invisible
+light, and many nebulæ are now found to be
+hardly recognizable, when photographs of them
+are compared with drawings. A conspicuous
+example of this is furnished by the great nebula
+in Andromeda, illustrated above in~\fig{40}.
+
+Photographs, however, do not always aid interpretation,
+for there are some which serve only
+to increase the chaos visible with the telescope.
+We may suspect, indeed, that the complete system
+of a nebula often contains masses of cold
+and photographically invisible gas, and in such
+cases it would seem that the true nature of the
+whole will always be concealed from us.
+\PageSep{345}
+
+Another group of strange celestial objects is
+that of the spiral nebulæ, whose forms irresistibly
+\index{Nebulae@Nebulæ, description of various}%
+suggest violent whirlpools of incandescent
+gas. Although in all probability the motion of
+the gas is very rapid, yet no change of form has
+been detected. We are here reminded of a
+rapid stream rushing past a post, where the form
+of the surface remains constant whilst the water
+itself is in rapid movement; and it seems reasonable
+to suppose that in these nebulæ it is
+only the lines of the flow of the gas which are
+visible. Again, there are other cases in which
+the telescopic view may be almost deceptive in
+its physical suggestions. Thus the Dumb-Bell
+\index{Dumb-bell nebula, description of photograph of}%
+nebula (27~Messier Vulpeculæ), as seen telescopically,
+might be taken as a good illustration of a
+nebula almost ready to split into two stars. If
+this were so, the rotation would be about an
+axis at right angles to the length of the nebula.
+But a photograph of this object shows that the
+system really consists of a luminous globe surrounded
+by a thick and less luminous ring, and
+that the opacity of the sides of the ring takes a
+bite, as it were, out of each side of the disk, and
+so gives it the apparent form of a dumb-bell.
+In this case the rotation must be about an axis
+at right angles to the ring, and therefore along
+the length of the dumb-bell. It is proper to
+add that Dr.~See is well aware of this, and does
+not refer to this nebula as a case of incipient
+fission.
+\PageSep{346}
+
+I have made these remarks in order to show
+that every theory of stellar evolution must be
+full of difficulty and uncertainty. According to
+our present knowledge Dr.~See's theory appears
+to have much in its favor, but we must await its
+confirmation or refutation from the results of
+future researches with the photographic plate,
+the spectroscope, and the telescope.
+
+\begin{Authorities}
+Mr.~G.~F. Becker (\Title{Amer.\ Jour.\ Science}, vol.~v. 1898, art.~xv.)\
+gives the following references to Kant's work: \Title{Sämmtliche
+Werke}, ed.~Hartenstein, 1868 (Tidal Friction and the Aging of
+the Earth), vol.~i.\ pp.~179--206; (Nebular Hypothesis), vol.~i.\
+pp.~207--345.
+
+Laplace, \Title{Système du Monde}, last appendix; the tidal retardation
+of the moon's rotation is only mentioned in the later
+editions.
+
+T.~J.~J. See, \Title{Die Entwickelung der Doppelstern-systeme}, ``Inaugural
+Dissertation,'' 1892. Schade, Berlin.
+
+T.~J.~J. See, \Title{Evolution of the Stellar Systems}, vol.~i.\ 1896.
+Nichols Press, Lynn, Massachusetts. Also a popular article,
+\Title{The Atlantic Monthly}, October, 1897.
+
+G.~H. Darwin, \Title{Tidal Friction~\dots\ and Evolution}, ``Phil.\ Trans.\
+\index{Darwin, G. H.!evolution of satellites}%
+Roy.\ Soc.,'' part~ii.\ 1881, p.~525.
+\end{Authorities}
+\index{Evolution of celestial systems|)}%
+\index{See, T. J. J.!theory of evolution of double stars|)}%
+\index{Stars!theory of evolution|)}%
+\PageSep{347}
+
+
+\Chapter[Saturn's Rings]{XX}
+{Saturn's Rings\protect\footnotemark}
+
+\footnotetext{Part of this chapter appeared as an article in \Title{Harper's
+ Magazine} for June,~1889.}
+
+\First{To} the naked eye Saturn appears as a brilliant
+\index{Saturn!description and picture|(}%
+star, which shines, without twinkling, with a
+yellowish light. It is always to be found very
+nearly in the ecliptic, moving slowly amongst
+the fixed stars at the rate of only thirteen degrees
+per annum. It is the second largest
+planet of the solar system, being only exceeded
+in size by the giant Jupiter. It weighs $91$~times
+as much as our earth, but, being as light as cork,
+occupies $690$~times the volume, and is nine times
+as great in circumference. Notwithstanding its
+great size it rotates around its axis far more
+rapidly than does the earth, its day being only
+$10\frac{1}{2}$~of our hours. It is ten times as far from
+the sun as we are, and its year, or time of revolution
+round the sun, is equal to thirty of our
+years. It was deemed by the early astronomers
+to be the planet furthest from the sun, but that
+was before the discovery by Herschel, at the
+end of the last century, of the further planet
+Uranus, and that of the still more distant Neptune
+by Adams and Leverrier in the year~1846.
+\PageSep{348}
+
+The telescope has shown that Saturn is attended
+by a retinue of satellites almost as numerous
+as, and closely analogous to, the planets
+circling round the sun. These moons are eight
+in number, are of the most various sizes, the
+largest as great as the planet Mars, and the
+smallest very small, and are equally diverse in
+respect of their distances from the planet. But
+besides its eight moons Saturn has another attendant
+absolutely unique in the heavens; it is
+girdled with a flat ring, which, like the planet
+itself, is only rendered visible to us by the
+illumination of sunlight. \Fig{41}, to which
+further reference is made below, shows the general
+appearance of the planet and of its ring.
+The theory of the physical constitution of that
+ring forms the subject of the present chapter.
+
+A system so rich in details, so diversified and
+so extraordinary, would afford, and doubtless
+has afforded, the subject for many descriptive
+essays; but description is not my present object.
+
+The existence of the ring of Saturn seems
+now a very commonplace piece of knowledge,
+and yet it is not $300$~years since the moons of
+Jupiter and Saturn were first detected, and since
+suspicion was first aroused that there was something
+altogether peculiar about the Saturnian
+system. These discoveries, indeed, depended
+entirely on the invention of the telescope. It
+may assist the reader to realize how necessary
+\PageSep{349}
+%[** TN: Oriented vertically in the original]
+\Figure{41}{The Planet Saturn}{jpg}
+\PageSep{350}
+the aid of that instrument was when I say that
+Saturn, when at his nearest to us, is the same in
+size as a sixpenny piece held up at a distance of
+$210$~yards.
+
+It was the celebrated Galileo who first invented
+\index{Galileo!Saturn's ring}%
+a combination of lenses such as is still
+used in our present opera-glasses, for the purpose
+of magnifying distant objects.
+
+In July of~1610 he began to examine Saturn
+with his telescope. His most powerful instrument
+only magnified $32$~times, and although
+such an enlargement should have amply sufficed
+to enable him to make out the ring, yet he persuaded
+himself that what he saw was a large
+bright disk, with two smaller ones touching it,
+one on each side. His lenses were doubtless
+imperfect, but the principal cause of his error
+must have been the extreme improbability of the
+existence of a ring girdling the planet. He
+wrote an account of what he had seen to the
+Grand Duke of Tuscany, Giuliano de'~Medici,
+and to others; he also published to the world an
+anagram which, when the letters were properly
+arranged, read as follows: ``Altissimum planetam
+tergeminum observavi'' (I have seen the
+furthest planet as triple), for it must be remembered
+that Saturn was then the furthest known
+planet.
+
+In 1612 Galileo again examined Saturn, and
+was utterly perplexed and discouraged to find
+\PageSep{351}
+his triple star replaced by a single disk. He
+writes, ``Is it possible that some mocking demon
+has deceived me?'' And here it may be well to
+remark that there are several positions in which
+Saturn's rings vanish from sight, or so nearly
+vanish as to be only visible with the most powerful
+modern telescopes. When the plane of the
+ring passes through the sun, only its very thin
+edge is illuminated; this was the case in~1612,
+when Galileo lost it; secondly, if the plane of
+the ring passes through the earth, we have only
+a very thin edge to look at; and thirdly, when
+the sun and the earth are on opposite sides of
+the ring, the face of the ring which is presented
+to us is in shadow, and therefore invisible.
+
+Some time afterwards Galileo's perplexity was
+increased by seeing that the planet had then a
+pair of arms, but he never succeeded in unraveling
+the mystery, and blindness closed his career
+as an astronomer in~1626.
+
+About thirty years after this, the great Dutch
+astronomer Huyghens, having invented a new
+\index{Huyghens, discovery of Saturn's ring}%
+sort of telescope (on the principle of our present
+powerful refractors), began to examine the planet
+and saw that it was furnished with two loops or
+handles. Soon after the ring disappeared; but
+when, in~1659, it came into view again, he at
+last recognized its true character, and announced
+that the planet was attended by a broad, flat
+ring.
+\PageSep{352}
+
+A few years later it was perceived that there
+were two rings, concentric with one another.
+The division, which may be easily seen in drawings
+of the planet, is still named after Cassini,
+\index{Cassini, discovery of division in Saturn's rings}%
+one of its discoverers. Subsequent observers
+have detected other less marked divisions.
+
+Nearly two centuries later, namely, in~1850,
+Bond in America and Dawes in England, independently
+\index{Bond, discovery of inner ring of Saturn}%
+\index{Dawes, discovery of inner ring of Saturn}%
+and within a fortnight of the same
+time, observed that inside of the well-known
+bright rings there is another very faint dark
+ring, which is so transparent that the edge of
+the planet is visible through it. There is some
+reason to believe that this ring has really become
+more conspicuous within the last $200$~years,
+so that it would not be right to attribute the
+lateness of its detection entirely to the imperfection
+of earlier observations.
+
+It was already discovered in the last century
+that the ring is not quite of the same thickness
+at all points of its circumference, that it is not
+strictly concentric with the planet, and that it
+revolves round its centre. Herschel, with his
+magnificent reflecting telescope, detected little
+beads on the outer ring, and by watching these
+he concluded that the ring completes its revolution
+in $10\frac{1}{2}$~hours.
+
+This sketch of the discovery and observation
+of Saturn's rings has been necessarily very incomplete,
+but we have perhaps already occupied
+too much space with it.
+\PageSep{353}
+
+\Fig{41} exhibits the appearance of Saturn and
+his ring. The drawing is by Bond of Harvard
+University, and is considered an excellent one.
+
+It is usual to represent the planets as they are
+seen through an astronomical telescope, that is
+\Figure{42}{Diagram of Saturn and his Rings}{png}
+to say, reversed. Thus in \fig{41} the south
+pole of the planet is at the top of the plate, and
+unless the telescope were being driven by clockwork,
+the planet would appear to move across
+the field of view from right to left.
+
+The plane of the ring is coincident with the
+equator of the planet, and both ring and equator
+are inclined to the plane of the planet's orbit at
+an angle of $27$~degrees.
+
+A whole essay might be devoted to the discussion
+of this and of other pictures, but we must
+confine ourselves to drawing attention to the
+well-marked split, called Cassini's division, and
+\PageSep{354}
+to the faint internal ring, through which the
+edge of the planet is visible.
+
+The scale on which the whole system is constructed
+is best seen in a diagram of concentric
+circles, showing the limits of the planet's body
+and of the successive rings. Such a diagram,
+with explanatory notes, is given in~\fig{42}.
+
+An explanation of the outermost circle, called
+\emph{Roche's limit}, will be given later. The following
+are the dimensions of the system:---
+\begin{center}
+\begin{tabular}{l>{\qquad}r}
+Equatorial diameter of planet & $73,000$ miles \\
+Interior diameter of dark ring & $93,000$ \Ditto{miles} \\
+Interior diameter of bright rings & $111,000$ \Ditto{miles} \\
+Exterior diameter of bright rings & $169,000$ \Ditto{miles}
+\end{tabular}
+\end{center}
+We may also remark that the radius of the
+limit of the rings is $2.38$~times the mean radius
+of the planet, whilst Roche's limit is $2.44$~such
+radii. The greatest thickness of the ring is uncertain,
+but it seems probable that it does not
+exceed $200$~or $300$~miles.
+
+The pictorial interest, as we may call it, of all
+this wonderful combination is obvious, but our
+curiosity is further stimulated when we reflect on
+the difficulty of reconciling the existence of this
+strange satellite with what we know of our own
+planet and of other celestial bodies.
+
+It may be admitted that no disturbance to our
+ordinary way of life would take place if Saturn's
+\index{Saturn!description and picture|)}%
+rings were annihilated, but, as Clerk-Maxwell
+has remarked, ``from a purely scientific point of
+\PageSep{355}
+view, they become the most remarkable bodies in
+the heavens, except, perhaps, those still less \emph{useful}
+bodies---the spiral nebulæ. When we have
+actually seen that great arch swung over the
+equator of the planet without any visible connection,
+we cannot bring our minds to rest. We
+cannot simply admit that such is the case, and
+describe it as one of the observed facts of nature,
+not admitting or requiring explanation. We
+must either explain its motion on the principles
+of mechanics, or admit that, in Saturnian realms,
+there can be motion regulated by laws which we
+are unable to explain.''
+
+I must now revert to the subject of \Ref{Chapter}{XVIII}.\
+and show how the investigations, there
+explained, bear on the system of the planet. We
+then imagined a liquid satellite revolving in a
+circular orbit about a liquid planet, and supposed
+that each of these two masses moved so as always
+to present the same face to the other. It was
+pointed out that each body must be somewhat
+flattened by its rotation round an axis at right
+angles to the plane of the orbit, and that the
+tidal attraction of each must deform the other.
+In the application of this theory to the system of
+Saturn it is not necessary to consider further the
+tidal action of the satellite on the planet, and we
+must concentrate our attention on the action of
+the planet on the satellite. We have found reason
+to suppose that the earth once raised enormous
+\PageSep{356}
+\index{Saturn!theory of ring|(}%
+tides in the moon, when her body was
+molten, and any planet must act in the same way
+on its satellite. When, as we now suppose, the
+satellite moves so as always to present the same
+face to the planet, the tide is fixed and degenerates
+into a permanent distortion of the equator
+of the satellite into an elliptic shape. If the
+satellite is very small compared with its planet,
+and if it is gradually brought closer and closer
+to the planet, the tide-generating force, which
+varies inversely as the cube of the distance, increases
+with great rapidity, and we shall find the
+satellite to assume a more and more elongated
+shape. When the satellite is not excessively
+small, the two bodies may be brought together
+until they actually touch, and form the hour-glass
+figure exhibited in \fig{39}, \PageRef{p.}{329}.
+
+The general question of the limiting proximity
+of a liquid planet and satellite which just insures
+stability is as yet unsolved. But it has been
+proved that there is one case in which instability
+sets in. Édouard Roche has shown that this approach
+\index{Roche, E.!theory of limit and Saturn's ring|(}%
+up to contact is not possible when the
+satellite is very small, for at a certain distance
+the tidal distortion of a small satellite becomes
+so extreme that it can no longer subsist as a
+single mass of fluid. He also calculated the
+form of the satellite when it is elongated as much
+as possible. \Fig{43} represents the satellite in
+its limiting form. We must suppose the planet
+\PageSep{357}
+about which it revolves to be a large globe, with
+its centre lying on the prolongation of the longest
+axis of the egg-like body in the direction
+of~$E$. As it revolves, the longest axis of the satellite
+always points straight towards its planet.
+The egg, though not strictly circular in girth, is
+\Figure[0.7]{43}{Roche's Figure of a Satellite when elongated
+to the utmost}{png}
+very nearly so. Thus another section at right
+angles to this one would be of nearly the same
+shape. One diameter of the girth is in fact only
+longer than the other by a seventeenth part.
+The shortest of the three axes of the slightly flattened
+egg is at right angles to the plane of the
+orbit in which the satellite revolves. The longest
+axis of the body is nearly twice as long as
+either of the two shorter ones; for if we take
+the longest as~$\DPchg{1000}{1,000}$, the other two would be $496$
+and~$469$. \Fig{43} represents a section through
+the two axes equal respectively to~$\DPchg{1000}{1,000}$ and to~$469$,
+so that we are here supposed to be looking
+at the satellite's orbit edgewise.
+\PageSep{358}
+
+But, as I have said, Roche determined not
+\index{Earth and moon!Roche's limit for}%
+only the shape of the satellite when thus elongated
+to the utmost possible extent, but also in
+its nearness to the planet, and he proved that if
+the planet and satellite be formed of matter of
+the same density, the centre of such a satellite
+must be at a distance from the planet's centre of
+$2\frac{11}{25}$~of the planet's radius. This distance of $2\frac{11}{25}$
+or $2.44$~of a planet's radius I call Roche's limit
+for that planet. The meaning of this is that inside
+of a circle drawn around a planet at a distance
+so proportionate to its radius no small
+satellite can circulate; the reason being that if
+a lump of matter were started to revolve about
+the planet inside of that circle, it would be torn
+to pieces under the action of the forces we have
+been considering. It is true that if the lump of
+matter were so small as to be more properly described
+as a stone than as a satellite, then the
+cohesive force of stone might be strong enough
+to resist the disruptive force. But the size for
+which cohesion is sufficient to hold a mass of
+matter together is small compared with the
+smallest satellite.
+
+I have said that Roche's limit as evaluated at
+$2.44$~radii is dependent on the assumption of
+equal densities in the satellite and planet. If
+the planet be denser than the satellite, Roche's
+limit is a larger multiple of the planet's radius,
+and if it be less dense the multiple is smaller.
+\PageSep{359}
+But the variation of distance is not great for
+considerable variations in the relative densities
+of the two bodies, the law being that the~$2.44$
+must be multiplied by the cube root of the ratio
+of the density of the planet to that of the satellite.
+If for example the planet be on the average
+of its whole volume twice as dense as the
+satellite, the limit is only augmented from $2.44$
+to $3$~times the planet's radius; and if it be half
+as dense, the $2.44$ is depressed to~$1.94$. Thus
+the variation of density of the planet from a
+half to twice that of the planet---that is to
+say, the multiplication of the smaller density by
+four---only changes Roche's limit from $2$ to $3$~radii.
+It follows from this that, within pretty
+wide limits of variation of relative densities,
+Roche's limit changes but little.
+
+The only relative density of planet and satellite
+that we know with accuracy is that of the
+earth and moon. Now the earth is more dense
+than the moon in the proportion of $8$~to~$5$; hence
+Roche's limit for the earth is the cube root of~$\frac{8}{5}$
+multiplied by~$2.44$, that is to say, it is $2.86$~times
+the earth's radius. It follows that if the moon
+were to revolve at a distance of less than $2.86$~radii,
+or $11,000$~miles, she would be torn to pieces
+by the earth's tidal force.
+
+If this result be compared with the conclusions
+drawn from the theory of tidal friction, it follows
+that at the earliest stage to which the moon was
+\PageSep{360}
+\index{Saturn!Roche's limit for}%
+traced, she could not have existed in her present
+form, but the matter which is now consolidated
+in the form of a satellite must then have been a
+mere swarm of loose fragments. Such fragments,
+if concentrated in one part of the orbit, would
+be nearly as efficient in generating tides in the
+planet as though they were agglomerated in the
+form of a satellite. Accordingly the action of
+tidal friction does not necessitate the agglomeration
+of the satellite. The origin and earliest history
+of the moon must always remain highly
+speculative, and it seems fruitless to formulate
+exact theories on the subject.\footnote
+ {Mr.~Nolan has criticised the theory of tidal friction from
+\index{Nolan, criticism of tidal theory of moon's origin}%
+ this point of view (\Title{Genesis of the Moon}, Melbourne, 1885; also
+ \Title{Nature}, Feb.~18 and July~29, 1886).}
+
+When we apply this reasoning to the other
+planets, exact data are wanting. The planet
+Mars resembles the earth in so many respects
+\index{Mars!Roche's limit}%
+that it is reasonable to suppose that there is much
+the same relationship between the densities of
+the planet and satellites as with us. As with the
+case of the earth and moon, this would bring
+Roche's limit to $2.86$~times the planet's radius.
+The satellite Phobos, however, revolves at a
+distance of $2.75$~radii of Mars; hence we are
+bound to suppose that the density of Phobos is
+a very little more nearly equal to that of Mars
+than in the case of the moon and earth; if
+it were not so, Phobos would be disrupted by
+\PageSep{361}
+tidal action. How interesting it will be if future
+generations shall cease to see the satellite Phobos,
+for they will then conclude that Phobos has been
+drawn within the charmed circle, and has been
+broken to pieces.
+
+In considering the planets Jupiter and Saturn,
+\index{Jupiter!Roche's limit for}%
+we are deprived of the indications which are useful
+in the case of Mars. The satellites are probably
+solid, and these planets are known to have
+a low mean density. Hence it is probable that
+Roche's limit is a somewhat smaller multiple than
+$2.44$~of the radii of Jupiter and Saturn. The
+only satellite which is in danger is the innermost
+and recently discovered satellite of Jupiter, which
+revolves at $2.6$~times the planet's mean radius,
+for with the same ratio of densities as obtains
+here the satellite would be broken up. This confirms
+the conclusion that the mean density of
+Jupiter is at least not greater than that of the
+satellite.
+
+We are also ignorant of the relative densities
+of Saturn and its satellites, and so in the figure
+Roche's limit is placed at $2.44$~times the planet's
+radius, corresponding to equal densities. But
+the density of the planet is very small, and therefore
+the limit is almost certainly slightly nearer
+to the planet than is shown.
+
+This system affords the only known instance
+where matter is clearly visible circulating round
+an attractive centre at a distance certainly less
+\PageSep{362}
+than the theoretical limit, and the belief seems
+justified that Saturn's rings consist of dust and
+fragments.
+
+Although Roche himself dismissed this matter
+in one or two sentences, he saw the full bearing
+of his remarks, and to do him justice we should
+date from~1848 the proof that Saturn's rings
+consist of meteoric stones.
+
+The theoretical limit lies just outside the limit
+of the rings, but we may suspect that the relative
+densities of the planet and satellite are such that
+the limit should be displaced to a distance just
+inside of the outer edge of the ring, because any
+solid satellite would almost necessarily have a
+mean density greater than that of the planet.
+
+Although Roche's paper was published about
+fifty years ago, it has only recently been mentioned
+in text-books and general treatises. Indeed,
+it has been stated that Bond was the first
+in modern times to suggest the meteoric constitution
+of the rings. His suggestion, based on
+telescopic evidence, was however made in~1851.
+
+\TB
+
+And now to explain how a Cambridge mathematician
+to whom reference was made above, in
+ignorance of Roche's work of nine years before,
+\index{Roche, E.!theory of limit and Saturn's ring|)}%
+arrived at the same conclusion. In~1857, Clerk-Maxwell,
+one of the most brilliant men of science
+who have taught in the University of Cambridge,
+and whose early death we still deplore, attacked
+\PageSep{363}
+\index{Instability!of Saturn's ring}%
+the problem of Saturn's rings in a celebrated
+essay, which gained for him what is called the
+Adams prize. Laplace had early in the century
+considered the theory that the ring is solid, and
+Maxwell first took up the question of the motion
+\index{Maxwell on Saturn's ring|(}%
+of such a solid ring at the point where it had
+been left. He determined what amount of
+weighting at one point of a solid uniform ring is
+necessary to insure its steady motion round the
+planet. He found that there must be a mass
+attached to the circumference of the ring weighing
+$4\frac{1}{2}$~times as much as the ring itself. In fact,
+the system becomes a satellite with a light ring
+attached to it.
+
+``As there is no appearance,'' he says, ``about
+the rings justifying a belief in so great an irregularity,
+the theory of the solidity of the rings
+becomes very improbable. When we come to
+consider the additional difficulty of the tendency
+of the fluid or loose parts of the ring to accumulate
+at the thicker parts, and thus to destroy that
+nice adjustment of the load on which the stability
+depends, we have another powerful argument
+against solidity. And when we consider the immense
+size of the rings and their comparative
+thinness, the absurdity of treating them as rigid
+bodies becomes self-evident. An iron ring of
+such a size would be not only plastic, but semi-fluid,
+under the forces which it would experience,
+and we have no reason to believe these rings to
+\PageSep{364}
+\index{Instability!of Saturn's ring}%
+be artificially strengthened with any material
+unknown on this earth.''
+
+The hypothesis of solidity being condemned,
+Maxwell proceeds to suppose that the ring is
+composed of a number of equal small satellites.
+This is a step towards the hypothesis of an indefinite
+number of meteorites of all sizes. The
+consideration of the motion of these equal satellites
+affords a problem of immense difficulty, for
+each satellite is attracted by all the others and
+by the planet, and they are all in motion.
+
+If they were arranged in a circle round the
+planet at equal distances, they might continue to
+revolve round the planet, provided that each
+satellite remained in its place with mathematical
+exactness. Let us consider that the proper place
+of each satellite is at the ends of the spokes of
+a revolving wheel, and then let us suppose that
+none of them is exactly in its place, some being
+a little too far advanced, some a little behind,
+some too near and some too far from the centre
+of the wheel---that is to say, from the planet---then
+we want to know whether they will swing
+to and fro in the neighborhood of their places,
+or will get further and further from their places,
+and whether the ring will end in confusion.
+
+Maxwell treated this problem with consummate
+skill, and showed that if the satellites were
+not too large, confusion would not ensue, but
+each satellite would oscillate about its proper
+place.
+\PageSep{365}
+\index{Stability!of Saturn's ring}%
+
+At any moment there are places where the
+satellites are crowded and others where they are
+spaced out, and he showed that the places of
+crowding and of spacing out will travel round
+the ring at a different speed from that with
+which the ring as a whole revolves. In other
+words, waves of condensation and of rarefaction
+are propagated round the ring as it rotates.
+
+He constructed a model, now in the laboratory
+at Cambridge, to exhibit these movements; it is
+pretty to observe the changes of the shape of the
+ring and of the crowding of the model satellites
+as they revolve.
+
+I cannot sum up the general conclusions at
+which Maxwell arrived better than by quoting
+his own words.
+
+In the summary of his paper he says:---
+
+``If the satellites are unequal, the propagation
+of waves will no longer be regular, but the disturbances
+of the ring will in this, as in the
+former case, produce only waves, and not growing
+confusion. Supposing the ring to consist,
+not of a single row of large-satellites, but of a
+cloud of evenly distributed unconnected particles,
+we found that such a cloud must have a
+very small density in order to be permanent, and
+that this is inconsistent with its outer and inner
+parts moving with the same angular velocity.
+Supposing the ring to be fluid and continuous,
+we found that it will necessarily be broken up
+into small portions.
+\PageSep{366}
+\index{Stability!of Saturn's ring}%
+
+``We conclude, therefore, that the rings must
+consist of disconnected particles; these may be
+either solid or liquid, but they must be independent.
+The entire system of rings must therefore
+consist either of a series of many concentric
+rings, each moving with its own velocity, and
+having its own system of waves, or else of a confused
+multitude of revolving particles, not arranged
+in rings, and continually coming into
+collision with each other.
+
+``Taking the first case, we found that in an
+indefinite number of possible cases the mutual
+perturbation of two rings, stable in themselves,
+might mount up in time to a destructive magnitude,
+and that such cases must continually occur
+in an extensive system like that of Saturn, the
+only retarding cause being the possible irregularity
+of the rings.
+
+``The result of long-continued disturbance
+was found to be the spreading out of the rings
+in breadth, the outer rings pressing outward,
+while the inner rings press inward.
+
+``The final result, therefore, of the mechanical
+theory is, that the only system of rings which
+can exist is one composed of an indefinite number
+of unconnected particles, revolving round the
+planet with different velocities according to their
+respective distances. These particles may be
+arranged in a series of narrow rings, or they may
+move through each other irregularly. In the
+\PageSep{367}
+\index{Keeler, spectroscopic examination of Saturn's ring|(}%
+first case the destruction of the system will be
+very slow, in the second case it will be more
+rapid, but there may be a tendency towards an
+arrangement in narrow rings, which may retard
+the process.
+
+``We are not able to ascertain by observation
+the constitution of the two outer divisions of the
+system of rings, but the inner ring is certainly
+transparent, for the limb (\ie~edge) of Saturn
+has been observed through it. It is also certain,
+that though the space occupied by the ring is
+transparent, it is not through the material particles
+of it that Saturn was seen, for his limb was
+observed without distortion; which shows that
+there was no refraction, and therefore that the
+rays did not pass through a medium at all, but
+between the solid or liquid particles of which the
+ring is composed. Here then we have an optical
+argument in favor of the theory of independent
+particles as the material of the rings. The
+two outer rings may be of the same nature, but
+not so exceedingly rare that a ray of light can
+pass through their whole thickness without encountering
+one of the particles.''
+\index{Maxwell on Saturn's ring|)}%
+
+\TB
+
+The last link in the chain of evidence has been
+furnished by recent observations made in America.
+If it can be proved that every part of the
+apparently solid ring moves round the planet's
+centre at a different rate, and that the speed at
+\PageSep{368}
+\index{Meteoric constitution of Saturn's ring}%
+\index{Spectroscopic proof of rotation of Saturn's ring}%
+each part is appropriate at its distance from the
+centre, the conclusion is inevitable that the ring
+consists of scattered fragments.
+
+Every one must have noticed that when a
+train passes at full speed with the whistle blowing,
+there is an abrupt fall in the pitch of the
+note. This change of note is only apparent to
+the stationary listener, and is caused by the
+crowding together of the waves of sound as the
+train approaches, and by their spacing out as it
+recedes. The same thing is true of light-waves,
+and if we could imagine a colored light to pass
+us at an almost inconceivable velocity it would
+change in tint as it passed.\footnote
+ {This statement is strictly correct only of monochromatic
+ light. I might, in the subsequent argument, have introduced
+ the limitation that the moving body shall emit only monochromatic
+ light. The qualification would, however, only complicate
+ the statement, and thus render the displacement of the lines of
+ the spectrum less easily intelligible.}
+Now there are certain
+lines in the spectrum of sunlight, and the
+shifting of their positions affords an excessively
+delicate measure of a change which, when magnified
+enormously, would produce a change of
+tint. For example, the sun is a rotating body,
+and when we look at its disk one edge is approaching
+us and the other is receding. The
+two edges are infinitesimally of different colors,
+and the change of tint is measurable by the displacement
+of the lines I have mentioned. In
+the same way Saturn's ring is illuminated by
+sunlight, and if different portions are moving at
+\PageSep{369}
+\index{Spectroscopic proof of rotation of Saturn's ring}%
+different velocities, those portions are infinitesimally
+of different colors. Now Professor Keeler,
+the present director of the Lick Observatory, has
+actually observed the reflected sunlight from the
+several parts of Saturn's ring, and he finds that
+the lines in the spectrum of the several parts
+are differently displaced. From measurement of
+these displacements he has concluded that every
+part of the ring moves at the same pace as if it
+were an independent satellite. The proof of the
+meteoric constitution of the ring is therefore
+\index{Meteoric constitution of Saturn's ring}%
+complete.
+
+It would be hard to find in science a more
+beautiful instance of arguments of the most
+diverse natures concentrating themselves on a
+definite and final conclusion.
+
+\begin{Authorities}
+Édouard Roche, \Title{La figure d'une masse fluide soumise à l'attraction
+\index{Roche, E.!stability of ellipsoid of}%
+d'un point éloigné}, ``Mém.\ Acad.\ de~Montpelier,'' vol.~i.\
+(Sciences), 1847--50.
+
+Maxwell, \Title{Stability of Saturn's Rings}, Macmillan, 1859.
+
+Keeler, \Title{Spectroscopic Proof of the Meteoric Constitution of
+Saturn's Rings}, ``Astrophysical Journal,'' May, 1895; see also
+\index{Keeler, spectroscopic examination of Saturn's ring|)}%
+\index{Saturn!theory of ring|)}%
+the same for June, 1895.
+
+Schwarzschild, \Title{Die Poincarésche Theorie des Gleichgewichts},
+\index{Schwarzschild!stability of Roche's ellipsoid}%
+``Annals of Munich Observatory,'' vol.~iii.\ 1896. He considers
+the stability of Roche's ellipsoid.
+\end{Authorities}
+\PageSep{370}
+%[Blank Page]
+\PageSep{371}
+\BackMatter
+\printindex
+\iffalse
+INDEX
+
+Abacus for reducing tidal observations#abacus, 217-220.
+
+Abbadie, tidal deflection of vertical#Abbadie, 143, 144.
+
+Aden, errors of tidal prediction at#Aden, 246.
+
+Adriatic, tide in#Adriatic, 186.
+
+Airy, Sir G. B.#Airy,
+ tides in rivers, 75;
+ attack on Laplace, 181;
+ cotidal chart, 188;
+ \Title{Tides and Waves}, 192.
+
+America, North, tide tables for#America, 222.
+
+Analysis, harmonic, of tide#analysis, 193-210.
+
+Andromeda, nebula in#Andromeda, 339.
+
+Annual and semi-annual tides#annual, 206.
+
+Arabian theories of tide#Arab, 77-79.
+
+Aristotle on tides#Aristotle, 81.
+
+Assyrian records of eclipses, 272.
+
+Atlantic, tide in#Atlantic, 186-188.
+
+Atmospheric pressure,
+ cause of seiches, 40;
+ distortion of soil by, 145, 146;
+ influence on tidal prediction, 242, 243.
+
+Atmospheric waves, Helmholtz on#Helmholtz, 48-51.
+
+Attraction,
+ of weight resting on elastic slab proportional to slope, 136, 137;
+ of tide calculated, 143.
+
+Baird, \Title{Manual for Tidal Observation}#Baird, 16.
+
+Bakhuyzen on tide due to variation of latitude#Bakhuyzen, 255, 256.
+
+Barnard, rotation of Jupiter's satellites#Barnard, 315.
+
+Barometric pressure. |see{Atmospheric pressure}. 0
+
+Becker, G. F., on Nebular Hypothesis#Becker, 334, 336-338.
+
+Bernoulli, Daniel, essay on tides#Bernoulli, 86, 88.
+
+Bertelli on Italian seismology#Bertelli, 126, 127.
+
+Bifilar. |see{Pendulum}. 0
+
+Borgen@Börgen, method of reducing tidal observations#Börgen, 217.
+
+Bond, discovery of inner ring of Saturn#Bond, 352.
+
+Bore,
+ definition, 59;
+ bore-shelter, 63;
+ diagram of rise in Tsien-Tang, 66;
+ pictures, 67;
+ rivers where found, 71;
+ causes, 72;
+ Chinese superstition, 68-70.
+
+Browne, E. G., Arabian theories of tide#Browne, 77-79.
+
+Cambridge, experiments with bifilar pendulum at#bifilar, 115-125.
+
+Canal,
+ theory of tide wave in, 165-167;
+ critical depth, 163-165;
+ tides in ocean partitioned into canals, 175;
+ canal in high latitude, 174-176.
+
+Capillarity of liquids, and Plateau's experiment#Plateau, 316-318.
+
+Cassini, discovery of division in Saturn's rings#Cassini, 352.
+
+Castel, Father, ridiculed by Voltaire#Castel, 295, 296.
+
+Cavalleri, essay on tides#Cavalleri, 86.
+
+Centripetal and centrifugal forces#Centripetal, 91-93.
+
+Chambers on possible existence of Martian satellites#Chambers, 296.
+\PageSep{372}
+
+Chandler, free nutation of earth, and variation of latitude#Chandler, 253-257.
+
+Chinese
+ superstition as to bore, 68-70;
+ theories of tide, 76, 77.
+
+Christie, A. S., tide due to variation of latitude#Christie, 255, 256.
+
+Constants, tidal, explained#constants, 195.
+
+Continents, trend of, possibly due to primeval tidal friction#continents, 308.
+
+Cotidal chart, 188;
+ for diurnal tide hitherto undetermined, 191, 192.
+
+Currents, tidal, in rivers#tidal current, 56.
+
+Curve, tide#tide curve,
+ irregularities in, 10-16;
+ at Bombay, 12;
+ partitioned into lunar time, 213.
+
+D'Abbadie. |see{Abbadie}. 0
+
+Darwin, G. H.#Darwin,
+ bifilar pendulum, 115-125;
+ harmonic analysis, 210;
+ tidal abacus, 217-220;
+ distortion of earth's surface by varying loads, 134-148;
+ rigidity of earth, 261, 262;
+ papers on tidal friction, 315;
+ hour-glass figure of rotating liquid, 328-332;
+ Jacobi's ellipsoid, 333;
+ evolution of satellites, 346.
+
+Darwin, Horace, bifilar pendulum#Horace, 115-125.
+
+Davis, method of presenting tide-generating force#Davis, 96, 97.
+
+Davison, history of bifilar and horizontal pendulums#Davison, 133.
+
+Dawes, discovery of inner ring of Saturn#Dawes, 352.
+
+Dawson coöperates in investigation of seiches#Dawson, 48.
+
+Day, change in length of, under tidal friction#day, 275, 276.
+
+Deflection of the vertical, 109-133;
+ experiments to measure, 115-125;
+ due to tide, 134-143.
+
+Deimos, a satellite of Mars#Deimos, 297.
+
+Denison, F. Napier, vibrations and seiches on lakes#Denison, 48-53.
+
+Density
+ of earth, law of internal, 302;
+ of planets determinable from their figures, 332, 333.
+
+Descartes, vortical theory of cosmogony#Descartes, 335.
+
+Dick, argument as to Martian satellites#Dick, 295.
+
+Dimple,
+ in soil, due to weight, 123;
+ form of, in elastic slab, 135.
+
+Distortion of soil
+ by weight, 123;
+ by varying loads, 134-148.
+
+Diurnal inequality
+ observed by Seleucus, 84, 85;
+ according to equilibrium theory, 156;
+ in Laplace's solution, 179;
+ in Atlantic, Pacific, and Indian Oceans, 180;
+ not shown in cotidal chart, 191;
+ in harmonic method, 205;
+ complicates prediction, 224, 225.
+
+Douglass, rotation of Jupiter's satellites#Douglass, 315.
+
+Dumb-bell nebula, description of photograph of#dumb-bell, 345.
+
+Dynamical theory of tide-wave, 163-181.
+
+Earth and moon#Earth,
+ diagram, 93;
+ rotation of, effects on tides, 177;
+ rigidity of, 256-260;
+ rotation retarded by tidal friction, 268;
+ figure of, 299;
+ adjustment of figure to suit change of rotation, 299-302;
+ internal density, 302;
+ probably once molten, 306;
+ distortion under primeval tidal friction, 307;
+ Roche's limit for, 358.
+
+Earthquakes,
+ a cause of seiches, 39;
+ microsisms and earth tremors, 125-127;
+ shock perceptible at great distance, 261.
+
+Ebb and flow defined#ebb, 56.
+
+Eccentricity of orbit
+ due to tidal friction, 313, 314;
+ theory of, in case of double stars, 342.
+\PageSep{373}
+
+Eclipses, ancient, and earth's rotation#eclipses, 272, 273.
+
+Ecliptic, obliquity of, due to tidal friction#ecliptic, 308-312.
+
+Eddies, tidal oscillation involves#eddies, 177.
+
+Ehlert, observation with horizontal pendulum#Ehlert, 132.
+
+Elastic distortion#elastic
+ of soil by weight, 123;
+ of earth by varying loads, 134-148;
+ calculation and illustration, 138-140;
+ by atmospheric pressure, 145-147.
+
+Elasticity of earth#elasticity, 254, 255.
+
+Elliptic tide, lunar#elliptic tide, 204.
+
+Ellipticity of earth's strata in excess for present rotation#ellipticity, 303, 304.
+
+Energy, tidal, utilization of#tidal energy, 73, 74.
+
+Equatorial canal, tide wave in#canal, 173.
+
+Equilibrium, figures of, of rotating liquid#equilibrium, 316-333.
+
+Equilibrium theory of tides#equilibrium, 149-162;
+ chart and law of tide, 151-153;
+ defects of, 160.
+
+Errors in tidal prediction#errors, 243-245.
+
+Establishment of port,
+ definition, 161, 162;
+ zero in equilibrium theory, 161;
+ shown in cotidal chart, 189.
+
+Estuary, annual meteorological tide in#estuary, 207, 208.
+
+Euler, essay on tides#Euler, 86.
+
+Europe, tides on coasts of#Europe, 188.
+
+Evolution of celestial systems, 334-346.
+
+Ferrel, tide-predicting instrument#Ferrel, 241.
+
+Figure of equilibrium
+ of ocean under tidal forces, 151-153;
+ of rotating liquid, 316-333.
+
+Figure of planets and their density, 332, 333.
+
+Fisher, Osmond, on molten interior of earth#Fisher, 262.
+
+Flow and ebb defined#ebb, 56.
+
+Forced oscillation,
+ principle of, 169, 170;
+ due to solar tide, possibly related to birth of moon, 282-284.
+
+Forced wave, explanation and contrast with free wave#forced wave, 164.
+
+Forces,
+ centripetal and centrifugal, 91-93;
+ tide-generating, 93-108;
+ numerical estimate, 109-111;
+ deflection of vertical by, 109-133;
+ figure of equilibrium under tidal, 151-153;
+ those of sun and moon compared, 156-158.
+
+Forel
+ on seiches, 17-38;
+ list of papers, 53, 54.
+
+Free oscillation contrasted with forced#free oscillation, 169, 170.
+
+Free wave, explanation and contrast with forced#free wave, 164.
+
+Friction of tides, 264-315.
+
+Galileo,
+ blames Kepler for his tidal theory, 85;
+ discovery of Jupiter's satellites, 291;
+ Saturn's ring, 350.
+
+Gauge, tide#tide gauge,
+ description of, 6-11;
+ site for, 14.
+
+Geneva,
+ seiches in lake, 17-28;
+ model of lake, 28.
+
+Geological evidence of earth's plasticity#plasticity, 300;
+ as to retardation of earth's rotation, 304-306.
+
+German method of reducing tidal observations#German, 217.
+
+Giles on Chinese theories of the tide#Giles, 76, 77.
+
+Gravity, variation according to latitude#gravity, 302, 303, 332.
+
+Greek
+ theory and description of tides, 81-85;
+ records of ancient eclipses, 272.
+
+Gulliver@\Title{Gulliver's Travels}, satire on mathematics#Gulliver, 292-295.
+
+Hall, Asaph, discovery of Martian satellites#Hall, 290-298.
+
+Hangchow, the bore at#Hangchow, 60-70.
+\PageSep{374}
+
+Harmonic analysis
+ initiated by Lord Kelvin, 87;
+ account of, 193-210.
+
+Height of tide#height
+ due to ideal satellite, 198;
+ at Portsmouth and at Aden, 225;
+ reduced by elastic yielding of earth, 259.
+
+Helmholtz
+ on atmospheric waves, 48-51;
+ on rotation of the moon, 286.
+
+Herschel, observations of twin nebulæ#Herschel, 344.
+
+High water
+ under moon in equilibrium theory, 160;
+ position in shallow and deep canals in dynamical theory, 171, 172.
+
+History
+ of tidal theories, 76-88;
+ of earth and moon, 278-286, 308-313.
+
+Hopkins on rigidity of earth#Hopkins, 258, 259.
+
+Horizontal tide-generating force, 107.
+
+Horizontal tide-generating force |see{also Pendulum}. 0
+
+Hough, S. S.#Hough,
+ frictional extinction of waves, 47;
+ dynamical solution of tidal problem, 181;
+ rigidity of earth, 254;
+ Chandler's nutation, 262.
+
+Hugli, bore on the#Hugli, 71.
+
+Huyghens, discovery of Saturn's ring#Huyghens, 351.
+
+Icelandic theory of tides, 79, 80.
+
+Indian Survey,
+ method of reducing tidal observations, 216, 217;
+ tide tables, 222.
+
+Instability,
+ nature of dynamical, and initial of moon's motion, 280-282;
+ of Saturn's ring, 363, 364.
+
+Interval from moon's transit to high water
+ in case of ideal satellite, 198;
+ at Portsmouth and at Aden, 225.
+
+Italian investigations in seismology, 125-130.
+
+Jacobi, figure of equilibrium of rotating liquid#Jacobi, 322-324.
+
+Japan, frequency of earthquakes#Japan, 130, 131.
+
+Jupiter,
+ satellites constantly face planet, 298;
+ figure and law of internal density, 333;
+ Roche's limit for, 361.
+
+Kant,
+ rotation of moon, 286;
+ nebular hypothesis, 334-339.
+
+Keeler, spectroscopic examination of Saturn's ring#Keeler, 367-369.
+
+Kelvin, Lord#Kelvin,
+ initiates harmonic analysis, 87, 199;
+ calculation of tidal attraction, 143;
+ tide predicting machine, 233;
+ rigidity of earth, 257-260;
+ denies adjustment of earth's figure to changed rotation, 301;
+ on geological time, 315.
+
+Kepler,
+ ideas concerning tides, 85, 86;
+ argument respecting Martian satellites, 291, 292.
+
+Krüger, figures of equilibrium of liquid#Krüger, 333.
+
+Lakes,
+ seiches in, 17-54;
+ mode of rocking in seiches, 24, 25;
+ vibrations, 41-53;
+ tides in, 182-185.
+
+Lamb, H., presentation of Laplace's theory#Lamb, 181.
+
+Laplace,
+ theory of tides, 86-88, 177-180;
+ on rotation of moon, 286, 287;
+ nebular hypothesis, 335-337.
+
+Lardner, possibility of Martian satellites#Lardner, 295.
+
+Latitude,
+ tidal wave in canal in high, 174-176;
+ periodic variations of, 251-256.
+
+Lege@Légé, constructor of tide-predicting machine#Légé, 233.
+
+Level of sea affected by atmospheric pressure#atmospheric pressure, 146.
+
+Limnimeter, a form of tide gauge#limnimeter, 24.
+
+Lowell, P., on rotations of Venus and Mercury#Lowell, 298, 299, 315.
+
+Low water. |see{High water}.
+\PageSep{375}
+
+Lubbock, Sir J., senior, on tides#Lubbock, 87.
+
+Lunar
+ tide-generating force compared with solar, 156-158;
+ tide, principal, 201;
+ elliptic tide, 204;
+ time, 213.
+
+Machine, tide-predicting#tide-predicting, 233, 241.
+
+Mackerel sky, evidence of air-waves#mackerel, 49.
+
+Maclaurin,
+ essay on tides, 86;
+ figure of equilibrium of rotating liquid, 322-324.
+
+Magnússon on Icelandic theories of tides#Magnússon, 79, 80.
+
+Marco Polo, resident of Hangchow#Marco, 70.
+
+Mars,
+ discovery of satellites, 290-298;
+ Roche's limit, 360.
+
+Maxwell on Saturn's ring#Maxwell, 363-367.
+
+Mediterranean Sea, tides in#Mediterranean, 185, 186.
+
+Mercury, rotation of#Mercury, 298, 299.
+
+Meteoric constitution of Saturn's ring#meteoric, 368, 369.
+
+Meteorological
+ tides, 206, 207;
+ conditions dependent on earth's rotation, 303.
+
+Microphone as a seismological instrument#microphone, 128-130.
+
+Microsisms, minute earthquakes#microsisms, 125-127.
+
+Mills worked by the tide#mills, 74, 75.
+
+Milne on seismology#Milne, 125, 130.
+
+Month, change in, under tidal friction#month, 275-277.
+
+Moon and earth,
+ diagram, 93;
+ tide-generating force compared with sun's, 156-158;
+ tide due to ideal, moving in equator, 193, 194;
+ ideal satellites replacing actual, 199, 200;
+ tidal prediction by reference to transit, 224-230;
+ retardation of motion by tidal friction, 269, 270;
+ origin of, 282, 283;
+ rotation annulled by tidal friction and present libration, 286;
+ inequality in motion indicates internal density of earth, 302, 303;
+ eccentricity of orbit increased by tidal friction, 313, 314.
+
+Moore, Captain#Moore,
+ illustrations of bore, 67;
+ survey of Tsien-Tang-Kiang, 60-70.
+
+Neap and spring tides#neap
+ in equilibrium theory, 159;
+ represented by principal lunar and solar tides, 204.
+
+Nebula in Andromeda#Andromeda, 339.
+
+Nebulae@Nebulæ, description of various#Nebulæ, 345.
+
+Nebular hypothesis, 334-339.
+
+Newcomb, S., theoretical explanation of Chandler's nutation#Newcomb, 254.
+
+Newton,
+ founder of tidal theory, 86;
+ theory of tide in equatorial canal, 172.
+
+Nolan, criticism of tidal theory of moon's origin#Nolan, 360.
+
+Nutation,
+ value of, indicates internal density of earth, 303;
+ Chandler's, 251-256.
+
+Obliquity of ecliptic, effects of tidal friction on#obliquity, 310-312.
+
+Observation,
+ methods of tidal, 6-14;
+ reduction of tidal, 211-220.
+
+Orbit
+ of moon and earth, 93-95;
+ of double stars, very eccentric, 313.
+
+Pacific Ocean, tide in, affects Atlantic#Pacific, 186, 187.
+
+Partial tides in harmonic method#partial tides, 199.
+
+Paschwitz, von Rebeur#Paschwitz,
+ on horizontal pendulum, 130-132;
+ tidal deflection of vertical at Wilhelmshaven, 144.
+
+Pendulum,
+ curves traced by, under tidal force, 111, 112;
+ bifilar, 115-125;
+ as seismological instrument, 126, 127;
+ horizontal, 130-132.
+\PageSep{376}
+
+Petitcodiac, bore in the#Petitcodiac, 71.
+
+Phobos, a satellite of Mars#Phobos, 297.
+
+Planetary figure of equilibrium of rotating liquid#planetary figure, 322.
+
+Planets,
+ rotation of some, annulled by tidal friction, 298;
+ figures and internal densities, 332, 333.
+
+Plasticity of earth under change of rotation#plasticity, 300-302.
+
+Plateau, experiment on figure of rotating globule#Plateau, 316-319.
+
+Plemyrameter, observation of seiches with#plemyrameter, 19-22.
+
+Poincaré,
+ law of interchange of stability, 326, 327;
+ figure of rotating liquid, 325, 327.
+
+Polibius on tides at Cadiz#Polibus, 83.
+
+Portsmouth, table of errors in tidal predictions#Portsmouth, 244.
+
+Posidonius on tides#Posidonius, 81-84.
+
+Precession, value of, indicates internal density of earth#internal density, 303.
+
+Predicting machine for tides, 233-241;
+ Ferrel's, 241.
+
+Prediction of tide,
+ due to ideal satellite, 200;
+ example at Aden, 226-230;
+ method of computing, 230-233;
+ errors in, 242-250.
+
+Pressure of atmosphere, elastic distortion of soil by#distortion, 145, 146.
+
+Principle of forced oscillations#forced oscillations, 169, 170.
+
+Rebeur. |see{Paschwitz}. 0
+
+Reduction of tidal observations, 211-220.
+
+Retardation of earth's rotation, 268.
+
+Rigidity of earth#rigidity, 256-260.
+
+Ripple mark in sand preserved in geological strata#ripple mark, 305.
+
+Rivers,
+ tide wave in, 55-59;
+ Airy on tide in, 75;
+ annual meteorological tide in, 206.
+
+Roberts, E., the tide-predicting machine#Roberts, 233.
+
+Roberts, I., photograph of nebula in Andromeda#Roberts, 339.
+
+Roche, E.#Roche,
+ ellipticity of internal strata of earth, 303;
+ theory of limit and Saturn's ring, 356-362;
+ stability of ellipsoid of, 369.
+
+Roman description of tides#Roman, 81-85.
+
+Rossi on Italian seismology#Rossi, 128-130.
+
+Rotating liquid, figures of equilibrium#rotating liquid, 316-333.
+
+Rotation
+ of earth involved in tidal problem, 177;
+ retarded by tidal friction, 268;
+ of moon annulled by tidal friction, 286;
+ of Mercury, Venus, and satellites of Jupiter and Saturn annulled by tidal friction, 298.
+
+Russell, observation of seiches in New South Wales#Russell, 47.
+
+Saint@St.\ Vénant on flow of solids#Vénant, 313.
+
+Satellites,
+ tide due to single equatorial, 195, 196;
+ ideal replacing sun and moon in harmonic analysis, 199, 200;
+ discovery of those of Mars, 290-298;
+ rotation of those of Jupiter and Saturn annulled, 298;
+ distribution of, in solar system, 339-341.
+
+Saturn,
+ satellites always face the planet, 298;
+ law of density and figure, 332;
+ description and picture, 347-354;
+ theory of ring, 356-369;
+ Roche's limit for, 360.
+
+Schedule for reducing tidal observations, 215, 216.
+
+Schiaparelli on rotation of Venus and Mercury#Schiaparelli, 298, 315.
+
+Schwarzschild,
+ exposition of Poincaré's theory, 333;
+ stability of Roche's ellipsoid, 369.
+
+Sea,
+ vibrations of, 44, 45;
+ level affected by atmospheric pressure, 146.
+\PageSep{377}
+
+See, T. J. J.,
+ eccentricity of orbits of double stars, 313;
+ theory of evolution of double stars, 342-346.
+
+Seiches,
+ definition, 17;
+ records of, 21;
+ longitudinal and transverse, 25-27;
+ periods of, 27;
+ causes of, 39, 40.
+
+Seine, bore in the#Seine, 71.
+
+Seismology, 133.
+
+Seleucus, observation of tides of Indian Ocean#Seleucus, 84, 85.
+
+Semidiurnal tide
+ in equilibrium theory, 153-156;
+ in harmonic method, 201-204.
+
+Severn, bore in the#Severn, 71.
+
+Slope of soil
+ due to elastic distortion, 136;
+ calculation and illustration of, 138-140.
+
+Solar
+ tide-generating force compared with lunar, 156-158;
+ principal tide, 202;
+ possible effect of tide in assisting birth of moon, 284, 285;
+ system, nebular hypothesis as to origin of, 334-339;
+ system, distribution of satellites in, 339-341.
+
+Spectroscopic proof of rotation of Saturn's ring#spectroscop, 368, 369.
+
+Spring and neap tides
+ in equilibrium theory, 159;
+ represented by principal lunar and solar tides, 203.
+
+Stability,
+ nature of dynamical, 280, 281;
+ of figures of equilibrium, 322, 323;
+ of Saturn's ring, 365, 366.
+
+Stars,
+ double, eccentricity of orbits, 313;
+ theory of evolution, 342-346.
+
+Storms a cause of seiches#storms, 39, 40.
+
+Strabo on tides#Strabo, 81-85.
+
+Stupart coöperates in investigation of seiches#Stupart, 48.
+
+Sun,
+ tide-generating force of, compared with that of moon, 156-158;
+ ideal, replacing real sun in harmonic analysis, 201;
+ possible influence of, in assisting birth of moon, 284, 285.
+
+Surface tension of liquids#surface tension, 317, 318.
+
+Swift, satire on mathematicians#Swift, 292-295.
+
+Synthesis of partial tides for prediction#synthesis, 230-233.
+
+Tables, tide#tide tables, 221-241;
+ method of calculating, 230-241;
+ amount of error in, 246, 247.
+
+Thomson, Sir W. |see{Kelvin}. 0
+
+Tidal problem. |see{Laplace, Harmonic Analysis, etc.} 0
+
+Tide,
+ definition, 1-3;
+ general description, 4-6.
+
+Tide, |see{also other headings; \eg\ for tide-generating force, |see{Force}}. 0
+
+Time,
+ lunar, 213;
+ requisite for evolution of moon, 285.
+
+Tisserand, Roche's investigations as to earth's figure#Tisserand, 315.
+
+Tremors, earth#tremors, 125.
+
+Tresca on flow of solids#Tresca, 300.
+
+Tromometer, a seismological instrument#tromometer, 126, 127.
+
+Tsien-Tang-Kiang, the bore in#Tsien, 60-70.
+
+United States Coast Survey,
+ method of reducing tidal observations, 217;
+ tide tables of, 222.
+
+Variation of latitude, 251-256.
+
+Vaucher, record of a great seiche at Geneva#Vaucher, 17.
+
+Venus, rotation of#Venus, 298, 299.
+
+Vertical. |see{Deflection}. 0
+
+Vibration of lakes, 41-53.
+
+Voltaire, satire on mathematicians, and Martian satellites#Voltaire, 295, 296.
+
+Vortical motion in oceanic tides#vorticial, 177, 178.
+\PageSep{378}
+
+Waves
+ in deep and shallow water, 29;
+ speed of, 31;
+ composition of, 33-37;
+ in atmosphere, 48-50;
+ forced and free, 164;
+ of tide in equatorial canal, 173;
+ in canal in high latitude, 174-176;
+ propagated northward in Atlantic, 186-188.
+
+Wharton, Sir W. J., illustration of bore#Wharton, 69.
+
+Whewell
+ on tides, 87;
+ empirical construction of tide tables, 87-90;
+ on cotidal charts, 188, 189.
+
+Wind,
+ a cause of seiches, 39;
+ vibrations of lakes due to, 41, 42;
+ a cause of meteorological tides, 206;
+ perturbation of, in tidal prediction, 242, 243.
+
+Woodward on variation of latitude#Woodward, 262.
+
+Wright, Thomas, on a theory of cosmogony#Wright, 335.
+
+Wye, bore in the#Wye, 71.
+\fi
+\PageSep{379}
+\clearpage
+\null\vfill
+\begin{center}
+\textgoth{The Riverside Press}
+
+\footnotesize
+CAMBRIDGE, MASSACHUSETTS, U. S. A. \\
+ELECTROTYPED AND PRINTED BY \\
+H. O. HOUGHTON AND CO.
+\end{center}
+\vfill
+%%%%%%%%%%%%%%%%%%%%%%%%% GUTENBERG LICENSE %%%%%%%%%%%%%%%%%%%%%%%%%%
+\FlushRunningHeads
+\PGLicense
+\begin{PGtext}
+End of the Project Gutenberg EBook of The Tides and Kindred Phenomena in the
+Solar System, by Sir George Howard Darwin
+
+*** END OF THIS PROJECT GUTENBERG EBOOK TIDES AND KINDRED PHENOMENA ***
+
+***** This file should be named 38722-t.tex or 38722-t.zip *****
+This and all associated files of various formats will be found in:
+ http://www.gutenberg.org/3/8/7/2/38722/
+
+Produced by Andrew D. Hwang, Bryan Ness and the Online
+Distributed Proofreading Team at http://www.pgdp.net (This
+file was produced from images generously made available
+by The Internet Archive/Canadian Libraries)
+
+
+Updated editions will replace the previous one--the old editions
+will be renamed.
+
+Creating the works from public domain print editions means that no
+one owns a United States copyright in these works, so the Foundation
+(and you!) can copy and distribute it in the United States without
+permission and without paying copyright royalties. Special rules,
+set forth in the General Terms of Use part of this license, apply to
+copying and distributing Project Gutenberg-tm electronic works to
+protect the PROJECT GUTENBERG-tm concept and trademark. Project
+Gutenberg is a registered trademark, and may not be used if you
+charge for the eBooks, unless you receive specific permission. If you
+do not charge anything for copies of this eBook, complying with the
+rules is very easy. You may use this eBook for nearly any purpose
+such as creation of derivative works, reports, performances and
+research. They may be modified and printed and given away--you may do
+practically ANYTHING with public domain eBooks. Redistribution is
+subject to the trademark license, especially commercial
+redistribution.
+
+
+
+*** START: FULL LICENSE ***
+
+THE FULL PROJECT GUTENBERG LICENSE
+PLEASE READ THIS BEFORE YOU DISTRIBUTE OR USE THIS WORK
+
+To protect the Project Gutenberg-tm mission of promoting the free
+distribution of electronic works, by using or distributing this work
+(or any other work associated in any way with the phrase "Project
+Gutenberg"), you agree to comply with all the terms of the Full Project
+Gutenberg-tm License (available with this file or online at
+http://gutenberg.net/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.net
+
+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.net),
+you must, at no additional cost, fee or expense to the user, provide a
+copy, a means of exporting a copy, or a means of obtaining a copy upon
+request, of the work in its original "Plain Vanilla ASCII" or other
+form. Any alternate format must include the full Project Gutenberg-tm
+License as specified in paragraph 1.E.1.
+
+1.E.7. Do not charge a fee for access to, viewing, displaying,
+performing, copying or distributing any Project Gutenberg-tm works
+unless you comply with paragraph 1.E.8 or 1.E.9.
+
+1.E.8. You may charge a reasonable fee for copies of or providing
+access to or distributing Project Gutenberg-tm electronic works provided
+that
+
+- You pay a royalty fee of 20% of the gross profits you derive from
+ the use of Project Gutenberg-tm works calculated using the method
+ you already use to calculate your applicable taxes. The fee is
+ owed to the owner of the Project Gutenberg-tm trademark, but he
+ has agreed to donate royalties under this paragraph to the
+ Project Gutenberg Literary Archive Foundation. Royalty payments
+ must be paid within 60 days following each date on which you
+ prepare (or are legally required to prepare) your periodic tax
+ returns. Royalty payments should be clearly marked as such and
+ sent to the Project Gutenberg Literary Archive Foundation at the
+ address specified in Section 4, "Information about donations to
+ the Project Gutenberg Literary Archive Foundation."
+
+- You provide a full refund of any money paid by a user who notifies
+ you in writing (or by e-mail) within 30 days of receipt that s/he
+ does not agree to the terms of the full Project Gutenberg-tm
+ License. You must require such a user to return or
+ destroy all copies of the works possessed in a physical medium
+ and discontinue all use of and all access to other copies of
+ Project Gutenberg-tm works.
+
+- You provide, in accordance with paragraph 1.F.3, a full refund of any
+ money paid for a work or a replacement copy, if a defect in the
+ electronic work is discovered and reported to you within 90 days
+ of receipt of the work.
+
+- You comply with all other terms of this agreement for free
+ distribution of Project Gutenberg-tm works.
+
+1.E.9. If you wish to charge a fee or distribute a Project Gutenberg-tm
+electronic work or group of works on different terms than are set
+forth in this agreement, you must obtain permission in writing from
+both the Project Gutenberg Literary Archive Foundation and Michael
+Hart, the owner of the Project Gutenberg-tm trademark. Contact the
+Foundation as set forth in Section 3 below.
+
+1.F.
+
+1.F.1. Project Gutenberg volunteers and employees expend considerable
+effort to identify, do copyright research on, transcribe and proofread
+public domain works in creating the Project Gutenberg-tm
+collection. Despite these efforts, Project Gutenberg-tm electronic
+works, and the medium on which they may be stored, may contain
+"Defects," such as, but not limited to, incomplete, inaccurate or
+corrupt data, transcription errors, a copyright or other intellectual
+property infringement, a defective or damaged disk or other medium, a
+computer virus, or computer codes that damage or cannot be read by
+your equipment.
+
+1.F.2. LIMITED WARRANTY, DISCLAIMER OF DAMAGES - Except for the "Right
+of Replacement or Refund" described in paragraph 1.F.3, the Project
+Gutenberg Literary Archive Foundation, the owner of the Project
+Gutenberg-tm trademark, and any other party distributing a Project
+Gutenberg-tm electronic work under this agreement, disclaim all
+liability to you for damages, costs and expenses, including legal
+fees. YOU AGREE THAT YOU HAVE NO REMEDIES FOR NEGLIGENCE, STRICT
+LIABILITY, BREACH OF WARRANTY OR BREACH OF CONTRACT EXCEPT THOSE
+PROVIDED IN PARAGRAPH 1.F.3. YOU AGREE THAT THE FOUNDATION, THE
+TRADEMARK OWNER, AND ANY DISTRIBUTOR UNDER THIS AGREEMENT WILL NOT BE
+LIABLE TO YOU FOR ACTUAL, DIRECT, INDIRECT, CONSEQUENTIAL, PUNITIVE OR
+INCIDENTAL DAMAGES EVEN IF YOU GIVE NOTICE OF THE POSSIBILITY OF SUCH
+DAMAGE.
+
+1.F.3. LIMITED RIGHT OF REPLACEMENT OR REFUND - If you discover a
+defect in this electronic work within 90 days of receiving it, you can
+receive a refund of the money (if any) you paid for it by sending a
+written explanation to the person you received the work from. If you
+received the work on a physical medium, you must return the medium with
+your written explanation. The person or entity that provided you with
+the defective work may elect to provide a replacement copy in lieu of a
+refund. If you received the work electronically, the person or entity
+providing it to you may choose to give you a second opportunity to
+receive the work electronically in lieu of a refund. If the second copy
+is also defective, you may demand a refund in writing without further
+opportunities to fix the problem.
+
+1.F.4. Except for the limited right of replacement or refund set forth
+in paragraph 1.F.3, this work is provided to you 'AS-IS' WITH NO OTHER
+WARRANTIES OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO
+WARRANTIES OF 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 http://www.pglaf.org.
+
+
+Section 3. Information about the Project Gutenberg Literary Archive
+Foundation
+
+The Project Gutenberg Literary Archive Foundation is a non profit
+501(c)(3) educational corporation organized under the laws of the
+state of Mississippi and granted tax exempt status by the Internal
+Revenue Service. The Foundation's EIN or federal tax identification
+number is 64-6221541. Its 501(c)(3) letter is posted at
+http://pglaf.org/fundraising. Contributions to the Project Gutenberg
+Literary Archive Foundation are tax deductible to the full extent
+permitted by U.S. federal laws and your state's laws.
+
+The Foundation's principal office is located at 4557 Melan Dr. S.
+Fairbanks, AK, 99712., but its volunteers and employees are scattered
+throughout numerous locations. Its business office is located at
+809 North 1500 West, Salt Lake City, UT 84116, (801) 596-1887, email
+business@pglaf.org. Email contact links and up to date contact
+information can be found at the Foundation's web site and official
+page at http://pglaf.org
+
+For additional contact information:
+ Dr. Gregory B. Newby
+ Chief Executive and Director
+ gbnewby@pglaf.org
+
+
+Section 4. Information about Donations to the Project Gutenberg
+Literary Archive Foundation
+
+Project Gutenberg-tm depends upon and cannot survive without wide
+spread public support and donations to carry out its mission of
+increasing the number of public domain and licensed works that can be
+freely distributed in machine readable form accessible by the widest
+array of equipment including outdated equipment. Many small donations
+($1 to $5,000) are particularly important to maintaining tax exempt
+status with the IRS.
+
+The Foundation is committed to complying with the laws regulating
+charities and charitable donations in all 50 states of the United
+States. Compliance requirements are not uniform and it takes a
+considerable effort, much paperwork and many fees to meet and keep up
+with these requirements. We do not solicit donations in locations
+where we have not received written confirmation of compliance. To
+SEND DONATIONS or determine the status of compliance for any
+particular state visit http://pglaf.org
+
+While we cannot and do not solicit contributions from states where we
+have not met the solicitation requirements, we know of no prohibition
+against accepting unsolicited donations from donors in such states who
+approach us with offers to donate.
+
+International donations are gratefully accepted, but we cannot make
+any statements concerning tax treatment of donations received from
+outside the United States. U.S. laws alone swamp our small staff.
+
+Please check the Project Gutenberg Web pages for current donation
+methods and addresses. Donations are accepted in a number of other
+ways including including checks, online payments and credit card
+donations. To donate, please visit: http://pglaf.org/donate
+
+
+Section 5. General Information About Project Gutenberg-tm electronic
+works.
+
+Professor Michael S. Hart is the originator of the Project Gutenberg-tm
+concept of a library of electronic works that could be freely shared
+with anyone. For thirty years, he produced and distributed Project
+Gutenberg-tm eBooks with only a loose network of volunteer support.
+
+
+Project Gutenberg-tm eBooks are often created from several printed
+editions, all of which are confirmed as Public Domain in the U.S.
+unless a copyright notice is included. Thus, we do not necessarily
+keep eBooks in compliance with any particular paper edition.
+
+
+Most people start at our Web site which has the main PG search facility:
+
+ http://www.gutenberg.net
+
+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.
+\end{PGtext}
+
+% %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% %
+% %
+% End of the Project Gutenberg EBook of The Tides and Kindred Phenomena in the
+% Solar System, by Sir George Howard Darwin %
+% %
+% *** END OF THIS PROJECT GUTENBERG EBOOK TIDES AND KINDRED PHENOMENA *** %
+% %
+% ***** This file should be named 38722-t.tex or 38722-t.zip ***** %
+% This and all associated files of various formats will be found in: %
+% http://www.gutenberg.org/3/8/7/2/38722/ %
+% %
+% %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% %
+
+\end{document}
+###
+@ControlwordReplace = (
+ ['\\ie', 'i.e.'],
+ ['\\eg', 'e.g.'],
+ ['\\NB', 'N.B.'],
+ ['\\ieme', '^{e}'],
+ ['\\iere', '^{re}'],
+ ['\\AD', 'A.D.'],
+ ['\\AM', 'A.M.'],
+ ['\\PM', 'P.M.'],
+ ['\\Contents', 'CONTENTS'],
+ ['\\Illustrations', 'LIST OF ILLUSTRATIONS'],
+ ['\\Preface', 'PREFACE'],
+ ['\\printindex', 'INDEX']
+ );
+
+@ControlwordArguments = (
+ ['\\Chapter', 0, 0, '', '', 1, 1, 'Chapter ', '. ', 1, 1, '', ''],
+ ['\\TranscribersNote', 1, 0, '', ''],
+ ['\\Signature', 1, 1, '', ' ', 1, 1, '', ''],
+ ['\\ToCChap', 1, 1, '', ' ', 1, 1, '', ''],
+ ['\\ToCSect', 1, 1, '', ' ', 1, 0, '', ''],
+ ['\\LoFFig', 1, 1, '', ' ', 1, 1, '', '', 1, 0, '', ''],
+ ['\\Graphic', 0, 0, '', '', 1, 0, '', '', 1, 0, '', ''],
+ ['\\Figure', 0, 0, '', '', 1, 1, '<Figure ', '>', 1, 0, '', '', 1, 0, '', ''],
+ ['\\TallFig', 0, 0, '', '', 1, 1, '<Figure ', '>', 1, 0, '', '', 1, 0, '', ''],
+ ['\\begin{Authorities}', 0, 0, 'Authorities. ', ''],
+ ['\\Paper', 1, 1, '', ''],
+ ['\\Ref', 1, 1, '', ' ', 1, 1, '', ''],
+ ['\\PageLabel', 0, 0, '', '', 1, 0, '', ''],
+ ['\\PageRef', 1, 1, '', ' ', 1, 1, '', ''],
+ ['\\First', 1, 1, '', ''],
+ ['\\bar', 1, 1, '', ''],
+ ['\\figref', 1, 1, '', ''],
+ ['\\fig', 1, 1, 'fig. ', ''],
+ ['\\Fig', 1, 1, 'Fig. ', ''],
+ ['\\Ref', 1, 1, '', ' ', 1, 1, '', ''],
+ ['\\DPtypo', 1, 0, '', '', 1, 1, '', ''],
+ ['\\DPchg', 1, 0, '', '', 1, 1, '', ''],
+ ['\\DPnote', 1, 0, '', '']
+ );
+
+$PageSeparator = qr/^\\PageSep/;
+$CustomClean = 'print "\\nCustom cleaning in progress...";
+my $cline = 0;
+ while ($cline <= $#file) {
+ $file[$cline] =~ s/--------[^\n]*\n//; # strip page separators
+ $cline++
+ }
+ print "done\\n";';
+###
+This is pdfTeX, Version 3.1415926-1.40.10 (TeX Live 2009/Debian) (format=pdflatex 2011.9.6) 30 JAN 2012 21:37
+entering extended mode
+ %&-line parsing enabled.
+**38722-t.tex
+(./38722-t.tex
+LaTeX2e <2009/09/24>
+Babel <v3.8l> and hyphenation patterns for english, usenglishmax, dumylang, noh
+yphenation, farsi, arabic, croatian, bulgarian, ukrainian, russian, czech, slov
+ak, danish, dutch, finnish, french, basque, ngerman, german, german-x-2009-06-1
+9, ngerman-x-2009-06-19, ibycus, monogreek, greek, ancientgreek, hungarian, san
+skrit, italian, latin, latvian, lithuanian, mongolian2a, mongolian, bokmal, nyn
+orsk, romanian, irish, coptic, serbian, turkish, welsh, esperanto, uppersorbian
+, estonian, indonesian, interlingua, icelandic, kurmanji, slovenian, polish, po
+rtuguese, spanish, galician, catalan, swedish, ukenglish, pinyin, loaded.
+(/usr/share/texmf-texlive/tex/latex/base/book.cls
+Document Class: book 2007/10/19 v1.4h Standard LaTeX document class
+(/usr/share/texmf-texlive/tex/latex/base/bk12.clo
+File: bk12.clo 2007/10/19 v1.4h Standard LaTeX file (size option)
+)
+\c@part=\count79
+\c@chapter=\count80
+\c@section=\count81
+\c@subsection=\count82
+\c@subsubsection=\count83
+\c@paragraph=\count84
+\c@subparagraph=\count85
+\c@figure=\count86
+\c@table=\count87
+\abovecaptionskip=\skip41
+\belowcaptionskip=\skip42
+\bibindent=\dimen102
+) (/usr/share/texmf-texlive/tex/latex/base/inputenc.sty
+Package: inputenc 2008/03/30 v1.1d Input encoding file
+\inpenc@prehook=\toks14
+\inpenc@posthook=\toks15
+(/usr/share/texmf-texlive/tex/latex/base/latin1.def
+File: latin1.def 2008/03/30 v1.1d Input encoding file
+)) (/usr/share/texmf-texlive/tex/latex/base/ifthen.sty
+Package: ifthen 2001/05/26 v1.1c Standard LaTeX ifthen package (DPC)
+) (/usr/share/texmf-texlive/tex/latex/amsmath/amsmath.sty
+Package: amsmath 2000/07/18 v2.13 AMS math features
+\@mathmargin=\skip43
+For additional information on amsmath, use the `?' option.
+(/usr/share/texmf-texlive/tex/latex/amsmath/amstext.sty
+Package: amstext 2000/06/29 v2.01
+(/usr/share/texmf-texlive/tex/latex/amsmath/amsgen.sty
+File: amsgen.sty 1999/11/30 v2.0
+\@emptytoks=\toks16
+\ex@=\dimen103
+)) (/usr/share/texmf-texlive/tex/latex/amsmath/amsbsy.sty
+Package: amsbsy 1999/11/29 v1.2d
+\pmbraise@=\dimen104
+) (/usr/share/texmf-texlive/tex/latex/amsmath/amsopn.sty
+Package: amsopn 1999/12/14 v2.01 operator names
+)
+\inf@bad=\count88
+LaTeX Info: Redefining \frac on input line 211.
+\uproot@=\count89
+\leftroot@=\count90
+LaTeX Info: Redefining \overline on input line 307.
+\classnum@=\count91
+\DOTSCASE@=\count92
+LaTeX Info: Redefining \ldots on input line 379.
+LaTeX Info: Redefining \dots on input line 382.
+LaTeX Info: Redefining \cdots on input line 467.
+\Mathstrutbox@=\box26
+\strutbox@=\box27
+\big@size=\dimen105
+LaTeX Font Info: Redeclaring font encoding OML on input line 567.
+LaTeX Font Info: Redeclaring font encoding OMS on input line 568.
+\macc@depth=\count93
+\c@MaxMatrixCols=\count94
+\dotsspace@=\muskip10
+\c@parentequation=\count95
+\dspbrk@lvl=\count96
+\tag@help=\toks17
+\row@=\count97
+\column@=\count98
+\maxfields@=\count99
+\andhelp@=\toks18
+\eqnshift@=\dimen106
+\alignsep@=\dimen107
+\tagshift@=\dimen108
+\tagwidth@=\dimen109
+\totwidth@=\dimen110
+\lineht@=\dimen111
+\@envbody=\toks19
+\multlinegap=\skip44
+\multlinetaggap=\skip45
+\mathdisplay@stack=\toks20
+LaTeX Info: Redefining \[ on input line 2666.
+LaTeX Info: Redefining \] on input line 2667.
+) (/usr/share/texmf-texlive/tex/latex/amsfonts/amssymb.sty
+Package: amssymb 2009/06/22 v3.00
+(/usr/share/texmf-texlive/tex/latex/amsfonts/amsfonts.sty
+Package: amsfonts 2009/06/22 v3.00 Basic AMSFonts support
+\symAMSa=\mathgroup4
+\symAMSb=\mathgroup5
+LaTeX Font Info: Overwriting math alphabet `\mathfrak' in version `bold'
+(Font) U/euf/m/n --> U/euf/b/n on input line 96.
+)) (/usr/share/texmf-texlive/tex/latex/base/alltt.sty
+Package: alltt 1997/06/16 v2.0g defines alltt environment
+) (/usr/share/texmf-texlive/tex/latex/footmisc/footmisc.sty
+Package: footmisc 2009/09/15 v5.5a a miscellany of footnote facilities
+\FN@temptoken=\toks21
+\footnotemargin=\dimen112
+\c@pp@next@reset=\count100
+\c@@fnserial=\count101
+Package footmisc Info: Declaring symbol style bringhurst on input line 855.
+Package footmisc Info: Declaring symbol style chicago on input line 863.
+Package footmisc Info: Declaring symbol style wiley on input line 872.
+Package footmisc Info: Declaring symbol style lamport-robust on input line 883.
+
+Package footmisc Info: Declaring symbol style lamport* on input line 903.
+Package footmisc Info: Declaring symbol style lamport*-robust on input line 924
+.
+) (/usr/share/texmf-texlive/tex/latex/tools/multicol.sty
+Package: multicol 2008/12/05 v1.6h multicolumn formatting (FMi)
+\c@tracingmulticols=\count102
+\mult@box=\box28
+\multicol@leftmargin=\dimen113
+\c@unbalance=\count103
+\c@collectmore=\count104
+\doublecol@number=\count105
+\multicoltolerance=\count106
+\multicolpretolerance=\count107
+\full@width=\dimen114
+\page@free=\dimen115
+\premulticols=\dimen116
+\postmulticols=\dimen117
+\multicolsep=\skip46
+\multicolbaselineskip=\skip47
+\partial@page=\box29
+\last@line=\box30
+\mult@rightbox=\box31
+\mult@grightbox=\box32
+\mult@gfirstbox=\box33
+\mult@firstbox=\box34
+\@tempa=\box35
+\@tempa=\box36
+\@tempa=\box37
+\@tempa=\box38
+\@tempa=\box39
+\@tempa=\box40
+\@tempa=\box41
+\@tempa=\box42
+\@tempa=\box43
+\@tempa=\box44
+\@tempa=\box45
+\@tempa=\box46
+\@tempa=\box47
+\@tempa=\box48
+\@tempa=\box49
+\@tempa=\box50
+\@tempa=\box51
+\c@columnbadness=\count108
+\c@finalcolumnbadness=\count109
+\last@try=\dimen118
+\multicolovershoot=\dimen119
+\multicolundershoot=\dimen120
+\mult@nat@firstbox=\box52
+\colbreak@box=\box53
+) (/usr/share/texmf-texlive/tex/latex/base/makeidx.sty
+Package: makeidx 2000/03/29 v1.0m Standard LaTeX package
+) (/usr/share/texmf-texlive/tex/latex/tools/indentfirst.sty
+Package: indentfirst 1995/11/23 v1.03 Indent first paragraph (DPC)
+) (/usr/share/texmf-texlive/tex/latex/was/icomma.sty
+Package: icomma 2002/03/10 v2.0 (WaS)
+) (/usr/share/texmf-texlive/tex/latex/tools/calc.sty
+Package: calc 2007/08/22 v4.3 Infix arithmetic (KKT,FJ)
+\calc@Acount=\count110
+\calc@Bcount=\count111
+\calc@Adimen=\dimen121
+\calc@Bdimen=\dimen122
+\calc@Askip=\skip48
+\calc@Bskip=\skip49
+LaTeX Info: Redefining \setlength on input line 76.
+LaTeX Info: Redefining \addtolength on input line 77.
+\calc@Ccount=\count112
+\calc@Cskip=\skip50
+) (/usr/share/texmf-texlive/tex/latex/tools/array.sty
+Package: array 2008/09/09 v2.4c Tabular extension package (FMi)
+\col@sep=\dimen123
+\extrarowheight=\dimen124
+\NC@list=\toks22
+\extratabsurround=\skip51
+\backup@length=\skip52
+) (/usr/share/texmf-texlive/tex/latex/yfonts/yfonts.sty
+Package: yfonts 2003/01/08 v1.3 (WaS)
+) (/usr/share/texmf-texlive/tex/latex/fancyhdr/fancyhdr.sty
+\fancy@headwidth=\skip53
+\f@ncyO@elh=\skip54
+\f@ncyO@erh=\skip55
+\f@ncyO@olh=\skip56
+\f@ncyO@orh=\skip57
+\f@ncyO@elf=\skip58
+\f@ncyO@erf=\skip59
+\f@ncyO@olf=\skip60
+\f@ncyO@orf=\skip61
+) (/usr/share/texmf-texlive/tex/latex/graphics/graphicx.sty
+Package: graphicx 1999/02/16 v1.0f Enhanced LaTeX Graphics (DPC,SPQR)
+(/usr/share/texmf-texlive/tex/latex/graphics/keyval.sty
+Package: keyval 1999/03/16 v1.13 key=value parser (DPC)
+\KV@toks@=\toks23
+) (/usr/share/texmf-texlive/tex/latex/graphics/graphics.sty
+Package: graphics 2009/02/05 v1.0o Standard LaTeX Graphics (DPC,SPQR)
+(/usr/share/texmf-texlive/tex/latex/graphics/trig.sty
+Package: trig 1999/03/16 v1.09 sin cos tan (DPC)
+) (/etc/texmf/tex/latex/config/graphics.cfg
+File: graphics.cfg 2009/08/28 v1.8 graphics configuration of TeX Live
+)
+Package graphics Info: Driver file: pdftex.def on input line 91.
+(/usr/share/texmf-texlive/tex/latex/pdftex-def/pdftex.def
+File: pdftex.def 2009/08/25 v0.04m Graphics/color for pdfTeX
+\Gread@gobject=\count113
+))
+\Gin@req@height=\dimen125
+\Gin@req@width=\dimen126
+) (/usr/share/texmf-texlive/tex/latex/caption/caption.sty
+Package: caption 2009/10/09 v3.1k Customizing captions (AR)
+(/usr/share/texmf-texlive/tex/latex/caption/caption3.sty
+Package: caption3 2009/10/09 v3.1k caption3 kernel (AR)
+\captionmargin=\dimen127
+\captionmargin@=\dimen128
+\captionwidth=\dimen129
+\caption@indent=\dimen130
+\caption@parindent=\dimen131
+\caption@hangindent=\dimen132
+)
+\c@ContinuedFloat=\count114
+) (/usr/share/texmf-texlive/tex/latex/geometry/geometry.sty
+Package: geometry 2008/12/21 v4.2 Page Geometry
+(/usr/share/texmf-texlive/tex/generic/oberdiek/ifpdf.sty
+Package: ifpdf 2009/04/10 v2.0 Provides the ifpdf switch (HO)
+Package ifpdf Info: pdfTeX in pdf mode detected.
+) (/usr/share/texmf-texlive/tex/generic/oberdiek/ifvtex.sty
+Package: ifvtex 2008/11/04 v1.4 Switches for detecting VTeX and its modes (HO)
+Package ifvtex Info: VTeX not detected.
+)
+\Gm@cnth=\count115
+\Gm@cntv=\count116
+\c@Gm@tempcnt=\count117
+\Gm@bindingoffset=\dimen133
+\Gm@wd@mp=\dimen134
+\Gm@odd@mp=\dimen135
+\Gm@even@mp=\dimen136
+\Gm@dimlist=\toks24
+(/usr/share/texmf-texlive/tex/xelatex/xetexconfig/geometry.cfg)) (/usr/share/te
+xmf-texlive/tex/latex/hyperref/hyperref.sty
+Package: hyperref 2009/10/09 v6.79a Hypertext links for LaTeX
+(/usr/share/texmf-texlive/tex/generic/ifxetex/ifxetex.sty
+Package: ifxetex 2009/01/23 v0.5 Provides ifxetex conditional
+) (/usr/share/texmf-texlive/tex/latex/oberdiek/hycolor.sty
+Package: hycolor 2009/10/02 v1.5 Code for color options of hyperref/bookmark (H
+O)
+(/usr/share/texmf-texlive/tex/latex/oberdiek/xcolor-patch.sty
+Package: xcolor-patch 2009/10/02 xcolor patch
+))
+\@linkdim=\dimen137
+\Hy@linkcounter=\count118
+\Hy@pagecounter=\count119
+(/usr/share/texmf-texlive/tex/latex/hyperref/pd1enc.def
+File: pd1enc.def 2009/10/09 v6.79a Hyperref: PDFDocEncoding definition (HO)
+) (/usr/share/texmf-texlive/tex/generic/oberdiek/etexcmds.sty
+Package: etexcmds 2007/12/12 v1.2 Prefix for e-TeX command names (HO)
+(/usr/share/texmf-texlive/tex/generic/oberdiek/infwarerr.sty
+Package: infwarerr 2007/09/09 v1.2 Providing info/warning/message (HO)
+)
+Package etexcmds Info: Could not find \expanded.
+(etexcmds) That can mean that you are not using pdfTeX 1.50 or
+(etexcmds) that some package has redefined \expanded.
+(etexcmds) In the latter case, load this package earlier.
+) (/etc/texmf/tex/latex/config/hyperref.cfg
+File: hyperref.cfg 2002/06/06 v1.2 hyperref configuration of TeXLive
+) (/usr/share/texmf-texlive/tex/latex/oberdiek/kvoptions.sty
+Package: kvoptions 2009/08/13 v3.4 Keyval support for LaTeX options (HO)
+(/usr/share/texmf-texlive/tex/generic/oberdiek/kvsetkeys.sty
+Package: kvsetkeys 2009/07/30 v1.5 Key value parser with default handler suppor
+t (HO)
+))
+Package hyperref Info: Option `hyperfootnotes' set `false' on input line 2864.
+Package hyperref Info: Option `bookmarks' set `true' on input line 2864.
+Package hyperref Info: Option `linktocpage' set `false' on input line 2864.
+Package hyperref Info: Option `pdfdisplaydoctitle' set `true' on input line 286
+4.
+Package hyperref Info: Option `pdfpagelabels' set `true' on input line 2864.
+Package hyperref Info: Option `bookmarksopen' set `true' on input line 2864.
+Package hyperref Info: Option `colorlinks' set `true' on input line 2864.
+Package hyperref Info: Hyper figures OFF on input line 2975.
+Package hyperref Info: Link nesting OFF on input line 2980.
+Package hyperref Info: Hyper index ON on input line 2983.
+Package hyperref Info: Plain pages OFF on input line 2990.
+Package hyperref Info: Backreferencing OFF on input line 2995.
+Implicit mode ON; LaTeX internals redefined
+Package hyperref Info: Bookmarks ON on input line 3191.
+(/usr/share/texmf-texlive/tex/latex/ltxmisc/url.sty
+\Urlmuskip=\muskip11
+Package: url 2006/04/12 ver 3.3 Verb mode for urls, etc.
+)
+LaTeX Info: Redefining \url on input line 3428.
+(/usr/share/texmf-texlive/tex/generic/oberdiek/bitset.sty
+Package: bitset 2007/09/28 v1.0 Data type bit set (HO)
+(/usr/share/texmf-texlive/tex/generic/oberdiek/intcalc.sty
+Package: intcalc 2007/09/27 v1.1 Expandable integer calculations (HO)
+) (/usr/share/texmf-texlive/tex/generic/oberdiek/bigintcalc.sty
+Package: bigintcalc 2007/11/11 v1.1 Expandable big integer calculations (HO)
+(/usr/share/texmf-texlive/tex/generic/oberdiek/pdftexcmds.sty
+Package: pdftexcmds 2009/09/23 v0.6 LuaTeX support for pdfTeX utility functions
+ (HO)
+(/usr/share/texmf-texlive/tex/generic/oberdiek/ifluatex.sty
+Package: ifluatex 2009/04/17 v1.2 Provides the ifluatex switch (HO)
+Package ifluatex Info: LuaTeX not detected.
+) (/usr/share/texmf-texlive/tex/generic/oberdiek/ltxcmds.sty
+Package: ltxcmds 2009/08/05 v1.0 Some LaTeX kernel commands for general use (HO
+)
+)
+Package pdftexcmds Info: LuaTeX not detected.
+Package pdftexcmds Info: \pdf@primitive is available.
+Package pdftexcmds Info: \pdf@ifprimitive is available.
+)))
+\Fld@menulength=\count120
+\Field@Width=\dimen138
+\Fld@charsize=\dimen139
+\Field@toks=\toks25
+Package hyperref Info: Hyper figures OFF on input line 4377.
+Package hyperref Info: Link nesting OFF on input line 4382.
+Package hyperref Info: Hyper index ON on input line 4385.
+Package hyperref Info: backreferencing OFF on input line 4392.
+Package hyperref Info: Link coloring ON on input line 4395.
+Package hyperref Info: Link coloring with OCG OFF on input line 4402.
+Package hyperref Info: PDF/A mode OFF on input line 4407.
+(/usr/share/texmf-texlive/tex/generic/oberdiek/atbegshi.sty
+Package: atbegshi 2008/07/31 v1.9 At begin shipout hook (HO)
+)
+\Hy@abspage=\count121
+\c@Item=\count122
+)
+*hyperref using driver hpdftex*
+(/usr/share/texmf-texlive/tex/latex/hyperref/hpdftex.def
+File: hpdftex.def 2009/10/09 v6.79a Hyperref driver for pdfTeX
+\Fld@listcount=\count123
+)
+\@indexfile=\write3
+\openout3 = `38722-t.idx'.
+
+Writing index file 38722-t.idx
+\TmpLen=\skip62
+LaTeX Font Info: Encoding `OT1' has changed to `T1' for symbol font
+(Font) `operators' in the math version `normal' on input line 594.
+
+LaTeX Font Info: Overwriting symbol font `operators' in version `normal'
+(Font) OT1/cmr/m/n --> T1/cmsc/m/n on input line 594.
+(./38722-t.aux)
+\openout1 = `38722-t.aux'.
+
+LaTeX Font Info: Checking defaults for OML/cmm/m/it on input line 613.
+LaTeX Font Info: ... okay on input line 613.
+LaTeX Font Info: Checking defaults for T1/cmr/m/n on input line 613.
+LaTeX Font Info: ... okay on input line 613.
+LaTeX Font Info: Checking defaults for OT1/cmr/m/n on input line 613.
+LaTeX Font Info: ... okay on input line 613.
+LaTeX Font Info: Checking defaults for OMS/cmsy/m/n on input line 613.
+LaTeX Font Info: ... okay on input line 613.
+LaTeX Font Info: Checking defaults for OMX/cmex/m/n on input line 613.
+LaTeX Font Info: ... okay on input line 613.
+LaTeX Font Info: Checking defaults for U/cmr/m/n on input line 613.
+LaTeX Font Info: ... okay on input line 613.
+LaTeX Font Info: Checking defaults for LY/yfrak/m/n on input line 613.
+LaTeX Font Info: ... okay on input line 613.
+LaTeX Font Info: Checking defaults for LYG/ygoth/m/n on input line 613.
+LaTeX Font Info: ... okay on input line 613.
+LaTeX Font Info: Checking defaults for PD1/pdf/m/n on input line 613.
+LaTeX Font Info: ... okay on input line 613.
+(/usr/share/texmf/tex/context/base/supp-pdf.mkii
+[Loading MPS to PDF converter (version 2006.09.02).]
+\scratchcounter=\count124
+\scratchdimen=\dimen140
+\scratchbox=\box54
+\nofMPsegments=\count125
+\nofMParguments=\count126
+\everyMPshowfont=\toks26
+\MPscratchCnt=\count127
+\MPscratchDim=\dimen141
+\MPnumerator=\count128
+\everyMPtoPDFconversion=\toks27
+)
+Package caption Info: Begin \AtBeginDocument code.
+Package caption Info: hyperref package is loaded.
+Package caption Info: End \AtBeginDocument code.
+*geometry auto-detecting driver*
+*geometry detected driver: pdftex*
+-------------------- Geometry parameters
+paper: class default
+landscape: --
+twocolumn: --
+twoside: true
+asymmetric: --
+h-parts: 9.03375pt, 307.14749pt, 9.03375pt
+v-parts: 4.15848pt, 423.22379pt, 6.23773pt
+hmarginratio: 1:1
+vmarginratio: 2:3
+lines: --
+heightrounded: --
+bindingoffset: 0.0pt
+truedimen: --
+includehead: true
+includefoot: true
+includemp: --
+driver: pdftex
+-------------------- Page layout dimensions and switches
+\paperwidth 325.215pt
+\paperheight 433.62pt
+\textwidth 307.14749pt
+\textheight 361.34999pt
+\oddsidemargin -63.23624pt
+\evensidemargin -63.23624pt
+\topmargin -68.11151pt
+\headheight 15.0pt
+\headsep 19.8738pt
+\footskip 30.0pt
+\marginparwidth 98.0pt
+\marginparsep 7.0pt
+\columnsep 10.0pt
+\skip\footins 10.8pt plus 4.0pt minus 2.0pt
+\hoffset 0.0pt
+\voffset 0.0pt
+\mag 1000
+\@twosidetrue \@mparswitchtrue
+(1in=72.27pt, 1cm=28.45pt)
+-----------------------
+(/usr/share/texmf-texlive/tex/latex/graphics/color.sty
+Package: color 2005/11/14 v1.0j Standard LaTeX Color (DPC)
+(/etc/texmf/tex/latex/config/color.cfg
+File: color.cfg 2007/01/18 v1.5 color configuration of teTeX/TeXLive
+)
+Package color Info: Driver file: pdftex.def on input line 130.
+)
+Package hyperref Info: Link coloring ON on input line 613.
+(/usr/share/texmf-texlive/tex/latex/hyperref/nameref.sty
+Package: nameref 2007/05/29 v2.31 Cross-referencing by name of section
+(/usr/share/texmf-texlive/tex/latex/oberdiek/refcount.sty
+Package: refcount 2008/08/11 v3.1 Data extraction from references (HO)
+)
+\c@section@level=\count129
+)
+LaTeX Info: Redefining \ref on input line 613.
+LaTeX Info: Redefining \pageref on input line 613.
+(./38722-t.out) (./38722-t.out)
+\@outlinefile=\write4
+\openout4 = `38722-t.out'.
+
+\AtBeginShipoutBox=\box55
+
+Overfull \hbox (3.10696pt too wide) in paragraph at lines 642--642
+[]\OT1/cmtt/m/n/8 *** START OF THIS PROJECT GUTENBERG EBOOK TIDES AND KINDRED P
+HENOMENA ***[]
+ []
+
+LaTeX Font Info: Try loading font information for T1+cmsc on input line 644.
+
+(/usr/share/texmf-texlive/tex/latex/taupin/t1cmsc.fd
+File: T1cmsc.fd 1995/12/20 v2.4f Standard LaTeX font definitions
+)
+LaTeX Font Info: Try loading font information for U+msa on input line 644.
+(/usr/share/texmf-texlive/tex/latex/amsfonts/umsa.fd
+File: umsa.fd 2009/06/22 v3.00 AMS symbols A
+)
+LaTeX Font Info: Try loading font information for U+msb on input line 644.
+(/usr/share/texmf-texlive/tex/latex/amsfonts/umsb.fd
+File: umsb.fd 2009/06/22 v3.00 AMS symbols B
+) [1
+
+
+
+{/var/lib/texmf/fonts/map/pdftex/updmap/pdftex.map}] [2
+
+] <./images/riverside.png, id=137, 78.0516pt x 104.0688pt>
+File: ./images/riverside.png Graphic file (type png)
+<use ./images/riverside.png> [1
+
+ <./images/riverside.png (PNG copy)>] [2
+
+] [3
+
+
+] [4] [5] [6
+
+
+] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17
+
+
+] [18] [19] [1
+
+
+
+
+
+] [2] [3] [4] [5] <./images/fig1.png, id=1253, 236.3229pt x 230.5413pt>
+File: ./images/fig1.png Graphic file (type png)
+<use ./images/fig1.png> [6] [7 <./images/fig1.png (PNG copy)>] <./images/fig2.p
+ng, id=1264, 235.6002pt x 164.7756pt>
+File: ./images/fig2.png Graphic file (type png)
+<use ./images/fig2.png> [8] [9 <./images/fig2.png (PNG copy)>] <./images/fig3.j
+pg, id=1275, 154.176pt x 74.679pt>
+File: ./images/fig3.jpg Graphic file (type jpg)
+<use ./images/fig3.jpg> <./images/fig4.jpg, id=1276, 154.176pt x 241.1409pt>
+File: ./images/fig4.jpg Graphic file (type jpg)
+<use ./images/fig4.jpg> [10] [11 <./images/fig3.jpg>] [12 <./images/fig4.jpg>]
+[13] <./images/fig5.jpg, id=1300, 501.875pt x 500.87125pt>
+File: ./images/fig5.jpg Graphic file (type jpg)
+<use ./images/fig5.jpg> <./images/fig6.png, id=1301, 238.491pt x 108.405pt>
+File: ./images/fig6.png Graphic file (type png)
+<use ./images/fig6.png> [14] [15 <./images/fig5.jpg>] [16 <./images/fig6.png (P
+NG copy)>] [17] [18
+
+
+] [19] <./images/fig7.png, id=1340, 248.6088pt x 106.2369pt>
+File: ./images/fig7.png Graphic file (type png)
+<use ./images/fig7.png> [20] [21 <./images/fig7.png (PNG copy)>] [22] <./images
+/fig8.png, id=1360, 244.9953pt x 104.7915pt>
+File: ./images/fig8.png Graphic file (type png)
+<use ./images/fig8.png> [23] [24 <./images/fig8.png (PNG copy)>] [25] <./images
+/fig10.png, id=1374, 233.4321pt x 132.9768pt>
+File: ./images/fig10.png Graphic file (type png)
+<use ./images/fig10.png> [26] [27 <./images/fig10.png (PNG copy)>] [28] [29] <.
+/images/fig11.png, id=1397, 239.2137pt x 135.1449pt>
+File: ./images/fig11.png Graphic file (type png)
+<use ./images/fig11.png> [30] [31 <./images/fig11.png (PNG copy)>] <./images/fi
+g12.png, id=1408, 242.1045pt x 66.4884pt>
+File: ./images/fig12.png Graphic file (type png)
+<use ./images/fig12.png> [32 <./images/fig12.png (PNG copy)>] [33] <./images/fi
+g13.png, id=1421, 238.491pt x 68.6565pt>
+File: ./images/fig13.png Graphic file (type png)
+<use ./images/fig13.png> <./images/fig13b.pdf, id=1422, 293.095pt x 431.6125pt>
+File: ./images/fig13b.pdf Graphic file (type pdf)
+<use ./images/fig13b.pdf> [34 <./images/fig13.png (PNG copy)>] [35] [36 <./imag
+es/fig13b.pdf>] [37] [38] [39] [40] [41] [42] [43] [44] <./images/fig14.png, id
+=1492, 71.5473pt x 389.5353pt>
+File: ./images/fig14.png Graphic file (type png)
+<use ./images/fig14.png>
+
+LaTeX Warning: Float too large for page by 3.9314pt on input line 2525.
+
+[45] [46] [47 <./images/fig14.png (PNG copy)>] [48] [49] [50] [51] [52] [53] [5
+4] [55] [56] [57
+
+
+] [58] [59] <./images/fig15.png, id=1572, 242.1045pt x 36.135pt>
+File: ./images/fig15.png Graphic file (type png)
+<use ./images/fig15.png> [60] [61 <./images/fig15.png (PNG copy)>] <./images/fi
+g16.png, id=1581, 255.1131pt x 114.9093pt>
+File: ./images/fig16.png Graphic file (type png)
+<use ./images/fig16.png> [62] [63 <./images/fig16.png (PNG copy)>] [64] <./imag
+es/fig17.jpg, id=1599, 524.96124pt x 310.15875pt>
+File: ./images/fig17.jpg Graphic file (type jpg)
+<use ./images/fig17.jpg> <./images/fig18.jpg, id=1601, 154.176pt x 91.542pt>
+File: ./images/fig18.jpg Graphic file (type jpg)
+<use ./images/fig18.jpg> [65] [66 <./images/fig17.jpg>] [67] [68 <./images/fig1
+8.jpg>] [69] <./images/fig19.jpg, id=1625, 422.57875pt x 642.4pt>
+File: ./images/fig19.jpg Graphic file (type jpg)
+<use ./images/fig19.jpg>
+
+LaTeX Warning: Float too large for page by 10.85399pt on input line 3278.
+
+[70] [71 <./images/fig19.jpg>] [72] [73] [74] [75] [76] [77] [78
+
+
+] [79] [80] [81] [82] [83] [84] [85] [86] [87] [88] [89] [90] [91] [92] [93
+
+
+] <./images/fig20.png, id=1745, 242.8272pt x 120.6909pt>
+File: ./images/fig20.png Graphic file (type png)
+<use ./images/fig20.png> [94] [95 <./images/fig20.png (PNG copy)>] [96] [97] [9
+8] <./images/fig21.png, id=1771, 233.4321pt x 128.6406pt>
+File: ./images/fig21.png Graphic file (type png)
+<use ./images/fig21.png> [99] [100 <./images/fig21.png (PNG copy)>] <./images/f
+ig22.png, id=1784, 250.7769pt x 146.7081pt>
+File: ./images/fig22.png Graphic file (type png)
+<use ./images/fig22.png> [101] [102 <./images/fig22.png (PNG copy)>] [103] [104
+] [105] [106] [107] <./images/fig23.jpg, id=1827, 154.176pt x 170.3163pt>
+File: ./images/fig23.jpg Graphic file (type jpg)
+<use ./images/fig23.jpg>
+Underfull \hbox (badness 2469) in paragraph at lines 4551--4563
+[]\OT1/cmr/m/n/12 Taking then the di-a-gram of tide-generating forces
+ []
+
+[108] [109 <./images/fig23.jpg>] [110] [111] <./images/fig24.png, id=1847, 238.
+491pt x 111.2958pt>
+File: ./images/fig24.png Graphic file (type png)
+<use ./images/fig24.png> [112
+
+
+] [113] [114 <./images/fig24.png (PNG copy)>] <./images/fig25.png, id=1866, 232
+.7094pt x 125.0271pt>
+File: ./images/fig25.png Graphic file (type png)
+<use ./images/fig25.png> [115] [116 <./images/fig25.png (PNG copy)>] [117] <./i
+mages/fig26.png, id=1882, 52.0344pt x 255.1131pt>
+File: ./images/fig26.png Graphic file (type png)
+<use ./images/fig26.png> [118] [119 <./images/fig26.png (PNG copy)>] [120] [121
+] [122] [123] [124] [125] [126] [127] [128] [129] [130] [131] [132] [133] [134]
+[135] [136] [137] <./images/fig27.png, id=1983, 247.8861pt x 117.8001pt>
+File: ./images/fig27.png Graphic file (type png)
+<use ./images/fig27.png> [138
+
+
+] [139 <./images/fig27.png (PNG copy)>] [140] [141] <./images/fig28.png, id=200
+8, 235.6002pt x 66.4884pt>
+File: ./images/fig28.png Graphic file (type png)
+<use ./images/fig28.png> [142] [143 <./images/fig28.png (PNG copy)>] [144] [145
+] [146] [147] [148] [149] [150] [151] [152] [153
+
+
+] [154] <./images/fig29.png, id=2081, 198.7425pt x 401.8212pt>
+File: ./images/fig29.png Graphic file (type png)
+<use ./images/fig29.png>
+
+LaTeX Warning: Float too large for page by 3.9314pt on input line 6094.
+
+[155] [156 <./images/fig29.png (PNG copy)>] [157] [158] [159] [160] [161] [162]
+[163] [164] [165] [166] [167
+
+
+] [168] [169] [170] [171] [172] [173] <./images/fig30.png, id=2187, 237.7683pt
+x 94.6737pt>
+File: ./images/fig30.png Graphic file (type png)
+<use ./images/fig30.png> [174] [175 <./images/fig30.png (PNG copy)>] [176] [177
+] [178] [179] [180] [181] [182] [183] [184] [185] [186
+
+
+] <./images/fig31.png, id=2253, 241.3818pt x 80.9424pt>
+File: ./images/fig31.png Graphic file (type png)
+<use ./images/fig31.png> [187] [188 <./images/fig31.png (PNG copy)>] [189] [190
+] [191] [192] <./images/fig32.jpg, id=2284, 410.53375pt x 642.4pt>
+File: ./images/fig32.jpg Graphic file (type jpg)
+<use ./images/fig32.jpg> [193] [194 <./images/fig32.jpg>] [195] [196
+
+
+] [197] [198] [199] [200] [201] [202] [203] [204] [205] [206] [207] [208] [209]
+[210] [211] [212] [213
+
+
+] [214] [215] [216] [217] [218] [219] <./images/fig33.jpg, id=2427, 386.44376pt
+ x 642.4pt>
+File: ./images/fig33.jpg Graphic file (type jpg)
+<use ./images/fig33.jpg>
+
+LaTeX Warning: Float too large for page by 3.9314pt on input line 8282.
+
+[220] [221 <./images/fig33.jpg>] [222] [223] [224
+
+
+] [225] [226] [227] [228] <./images/fig34.png, id=2477, 244.2726pt x 247.8861pt
+>
+File: ./images/fig34.png Graphic file (type png)
+<use ./images/fig34.png> [229] [230 <./images/fig34.png (PNG copy)>] [231] [232
+] [233] [234] [235] [236] <./images/fig35.png, id=2517, 236.3229pt x 190.7928pt
+>
+File: ./images/fig35.png Graphic file (type png)
+<use ./images/fig35.png> [237] [238 <./images/fig35.png (PNG copy)>] [239] [240
+] [241] [242] [243] [244] [245
+
+
+] [246] [247] [248] [249] [250] [251] [252] [253] [254] [255
+
+
+] [256] [257] [258] [259] [260] [261] [262] [263] [264] [265] [266] [267
+
+
+] <./images/fig36.png, id=2682, 235.6002pt x 104.7915pt>
+File: ./images/fig36.png Graphic file (type png)
+<use ./images/fig36.png> [268] [269 <./images/fig36.png (PNG copy)>] [270] [271
+] [272] [273] [274] [275] [276] [277] [278] [279] [280] [281] [282] [283] [284]
+[285] [286] [287] [288] [289] [290] [291] [292
+
+
+] [293] [294] [295] [296] [297] [298] [299] [300] [301] [302] [303] [304] [305]
+[306] [307] [308] [309] [310] [311] [312] [313] [314] [315] [316] [317] [318
+
+
+] [319] [320] [321] [322] [323] <./images/fig37.png, id=2971, 235.6002pt x 140.
+9265pt>
+File: ./images/fig37.png Graphic file (type png)
+<use ./images/fig37.png> [324] [325 <./images/fig37.png (PNG copy)>] <./images/
+fig38.png, id=2982, 239.9364pt x 130.8087pt>
+File: ./images/fig38.png Graphic file (type png)
+<use ./images/fig38.png> [326] [327 <./images/fig38.png (PNG copy)>] [328] [329
+] [330] <./images/fig39.png, id=3008, 247.1634pt x 203.8014pt>
+File: ./images/fig39.png Graphic file (type png)
+<use ./images/fig39.png> [331 <./images/fig39.png (PNG copy)>] [332] [333] [334
+] [335] [336] [337
+
+
+] [338] [339] [340] <./images/fig40.jpg, id=3063, 431.6125pt x 642.4pt>
+File: ./images/fig40.jpg Graphic file (type jpg)
+<use ./images/fig40.jpg> [341] [342 <./images/fig40.jpg>] [343] [344] [345] [34
+6] [347] [348] [349] [350] [351
+
+
+] <./images/fig41.jpg, id=3121, 154.176pt x 78.5334pt>
+File: ./images/fig41.jpg Graphic file (type jpg)
+<use ./images/fig41.jpg> [352] [353 <./images/fig41.jpg>] [354] [355] <./images
+/fig42.png, id=3143, 238.491pt x 123.5817pt>
+File: ./images/fig42.png Graphic file (type png)
+<use ./images/fig42.png> [356] [357 <./images/fig42.png (PNG copy)>] [358] [359
+] <./images/fig43.png, id=3169, 236.3229pt x 83.8332pt>
+File: ./images/fig43.png Graphic file (type png)
+<use ./images/fig43.png> [360] [361 <./images/fig43.png (PNG copy)>] [362] [363
+] [364] [365] [366] [367]
+Underfull \hbox (badness 1087) in paragraph at lines 13273--13276
+[]\OT1/cmr/m/n/12 I can-not sum up the gen-eral con-clu-sions at which
+ []
+
+[368] [369] [370] [371] [372] [373] (./38722-t.ind [374
+
+
+
+
+
+] [375] [376] [377] [378] [379] [380] [381] [382] [383] [384] [385] [386] [387]
+[388] [389] [390]) [391
+
+
+
+]
+Overfull \hbox (15.85715pt too wide) in paragraph at lines 14256--14256
+[]\OT1/cmtt/m/n/8 End of the Project Gutenberg EBook of The Tides and Kindred P
+henomena in the[]
+ []
+
+[1
+
+
+
+
+]
+Overfull \hbox (3.10696pt too wide) in paragraph at lines 14330--14330
+[]\OT1/cmtt/m/n/8 1.C. The Project Gutenberg Literary Archive Foundation ("the
+ Foundation"[]
+ []
+
+[2]
+Overfull \hbox (3.10696pt too wide) in paragraph at lines 14335--14335
+[]\OT1/cmtt/m/n/8 located in the United States, we do not claim a right to prev
+ent you from[]
+ []
+
+
+Overfull \hbox (3.10696pt too wide) in paragraph at lines 14340--14340
+[]\OT1/cmtt/m/n/8 freely sharing Project Gutenberg-tm works in compliance with
+the terms of[]
+ []
+
+[3]
+Overfull \hbox (3.10696pt too wide) in paragraph at lines 14403--14403
+[]\OT1/cmtt/m/n/8 posted on the official Project Gutenberg-tm web site (www.gut
+enberg.net),[]
+ []
+
+[4] [5] [6] [7] [8] [9] [10] (./38722-t.aux)
+
+ *File List*
+ book.cls 2007/10/19 v1.4h Standard LaTeX document class
+ bk12.clo 2007/10/19 v1.4h Standard LaTeX file (size option)
+inputenc.sty 2008/03/30 v1.1d Input encoding file
+ latin1.def 2008/03/30 v1.1d Input encoding file
+ ifthen.sty 2001/05/26 v1.1c Standard LaTeX ifthen package (DPC)
+ amsmath.sty 2000/07/18 v2.13 AMS math features
+ amstext.sty 2000/06/29 v2.01
+ amsgen.sty 1999/11/30 v2.0
+ amsbsy.sty 1999/11/29 v1.2d
+ amsopn.sty 1999/12/14 v2.01 operator names
+ amssymb.sty 2009/06/22 v3.00
+amsfonts.sty 2009/06/22 v3.00 Basic AMSFonts support
+ alltt.sty 1997/06/16 v2.0g defines alltt environment
+footmisc.sty 2009/09/15 v5.5a a miscellany of footnote facilities
+multicol.sty 2008/12/05 v1.6h multicolumn formatting (FMi)
+ makeidx.sty 2000/03/29 v1.0m Standard LaTeX package
+indentfirst.sty 1995/11/23 v1.03 Indent first paragraph (DPC)
+ icomma.sty 2002/03/10 v2.0 (WaS)
+ calc.sty 2007/08/22 v4.3 Infix arithmetic (KKT,FJ)
+ array.sty 2008/09/09 v2.4c Tabular extension package (FMi)
+ yfonts.sty 2003/01/08 v1.3 (WaS)
+fancyhdr.sty
+graphicx.sty 1999/02/16 v1.0f Enhanced LaTeX Graphics (DPC,SPQR)
+ keyval.sty 1999/03/16 v1.13 key=value parser (DPC)
+graphics.sty 2009/02/05 v1.0o Standard LaTeX Graphics (DPC,SPQR)
+ trig.sty 1999/03/16 v1.09 sin cos tan (DPC)
+graphics.cfg 2009/08/28 v1.8 graphics configuration of TeX Live
+ pdftex.def 2009/08/25 v0.04m Graphics/color for pdfTeX
+ caption.sty 2009/10/09 v3.1k Customizing captions (AR)
+caption3.sty 2009/10/09 v3.1k caption3 kernel (AR)
+geometry.sty 2008/12/21 v4.2 Page Geometry
+ ifpdf.sty 2009/04/10 v2.0 Provides the ifpdf switch (HO)
+ ifvtex.sty 2008/11/04 v1.4 Switches for detecting VTeX and its modes (HO)
+geometry.cfg
+hyperref.sty 2009/10/09 v6.79a Hypertext links for LaTeX
+ ifxetex.sty 2009/01/23 v0.5 Provides ifxetex conditional
+ hycolor.sty 2009/10/02 v1.5 Code for color options of hyperref/bookmark (HO
+)
+xcolor-patch.sty 2009/10/02 xcolor patch
+ pd1enc.def 2009/10/09 v6.79a Hyperref: PDFDocEncoding definition (HO)
+etexcmds.sty 2007/12/12 v1.2 Prefix for e-TeX command names (HO)
+infwarerr.sty 2007/09/09 v1.2 Providing info/warning/message (HO)
+hyperref.cfg 2002/06/06 v1.2 hyperref configuration of TeXLive
+kvoptions.sty 2009/08/13 v3.4 Keyval support for LaTeX options (HO)
+kvsetkeys.sty 2009/07/30 v1.5 Key value parser with default handler support
+(HO)
+ url.sty 2006/04/12 ver 3.3 Verb mode for urls, etc.
+ bitset.sty 2007/09/28 v1.0 Data type bit set (HO)
+ intcalc.sty 2007/09/27 v1.1 Expandable integer calculations (HO)
+bigintcalc.sty 2007/11/11 v1.1 Expandable big integer calculations (HO)
+pdftexcmds.sty 2009/09/23 v0.6 LuaTeX support for pdfTeX utility functions (
+HO)
+ifluatex.sty 2009/04/17 v1.2 Provides the ifluatex switch (HO)
+ ltxcmds.sty 2009/08/05 v1.0 Some LaTeX kernel commands for general use (HO)
+
+atbegshi.sty 2008/07/31 v1.9 At begin shipout hook (HO)
+ hpdftex.def 2009/10/09 v6.79a Hyperref driver for pdfTeX
+supp-pdf.mkii
+ color.sty 2005/11/14 v1.0j Standard LaTeX Color (DPC)
+ color.cfg 2007/01/18 v1.5 color configuration of teTeX/TeXLive
+ nameref.sty 2007/05/29 v2.31 Cross-referencing by name of section
+refcount.sty 2008/08/11 v3.1 Data extraction from references (HO)
+ 38722-t.out
+ 38722-t.out
+ t1cmsc.fd
+ umsa.fd 2009/06/22 v3.00 AMS symbols A
+ umsb.fd 2009/06/22 v3.00 AMS symbols B
+./images/riverside.png
+./images/fig1.png
+./images/fig2.png
+./images/fig3.jpg
+./images/fig4.jpg
+./images/fig5.jpg
+./images/fig6.png
+./images/fig7.png
+./images/fig8.png
+./images/fig10.png
+./images/fig11.png
+./images/fig12.png
+./images/fig13.png
+./images/fig13b.pdf
+./images/fig14.png
+./images/fig15.png
+./images/fig16.png
+./images/fig17.jpg
+./images/fig18.jpg
+./images/fig19.jpg
+./images/fig20.png
+./images/fig21.png
+./images/fig22.png
+./images/fig23.jpg
+./images/fig24.png
+./images/fig25.png
+./images/fig26.png
+./images/fig27.png
+./images/fig28.png
+./images/fig29.png
+./images/fig30.png
+./images/fig31.png
+./images/fig32.jpg
+./images/fig33.jpg
+./images/fig34.png
+./images/fig35.png
+./images/fig36.png
+./images/fig37.png
+./images/fig38.png
+./images/fig39.png
+./images/fig40.jpg
+./images/fig41.jpg
+./images/fig42.png
+./images/fig43.png
+ 38722-t.ind
+ ***********
+
+ )
+Here is how much of TeX's memory you used:
+ 8940 strings out of 493848
+ 125292 string characters out of 1152824
+ 218123 words of memory out of 3000000
+ 10981 multiletter control sequences out of 15000+50000
+ 23054 words of font info for 76 fonts, out of 3000000 for 9000
+ 714 hyphenation exceptions out of 8191
+ 37i,14n,44p,330b,1043s stack positions out of 5000i,500n,10000p,200000b,50000s
+{/usr/share/texmf/fonts/enc/dvips/cm-super/cm-super-t1.enc}</usr/share/texmf-
+texlive/fonts/type1/public/amsfonts/cm/cmbx12.pfb></usr/share/texmf-texlive/fon
+ts/type1/public/amsfonts/cm/cmcsc10.pfb></usr/share/texmf-texlive/fonts/type1/p
+ublic/amsfonts/cm/cmmi10.pfb></usr/share/texmf-texlive/fonts/type1/public/amsfo
+nts/cm/cmmi12.pfb></usr/share/texmf-texlive/fonts/type1/public/amsfonts/cm/cmmi
+5.pfb></usr/share/texmf-texlive/fonts/type1/public/amsfonts/cm/cmmi7.pfb></usr/
+share/texmf-texlive/fonts/type1/public/amsfonts/cm/cmmi8.pfb></usr/share/texmf-
+texlive/fonts/type1/public/amsfonts/cm/cmr10.pfb></usr/share/texmf-texlive/font
+s/type1/public/amsfonts/cm/cmr12.pfb></usr/share/texmf-texlive/fonts/type1/publ
+ic/amsfonts/cm/cmr17.pfb></usr/share/texmf-texlive/fonts/type1/public/amsfonts/
+cm/cmr7.pfb></usr/share/texmf-texlive/fonts/type1/public/amsfonts/cm/cmr8.pfb><
+/usr/share/texmf-texlive/fonts/type1/public/amsfonts/cm/cmsy10.pfb></usr/share/
+texmf-texlive/fonts/type1/public/amsfonts/cm/cmsy7.pfb></usr/share/texmf-texliv
+e/fonts/type1/public/amsfonts/cm/cmsy8.pfb></usr/share/texmf-texlive/fonts/type
+1/public/amsfonts/cm/cmti10.pfb></usr/share/texmf-texlive/fonts/type1/public/am
+sfonts/cm/cmti12.pfb></usr/share/texmf-texlive/fonts/type1/public/amsfonts/cm/c
+mtt10.pfb></usr/share/texmf-texlive/fonts/type1/public/amsfonts/cm/cmtt8.pfb></
+usr/share/texmf/fonts/type1/public/cm-super/sfcc0500.pfb></usr/share/texmf/font
+s/type1/public/cm-super/sfcc0600.pfb></usr/share/texmf/fonts/type1/public/cm-su
+per/sfcc0700.pfb></usr/share/texmf/fonts/type1/public/cm-super/sfcc0800.pfb></u
+sr/share/texmf/fonts/type1/public/cm-super/sfcc1000.pfb></usr/share/texmf/fonts
+/type1/public/cm-super/sfcc1095.pfb></usr/share/texmf/fonts/type1/public/cm-sup
+er/sfcc1200.pfb></usr/share/texmf-texlive/fonts/type1/public/gothic/ygoth.pfb>
+Output written on 38722-t.pdf (422 pages, 3217612 bytes).
+PDF statistics:
+ 4299 PDF objects out of 5155 (max. 8388607)
+ 1042 named destinations out of 1200 (max. 500000)
+ 453 words of extra memory for PDF output out of 10000 (max. 10000000)
+
diff --git a/38722-t/images/fig1.png b/38722-t/images/fig1.png
new file mode 100644
index 0000000..5ed2f02
--- /dev/null
+++ b/38722-t/images/fig1.png
Binary files differ
diff --git a/38722-t/images/fig10.png b/38722-t/images/fig10.png
new file mode 100644
index 0000000..41286ca
--- /dev/null
+++ b/38722-t/images/fig10.png
Binary files differ
diff --git a/38722-t/images/fig11.png b/38722-t/images/fig11.png
new file mode 100644
index 0000000..66ed523
--- /dev/null
+++ b/38722-t/images/fig11.png
Binary files differ
diff --git a/38722-t/images/fig12.png b/38722-t/images/fig12.png
new file mode 100644
index 0000000..aba5ba7
--- /dev/null
+++ b/38722-t/images/fig12.png
Binary files differ
diff --git a/38722-t/images/fig13.png b/38722-t/images/fig13.png
new file mode 100644
index 0000000..9583044
--- /dev/null
+++ b/38722-t/images/fig13.png
Binary files differ
diff --git a/38722-t/images/fig13b.pdf b/38722-t/images/fig13b.pdf
new file mode 100644
index 0000000..576ad7d
--- /dev/null
+++ b/38722-t/images/fig13b.pdf
Binary files differ
diff --git a/38722-t/images/fig14.png b/38722-t/images/fig14.png
new file mode 100644
index 0000000..97d39d6
--- /dev/null
+++ b/38722-t/images/fig14.png
Binary files differ
diff --git a/38722-t/images/fig15.png b/38722-t/images/fig15.png
new file mode 100644
index 0000000..5f4b8a4
--- /dev/null
+++ b/38722-t/images/fig15.png
Binary files differ
diff --git a/38722-t/images/fig16.png b/38722-t/images/fig16.png
new file mode 100644
index 0000000..dce4ee1
--- /dev/null
+++ b/38722-t/images/fig16.png
Binary files differ
diff --git a/38722-t/images/fig17.jpg b/38722-t/images/fig17.jpg
new file mode 100644
index 0000000..ccd977e
--- /dev/null
+++ b/38722-t/images/fig17.jpg
Binary files differ
diff --git a/38722-t/images/fig18.jpg b/38722-t/images/fig18.jpg
new file mode 100644
index 0000000..8e861ab
--- /dev/null
+++ b/38722-t/images/fig18.jpg
Binary files differ
diff --git a/38722-t/images/fig19.jpg b/38722-t/images/fig19.jpg
new file mode 100644
index 0000000..43fba80
--- /dev/null
+++ b/38722-t/images/fig19.jpg
Binary files differ
diff --git a/38722-t/images/fig2.png b/38722-t/images/fig2.png
new file mode 100644
index 0000000..998fd5a
--- /dev/null
+++ b/38722-t/images/fig2.png
Binary files differ
diff --git a/38722-t/images/fig20.png b/38722-t/images/fig20.png
new file mode 100644
index 0000000..af6c94f
--- /dev/null
+++ b/38722-t/images/fig20.png
Binary files differ
diff --git a/38722-t/images/fig21.png b/38722-t/images/fig21.png
new file mode 100644
index 0000000..11f6595
--- /dev/null
+++ b/38722-t/images/fig21.png
Binary files differ
diff --git a/38722-t/images/fig22.png b/38722-t/images/fig22.png
new file mode 100644
index 0000000..ca6849a
--- /dev/null
+++ b/38722-t/images/fig22.png
Binary files differ
diff --git a/38722-t/images/fig23.jpg b/38722-t/images/fig23.jpg
new file mode 100644
index 0000000..e54141b
--- /dev/null
+++ b/38722-t/images/fig23.jpg
Binary files differ
diff --git a/38722-t/images/fig24.png b/38722-t/images/fig24.png
new file mode 100644
index 0000000..9f79882
--- /dev/null
+++ b/38722-t/images/fig24.png
Binary files differ
diff --git a/38722-t/images/fig25.png b/38722-t/images/fig25.png
new file mode 100644
index 0000000..90ffee6
--- /dev/null
+++ b/38722-t/images/fig25.png
Binary files differ
diff --git a/38722-t/images/fig26.png b/38722-t/images/fig26.png
new file mode 100644
index 0000000..3492dca
--- /dev/null
+++ b/38722-t/images/fig26.png
Binary files differ
diff --git a/38722-t/images/fig27.png b/38722-t/images/fig27.png
new file mode 100644
index 0000000..1a7feb7
--- /dev/null
+++ b/38722-t/images/fig27.png
Binary files differ
diff --git a/38722-t/images/fig28.png b/38722-t/images/fig28.png
new file mode 100644
index 0000000..89719bb
--- /dev/null
+++ b/38722-t/images/fig28.png
Binary files differ
diff --git a/38722-t/images/fig29.png b/38722-t/images/fig29.png
new file mode 100644
index 0000000..802e29c
--- /dev/null
+++ b/38722-t/images/fig29.png
Binary files differ
diff --git a/38722-t/images/fig3.jpg b/38722-t/images/fig3.jpg
new file mode 100644
index 0000000..b04fef5
--- /dev/null
+++ b/38722-t/images/fig3.jpg
Binary files differ
diff --git a/38722-t/images/fig30.png b/38722-t/images/fig30.png
new file mode 100644
index 0000000..83cef6e
--- /dev/null
+++ b/38722-t/images/fig30.png
Binary files differ
diff --git a/38722-t/images/fig31.png b/38722-t/images/fig31.png
new file mode 100644
index 0000000..9ed6c5c
--- /dev/null
+++ b/38722-t/images/fig31.png
Binary files differ
diff --git a/38722-t/images/fig32.jpg b/38722-t/images/fig32.jpg
new file mode 100644
index 0000000..a751f1d
--- /dev/null
+++ b/38722-t/images/fig32.jpg
Binary files differ
diff --git a/38722-t/images/fig33.jpg b/38722-t/images/fig33.jpg
new file mode 100644
index 0000000..0585f90
--- /dev/null
+++ b/38722-t/images/fig33.jpg
Binary files differ
diff --git a/38722-t/images/fig34.png b/38722-t/images/fig34.png
new file mode 100644
index 0000000..481e21d
--- /dev/null
+++ b/38722-t/images/fig34.png
Binary files differ
diff --git a/38722-t/images/fig35.png b/38722-t/images/fig35.png
new file mode 100644
index 0000000..9f52a7c
--- /dev/null
+++ b/38722-t/images/fig35.png
Binary files differ
diff --git a/38722-t/images/fig36.png b/38722-t/images/fig36.png
new file mode 100644
index 0000000..7e522be
--- /dev/null
+++ b/38722-t/images/fig36.png
Binary files differ
diff --git a/38722-t/images/fig37.png b/38722-t/images/fig37.png
new file mode 100644
index 0000000..48b9fb7
--- /dev/null
+++ b/38722-t/images/fig37.png
Binary files differ
diff --git a/38722-t/images/fig38.png b/38722-t/images/fig38.png
new file mode 100644
index 0000000..534e5b9
--- /dev/null
+++ b/38722-t/images/fig38.png
Binary files differ
diff --git a/38722-t/images/fig39.png b/38722-t/images/fig39.png
new file mode 100644
index 0000000..b9f2846
--- /dev/null
+++ b/38722-t/images/fig39.png
Binary files differ
diff --git a/38722-t/images/fig4.jpg b/38722-t/images/fig4.jpg
new file mode 100644
index 0000000..001acac
--- /dev/null
+++ b/38722-t/images/fig4.jpg
Binary files differ
diff --git a/38722-t/images/fig40.jpg b/38722-t/images/fig40.jpg
new file mode 100644
index 0000000..4f20a32
--- /dev/null
+++ b/38722-t/images/fig40.jpg
Binary files differ
diff --git a/38722-t/images/fig41.jpg b/38722-t/images/fig41.jpg
new file mode 100644
index 0000000..c69e363
--- /dev/null
+++ b/38722-t/images/fig41.jpg
Binary files differ
diff --git a/38722-t/images/fig42.png b/38722-t/images/fig42.png
new file mode 100644
index 0000000..08c72b4
--- /dev/null
+++ b/38722-t/images/fig42.png
Binary files differ
diff --git a/38722-t/images/fig43.png b/38722-t/images/fig43.png
new file mode 100644
index 0000000..87534ea
--- /dev/null
+++ b/38722-t/images/fig43.png
Binary files differ
diff --git a/38722-t/images/fig5.jpg b/38722-t/images/fig5.jpg
new file mode 100644
index 0000000..2a26102
--- /dev/null
+++ b/38722-t/images/fig5.jpg
Binary files differ
diff --git a/38722-t/images/fig6.png b/38722-t/images/fig6.png
new file mode 100644
index 0000000..48364ee
--- /dev/null
+++ b/38722-t/images/fig6.png
Binary files differ
diff --git a/38722-t/images/fig7.png b/38722-t/images/fig7.png
new file mode 100644
index 0000000..5dca6fe
--- /dev/null
+++ b/38722-t/images/fig7.png
Binary files differ
diff --git a/38722-t/images/fig8.png b/38722-t/images/fig8.png
new file mode 100644
index 0000000..082d2ae
--- /dev/null
+++ b/38722-t/images/fig8.png
Binary files differ
diff --git a/38722-t/images/riverside.png b/38722-t/images/riverside.png
new file mode 100644
index 0000000..3df4a17
--- /dev/null
+++ b/38722-t/images/riverside.png
Binary files differ
diff --git a/38722-t/old/38722-t.tex b/38722-t/old/38722-t.tex
new file mode 100644
index 0000000..d6d53d0
--- /dev/null
+++ b/38722-t/old/38722-t.tex
@@ -0,0 +1,15600 @@
+% %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% %
+% %
+% The Project Gutenberg EBook of The Tides and Kindred Phenomena in the %
+% Solar System, by Sir George Howard Darwin %
+% %
+% 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.net %
+% %
+% %
+% Title: The Tides and Kindred Phenomena in the Solar System %
+% The Substance of Lectures Delivered in 1897 at the Lowell %
+% Institute, Boston, Massachusetts %
+% %
+% Author: Sir George Howard Darwin %
+% %
+% Release Date: January 31, 2012 [EBook #38722] %
+% %
+% Language: English %
+% %
+% Character set encoding: ISO-8859-1 %
+% %
+% *** START OF THIS PROJECT GUTENBERG EBOOK TIDES AND KINDRED PHENOMENA ***
+% %
+% %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% %
+
+\def\ebook{38722}
+%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
+%% %%
+%% Packages and substitutions: %%
+%% %%
+%% book: Required. %%
+%% inputenc: Standard DP encoding. Required. %%
+%% %%
+%% ifthen: Logical conditionals. Required. %%
+%% %%
+%% amsmath: AMS mathematics enhancements. Required. %%
+%% amssymb: Additional mathematical symbols. Required. %%
+%% %%
+%% alltt: Fixed-width font environment. Required. %%
+%% %%
+%% multicol: Multicolumn environment for index. Required. %%
+%% makeidx: Indexing. Required. %%
+%% %%
+%% footmisc: Extended footnote capabilities. Required. %%
+%% %%
+%% indentfirst: Indent first word of each sectional unit. Required. %%
+%% icomma: Make the comma a decimal separator in math. Required. %%
+%% %%
+%% calc: Length calculations. Required. %%
+%% %%
+%% array: Enhanced arrays. Required. %%
+%% %%
+%% yfonts: Gothic font on title page. Optional. %%
+%% %%
+%% fancyhdr: Enhanced running headers and footers. Required. %%
+%% %%
+%% graphicx: Standard interface for graphics inclusion. Required. %%
+%% caption: Caption customization. Required. %%
+%% %%
+%% geometry: Enhanced page layout package. Required. %%
+%% hyperref: Hypertext embellishments for pdf output. Required. %%
+%% %%
+%% %%
+%% Producer's Comments: %%
+%% %%
+%% Changes are noted in this file in multiple ways. %%
+%% 1. \DPnote{} for in-line `placeholder' notes. %%
+%% 2. \DPtypo{}{} for typographical corrections, showing original %%
+%% and replacement text side-by-side. %%
+%% 3. \DPchg{}{} for stylistic changes made for consistency. %%
+%% 4. [** TN: Note]s for lengthier or stylistic comments. %%
+%% %%
+%% %%
+%% Compilation Flags: %%
+%% %%
+%% The following behavior may be controlled by boolean flags. %%
+%% %%
+%% ForPrinting (false by default): %%
+%% Compile a print-optimized PDF file. Set to false for screen- %%
+%% optimized file (pages cropped, one-sided, blue hyperlinks). %%
+%% %%
+%% %%
+%% PDF pages: 422 (if ForPrinting set to false) %%
+%% PDF page size: 4.5 x 6in (non-standard) %%
+%% %%
+%% Images: 1 pdf diagram, 32 png images, 11 jpg images. %%
+%% %%
+%% Summary of log file: %%
+%% * Two underfull hboxes. %%
+%% %%
+%% %%
+%% Compile History: %%
+%% %%
+%% January, 2012: adhere (Andrew D. Hwang) %%
+%% texlive2007, GNU/Linux %%
+%% %%
+%% Command block: %%
+%% %%
+%% pdflatex x2 %%
+%% makeindex %%
+%% pdflatex x2 %%
+%% %%
+%% %%
+%% January 2012: pglatex. %%
+%% Compile this project with: %%
+%% pdflatex 38722-t.tex ..... TWO times %%
+%% makeindex 38722-t.idx %%
+%% pdflatex 38722-t.tex ..... TWO times %%
+%% %%
+%% pdfTeX, Version 3.1415926-1.40.10 (TeX Live 2009/Debian) %%
+%% %%
+%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
+\listfiles
+\documentclass[12pt]{book}[2005/09/16]
+
+%%%%%%%%%%%%%%%%%%%%%%%%%%%%% PACKAGES %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
+\usepackage[latin1]{inputenc}[2006/05/05]
+
+\usepackage{ifthen}[2001/05/26] %% Logical conditionals
+
+\usepackage{amsmath}[2000/07/18] %% Displayed equations
+\usepackage{amssymb}[2002/01/22] %% and additional symbols
+
+\usepackage{alltt}[1997/06/16] %% boilerplate, credits, license
+
+ %% extended footnote capabilities
+\usepackage[perpage]{footmisc}[2005/03/17]
+
+\usepackage{multicol}[2006/05/18]
+\usepackage{makeidx}[2000/03/29]
+
+\usepackage{indentfirst}[1995/11/23]
+\usepackage{icomma}[2002/03/10]
+
+\usepackage{calc}[2005/08/06]
+
+\usepackage{array}
+
+\IfFileExists{yfonts.sty}{%
+ \usepackage{yfonts}[2003/01/08]%
+}{%
+ \providecommand{\textgoth}[1]{\textbf{#1}}%
+}
+
+% for running heads
+\usepackage{fancyhdr}
+
+\usepackage{graphicx}[1999/02/16]%% For diagrams
+\usepackage[labelformat=empty,justification=centerlast,textfont={sc,footnotesize}]{caption}[2007/01/07]
+
+%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
+%%%% Interlude: Set up PRINTING (default) or SCREEN VIEWING %%%%
+%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
+
+% ForPrinting=true (default) false
+% Asymmetric margins Symmetric margins
+% Black hyperlinks Blue hyperlinks
+% Start Preface, ToC, etc. recto No blank verso pages
+%
+\newboolean{ForPrinting}
+%% UNCOMMENT the next line for a PRINT-OPTIMIZED VERSION of the text %%
+%\setboolean{ForPrinting}{true}
+
+%% Initialize values to ForPrinting=false
+\newcommand{\ChapterSpace}{}
+\newcommand{\Margins}{hmarginratio=1:1} % Symmetric margins
+\newcommand{\HLinkColor}{blue} % Hyperlink color
+\newcommand{\PDFPageLayout}{SinglePage}
+\newcommand{\TransNote}{Transcriber's Note}
+\newcommand{\TransNoteCommon}{%
+ Minor typographical corrections and presentational changes have been
+ made without comment. All changes are detailed in the \LaTeX\ source
+ file, which may be downloaded from
+ \begin{center}
+ \texttt{www.gutenberg.org/ebooks/\ebook}.
+ \end{center}
+ \bigskip
+}
+\newcommand{\TransNoteText}{%
+ \TransNoteCommon
+
+ This PDF file is optimized for screen viewing, but may easily be
+ recompiled for printing. Please consult the preamble of the \LaTeX\
+ source file for instructions.
+}
+%% Re-set if ForPrinting=true
+\ifthenelse{\boolean{ForPrinting}}{%
+ \renewcommand{\ChapterSpace}{\vspace*{1in}}
+ \renewcommand{\Margins}{hmarginratio=2:3} % Asymmetric margins
+ \renewcommand{\HLinkColor}{black} % Hyperlink color
+ \renewcommand{\PDFPageLayout}{TwoPageRight}
+ \renewcommand{\TransNoteText}{%
+ \TransNoteCommon
+
+ This PDF file is optimized for printing, but may easily be
+ recompiled for screen viewing. Please consult the preamble of the
+ \LaTeX\ source file for instructions.
+ }
+}{% If ForPrinting=false, don't skip to recto
+ \renewcommand{\cleardoublepage}{\clearpage}
+ \raggedbottom
+}
+%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
+%%%% End of PRINTING/SCREEN VIEWING code; back to packages %%%%
+%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
+
+\ifthenelse{\boolean{ForPrinting}}{%
+ \setlength{\paperwidth}{8.5in}%
+ \setlength{\paperheight}{11in}%
+ \usepackage[body={4.25in,7.5in},\Margins]{geometry}[2002/07/08]
+}{%
+ \setlength{\paperwidth}{4.5in}%
+ \setlength{\paperheight}{6in}%
+ \usepackage[body={4.25in,5in},\Margins,includeheadfoot]{geometry}[2002/07/08]
+}
+
+\providecommand{\ebook}{00000} % Overridden during white-washing
+\usepackage[pdftex,
+ hyperfootnotes=false,
+ pdftitle={The Project Gutenberg eBook \#\ebook: The tides and kindred phenomena in the solar system},
+ pdfauthor={George Howard Darwin},
+ pdfkeywords={Andrew D. Hwang, Bryan Ness,
+ Project Gutenberg Online Distributed Proofreading Team,
+ Internet Archive/American Libraries},
+ pdfstartview=Fit, % default value
+ pdfstartpage=1, % default value
+ pdfpagemode=UseNone, % default value
+ bookmarks=true, % default value
+ linktocpage=false, % default value
+ pdfpagelayout=\PDFPageLayout,
+ pdfdisplaydoctitle,
+ pdfpagelabels=true,
+ bookmarksopen=true,
+ bookmarksopenlevel=1,
+ colorlinks=true,
+ linkcolor=\HLinkColor]{hyperref}[2007/02/07]
+
+%%%% Fixed-width environment to format PG boilerplate %%%%
+\newenvironment{PGtext}{%
+\begin{alltt}
+\fontsize{8.1}{9.2}\ttfamily\selectfont}%
+{\end{alltt}}
+
+%%%% Global style parameters %%%%
+\makeindex
+
+% Running heads
+\renewcommand{\headrulewidth}{0pt}
+\setlength{\headheight}{15pt}
+
+% Float placement
+\renewcommand{\floatpagefraction}{0.75}
+\renewcommand{\topfraction}{0.95}
+\renewcommand{\bottomfraction}{0.95}
+
+% Misc.
+\widowpenalty=3000
+\setlength{\arraycolsep}{2pt}
+
+% Loosen horizontal spacing
+\setlength{\emergencystretch}{1.5em}
+\newcommand{\Loosen}{\spaceskip 0.375em plus 0.75em minus 0.25em}
+
+% "Scratch pad" for length calculations
+\newlength{\TmpLen}
+
+%% Parametrized vertical space %%
+\newcommand{\Strut}[1][12pt]{\rule{0pt}{#1}}
+
+%%%% Corrections and in-line transcriber's notes %%%%
+% In-line notes
+\newcommand{\DPnote}[1]{}
+% Errors
+\newcommand{\DPtypo}[2]{#2}
+% Stylistic changes
+\newcommand{\DPchg}[2]{#2}
+
+%%%% Running heads %%%%
+\newcommand{\FlushRunningHeads}{%
+ \clearpage
+ \pagestyle{fancy}
+ \fancyhf{}
+ \cleardoublepage
+ \thispagestyle{empty}
+
+ \ifthenelse{\boolean{ForPrinting}}
+ {\fancyhead[RO,LE]{\thepage}}
+ {\fancyhead[R]{\thepage}}
+}
+
+
+\newcommand{\SetCenterHeads}[2][]{%
+ \ifthenelse{\equal{#1}{}}{%
+ \fancyhead[C]{{\small \MakeUppercase{#2}}}%
+ }{%
+ \fancyhead[CE]{{\small \MakeUppercase{#1}}}%
+ \fancyhead[CO]{{\small \MakeUppercase{#2}}}%
+ }%
+}
+
+\newcommand{\BookMark}[2]{%
+ \phantomsection%
+ \pdfbookmark[#1]{#2}{#2}%
+}
+
+%%%% Major document divisions %%%%
+\newcommand{\FrontMatter}{%
+ \cleardoublepage
+ \frontmatter
+ \BookMark{-1}{Front Matter}
+}
+\newcommand{\PGBoilerPlate}{%
+ \pagenumbering{Alph}
+ \pagestyle{empty}
+ \BookMark{0}{PG Boilerplate}
+}
+\newcommand{\MainMatter}{%
+ \FlushRunningHeads
+ \mainmatter
+ \BookMark{-1}{Main Matter}
+}
+\newcommand{\BackMatter}{%
+ \FlushRunningHeads
+ \backmatter
+ \BookMark{-1}{Back Matter}
+}
+\newcommand{\PGLicense}{%
+ \FlushRunningHeads
+ \pagenumbering{Roman}
+ \BookMark{0}{PG License}
+ \SetCenterHeads{License}
+}
+
+\newcommand{\TranscribersNote}[1]{%
+ \begin{minipage}{0.85\textwidth}
+ \small
+ \BookMark{0}{Transcriber's Note}
+ \subsection*{\centering\normalfont\scshape\normalsize\TransNote}
+ #1
+ \end{minipage}
+}
+
+%%%% Table of Contents %%%%
+\newcommand{\Contents}{%
+ \FlushRunningHeads
+ \SetCenterHeads{Contents}
+ \BookMark{0}{Contents}
+ \thispagestyle{empty}
+ \Section{Contents}
+}
+
+\newcommand{\Illustrations}{%
+ \FlushRunningHeads
+ \SetCenterHeads{List of Illustrations}
+ \BookMark{0}{List of Illustrations}
+ \thispagestyle{empty}
+ \Section{List of Illustrations}
+}
+
+% Chapter entries
+\newcommand{\ToCChap}[2]{%
+ \subsection*{\centering\normalfont\small CHAPTER #1}
+ \subsubsection*{\centering\normalfont\footnotesize #2}
+}
+
+\newcommand{\ToCFont}{\small\raggedright}
+
+% Section(s) entries
+\newcommand{\ToCSect}[2]{%
+ \settowidth{\TmpLen}{ #2}%
+ \par\noindent\strut\parbox[b]{\textwidth-\TmpLen}{\ToCFont\hangindent2em%
+ #1~\dotfill}\makebox[\TmpLen][r]{#2}%
+}
+
+% Macro discards 3rd argument (original page number)
+\newcommand{\LoFFig}[3]{%
+%[** TN: Figures 8, 9 bundled together, Fig 13b needs ad hoc handling]
+ \smallskip\noindent\strut\ifthenelse{\equal{#1}{8, 9}}{% Fig 8--9
+ \settowidth{\TmpLen}{ \PgNo[fig]{8}}%
+ \parbox[b]{\textwidth-\TmpLen}{\ToCFont\hangindent3em%
+ \PadTxt[r]{99}{8}, 9.~\textsc{#2} \dotfill}\makebox[\TmpLen][r]{\PgNo[fig]{8}}%
+ }{% Fig 13b
+ \ifthenelse{\equal{#1}{}}{%
+ \settowidth{\TmpLen}{ \PgNo[fig]{13b}}%
+ \parbox[b]{\textwidth-\TmpLen}{\ToCFont\hangindent3em%
+ \PadTxt[r]{99.}{} \textsc{#2} \dotfill}\makebox[\TmpLen][r]{\PgNo[fig]{13b}}%
+ }{% Generic handling
+ \settowidth{\TmpLen}{ \PgNo[fig]{#1}}%
+ \parbox[b]{\textwidth-\TmpLen}{\ToCFont\hangindent3em%
+ \PadTxt[r]{99}{#1}.~\textsc{#2} \dotfill}\makebox[\TmpLen][r]{\PgNo[fig]{#1}}%
+ }%
+ }%
+}
+
+% Page numbers
+% ** Approximate by default (points to original page separators)
+\newcommand{\PgNo}[2][page]{\pageref{#1:#2}}
+
+\newcommand{\PgNos}[3][--]{%
+ \ifthenelse{\equal{\pageref{page:#2}}{\pageref{page:#3}}}{%
+ \PgNo{#2}%
+ }{%
+ \ifthenelse{\equal{#1}{--}}{%
+ \PgNo{#2}#1\PgNo{#3}%
+ }{%
+ \PgNo{#2}\PadTxt[l]{--}{#1}\PgNo{#3}%
+ }%
+ }%
+}
+
+%% Index formatting
+\makeindex
+\makeatletter
+\renewcommand{\@idxitem}{\par\hangindent 30\p@\global\let\idxbrk\nobreak}
+\renewcommand\subitem{\idxbrk\@idxitem \hspace*{12\p@}\let\idxbrk\relax}
+\renewcommand{\indexspace}{\par\penalty-3000 \vskip 10pt plus5pt minus3pt\relax}
+
+\renewenvironment{theindex}{%
+ \setlength\columnseprule{0.5pt}\setlength\columnsep{18pt}%
+ \FlushRunningHeads
+ \SetCenterHeads{Index}
+ \BookMark{0}{Index}
+ \begin{multicols}{2}[\Section{Index}\small]% ** N.B. font size
+ \setlength\parindent{0pt}\setlength\parskip{0pt plus 0.3pt}%
+ \thispagestyle{empty}\let\item\@idxitem\raggedright%
+ }{%
+ \end{multicols}\FlushRunningHeads
+}
+\makeatother
+
+%%%% Document Sectioning %%%%
+\newcommand{\Section}[1]{\section*{\centering\normalfont\MakeUppercase{#1}}}
+
+\newcommand{\Preface}{%
+ \FlushRunningHeads
+ \SetCenterHeads{Preface}
+ \Section{Preface}}
+
+% \Chapter[Running head]{Number}{Heading title}
+\newcommand{\Chapter}[3][]{%
+ \FlushRunningHeads
+ \phantomsection
+ \label{chapter:#2}
+ \thispagestyle{plain}
+ \ifthenelse{\equal{#1}{}}{%
+ \SetCenterHeads[Chapter #2]{#3}%
+ \BookMark{0}{Ch. #2: #3}%
+ }{%
+ \SetCenterHeads[Chapter #2]{#1}%
+ \BookMark{0}{Ch. #2: #1}%
+ }
+ \ChapterSpace
+ \ifthenelse{\equal{#2}{I}}{%
+ \section*{\centering\normalfont\Large THE TIDES}
+ }{}
+ \ifthenelse{\equal{#2}{III}\or\equal{#2}{XVI}}{
+ \section*{\centering\normalfont CHAPTER #2\footnotemark}
+ }{%
+ \section*{\centering\normalfont CHAPTER #2}
+ }
+ \subsection*{\centering\normalfont\small\MakeUppercase{#3}}
+}
+
+\newcommand{\Signature}[2]{%
+ \null\hfill#1\hspace*{\parindent}\\
+ \hspace*{\parindent}#2%
+}
+
+%%%% Misc. textual macros %%%%
+\newcommand{\First}[1]{\textsc{#1}}
+\newcommand{\Title}[1]{\textit{#1}}
+
+\newcommand{\eg}{e.\;g.}
+\newcommand{\ie}{i.\;e.}
+\newcommand{\NB}{N.\;B.}
+
+\renewcommand{\bar}[1]{%
+ \ifthenelse{\equal{#1}{i}}{%
+ \ensuremath{\overline{\PadTxt{ }{\i}}}%
+ }{%
+ \ensuremath{\overline{\PadTxt{r}{#1}}}%
+ }%
+}
+
+\newcommand{\ieme}{\textsuperscript{e}}
+\newcommand{\iere}{\textsuperscript{re}}
+
+% For small caps A.M. and P.M. Usage: 11:30 o'clock, \PM,
+% If followed by a space, use a backslash, e.g. By 6:00~\AM\ the sun will rise.
+\newcommand{\SCAbbrev}[3]{%
+ \ifthenelse{\equal{#3}{.}}%
+ {\textsc{\MakeLowercase{#1.#2}.}}%
+ {\textsc{\MakeLowercase{#1.#2}.}\@#3}%
+}
+
+\newcommand{\AM}[1]{\SCAbbrev{A}{M}{#1}}
+\newcommand{\PM}[1]{\SCAbbrev{P}{M}{#1}}
+
+\newcommand{\AD}[1]{\SCAbbrev{A}{D}{#1}}
+
+%%%% Illustrations %%%%
+% Inclusion wrapper: \Graphics[ext]{width}{filename}
+\newcommand{\Graphic}[3][png]{\includegraphics[width=#2]{./images/#3.#1}}
+
+%\Figure[width]{Figure number}{Caption}{ext}
+\newcommand{\Figure}[4][1.0]{%
+ \begin{figure}[hbtp!]
+ \centering
+ \Graphic[#4]{#1\textwidth}{fig#2}
+ \ifthenelse{\equal{#2}{8}}{%
+ \caption{Fig.~8, 9.---#3}
+ \label{fig:8}\label{fig:9}
+ }{%
+ \ifthenelse{\not\equal{#3}{}}{%
+ \caption{Fig.~#2.---#3}%
+ }{%
+ \caption{Fig.~#2.}%
+ }%
+ \label{fig:#2}
+ }
+ \end{figure}\ignorespaces%
+}
+
+\newcommand{\TallFig}[4][0.95]{%
+ \begin{figure}[hbtp!]
+ \centering
+ \includegraphics[height=#1\textheight]{./images/fig#2.#4}%
+ \caption{Fig.~#2.---#3}%
+ \label{fig:#2}
+ \end{figure}\ignorespaces%
+}
+
+\newenvironment{Authorities}[1][Authorities]{%
+\subsection*{\centering\normalfont\small\scshape #1.}
+\small}{\normalsize}
+
+\newcommand{\Journal}{\medskip}
+\newcommand{\Paper}[1]{\hspace*{\parindent}\hangindent3\parindent#1\par}
+
+\newenvironment{Remark}{\medskip\footnotesize}{\medskip\normalsize}
+
+%%%% Cross-referencing %%%%
+
+%% Anchors
+\newcommand{\PageSep}[1]{\PageLabel{#1}\ignorespaces}
+\newcommand{\PageLabel}[2][page]{\phantomsection\label{#1:#2}}
+
+%% Links
+% Used only once; \PageRef{p.}{329}
+\newcommand{\PageRef}[2]{\hyperref[pg:#2]{#1~\pageref*{pg:#2}}}
+
+\newcommand{\figref}[1]{\hyperref[fig:#1]{#1}}
+\newcommand{\fig}[1]{\hyperref[fig:#1]{fig.~#1}}
+\newcommand{\Fig}[1]{\hyperref[fig:#1]{Fig.~#1}}
+
+% Chapter/appendix reference
+\newcommand{\Ref}[2]{%
+ \ifthenelse{\not\equal{#1}{}}{%
+ \hyperref[chapter:#2]{#1~#2}%
+ }{%
+ \hyperref[chapter:#2]{#2}%
+ }%
+}
+
+%%%% Typographical conveniences %%%%
+\newcommand{\TB}{\bigskip}
+\newcommand{\Seg}[3][]{\ifthenelse{\not\equal{#1}{}}{#1\;}{}#2\;#3}
+
+\newcommand{\Z}{\phantom{0}}
+
+\newcommand{\TEntry}[2][1.25in]{%
+ \parbox[c]{#1}{\centering\footnotesize #2}%
+}
+
+\newcommand{\ColHead}[2][1.25in]{%
+ \TEntry[#1]{\medskip#2\medskip}%
+}
+
+\newcommand{\Dash}{\text{---}}
+\newcommand{\mm}{^{\text{m}}}
+
+% \PadTxt[#1]{#2}{#3} sets #3 in a box of width #2, aligned at #1 (default [c])
+\newcommand{\PadTxt}[3][c]{%
+ \settowidth{\TmpLen}{\text{#2}}%
+ \makebox[\TmpLen][#1]{#3}%
+}
+
+\newcommand{\Ditto}[1]{\PadTxt{#1}{``}}%''
+
+%%%% Misc. mathematical setup %%%%
+\SetSymbolFont{operators}{normal}{T1}{cmsc}{m}{n}
+%% Upright small-cap letters in math mode
+\DeclareMathSymbol{A}{\mathalpha}{operators}{`a}
+\DeclareMathSymbol{B}{\mathalpha}{operators}{`b}
+\DeclareMathSymbol{C}{\mathalpha}{operators}{`c}
+\DeclareMathSymbol{D}{\mathalpha}{operators}{`d}
+\DeclareMathSymbol{E}{\mathalpha}{operators}{`e}
+\DeclareMathSymbol{F}{\mathalpha}{operators}{`f}
+\DeclareMathSymbol{G}{\mathalpha}{operators}{`g}
+\DeclareMathSymbol{I}{\mathalpha}{operators}{`i}
+\DeclareMathSymbol{M}{\mathalpha}{operators}{`m}
+\DeclareMathSymbol{P}{\mathalpha}{operators}{`p}
+\DeclareMathSymbol{V}{\mathalpha}{operators}{`v}
+
+% Handle degree symbols and centered dots as Latin-1 characters
+\DeclareInputText{176}{\ifmmode{{}^\circ}\else\textdegree\fi}
+\DeclareInputText{183}{\ifmmode\cdot\else\textperiodcentered\fi}
+
+%%%%%%%%%%%%%%%%%%%%%%%% START OF DOCUMENT %%%%%%%%%%%%%%%%%%%%%%%%%%
+\begin{document}
+\FrontMatter
+%%%% PG BOILERPLATE %%%%
+\PGBoilerPlate
+\begin{center}
+\begin{minipage}{\textwidth}
+\small
+\begin{PGtext}
+The Project Gutenberg EBook of The Tides and Kindred Phenomena in the
+Solar System, by Sir George Howard Darwin
+
+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.net
+
+
+Title: The Tides and Kindred Phenomena in the Solar System
+ The Substance of Lectures Delivered in 1897 at the Lowell
+ Institute, Boston, Massachusetts
+
+Author: Sir George Howard Darwin
+
+Release Date: January 31, 2012 [EBook #38722]
+
+Language: English
+
+Character set encoding: ISO-8859-1
+
+*** START OF THIS PROJECT GUTENBERG EBOOK TIDES AND KINDRED PHENOMENA ***
+\end{PGtext}
+\end{minipage}
+\end{center}
+\clearpage
+
+%%%% Credits and transcriber's note %%%%
+\begin{center}
+\begin{minipage}{\textwidth}
+\begin{PGtext}
+Produced by Andrew D. Hwang, Bryan Ness and the Online
+Distributed Proofreading Team at http://www.pgdp.net (This
+file was produced from images generously made available
+by The Internet Archive/Canadian Libraries)
+\end{PGtext}
+\end{minipage}
+\vfill
+\TranscribersNote{\TransNoteText}
+\end{center}
+%%%%%%%%%%%%%%%%%%%%%%%%%%% FRONT MATTER %%%%%%%%%%%%%%%%%%%%%%%%%%
+\cleardoublepage
+\pagenumbering{roman}
+\PageSep{iii}
+\begin{center}
+\textbf{\Huge THE TIDES}
+\vfil
+\textbf{\large AND KINDRED PHENOMENA IN THE \\[4pt]
+SOLAR SYSTEM}
+\vfil
+{\footnotesize THE SUBSTANCE OF LECTURES DELIVERED \\[4pt]
+IN 1897 AT THE LOWELL INSTITUTE, \\[4pt]
+BOSTON, MASSACHUSETTS}
+\vfil
+BY
+\vfil
+\textbf{\Large GEORGE HOWARD DARWIN}
+\vfil
+\scriptsize PLUMIAN PROFESSOR AND FELLOW OF TRINITY COLLEGE IN THE \\[4pt]
+UNIVERSITY OF CAMBRIDGE
+\vfil\vfil
+\Graphic[png]{1.25in}{riverside}
+\vfil\vfil\vfil
+\footnotesize BOSTON AND NEW YORK \\
+\normalsize HOUGHTON, MIFFLIN AND COMPANY \\
+\textgoth{The Riverside Press, Cambridge} \\
+1899
+\end{center}
+\PageSep{iv}
+\clearpage
+\null\vfill
+\begin{center}
+\footnotesize
+COPYRIGHT, 1898, BY GEORGE HOWARD DARWIN \\
+ALL RIGHTS RESERVED
+\end{center}
+\vfill
+\PageSep{v}
+
+
+\Preface
+
+\First{In} 1897 I delivered a course of lectures on
+the Tides at the Lowell Institute in Boston,
+Massachusetts, and this book contains the substance
+of what I then said. The personal form
+of address appropriate to a lecture is, I think,
+apt to be rather tiresome in a book, and I have
+therefore taken pains to eliminate all traces of
+the lecture from what I have written.
+
+A mathematical argument is, after all, only
+organized common sense, and it is well that men
+of science should not always expound their work
+to the few behind a veil of technical language,
+but should from time to time explain to a larger
+public the reasoning which lies behind their
+mathematical notation. To a man unversed in
+popular exposition it needs a great effort to shell
+away the apparatus of investigation and the
+technical mode of speech from the thing behind
+it, and I owe a debt of gratitude to Mr.~Lowell,
+trustee of the Institute, for having afforded me
+the occasion for making that effort.
+\PageSep{vi}
+
+It is not unlikely that the first remark of
+many who see my title will be that so small a
+subject as the Tides cannot demand a whole volume;
+but, in fact, the subject branches out in
+so many directions that the difficulty has been to
+attain to the requisite compression of my matter.
+Many popular works on astronomy devote a few
+pages to the Tides, but, as far as I know, none
+of these books contain explanations of the practical
+methods of observing and predicting the
+Tides, or give any details as to the degree of
+success attained by tidal predictions. If these
+matters are of interest, I invite my readers not
+to confine their reading to this preface. The
+later chapters of this book are devoted to the
+consideration of several branches of speculative
+Astronomy, with which the theory of the Tides
+has an intimate relationship. The problems involved
+in the origin and history of the solar
+and of other celestial systems have little bearing
+upon our life on the earth, yet these questions
+can hardly fail to be of interest to all those
+whose minds are in any degree permeated by
+the scientific spirit.
+
+I think that there are many who would like to
+understand the Tides, and will make the attempt
+to do so provided the exposition be sufficiently
+\PageSep{vii}
+simple and clear; it is to such readers I address
+this volume. It is for them to say how far I
+have succeeded in rendering these intricate subjects
+interesting and intelligible, but if I have
+failed it has not been for lack of pains.
+
+The figures and diagrams have, for the most
+part, been made by Mr.~Edwin Wilson of Cambridge,
+but I have to acknowledge the courtesy
+of the proprietors of \Title{Harper's}, the \Title{Century},
+and the \Title{Atlantic Monthly} magazines, in supplying
+me with some important illustrations.
+
+A considerable portion of \Ref{Chapter}{III}.\ on the
+``Bore'' is to appear as an article in the \Title{Century
+Magazine} for October, 1898, and the reproductions
+of Captain Moore's photographs of the
+``Bore'' in the Tsien-Tang-Kiang have been
+prepared for that article. The \Title{Century} has also
+kindly furnished the block of Dr.~Isaac Roberts's
+remarkable photograph of the great nebula in
+the constellation of Andromeda; it originally
+appeared in an article on Meteorites in the number
+for October,~1890. The greater portion of
+the text and the whole of the illustrations of
+\Ref{Chapter}{XX}.\ were originally published in \Title{Harper's
+Magazine} for June,~1889. Lastly, portions
+of Chapters \Ref{}{XV}.~and~\Ref{}{XVI}.\ appeared in
+the \Title{Atlantic Monthly} for April, 1898, published
+\PageSep{viii}
+by Messrs.\ Houghton, Mifflin~\&~Co., who also
+make themselves responsible for the publication
+of the American edition of this book.
+
+In conclusion, I wish to take this opportunity
+of thanking my American audience for the cordiality
+of their reception, and my many friends
+across the Atlantic for their abundant hospitality
+and kindness.
+
+\Signature{G. H. DARWIN.}
+{\textsc{Cambridge}, \textit{August}, 1898.}
+\PageSep{ix}
+
+
+\Contents
+
+\ToCChap{I}
+{TIDES AND METHODS OF OBSERVATION}
+
+\ToCSect{Definition of tide}{\PgNos{1}{3}}
+\ToCSect{Oceanic tides}{\PgNos[,]{4}{5}}
+\ToCSect{Methods of observation}{\PgNo{6}}
+\ToCSect{Tide-gauge}{\PgNos{7}{12}}
+\ToCSect{Tide-curve}{\PgNo{12}}
+\ToCSect{Site for tide-gauge}{\PgNo{13}}
+\ToCSect{Irregularities in tide-curve}{\PgNos[,]{14}{15}}
+\ToCSect{Authorities}{\PgNo{16}}
+
+
+\ToCChap{II}
+{SEICHES IN LAKES}
+
+\ToCSect{Meaning of seiche}{\PgNo{17}}
+\ToCSect{Uses of scientific apparatus}{\PgNo{18}}
+\ToCSect{Forel's plemyrameter}{\PgNos[,]{19}{20}}
+\ToCSect{Records of the level of the lake}{\PgNo{20}}
+\ToCSect{Interpretation of record}{\PgNos{21}{23}}
+\ToCSect{Limnimeter}{\PgNo{24}}
+\ToCSect{Mode of oscillation in seiches}{\PgNos{25}{28}}
+\ToCSect{Wave motion in deep and in shallow water}{\PgNos{29}{32}}
+\ToCSect{Composition of waves}{\PgNos{32}{36}}
+\ToCSect{Periods of seiches}{\PgNos[,]{37}{38}}
+\ToCSect{Causes of seiches}{\PgNos[,]{39}{40}}
+\ToCSect{Vibrations due to wind and to steamers}{\PgNos{41}{47}}
+\ToCSect{Aerial waves and their action on lakes and on the sea}{\PgNos{48}{53}}
+\ToCSect{Authorities}{\PgNos[,]{53}{54}}
+\PageSep{x}
+
+
+\ToCChap{III}
+{TIDES IN RIVERS---TIDE MILLS}
+
+\ToCSect{Definition of ebb and flow}{\PgNo{56}}
+\ToCSect{Tidal currents in rivers}{\PgNos{56}{58}}
+\ToCSect{Progressive change of wave in shallow water}{\PgNos[,]{58}{59}}
+\ToCSect{The bore}{\PgNo{59}}
+\ToCSect{Captain Moore's survey of the Tsien-Tang-Kiang}{\PgNos{60}{64}}
+\ToCSect{Diagram of water-levels during the flow}{\PgNos[,]{64}{65}}
+\ToCSect{Chinese superstition}{\PgNos{68}{71}}
+\ToCSect{Pictures of the bore}{\PgNo{69}}
+\ToCSect{Other cases of bores}{\PgNo{71}}
+\ToCSect{Causes of the bore}{\PgNo{72}}
+\ToCSect{Tidal energy}{\PgNos[,]{73}{74}}
+\ToCSect{Tide mills}{\PgNos[,]{74}{75}}
+\ToCSect{Authorities}{\PgNo{75}}
+
+
+\ToCChap{IV}
+{HISTORICAL SKETCH}
+
+\ToCSect{Theories of the Chinese}{\PgNos[,]{76}{77}}
+\ToCSect{Theories of the Arabs}{\PgNos{77}{79}}
+\ToCSect{Theories of the Norsemen}{\PgNos[,]{79}{80}}
+\ToCSect{Writings of Posidonius and Strabo}{\PgNos{80}{84}}
+\ToCSect{Seleucus the Babylonian on the diurnal inequality}{\PgNos[,]{84}{85}}
+\ToCSect{Galileo and Kepler}{\PgNo{85}}
+\ToCSect{Newton and his successors}{\PgNos{86}{88}}
+\ToCSect{Empirical method of tidal prediction}{\PgNos{88}{90}}
+\ToCSect{Authorities}{\PgNo{90}}
+
+
+\ToCChap{V}
+{TIDE-GENERATING FORCE}
+
+\ToCSect{Inertia and centrifugal force}{\PgNos{91}{93}}
+\ToCSect{Orbital motion of earth and moon}{\PgNos{93}{95}}
+\ToCSect{Tide-generating force}{\PgNos{96}{100}}
+\ToCSect{Law of its dependence on the moon's distance}{\PgNos{101}{103}}
+\ToCSect{Earth's rotation}{\PgNos[,]{103}{104}}
+\PageSep{xi}
+\ToCSect{Second explanation of tide-generating force}{\PgNos[,]{104}{105}}
+\ToCSect{Horizontal tide-generating force}{\PgNos[,]{105}{106}}
+\ToCSect{Successive changes of force in the course of a day}{\PgNos[,]{107}{108}}
+\ToCSect{Authorities}{\PgNo{108}}
+
+
+\ToCChap{VI}
+{DEFLECTION OF THE VERTICAL}
+
+\ToCSect{Deflection of a pendulum by horizontal tidal force}{\PgNos{109}{111}}
+\ToCSect{Path pursued by a pendulum under tidal force}{\PgNos{111}{113}}
+\ToCSect{Object of measuring the deflection of a pendulum}{\PgNos{113}{115}}
+\ToCSect{Attempt to measure deflection by bifilar pendulum}{\PgNos{115}{125}}
+\ToCSect{Microsisms}{\PgNos{125}{127}}
+\ToCSect{The microphone as a seismological instrument}{\PgNos{127}{130}}
+\ToCSect{Paschwitz's work with the horizontal pendulum}{\PgNos[,]{131}{132}}
+\ToCSect{Supposed measurement of the lunar deflection of gravity}{\PgNo{132}}
+\ToCSect{Authorities}{\PgNo{133}}
+
+
+\ToCChap{VII}
+{THE ELASTIC DISTORTION OF THE EARTH'S SURFACE BY
+VARYING LOADS}
+
+\ToCSect{Distortion of an elastic surface by superincumbent load}{\PgNos{134}{137}}
+\ToCSect{Application of the theory to the earth}{\PgNos[,]{137}{138}}
+\ToCSect{Effects of tidal load}{\PgNos{138}{140}}
+\ToCSect{Probable deflections at various distances from the coast}{\PgNos{140}{143}}
+\ToCSect{Deflections observed by M.~d'Abbadie and by Dr.~Paschwitz}{\PgNos{143}{145}}
+\ToCSect{Effects of atmospheric pressure on the earth's surface}{\PgNos{145}{147}}
+\ToCSect{Authorities}{\PgNo{148}}
+
+
+\ToCChap{VIII}
+{EQUILIBRIUM THEORY OF TIDES}
+
+\ToCSect{Explanation of the figure of equilibrium}{\PgNos{149}{151}}
+\ToCSect{Map of equilibrium tide}{\PgNos{151}{153}}
+\ToCSect{Tides according to the equilibrium theory}{\PgNos{153}{156}}
+\ToCSect{Solar tidal force compared with lunar}{\PgNos{156}{158}}
+\PageSep{xii}
+\ToCSect{Composition of lunar and solar tides}{\PgNos[,]{158}{159}}
+\ToCSect{Points of disagreement between theory and fact}{\PgNos{159}{162}}
+\ToCSect{Authorities}{\PgNo{162}}
+
+
+\ToCChap{IX}
+{DYNAMICAL THEORY OF THE TIDE WAVE}
+
+\ToCSect{Free and forced waves in an equatorial canal}{\PgNos{163}{165}}
+\ToCSect{Critical depth of canal}{\PgNos{165}{167}}
+\ToCSect{General principle as to free and forced oscillations}{\PgNos{167}{174}}
+\ToCSect{Inverted and direct oscillation}{\PgNos[,]{172}{173}}
+\ToCSect{Canal in latitude~$60°$}{\PgNos[,]{174}{175}}
+\ToCSect{Tides where the planet is partitioned into canals}{\PgNo{175}}
+\ToCSect{Removal of partitions; vortical motion of the water}{\PgNos[,]{176}{177}}
+\ToCSect{Critical latitude where the rise and fall vanish}{\PgNos[,]{177}{178}}
+\ToCSect{Diurnal inequality}{\PgNos{178}{180}}
+\ToCSect{Authorities}{\PgNo{181}}
+
+
+\ToCChap{X}
+{TIDES IN LAKES---COTIDAL CHART}
+
+\ToCSect{The tide in a lake}{\PgNos{182}{185}}
+\ToCSect{The Mediterranean Sea}{\PgNos[,]{185}{186}}
+\ToCSect{Derived tide of the Atlantic}{\PgNos{186}{188}}
+\ToCSect{Cotidal chart}{\PgNos{188}{192}}
+\ToCSect{Authorities}{\PgNo{192}}
+
+
+\ToCChap{XI}
+{HARMONIC ANALYSIS OF THE TIDE}
+
+\ToCSect{Tide in actual oceans due to single equatorial satellite}{\PgNos{193}{196}}
+\ToCSect{Substitution of ideal satellites for the moon}{\PgNos{197}{199}}
+\ToCSect{Partial tide due to each ideal satellite}{\PgNos{199}{201}}
+\ToCSect{Three groups of partial tides}{\PgNo{201}}
+\ToCSect{Semidiurnal group}{\PgNos{201}{204}}
+\ToCSect{Diurnal group}{\PgNos{204}{206}}
+\ToCSect{Meteorological tides}{\PgNos[,]{206}{207}}
+\ToCSect{Shallow water tides}{\PgNos{207}{210}}
+\ToCSect{Authorities}{\PgNo{210}}
+\PageSep{xiii}
+
+
+\ToCChap{XII}
+{REDUCTION OF TIDAL OBSERVATIONS}
+
+\ToCSect{Method of singling out a single partial tide}{\PgNos{211}{214}}
+\ToCSect{Variety of plans adopted}{\PgNos{214}{217}}
+\ToCSect{Tidal abacus}{\PgNos{217}{220}}
+\ToCSect{Authorities}{\PgNo{220}}
+
+
+\ToCChap{XIII}
+{TIDE TABLES}
+
+\ToCSect{Definition of special and general tables}{\PgNo{221}}
+\ToCSect{Reference to moon's transit}{\PgNos[,]{222}{223}}
+\ToCSect{Examples at Portsmouth and at Aden}{\PgNos{223}{228}}
+\ToCSect{General inadequacy of tidal information}{\PgNos[,]{229}{230}}
+\ToCSect{Method of calculating tide tables}{\PgNos{230}{233}}
+\ToCSect{Tide-predicting machine}{\PgNos{233}{241}}
+\ToCSect{Authorities}{\PgNo{241}}
+
+
+\ToCChap{XIV}
+{THE DEGREE OF ACCURACY OF TIDAL PREDICTION}
+
+\ToCSect{Effects of wind and barometric pressure}{\PgNos[,]{242}{243}}
+\ToCSect{Errors at Portsmouth}{\PgNos[,]{243}{244}}
+\ToCSect{Errors at Aden}{\PgNos{245}{249}}
+\ToCSect{Authorities}{\PgNo{250}}
+
+
+\ToCChap{XV}
+{CHANDLER'S NUTATION---THE RIGIDITY OF THE EARTH}
+
+\ToCSect{Nutation of the earth and variation of latitude}{\PgNos{251}{254}}
+\ToCSect{Elasticity of the earth}{\PgNos[,]{254}{255}}
+\ToCSect{Tide due to variation of latitude}{\PgNos[,]{255}{256}}
+\ToCSect{Rigidity of the earth}{\PgNos{256}{260}}
+\ToCSect{Transmission of earthquake shocks}{\PgNos[,]{261}{262}}
+\ToCSect{Authorities}{\PgNos[,]{262}{263}}
+\PageSep{xiv}
+
+
+\ToCChap{XVI}
+{TIDAL FRICTION}
+
+\ToCSect{Friction retards the tide}{\PgNos{264}{267}}
+\ToCSect{Retardation of planetary rotation}{\PgNos{267}{269}}
+\ToCSect{Reaction on the satellite}{\PgNos{269}{272}}
+\ToCSect{Ancient eclipses of the sun}{\PgNos[,]{272}{273}}
+\ToCSect{Law of variation of tidal friction with moon's distance}{\PgNos{273}{275}}
+\ToCSect{Transformations of the month and of the day}{\PgNos{275}{280}}
+\ToCSect{Initial and final conditions of motion}{\PgNos[,]{280}{281}}
+\ToCSect{Genesis of the moon}{\PgNos{281}{285}}
+\ToCSect{Minimum time requisite for the evolution}{\PgNos[,]{285}{286}}
+\ToCSect{Rotation of the moon}{\PgNos[,]{286}{287}}
+\ToCSect{The month ultimately to be shorter than the day}{\PgNos{287}{289}}
+
+
+\ToCChap{XVII}
+{TIDAL FRICTION (\textit{continued})}
+
+\ToCSect{Discovery of the Martian satellites}{\PgNos{290}{298}}
+\ToCSect{Rotation of Mercury, of Venus, and of the Jovian satellites}{\PgNos[,]{298}{299}}
+\ToCSect{Adaptation of the earth's figure to changed rotation}{\PgNos{299}{302}}
+\ToCSect{Ellipticity of the internal strata of the earth}{\PgNos{302}{304}}
+\ToCSect{Geological evidence}{\PgNos{304}{306}}
+\ToCSect{Distortion of a plastic planet and trend of continents}{\PgNos{306}{308}}
+\ToCSect{Obliquity of the ecliptic}{\PgNos{308}{312}}
+\ToCSect{Eccentricity of lunar orbit}{\PgNos[,]{312}{313}}
+\ToCSect{Eccentricity of the orbits of double stars}{\PgNo{313}}
+\ToCSect{Plane of the lunar orbit}{\PgNos[,]{313}{314}}
+\ToCSect{Short summary}{\PgNos[,]{314}{315}}
+\ToCSect{Authorities}{\PgNo{315}}
+
+
+\ToCChap{XVIII}
+{THE FIGURES OF EQUILIBRIUM OF A ROTATING MASS OF
+LIQUID}
+
+\ToCSect{Plateau's experiment}{\PgNos{316}{319}}
+\ToCSect{Stability of a celestial sphere of liquid}{\PgNos{319}{321}}
+\ToCSect{The two ellipsoids of Maclaurin and that of Jacobi}{\PgNos{321}{323}}
+\PageSep{xv}
+\ToCSect{Transitions with change of rotation}{\PgNos[,]{323}{324}}
+\ToCSect{Coalescence of Jacobi's with Maclaurin's ellipsoid}{\PgNos{324}{326}}
+\ToCSect{Poincaré's law of stability and coalescence}{\PgNos[,]{326}{327}}
+\ToCSect{Poincaré's pear-shaped figure}{\PgNos[,]{327}{328}}
+\ToCSect{Hour-glass figure of equilibrium}{\PgNos{328}{332}}
+\ToCSect{Figures of planets}{\PgNos[,]{332}{333}}
+\ToCSect{Authorities}{\PgNo{333}}
+
+
+\ToCChap{XIX}
+{THE EVOLUTION OF CELESTIAL SYSTEMS}
+
+\ToCSect{The Nebular Hypothesis}{\PgNos{334}{338}}
+\ToCSect{Nebula in Andromeda}{\PgNos[,]{338}{339}}
+\ToCSect{Distribution of satellites in the solar system}{\PgNos{339}{341}}
+\ToCSect{Genesis of celestial bodies by fission}{\PgNo{342}}
+\ToCSect{Dr.~See's speculations as to systems of double stars}{\PgNos{342}{344}}
+\ToCSect{Diversity of celestial bodies}{\PgNos{344}{346}}
+\ToCSect{Authorities}{\PgNo{346}}
+
+
+\ToCChap{XX}
+{SATURN'S RINGS}
+
+\ToCSect{Description}{\PgNos[,]{347}{348}}
+\ToCSect{Discovery of Saturn's rings}{\PgNos{348}{352}}
+\ToCSect{Diagram of the rings}{\PgNos{353}{356}}
+\ToCSect{Roche's investigation}{\PgNos[,]{356}{357}}
+\ToCSect{Roche's limit}{\PgNos{358}{360}}
+\ToCSect{The limit for the several planets}{\PgNos[,]{360}{361}}
+\ToCSect{Meteoric constitution of Saturn's rings}{\PgNos[,]{361}{362}}
+\ToCSect{Maxwell's investigations}{\PgNos{362}{367}}
+\ToCSect{Spectroscopic examination of the rings}{\PgNos{367}{369}}
+\ToCSect{Authorities}{\PgNo{369}}
+\PageSep{xvi}
+%[Blank Page]
+\PageSep{xvii}
+
+
+% [** TN: Plates (Figs. 23, 33, and 40) listed separately in the original;
+% omitted separate "FULL-PAGE" and "IN TEXT" headings, folded plate
+% references into main illustration list.]
+% LIST OF ILLUSTRATIONS
+\Illustrations
+
+\LoFFig{1}{Well for Tide-Gauge}{7}
+\LoFFig{2}{Pipe of Tide-Gauge}{9}
+\LoFFig{3}{Indian Tide-Gauge}{10}
+\LoFFig{4}{Légé's Tide-Gauge}{11}
+\LoFFig{5}{Bombay Tide-Curve from Noon, April 22, to Noon, April 30, 1884}{14}
+\LoFFig{6}{Sites for a Tide-Gauge}{15}
+\LoFFig{7}{Plemyrameter}{20}
+\LoFFig{8, 9}{Records of Seiches at Évian}{23}
+\LoFFig{10}{Map of Lake of Geneva}{26}
+\LoFFig{11}{Wave in Deep Water}{30}
+\LoFFig{12}{Wave in Shallow Water}{31}
+\LoFFig{13}{Simple Wave}{33}
+\LoFFig{}{Composition of Two Equal and Opposite Waves}{34}
+\LoFFig{14}{Vibrations due to Steamers}{45}
+\LoFFig{15}{Progressive Change of a Wave in Shallow Water}{59}
+\LoFFig{16}{Chart of the Estuary of the Tsien-Tang-Kiang}{61}
+\LoFFig{17}{Bore-Shelter on the Tsien-Tang-Kiang}{64}
+\LoFFig{18}{Diagram of the Flow of the Tide on the Tsien-Tang-Kiang}{66}
+\LoFFig{19}{Pictures of the Bore on the Tsien-Tang-Kiang}{69}
+\LoFFig{20}{Earth and Moon}{93}
+\PageSep{xviii}
+\LoFFig{21}{Revolution of a Body without Rotation}{98}
+\LoFFig{22}{Tide-generating Force}{100}
+\LoFFig{23}{Horizontal Tide-generating Force}{106}
+\LoFFig{24}{Deflection of a Pendulum; the Moon and Observer on the Equator}{111}
+\LoFFig{25}{Deflection of a Pendulum; the Moon in N. Declination~$15°$,
+the Observer in N. Latitude~$30°$}{112}
+\LoFFig{26}{Bifilar Pendulum}{115}
+\LoFFig{27}{Form of Dimple in an Elastic Surface}{135}
+\LoFFig{28}{Distortion of Land and Sea-Bed by Tidal Load}{139}
+\LoFFig{29}{Chart of Equilibrium Tides}{152}
+\LoFFig{30}{Forced Oscillations of a Pendulum}{171}
+\LoFFig{31}{The Tide in a Lake}{184}
+\LoFFig{32}{Chart of Cotidal Lines}{190}
+\LoFFig{33}{Tidal Abacus}{218}
+\LoFFig{34}{Curves of Intervals and Heights at Portsmouth and at Aden}{227}
+\LoFFig{35}{Diagram of Tide-predicting Instrument}{235}
+\LoFFig{36}{Frictionally Retarded Tide}{266}
+\LoFFig{37}{Maclaurin's and Jacobi's Ellipsoids of Equilibrium}{323}
+\LoFFig{38}{Figures of Equilibrium}{325}
+\LoFFig{39}{Hour-glass Figure of Equilibrium}{329}
+\LoFFig{40}{Nebula in Andromeda}{339}
+\LoFFig{41}{The Planet Saturn}{349}
+\LoFFig{42}{Diagram of Saturn and his Rings}{353}
+\LoFFig{43}{Roche's Figure of a Satellite when elongated to the utmost}{357}
+
+\MainMatter
+\PageSep{1}
+\index{Barometric pressure|see{Atmospheric pressure}}%
+\index{Bifilar|see{Pendulum}}%
+\index{D'Abbadie|see{Abbadie}}%
+\index{Horizontal tide-generating force|see{also Pendulum}}%
+\index{Rebeur|see{Paschwitz}}%
+\index{Thomson, Sir W.|see{Kelvin}}%
+\index{Tidal problem|see{Laplace, Harmonic Analysis, etc.}}%
+\index{Tide|see{also other headings; \eg\ for tide-generating force \textit{see} Force}}%
+\index{Vertical|see{Deflection}}%
+% [** TN: Text printed by \Chapter macro]
+% THE TIDES
+
+
+\Chapter{I}
+{Tides and Methods of Observation}
+
+\First{The} great wave caused by an earthquake is
+\index{Tide!definition|(}%
+often described in the newspapers as a tidal
+wave, and the same name is not unfrequently
+applied to such a short series of enormous waves
+as is occasionally encountered by a ship in the
+open sea. We must of course use our language
+in the manner which is most convenient, but as
+in this connection the adjective ``tidal'' implies
+simply greatness and uncommonness, the use of
+the term in such a sense cannot be regarded as
+appropriate.
+
+The word ``tidal'' should, I think, only be
+used when we are referring to regular and persistent
+alternations of rise and fall of sea-level.
+Even in this case the term may perhaps be used
+in too wide a sense, for in many places there is a
+regular alternation of the wind, which blows in-shore
+during the day and out during the night
+with approximate regularity, and such breezes
+\PageSep{2}
+alternately raise and depress the sea-level, and
+thus produce a sort of tide. Then in the Tropics
+there is a regularly alternating, though small,
+periodicity in the pressure of the atmosphere,
+which is betrayed by an oscillation in the height
+of the barometer. Now the ocean will respond
+to the atmospheric pressure, so that the sea-level
+will fall with a rising barometer, and rise
+with a falling barometer. Thus a regularly
+periodic rise and fall of the sea-level must result
+from this cause also. Again, the melting of the
+snows in great mountain ranges, and the annual
+variability in rainfall and evaporation, produce
+approximately periodic changes of level in the
+estuaries of rivers, and although the period of
+these changes is very long, extending as they do
+over the whole year, yet from their periodicity
+they partake of the tidal character.
+
+These changes of water level are not, however,
+tides in the proper sense of the term, and a true
+tide can only be adequately defined by reference
+to the causes which produce it. A tide, in fact,
+means a rising and falling of the water of the
+ocean caused by the attractions of the sun and
+moon.
+
+Although true tides are due to astronomical
+causes, yet the effects of regularly periodic winds,
+variation of atmospheric pressure, and rainfall
+are so closely interlaced with the true tide that
+in actual observation of the sea it is necessary to
+\PageSep{3}
+consider them both together. It is accordingly
+practically convenient to speak of any regular
+alternation of sea-level, due to the wind and to
+the other influences to which I have referred, as
+a Meteorological Tide. The addition of the adjective
+``meteorological'' justifies the use of the
+term ``tide'' in this connection.
+
+We live at the bottom of an immense sea of
+air, and if the attractions of the sun and moon
+affect the ocean, they must also affect the air.
+This effect will be shown by a regular rise and
+fall in the height of the barometer. Although
+such an effect is undoubtedly very small, yet it
+is measurable. The daily heating of the air by
+the sun, and its cooling at night, produce marked
+alternations in the atmospheric pressure, and this
+effect may by analogy be called an atmospheric
+meteorological tide.
+
+The attractions of the moon and sun must certainly
+act not only on the sea, but also on the
+solid earth; and, since the earth is not perfectly
+rigid or stiff, they must produce an alternating
+change in its shape. Even if the earth is now
+so stiff that the changes in its shape escape
+detection through their minuteness, yet such
+changes of shape must exist. There is much
+evidence to show that in the early stages of their
+histories the planets consisted largely or entirely
+of molten rock, which must have yielded to tidal
+influences. I shall, then, extend the term ``tide''
+\index{Tide!definition|)}%
+\PageSep{4}
+so as to include such alternating deformations of
+a solid and elastic, or of a molten and plastic,
+globe. These corporeal tides will be found to
+lead us on to some far-reaching astronomical
+speculations. The tide, in the sense which I
+\index{Tide!general description|(}%
+have attributed to the term, covers a wide field
+of inquiry, and forms the subject of the present
+volume.
+
+I now turn to the simplest and best known
+form of tidal phenomena. When we are at the
+seashore, or on an estuary, we see that the water
+rises and falls nearly twice a day. To be more
+exact, the average interval from one high water
+to the next is twelve hours twenty-five minutes,
+and so high water falls later, according to the
+clock, by twice twenty-five minutes, or by fifty
+minutes, on each successive day. Thus if high
+water falls to-day at noon, it will occur to-morrow
+at ten minutes to one. Before proceeding, it
+may be well to remark that I use high water and
+low water as technical terms. In common parlance
+the level of water may be called high or
+low, according as whether it is higher or lower
+than usual. But when the level varies periodically,
+there are certain moments when it is highest
+and lowest, and these will be referred to as
+the times of high and low water, or of high and
+of low tide. In the same way I shall speak of
+the heights at high and low water, as denoting
+the water-level at the moments in question.
+\PageSep{5}
+
+The most elementary observations would show
+that the time of high water has an intimate relationship
+to the moon's position. The moon, in
+fact, passes the meridian on the average fifty
+minutes later on each succeeding day, so that if
+high water occurs so many hours after the moon
+is due south on any day, it will occur on any
+other day about the same number of hours after
+the moon was south. This rule is far from being
+exact, for it would be found that the interval
+from the moon's passage to high water differs
+considerably according to the age of the moon.
+I shall not, however, attempt to explain at present
+how this rough rule as to the time of high
+water must be qualified, so as to convert it into
+an accurate statement.
+
+But it is not only the hour of high water which
+changes from day to day, for the height to which
+the water rises varies so conspicuously that the
+fact could not escape the notice of even the most
+casual observer. It would have been necessary
+to consult a clock to discover the law by which
+the hour of high water changes from day to day;
+but at the seashore it would be impossible to
+avoid noticing that some rocks or shoals which
+are continuously covered by the sea at one part
+of a fortnight are laid bare at others. It is, in
+fact, about full and new moon that the range
+from low to high water is greatest, and at the
+moon's first or third quarter that the range is
+\PageSep{6}
+least. The greater tides are called ``springs,''
+and the smaller ``neaps.''
+
+The currents produced in the sea by tides are
+\index{Gauge, tide!description of|(}%
+often very complicated where the open sea is
+broken by islands and headlands, and the knowledge
+of tidal currents at each place is only to be
+gained by the practical experience of the pilot.
+Indeed, in the language of sailors, the word
+``tide'' is not unfrequently used as meaning
+tidal current, without reference to rise and fall.
+These currents are often of great violence, and
+vary from hour to hour as the water rises and
+falls, so that the pilot requires to know how the
+water stands in-shore in order to avail himself of
+his practical knowledge of how the currents will
+make in each place. A tide table is then of
+\index{Tide!general description|)}%
+much use, even at places where the access to a
+harbor is not obstructed by a bar or shoal. It
+is, of course, still more important for ships to
+have a correct forecast of the tides where the
+entrance to the harbor is shallow.
+
+I have now sketched in rough outline some
+of the peculiarities of the tides, and it will have
+become clear that the subject is a complicated
+one, not to be unraveled without regular observation.
+\index{Observation!methods of tidal|(}%
+I shall, therefore, explain how tides are
+observed scientifically, and how the facts are collected
+upon which the scientific treatment of the
+tides is based.
+
+The rise and fall of the sea may, of course, be
+\PageSep{7}
+roughly estimated by observing the height of the
+water on posts or at jetties, which jut out into
+moderately deep water. But as the sea is continually
+disturbed by waves, observations of this
+kind are not susceptible of accuracy, and for
+\Figure{1}{Well for Tide-Gauge}{png}
+scientific purposes more elaborate apparatus is
+required. The exact height of the water can
+only be observed in a place to which the sea has
+a moderately free access, but where the channel
+is so narrow as to prevent the waves from sensibly
+disturbing the level of the water. This result
+\PageSep{8}
+is obtained in a considerable variety of ways,
+but one of them may be described as typical of
+all.
+
+A well (\fig{1}) about two feet in diameter is
+dug to a depth of several feet below the lowest
+tide and in the neighborhood of deep water. The
+well is lined with iron, and a two-inch iron pipe
+runs into the well very near its bottom, and passes
+down the shore to the low-water line. Here it is
+joined to a flexible pipe running out into deep
+water, and ending with a large rose pierced with
+many holes, like that of a watering can. The
+rose (\fig{2}) is anchored to the bottom of the
+sea, and is suspended by means of a buoy, so as
+to be clear of the bottom. The tidal water can
+thus enter pretty freely into the well, but the
+passage is so narrow that the wave motion is not
+transmitted into the well. Inside the well there
+floats a water-tight copper cylinder, weighted at
+the bottom so that it floats upright, and counter-poised
+so that it only just keeps its top clear of
+the water. To the top of the float there is fastened
+a copper tape or wire, which runs up to
+the top of the well and there passes round a
+wheel. Thus as the water rises and falls this
+wheel turns backwards and forwards.
+
+It is hardly necessary to describe in detail the
+simple mechanism by which the turning of this
+wheel causes a pencil to move backwards and forwards
+in a straight line. The mechanism is,
+\PageSep{9}
+however, such that the pencil moves horizontally
+backwards and forwards by exactly the same
+amount as the water rises or falls in the well; or,
+\Figure{2}{Pipe of Tide-Gauge}{png}
+if the rise and fall of the tide is considerable,
+the pencil only moves by half as much, or one
+third, or even one tenth as much as the water.
+At each place a scale of reduction is so chosen
+as to bring the range of motion of the pencil
+within convenient limits. We thus have a pencil
+which will draw the rise and fall of the tide
+on the desired scale.
+
+It remains to show how the times of the rise
+and fall are indicated. The end of the pencil
+touches a sheet of paper which is wrapped round
+a drum about five feet long and twenty-four
+\PageSep{10}
+\index{Curve, tide!irregularities in|(}%
+inches in circumference. If the drum were kept
+still the pencil would simply draw a straight line
+to and fro along the length of the drum as the
+water rises and falls. But the drum is kept
+turning by clockwork, so that it makes exactly
+one revolution in twenty-four hours. Since the
+drum is twenty-four inches round, each inch of
+circumference corresponds to one hour. If the
+water were at rest the pencil would simply draw
+a circle round the paper, and the beginning and
+ending of the line would join, whilst if the drum
+remained still and the water moved, the pencil
+\Figure{3}{Indian Tide-Gauge}{jpg}
+would draw a straight line along the length of
+the cylinder; but when both drum and water
+are in motion, the pencil draws a curve on the
+cylinder from which the height of water may be
+read off at any time in each day and night. At
+the end of twenty-four hours the pencil has returned
+to the same part of the paper from which
+\PageSep{11}
+\Figure[0.7]{4}{Légé's Tide-Gauge}{jpg}
+\index{Gauge, tide!description of|)}%
+\PageSep{12}
+it started, and it might be thought that there
+would be risk of confusion between the tides of
+to-day and those of yesterday. But since to-day
+the tides happen about three quarters of an hour
+later than yesterday, it is found that the lines
+keep clear of one another, and, in fact, it is
+usual to allow the drum to run for a fortnight
+before changing the paper, and when the old
+sheet is unwrapped from the drum, there is
+written on it a tidal record for a fortnight.
+
+The instrument which I have described is
+called a ``tide-gauge,'' and the paper a ``tide-curve.''
+As I have already said, tide-gauges
+may differ in many details, but this description
+will serve as typical of all. Another form of
+tide-gauge is shown in \fig{4}; here a continuous
+sheet of paper is placed over the drum, so that
+there is no crossing of the curves, as in the first
+example. Yet another form, designed by Lord
+Kelvin, is shown on p.~170 of vol.~iii.\ of his
+``Popular Lectures.''
+
+The actual record for a week is exhibited in
+\index{Curve, tide!at Bombay}%
+\fig{5}, on a reduced scale. This tide-curve was
+drawn at Bombay by a tide-gauge of the pattern
+first described. When the paper was wrapped
+on the drum, the right edge was joined to the
+left, and now that it is unwrapped the curve
+must be followed out of the paper on the left
+and into it again on the right. The figure
+shows that spring tide occurred on April~26,
+\PageSep{13}
+1884; the preceding neap tide was on the~18th,
+and is not shown. It may be noticed that the
+law of the tide is conspicuously different from
+that which holds good on the coast of England,
+for the two successive high or low waters which
+occur on any day have very different heights.
+Thus, for example, on April~26 low water occurred
+at 5.50~\PM, and the water fell to $5$~ft.\
+$2$~in., whereas the next low water, occurring at
+5.45~\AM\ of the~27th, fell to $1$~ft.\ $3$~in., the
+heights being in both cases measured from a
+certain datum. When we come to consider the
+theory of the tides the nature of this irregularity
+will be examined.
+
+The position near the seashore to be chosen
+for the erection of the tide-gauge is a matter of
+much importance. The choice of a site is generally
+limited by nature, for it should be near
+the open sea, should be sheltered from heavy
+weather, and deep water must be close at hand
+even at low tide.
+
+In the sketch map shown in \fig{6} a site such
+as~$A$ is a good one when the prevailing wind
+blows in the direction of the arrow. A position
+such as~$B$, although well sheltered from heavy
+seas, is not so good, because it is found that
+tide-curves drawn at~$B$ would be much zigzagged.
+These zigzags appear in the Bombay curves,
+although at Bombay they are usually very
+smooth ones.
+\PageSep{14}
+
+These irregularities in the tide-curve are not
+due to tides, and as the object of the observation
+\index{Observation!methods of tidal|)}%
+is to determine the nature of the tides it is
+\Figure{5}{Bombay Tide-Curve from Noon, April~22,
+to Noon, April~30, 1884}{jpg}
+desirable to choose a site for the gauge where
+\index{Gauge, tide!site for}%
+the zigzags shall not be troublesome; but it is
+not always easy to foresee the places which will
+furnish smooth tide-curves.
+
+Most of us have probably at some time or
+other made a scratch on the sand by the seashore,
+\PageSep{15}
+and watched the water rise over it. We
+generally make our mark on the sand at the
+furthest point, where the wash of a rather large
+wave has brought up the water. For perhaps
+five or ten minutes no wave brings the water up
+as far as the mark, and one begins to think that
+it was really an extraordinarily large wave which
+was marked, although it did not seem so at the
+time. Then a wave brings up the water far over
+the mark, and immediately all the waves submerge
+it. This little observation simply points
+\Figure{6}{Sites for a Tide-Gauge}{png}
+to the fact that the tide is apt to rise by jerks,
+and it is this irregularity of rise and fall which
+marks the notches in the tide-curves to which I
+have drawn attention.
+
+Now in scientific matters it is well to follow
+up the clues afforded by such apparently insignificant
+facts as this. An interesting light is
+indeed thrown on the origin of these notches on
+tide-curves by an investigation, not very directly
+\PageSep{16}
+connected with tidal observation, on which I shall
+make a digression in the following chapter.
+\index{Curve, tide!irregularities in|)}%
+
+\begin{Authorities}
+Baird's \Title{Manual for Tidal Observations} (Taylor \& Francis,
+\index{Baird, \Title{Manual for Tidal Observation}}%
+1886). Price 7\textit{s.}~6\textit{d.} Figs.\ \figref{1},~\figref{2},~\figref{3},~\figref{6} are reproduced from this
+work.
+
+The second form of tide-gauge shown in \fig{4} is made by
+Messrs.\ Légé, and is reproduced from a woodcut kindly provided
+by them.
+
+Sir William Thomson's (Lord Kelvin's) \Title{Popular Lectures and
+Addresses}, vol.~iii. (Macmillan,~1891).
+\end{Authorities}
+\PageSep{17}
+
+
+\Chapter{II}
+{Seiches in Lakes}
+
+\First{It} has been known for nearly three centuries
+\index{Lakes!seiches in|(}%
+\index{Seiches!definition}%
+that the water of the Lake of Geneva is apt to
+\index{Geneva!seiches in lake|(}%
+rise and fall by a few inches, sometimes irregularly
+and sometimes with more or less regularity;
+and the same sort of oscillation has been observed
+in other Swiss lakes. These quasi-tides,
+called seiches, were until recently supposed only
+to occur in stormy weather, but it is now known
+that small seiches are of almost daily occurrence.\footnote
+ {The word ``seiche'' is a purely local one. It has been alleged
+ to be derived from ``sèche,'' but I can see no reason for
+ associating dryness with the phenomenon.}
+
+Observations were made by Vaucher in the
+\index{Vaucher, record of a great seiche at Geneva}%
+last century on the oscillations of the Lake of
+Geneva, and he gave an account of a celebrated
+seiche in the year~1600, when the water oscillated
+through three or four feet; but hardly any
+systematic observation had been undertaken when
+Professor Forel, of Lausanne, attacked the subject,
+\index{Forel!on seiches|(}%
+and it is his very interesting observations
+which I propose to describe.
+
+Doctor Forel is not a mathematician, but is
+\PageSep{18}
+rather a naturalist of the old school, who notes any
+interesting fact and then proceeds carefully to
+investigate its origin. His papers have a special
+charm in that he allows one to see all the workings
+of his mind, and tells of each difficulty as it
+arose and how he met it. To those who like to
+read of such work, almost in the form of a narrative,
+I can strongly recommend these papers,
+which afford an admirable example of research
+thoroughly carried out with simple appliances.
+
+People are nowadays too apt to think that
+science can only be carried to perfection with
+elaborate appliances, and yet it is the fact that
+many of the finest experiments have been made
+with cardboard, cork, and sealing-wax. The principal
+reason for elaborate appliances in the laboratories
+of universities is that a teacher could not
+deal with a large number of students if he had
+to show each of them how to make and set up
+his apparatus, and a student would not be able
+to go through a large field of study if he had to
+spend days in preparation. Great laboratories
+have, indeed, a rather serious defect, in that they
+tend to make all but the very best students helpless,
+and thus to dwarf their powers of resource
+and inventiveness. The mass of scientific work
+is undoubtedly enormously increased by these
+institutions, but the number of really great investigators
+seems to remain almost unaffected by
+them. But I must not convey the impression
+\PageSep{19}
+that, in my opinion, great laboratories are not
+useful. It is obvious, indeed, that without them
+science could not be taught to large numbers of
+students, and, besides, there are many investigations
+in which every possible refinement of apparatus
+is necessary. But I do say that the
+number of great investigators is but little increased
+by laboratories, and that those who are
+interested in science, but yet have not access to
+laboratories, should not give up their study in
+despair.
+
+Doctor Forel's object was, in the first instance,
+to note the variations of the level of the lake, after
+obliterating the small ripple of the waves on the
+surface. The instrument used in his earlier investigations
+was both simple and delicate. Its principle
+was founded on casual observation at the port
+of Morges, where there happens to be a breakwater,
+pierced by a large ingress for ships and a small
+one for rowing boats. He accidentally noticed
+that at the small passage there was always a current
+setting either inwards or outwards, and it
+occurred to him that such a current would form
+a very sensitive index of the rise and fall of the
+water in the lake. He therefore devised an instrument,
+illustrated in \fig{7}, and called by him
+a plemyrameter, for noting currents of even the
+\index{Plemyrameter, observation of seiches with|(}%
+most sluggish character. Near the shore he
+made a small tank, and he connected it with the
+lake by means of an india-rubber siphon pipe of
+\PageSep{20}
+small bore. Where the pipe crossed the edge
+of the tank he inserted a horizontal glass tube
+of seven millimetres diameter, and in that tube
+he put a float of cork, weighted with lead so that
+it should be of the same density as water. At
+the ends of the glass tube there were stops, so
+that the float could not pass out of it. When
+the lake was higher than the tank, the water ran
+\Figure{7}{Plemyrameter}{png}
+through the siphon pipe from lake to tank, and
+the float remained jammed in the glass tube
+against the stop on the side towards the tank;
+and when the lake fell lower than the tank, the
+float traveled slowly to the other end and remained
+there. The siphon pipe being small, the
+only sign of the waves in the lake was that the
+float moved with slight jerks, instead of uniformly.
+Another consequence of the smallness
+of the tube was that the amount of water which
+could be delivered into the tank or drawn out of
+it in one or two hours was so small that it might
+\PageSep{21}
+practically be neglected, so that the water level
+in the tank might be considered as invariable.
+
+This apparatus enabled Forel to note the rise
+and fall of the water, and he did not at first
+attempt to measure the height of rise and fall,
+as it was the periodicity in which he was principally
+interested.
+
+In order to understand the record of observations,
+\index{Seiches!records of}%
+it must be remembered that when the
+float is towards the lake, the water in the tank
+stands at the higher level, and when the float
+is towards the tank the lake is the higher. In
+the diagrams, of which \fig{8} is an example, the
+straight line is divided into a scale of hours and
+minutes. The zigzag line gives the record, and
+the lower portions represent that the water of
+the lake was below the tank, and the upper line
+that it was above the tank. The fact that the
+float only moved slowly across from end to end
+of the glass tube, is indicated by the slope of
+the lines, which join the lower and upper portions
+of the zigzags. Then on reading \fig{8} we
+see that from $2$~hrs.\ $1$~min.\ to $2$~hrs.\ $4$~mins.\ the
+water was high and the float was jammed against
+the tank end of the tube, because there was a
+current from the lake to the tank. The float
+then slowly left the tank end and traveled
+across, so that at $2$~hrs.\ $5$~mins.\ the water was
+low in the lake. It continued, save for transient
+changes of level, to be low until $2$~hrs.\ $30$~mins.,
+\PageSep{22}
+when it rose again. Further explanation seems
+unnecessary, as it should now be easy to read
+this diagram, and that shown in~\fig{9}.
+\index{Plemyrameter, observation of seiches with|)}%
+
+The sharp pinnacles indicate alternations of
+level so transient that the float had not time to
+travel across from one end of the glass tube
+to the other, before the current was reversed.
+These pinnacles may be disregarded for the
+present, since we are only considering seiches of
+considerable period.
+
+These two diagrams are samples of hundreds
+which were obtained at various points on the
+shores of Geneva, and of other lakes in Switzerland.
+In order to render intelligible the method
+by which Forel analyzed and interpreted these
+records, I must consider \fig{8} more closely. In
+this case it will be noticed that the record shows
+a long high water separated from a long low
+water by two pinnacles with flat tops. These
+pieces at the ends have an interesting significance.
+When the water of the lake is simply
+oscillating with a period of about an hour we
+have a trace of the form shown in~\fig{9}. But
+when there exists concurrently with this another
+oscillation, of much smaller range and of short
+period, the form of the trace will be changed.
+When the water is high in consequence of the
+large and slow oscillation, the level of the lake
+cannot be reduced below that of the tank by
+the small short oscillation, and the water merely
+\PageSep{23}
+stands a little higher or a little lower, but always
+remains above the level of the tank, so that the
+trace continues on the higher level. But when,
+in course of the changes of the large oscillation,
+the water has sunk to near the mean level of the
+%[** TN: Figures 8 and 9 combined; cross-refs handled by preamble code]
+\Figure{8}{Records of Seiches at Évian}{png}
+lake, the short oscillation will become manifest,
+and so it is only at the ends of the long flat
+pieces that we shall find evidence of the quick
+oscillation.
+
+Thus, in these two figures there was in one
+case only one sort of wave, and in the other
+there were two simultaneous waves. These
+records are amongst the simplest of those obtained
+by Forel, and yet even here the oscillations
+of the water were sufficiently complicated.
+It needed, indeed, the careful analysis of many
+records to disentangle the several waves and to
+determine their periods.
+
+After having studied seiches with a plemyrameter
+for some time, Forel used another form of
+\PageSep{24}
+apparatus, by which he could observe the amplitude
+of the waves as well as their period. His
+apparatus was, in fact, a very delicate tide-gauge,
+which he called a limnimeter. The only
+\index{Limnimeter, a form of tide gauge}%
+difference between this instrument and the one
+already described as a tide-gauge is that the
+drum turned much more rapidly, so that five
+feet of paper passed over the drum in twenty-four
+hours, and that the paper was comparatively
+narrow, the range of the oscillation being
+small. The curve was usually drawn on the full
+scale, but it could be quickly reduced to half
+scale when large seiches were under observation.
+
+It would be impossible in a book of this kind
+\index{Lakes!mode of rocking in seiches}%
+to follow Forel in the long analysis by which he
+interpreted his curves. He speaks thus of the
+complication of simultaneous waves: ``All these
+oscillations are embroidered one on the other
+and interlace their changes of level. There is
+here matter to disturb the calmest mind. I
+must have a very stout faith in the truth of my
+hypothesis to persist in maintaining that, in the
+midst of all these waves which cross and mingle,
+there is, nevertheless, a recognizable rhythm.
+This is, however, what I shall try to prove.''\footnote
+ {\Title{Deuxième Étude}, p.~544.}
+The hypothesis to which he here refers, and
+triumphantly proves, is that seiches consist of a
+rocking of the whole water of the lake about
+fixed lines, just as by tilting a trough the water
+\PageSep{25}
+in it may be set swinging, so that the level at
+the middle remains unchanged, while at the two
+ends the water rises and falls alternately.
+
+In another paper he remarks: ``If you will
+\index{Lakes!mode of rocking in seiches}%
+follow and study with me these movements you
+will find a great charm in the investigation.
+When I see the water rising and falling on the
+shore at the end of my garden I have not before
+me a simple wave which disturbs the water of
+the bay of Morges, but I am observing the manifestation
+of a far more important phenomenon.
+It is the whole water of the lake which is rocking.
+It is a gigantic impulse which moves the
+whole liquid mass of Leman throughout its
+length, breadth, and depth\dots. It is probable
+that the same thing would be observed in far
+larger basins of water, and I feel bound to
+recognize in the phenomenon of seiches the
+\index{Seiches!longitudinal and transverse|(}%
+grandest oscillatory movement which man can
+study on the face of our globe.''\footnote
+ {\Title{Les Seiches, Vagues d'Oscillation}, p.~11.}
+
+It will now be well to consider the map of
+Geneva in~\fig{10}. Although the lake somewhat
+resembles the arc of a circle, the curvature of
+its shores will make so little difference in the
+nature of the swinging of the water that we
+may, in the first instance, consider it as practically
+straight.
+
+Forel's analysis of seiches led him to conclude
+that the oscillations were of two kinds, the longitudinal
+\PageSep{26}
+and the transverse. In the longitudinal
+seiche the water rocks about a line drawn across
+the lake nearly through Morges, and the water
+at the east end of the lake rises when that at
+the west falls, and vice versa. The line about
+which the water rocks is called a node, so that in
+this case there is one node at the middle of the
+lake. This sort of seiche is therefore called a
+uninodal longitudinal seiche. The period of the
+\Figure{10}{Map of Lake of Geneva}{png}
+oscillation is the time between two successive
+high waters at any place, and it was found to be
+seventy-three minutes, but the range of rise and
+fall was very variable. There are also longitudinal
+seiches in which there are two nodes,
+dividing the lake into three parts, of which the
+central one is twice as long as the extreme parts;
+such an oscillation is called a binodal longitudinal
+seiche. In this mode the water at the middle
+\PageSep{27}
+of the lake is high when that at the two ends
+is low, and vice versa; the period is thirty-five
+minutes.
+
+Other seiches of various periods were observed,
+\index{Seiches!periods of}%
+some of which were no doubt multinodal. Thus
+in a trinodal seiche, the nodes divide the lake
+into four parts, of which the two central ones
+are each twice as long as the extreme parts. If
+there are any number of nodes, their positions
+are such that the central portion of the lake is
+divided into equal lengths, and the terminal
+parts are each of half the length of the central
+part or parts. This condition is necessary in
+order that the ends of the lake may fall at places
+where there is no horizontal current. In all such
+modes of oscillation the places where the horizontal
+current is evanescent are called loops, and
+these are always halfway between the nodes,
+where there is no rise and fall.
+
+A trinodal seiche should have a period of
+about twenty-four minutes, and a quadrinodal
+seiche should oscillate in about eighteen minutes.
+The periods of these quicker seiches would, no
+doubt, be affected by the irregularity in the form
+and depth of the lake, and it is worthy of notice
+that Forel observed at Morges seiches with
+periods of about twenty minutes and thirty minutes,
+which he conjectured to be multinodal.
+
+The second group of seiches were transverse,
+\index{Seiches!longitudinal and transverse|)}%
+being observable at Morges and Évian. It was
+\PageSep{28}
+clear that these oscillations, of which the period
+was about ten minutes, were transversal, because
+at the moment when the water was highest at
+Morges it was lowest at Évian, and vice versa.
+As in the case of the longitudinal seiches, the
+principal oscillation of this class was uninodal,
+but the node was, of course, now longitudinal to
+the lake. The irregularity in the width and
+depth of the lake must lead to great diversity of
+period in the transverse seiches appropriate to
+the various parts of the lake. The transverse
+seiches at one part of the lake must also be
+transmitted elsewhere, and must confuse the
+seiches appropriate to other parts. Accordingly
+there is abundant reason to expect oscillations of
+such complexity as to elude complete explanation.
+
+The great difficulty of applying deductive
+reasoning to the oscillations of a sheet of water
+of irregular outline and depth led Forel to construct
+a model of the lake. By studying the
+waves in his model he was able to recognize
+many of the oscillations occurring in the real
+lake, and so obtained an experimental confirmation
+of his theories, although the periods of
+oscillation in the model of course differed enormously
+from those observed in actuality.
+
+The theory of seiches cannot be considered as
+demonstrated, unless we can show that the water
+of such a basin as that of Geneva is capable of
+\index{Geneva!seiches in lake|)}%
+\index{Geneva!model of lake}%
+\PageSep{29}
+swinging at the rates observed. I must, therefore,
+now explain how it may be proved that the
+periods of the actual oscillations agree with the
+facts of the case.
+
+As a preliminary let us consider the nature of
+wave motion. There are two very distinct cases
+of the undulatory motion of water, which nevertheless
+graduate into one another. The distinction
+lies in the depth of the water compared with
+the length of the wave, measured from crest to
+crest, in the direction of wave propagation. The
+wave-length may be used as a measuring rod,
+and if the depth of the water is a small fraction
+of the wave-length, it must be considered shallow,
+but if its depth is a multiple of the wave-length,
+it will be deep. The two extremes of
+course graduate into one another.
+
+In a wave in deep water the motion dies out
+pretty rapidly as we go below the surface, so that
+when we have gone down half a wave-length
+below the surface, the motion is very small. In
+shallow water, on the other hand, the motion extends
+quite to the bottom, and in water which is
+neither deep nor shallow, the condition of affairs
+is intermediate. The two figures, \figref{11}~and~\figref{12},
+show the nature of the movement in the two
+classes of waves. In both cases the dotted lines
+\index{Waves!in deep and shallow water}%
+show the position of the water when at rest, and
+the full lines show the shapes assumed by the
+rectangular blocks marked out by the dotted
+\PageSep{30}
+lines, when wave motion is disturbing the water.
+It will be observed that in the deep water, as
+shown in \fig{11}, the rectangular blocks change
+their shape, rise and fall, and move to and fro.
+Taking the topmost row of rectangles, each block
+of water passes successively in time through all
+the forms and positions shown by the top row
+of quasi-parallelograms. So also the successive
+changes of the second row of blocks are indicated
+by the second strip, and the third and the fourth
+indicate the same. The changes in the bottom
+\Figure[0.7]{11}{Wave in Deep Water}{png}
+row are relatively very small both as to shape
+and as to displacement, so that it did not seem
+worth while to extend the figure to a greater
+depth.
+
+Turning now to the wave in shallow water in~\fig{12},
+we see that each of the blocks is simply
+displaced sideways and gets thinner or more
+\PageSep{31}
+squat as the wave passes along. Now, I say that
+we may roughly classify the water as being deep
+with respect to wave motion when its depth is
+more than half a wave-length, and as being shallow
+when it is less. Thus the same water may
+be shallow for long waves and deep for short
+\index{Waves!speed of}%
+ones. For example, the sea is very shallow for
+\Figure[0.7]{12}{Wave in Shallow Water}{png}
+the great wave of the oceanic tide, but it is very
+deep even for the largest waves of other kinds.
+Deepness and shallowness are thus merely relative
+to wave-length.
+
+The rate at which a wave moves can be exactly
+calculated from mathematical formulæ,
+from which it appears that in the deep sea a
+wave $63$~metres in length travels at $36$~kilometres
+per hour, or, in British measure, a wave of
+$68$~yards in length travels $22\frac{1}{2}$~miles an hour.
+Then, the rule for other waves is that the speed
+varies as the square root of the wave-length, so
+that a wave $16$~metres long---that is, one quarter
+of $63$~metres---travels at $18$~kilometres an
+hour, which is half of $36$~kilometres an hour.
+Or if its length were $7$~metres, or one ninth as
+\PageSep{32}
+long, it would travel at $12$~kilometres an hour,
+or one third as quick.
+
+Although the speed of waves in deep water
+depends on wave-length, yet in shallow water the
+speed is identical for waves of all lengths, and
+depends only on the depth of the water. In
+water $10$~metres deep, the calculated velocity of
+a wave is $36$~kilometres an hour; or if the water
+were $2\frac{1}{2}$~metres deep (quarter of $10$~metres), it
+would travel $18$~kilometres (half of $36$~kilometres)
+an hour; the law of variation being that
+the speed of the wave varies as the square root
+of the depth. For water that is neither deep nor
+shallow, the rate of wave propagation depends
+both on depth and on wave-length, according to
+a law which is somewhat complicated.
+
+In the case of seiches, the waves are very long
+compared with the depth, so that the water is to
+be considered as shallow; and here we know
+that the speed of propagation of the wave depends
+only on depth. The average depth of the
+Lake of Geneva may be taken as about $150$~metres,
+and it follows that the speed of a long wave
+in the lake is about $120$~kilometres an hour.
+
+In order to apply this conclusion to the study
+of seiches, we have to consider what is meant by
+the composition of two waves. If I take the
+series of numbers
+\[
+\text{\&c.}\quad
+100\quad 71\quad 0\quad -71\quad -100\quad -71\quad 0\quad 71\quad 100\quad
+\text{\&c.}
+\]
+and plot out, at equal distances, a figure of
+\PageSep{33}
+\index{Waves!composition of|(}%
+heights proportional to these numbers, setting
+off the positive numbers above and the negative
+numbers below a horizontal line, I get the simple
+wave line shown in~\fig{13}. Now, if this
+wave is traveling to the right, the same series of
+numbers will represent the wave at a later time,
+when they are all displaced towards the right, as
+in the dotted line.
+
+Now turn to the following schedule of numbers,
+and consider those which are written in the
+top row of each successive group of three rows.
+The columns represent equidistant spaces, and
+the rows equidistant times. The first set of
+numbers, $-100$,~$-71$,~$0$,~\&c., are those which
+were plotted out as a wave in~\fig{13}; in the top
+\Figure[0.7]{13}{Simple Wave}{png}
+row of the second group they are the same, but
+moved one space to the right, so that they represent
+the movement of the wave to the right in
+one interval of time. In the top row of each
+successive group the numbers are the same, but
+always displaced one more space to the right;
+they thus represent the successive positions of a
+\PageSep{34}
+\begin{figure}[hp!]
+\centering
+\ifthenelse{\boolean{ForPrinting}}{%
+ \makebox[0pt][c]{\includegraphics[height=0.9\textheight]{./images/fig13b.pdf}}
+}{
+ \includegraphics[width=0.75\textwidth]{./images/fig13b.pdf}
+}
+\caption{Composition of Two Equal and Opposite Waves}
+\label{fig:13b}
+\iffalse
+-100 -71 0 71 100 71 0 -71 -100
+
+-100 -71 0 71 100 71 0 -71 -100
+[*--]
+-200 -142 0 142 200 142 0 -142 -200
+[*--]
+ -71 -100 -71 0 71 100 71 0 -71
+
+ -71 0 71 100 71 0 -71 -100 -71
+[*--]
+-142 -100 0 100 142 100 0 -100 -142
+[*--]
+ 0 -71 -100 -71 0 71 100 71 0
+
+ 0 71 100 71 0 -71 -100 -71 0
+[*--]
+ 0 0 0 0 0 0 0 0 0
+[*--]
+ 71 0 -71 -100 -71 0 71 100 71
+
+ 71 100 71 0 -71 -100 -71 0 71
+[*--]
+ 142 100 0 -100 -142 -100 0 100 142
+[*--]
+ 100 71 0 -71 -100 -71 0 71 100
+
+ 100 71 0 -71 -100 -71 0 71 100
+[*--]
+ 200 142 0 -142 -200 -142 0 142 200
+[*--]
+ 71 100 71 0 -71 -100 -71 0 71
+
+ 71 0 -71 -100 -71 0 71 100 71
+[*--]
+ 142 100 0 -100 -142 -100 0 100 142
+[*--]
+ 0 71 100 71 0 -71 -100 -71 0
+
+ 0 -71 -100 -71 0 71 100 71 0
+[*--]
+ 0 0 0 0 0 0 0 0 0
+[*--]
+ -71 0 71 100 71 0 -71 -100 -71
+
+ -71 -100 -71 0 71 100 71 0 -71
+[*--]
+-142 -100 0 100 142 100 0 -100 -142
+[*--]
+-100 \DPtypo{71}{-71} 0 71 100 71 0 -71 -100
+
+-100 -71 0 71 100 71 0 -71 -100
+[*--]
+-200 -142 0 142 200 142 0 -142 -200
+[*--]
+\fi
+\end{figure}
+\PageSep{35}
+wave moving to the right. The table ends in
+the same way as it begins, so that in eight of
+these intervals of time the wave has advanced
+through a space equal to its own length.
+
+If we were to invert these upper figures, so
+that the numbers on the right are exchanged
+with those on the left, we should have a series of
+numbers representing a wave traveling to the
+left. Such numbers are shown in the second
+row in each group.
+
+When these two waves coëxist, the numbers
+must be compounded together by addition, and
+then the result is the series of numbers written
+in the third rows. These numbers represent the
+resultant of a wave traveling to the right, and of
+an equal wave traveling simultaneously to the
+left.
+
+It may be well to repeat that the first row of
+each group represents a wave moving to the
+right, the second row represents a wave moving
+to the left, and the third represents the resultant
+of the two. Now let us consider the nature
+of this resultant motion; the third and the
+seventh columns of figures are always zero, and
+therefore at these two places the water neither
+rises nor falls,---they are, in fact, nodes. If the
+schedule were extended indefinitely both ways,
+exactly halfway between any pairs of nodes
+there would be a loop, or line across which there
+is no horizontal motion. In the schedule, as it
+\PageSep{36}
+stands, the first, fifth, and ninth columns are
+loops.
+
+At the extreme right and at the extreme left
+the resultant numbers are the same, and represent
+a rise of the water from $-200$ to~$+200$,
+and a subsequent fall to~$-200$ again. If these
+nine columns represent the length of the lake,
+the motion is that which was described as binodal,
+for there are two nodes dividing the lake
+into three parts, there is a loop at each end, and
+when the water is high in the middle it is low at
+the ends, and vice versa. It follows that two
+equal waves, each as long as the lake, traveling
+in opposite directions, when compounded together
+give the motion which is described as the
+binodal longitudinal seiche.
+
+Now let us suppose that only five columns of
+the table represent the length of the lake. The
+resultant numbers, which again terminate at
+each end with a loop, are:---
+\begin{align*}
+-200&& -142&& 0&& 142&& 200& \displaybreak[0] \\
+-142&& -100&& 0&& 100&& 142& \displaybreak[0] \\
+ 0&& 0&& 0&& 0&& 0& \displaybreak[0] \\
+ 142&& 100&& 0&& -100&& -142& \displaybreak[0] \\
+ 200&& 142&& 0&& -142&& -200& \displaybreak[0] \\
+ 142&& 100&& 0&& -100&& -142& \displaybreak[0] \\
+ 0&& 0&& 0&& 0&& 0& \displaybreak[0] \\
+-142&& -100&& 0&& 100&& 142& \displaybreak[0] \\
+-200&& -142&& 0&& 142&& 200&
+\end{align*}
+
+Since the middle column consists of zero
+throughout, the water neither rises nor falls
+\PageSep{37}
+there, and there is a node at the middle. Again,
+since the numbers at one end are just the same
+as those at the other, but reversed as to positive
+and negative, when the water is high at one end
+it is low at the other. The motion is, in fact, a
+simple rocking about the central line, and is that
+described as the uninodal longitudinal seiche.
+
+The motion is here again the resultant of two
+equal waves moving in opposite directions, and
+the period of the oscillation is equal to the time
+which either simple wave takes to travel through
+its own length. But the length of the wave is
+now twice that of the lake. Hence it follows
+that the period of the rocking motion is the
+time occupied by a wave in traveling twice the
+length of the lake. We have already seen that
+in shallow water the rate at which a wave moves
+is independent of its length and depends only
+on the depth of the water, and that in water of
+the same depth as the Lake of Geneva the wave
+travels $120$~kilometres an hour. The Lake of
+Geneva is $70$~kilometres long, so that the two
+waves, whose composition produces a simple rocking
+\index{Waves!composition of|)}%
+of the water, must each of them have a
+length of $140$~kilometres. Hence it follows that
+the period of a simple rocking motion, with one
+node in the middle of the Lake of Geneva, will
+be almost exactly $\frac{140}{120}$~of an hour, or $70$~minutes.
+Forel, in fact, found the period to be $73$~minutes.
+He expresses this result by saying that
+\PageSep{38}
+a uninodal longitudinal seiche in the Lake of
+Geneva has a period of $73$~minutes. His observations
+also showed him that the period of a
+binodal seiche was $35$~minutes. It follows from
+the previous discussion that when there are two
+nodes the period of the oscillation should be
+half as long as when there is one node. Hence,
+we should expect that the period would be
+about $36$~or $37$~minutes, and the discrepancy
+between these two results may be due to the
+fact that the formula by which we calculate the
+period of a binodal seiche would require some
+correction, because the depth of the lake is not
+so very small compared with the length of these
+shorter waves.
+
+It is proper to remark that the agreement
+between the theoretical and observed periods is
+suspiciously exact. The lake differs much in
+depth in different parts, and it is not quite certain
+what is the proper method of computing
+the average depth for the determination of the
+period of a seiche. It is pretty clear, in fact,
+that the extreme closeness of the agreement is
+accidentally due to the assumption of a round
+number of metres as the average depth of the
+lake. The concordance between theory and observation
+must not, however, be depreciated too
+much, for it is certain that the facts of the case
+agree well with what is known of the depth of
+the lake.
+\index{Forel!on seiches|)}%
+\PageSep{39}
+
+The height of the waves called ``seiches'' is
+\index{Seiches!causes of}%
+\index{Storms a cause of seiches}%
+very various. I have mentioned an historical
+seiche which had a range of as much as four
+feet, and Forel was able by his delicate instruments
+still to detect them when they were only a
+millimetre or a twenty-fifth of an inch in height.
+It is obvious, therefore, that whatever be the
+cause of seiches, that cause must vary widely
+in intensity. According to Forel, seiches arise
+from several causes. It is clear that anything
+which heaps up the water at one end of the
+lake, and then ceases to act, must tend to produce
+an oscillation of the whole. Now, a rise
+of water level at one end or at one side of the
+lake may be produced in various ways. Some,
+and perhaps many, seiches are due to the tilting
+of the whole lake bed by minute earthquakes.
+\index{Earthquakes!a cause of seiches}%
+Modern investigations seem to show that this is
+a more fertile cause than Forel was disposed to
+allow, and it would therefore be interesting to
+see the investigation of seiches repeated with the
+aid of delicate instruments for the study of
+earthquakes, some of which will be described in
+\Ref{Chapter}{VI}. I suspect that seiches would be
+observed at times when the surface of the earth
+is much disturbed.
+
+The wind is doubtless another cause of seiches.
+\index{Wind!a cause of seiches}%
+When it blows along the lake for many hours in
+one direction, it produces a superficial current,
+and heaps up the water at the end towards
+\PageSep{40}
+\index{Atmospheric pressure!cause of seiches}%
+which it is blowing. If such a wind ceases
+somewhat suddenly, a seiche will certainly be
+started, and will continue for hours until it dies
+out from the effects of the friction of the water
+on the lake bottom. Again, the height of the
+barometer will often differ slightly at different
+parts of the lake, and the water will respond, just
+as does the mercury, to variations of atmospheric
+pressure. About a foot of rise of water should
+correspond to an inch of difference in the height
+of barometer. The barometric pressure cannot
+be quite uniform all over the Lake of Geneva,
+and although the differences must always be
+exceedingly small, yet it is impossible to doubt
+that this cause, combined probably with wind,
+will produce many seiches. I shall return later
+\index{Seiches!causes of}%
+to the consideration of an interesting speculation
+as to the effects of barometric pressure on
+the oscillation of lakes and of the sea. Lastly,
+Forel was of opinion that sudden squalls or local
+storms were the most frequent causes of seiches.
+\index{Storms a cause of seiches}%
+I think that he much overestimated the efficiency
+of this cause, because his theory of the path of
+the wind in sudden and local storms is one that
+would hardly be acceptable to most meteorologists.
+
+Although, then, it is possible to indicate causes
+competent to produce seiches, yet we cannot as
+yet point out the particular cause for any individual
+seiche. The complication of causes is so
+\PageSep{41}
+\index{Lakes!vibrations|(}%
+\index{Vibration of lakes|(}%
+great that this degree of uncertainty will probably
+never be entirely removed.
+
+But I have not yet referred to the point which
+justifies this long digression on seiches in a book
+on the tides. The subject was introduced by
+the irregularities in the line traced by the tide-gauge
+at Bombay, which indicated that there
+are oscillations of the water with periods ranging
+from two minutes to a quarter of an hour or
+somewhat longer. Now these zigzags are not
+found in the sea alone, for Forel observed on
+the lake oscillations of short period, which resembled
+seiches in all but the fact of their more
+rapid alternations. Some of these waves are
+perhaps multinodal seiches, but it seems that
+they are usually too local to be true seiches
+affecting the whole body of the lake at one time.
+Forel calls these shorter oscillations ``vibrations,''
+thus distinguishing them from proper seiches.
+A complete theory of the so-called vibrations
+has not yet been formulated, although, as I shall
+show below, a theory is now under trial which
+serves to explain, at least in part, the origin of
+vibrations.
+
+Forel observed with his limnimeter or tide-gauge
+that when there is much wind, especially
+\index{Wind!vibrations of lakes due to}%
+from certain quarters, vibrations arise which are
+quite distinct from the ordinary visible wave
+motion. The period of the visible waves on the
+\PageSep{42}
+Lake of Geneva is from $4$~to $5$~seconds,\footnote
+ {I observed when it was blowing half a gale on Ullswater, in
+ Cumberland, that the waves had a period of about a second.}
+whereas
+vibrations have periods ranging from $45$~seconds
+to $4$~minutes. Thus there is a clear line separating
+waves from vibrations. Forel was unable
+to determine what proportion of the area of the
+lake is disturbed by vibrations at any one time,
+and although their velocity was not directly observed,
+there can be no doubt that these waves
+are propagated at a rate which corresponds to
+their length and to the depth of the water. I
+have little doubt but that the inequalities which
+produce notches in a tide-curve have the same
+origin as vibrations on lakes.
+
+It is difficult to understand how a wind, whose
+\index{Wind!vibrations of lakes due to}%
+only visible effect is short waves, can be responsible
+for raising waves of a length as great as a
+thousand yards or a mile, and yet we are driven
+to believe that this is the case. But Forel also
+found that steamers produce vibrations exactly
+like those due to wind. The resemblance was
+indeed so exact that vibrations due to wind
+could only be studied at night, when it was
+known that no steamers were traveling on the
+lake, and, further, the vibrations due to steamers
+could only be studied when there was no wind.
+
+His observations on the steamer vibrations are
+amongst the most curious of all his results.
+When a boat arrives at the pier at Morges, the
+\PageSep{43}
+water rises slowly by about $5$~to $8$~millimetres,
+and then falls in about $20$~to $30$~seconds. The
+amount and the rapidity of the rise and fall
+vary with the tonnage of the boat and with the
+rate of her approach. After the boat has passed,
+the trace of the limnimeter shows irregularities
+with sharp points, the variations of height ranging
+from about two to five millimetres, with a
+period of about two minutes. These vibrations
+continue to be visible during two to three hours
+after the boat has passed. As these boats travel
+at a speed of $20$~kilometres an hour, the vibrations
+persist for a long time after any renewal
+of them by the boat has ceased. These vibrations
+are called by Forel ``the subsequent steamer
+vibrations.''
+
+That the agitation of the water should continue
+for more than two hours is very remarkable,
+and shows the delicacy of the method of
+observation. But it seems yet more strange
+that, when a boat is approaching Morges, the
+vibrations should be visible during $25$~minutes
+before she reaches the pier. These he calls
+``antecedent steamer vibrations.'' They are
+more rapid than the subsequent ones, having a
+period of a minute to a minute and a quarter.
+Their height is sometimes two millimetres (a
+twelfth of an inch), but they are easily detected
+when less than one millimetre in height. It
+appears that these antecedent vibrations are first
+\PageSep{44}
+noticeable when the steamer rounds the mole of
+Ouchy, when she is still at a distance of $10$~kilometres.
+As far as one can judge from the speed
+at which waves are transmitted in the Lake of
+Geneva, the antecedent vibrations, which are
+noticed $25$~minutes before the arrival of the
+boat, must have been generated when she was at
+a distance of $12$~kilometres from Morges. \Fig{14}
+gives an admirable tracing of these steamer
+vibrations.\footnote
+ {From \Title{Les Seiches, Vagues d'Oscillation fixe des Lacs}, 1876.}
+
+In this figure the line~$\Seg{a}{a'}$ was traced between
+two and three o'clock in the morning, and shows
+scarcely any sign of perturbation. Between
+three and eight o'clock in the morning no observations
+were taken, but the record begins again
+at eight o'clock. The portion marked~$\Seg{b}{b'}$ shows
+weak vibrations, probably due to steamers passing
+along the coast of Savoy. The antecedent
+vibrations, produced by a steamer approaching
+Morges, began about the time of its departure
+from Ouchy, and are shown at~$\Seg{c}{c'}$. The point~$d$
+shows the arrival of this boat at Morges, and
+$d'$~shows the effect of another boat coming from
+Geneva. The portion marked~$\Seg[e]{e}{e}$ shows the
+subsequent steamer vibrations, which were very
+clear during more than two hours after the boats
+had passed.
+
+Dr.~Forel was aware that similar vibrations occur
+in the sea, for he says: ``What are these
+\index{Sea!vibrations of}%
+\PageSep{45}
+oscillations with periods of
+$5$,~$10$, $20$, or $100$~minutes,
+which are sometimes irregular?
+Are they analogous to
+our seiches? Not if we define
+seiches as uninodal oscillations,
+for it is clear that
+if, in a closed basin of $70$~kilometres
+in length, uninodal
+seiches have a period of
+$73$~minutes, in the far greater
+basin of the Mediterranean,
+or of the ocean, a uninodal
+wave of oscillation must have
+a much longer period. They
+resemble much more closely
+what I have called vibrations,
+and, provisionally, I
+shall call them by the name
+of `vibrations of the sea.' I
+\index{Sea!vibrations of}%
+venture to invite men of science
+who live on the seacoast
+to follow this study.
+It presents a fine subject for
+research, either in the interpretation
+of the phenomenon
+or in the establishment of
+the relations between these
+movements and meteorological
+conditions.''\footnote
+ {\Title{Seiches et Vibrations des Lacs et de la Mer}, 1879, p.~5.}
+%[** TN: Figure wrapped in the original]
+\TallFig{14}{Vibrations due to Steamers}{png}
+\PageSep{46}
+
+These vibrations are obviously due to the wind
+or to steamers, but it is a matter of no little surprise
+that such insignificant causes should produce
+even very small waves of half a mile to a
+mile in length.
+
+The manner in which this is brought about is
+undoubtedly obscure, yet it is possible to obtain
+some sort of insight into the way in which these
+long waves arise. When a stone falls into calm
+water waves of all sorts of lengths are instantaneously
+generated, and the same is true of
+any other isolated disturbance. Out of all these
+waves the very long ones and the very short
+ones are very small in height. Theoretically,
+waves of infinitely great and of infinitely small
+lengths, yet in both cases of infinitely small
+heights, are generated at the instant of the impulse,
+but the waves of enormous length and
+those of very small length are of no practical
+importance, and we need only consider the moderate
+waves. For the shorter of these the water
+is virtually deep, and so they will each travel
+outwards at a pace dependent on length, the
+longer ones outstripping the shorter ones. But
+for the longer waves the water will be shallow,
+and they will all travel together. Thus the general
+effect at a distance is the arrival of a long
+wave first, followed by an agitated rippling.
+The point which we have to note is that an isolated
+disturbance will generate long waves and
+\PageSep{47}
+that they will run ahead of the small ones. It
+is important also to observe that the friction of
+the water annuls the oscillation in the shorter
+waves more rapidly than it does that of the
+longer ones, and therefore the long waves are
+more persistent. Now we may look at the disturbance
+due to a steamer or to the wind as consisting
+of a succession of isolated disturbances,
+each of which will create long waves outstripping
+the shorter ones. These considerations afford a
+sort of explanation of what is observed, but I do
+not understand how it is that the separation of
+the long from the short waves is so complete, nor
+what governs the length of the waves, nor have
+I made any attempt to evaluate the greater rapidity
+of decrease of short waves than long ones.\footnote
+ {See, however, S.~S. Hough, \Title{Proc.\ Lond.\ Math.\ Soc.}, xxviii.\
+\index{Hough, S. S.!frictional extinction of waves}%
+ p.~276.}
+It must then be left to future investigators to
+elucidate these points.
+
+\TB
+
+The subject of seiches and vibrations clearly
+affords an interesting field for further research.
+The seiches of Lake George in New South Wales
+have been observed by Mr.~Russell, the government
+\index{Russell, observation of seiches in New South Wales}%
+astronomer at Sydney; but until last year
+they do not seem to have been much studied on
+any lakes outside of Switzerland. The great
+lakes of North America are no doubt agitated by
+seiches on a much larger scale than those on the
+\PageSep{48}
+comparatively small basin of Geneva. This idea
+appears to have struck Mr.~Napier Denison of
+\index{Denison, F. Napier, vibrations and seiches on lakes|(}%
+Toronto, and he has been so fortunate as to enlist
+the interest of Mr.~Bell Dawson, the chief of
+\index{Dawson coöperates in investigation of seiches}%
+the Canadian Tidal Survey, and of Mr.~Stupart,
+\index{Stupart coöperates in investigation of seiches}%
+the director of the Meteorological Department.
+Mr.~Denison's attention has been, in the first instance,
+principally directed towards those notches
+in tide-curves which have afforded the occasion
+for the present discussion of this subject. He
+has made an interesting suggestion as to the
+origin of these oscillations, which I will now
+explain.
+
+The wind generally consists of a rather shallow
+\index{Waves!in atmosphere|(}%
+current, so that when it is calm at the earth's
+surface there is often a strong wind at the top
+of a neighboring mountain; or the wind aloft
+may blow from a different quarter from that below.
+If we ascend a mountain or go up in a
+balloon, the temperature of the air falls on the
+average by a certain definite number of degrees
+per thousand feet. But the normal rate of fall
+of temperature is generally interrupted on passing
+into an upper current, which blows from a
+different direction. This abrupt change of temperature
+corresponds with a sudden change of
+density, so that the upper layer of air must be
+regarded as a fluid of different density from that
+of the lower air, over which it slides.
+
+Now Helmholtz has pointed out that one layer
+\index{Atmospheric waves, Helmholtz on|(}%
+\index{Helmholtz!on atmospheric waves|(}%
+\PageSep{49}
+of fluid cannot slide over another, without generating
+waves at the surface of separation. We
+are familiar with this fact in the case of sea-waves
+generated by wind. A mackerel sky
+\index{Mackerel sky, evidence of air-waves}%
+proves also the applicability to currents of air of
+Helmholtz's observation. In this case the moisture
+of the air is condensed into clouds at the
+crests of the air waves, and reabsorbed in the
+hollows, so that the clouds are arranged in a visible
+ripple-mark. A mackerel sky is not seen in
+stormy weather, for it affords proof of the existence
+of an upper layer of air sliding with only
+moderate velocity over a lower layer. The distance
+from crest to crest must be considerable
+as measured in yards, yet we must regard the
+mackerel sky as a mere ripple formed by a slow
+relative velocity of the two layers. If this is so,
+it becomes of interest to consider what wave-lengths
+may be expected to arise when the upper
+current is moving over the lower with a speed of
+perhaps a hundred miles an hour. The problem
+is not directly soluble, for even in the case of
+sea-waves it is impossible to predict the wave-lengths.
+We do know, however, that the duration
+of the wind and the size of the basin are
+material circumstances, and that in gales in the
+open ocean the waves attain a very definite magnitude.
+
+Although the problem involved is not a soluble
+one, yet Helmholtz has used the analogy of
+\PageSep{50}
+oceanic waves for an approximate determination
+of the sizes of the atmospheric ones. His
+method is a very fertile one in many complex
+physical investigations, where an exact solution
+is not attainable. The method may be best illustrated
+by one or two simple cases.
+
+It is easy for the mathematician to prove that
+the period of a swing of a simple pendulum must
+vary as the square root of its length. The proof
+does not depend on the complete solution of the
+problem, so that even if it were insoluble he
+would still be sure of the correctness of his conclusion.
+If, then, a given pendulum is observed
+to swing in a certain period, it is certain that a
+similar pendulum of four times the length will
+take twice as long to perform its oscillation. In
+the same way, the engine power required for a
+ship is determinable from experiments on the
+resistance suffered by a small model when towed
+through the water. The correct conclusion is
+discovered in this case, although it is altogether
+impossible to discover the resistance of a ship
+by \textit{à~priori} reasoning.
+
+The wave motion at the surface separating
+two fluids of different densities presents another
+problem of the same kind, and if the result is
+known in one case, it can be confidently predicted
+in another. Now oceanic waves generated
+\index{Waves!in atmosphere|)}%
+by wind afford the known case, and Helmholtz
+has thence determined by analogy the
+\PageSep{51}
+lengths of the atmospheric waves which must
+exist aloft. By making plausible suppositions
+as to the densities of the two layers of air and
+as to their relative velocity, he has shown that
+sea-waves of ten yards in length will correspond
+with air-waves of as much as twenty miles. A
+wave of this length would cover the whole sky,
+and might have a period of half an hour. It is
+clear then that mackerel sky will disappear in
+stormy weather, because we are too near to the
+crests and furrows to observe the orderly arrangement
+of the clouds.
+
+Although the waves are too long to be seen as
+such, yet the unsteadiness of the barometer in a
+gale of wind affords evidence of the correctness
+of this theory. In fact, when the crest of denser
+air is over the place of observation the barometer
+rises, and it falls as the hollow passes. The
+waves in the continuous trace of the barometer
+have some tendency to regularity, and have
+periods of from ten minutes to half an hour.
+The analogy seems to be pretty close with the
+confused and turbulent sea often seen in a gale
+of wind in the open ocean.\footnote
+ {A gust of wind will cause the barometer to vary, without a
+ corresponding change in the density of the air. It is not therefore
+ safe to interpret the oscillations of the barometer as being
+ due entirely to true changes of pressure. If, however, the intermittent
+ squalls in a gale are connected with the waves aloft,
+ the waviness of the barometric trace would still afford signals
+ of the passage of crests and hollows above.}
+\index{Atmospheric waves, Helmholtz on|)}%
+\index{Helmholtz!on atmospheric waves|)}%
+\PageSep{52}
+
+Mr.~Denison's application of this theory consists
+in supposing that the vibrations of the sea
+and of lakes are the response of the water to
+variations in the atmospheric pressure. The sea,
+being squeezed down by the greater pressure,
+should fall as the barometer rises, and conversely
+should rise as the barometer falls. He is engaged
+in a systematic comparison of the simultaneous
+excursions of the water and of the barometer
+on Lake Huron. Thus far the evidence
+seems decidedly favorable to the theory. He
+concludes that when the water is least disturbed,
+so also is the barometric trace; and that when
+the undulations of the lake become large and
+rapid, the atmospheric waves recorded by the
+barometer have the same character. There is
+also a considerable degree of correspondence
+between the periods of the two oscillations. The
+smaller undulations of the water correspond with
+the shorter air-waves, and are magnified as they
+run into narrower and shallower places, so as to
+make conspicuous ``vibrations.''
+
+It is interesting to note that the vibrations of
+the water have a tendency to appear before those
+in the barometer, so that they seem to give a
+warning of approaching change of weather. It
+is thus not impossible that we here have the
+foreshadowing of a new form of meteorological
+instrument, which may be of service in the forecasting
+of the weather.
+\PageSep{53}
+
+I must, however, emphasize that these conclusions
+are preliminary and tentative, and that
+much observation will be needed before they can
+be established as definite truths. Whatever
+may be the outcome, the investigation appears
+promising, and it is certainly already interesting.
+\index{Denison, F. Napier, vibrations and seiches on lakes|)}%
+\index{Lakes!vibrations|)}%
+\index{Vibration of lakes|)}%
+
+\begin{Authorities}
+Papers by Dr.~Forel on Seiches.
+\index{Forel!list of papers}%
+
+\Journal ``Bibliothèque Universelle, Archives des Sciences physiques
+et naturelles,'' Geneva:---
+
+\Paper{\Title{Formule des Seiches}, 1876.}
+
+\Paper{\Title{Limnimètre Enregistreur}, 1876.}
+
+\Paper{\Title{Essai monographique}, 1877.}
+
+\Paper{\Title{Causes des Seiches}, \DPchg{Sept.~15}{15~Sept.}, 1878.}
+
+\Paper{\Title{Limnographe}, 15~Déc., 1878.}
+
+\Paper{\Title{Seiche du \emph{20}~Février, \emph{1879}}, 15~Avril, 1879.}
+
+\Paper{\Title{Seiches dicrotes}, 15~Jan., 1880.}
+
+\Paper{\Title{Formules des Seiches}, 15~Sept., 1885.}
+
+\Journal ``Bulletin de la Soc.\ Vaudoise des Sciences naturelles:''---
+
+\Paper{\Title{Première Étude}, 1873.}
+
+\Paper{\Title{Deuxième Étude}, 1875.}
+
+\Paper{\Title{Limnimétrie du Lac Léman}. I\iere~Série. Bull.~xiv.\ 1877.
+II\ieme~Série. Bull.~xv. III\ieme~Série. Bull.~xv. 1879.}
+
+\Journal ``Actes de la Soc.\ helv.\ Andermatt:''---
+
+\Paper{\Title{Les Seiches, Vagues d'Oscillation}, 1875.}
+
+\Journal ``Association Française pour l'avancement,'' etc.:---
+
+\Paper{\Title{Seiches et Vibrations}, Congrès de Montpelier, 1879.}
+
+\Journal ``Annales de Chimie et de Physique:''---
+
+\Paper{\Title{Les Seiches, Vagues d'Oscillation}, 1876.}
+
+\Paper{\Title{Un Limnimètre Enregistreur}, 1876.}
+
+\TB
+
+Helmholtz, Sitzungsberichte der Preuss.\ Akad.\ der Wissenschaft,
+July~25, 1889; transl.\ by Abbe in \Title{Smithsonian Reports}.
+\PageSep{54}
+
+F.~Napier Denison:---
+
+\Paper{\Title{Secondary Undulations~\dots\ found in Tide-Gauges.} ``Proc.\
+Canadian Institute,'' Jan.~16, 1897.}
+
+\Paper{\Title{The Great Lakes as a Sensitive Barometer.} ``Proc.\ Canadian
+Institute,'' Feb.~6, 1897.}
+\index{Lakes!seiches in|)}%
+
+\Paper{Same title, but different paper, ``Canadian Engineer,'' Oct.\
+and Nov., 1897.}
+\end{Authorities}
+\index{Forel!list of papers}%
+\PageSep{55}
+
+
+%[** TN: Footnote mark handled by \Chapter logic]
+\Chapter{III}
+{Tides in Rivers---Tide Mills}
+
+\footnotetext{The account of the bore in this chapter appeared as an
+ article in the \Title{Century Magazine} for August,~1898. The illustrations
+ then used are now reproduced, through the courtesy of
+ the proprietors.}
+
+\First{Since} most important towns are situated on
+\index{Rivers!tide wave in|(}%
+rivers or on estuaries, a large proportion of our
+tidal observations relates to such sites. I shall
+therefore now consider the curious, and at times
+very striking phenomena which attend the rise
+and fall of the tide in rivers.
+
+The sea resembles a large pond in which the
+water rises and falls with the oceanic tide, and a
+river is a canal which leads into it. The rhythmical
+rise and fall of the sea generate waves
+which would travel up the river, whatever were
+the cause of the oscillation of the sea. Accordingly,
+a tide wave in a river owes its origin
+directly to the tide in the sea, which is itself
+produced by the tidal attractions of the sun and
+moon.
+
+We have seen in \Ref{Chapter}{II}.\ that long waves
+progress in shallow water at a speed which depends
+only on the depth of the water, and that
+\PageSep{56}
+waves are to be considered as long when their
+length is at least twice the depth of the water.
+Now the tide wave in a river is many hundreds
+of times as long as the depth, and it must therefore
+progress at a speed dependent only on the
+depth. That speed is very slow compared with
+the motion of the great tide wave in the open
+ocean.
+
+The terms ``ebb'' and ``flow'' are applied to
+\index{Ebb and flow defined}%
+\index{Flow and ebb defined}%
+tidal currents. The current ebbs when the
+water is receding from the land seaward, and
+flows when it is approaching the shore. On the
+open seacoast the water ebbs as the water-level
+falls, and it flows as the water rises. Thus at
+high and low tide the water is neither flowing
+landward nor ebbing seaward, and we say that
+it is slack or dead. In this case ebb and flow
+are simultaneous with rise and fall, and it is not
+uncommon to hear the two terms used synonymously;
+but we shall see that this usage is incorrect.
+
+I begin by considering the tidal currents in a
+river of uniform depth, so sluggish in its own
+proper current that it may be considered as a
+stagnant canal, and the only currents to be considered
+are tidal currents. At any point on the
+river bank there is a certain mean height of
+water, such that the water rises as much above
+that level at high water as it falls below it at
+low water. The law of tidal current is, then,
+\index{Currents, tidal, in rivers}%
+\PageSep{57}
+very simple. Whenever the water stands above
+the mean level the current is up-stream and progresses
+along with the tide wave; and whenever
+it stands below mean level the current is down-stream
+and progresses in the direction contrary
+to the tide wave. Since the current is up-stream
+when the water is higher than the mean, and
+down-stream when it is lower, it is obvious that
+when it stands exactly at mean level the current
+is neither up nor down, and the water is slack
+or dead. Also, at the moment of high water
+the current is most rapid up-stream, and at low
+water it is most rapid down-stream. Hence the
+tidal current ``flows'' for a long time after high
+water has passed and when the water-level is
+falling, and ``ebbs'' for a long time after low
+water and when the water-level is rising.
+
+The law of tidal currents in a uniform canal
+communicating with the sea is thus very different
+from that which holds on an open seacoast,
+where slack water occurs at high and at low
+water, instead of at mean water. But rivers
+gradually broaden and become deeper as they
+approach the coast, and therefore the tidal currents
+in actual estuaries must be intermediate
+between the two cases of the open seacoast and
+the uniform canal.
+
+A river has also to deliver a large quantity of
+water into the sea in the course of a single tidal
+oscillation, and its own proper current is superposed
+\PageSep{58}
+on the tidal currents. Hence in actual
+rivers the resultant current continues to flow up
+stream after high water is reached, with falling
+water-level, but ceases flowing before mean water-level
+is reached, and the resultant current ebbs
+down-stream after low water, and continues to
+ebb with the rising tide until mean water is
+reached, and usually for some time afterward.
+The downward stream, in fact, lasts longer than
+the upward one. The moments at which the
+currents change will differ in each river according
+to the depth, the rise and fall of the tide at
+the mouth, and the amount of water delivered
+by the river. An obvious consequence of this
+is that in rivers the tide rises quicker than it
+falls, so that a shorter time elapses between low
+water and high water than between high water
+and low water.
+
+The tide wave in a river has another peculiarity
+of which I have not yet spoken. The complete
+theory of waves would be too technical for a book
+of this sort, and I must ask the reader to accept
+as a fact that a wave cannot progress along a
+river without changing its shape. The change
+is such that the front slope of the wave gradually
+gets steeper, and the rear slope becomes more
+gradual. This is illustrated in~\fig{15}, which
+shows the progress of a train of waves in shallow
+water as calculated theoretically. If the
+steepening of the advancing slope of a wave
+\PageSep{59}
+were carried to an extreme, the wave would present
+the form of a wall of water; but the mere
+advance of a wave into shallow water would by
+itself never suffice to produce so great a change
+of form without the concurrence of the natural
+\Figure{15}{Progressive Change of a Train
+of Waves in Shallow Water}{png}
+stream of the river. The downward current in
+the river has, in fact, a very important influence
+in heading the sea-water back, and this coöperates
+with the natural change in the shape of a
+wave as it runs into shallow water, so as to exaggerate
+the steepness of the advancing slope of
+the wave.
+
+There are in the estuaries of many rivers
+\index{Rivers!tide wave in|)}%
+broad flats of mud or sand which are nearly dry
+at low water, and in such situations the tide not
+unfrequently rises with such great rapidity that
+the wave assumes the form of a wall of water.
+This sort of tide wave is called a ``bore,'' and in
+\index{Bore!definition}%
+French \textit{mascaret}. Notwithstanding the striking
+nature of the phenomenon, very little has been
+published on the subject, and I know of only one
+series of systematic observations of the bore.
+As the account to which I refer is contained in
+the official publications of the English Admiralty,
+it has probably come under the notice of only a
+\PageSep{60}
+small circle of readers. But the experiences of
+the men engaged in making these observations
+were so striking that an account of them should
+prove of interest to the general public. I have,
+moreover, through the kindness of Admiral Sir
+William Wharton and of Captain Moore, the
+\index{Moore, Captain!survey of Tsien-Tang-Kiang|(}%
+advantage of supplementing verbal description
+by photographs.
+
+The estuary on which the observations were
+made is that of the Tsien-Tang-Kiang, a considerable
+\index{Tsien-Tang-Kiang, the bore in|(}%
+river which flows into the China Sea about
+sixty miles south of the great Yang-Tse-Kiang.
+At most places the bore occurs only intermittently,
+but in this case it travels up the river at
+every tide. The bore may be observed within
+seventy miles of Shanghai, and within an easy
+walk of the great city of Hangchow; and yet
+\index{Hangchow, the bore at|(}%
+nothing more than a mere mention of it is to be
+found in any previous publication.
+
+In 1888 Captain Moore, R.~N., in command
+of Her Majesty's surveying ship Rambler,
+thought that it was desirable to make a thorough
+survey of the river and estuary. He returned
+to the same station in~1892; and the account
+which I give of his survey is derived from reports
+drawn up after his two visits. The annexed
+sketch-map shows the estuary of the
+Tsien-Tang, and the few places to which I shall
+have occasion to refer are marked thereon.
+
+On the morning of September~19, 1888, the
+\PageSep{61}
+Rambler was moored near an island, named
+after the ship, to the southwest of Chapu Bay;
+and on the~20th the two steam cutters Pandora
+and Gulnare, towing the sailing cutter
+\Figure{16}{Chart of the Estuary of the Tsien-Tang-Kiang}{png}
+Brunswick, left the ship with instruments for
+observing and a week's provisions.
+
+Captain Moore had no reason to suspect that
+the tidal currents would prove dangerous out
+in the estuary, and he proposed to go up the
+estuary about thirty miles to Haining, and then
+follow the next succeeding bore up-stream to
+Hangchow. Running up-stream with the flood,
+all went well until about~11.30, when they were
+about fifteen miles southwest by west of Kanpu.
+The leading boat, the Pandora, here grounded,
+and anchored quickly, but swung round violently
+as far as the keel would let her. The other
+boats, being unable to stop, came up rapidly;
+and the Gulnare, casting off the Brunswick,
+\PageSep{62}
+struck the Pandora, and then drove on to and
+over the bank, and anchored. The boats soon
+floated in the rising flood, and although the engines
+of the steam cutters were kept going
+full speed, all three boats dragged their anchors
+in an eleven-knot stream. When the flood
+slackened, the three boats pursued their course
+to the mouth of the river, where they arrived
+about 4~\PM. The ebb was, however, so violent
+that they were unable to anchor near one another.
+Their positions were chosen by the advice of
+some junkmen, who told Captain Moore, very
+erroneously as it turned out, that they would be
+safe from the night bore.
+
+The night was calm, and at~11.29 the murmur
+of the bore was heard to the eastward; it could
+be seen at~11.55, and passed with a roar at~12.20,
+well over toward the opposite bank, as predicted
+by the Chinese. The danger was now supposed
+to be past; but at~1~\AM\ a current of extreme
+violence caught the Pandora, and she had
+much difficulty to avoid shipwreck. In the
+morning it was found that her rudder-post and
+propeller-guard were broken, and the Brunswick
+and Gulnare were nowhere to be seen.
+They had, in fact, been in considerable danger,
+and had dragged their anchors three miles up
+the river. At 12.20~\AM\ they had been struck
+by a violent rush of water in a succession of big
+ripples. In a few moments they were afloat in
+\PageSep{63}
+an eight-knot current; in ten minutes the water
+rose nine feet, and the boats began to drag their
+anchors, although the engines of the Gulnare
+were kept going full speed. After the boats had
+dragged for three miles, the rush subsided, and
+when the anchor was hove up the pea and the
+greater part of the chain were as bright as polished
+silver.
+
+This account shows that all the boats were in
+imminent danger, and that great skill was needed
+to save them. After this experience and warning,
+the survey was continued almost entirely
+from the shore.
+
+The junks which navigate the river are well
+aware of the dangers to which the English boats
+were exposed, and they have an ingenious method
+of avoiding them. At various places on the
+bank of the river there are shelter platforms, of
+which I show an illustration in~\fig{17}. Immediately
+after the passing of the bore the
+\index{Bore!bore-shelter}%
+junks run up-stream with the after-rush and
+make for one of these shelters, where they allow
+themselves to be left stranded on the raised
+platform shown in the picture. At the end of
+this platform there is a sort of round tower
+jutting out into the stream. The object of this
+is to deflect the main wave of the bore so as to
+protect the junks from danger. After the passage
+of the bore, the water rises on the platform
+very rapidly, but the junks are just able to float
+\PageSep{64}
+in safety. Captain Moore gives a graphic account
+of the spectacle afforded by the junks as
+they go up-stream, and describes how on one
+occasion he saw no less than thirty junks swept
+\Figure{17}{Bore-Shelter on the Tsien-Tang-Kiang}{jpg}
+up in the after-rush, at a rate of ten knots, past
+the town of Haining toward Hangchow, with all
+sail set but with their bows in every direction.
+
+Measurements of the water-level were made
+in the course of the survey, and the results, in
+the form of a diagram, \fig{18}, exhibit the nature
+of the bore with admirable clearness. The
+observations of water-level were taken simultaneously
+at three places, viz., Volcano Island
+in the estuary, Rambler Island near the mouth
+of the river, and Haining, twenty-six miles up
+the river. In the figure, the distance between
+\PageSep{65}
+the lines marked Rambler and Volcano represents
+fifty-one miles, and that between Rambler
+and Haining twenty-six miles. The vertical
+scales show the height of water, measured in
+feet, above and below the mean level of the
+water at these three points. The lines joining
+these vertical scales, marked with the hours of
+the clock, show the height of the water simultaneously.
+The hour of~8.30 is indicated by
+the lowest line it shows that the water was
+one foot below mean level at Volcano Island,
+twelve feet below at Rambler Island, and eight
+feet below at Haining. Thus the water sloped
+down from Haining to Rambler, and from Volcano
+to Rambler; the water was running up the
+estuary toward Rambler Island, and down the
+estuary to the same point. At 9~and at~9.30
+there was no great change, but the water had
+risen two or three feet at Volcano Island and at
+Rambler Island. By ten~o'clock the water was
+rising rapidly at Rambler Island, so that there
+was a nearly uniform slope up the river from
+Volcano Island to Haining. The rise at Rambler
+Island then continued to be very rapid,
+while the water at Haining remained almost
+stationary. This state of affairs went on until
+midnight, by which time the water had risen
+twenty-one feet at Rambler Island, and about
+six feet at Volcano Island, but had not yet risen
+at all at Haining. No doubt through the whole
+\PageSep{66}
+of this time the water was running down the
+river from Haining towards its mouth. It is
+clear that this was a state of strain which could
+not continue long, for there was over twenty
+feet of difference of level between Rambler
+Island, outside, and Haining, in the river. Almost
+exactly at midnight the strain broke down
+and the bore started somewhere between Rambler
+\index{Bore!diagram of rise in Tsien-Tang}%
+Island and Kanpu, and rushed up the river
+in a wall of water twelve feet high. This result
+is indicated in the figure by the presence of two
+lines marked ``midnight.'' After the bore had
+\Figure{18}{Diagram of the Flow of the Tide on
+the Tsien-Tang-Kiang}{jpg}
+passed there was an after-rush that carried the
+water up eight feet more. It was on this that
+the junks were swept up the stream, as already
+described. At~1.30 the after-rush was over,
+\PageSep{67}
+but the water was still somewhat higher at
+Rambler Island than at Haining, and a gentle
+current continued to set up-stream. The water
+then began to fall at Rambler Island, while it
+continued to rise at Haining up to three o'clock.
+At this point the ebb of the tide sets in. I do
+not reproduce the figure which exhibits the fall
+of the water in the ebbing tide, for it may suffice
+to say that there is no bore down-stream,
+\index{Bore!pictures}%
+although there is at one time a very violent
+current.
+
+In 1892 Captain Moore succeeded, with considerable
+\index{Moore, Captain!illustrations of bore}%
+difficulty, in obtaining photographs of
+the bore as it passed Haining. They tell more
+of the violence of the wave than could be conveyed
+by any amount of description. The photographs,
+reproduced in~\fig{19}, do not, however,
+show that the broken water in the rear of the
+crest is often disturbed by a secondary roller, or
+miniature wave, which leaps up, from time to
+time, as if struck by some unseen force, and disappears
+in a cloud of spray. These breakers
+were sometimes twenty to thirty feet above the
+level of the river in front of the bore.
+
+The upper of these pictures is from a photograph,
+taken at a height of twenty-seven feet
+above the river, as the bore passed Haining on
+October~10, 1892. The height of this bore was
+eleven feet. The lower pictures, also taken at
+Haining, represent the passage of the bore on
+\PageSep{68}
+October~9, 1892. The first of these photographs
+was taken at 1.29~\PM, and the second
+represents the view only one minute later.
+
+The Chinese regard the bore with superstitious
+\index{Bore!Chinese superstition|(}%
+\index{Chinese!superstition as to bore|(}%
+reverence, and their explanation, which I quote
+from Captain Moore's report, is as follows:
+``Many hundred years ago there was a certain
+general who had obtained many victories over
+the enemies of the Emperor, and who, being
+constantly successful and deservedly popular
+among his countrymen, excited the jealousy of
+his sovereign, who had for some time observed
+with secret wrath his growing influence. The
+Emperor accordingly caused him to be assassinated
+and thrown into the Tsien-Tang-Kiang,
+where his spirit conceived the idea of revenging
+itself by bringing the tide in from the ocean in
+such force as to overwhelm the city of Hangchow,
+then the magnificent capital of the empire.
+As my interpreter, who has been for some years
+in America, put it, `his sowl felt a sort of ugly-like
+arter the many battles he had got for the
+Emperor.' The spirit so far succeeded as to
+flood a large portion of the country, when the
+Emperor, becoming alarmed at the distress and
+loss of property occasioned, endeavored to enter
+into a sort of compact with it by burning paper
+and offering food upon the sea-wall. This, however,
+did not have the desired effect, as the high
+tide came in as before; and it was at last determined
+\PageSep{69}
+\ifthenelse{\boolean{ForPrinting}}{%
+ \TallFig[0.7]{19}{Pictures of the Bore on the Tsien-Tang-Kiang}{jpg}
+}{%
+ \Figure[0.75]{19}{Pictures of the Bore on the Tsien-Tang-Kiang}{jpg}
+}
+\index{Wharton, Sir W. J., illustration of bore}%
+\PageSep{70}
+to erect a pagoda at the spot where the
+worst breach in the embankment had been made.
+Hence the origin of the Bhota Pagoda. A
+pagoda induces the good \textit{fungshui}, or spirit.
+After it was built the flood tide, though it still
+continued to come in the shape of a bore, did
+not flood the country as before.''
+
+We ``foreign devils'' may take the liberty of
+suspecting that the repairs to the embankment
+had also some share in this beneficial result.
+
+This story is remarkable in that it refers to
+the reign of an Emperor whose historical existence
+is undoubted. It thus differs from many
+of the mythical stories which have been invented
+by primitive peoples to explain great natural
+phenomena. There is good reason to suppose,
+in fact, that this bore had no existence some centuries
+ago; for Marco Polo, in the thirteenth
+\index{Marco Polo, resident of Hangchow}%
+century, stayed about a year and a half at
+Hangchow, and gives so faithful and minute
+\index{Hangchow, the bore at|)}%
+an account of that great town that it is almost
+impossible to believe that he would have omitted
+to notice a fact so striking. But the Emperor
+referred to in the Chinese legend reigned some
+centuries before the days of Marco Polo, so that
+we have reason to believe that the bore is intermittent.
+\index{Bore!Chinese superstition|)}%
+I have also learned from Captain
+Moore himself that at the time of the great
+\index{Moore, Captain!survey of Tsien-Tang-Kiang|)}%
+Taiping rebellion, the suppression of which was
+principally due to ``Chinese'' Gordon, the intensity
+\index{Chinese!superstition as to bore|)}%
+\index{Tsien-Tang-Kiang, the bore in|)}%
+\PageSep{71}
+of the bore was far less than it is to-day.
+\index{Bore!rivers where found}%
+This shows that the bore is liable to great variability,
+according as the silting of the estuary
+changes.
+
+The people at Haining still continue to pay
+religious reverence to the bore, and on one of
+the days when Captain Moore was making observations
+some five or six thousand people assembled
+on the river-wall to propitiate the god of
+the waters by throwing in offerings. This was
+the occasion of one of the highest bores at spring
+tide, and the rebound of the bore from the sea-wall,
+and the sudden heaping up of the waters
+as the flood conformed to the narrow mouth of
+the river, here barely a mile in width at low
+water, was a magnificent spectacle. A series of
+breakers were formed on the back of the advancing
+flood, which for over five minutes were not
+less than twenty-five feet above the level of the
+river in front of the bore. On this occasion
+Captain Moore made a rough estimate that a
+million and three quarters of tons of water passed
+the point of observation in one minute.
+
+The bore of which I have given an account is
+perhaps the largest known; but relatively small
+ones are to be observed on the Severn and Wye
+\index{Severn, bore in the}%
+\index{Wye, bore in the}%
+in England, on the Seine in France, on the Petitcodiac
+\index{Petitcodiac, bore in the}%
+\index{Seine, bore in the}%
+in Canada, on the Hugli in India, and
+\index{Hugli, bore on the}%
+doubtless in many other places. In general,
+however, it is only at spring tides and with certain
+\PageSep{72}
+winds that the phenomenon is at all striking.
+In September,~1897, I was on the banks of the
+Severn at spring tide; but there was no proper
+bore, and only a succession of waves up-stream,
+\index{Bore!causes}%
+and a rapid rise of water-level.
+
+I have shown, at the beginning of this chapter,
+that the heading back of the sea water by
+the natural current of a river, and the progressive
+change of shape of a wave in shallow water combine
+to produce a rapid rise of the tide in rivers.
+But the explanation of the bore, as resulting
+from these causes, is incomplete, because it leaves
+their relative importance indeterminate, and
+serves rather to explain a rapid rise than an absolutely
+sudden one. I think that it would be
+impossible, from the mere inspection of an estuary,
+to say whether there would be a bore there;
+we could only say that the situation looked
+promising or the reverse.
+
+The capriciousness of the appearance of the
+bore proves in fact that it depends on a very nice
+balance between conflicting forces, and the irregularity
+in the depth and form of an estuary renders
+the exact calculation of the form of the
+rising tide an impossibility. It would be easy
+to imitate the bore experimentally on a small
+scale; but, as in many other physical problems,
+we must rest satisfied with a general comprehension
+of the causes which produce the observed
+result.
+\PageSep{73}
+
+The manner in which the Chinese avail themselves
+of the after-rush for ascending the river
+affords an illustration of the utilization by mankind
+of tidal energy. In going up-stream, a
+\index{Energy, tidal, utilization of}%
+barge, say of one hundred tons, may rise some
+twenty or thirty feet. There has, then, been
+done upon that barge a work of from two to
+three thousand foot-tons. Whence does this
+energy come? Now, I say that it comes from
+the rotation of the earth; for we are making the
+tide do the work for us, and thus resisting the
+tidal movement. But resistance to the tide has
+the effect of diminishing the rate at which the
+earth is spinning round. Hence it is the earth's
+rotation which carries the barge up the river, and
+we are retarding the earth's rotation and making
+the day infinitesimally longer by using the tide
+in this way. This resistance is of an analogous
+character to that due to tidal friction, the consideration
+of which I must defer to a future
+chapter, as my present object is to consider the
+uses which may be made of tidal energy.
+
+It has been supposed by many that when the
+coal supply of the world has been exhausted we
+shall fall back on the tides to do our work. But
+a little consideration will show that although this
+source of energy is boundless, there are other far
+more accessible funds on which to draw.
+
+I saw some years ago a suggestion that the
+rise and fall of old hulks on the tide would afford
+\PageSep{74}
+\index{Energy, tidal, utilization of}%
+serviceable power. If we picture to ourselves the
+immense weight of a large ship, we may be deluded
+for a moment into agreement with this
+project, but numerical calculation soon shows its
+futility. The tide takes about six hours to rise
+from low water to high water, and the same
+period to fall again. Let us suppose that the
+water rises ten feet, and that a hulk of $10,000$
+tons displacement is floating on it; then it is
+easy to show that only twenty horse-power will
+be developed by its rise and fall. We should
+then require ten such hulks to develop as much
+work as would be given by a steam engine of
+very moderate size, and the expense of the installation
+would be far better bestowed on water-wheels
+in rivers or on wind-mills. I am glad to
+\index{Mills worked by the tide}%
+say that the projector of this scheme gave it up
+when its relative insignificance was pointed out
+to him. It is the only instance of which I ever
+heard where an inventor was deterred by the impracticability
+of his plan.
+
+We may, then, fairly conclude that, with existing
+mechanical appliances, the attempt to utilize
+the tide on an open coast is futile. But
+where a large area of tidal water can be easily
+trapped at high water, its fall may be made to
+work mill-wheels or turbines with advantage.
+The expense of building long jetties to catch the
+water is prohibitive, and therefore tide mills are
+only practicable where there exists an easily
+\PageSep{75}
+adaptable configuration of shoals in an estuary.
+There are, no doubt, many such mills in the
+\index{Mills worked by the tide}%
+world, but the only one which I happen to have
+seen is at Bembridge, in the Isle of Wight. At
+this place embankments formed on the natural
+shoals are furnished with lock-gates, and inclose
+many acres of tidal water. The gates open automatically
+with the rising tide, and the incipient
+outward current at the turn of the tide closes
+the gates again, so that the water is trapped.
+The water then works a mill wheel of moderate
+size. When we reflect on the intermittence of
+work from low water to high water and the great
+inequality of work with springs and neaps, it
+may be doubted whether this mill is worth the
+expense of retaining the embankments and lock-gates.
+
+We see then that, notwithstanding the boundless
+energy of the tide, rivers and wind and fuel
+are likely for all time to be incomparably more
+important for the use of mankind.
+
+\begin{Authorities}
+On waves in rivers see Airy's article on \Title{Tides and Waves} in
+\index{Airy, Sir G. B.!tides in rivers}%
+\index{Rivers!Airy on tide in}%
+the ``Encyclopædia Metropolitana.'' Some of his results will
+also be found in the article \Title{Tides} in the ``Encyclopædia Britannica.''
+
+Commander Moore, R.~N., \Title{Report on the Bore of the Tsien-Tang-Kiang.}
+Sold by Potter, Poultry, London,~1888.
+
+\Title{Further Report},~\&c., by the same author and publisher,~1893.
+\end{Authorities}
+\PageSep{76}
+
+
+\Chapter{IV}
+{Historical Sketch}
+
+\First{I Cannot} claim to have made extensive investigations
+\index{History!of tidal theories|(}%
+as to the ideas of mankind at different
+periods on the subject of the tides, but I propose
+in the present chapter to tell what I have
+been able to discover.
+
+No doubt many mythologies contain stories
+explanatory of the obvious connection between
+the moon and the tide. But explanations, professing
+at least to be scientific, would have been
+brought forward at periods much later than
+those when the mythological stories originated,
+and I shall only speak of the former.
+
+I have to thank my colleagues at Cambridge
+for the translations from the Chinese, Arabic,
+\index{Chinese!theories of tide}%
+Icelandic, and classical literatures of such passages
+as they were able to discover.
+
+I learn from Professor Giles that Chinese
+\index{Giles on Chinese theories of the tide}%
+writers have suggested two causes for the tides:
+first, that water is the blood of the earth, and
+that the tides are the beating of its pulse; and
+secondly, that the tides are caused by the earth
+breathing. Ko~Hung, a writer of the fourth
+century of our era, gives a somewhat obscure
+\PageSep{77}
+\index{Chinese!theories of tide}%
+\index{Giles on Chinese theories of the tide}%
+explanation of spring and neap tides. He says
+that every month the sky moves eastward and
+then westward, and hence the tides are greater
+and smaller alternately. Summer tides are said
+to be higher than winter tides, because in summer
+the sun is in the south and the sky is $15,000$~li
+($5,000$~miles) further off, and therefore in
+summer the female or negative principle in nature
+is weak, and the male or positive principle
+strong.
+
+In China the diurnal inequality is such that
+in summer the tide rises higher in the daytime
+than in the night, whilst the converse is true
+in winter. I suggest that this fact affords the
+justification for the statement that the summer
+tides are great.
+
+\TB
+
+Mr.~E.~G. Browne has translated for me the
+\index{Arabian theories of tide|(}%
+\index{Browne, E. G., Arabian theories of tide|(}%
+following passage from the ``Wonders of Creation''
+of Zakariyy\bar{a} ibn Muhammad ibn Mahm\bar{u}d
+al Qazv\bar{i}n\bar{i}, who died in \AD~1283.\footnote
+ {Wüstenfeld's edition, pp.~103,~104.}
+
+``Section treating of certain wonderful conditions
+of the sea.
+
+``Know that at different periods of the four
+seasons, and on the first and last days of the
+months, and at certain hours of the night and
+day, the seas have certain conditions as to the
+rising of their waters and the flow and agitation
+thereof.
+\PageSep{78}
+
+``As to the rising of the waters, it is supposed
+that when the sun acts on them it rarefies them,
+and they expand and seek a space ampler than
+that wherein they were before, and the one part
+repels the other in the five directions eastwards,
+westwards, southwards, northwards, and upwards,
+and there arise at the same time various winds
+on the shores of the sea. This is what is said
+as to the cause of the rising of the waters.
+
+``As for the flow of certain seas at the time
+of the rising of the moon, it is supposed that at
+the bottom of such seas there are solid rocks
+and hard stones, and that when the moon rises
+over the surface of such a sea, its penetrating
+rays reach these rocks and stones which are at
+the bottom, and are then reflected back thence;
+and the waters are heated and rarefied and seek
+an ampler space and roll in waves towards the
+seashore~\dots\ and so it continues as long as
+the moon shines in mid-heaven. But when she
+begins to decline, the boiling of the waters
+ceases, and the particles cool and become dense
+and return to their state of rest, and the currents
+run according to their wont. This goes
+on until the moon reaches the western horizon,
+when the flow begins again, as it did when the
+moon was in the eastern horizon. And this
+flow continues until the moon is at the middle
+of the sky below the horizon, when it ceases.
+Then when the moon comes upward, the flow
+\PageSep{79}
+begins again until she reaches the eastern horizon.
+This is the account of the flow and ebb
+of the sea.
+
+``The agitation of the sea resembles the agitation
+of the humours in men's bodies, for verily
+as thou seest in the case of a sanguine or bilious
+man,~\&c., the humours stirring in his body, and
+then subsiding little by little; so likewise the
+sea has matters which rise from time to time as
+they gain strength, whereby it is thrown into
+violent commotion which subsides little by little.
+And this the Prophet (on whom be the blessings
+of God and his peace) hath expressed in a poetical
+manner, when he says: `Verily the Angel,
+who is set over the seas, places his foot in the
+sea and thence comes the flow; then he raises it
+and thence comes the ebb.'\,''
+\index{Arabian theories of tide|)}%
+\index{Browne, E. G., Arabian theories of tide|)}%
+
+\TB
+
+Mr.~Magnússon has kindly searched the old
+\index{Icelandic theory of tides}%
+\index{Magnússon on Icelandic theories of tides}%
+Icelandic literature for references to the tides.
+In the Rimbegla he finds this passage:---
+
+``Beda the priest says that the tides follow
+the moon, and that they ebb through her blowing
+on them, but wax in consequence of her
+movement.''
+
+And again:---
+
+``(At new moon) the moon stands in the way
+of the sun and prevents him from drying up the
+sea; she also drops down her own moisture.
+For both these reasons, at every new moon, the
+\PageSep{80}
+\index{Icelandic theory of tides}%
+\index{Magnússon on Icelandic theories of tides}%
+ocean swells and makes those tides which we call
+spring tides. But when the moon gets past the
+sun, he throws down some of his heat upon
+the sea, and diminishes thereby the fluidity of
+the water. In this way the tides of the sea
+are diminished.''
+
+In another passage the author writes:---
+
+``But when the moon is opposite to the sun,
+the sun heats the ocean greatly, and as nothing
+impedes that warmth, the ocean boils and the
+sea flood is more impetuous than before---just
+as one may see water rise in a kettle when it
+boils violently. This we call spring tide.''
+
+There seems to be a considerable inconsistency
+in explaining one spring tide by the interception
+of the sun's heat by the moon, and the next one
+by the excess of that heat.
+
+But it is not necessary to search ancient literature
+for grotesque theories of the tides. In
+1722 E.~Barlow, gentleman, in ``An Exact Survey
+of the Tide,''\footnote
+ {``The Second Edition, with Curious Maps.'' (London: John
+ Hooke, 1722.)}
+attributes it to the pressure
+of the moon on the atmosphere. And theories
+not less absurd have been promulgated during
+the last twenty years.
+
+\TB
+
+The Greeks and Romans, living on the shores
+of the Mediterranean, had not much occasion to
+learn about the tide, and the passages in classical
+\PageSep{81}
+\index{Greek!theory and description of tides|(}%
+\index{Roman description of tides|(}%
+literature which treat of this matter are but
+few. But where the subject is touched on we
+see clearly their great intellectual superiority over
+those other peoples, whose ideas have just been
+quoted.
+
+The only author who treats of the tide in any
+\index{Strabo on tides|(}%
+detail is Posidonius, and we have to rely for our
+\index{Posidonius on tides|(}%
+knowledge of his work entirely on quotations
+from him by Strabo.\footnote
+ {My attention was drawn to Strabo by a passage in Sir W.
+ Thomson's (Lord Kelvin's) Popular Lectures, \Title{The Tides}, vol.~ii.
+ I have to thank Mr.~Duff for the translations which follow from
+ Strabo and Posidonius. The work consulted was Bake's \Title{Posidonius}
+ (Leiden,~1810), but Mr.~Duff tells me that the text is very
+ corrupt in some places, and he has therefore also consulted a
+ more recent text.}
+
+Posidonius says that Aristotle attributed the
+\index{Aristotle on tides}%
+flow and ebb of the sea at Cadiz to the mountainous
+formation of the coast, but he very justly
+pronounces this to be nonsense, particularly as
+the coast of Spain is flat and sandy. He himself
+attributes the tides to the moon's influence, and
+the accuracy of his observations is proved by the
+following interesting passage from Strabo:\footnotemark---
+\footnotetext{Teubner's \Title{Strabo},~i.\ p.~236.}
+
+``Posidonius says that the movement of the
+ocean observes a regular series like a heavenly
+body, there being a daily, monthly, and yearly
+movement according to the influence of the
+moon. For when the moon is above the (eastern)
+horizon by the distance of one sign of the
+zodiac (\ie~$30°$) the sea begins to flow, and encroaches
+\PageSep{82}
+visibly on the land until the moon
+reaches the meridian. When she has passed the
+meridian, the sea in turn ebbs gradually, until
+the moon is above the western horizon by the
+distance of one sign of the zodiac. The sea then
+remains motionless while the moon is actually
+setting, and still more so (\textit{sic}) so long as the
+moon is moving beneath the earth as far as a
+sign of the zodiac beneath the horizon. Then
+the sea again advances until the moon has
+reached the meridian below the earth; and retreats
+while the moon is moving towards the east,
+until she is the distance of a sign of the zodiac
+below the horizon; it remains at rest until the
+moon is the same distance above the horizon, and
+then begins to flow again. Such is the daily
+movement of the tides, according to Posidonius.
+
+``As to their monthly movement, he says that
+the ebbs are greatest at the conjunctions [of
+the sun and moon], and then grow less until the
+time of half moon, and increase again until the
+time of full moon, and grow less again until
+the moon has waned to half. Then the increase
+of the tide follows until the conjunction. But
+the increases last longer and come quicker [this
+phrase is very obscure].
+
+``The yearly movements of the tides he says
+he learned from the people of Cadiz. They told
+him that the ebb and flow alike were greatest at
+the summer solstice. He guesses for himself
+\PageSep{83}
+that the tides grow less from the solstice to the
+equinox, and then increase between the equinox
+and the winter solstice, and then grow less until
+the spring equinox, and greater until the summer
+solstice.''
+
+This is an excellent account of the tides at
+Cadiz, but I doubt whether there is any foundation
+\index{Polibius on tides at Cadiz}%
+for that part which was derived from hearsay.
+Lord Kelvin remarks, however, that it is interesting
+to note that inequalities extending over
+the year should have been recognized.
+
+Strabo also says that there was a spring near
+Cadiz in which the water rose and fell, and that
+this was believed by the inhabitants, and by
+Polybius, to be due to the influence of the ocean
+tide, but Posidonius was not of this opinion.
+Strabo says:---
+
+``Posidonius denies this explanation. He says
+there are two wells in the precinct of Hercules at
+Cadiz, and a third in the city. Of the two former
+the smaller runs dry while people are drawing
+water from it, and when they stop drawing water
+it fills again; the larger continues to supply
+water all day, but, like all other wells, it falls
+during the day but is replenished at night, when
+the drawing of water has ceased. But since the
+ebb tide often coincides with the replenishing of
+the well, therefore, says Posidonius, the idle story
+of the tidal influence has been believed by the
+inhabitants.''
+\PageSep{84}
+
+Since the wells follow the sun, whilst the tide
+follows the moon, the criticism of Posidonius is
+a very just one. But Strabo blames him for
+distrusting the Cadizians in a simple matter of
+everyday experience, whilst accepting their evidence
+as to an annual inequality in the tides.
+
+There is another very interesting passage in
+Strabo, the meaning of which was obviously unknown
+to the Dutch commentator Bake---and
+indeed must necessarily have been unintelligible
+to him at the time when he wrote, on account of
+the then prevailing ignorance of tidal phenomena
+in remoter parts of the world. Strabo
+writes:---
+
+``Anyhow Posidonius says that Seleucus of
+\index{Diurnal inequality!observed by Seleucus}%
+\index{Posidonius on tides|)}%
+\index{Seleucus, observation of tides of Indian Ocean}%
+the Red Sea [also called the Babylonian] declares
+that there is a certain irregularity and regularity
+in these phenomena [the tides], according
+to the different positions [of the moon] in the
+zodiac. While the moon is in the equinoctial
+signs, the phenomena are regular; but while she
+is in the signs of the solstices, there is irregularity
+both in the height and speed of the tides,
+and in the other signs there is regularity or the
+reverse in proportion to their nearness to the solstices
+or to the equinoxes.''
+
+Now let us consider the meaning of this.
+When the moon is in the equinoxes she is on
+the equator, and when she is in the solstices she
+is at her maximum distances to the north or
+\PageSep{85}
+south of the equator---or, as astronomers say, in
+her greatest north or south declination. Hence
+Seleucus means that, when the moon is on the
+\index{Seleucus, observation of tides of Indian Ocean}%
+equator, the tides follow one another, with two
+equal high and low waters a day; but when she
+is distant from the equator, the regular sequence
+is interrupted. In other words, the diurnal
+inequality (which I shall explain in a later chapter)
+vanishes when the moon is on the equator,
+and is at its maximum when the declination is
+greatest. This is quite correct, and since the
+diurnal inequality is almost evanescent in the
+\index{Diurnal inequality!observed by Seleucus}%
+Atlantic, whilst it is very great in the Indian
+Ocean, especially about Aden, it is clear that
+Seleucus had watched the sea there, just as we
+should expect him to do from his place of origin.
+
+\TB
+
+Many centuries elapsed after the classical
+period before any scientific thought was bestowed
+on the tides. Kepler recognized the
+\index{Kepler!ideas concerning tides}%
+tendency of the water on the earth to move
+towards the sun and the moon, but he was unable
+to submit his theory to calculation. Galileo
+\index{Galileo!blames Kepler for his tidal theory}%
+expresses his regret that so acute a man as
+Kepler should have produced a theory, which
+appeared to him to reintroduce the occult qualities
+of the ancient philosophers. His own explanation
+referred the phenomenon to the rotation
+of the earth, and he considered that it afforded
+a principal proof of the Copernican system.
+\index{Greek!theory and description of tides|)}%
+\index{Roman description of tides|)}%
+\index{Strabo on tides|)}%
+\PageSep{86}
+\index{Kepler!ideas concerning tides}%
+
+The theory of tide-generating force which will
+be set forth in \Ref{Chapter}{V}.\ is due to Newton,
+\index{Newton!founder of tidal theory}%
+who expounded it in his ``Principia'' in~1687.
+His theory affords the firm basis on which all
+subsequent work has been laid.
+
+In 1738 the Academy of Sciences of Paris
+offered the theory of the tides as the subject for
+a prize. The authors of four essays received
+prizes, viz., Daniel Bernoulli, Euler, Maclaurin,
+\index{Bernoulli, Daniel, essay on tides}%
+\index{Euler, essay on tides}%
+\index{Maclaurin!essay on tides}%
+and Cavalleri. The first three adopted, not only
+\index{Cavalleri, essay on tides}%
+the theory of gravitation, but also Newton's
+theory to its fullest extent. A considerable
+portion of Bernoulli's work is incorporated in
+the account of the theory of the tides which I
+shall give later. The essays of Euler and Maclaurin
+contained remarkable advances in mathematical
+knowledge, but did not add greatly to
+the theory of the tides. The Jesuit priest
+Cavalleri adopted the theory of vortices to explain
+the tides, and it is not worth while to
+follow him in his erroneous and obsolete speculations.
+
+Nothing of importance was added to our
+knowledge until the great French mathematician
+Laplace took up the subject in~1774. It was he
+\index{Laplace!theory of tides|(}%
+who for the first time fully recognized the difficulty
+of the problem, and showed that the earth's
+rotation is an essential feature in the conditions.
+The actual treatment of the tidal problem is in
+effect due to Laplace, although the mode of
+\PageSep{87}
+presentment of the theory has come to differ
+considerably from his.
+
+Subsequently to Laplace, the most important
+workers in this field have been Sir John Lubbock
+\index{Lubbock, Sir J., senior, on tides}%
+senior, Whewell, Airy, and Lord Kelvin.
+\index{Kelvin, Lord!initiates harmonic analysis}%
+\index{Whewell!on tides}%
+\index{Whewell!empirical construction of tide tables|(}%
+The work of Lubbock and Whewell is chiefly
+remarkable for the coördination and analysis of
+enormous masses of data at various ports, and
+the construction of trustworthy tide tables.
+Airy contributed an important review of the
+whole tidal theory. He also studied profoundly
+the theory of waves in canals, and considered
+the effects of frictional resistances on the progress
+of tidal and other waves.
+
+Lord Kelvin initiated a new and powerful
+method of considering tidal oscillations. His
+method possesses a close analogy with that already
+used in discussing the irregularities in the
+motions of the moon and planets. His merit
+consists in the clear conception that the plan of
+procedure which has been so successful in the
+one case would be applicable to the other. The
+difference between the laws of the moon's motion
+and those of tidal oscillations is, however,
+so great that there is scarcely any superficial
+resemblance between the two methods. This
+so-called ``harmonic analysis'' of the tides is
+\index{Harmonic analysis!initiated by Lord Kelvin}%
+daily growing in favor in the eyes of men of
+science, and is likely to supersede all the older
+methods. I shall explain it in a future chapter.
+\PageSep{88}
+
+Amongst all the grand work which has been
+bestowed on this difficult subject, Newton stands
+out first, and next to him we must rank Laplace.
+However original any future contribution to the
+science of tides may be, it would seem as though
+it must perforce be based on the work of these
+two. The exposition which I shall give hereafter
+of the theory of oceanic tides is based on
+the work of Newton, Bernoulli, Laplace, and
+\index{Bernoulli, Daniel, essay on tides}%
+\index{Laplace!theory of tides|)}%
+Kelvin, in proportions of which it would be
+difficult to assign the relative importance.
+
+\TB
+
+The connection between the moon and the
+tide is so obvious that long before the formulation
+of a satisfactory theory fairly accurate predictions
+of the tides were made and published.
+On this head Whewell\footnote
+ {\Title{History of the Inductive Sciences}, 1837, vol.~ii.\ p.~248 \textit{et~seq.}}
+\index{History!of tidal theories|)}%
+has the following interesting
+passage:---
+
+``The course which analogy would have recommended
+for the cultivation of our knowledge of
+tides would have been to ascertain by an analysis
+of long series of observations, the effects of
+changes in the time of transit, parallax, and
+declination of the moon, and thus to obtain the
+laws of phenomena; and then to proceed to
+investigate the laws of causation.
+
+``Though this was not the course followed by
+mathematical theorists, it was really pursued by
+those who practically calculated tide tables; and
+\PageSep{89}
+the application of knowledge to the useful purposes
+of life, being thus separated from the
+promotion of the theory, was naturally treated
+as a gainful property, and preserved by secrecy.
+\dots~Liverpool, London, and other places, had
+their tide tables, constructed by undivulged
+methods, which methods, in some instances at
+least, were handed down from father to son for
+several generations as a family possession; and
+the publication of new tables accompanied by a
+statement of the mode of calculation was resented
+as an infringement of the rights of property.
+
+``The mode in which these secret methods
+were invented was that which we have pointed
+out,---the analysis of a considerable series of
+observations. Probably the best example of this
+was afforded by the Liverpool tide tables. These
+were deduced by a clergyman named Holden,
+from observations made at that port by a harbor
+master of the name of Hutchinson, who was
+led, by a love of such pursuits, to observe the
+tides for above twenty years, day and night.
+Holden's tables, founded on four years of these
+observations, were remarkably accurate.
+
+``At length men of science began to perceive
+that such calculations were part of their business;
+and that they were called upon, as the
+guardians of the established theory of the universe,
+to compare it in the greatest possible
+\PageSep{90}
+detail with the facts. Mr.~Lubbock was the
+first mathematician who undertook the extensive
+labors which such a conviction suggested. Finding
+that regular tide observations had been made
+at the London docks from~1795, he took nineteen
+years of these (purposely selecting the
+length of the cycle of the motions of the lunar
+orbit), and caused them (in~1831) to be analyzed
+by Mr.~Dessiou, an expert calculator. He thus
+obtained tables for the effect of the moon's
+declination, parallax, and hour of transit, on the
+tides; and was enabled to produce tide tables
+founded upon the data thus obtained. Some
+mistakes in these as first published (mistakes unimportant
+as to the theoretical value of the work)
+served to show the jealousy of the practical tide
+table calculators, by the acrimony with which the
+oversights were dwelt upon; but in a very few
+years the tables thus produced by an open and scientific
+process were more exact than those which
+resulted from any of the secrets; and thus practice
+was brought into its proper subordination to
+theory.''
+\index{Whewell!empirical construction of tide tables|)}%
+
+\begin{Authorities}
+The history from Galileo to Laplace is to be found in the
+\Title{Mécanique Céleste} of Laplace, book~xiii, chapter~i.
+
+The other authorities are quoted in the text or in footnotes.
+\end{Authorities}
+\PageSep{91}
+
+
+\Chapter{V}
+{Tide-generating Force}
+
+\First{It} would need mathematical reasoning to fully
+\index{Centripetal and centrifugal forces|(}%
+explain how the attractions of the sun and moon
+give rise to tide-generating forces. But as this
+\index{Forces!centripetal and centrifugal|(}%
+book is not intended for the mathematician, I
+must endeavor to dispense with technical language.
+
+A body in motion will move in a straight line,
+unless it is deflected from its straight path by
+some external force, and the resistance to the
+deflection is said to be due to inertia. The motion
+of the body then is equivalent in its effect
+to a force which opposes the deflection due to
+the external force, and in many cases it is permissible
+to abstract our attention from the motion
+of the system and to regard it as at rest, if
+at the same time we introduce the proper ideal
+forces, due to inertia, so that they shall balance
+the action of the real external forces.
+
+If I tie a string to a stone and whirl it round,
+the string is thrown into a state of tension. The
+natural tendency of the stone, at each instant, is
+to move onward in a straight line, but it is continuously
+deflected from its straight path by the
+\PageSep{92}
+tension of the string. In this case the ideal
+force, due to inertia, whereby the stone resists
+its continuous deflection, is called centrifugal
+force. This force is in reality only a substitute
+for the motion, but if we withdraw our attention
+from the motion, it may be regarded as a reality.
+
+The centrifugal force is transmitted to my
+hand through the string, and I thus experience
+an outward or centrifugal tendency. But the
+stone itself is continually pulled inward by the
+string, and the force is called centripetal. When
+a string is under tension, as in this experiment,
+it is subject to equal and opposite forces, so that
+the tension implies the existence of a pair of
+forces, one towards and the other away from the
+centre of rotation. The force is to be regarded
+as away from the centre when we consider the
+sensation of the whirler, and as towards the centre
+when we consider the thing whirled. A similar
+double view occurs in commerce, where a
+transaction which stands on the credit side in the
+books of one merchant appears on the debit side
+in the books of the other.
+
+This simple experiment exemplifies the mechanism
+by which the moon is kept revolving round
+the earth. There is not of course any visible
+connection between the two bodies, but an invisible
+bond is provided by the attraction of gravity,
+which replaces the string which unites the
+stone to the hand. The moon, then, whirls
+\PageSep{93}
+\index{Forces!tide-generating|(}%
+\index{Orbit!of moon and earth|(}%
+round the earth at just such a rate and at just
+\index{Earth and moon!diagram}%
+such a distance, that her resistance to circular
+motion, called centrifugal force, is counterbalanced
+by the centripetal tendency of gravity. If
+\index{Centripetal and centrifugal forces|)}%
+she were nearer to us the attraction of gravity
+would be greater, and she would have to go
+round the earth faster, so as to make enough
+centrifugal force to counterbalance the greater
+\Figure[0.8]{20}{Earth and Moon}{png}
+gravity. The converse would be true, and the
+moon would go round slower, if she were further
+from us.
+
+The moon and the earth go round the sun in
+companionship once in a year, but this annual
+motion does not affect the interaction between
+them, and we may put aside the orbital motion
+of the earth, and suppose the moon and earth to
+\index{Moon and earth!diagram}%
+be the only pair of bodies in existence. When
+the principle involved in a purely lunar tide is
+grasped, the action of the sun in producing a
+\index{Forces!centripetal and centrifugal|)}%
+\PageSep{94}
+solar tide will become obvious. But the analogy
+of the string and stone is imperfect in one
+respect where the distinction is important; the
+moon, in fact, does not revolve exactly about
+the earth, but about the centre of gravity of
+the earth and moon. The earth is eighty times
+as heavy as the moon, and so this centre of gravity
+is not very far from the earth's centre. The
+upper part of \fig{20} is intended to represent a
+planet and its satellite; the lower part shows
+the earth and the moon in their true proportions.
+The upper figure is more convenient for
+our present argument, and the planet and satellite
+may be described as the earth and the moon,
+notwithstanding the exaggeration of their relative
+proportions. The point~$G$ is the centre of
+gravity of the two, and the axis about which
+they revolve passes through~$G$. This point is
+sufficiently near to the centre of the earth to
+permit us, for many purposes, to speak of the
+moon as revolving round the earth. But in the
+present case we must be more accurate and must
+regard the moon and earth as revolving round~$G$,
+their centre of gravity. The moon and earth
+are on opposite sides of this point, and describe
+circles round it. The distance of the moon's
+centre from~$G$ is $237,000$~miles, whilst that of
+the earth's centre is only $\DPchg{3000}{3,000}$~miles in the opposite
+direction. The $\DPchg{3000}{3,000}$~and $237,000$~miles
+together make up the $240,000$~miles which separate
+the centres of the two bodies.
+\PageSep{95}
+
+A system may now be devised so as to resemble
+the earth and moon more closely than that
+of the string and stone with which I began. If
+a large stone and a small one are attached to one
+another by a light and stiff rod, the system can
+be balanced horizontally about a point in the rod
+called the centre of gravity~$G$. The two weights
+may then be set whirling about a pivot at~$G$, so
+that the rod shall always be horizontal. In consequence
+of the rotation the rod is brought into
+a state of stress, just as was the string in the
+first example, and the centripetal stress in the
+rod exactly counterbalances the centrifugal force.
+The big and the little stones now correspond to
+the earth and the moon, and the stress in the rod
+plays the same part as the invisible bond of
+gravity between the earth and the moon. Fixing
+our attention on the smaller stone or moon
+at the end of the longer arm of the rod, we see
+that the total centrifugal force acting on the
+moon, as it revolves round the centre of gravity,
+is equal and opposite to the attraction of the
+earth on the moon. On considering the short
+arm of the rod between the pivot and the big
+stone, we see also that the centrifugal force acting
+on the earth is equal and opposite to the
+attraction of the moon on it. In this experiment
+as well as in the former one, we consider
+the total of centrifugal force and of attraction,
+but every particle of both the celestial bodies is
+\index{Orbit!of moon and earth|)}%
+\PageSep{96}
+\index{Davis, method of presenting tide-generating force}%
+acted on by these forces, and accordingly a
+closer analysis is necessary.
+
+It will now simplify matters if we make a supposition
+which departs from actuality, introducing
+the true conditions at a later stage in the
+argument.
+
+The earth's centre describes a circle about the
+centre of gravity~$G$, with a radius of $\DPchg{3000}{3,000}$~miles,
+and the period of the revolution is of course one
+month. Now whilst this motion of revolution
+of the earth's centre continues, let it be supposed
+that the diurnal rotation is annulled. As this
+is a mode of revolution which differs from that
+of a wheel, it is well to explain exactly what is
+meant by the annulment of the diurnal rotation.
+This is illustrated in~\fig{21}, which shows the
+successive positions assumed by an arrow in revolution
+without rotation. The shaft of the arrow
+always remains parallel to the same direction in
+space, and therefore it does not rotate, although
+the whole arrow revolves. It is obvious that every
+particle of the arrow describes a circle of the
+same radius, but that the circles described by
+them are not concentric. The circles described
+by the point and by the base of the arrow are
+shown in the figure, and their centres are separated
+by a distance equal to the length of the
+arrow. Now the centrifugal force on a revolving
+particle acts along the radius of the circle described,
+and in this case the radii of the circles
+\PageSep{97}
+described by any two particles in the arrow are
+always parallel. The parallelism of the centrifugal
+forces at the two ends of the arrow is
+indicated in the figure. Then again, the centrifugal
+force must everywhere be equal as well
+as parallel, because its intensity depends both on
+the radius and on the speed of revolution, and
+these are the same for every part. It follows
+that if a body revolves without rotation, every
+part of it is subject to equal and parallel centrifugal
+forces. The same must therefore be
+true of the earth when deprived of diurnal rotation.
+Accordingly every particle of the idealized
+non-rotating earth is continuously subject to
+equal and parallel centrifugal forces, in consequence
+of the revolution of the earth's centre
+in its monthly orbit with a radius of $\DPchg{3000}{3,000}$~miles.\footnote
+ {I owe the suggestion of this method of presenting the origin
+ of tide-generating force to Professor Davis of Harvard
+\index{Davis, method of presenting tide-generating force}%
+ University.}
+
+We have seen that the total of centrifugal
+force acting on the whole earth must be just
+such as to balance the total of the centripetal
+forces due to the moon's attraction. If, then,
+the attractional forces, acting on every particle
+of the earth, were also equal and parallel, there
+would be a perfect balance throughout. We
+shall see, however, that although there is a perfect
+balance on the whole, there is not that uniformity
+\PageSep{98}
+which would render the balance perfect
+at every particle.
+
+As far as concerns the totality of the attraction
+the analogy is complete between the larger
+stone, revolving at the end of the shorter arm
+of the rod, and the earth revolving in its small
+\Figure[0.7]{21}{Revolution of a Body without Rotation}{png}
+orbit round~$G$. But a difference arises when we
+compare the distribution of the tension of the
+rod with that of the lunar attraction; for the
+rod only pulls at the stone at the point where it
+is attached to it, whereas the moon attracts every
+particle of the earth. She does not, however,
+attract every particle with equal force, for she
+pulls the nearer parts more strongly than the
+further, as is obvious from the nature of the law
+of gravitation. The earth's centre is distant
+sixty times its radius from the moon, so that the
+nearest and furthest parts are distant fifty-nine
+\PageSep{99}
+and sixty-one radii respectively. Hence the attractions
+at the nearest and furthest parts differ
+only a little from the average, namely, that at
+the centre; but it is just these small differences
+which are important in this matter.
+
+Since on the whole the attractions and the centrifugal
+forces are equal and opposite, and since
+the centrifugal forces acting on the non-rotating
+earth are equal and parallel at every part, and
+since the attraction at the earth's centre is the
+average attraction, it follows that where the attraction
+is stronger than the average it overbalances
+the centrifugal force, and where it is weaker
+it is overbalanced thereby.
+
+The result of the contest between the two sets
+of forces is illustrated in~\fig{22}. The circle
+represents a section of the earth, and the moon
+is a long way off in the direction~$M$.
+
+Since the moon revolves round the earth,
+whilst the earth is still deprived of rotation, the
+figure only shows the state of affairs at a definite
+instant of time. The face which the earth exhibits
+to the moon is always changing, and the
+moon returns to the same side of the earth only
+at the end of the month. Hence the section of
+the earth shown in this figure always passes
+through the moon, while it is continually shifting
+with respect to the solid earth. The arrows in
+the figure show by their directions and lengths
+the magnitudes and directions of the overbalance
+\PageSep{100}
+in the contest between centrifugal and centripetal
+tendencies. The point~$V$ in the figure is
+the middle of the hemisphere, which at the moment
+portrayed faces full towards the moon. It
+\Figure[0.7]{22}{Tide-Generating Force}{png}
+is the middle of the round disk which the man in
+the moon looks at. The middle of the face invisible
+to the man in the moon is at~$I$. The
+point of the earth which is only fifty-nine earth's
+radii from the moon is at~$V$. Here attraction
+overbalances centrifugal force, and this is indicated
+by an arrow pointing towards the moon.
+The point distant sixty-one earth's radii from
+the moon is at~$I$, and attraction is here overbalanced,
+as indicated by the arrow pointing away
+from the moon.
+
+I shall have to refer hereafter to the intensities
+\PageSep{101}
+of these forces, and will therefore here pause
+to make some numerical calculations.
+
+The moon is distant from the earth's centre
+sixty times the earth's radius, and the attraction
+of gravity varies inversely as the square of the
+distance. Hence we may take $\frac{1}{60^{2}}$ or $\frac{1}{3,600}$ as a
+measure of the intensity of the moon's attraction
+at the earth's centre. The particle which occupies
+the centre of the earth is also that particle
+which is at the average distance of all the particles
+constituting the earth's mass. Hence $\frac{1}{60^{2}}$ or
+$\frac{1}{3,600}$ may be taken as a measure of the average
+attraction of the moon on every particle of the
+earth.
+
+Now the point~$V$ is only distant fifty-nine
+earth's radii from the moon, and therefore, on
+the same scale, the moon attraction is measured
+by $\frac{1}{59^{2}}$ or~$\frac{1}{3,481}$.
+
+The attraction therefore at~$V$ exceeds the average
+by $\frac{1}{59^{2}} - \frac{1}{60^{2}}$, or $\frac{1}{3,481} - \frac{1}{3,600}$. It will be well to
+express these results in decimals; now $\frac{1}{3,481}$ is
+$.000,287,27$, and $\frac{1}{3,600}$ is $.000,277,78$, so that the
+difference is~$.000,009,49$. It is important to
+notice that $\frac{2}{60^{3}}$ or $\frac{2}{216,000}$ is equal to~$.000,009,26$;
+so that the difference is nearly equal to~$\frac{2}{60^{3}}$.
+
+{\Loosen Again, the point~$I$ is distant sixty-one earth's
+radii from the moon, and the moon's attraction
+there is to be measured by $\frac{1}{61^{2}}$ or $\frac{1}{3,721}$. The attraction
+at~$I$ therefore falls below the average by
+$\frac{1}{60^{2}} - \frac{1}{61^{2}}$, or $\frac{1}{3,600} - \frac{1}{3,721}$; that is, by~$.000,277,78 - .000,268,75$,
+\PageSep{102}
+which is equal to~$.000,009,03$.
+This again does not differ much from~$\frac{2}{60^{3}}$.}
+
+These calculations show that the excess of the
+actual attraction at~$V$ above the average attraction
+is nearly equal to the excess of the average
+above the actual attraction at~$I$. These two
+excesses only differ from one another by $5$~per
+cent.\ of either, and they are both approximately
+equal to~$\frac{2}{60^{3}}$ on the adopted scale of measurement.
+
+The use of any particular scale of measurement
+is not material to this argument, and we
+should always find that the two excesses are
+nearly equal to one another. And further, if
+the moon were distant from the earth by any
+other number of earth's radii, we should find
+that the two excesses are each nearly equal to $2$
+divided by the cube of that number.\footnote
+ {\Loosen This argument is very easily stated in algebraic notation.
+ If $x$~be the number of earth's radii at which the moon is
+ placed, the points $V$~and~$I$ are respectively distant $x - 1$ and
+ $x + 1$ radii. Now $(x - 1)^{2}$~is nearly equal to~$x^{2} - 2x$ or to
+ $x^{2}(1 - \frac{2}{x})$, and therefore $\frac{1}{(x - 1)^{2}}$~is nearly equal to~$\frac{1}{x^{2}(1 - \frac{2}{x})}$, which is
+ nearly equal to~$\frac{1}{x^{2}}(1 + \frac{2}{x})$. Hence $\frac{1}{(x - 1)^{2}} - \frac{1}{x^{2}}$~is nearly equal to~$\frac{2}{x^{3}}$.
+ By a similar argument $(x + 1)^{2}$~is nearly equal to~$x^{2}(1 + \frac{2}{x})$,
+ and $\frac{1}{(x + 1)^{2}}$~is nearly equal to~$\frac{1}{x^{2}}(1 - \frac{2}{x})$; so that $\frac{1}{x^{2}} - \frac{1}{(x + 1)^{2}}$~is nearly
+ equal to~$\frac{2}{x^{3}}$.}
+
+We conclude then that the two overbalances
+at $V$~and~$I$, which will be called tide-generating
+forces, are nearly equal to one another, and vary
+\PageSep{103}
+inversely as the cube of the distance of the moon
+from the earth.
+
+The fact of the approximate equality of the
+overbalance or excess on the two sides of the
+earth is noted in the figure by two arrows at $V$
+and $I$ of equal lengths. The argument would
+be a little more complicated, if I were to attempt
+to follow the mathematician in his examination
+of the whole surface of the earth, and to trace
+from point to point how the balance between
+the opposing forces turns. The reader must
+accept the results of such an analysis as shown
+in \fig{22} by the directions and lengths of the
+arrows.
+
+We have already seen that the forces at $V$ and~$I$,
+the middles of the faces of the earth which
+are visible and invisible to the man in the moon,
+are directed away from the earth's centre. The
+edges of the earth's disk as seen from the moon
+are at $D$ and~$D$, and here the arrows point inwards
+to the earth's centre and are half as long
+as those at $V$ and~$I$. At intermediate points,
+they are intermediate both in size and direction.
+
+The only point in which the system considered
+differs from actuality is that the earth has
+been deprived of rotation. But this restriction
+may be removed, for, when the earth rotates
+once in $24$~hours, no difference is made in the
+forces which I have been trying to explain,
+\PageSep{104}
+although of course the force of gravity and the
+shape of the planet are affected by the rotation.
+This figure is called a diagram of tide-generating
+forces, because the tides of the ocean are due to
+the action of this system of forces.
+
+The explanation of tide-generating force is
+the very kernel of our subject, and, at the risk
+of being tedious, I shall look at it from a slightly
+different point of view. If every particle of the
+earth and of the ocean were acted on by equal
+and parallel forces, the whole system would
+move together and the ocean would not be displaced
+relatively to the earth; we should say
+that the ocean was at rest. If the forces were
+not quite equal and not quite parallel, there
+would be a slight residual effect tending to make
+the ocean move relatively to the solid earth. In
+other words, any defect from equality and parallelism
+in the forces would cause the ocean to
+move on the earth's surface.
+
+The forces which constitute the departure
+from equality and parallelism are called ``tide-generating
+forces,'' and it is this system which
+is indicated by the arrows in~\fig{22}. Tide-generating
+force is, in fact, that force which,
+superposed on the average force, makes the actual
+force. The average direction of the forces
+which act on the earth, as due to the moon's
+attraction, is along the line joining the earth's
+centre to the moon's centre, and its average
+\PageSep{105}
+intensity is equal to the force at the earth's
+centre.
+
+Now at~$V$ the actual force is straight towards~$M$,
+in the same direction as the average, but of
+greater intensity. Hence we find an arrow
+directed towards~$M$, the moon. At~$I$, the actual
+force is again in the same direction as, but of
+less intensity than, the average, and the arrow is
+directed away from~$M$, the moon. At~$D$, the
+actual force is almost exactly of the same intensity
+as the average, but it is not parallel thereto,
+and we must insert an inward force as shown by
+the arrow, so that when this is compounded with
+the average force we may get a total force in
+the right direction.
+
+Now let us consider how these forces tend to
+affect an ocean lying on the surface of the earth.
+The moon is directly over the head of an inhabitant
+of the earth, that is to say in his zenith,
+when he is at~$V$; she is right under his feet in
+the nadir when he is at~$I$; and she is in the
+observer's horizon, either rising or setting, when
+he is anywhere on the circle~$D$. When the
+inhabitant is at~$V$ or at~$I$ he finds that the tide-generating
+force is towards the zenith; when he
+is anywhere on the circle~$D$ he finds it towards
+the nadir. At other places he finds it directed
+towards or away from some point in the sky,
+except along two circles halfway between $V$ and~$D$,
+or between $I$ and~$D$, where the tide-generating
+\PageSep{106}
+force is level along the earth's surface, and may
+be called horizontal.
+
+A vertical force cannot make things move
+sideways, and so the sea will not be moved horizontally
+by it. The vertical part of the tide-generating
+force is not sufficiently great to
+overcome gravity, but will have the effect of
+making the water appear lighter or heavier. It
+will not, however, be effective in moving the
+water, since the water must remain in contact
+with the earth. We want, then, to omit the
+vertical part of the force and leave behind only
+the horizontal part, by which I mean a force
+which, to an observer on the earth's surface, is
+not directed either upwards or downwards, but
+along the level to any point of the compass.
+
+If there be a force acting at any point of the
+earth's surface, and directed upwards or downwards
+away from or towards some point in the
+sky other than the zenith, it may be decomposed
+into two forces, one vertically upwards or downwards,
+and another along the horizontal surface.
+Now as concerns the making of the tides,
+no attention need be paid to that part which
+is directed straight up or down, and the only
+important part is that along the surface,---the
+horizontal portion.
+
+Taking then the diagram of tide-generating
+forces in~\fig{22}, and obliterating the upward
+and downward portions of the force, we are left
+%% Plate 1
+\Figure[0.9]{23}{Horizontal Tide-Generating Force}{jpg}
+%% Facing page
+%[Blank Page]
+\PageSep{107}
+\index{Horizontal tide-generating force}%
+with a system of forces which may be represented
+by the arrows in the perspective picture of horizontal
+tide-generating force shown in~\fig{23}.
+
+If we imagine an observer to wander over the
+earth, $V$~is the place at which the moon is vertically
+over his head, and the circle~$D$, shown by
+the boundary of the shadow, passes through all
+the places at which the moon is in the horizon,
+just rising or setting. Then there is no horizontal
+force where the moon is over his head or under
+his feet, or where the moon is in his horizon
+either rising or setting, but everywhere else there
+is a force directed along the surface of the earth
+in the direction of the point at which the moon
+is straight overhead or underfoot.
+
+Now suppose $P$ to be the north pole of the
+earth, and that the circle $A_{1}$,~$A_{2}$, $A_{3}$, $A_{4}$,~$A_{5}$ is a
+parallel of latitude---say the latitude of London.
+Then if we watch our observer from external
+space, he first puts in an appearance on the picture
+at~$A_{1}$, and is gradually carried along to~$A_{2}$
+by the earth's rotation, and so onwards. Just before
+he comes to~$A_{2}$, the moon is due south of him,
+and the tide-generating force is also south, but
+not very large. It then increases, so that nearly
+three hours later, when he has arrived at~$A_{3}$, it
+is considerably greater. It then wanes, and
+when he is at~$A_{4}$ the moon is setting and the
+force is nil. After the moon has set, the force
+is directed towards the moon's antipodes, and it
+\PageSep{108}
+is greatest about three hours after moonset, and
+vanishes when the moon, still being invisible, is
+on the meridian.
+
+It must be obvious from this discussion that
+the lunar horizontal tide-generating force will
+differ, both as to direction and magnitude, according
+to the position of the observer on the
+earth and of the moon in the heavens, and that
+it can only be adequately stated by means of
+mathematical formulæ. I shall in the following
+chapter consider the general nature of the
+changes which the forces undergo at any point
+\index{Forces!tide-generating|)}%
+on the earth's surface.
+
+But before passing on to that matter it should
+be remarked that if the earth and sun had been
+the only pair of bodies in existence the whole of
+the argument would have applied equally well.
+Hence it follows that there is also a solar tide-generating
+force, which in actuality coëxists
+with the lunar force. I shall hereafter show
+how the relative importance of these two influences
+is to be determined.
+
+\begin{Authorities}
+Any mathematical work on the theory of the tides; for example,
+Thomson and Tait's \Title{Natural Philosophy}, Lamb's \Title{Hydrodynamics},
+Bassett's \Title{Hydrodynamics}, article \Title{Tides}, ``Encycl.\ Britan.,''
+Laplace's \Title{Mécanique Céleste},~\&c.
+\end{Authorities}
+\PageSep{109}
+
+
+\Chapter{VI}
+{Deflection of the Vertical}
+
+\First{The} intensity of tide-generating force is to be
+\index{Deflection of the vertical|(}%
+estimated by comparison with some standard, and
+it is natural to take as that standard the force of
+gravity at the earth's surface. Gravity acts in a
+vertical direction, whilst that portion of the tidal
+force which is actually efficient in disturbing the
+ocean is horizontal. Now the comparison between
+a small horizontal force and gravity is
+easily effected by means of a pendulum. For if
+the horizontal force acts on a suspended weight,
+the pendulum so formed will be deflected from
+the vertical, and the amount of deflection will
+measure the force in comparison with gravity.
+A sufficiently sensitive spirit level would similarly
+show the effect of a horizontal force by the
+displacement of the bubble. When dealing with
+tidal forces the displacements of either the pendulum
+\index{Forces!numerical estimate|(}%
+\index{Forces!deflection of vertical by|(}%
+or the level must be exceedingly minute,
+but, if measurable, they will show themselves as
+a change in the apparent direction of gravity.
+Accordingly a disturbance of this kind is often
+described as a deflection of the vertical.
+
+The maximum horizontal force due to the
+\PageSep{110}
+moon may be shown by a calculation, which involves
+the mass and distance of the moon, to
+have an intensity of $\frac{1}{11,660,000}$ of gravity.\footnote
+ {It does not occur to me that there is any very elementary
+ method of computing the maximum horizontal tidal force, but it
+ is easy to calculate the vertical force at the points $V$~or~$I$ in~\fig{22}.
+
+ {\Loosen The moon weighs $\frac{1}{80}$~of the earth, and has a radius $\frac{1}{4}$~as large.
+ Hence lunar gravity on the moon's surface is~$\frac{1}{80} × 4^{2}$, or $\frac{1}{5}$~of
+ terrestrial gravity at the earth's surface. The earth's radius is
+ $4,000$~miles and the moon's distance from the earth's centre
+ $240,000$~miles. Hence her distance from the nearer side of the
+ earth is $236,000$~miles. Therefore lunar gravity at the earth's
+ centre is $\frac{1}{5} × \frac{1}{240^{2}}$~of terrestrial gravity, and lunar gravity at the
+ point~$V$ is $\frac{1}{5} × \frac{1}{236^{2}}$~of the same. Therefore the tidal force at~$V$
+ is $\frac{1}{5} × \frac{1}{236^{2}} - \frac{1}{5} × \frac{1}{240^{2}}$~of terrestrial gravity. On multiplying the
+ squares of~$236$ and of~$240$ by~$5$, we find that this difference is
+ $\frac{1}{278,480} - \frac{1}{288,000}$. If these fractions are reduced to decimals
+ and the subtraction is performed, we find that the force at~$V$
+ is $.000,000,118,44$~of terrestrial gravity. When this decimal is
+ written as a fraction, we find the result to be $\frac{1}{8,450,000}$~of
+ gravity.}
+
+ Now it is the fact, although I do not see how to prove it in an
+ equally elementary manner, that the maximum horizontal tide-generating
+ force has an intensity equal to $\frac{3}{4}$~of the vertical force
+ at $V$~or~$I$. To find $\frac{3}{4}$~of the above fraction we must augment the
+ denominator by one third part. Hence the maximum horizontal
+ force is $\frac{1}{11,260,000}$~of gravity. This number does not agree exactly
+ with that given in the text; the discrepancy is due to the
+ fact that round numbers have been used to express the sizes and
+ distance apart of the earth and the moon, and their relative
+ masses.}
+Such a
+force must deflect the bob of a pendulum by the
+same fraction of the length of the cord by which
+it is suspended. If therefore the string were $10$~metres
+or $33$~feet in length, the maximum deflection
+of the weight would be $\frac{1}{11,660,000}$~of $10$~metres,
+\PageSep{111}
+or $\frac{1}{1,166}$~of a millimetre. In English measure this
+is $\frac{1}{29,000}$~of an inch. But the tidal force is reversed
+in direction about every six hours, so that the
+pendulum will depart from its mean direction by
+\index{Pendulum!curves traced by, under tidal force}%
+as much in the opposite direction. Hence the
+\Figure[0.8]{24}{Deflection of a Pendulum; the Moon and
+Observer on the Equator}{png}
+excursion to and fro of the pendulum under the
+lunar influence will be $\frac{1}{14,500}$~of an inch. With a
+pendulum one metre, or $3$~ft.\ $3$~in.\ in length,
+the range of motion of the pendulum bob is
+$\frac{1}{145,000}$~of an inch. For any pendulum of manageable
+length this displacement is so small, that it
+seems hopeless to attempt to measure it by direct
+observation. Nevertheless the mass and distance
+of the moon and the intensity of gravity being
+known with a considerable degree of accuracy, it
+is easy to calculate the deflection of the vertical
+at any time.
+\index{Forces!numerical estimate|)}%
+
+The curves which are traced out by a pendulum
+present an infinite variety of forms, corresponding
+\PageSep{112}
+to various positions of the observer on
+the earth and of the moon in the heavens. Two
+illustrations of these curves must suffice. \Fig{24}
+shows the case when the moon is on the celestial
+equator and the observer on the terrestrial
+equator. The path is here a simple ellipse,
+which is traversed twice over in the lunar day by
+the pendulum. The hours of the lunar day at
+\index{Pendulum!curves traced by, under tidal force}%
+which the bob occupies successive positions are
+marked on the curve.
+
+If the larger ellipse be taken to show the displacement
+of a pendulum when the sun and
+\Figure[0.8]{25}{Deflection of a Pendulum; the Moon in N.
+Declination~$15°$, the Observer in N. Latitude~$30°$}{png}
+moon coöperate at spring tide, the smaller one
+will show its path at the time of neap tide.
+
+In \fig{25} the observer is supposed to be in
+latitude~$30°$, whilst the moon stands $15°$~N. of
+the equator; in this figure no account is taken
+\PageSep{113}
+of the sun's force. Here also the hours are
+marked at the successive positions of the pendulum,
+which traverses this more complex curve
+only once in the lunar day. These curves are
+somewhat idealized, for they are drawn on the
+hypothesis that the moon does not shift her
+position in the heavens. If this fact were taken
+into account, we should find that the curve
+would not end exactly where it began, and that
+the character of the curve would change slowly
+from day to day.
+
+But even after the application of a correction
+for the gradual shift of the moon in the heavens,
+the curves would still be far simpler than in actuality,
+because the sun's influence has been left
+out of account. It has been remarked in the
+last chapter that the sun produces a tide-generating
+force, and it must therefore produce a
+deflection of the vertical. Although the solar
+deflection is considerably less than the lunar, yet
+it would serve to complicate the curve to a great
+degree, and it must be obvious then that when
+the full conditions of actuality are introduced
+the path of the pendulum will be so complicated,
+that mathematical formulæ are necessary for
+complete representation.
+
+Although the direct observation of the tidal
+deflection of the vertical would be impossible
+even by aid of a powerful microscope, yet several
+attempts have been made by more or less
+\PageSep{114}
+indirect methods. I have just pointed out that
+the path of a pendulum, although drawn on an
+ultra-microscopic scale, can be computed with a
+high degree of accuracy. It may then occur to
+the reader that it is foolish to take a great deal
+of trouble to measure a displacement which is
+scarcely measurable, and which is already known
+with fair accuracy. To this it might be answered
+that it would be interesting to watch the direct
+gravitational effects of the moon on the earth's
+surface. But such an interest does not afford
+the principal grounds for thinking that this
+attempted measurement is worth making. If the
+solid earth were to yield to the lunar attraction
+with the freedom of a perfect fluid, its surface
+would always be perpendicular to the direction
+of gravity at each instant of time. Accordingly
+a pendulum would then always hang perpendicularly
+to the average surface of the earth, and so
+there would be no displacement of the pendulum
+with reference to the earth's surface. If, then,
+the solid earth yields partially to the lunar attraction,
+the displacements of a pendulum must be
+of smaller extent relatively to the earth than if
+the solid earth were absolutely rigid. I must
+therefore correct my statement as to our knowledge
+of the path pursued by a pendulum, and
+say that it is known if the earth is perfectly
+unyielding. The accurate observation of the
+movement of a pendulum under the influence of
+\PageSep{115}
+the moon, and the comparison of the observed
+oscillation, with that computed on the supposition
+that the earth is perfectly stiff, would afford
+the means of determining to
+what extent the solid earth is
+yielding to tidal forces. Such
+a result would be very interesting
+as giving a measure of the
+stiffness of the earth as a whole.
+
+I must pass over the various
+\index{Cambridge, experiments with bifilar pendulum at|(}%
+\index{Darwin, G. H.!bifilar pendulum|(}%
+\index{Darwin, Horace, bifilar pendulum|(}%
+\index{Deflection of the vertical!experiments to measure|(}%
+\index{Pendulum!bifilar|(}%
+earlier attempts to measure the
+lunar attraction, and will only
+explain the plan, although it
+was abortive, used in~1879 by
+my brother Horace and myself.
+
+Our object was to measure
+the ultra-microscopic displacements
+of a pendulum with reference
+to the ground on which it
+stood. The principle of the apparatus
+used for this purpose is
+due to Lord Kelvin; it is very
+simple, although the practical
+application of it was not easy.
+
+%[** TN: Figure wrapped in the original]
+\Figure[0.2]{26}{Bifilar Pendulum}{png}
+
+\Fig{26} shows diagrammatically, and not drawn
+to scale, a pendulum~$\Seg{A}{B}$ hanging by two wires.
+At the foot of the pendulum there is a support~$C$
+attached to the stand of the pendulum; $D$~is a
+small mirror suspended by two silk fibres, one
+being attached to the bottom of the pendulum~$B$
+\PageSep{116}
+and the other to the support~$C$. When the
+two fibres are brought very close together, any
+movement of the pendulum perpendicular to the
+plane of the mirror causes the mirror to turn
+through a considerable angle. The two silk
+fibres diverge from one another, but if two vertical
+lines passing through the two points of suspension
+are $\frac{1}{1,000}$~of an inch apart, then when the
+pendulum moves one of these points through a
+millionth of an inch, whilst the other attached to~$C$
+remains at rest, the mirror will turn through
+an angle of more than three minutes of arc.
+A lamp is placed opposite to the mirror, and
+the image of the lamp formed by reflection in
+the mirror is observed. A slight rotation of the
+mirror corresponds to an almost infinitesimal
+motion of the pendulum, and even excessively
+small movements of the mirror are easily detected
+by means of the reflected image of the light.
+
+In our earlier experiments the pendulum was
+hung on a solid stone gallows; and yet, when
+the apparatus was made fairly sensitive, the image
+of the light danced and wandered incessantly.
+Indeed, the instability was so great that
+the reflected image wandered all across the room.
+We found subsequently that this instability was
+due both to changes of temperature in the stone
+gallows, and to currents in the air surrounding
+the pendulum.
+
+To tell of all the difficulties encountered
+\PageSep{117}
+might be as tedious as the difficulties themselves,
+so I shall merely describe the apparatus in its
+ultimate form. The pendulum was suspended,
+as shown in~\fig{26}, by two wires; the two wires
+being in an east and west plane, the pendulum
+could only swing north and south. It was hung
+inside a copper tube, just so wide that the solid
+copper cylinder, forming the pendulum bob, did
+not touch the sides of the tube. A spike projected
+from the base of the pendulum bob
+through a hole in the bottom of the tube. The
+mirror was hung in a little box, with a plate-glass
+front, which was fastened to the bottom of the
+copper tube. The only communication between
+the tube and the mirror-box was by the hole
+through which the spike of the pendulum projected,
+but the tube and mirror-box together
+formed a water-tight vessel, which was filled with
+a mixture of spirits of wine and boiled water.
+The object of the fluid was to steady the
+mirror and the pendulum, while allowing its
+slower movements to take place. The water was
+boiled to get rid of air in it, and the spirits of
+wine was added to increase the resistance of the
+fluid, for it is a remarkable fact that a mixture
+of spirits and water has considerably more viscosity
+or stickiness than either pure spirits or
+pure water.
+
+The copper tube, with the pendulum and mirror-box,
+was supported on three legs resting on
+\PageSep{118}
+a block of stone weighing a ton, and this stood
+on the native gravel in a north room in the laboratory
+at Cambridge. The whole instrument
+was immersed in a water-jacket, which was furnished
+with a window near the bottom, so that
+the little mirror could be seen from outside. A
+water ditch also surrounded the stone pedestal,
+and the water jacketing of the whole instrument
+made the changes of temperature very slow.
+
+A gas jet, only turned up at the moment of
+observation, furnished the light to be observed
+by reflection in the little mirror. The gas
+burner could be made to travel to and fro along
+a scale in front of the instrument. In the preliminary
+description I have spoken of the motion
+of the image of a fixed light, but it clearly
+amounts to the same thing if we measure the
+motion of the light, keeping the point of observation
+fixed. In our instrument the image of
+the movable gas jet was observed by a fixed telescope
+placed outside of the room. A bright
+light was unfortunately necessary, because there
+was a very great loss of light in the passages to
+and fro through two pieces of plate glass and a
+considerable thickness of water.
+
+Arrangements were made by which, without
+entering the room, the gas jet could be turned
+up and down, and could be made to move to and
+fro in the room in an east and west direction,
+until its image was observed in the telescope.
+\PageSep{119}
+There were also adjustments by which the two
+silk fibres from which the mirror hung could be
+brought closer together or further apart, thus
+making the instrument more or less sensitive.
+There was also an arrangement by which the image
+of the light could be brought into the field
+of view, when it had wandered away beyond the
+limits allowed for by the traverse of the gas jet.
+
+When the instrument was in adjustment, an
+observation consisted of moving the gas jet until
+its image was in the centre of the field of
+view of the telescope; a reading of the scale, by
+another telescope, determined the position of the
+gas jet to within about a twentieth of an inch.
+
+The whole of these arrangements were arrived
+at only after laborious trials, but all the precautions
+were shown by experience to be necessary,
+and were possibly even insufficient to guard the
+instrument from the effects of changes of temperature.
+I shall not explain the manner in which
+we were able to translate the displacements of
+the gas jet into displacements of the pendulum.
+It was not very satisfactory, and only gave approximate
+results. A subsequent form of an
+instrument of this kind, designed by my brother,
+has been much improved in this respect. It was
+he also who designed all the mechanical appliances
+in the experiment of which I am speaking.
+
+It may be well to reiterate that the pendulum
+was only free to move north and south, and that
+\PageSep{120}
+our object was to find how much it swung. The
+east and west motion of a pendulum is equally
+interesting, but as we could not observe both
+displacements at the same time, we confined our
+attention in the first instance to the northerly
+and southerly movements.
+
+When properly adjusted the apparatus was so
+sensitive that, if the bob of the pendulum moved
+through $\frac{1}{40,000}$~of a millimetre, that is, a millionth
+part of an inch, we could certainly detect the
+movement, for it corresponded to a twentieth
+of an inch in our scale of position of the gas
+jet. When the pendulum bob moved through
+this amount, the wires of the pendulum turned
+through one two-hundredth of a second of arc;
+this is the angle subtended by one inch at $770$~miles
+distance. I do not say that we could actually
+measure with this degree of refinement, but
+we could detect a change of that amount. In
+view of the instability of the pendulum, which
+still continued to some extent, it may be hard to
+gain credence for the statement that such a small
+deflection was a reality, so I will explain how we
+were sure of our correctness.
+
+In setting up the apparatus, work had to be
+conducted inside the room, and some preliminary
+observations of the reflected image of a stationary
+gas jet were made without the use of the telescope.
+The scale on which the reflected spot
+of light fell was laid on the ground at about
+\PageSep{121}
+seven feet from the instrument; in order to
+watch it I knelt on the pavement behind the
+scale, and leant over it. I was one day watching
+on the scale the spot of light which revealed
+the motion of the pendulum, and, being tired
+with kneeling, supported part of my weight on
+my hands a few inches in front of the scale.
+The place where my hands rested was on the
+bare earth, from which a paving stone had been
+removed. I was surprised to find quite a large
+change in the reading. It seemed at first incredible
+that my change of position was the cause,
+but after several trials I found that light pressure
+with one hand was quite sufficient to produce
+an effect. It must be remembered that this was
+not simply a small pressure delivered on the bare
+earth at, say, seven feet distance, but it was the
+difference of effect produced by the same pressure
+at seven feet and six feet; for, of course,
+the change only consisted in the distribution of
+the weight of a small portion of my body.
+
+It is not very easy to catch the telescopic image
+of a spot of light reflected from a mirror of
+the size of a shilling. Accordingly, in setting
+up our apparatus, we availed ourselves of this result,
+for we found that the readiest way of bringing
+the reflected image into the telescopic field
+of view was for one of us to move slowly about
+the room, until the image of the light was
+brought, by the warping of the soil due to his
+\PageSep{122}
+weight, into the field of view of the telescope.
+He then placed a heavy weight on the floor
+where he had been standing; this of course
+drove the image out of the field of view, but
+after he had left the room the image of the flame
+was found to be in the field.
+
+We ultimately found, even when no special
+pains had been taken to render the instrument
+sensitive, that if one of us was in the room, and
+stood at about sixteen feet south of the instrument
+with his feet about a foot apart, and slowly
+shifted his weight from one foot to the other, a
+distinct change was produced in the image of the
+gas flame, and of course in the position of the
+little mirror, from which the image was derived
+by reflection. It may be well to consider for
+a moment the meaning of this result. If one
+presses with a finger on a flat slab of jelly, a sort
+of dimple is produced, and if a pin were sticking
+upright in the jelly near the dimple, it would tilt
+slightly towards the finger. Now this is like
+what we were observing, for the jelly represents
+the soil, and the tilt of the pin corresponds to
+that of the pendulum. But the scale of the displacement
+is very different, for our pendulum
+stood on a block of stone weighing nearly a ton,
+which rested on the native gravel at two feet below
+the level of the floor, and the slabs of the
+floor were removed from all round the pendulum.
+The dimple produced by a weight of $140$~lbs.\ on
+\PageSep{123}
+\index{Distortion of soil!by weight}%
+\index{Elastic distortion!of soil by weight}%
+the stone paved floor must have been pretty
+small, and the slope of the sides of that dimple
+\index{Dimple!in soil, due to weight}%
+at sixteen feet must have been excessively slight;
+but we were here virtually observing the change
+of slope at the instrument, when the centre of
+the dimple was moved from a distance of fifteen
+feet to sixteen feet.
+
+It might perhaps be thought that all observation
+would be rendered impossible by the street
+traffic and by the ordinary work of the laboratory.
+But such disturbances only make tremors
+of very short period, and the spirits and water
+damped out quick oscillations so thoroughly, that
+no difference could be detected in the behavior
+of the pendulum during the day and during the
+night. Indeed, we found that a man could stand
+close to the instrument and hit the tub and pedestal
+smart blows with a stick, without producing
+any sensible effect. But it was not quite easy to
+try this experiment, because there was a considerable
+disturbance on our first entering the room;
+and when this had subsided small movements of
+the body produced a sensible deflection, by slight
+changes in the distribution of the experimenter's
+weight.
+
+It is clear that we had here an instrument of
+amply sufficient delicacy to observe the lunar
+tide-generating force, and yet we completely
+failed to do so. The pendulum was, in fact,
+always vacillating and changing its position by
+\PageSep{124}
+many times the amount of the lunar effect which
+we sought to measure.
+
+An example will explain how this was: A series
+of frequent readings were taken from July
+21st to~25th, 1881, with the pendulum arranged
+to swing north and south. We found that there
+was a distinct diurnal period, with a maximum at
+noon, when the pendulum bob stood furthest
+northward. The path of the pendulum was interrupted
+by many minor zigzags, and it would
+sometimes reverse its motion for an hour together.
+But the diurnal oscillation was superposed on a
+gradual drift of the pendulum, for the mean
+diurnal position traveled slowly southward. Indeed,
+in these four days the image disappeared
+from the scale three times over, and was brought
+back into the field of view three times by the
+appliance for that purpose. On the night between
+the 24th and~25th the pendulum took an
+abrupt turn northward, and the scale reading
+was found, on the morning of the~25th, nearly
+at the opposite end of the scale from that towards
+which it had been creeping for four days
+previously.
+
+Notwithstanding all our precautions the pendulum
+was never at rest, and the image of the
+flame was always trembling and dancing, or waving
+slowly to and fro. In fact, every reading of
+our scale had to be taken as the mean of the
+excursions to right and left. Sometimes for two
+\PageSep{125}
+or three days together the dance of the image
+would be very pronounced, and during other
+days it would be remarkably quiescent.
+
+The origin of these tremors and slower movements
+\index{Earthquakes!microsisms and earth tremors|(}%
+\index{Italian investigations in seismology|(}%
+\index{Tremors, earth}%
+is still to some extent uncertain. Quite
+recent investigations by Professor Milne seem to
+\index{Milne on seismology}%
+show that part of them are produced by currents
+in the fluid surrounding the pendulum, that
+others are due to changes in the soil of a very
+local character, and others again to changes
+affecting a considerable tract of soil. But when
+all possible allowance is made for these perturbations,
+it remains certain that a large proportion
+of these mysterious movements are due to minute
+earthquakes.
+
+Some part of the displacements of our pendulum
+was undoubtedly due to the action of the
+moon, but it was so small a fraction of the whole,
+that we were completely foiled in our endeavor
+to measure it.\footnote
+ {Since the date of our experiment the bifilar pendulum has
+ been perfected by my brother, and it is now giving continuous
+ photographic records at several observatories. It is now made
+ to be far less sensitive than in our original experiment, and no
+ attempt is made to detect the direct effect of the moon.}
+\index{Cambridge, experiments with bifilar pendulum at|)}%
+\index{Darwin, G. H.!bifilar pendulum|)}%
+\index{Darwin, Horace, bifilar pendulum|)}%
+\index{Deflection of the vertical!experiments to measure|)}%
+\index{Pendulum!bifilar|)}%
+
+The minute earthquakes of which I have
+\index{Microsisms, minute earthquakes|(}%
+spoken are called by Italian observers microsisms,
+and this name has been very generally
+adopted. The literature on the subject of seismology
+is now very extensive, and it would be
+out of place to attempt to summarize here the
+\PageSep{126}
+conclusions which have been drawn from observation.
+I may, however, permit myself to add a
+few words to indicate the general lines of the research,
+which is being carried on in many parts
+of the world.
+
+Italy is a volcanic country, and the Italians
+have been the pioneers in seismology. Their
+observations have been made by means of pendulums
+of various lengths, and with instruments
+of other forms, adapted for detecting vertical
+movements of the soil. The conclusions at
+which Father Bertelli arrived twenty years ago
+\index{Bertelli on Italian seismology}%
+may be summarized as follows:---
+
+The oscillation of the pendulum is generally
+\index{Pendulum!as seismological instrument}%
+parallel to valleys or chains of mountains in the
+neighborhood. The oscillations are independent
+of local tremors, velocity and direction of wind,
+rain, change of temperature, and atmospheric
+electricity.
+
+Pendulums of different lengths betray the
+movements of the soil in different manners, according
+to the agreement or disagreement of
+their natural periods of oscillation with the period
+of the terrestrial vibrations.
+
+The disturbances are not strictly simultaneous
+in the different towns of Italy, but succeed one
+another at short intervals.
+
+After earthquakes the ``tromometric'' or microseismic\DPnote{** [sic]}
+\index{Tromometer, a seismological instrument}%
+movements are especially apt to be in
+a vertical direction. They are always so when
+\PageSep{127}
+\index{Bertelli on Italian seismology}%
+the earthquake is local, but the vertical movements
+are sometimes absent when the shock
+occurs elsewhere. Sometimes there is no movement
+at all, even when the shock occurs quite
+close at hand.
+
+The positions of the sun and moon appear to
+have some influence on the movements of the
+pendulum, but the disturbances are especially
+\index{Pendulum!as seismological instrument}%
+frequent when the barometer is low.
+
+The curves of ``the monthly means of the
+tromometric movement'' exhibit the same forms
+in the various towns of Italy, even those which
+are distant from one another.
+
+The maximum of disturbance occurs near the
+winter solstice and the minimum near the summer
+solstice.
+
+At Florence a period of earthquakes is presaged
+\index{Earthquakes!microsisms and earth tremors|)}%
+\index{Tromometer, a seismological instrument}%
+by the magnitude and frequency of oscillatory
+movements in a vertical direction. These
+movements are observable at intervals and during
+several hours after each shock.
+
+Some very curious observations on microsisms
+\index{Microsisms, minute earthquakes|)}%
+have also been made in Italy with the microphone,
+by which very slight movements of the
+soil are rendered audible.
+
+Cavaliere de Rossi, of Rome, has established a
+``geodynamic'' observatory in a cave $700$~metres
+above the sea at Rocca di Papa, on the external
+slope of an extinct volcano.
+
+At this place, remote from all carriages and
+\PageSep{128}
+\index{Rossi on Italian seismology|(}%
+roads, he placed his microphone at a depth of $20$~metres
+\index{Microphone as a seismological instrument|(}%
+below the ground. It was protected
+against insects by woolen wrappings. Carpet
+was spread on the floor of the cave to deaden
+the noise from particles of stone which might
+possibly fall. Having established his microphone,
+he waited till night, and then heard noises which
+he says revealed ``natural telluric phenomena.''
+The sounds which he heard he describes as
+``roarings, explosions occurring isolated or in
+volleys, and metallic or bell-like sounds'' (\textit{fremiti,
+scopii isolati o di moschetteria, e suoni-metallici
+o di campana}). They all occurred
+mixed indiscriminately, and rose to maxima at
+irregular intervals. By artificial means he was
+able to cause noises which he calls ``rumbling (?)
+or crackling'' (\textit{rullo o crepito}). The roaring
+(\textit{fremito}) was the only noise which he could reproduce
+artificially, and then only for a moment.
+It was done by rubbing together the conducting
+wires, ``in the same manner as the rocks must
+rub against one another when there is an earthquake.''
+
+A mine having been exploded in a quarry at
+some distance, the tremors in the earth were
+audible in the microphone for some seconds
+subsequently.
+
+There was some degree of coincidence between
+the agitation of the pendulum-seismograph and
+the noises heard with the microphone.
+\PageSep{129}
+
+At a time when Vesuvius became active,
+Rocca di Papa was agitated by microsisms, and
+the shocks were found to be accompanied by the
+very same microphonic noises as before. The
+noises sometimes became ``intolerably loud;''
+especially on one occasion in the middle of the
+night, half an hour before a sensible earthquake.
+The agitation of the microphone corresponded
+exactly with the activity of Vesuvius.
+
+Rossi then transported his microphone to
+Palmieri's Vesuvian observatory, and worked in
+conjunction with him. He there found that
+each class of shock had its corresponding noise.
+The sussultorial shocks, in which I conceive the
+movement of the ground is vertically up and
+down, gave the volleys of musketry (\textit{i~colpi di
+moschetteria}), and the undulatory shocks gave
+the roarings (\textit{i~fremiti}). The two classes of
+noises were sometimes mixed up together.
+
+Rossi makes the following remarks: ``On
+Vesuvius I was put in the way of discovering
+that the simple fall and rise in the ticking which
+occurs with the microphone [\textit{battito del orologio
+unito al microfono}] (a phenomenon observed
+by all, and remaining inexplicable to all) is a
+consequence of the vibration of the ground.''
+This passage alone might perhaps lead one to
+suppose that clockwork was included in the circuit;
+but that this was not the case, and that
+``ticking'' is merely a mode of representing a
+\PageSep{130}
+\index{Paschwitz, von Rebeur!on horizontal pendulum|(}%
+\index{Pendulum!horizontal|(}%
+natural noise is proved by the fact that he subsequently
+says that he considers the ticking to
+be ``a telluric phenomenon.''
+
+Rossi then took the microphone to the Solfatara
+\index{Rossi on Italian seismology|)}%
+of Pozzuoli, and here, although no sensible
+tremors were felt, the noises were so loud as
+to be heard simultaneously by all the people in
+the room. The ticking was quite masked by
+other natural noises. The noises at the Solfatara
+were imitated by placing the microphone
+on the lid of a vessel of boiling water. Other
+seismic noises were then imitated by placing the
+microphone on a marble slab, and scratching
+\index{Microphone as a seismological instrument|)}%
+and tapping the under surface of it.
+
+The observations on Vesuvius led him to the
+conclusion that the earthquake oscillations have
+sometimes fixed ``nodes,'' for there were places
+on the mountain where no effects were observed.
+There were also places where the movement was
+intensified, and hence it may be concluded that
+the centre of disturbance may sometimes be very
+distant, even when the observed agitation is
+considerable.
+
+At the present time perhaps the most distinguished
+investigator in seismology is Professor
+Milne, formerly of the Imperial College of Engineering
+\index{Milne on seismology}%
+at Tokyo. His residence in Japan gave
+\index{Japan, frequency of earthquakes}%
+him peculiar opportunities of studying earthquakes,
+for there is, in that country, at least one
+earthquake per diem of sufficient intensity to
+\index{Italian investigations in seismology|)}%
+\PageSep{131}
+\index{Japan, frequency of earthquakes}%
+affect a seismometer. The instrument of which
+he now makes most use is called a horizontal
+pendulum. The principle involved in it is old,
+but it was first rendered practicable by von
+Rebeur-Paschwitz, whose early death deprived
+the world of a skillful and enthusiastic investigator.
+
+The work of Paschwitz touches more closely
+on our present subject than that of Milne, because
+he made a gallant attempt to measure the
+moon's tide-generating force, and almost persuaded
+himself that he had done so.
+
+The horizontal pendulum is like a door in its
+mode of suspension. If a doorpost be absolutely
+vertical, the door will clearly rest in any
+position, but if the post be even infinitesimally
+tilted the door naturally rests in one definite
+position. A very small shift of the doorpost is
+betrayed by a considerable change in the position
+of the door. In the pendulum the door is
+replaced by a horizontal boom, and the hinges
+by steel points resting in agate cups, but the
+principle is the same.
+
+The movement of the boom is detected and
+registered photographically by the image of a
+light reflected from certain mirrors. Paschwitz
+made systematic observations with his pendulum
+at Wilhelmshaven, Potsdam, Strassburg, and
+Orotava. He almost convinced himself at one
+time that he could detect, amidst the wanderings
+\PageSep{132}
+of the curves of record, a periodicity corresponding
+to the direct effect of the moon's action.
+But a more searching analysis of his results left
+the matter in doubt. Since his death the observations
+at Strassburg have been continued by
+M.~Ehlert. His results show an excellent consistency
+\index{Ehlert, observation with horizontal pendulum}%
+with those of Paschwitz, and are therefore
+\index{Paschwitz, von Rebeur!on horizontal pendulum|)}%
+confirmatory of the earlier opinion of the
+latter. I am myself disposed to think that the
+detection of the lunar attraction is a reality, but
+the effect is so minute that it cannot yet be
+relied on to furnish a trustworthy measurement
+of the amount of the yielding of the solid earth
+to tidal forces.
+
+It might be supposed that doubt could hardly
+arise as to whether or not the direct effect of
+the moon's attraction had been detected. But
+I shall show in the next chapter that at many
+places the tidal forces must exercise in an indirect
+manner an effect on the motion of a pendulum
+\index{Pendulum!horizontal|)}%
+much greater than the direct effect.
+
+It was the consideration of this indirect effect,
+and of other concomitants, which led us to
+abandon our attempted measurement, and to
+conclude that all endeavors in that direction
+were doomed to remain for ever fruitless. I can
+but hope that a falsification of our forecast by
+M.~Ehlert and by others may be confirmed.
+\PageSep{133}
+
+\begin{Authorities}
+G.~H. Darwin and Horace Darwin, ``Reports to the British
+Association for the Advancement of Science:''---
+
+\Title{Measurement of the Lunar Disturbance of Gravity.} York
+meeting, 1881, pp.~93--126.
+
+\Title{Second Report on the same}, with appendix. Southampton
+meeting, 1882, pp.~95--119.
+
+E.~von Rebeur-Paschwitz, \Title{Das Horizontalpendel}.
+
+``Nova Acta Leop.\ Carol.\ Akad.,'' 1892, vol.~lx.\ no.~1, p.~213;
+also ``Brit.\ Assoc.\ Reports,''~1893.
+
+E.~von Rebeur-Paschwitz, \Title{Ueber Horizontalpendel-Beobach\-tungen
+in Wilhelmshaven, Potsdam und Puerto Orotava auf Tenerifa}.
+
+``Astron.\ Nachrichten,'' vol.~cxxx.\ pp.~194--215.
+
+R.~Ehlert, \Title{Horizontalpendel-Beobachtungen}.
+
+``Beiträge zur Geophysik,'' vol.~iii.\ Part~I., 1896.
+
+C.~Davison, \Title{History of the Horizontal and Bifilar Pendulums}.
+\index{Davison, history of bifilar and horizontal pendulums}%
+
+``Appendix to Brit.\ Assoc.\ Report on Earth Tremors.'' Ipswich
+meeting, 1895, pp.~184--192.
+
+``British Association Reports of Committees.''
+
+\Title{On Earth Tremors}, 1891--95 (the first being purely formal).
+
+\Title{On Seismological Investigation}, 1896.
+
+The literature on Seismology is very extensive, and would
+\index{Seismology}%
+need a considerable index; the reader may refer to \Title{Earthquakes}
+and to \Title{Seismology} by John Milne. Both works form volumes in
+the International Scientific Series, published by Kegan Paul,
+Trench, Trübner \&~Co.
+\end{Authorities}
+\index{Deflection of the vertical|)}%
+\index{Forces!deflection of vertical by|)}%
+\PageSep{134}
+
+
+\Chapter[Distortion of the Earth's Surface]{VII}
+{The Elastic Distortion of the Earth's Surface
+by Varying Loads}
+
+\First{When} the tide rises and falls on the seacoast,
+\index{Darwin, G. H.!distortion of earth's surface by varying loads|(}%
+\index{Deflection of the vertical!due to tide|(}%
+\index{Distortion of soil!by varying loads|(}%
+\index{Elastic distortion!of earth by varying loads|(}%
+many millions of tons of water are brought alternately
+nearer and further from the land. Accordingly
+a pendulum suspended within a hundred
+miles or so of a seacoast should respond to the
+attraction of the sea water, swinging towards the
+sea at high water, and away from it at low water.
+Since the rise and fall has a lunar periodicity the
+pendulum should swing in the same period, even
+if the direct attraction of the moon did not affect
+it. But, as I shall now show, the problem is
+further confused by another effect of the varying
+tidal load.
+
+We saw in \Ref{Chapter}{VI}.\ how a weight resting
+on the floor in the neighborhood of our pendulum
+produced a dimple by which the massive
+stone pedestal of our instrument was tilted over.
+Now as low tide changes to high tide the position
+of an enormous mass of water is varied with
+respect to the land. Accordingly the whole
+coast line must rock to and fro with the varying
+tide. We must now consider the nature of the
+\PageSep{135}
+distortion of the soil produced in this way. The
+mathematical investigation of the form of the
+dimple in a horizontal slab of jelly or other elastic
+\index{Dimple!form of, in elastic slab}%
+material, due to pressure at a single point,
+shows that the slope at any place varies inversely
+as the square of the distance from the centre.
+That is to say, if starting from any point we
+proceed to half our original distance, we shall
+find four times as great a slope, and at one third
+\Figure[0.7]{27}{Form of Dimple in an Elastic Surface}{png}
+of the original distance the slope will be augmented
+ninefold.
+
+The theoretical form of dimple produced by
+pressure at a single mathematical point is shown
+in~\fig{27}. The slope is exaggerated so as to
+render it visible, and since the figure is drawn on
+the supposition that the pressure is delivered at
+a mathematical point, the centre of the dimple
+is infinitely deep. If the pressure be delivered
+by a blunt point, the slope at a little distance
+\PageSep{136}
+will be as shown, but the centre will not be infinitely
+deep. If therefore we pay no attention to
+the very centre, this figure serves to illustrate
+the state of the case. When the dimple is produced
+by the pressure of a weight, that weight,
+being endowed with gravitation, attracts any
+other body with a force varying inversely as the
+square of the distance. It follows, therefore,
+that the slope of the dimple is everywhere exactly
+\index{Slope of soil!due to elastic distortion}%
+proportional to the gravitational attraction
+\index{Attraction!of weight resting on elastic slab proportional to slope}%
+of the weight. Since this is true of a single
+weight, it is true of a group of weights, each
+producing its own dimple by pressure and its own
+attraction, strictly proportional to one another.
+Thus the whole surface is deformed by the superposition
+of dimples, and the total attraction is
+the sum of all the partial attractions.
+
+Let us then imagine a very thick horizontal
+slab of glass supporting any weights at any parts
+of its surface. The originally flat surface of the
+slab will be distorted into shallow valleys and
+low hills, and it is clear that the direct attraction
+of the weights will everywhere be exactly proportional
+to the slopes of the hillsides; also the
+direction of the greatest slope at each place must
+agree with the direction of the attraction. The
+direct attraction of the weights will deflect a
+pendulum from the vertical, and the deflection
+must be exactly proportional to the slope produced
+by the pressure of the weights. It may
+\PageSep{137}
+be proved that if the slab is made of a very stiff
+glass the angular deflection of the pendulum
+under the influence of attraction will be one fifth
+\index{Attraction!of weight resting on elastic slab proportional to slope}%
+of the slope of the hillside; if the glass were
+of the most yielding kind, the fraction would be
+one eighth. The fraction depends on the degree
+of elasticity of the material, and the stiffer it is
+the larger the fraction.
+
+The observation of a pendulum consists in
+noting its change of position with reference to
+the surface of the soil; hence the slope of the
+soil, and the direct attraction of the weight
+which causes that slope, will be absolutely fused
+together, and will be indistinguishable from one
+another.
+
+Now, this conclusion may be applied to the
+tidal load, and we learn that, if rocks are of the
+same degree of stiffness as glass of medium
+quality, the direct attraction of the tidal load
+produces one sixth of the apparent deflection of
+a pendulum produced by the tilting of the soil.
+
+If any one shall observe a pendulum, within
+say a hundred miles of the seacoast, and shall
+detect a lunar periodicity in its motion, he can
+only conclude that what he observes is partly
+due to the depression and tilting of the soil,
+partly to attraction of the sea water, and partly
+to the direct attraction of the moon. Calculation
+indicates that, with the known average elasticity
+of rock, the tilting of the soil is likely to
+\PageSep{138}
+\index{Elastic distortion!calculation and illustration|(}%
+\index{Slope of soil!calculation and illustration of|(}%
+be far greater than the other two effects combined.
+Hence, if the direct attraction of the
+moon is ever to be measured, it will first be
+necessary to estimate and to allow for other important
+oscillations with lunar periodicity. The
+difficulty thus introduced into this problem is so
+serious that it has not yet been successfully met.
+It may perhaps some day be possible to distinguish
+the direct effects of the moon's tidal attraction
+from the indirect effects, but I am not
+very hopeful of success in this respect. It was
+pointed out in \Ref{Chapter}{VI}.\ that there is some
+reason to think that a lunar periodicity in the
+swing of a pendulum has been already detected,
+and if this opinion is correct, the larger part of
+the deflection was probably due to these indirect
+effects.
+
+The calculation of the actual tilting of the
+coast line by the rising tide would be excessively
+complex even if accurate estimates were obtainable
+of the elasticity of the rock and of the tidal
+load. It is, however, possible to formulate a
+soluble problem of ideal simplicity, which will
+afford us some idea of the magnitude of the
+results occurring in nature.
+
+In the first place, we may safely suppose the
+earth to be flat, because the effect of the tidal
+load is quite superficial, and the curvature of the
+earth is not likely to make much difference in
+the result. In the second place, it greatly simplifies
+\PageSep{139}
+the calculation to suppose the ocean to
+consist of an indefinite number of broad canals,
+separated from one another by broad strips of
+land of equal breadth. Lastly, we shall suppose
+that each strip of sea rocks about its middle line,
+so that the water oscillates as in a seiche of the
+Lake of Geneva; thus, when it is high water
+on the right-hand coast of a strip of sea, it is
+low water on the left-hand coast, and vice versa.
+We have then to determine the change of shape
+of the ocean-bed and of the land, as the tide
+rises and falls. The problem as thus stated is
+\Figure{28}{Distortion of Land and Sea-bed by Tidal Load}{png}
+vastly simpler than in actuality, yet it will suffice
+to give interesting indications of what must
+occur in nature.
+
+The figure~\figref{28} shows the calculated result, the
+slopes being of course enormously exaggerated.
+The straight line represents the level surface of
+land and sea before the tidal oscillation begins,
+the shaded part being the land and the dotted
+part the sea. Then the curved line shows the
+form of the land and of the sea-bed, when it is
+low water at the right of the strip of land and high
+\PageSep{140}
+water at the left. The figure would be reversed
+when the high water interchanges position
+with the low water. Thus both land and sea
+rock about their middle lines, but the figure
+shows that the strip of land remains nearly flat
+although not horizontal, whilst the sea-bed becomes
+somewhat curved.
+
+It will be noticed that there is a sharp nick at
+the coast line. This arises from the fact that
+deep water was assumed to extend quite up to
+the shore line; if, however, the sea were given
+a shelving shore, as in nature, the sharp nick
+would disappear, although the form of the distorted
+rocks would remain practically unchanged
+elsewhere.
+
+Thus far the results have been of a general
+character, and we have made no assumptions as
+to the degree of stiffness of the rock, or as to
+the breadths of the oceans and continents. Let
+us make hypotheses which are more or less
+plausible. At many places on the seashore the
+tide ranges through twenty or thirty feet, but
+these great tides only represent the augmentation
+of the tide-wave as it runs into shallow
+water, and it would not be fair to suppose our
+tide to be nearly so great. In order to be moderate,
+I will suppose the tide to have a range of
+$160$~centimetres, or, in round numbers, about $5$~feet.
+Then, at the high-water side of the sea,
+the water is raised by $80$~centimetres, and at the
+\index{Elastic distortion!calculation and illustration|)}%
+\index{Slope of soil!calculation and illustration of|)}%
+\PageSep{141}
+low-water side it is depressed by the same
+amount. The breadth of the Atlantic is about
+$4,000$~or $5,000$~miles. I take then, the breadth of
+the oceans and of the continents as $3,900$~miles,
+or $6,280$~kilometres. Lastly, as rocks are usually
+stiffer than glass, I take the rock bed to
+be twice as stiff as the most yielding glass, and
+quarter as stiff again as the stiffest glass; this
+assumption as to the elasticity of rock makes the
+attraction at any place one quarter of the slope.
+For a medium glass we found the fraction to be
+about one sixth. These are all the data required
+for determining the slope.
+
+It is of course necessary to have a unit of
+measurement for the slope of the surface. Now
+a second of arc is the name for the angular
+magnitude of an inch seen at $3\frac{1}{4}$~miles, and accordingly
+a hundredth of a second of arc, usually
+written~$0''.01$, is the angular magnitude of
+an inch seen at $325$~miles; the angles will then
+be measured in hundredths of seconds.
+
+Before the tide rises, the land and sea-bed
+are supposed to be perfectly flat and horizontal.
+Then at high tides the slopes on the land are as
+follows:---
+\[
+\begin{array}{r@{\ }lc}
+\multicolumn{2}{c}{\ColHead{Distance from high-water mark}} &
+\multicolumn{1}{c}{\ColHead[2in]{Slope of the land measured in hundredths of seconds of arc}} \\
+ 10 &\text{metres} & 10 \\
+100 &\text{metres} & \Z8 \\
+ 1 &\text{kilometre} & \Z6 \\
+ 10 &\text{kilometres} & \Z4 \\
+ 20 &\text{kilometres} & \Z3\rlap{$\frac{1}{2}$} \\
+100 &\text{kilometres} & \Z2
+\end{array}
+\]
+\PageSep{142}
+The slope is here expressed in hundredths of a
+second of arc, so that at $100$~kilometres from the
+coast, where the slope is~$2$, the change of plane
+amounts to the angle subtended by one inch at
+$162$~miles.
+
+When high water changes to low water, the
+slopes are just reversed, hence the range of
+change of slope is represented by the doubles of
+these angles. If the change of slope is observed
+by some form of pendulum, allowance must be
+made for the direct attraction of the sea, and it
+appears that with the supposed degree of stiffness
+of rock these angles of slope must be augmented
+in the proportion of $5$~to~$4$. Thus, we
+double the angles to allow of change from high
+to low water, and augment the numbers as $5$ is
+to~$4$, to allow for the direct attraction of the sea.
+Finally we find results which may be arranged
+in the following tabular form:---
+\[
+\begin{array}{r@{\ }lc}
+\multicolumn{2}{c}{\ColHead{Distance from high-water mark}} &
+\multicolumn{1}{c}{\ColHead[1.75in]{Apparent range of deflection of the vertical}} \\
+ 10 &\text{metres} & 0''.25\Z \\
+100 &\text{metres} & 0''.20\Z \\
+ 1 &\text{kilometre} & 0''.15\Z \\
+ 10 &\text{kilometres} & 0''.10\Z \\
+ 20 &\text{kilometres} & 0''.084 \\
+100 &\text{kilometres} & 0''.050
+\end{array}
+\]
+At the centre of the continent, $1,950$~miles from
+the coast, the range will be~$0''.012$.
+
+If all the assumed data be varied, the ranges
+of the slopes are easily calculable, but these
+\PageSep{143}
+results may be taken as fairly representative, although
+perhaps somewhat underestimated. Lord
+Kelvin has made an entirely independent estimate
+\index{Kelvin, Lord!calculation of tidal attraction}%
+of the probable deflection of a pendulum
+by the direct attraction of the sea at high tide.
+\index{Attraction!of tide calculated}%
+He supposes the tide to have a range of $10$~feet
+from low water to high water, and he then estimates
+the attraction of a slab of water $10$~feet
+thick, $50$~miles broad perpendicular to the coast,
+and $100$~miles long parallel to the coast, on a
+plummet $100$~yards from low-water mark and
+opposite the middle of the $100$~miles. This
+would, he thinks, very roughly represent the
+state of things at St.~Alban's Head, in England.
+He finds the attraction such as to deflect the
+plumb-line, as high water changes to low water,
+by a twentieth of a second of arc. The general
+law as to the proportionality of slope to
+attraction shows that, with our supposed degree
+of stiffness of rock, the apparent deflection of a
+plumb-line, due to the depression of the coast
+and the attraction of the sea as high water
+changes to low water, will then be a quarter of a
+second of arc. Postulating a smaller tide, but
+spread over a wider area, I found the result
+would be a fifth of a second; thus the two results
+present a satisfactory agreement.
+
+This speculative investigation receives confirmation
+from observation. The late M.~d'Abbadie
+\index{Abbadie, tidal deflection of vertical}%
+established an observatory at his château of
+\index{Deflection of the vertical!due to tide|)}%
+\PageSep{144}
+\index{Abbadie, tidal deflection of vertical}%
+Abbadia, close to the Spanish frontier and within
+a quarter of a mile of the Bay of Biscay. Here
+he constructed a special form of instrument for
+detecting small changes in the direction of gravity.
+Without going into details, it may suffice
+to state that he compared a fixed mark with its
+image formed by reflection from a pool of mercury.
+He took $359$~special observations at the
+times of high and low tide in order to see, as he
+says, whether the water exercised an attraction
+on the pool of mercury, for it had not occurred
+to him that the larger effect would probably
+arise from the bending of the rock. He found
+that in $243$~cases the pool of mercury was tilted
+towards the sea at high water or away from it at
+low water; in $59$~cases there was no apparent
+effect, and in the remaining $57$~cases the action
+was inverted. The observations were repeated
+later by his assistant in the case of $71$~successive
+high waters\footnote
+ {Presumably the observation at one high water was defective.}
+and $73$~low waters, and he also
+found that in about two thirds of the observations
+the sea seemed to exercise its expected
+influence. We may, I think, feel confident that
+on the occasions where no effect or a reversal
+was perceived, it was annulled or reversed by a
+warping of the soil, such as is observed with
+seismometers.
+
+Dr.~von Rebeur-Paschwitz also noted deflections
+\index{Paschwitz, von Rebeur!tidal deflection of vertical at Wilhelmshaven}%
+due to the tide at Wilhelmshaven in Germany.
+\PageSep{145}
+\index{Atmospheric pressure!distortion of soil by}%
+\index{Elastic distortion!by atmospheric pressure|(}%
+\index{Pressure of atmosphere, elastic distortion of soil by}%
+The deflection was indeed of unexpected
+magnitude at this place, and this may probably
+be due to the peaty nature of the soil, which
+renders it far more yielding than if the observatory
+were built on rock.
+
+This investigation has another interesting application,
+for the solid earth has to bear another
+varying load besides that of the tide. The
+atmosphere rests on the earth and exercises a
+variable pressure, as shown by the varying
+height of the barometer. The variation of
+pressure is much more considerable than one
+would be inclined to suspect off-hand. The
+height of the barometer ranges through nearly
+two inches, or say five centimetres; this means
+that each square yard of soil supports a weight
+greater by $1,260$~lbs.\ when the barometer is very
+high, than when it is very low. If we picture
+to ourselves a field loaded with half a ton to
+each square yard, we may realize how enormous
+is the difference of pressure in the two cases.
+
+In order to obtain some estimate of the effects
+of the changing pressure, I will assume, as before,
+that the rocks are a quarter as stiff again
+as the stiffest glass. On a thick slab of this
+material let us imagine a train of parallel waves
+of air, such that at the crests of the waves the
+barometer is $5$~centimetres higher than at the
+hollow. Our knowledge of the march of barometric
+gradients on the earth's surface makes it
+\PageSep{146}
+plausible to assume that it is $1,500$~miles from
+the line of highest to that of lowest pressure.
+Calculation then shows that the slab is distorted
+into parallel ridges and valleys, and that the
+tops of the ridges are $9$~centimetres, or $3\frac{1}{2}$~inches,
+higher than the hollows. Although the actual
+distribution of barometric pressures is not of this
+simple character, yet this calculation shows, with
+a high degree of probability, that when the
+barometer is very high we are at least $3$~inches
+nearer the earth's centre than when it is very
+low.
+
+The consideration of the effects of atmospheric
+\index{Atmospheric pressure!distortion of soil by}%
+\index{Level of sea affected by atmospheric pressure}%
+\index{Pressure of atmosphere, elastic distortion of soil by}%
+pressure leads also to other curious conclusions.
+I have remarked before that the sea must respond
+\index{Sea!level affected by atmospheric pressure}%
+to barometric pressure, being depressed
+by high and elevated by low pressure. Since a
+column of water $68$~centimetres ($2$~ft.\ $3$~in.)\ in
+height weighs the same as a column, with the
+same cross section, of mercury, and $5$~centimetres
+in height, the sea should be depressed by $68$~centimetres
+under the very high barometer as compared
+with the very low barometer. But the
+height of the water can only be determined with
+reference to the land, and we have seen that the
+land must be depressed by $9$~centimetres. Hence
+the sea would be apparently depressed by only $59$~centimetres.
+
+It is probable that, in reality, the larger barometric
+inequalities do not linger quite long
+\PageSep{147}
+enough over particular areas to permit the sea to
+attain everywhere its due slope, and therefore the
+full difference of water level can only be attained
+occasionally. On the other hand the elastic compression
+\index{Elastic distortion!by atmospheric pressure|)}%
+of the ground must take place without
+sensible delay. Thus it seems probable that this
+compression must exercise a very sensible effect
+in modifying the apparent depression or elevation
+of the sea under high and low barometer.
+
+If delicate observations are made with some
+form of pendulum, the air waves and the consequent
+distortions of the soil should have a sensible
+effect on the instrument. In the ideal case
+which I have described above, it appears that
+the maximum apparent deflection of the plumb-line
+would be $\frac{1}{90}$~of a second of arc; this would
+be augmented to $\frac{1}{70}$~of a second by the addition
+of the true deflection, produced by the attraction
+of the air. Lastly, since the slope and attraction
+would be absolutely reversed when the air wave
+assumed a different position with respect to the
+observer, it is clear that the range of apparent
+oscillation of the pendulum might amount to
+$\frac{1}{35}$~of a second of arc.
+
+This oscillation is actually greater than that
+due to the direct tidal force of the moon acting
+on a pendulum suspended on an ideally unyielding
+earth. Accordingly we have yet another
+reason why the direct measurement of the tidal
+force presents a problem of the extremest difficulty.
+\PageSep{148}
+
+\begin{Authorities}
+G.~H. Darwin, \Title{Appendix to the Second Report on Lunar Disturbance
+of Gravity}. ``Brit.\ Assoc.\ Reports.'' Southampton,~1882.
+
+Reprint of the same in the ``Philosophical Magazine.''
+
+d'Abbadie, \Title{Recherches sur la verticale}. ``Ann.\ de~la Soc.\ Scient.\
+de~Bruxelles,'' 1881.
+
+von Rebeur-Paschwitz, \Title{Das Horizontalpendel}. ``Nova Acta
+K.~Leop.\ Car.\ Akad.,'' Band~60, No.~1, 1892.
+\end{Authorities}
+\index{Darwin, G. H.!distortion of earth's surface by varying loads|)}%
+\index{Distortion of soil!by varying loads|)}%
+\index{Elastic distortion!of earth by varying loads|)}%
+\PageSep{149}
+
+
+\Chapter{VIII}
+{Equilibrium Theory of Tides}
+
+\First{It} is clearly necessary to proceed step by step
+\index{Equilibrium theory of tides|(}%
+towards the actual conditions of the tidal problem,
+and I shall begin by supposing that the
+oceans cover the whole earth, leaving no dry
+land. It has been shown in \Ref{Chapter}{V}.\ that the
+tidal force is the resultant of opposing centrifugal
+and centripetal forces. The motion of the
+system is therefore one of its most essential features.
+We may however imagine a supernatural
+being, who carries the moon round the earth and
+makes the earth rotate at the actual relative
+speeds, but with indefinite slowness as regards
+absolute time. This supernatural being is further
+to have the power of maintaining the tidal forces
+at exactly their present intensities, and with their
+actual relationship as regards the positions of
+the moon and earth. Everything, in fact, is to
+remain as in reality, except time, which is to be
+indefinitely protracted. The question to be considered
+is as to the manner in which the tidal
+forces will cause the ocean to move on the slowly
+revolving earth.
+
+It appears from \fig{23} that the horizontal
+\PageSep{150}
+tidal force acts at right angles to the circle, where
+the moon is in the horizon, just rising or just
+setting, towards those two points, $V$~and~$I$, where
+the moon is overhead in the zenith, or underfoot
+in the nadir. The force will clearly generate
+currents in the water away from the circle of
+moonrise and moonset, and towards $V$~and~$I$.
+The currents will continue to flow until the water
+level is just so much raised above the primitive
+surface at $V$~and~$I$, and depressed along the circle,
+that the tendency to flow downhill towards
+the circle is equal to the tendency to flow uphill
+under the action of the tide-generating force.
+When the currents have ceased to flow, the figure
+of the ocean has become elongated, or egg-shaped
+with the two ends alike, and the longer
+axis of the egg is pointed at the moon. When
+this condition is attained the system is at rest or
+in equilibrium, and the technical name for the
+egg-like form is a ``prolate ellipsoid of revolution''---``prolate''
+because it is elongated, and
+``of revolution'' because it is symmetrical with
+respect to the line pointing at the moon. Accordingly
+the mathematician says that the figure
+of equilibrium under tide-generating force is a
+prolate ellipsoid of revolution, with the major
+axis directed to the moon.
+
+It has been supposed that the earth rotates and
+that the moon revolves, but with such extreme
+slowness that the ocean currents have time
+\PageSep{151}
+\index{Figure of equilibrium!of ocean under tidal forces|(}%
+\index{Forces!figure of equilibrium under tidal|(}%
+enough to bring the surface to its form of equilibrium,
+\index{Equilibrium theory of tides!chart and law of tide|(}%
+at each moment of time. If the time be
+sufficiently protracted, this is a possible condition
+of affairs. It is true that with the earth spinning
+at its actual rate, and with the moon revolving
+as in nature, the form of equilibrium can
+never be attained by the ocean; nevertheless it
+is very important to master the equilibrium
+theory.
+
+\Fig{29} represents the world in two hemispheres,
+as in an ordinary atlas, with parallels
+of latitude drawn at $15°$~apart. At the moment
+represented, the moon is supposed to be in the
+zenith at $15°$~of north latitude, in the middle of
+the right-hand hemisphere. The diametrically
+opposite point is of course at $15°$~of south latitude,
+in the middle of the other hemisphere.
+These are the two points $V$~and~$I$ of figs.\ \figref{22}~and~\figref{23},
+towards which the water is drawn, so that the
+vertices of the ellipsoid are at these two spots.
+A scale of measurement must be adopted for
+estimating the elevation of the water above, and
+its depression below the original undisturbed surface
+of the globe. It will be convenient to measure
+the elevation at these two spots by the
+number~$2$. A series of circles are drawn round
+these points, but one of them is, of necessity,
+presented as partly in one hemisphere and partly
+in the other. In the map they are not quite concentric
+with the two spots, but on the actual
+\PageSep{152}
+\TallFig{29}{Chart of Equilibrium Tide}{png}
+\PageSep{153}
+\index{Semidiurnal tide!in equilibrium theory|(}%
+globe they would be so. These circles show
+where, on the adopted scale of measurement, the
+elevation of height is successively $1\frac{1}{2}$,~$1$,~$\frac{1}{2}$. The
+fourth circle, marked in chain dot, shows where
+there is no elevation or depression above the original
+surface. The next succeeding and dotted
+circle shows where there is a depression of~$\frac{1}{2}$, and
+the last dotted line is the circle of lowest water
+where the depression is~$1$; it is the circle~$\Seg{D}{D}$ of~\fig{22},
+and the circle of the shadow in~\fig{23}.
+
+The elevation above the original spherical surface
+at the vertices or highest points is just twice
+as great as the greatest depression. But the
+greatest elevation only occurs at two points,
+whereas the greatest depression is found all along
+a circle round the globe. The horizontal tide-generating
+force is everywhere at right angles to
+these circles, and the present figure is in effect a
+reproduction, in the form of a map, of the perspective
+picture in~\fig{23}.
+
+Now as the earth turns from west to east, let
+us imagine a man standing on an island in the
+otherwise boundless sea, and let us consider what
+he will observe. Although the earth is supposed
+to be revolving very slowly, we may still call the
+twenty-fourth part of the time of its rotation an
+hour. The man will be carried by the earth's
+rotation along some one of the parallels of latitude.
+If, for example, his post of observation is
+in latitude $30°$~N., he will pass along the second
+\index{Equilibrium theory of tides!chart and law of tide|)}%
+\index{Figure of equilibrium!of ocean under tidal forces|)}%
+\index{Forces!figure of equilibrium under tidal|)}%
+\PageSep{154}
+parallel to the north of the equator. This parallel
+cuts several of the circles which indicate the
+elevation and depression of the water, and therefore
+he will during his progress pass places where
+the water is shallower and deeper alternately, and
+he would say that the water was rising and falling
+rhythmically. Let us watch his progress
+across the two hemispheres, starting from the
+extreme left. Shortly after coming into view he
+is on the dotted circle of lowest water, and he
+says it is low tide. As he proceeds the water
+rises, slowly at first and more rapidly later, until
+he is in the middle of the hemisphere; he arrives
+there six hours later than when we first began to
+watch him. It will have taken him about $5\frac{1}{2}$~hours
+to pass from low water to high water. At
+low water he was depressed by~$1$ below the original
+level, and at high water he is raised by~$\frac{1}{2}$
+above that level, so that the range from low
+water to high water is represented by~$1\frac{1}{2}$. After
+the passage across the middle of the hemisphere,
+the water level falls, and after about $5\frac{1}{2}$~hours
+more the water is again lowest, and the depression
+is measured by~$1$ on the adopted scale.
+Soon after this he passes out of this hemisphere
+into the other one, and the water rises again
+until he is in the middle of that hemisphere.
+But this time he passes much nearer to the vertex
+of highest water than was the case in the other
+hemisphere, so that the water now rises to a
+\PageSep{155}
+height represented by about~$1\frac{4}{5}$. In this half of
+his daily course the range of tide is from $1$~below
+to $1\frac{4}{5}$~above, and is therefore~$2\frac{4}{5}$, whereas before
+it was only~$1\frac{1}{2}$. The fact that the range of two
+successive tides is not the same is of great importance
+in tidal theory; it is called the diurnal
+inequality of the tide.
+
+It will have been noticed that in the left hemisphere
+the range of fall below the original spherical
+surface is greater than the range of rise
+above it; whereas in the right hemisphere the
+rise is greater than the fall. Mean water mark
+is such that the tide falls on the average as much
+below it as it rises above it, but in this case the
+rise and fall have been measured from the originally
+undisturbed surface. In fact the mean
+level of the water, in the course of a day, is not
+identical with the originally undisturbed surface,
+although the two levels do not differ much from
+one another.
+
+The reader may trace an imaginary observer
+in his daily progress along any other parallel of
+latitude, and will find a similar series of oscillations
+in the ocean; each latitude will, however,
+present its own peculiarities. Then again the
+moon moves in the heavens. In \fig{29} she has
+been supposed to be $15°$~north of the equator,
+but she might have been yet further northward,
+or on the equator, or to the south of it. Her
+extreme range is in fact $28°$~north or south of
+\PageSep{156}
+\index{Forces!those of sun and moon compared|(}%
+\index{Lunar!tide-generating force compared with solar|(}%
+\index{Moon and earth!tide-generating force compared with sun's|(}%
+\index{Solar!tide-generating force compared with lunar|(}%
+the equator. To represent each such case a new
+map would be required, which would, however,
+only differ from this one by the amount of displacement
+of the central spots from the equator.
+
+It is obvious that the two hemispheres in \fig{29}
+are exactly alike, save that they are inverted
+with respect to north and south; the right hemisphere
+is in fact the same as the left upside down.
+It is this inversion which causes the two successive
+tides to be unlike one another, or, in other
+words, gives rise to the diurnal inequality. But
+\index{Diurnal inequality!according to equilibrium theory}%
+there is one case where inversion makes no difference;
+this is when the central spot is on the
+equator in the left hemisphere, for its inversion
+then makes the right hemisphere an exact reproduction
+of the left one. In this case therefore
+the two successive tides are exactly alike, and
+there is no diurnal inequality. Hence the diurnal
+inequality vanishes when the moon is on the
+equator.
+
+Our figure exhibits another important point,
+for it shows that the tide has the greater range
+in that hemisphere where the observer passes
+nearest to one of the two central spots. That is
+to say, the higher tide occurs in that half of the
+daily circuit in which the moon passes nearest to
+the zenith or to the nadir of the observer.
+
+Thus far I have supposed the moon to exist
+alone, but the sun also acts on the ocean according
+\index{Sun!tide-generating force of, compared with that of moon|(}%
+to similar laws, although with less intensity.
+\index{Semidiurnal tide!in equilibrium theory|)}%
+\PageSep{157}
+We must now consider how the relative strengths
+of the actions of the two bodies are to be determined.
+It was indicated in \Ref{Chapter}{V}.\ that
+tide-generating force varies inversely as the cube
+of the distance from the earth of the tide-generating
+body. The force of gravity varies inversely
+as the square of the distance, so that, as
+we change the distance of the attracting body,
+tidal force varies with much greater rapidity than
+does the direct gravitational attraction. Thus if
+the moon stood at half her present distance from
+the earth, her tide-generating force would be $8$~times
+as great, whereas her direct attraction would
+only be multiplied $4$~times. It is also obvious
+that if the moon were twice as heavy as in reality,
+her tide-generating force would be doubled;
+and if she were half as heavy it would be halved.
+Hence we conclude that tide-generating force
+varies directly as the mass of the tide-generating
+body, and inversely as the cube of the distance.
+
+The application of this law enables us to compare
+the sun's tidal force with that of the moon.
+The sun is $25,500,000$~times as heavy as the
+moon, so that, on the score of mass, the solar
+tidal force should be $25\frac{1}{2}$~million times greater
+than that of the moon. But the sun is $389$~times
+as distant as the moon. And since the
+cube of~$389$ is about $59$~millions, the solar tidal
+force should be $59$~million times weaker than
+that of the moon, on the score of distance.
+\PageSep{158}
+
+We have, then, a force which is $25\frac{1}{2}$~million
+times stronger on account of the sun's greater
+weight, and $59$~million times weaker on account
+of his greater distance; it follows that the sun's
+tide-generating force is $25\frac{1}{2}$-$59$ths, or a little
+less than half of that of the moon.
+
+We conclude then that if the sun acted alone
+on the water, the degree of elongation or distortion
+of the ocean, when in equilibrium, would
+be a little less than half of that due to the moon
+alone. When both bodies act together, the distortion
+of the surface due to the sun is superposed
+on that due to the moon, and a terrestrial
+observer perceives only the total or sum of the
+two effects.
+
+When the sun and moon are together on the
+same side of the earth, or when they are diametrically
+opposite, the two distortions conspire
+together, and the total tide will be half as great
+again as that due to the moon alone, because
+the solar tide is added to the lunar tide. And
+when the sun and moon are at right angles to
+\index{Sun!tide-generating force of, compared with that of moon|)}%
+one another, the two distortions are at right
+angles, and the low water of the solar tide conspires
+with the high water of the lunar tide.
+The composite tide has then a range only half as
+great as that due to the moon alone, because the
+solar tide, which has a range of about half that
+\index{Solar!tide-generating force compared with lunar|)}%
+of the lunar tide, is deducted from the lunar
+\index{Lunar!tide-generating force compared with solar|)}%
+tide. Since one and a half is three times a half,
+\index{Forces!those of sun and moon compared|)}%
+\index{Moon and earth!tide-generating force compared with sun's|)}%
+\PageSep{159}
+\index{Spring and neap tides!in equilibrium theory}%
+it follows that when the moon and sun act together
+the range of tide is three times as great
+as when they act adversely. The two bodies
+are together at change of moon and opposite at
+full moon. In both of these positions their
+actions conspire; hence at the change and the
+full of moon the tides are at their largest, and
+are called spring tides. When the two bodies
+are at right angles to one another, it is half
+moon, either waxing or waning, the tides have
+their smallest range, and are called neap tides.
+\index{Neap and spring tides!in equilibrium theory}%
+
+The observed facts agree pretty closely with
+this theory in several respects, for spring tide
+occurs about the full and change of moon, neap
+tide occurs at the half moon, and the range at
+springs is usually about three times as great as
+that at neaps. Moreover, the diurnal inequality
+conforms to the theory in vanishing when the
+moon is on the equator, and rising to a maximum
+when the moon is furthest north or south. The
+amount of the diurnal inequality does not, however,
+agree with theory, and in many places the
+tide which should be the greater is actually the
+less.
+
+The theory which I have sketched is called
+the Equilibrium Theory of the Tides, because
+it supposes that at each moment the ocean is
+in that position of rest or equilibrium which it
+would attain if indefinite time were allowed.
+The general agreement with the real phenomena
+\PageSep{160}
+\index{Equilibrium theory of tides!defects of}%
+proves the theory to have much truth about it,
+but a detailed comparison with actuality shows
+that it is terribly at fault. The lunar and solar
+tidal ellipsoids were found to have their long
+axes pointing straight towards the tide-generating
+bodies, and, therefore, at the time when the
+moon and sun pull together, it ought to be high
+water just when they are due south. In other
+words, at full and change of moon, it ought
+to be high water exactly at noon and at midnight.
+\index{High water!under moon in equilibrium theory}%
+Now observation at spring tides shows
+that at most places this is utterly contradictory
+to fact.
+
+It is a matter of rough observation that the
+tides follow the moon's course, so that high
+water always occurs about the same number of
+hours after the moon is due south. This rule
+has no pretension to accuracy, but it is better
+than no rule at all. Now at change and full of
+the moon, the moon crosses the meridian at the
+same hour of the clock as the sun, for at change
+of moon they are together, and at full moon
+they are twelve hours apart. Hence the hour
+of the clock at which high water occurs at
+change and full of moon is in effect a statement
+of the number of hours which elapse after the
+moon's passage of the meridian up to high
+water. This clock time affords a rough rule for
+the time of high water at any other phase of the
+moon; if, for example, it is high water at eight
+\PageSep{161}
+o'clock at full and change, approximately eight
+hours will always elapse after the moon's passage
+until high water occurs. Mariners call the clock
+time of high water at change and full of moon
+\index{Establishment of port!definition}%
+\index{Establishment of port!zero in equilibrium theory}%
+``the establishment of the port,'' because it
+establishes a rough rule of the tide at all other
+times.
+
+According to the equilibrium theory, high
+water falls at noon and midnight at full and
+change of moon, or in the language of the mariner
+the establishment of all ports should be
+zero. But observation shows that the establishment
+at actual ports has all sorts of values, and
+that in the Pacific Ocean (where the tidal forces
+have free scope) it is at least much nearer to six
+hours than to zero. High water cannot be more
+than six hours before or after noon or midnight
+on the day of full or change of moon, because if
+it occurs more than six hours after one noon, it
+is less than six hours before the following midnight;
+hence the establishment of any port
+cannot possibly be more than six hours before or
+after. Accordingly, the equilibrium theory is
+nearly as much wrong as possible, in respect to
+the time of high water. In fact, in many places
+it is nearly low water at the time that the equilibrium
+theory predicts high water.
+
+It would seem then as if the tidal action of
+the moon was actually to repel the water instead
+of attracting it, and we are driven to ask whether
+\PageSep{162}
+\index{Establishment of port!definition}%
+this result can possibly be consistent with the
+theory of universal gravitation.
+
+The existence of continental barriers across
+the oceans must obviously exercise great influence
+on the tides, but this fact can hardly be
+responsible for a reversal of the previsions of the
+equilibrium theory. It was Newton who showed
+that a depression of the ocean under the moon
+is entirely consistent with the theory of gravitation.
+In the following chapter I shall explain
+Newton's theory, and show how it explains the
+discrepancy which we have found between the
+equilibrium theory and actuality.
+
+\begin{Authorities}
+An exposition of the equilibrium theory will be found in any
+mathematical work on the subject, or in the article \Title{Tides} in the
+``Encyclopædia Britannica.''
+\end{Authorities}
+\index{Equilibrium theory of tides|)}%
+\PageSep{163}
+
+
+\Chapter{IX}
+{Dynamical Theory of the Tide Wave}
+
+\First{The} most serious difficulties in the complete
+\index{Dynamical theory of tide-wave|(}%
+tidal problem do not arise in a certain special
+case which was considered by Newton. His supposition
+was that the sea is confined to a canal
+\index{Canal!critical depth|(}%
+circling the equator, and that the moon and sun
+move exactly in the equator.
+
+An earthquake or any other gigantic impulse
+may be supposed to generate a great wave in this
+equatorial canal. The rate of progress of such
+a wave is dependent on the depth of the canal
+only, according to the laws sketched in \Ref{Chapter}{II}.,
+and the earth's rotation and the moon's attraction
+make no sensible difference in its speed
+of transmission. If, for example, the canal were
+$5$~kilometres ($3$~miles) in depth, such a great
+wave would travel $796$~kilometres ($500$~miles)
+per~hour. If the canal were shallower the speed
+would be less than this; if deeper, greater.
+Now there is one special depth which will be
+found to have a peculiar importance in the theory
+of the tide, namely, where the canal is $13\frac{3}{4}$~miles
+deep. In this case the wave travels $1,042$~miles
+an hour, so that it would complete the
+\PageSep{164}
+$25,000$~miles round the earth in exactly $24$~hours.
+It is important to note that if the depth of the
+equatorial canal be less than $13\frac{3}{4}$~miles, a wave
+takes more than a day to complete the circuit of
+the earth, and if the depth be greater the circuit
+is performed in less than a day.
+
+The great wave, produced by an earthquake or
+other impulse, is called a ``free wave,'' because
+\index{Free wave, explanation and contrast with forced}%
+\index{Waves!forced and free}%
+when once produced it travels free from the action
+of external forces, and would persist forever,
+were it not for the friction to which water is
+necessarily subject. But the leading characteristic
+of the tide wave is that it is generated and
+kept in action by continuous forces, which act
+on the fluid throughout all time. Such a wave
+is called a ``forced wave,'' because it is due to
+\index{Forced wave, explanation and contrast with free wave}%
+the continuous action of external forces. The
+rate at which the tide wave moves is moreover
+dependent only on the rate at which the tidal
+forces travel over the earth, and not in any degree
+on the depth of the canal. It is true that
+the depth of the canal exercises an influence on
+the height of the wave generated by the tidal
+forces, but the wave itself must always complete
+the circuit of the earth in a day, because the
+earth turns round in that period.
+
+We must now contrast the progress of any
+long ``free wave'' in the equatorial canal with
+that of the ``forced'' tide wave. I may premise
+that it will here be slightly more convenient to
+\PageSep{165}
+consider the solar instead of the lunar tide. The
+lunar wave is due to a stronger tide-generating
+force, and since the earth takes $24$~hours $50$~minutes
+to turn round with respect to the moon, that
+is the time which the lunar tide wave takes to
+complete the circuit of the earth; but these differences
+are not material to the present argument.
+The earth turns with respect to the sun
+in exactly one day, or as we may more conveniently
+say, the sun completes the circuit of the
+earth in that time. Therefore the solar tidal
+influence travels over the surface of the earth
+at the rate of $1,042$~miles an hour. Now this is
+exactly the pace at which a ``free wave'' travels
+in a canal of a depth of $13\frac{3}{4}$~miles; accordingly
+\index{Canal!theory of tide wave in|(}%
+in such a canal any long free wave just keeps
+pace with the sun.
+
+We have seen in \Ref{Chapter}{V}.\ that the solar
+tide-generating force \emph{tends} to make a wave crest,
+at those points of the earth's circumference where
+it is noon and midnight. At each moment of
+time the sun is generating a new wave, and after
+it is generated that wave travels onwards as a
+free wave. If therefore the canal has a depth
+\index{Canal!critical depth|)}%
+of $13\frac{3}{4}$~miles, each new wave, generated at each
+moment of time, keeps pace with the sun, and
+the summation of them all must build up two
+enormous wave crests at opposite sides of the
+earth.
+
+If the velocity of a free wave were absolutely
+\PageSep{166}
+the same whatever were its height, the crests of
+the two tide waves would become infinite in
+height. As a fact the rate of progress of a wave
+is somewhat influenced by its height, and therefore,
+when the waves get very big, they will
+cease to keep pace exactly with the sun, and
+the cause for continuous exaggeration of their
+heights will cease to exist. We may, however,
+express this conclusion by saying that, when the
+canal is $13\frac{3}{4}$~miles deep, the height of the tide
+wave becomes mathematically infinite. This does
+not mean that mathematicians assert that the
+wave would really become infinite, but only that
+the simple method of treatment which supposes
+the wave velocity to depend only on the depth
+of water becomes inadequate. If the ocean was
+really confined to an equatorial canal, of this exact
+depth, the tides would be of very great
+height, and the theory would be even more complex
+than it is. It is, however, hardly necessary
+to consider this special case in further detail.
+
+We conclude then that for the depth of $13\frac{3}{4}$~miles,
+the wave becomes infinite in height, in
+the qualified sense of infinity which I have described.
+We may feel sure that the existence
+of the quasi-infinite tide betokens that the behavior
+of the water in a canal shallower than
+$13\frac{3}{4}$~miles differs widely from that in a deeper
+one. It is therefore necessary to examine into
+the essential point in which the two cases differ
+\PageSep{167}
+from one another. In the shallower canal a free
+wave covers less than $25,000$~miles a day, and
+thus any wave generated by the sun would tend
+to be left behind by him. On the other hand,
+in the deeper canal a free wave would outstrip
+the sun, and a wave generated by the sun tends
+to run on in advance of him. But these are
+only tendencies, for in both the shallower and
+the deeper canal the actual tide wave exactly
+keeps pace with the sun.
+
+It would be troublesome to find out what
+would happen if we had the water in the canal
+at rest, and were suddenly to start the sun to
+work at it; and it is fortunately not necessary
+to attempt to do so. It is, however, certain that
+for a long time the motion would be confused,
+but that the friction of the water would finally
+produce order out of chaos, and that ultimately
+there would be a pair of antipodal tide crests
+traveling at the same pace as the sun. Our
+task, then, is to discover what that final state of
+motion may be, without endeavoring to unravel
+the preliminary chaos.
+
+Let us take a concrete case, and suppose our
+canal to be $3$~miles deep, in which we have seen
+that a free wave will travel $500$~miles an hour.
+Suppose, then, we start a long free wave in the
+equatorial canal of $3$~miles deep, with two crests
+\index{Canal!theory of tide wave in|)}%
+$12,500$~miles apart, and therefore antipodal to
+one another. The period of a wave is the time
+\PageSep{168}
+between the passage of two successive crests
+past any fixed point. In this case the crests
+are antipodal to one another, and therefore the
+wave length is $12,500$~miles, and the wave
+travels $500$~miles an hour, so that the period of a
+free wave is $25$~hours. But the tide wave keeps
+pace with the sun, traveling $1,042$~miles an
+hour, and there are two antipodal crests, $12,500$~miles
+apart; hence, the time between the passage
+of successive tide crests is $12$~hours.
+
+In this case a free wave would have a period
+of $25$~hours, and the tide wave, resulting from
+the action of solar tide-generating force, has a
+period of $12$~hours. The contrast then lies between
+the free wave, with a period of $25$~hours,
+and the forced wave, with a period of $12$~hours.
+
+For any other depth of ocean the free wave
+will have another period depending on the depth,
+but the period of the forced wave is always $12$~hours,
+because it depends on the sun. If the
+ocean be shallower than $3$~miles, the free period
+will be greater than $25$~hours, and, if deeper,
+less than $25$~hours. But if the ocean be deepened
+to $13\frac{3}{4}$~miles, the free wave travels at the
+same pace as the forced wave, and therefore the
+two periods are coincident. For depths greater
+than $13\frac{3}{4}$~miles, the period of the free wave is
+less than that of the forced wave; and the
+converse is true for canals less than $13\frac{3}{4}$~miles in
+depth.
+\PageSep{169}
+\index{Forced oscillation!principle of}%
+\index{Free oscillation contrasted with forced}%
+\index{Principle of forced oscillations}%
+
+Now let us generalize this conception; we
+have a system which if disturbed and left to
+itself will oscillate in a certain period, called the
+free period. Periodic disturbing forces act on
+this system and the period of the disturbance
+is independent of the oscillating system itself.
+The period of the disturbing forces is called the
+forced period. How will such a system swing,
+when disturbed with this forced periodicity?
+
+A weight tied to the end of a string affords
+an example of a very simple system capable of
+oscillation, and the period of its free swing depends
+on the length of the string only. I will
+suppose the string to be $3$~feet, $3$~inches, or one
+metre in length, so that the period of the swing
+from right to left, or from left to right is one
+second.\footnote
+ {A pendulum of one metre in length is commonly called a
+ seconds-pendulum, although its complete period is two seconds.}
+If, holding the string, I move my
+hand horizontally to and fro through a small
+distance with a regular periodicity, I set the
+pendulum a-swinging. The period of the movement
+of my hand is the forced period, and the
+free period is two seconds, being the time occupied
+by a metre-long pendulum in moving from
+right to left and back again to right. If I time
+the to and fro motion of my hand so that its
+period from right to left, or from left to right,
+is exactly one second, the excursions of the pendulum
+bob grow greater and greater without limit,
+\PageSep{170}
+\index{Forced oscillation!principle of}%
+\index{Free oscillation contrasted with forced}%
+\index{Principle of forced oscillations}%
+because the successive impulses are stored up in
+the pendulum, which swings further and further
+with each successive impulse. This case is
+exactly analogous with the quasi-infinite tides
+which would arise in a canal $13\frac{3}{4}$~miles deep, and
+here also this case is critical, separating two
+modes of oscillation of the pendulum of different
+characters.
+
+Now when the hand occupies more than one
+second in moving from right to left, the forced
+period is greater than the free period of the
+pendulum; and when the system is swinging
+steadily, it will be observed that the excursion
+of the hand agrees in direction with the excursion
+of the pendulum, so that when the hand is
+furthest to the right so is also the pendulum,
+and vice versa. When the period of the force
+is greater than the free period of the system, at
+the time when the force tends to make the pendulum
+move to the right, it is furthest to the
+right. The excursions of the pendulum agree
+in direction with that of the hand.
+
+Next, when the hand occupies less than one
+second to move from right to left or from left
+to right, the forced period is less than the free
+period, and it will be found that when the hand
+is furthest to the right the pendulum is furthest
+to the left. The excursions of the pendulum
+are opposite in direction from those of the hand.
+These two cases are illustrated by~\fig{30}, which
+\PageSep{171}
+\index{High water!position in shallow and deep canals in dynamical theory}%
+will, perhaps, render my meaning more obvious.
+We may sum up this discussion by saying that
+in the case of a slowly varying disturbing force,
+the oscillation and the force are consentaneous,
+but that with a quickly oscillating force, the
+oscillation is exactly inverted with respect to the
+force.
+
+Now, this simple case illustrates a general
+dynamical principle, namely, that if a system
+\Figure{30}{Forced Oscillations of a Pendulum}{png}
+capable of oscillating with a certain period is
+acted on by a periodic force, when the period of
+the force is greater than the natural free period
+of the system, the oscillations of the system
+agree with the oscillations of the force; but if
+the period of the force is less than the natural
+free period of the system the oscillations are
+inverted with reference to the force.
+
+This principle may be applied to the case of
+the tides in the canal. When the canal is more
+than $13\frac{3}{4}$~miles deep, the period of the sun's disturbing
+force is $12$~hours and is greater than the
+\PageSep{172}
+\index{Newton!theory of tide in equatorial canal}%
+natural free period of the oscillation, because a
+free wave would go more than half round the
+earth in $12$~hours. We conclude, then, that when
+the tide-generating forces are trying to make it
+high water, it will be high water. It has been
+\index{High water!position in shallow and deep canals in dynamical theory}%
+shown that these forces are tending to make high
+water immediately under the sun and at its antipodes,
+and there accordingly will the high water
+be. In this case the tide is said to be direct.
+
+But when the canal is less than $13\frac{3}{4}$~miles
+deep, the sun's disturbing force has, as before, a
+period of $12$~hours, but the period of the free
+wave is more than $12$~hours, because a free wave
+would take more than $12$~hours to get half round
+the earth. Thus the general principle shows
+that where the forces are trying to make high
+water, there will be low water, and vice versa.
+Here, then, there will be low water under the sun
+and at its antipodes, and such a tide is said to
+be inverted, because the oscillation is the exact
+inversion of what would be naturally expected.
+
+All the oceans on the earth are very much
+shallower than fourteen miles, and so, at least
+near the equator, the tides ought to be inverted.
+The conclusion of the equilibrium theory will
+therefore be the exact opposite of the truth, near
+the equator.
+
+This argument as to the solar tide requires
+but little alteration to make it applicable to the
+lunar tide. In fact the only material difference
+\PageSep{173}
+\index{Waves!of tide in equatorial canal}%
+in the conditions is that the period of the lunar
+tide is $12$~hours $25$~minutes, instead of $12$~hours,
+and so the critical depth of an equatorial canal,
+\index{Equatorial canal, tide wave in}%
+which would allow the lunar tide to become
+quasi-infinite, is a little less than that for the
+solar tide. This depth for the lunar tide is in
+fact nearly $13$~miles.\footnote
+ {It is worthy of remark that if the canal had a depth of between
+ $13\frac{3}{4}$ and $13$~miles, the solar tides would be inverted, and
+ the lunar tides would be direct. We should then, at the equator,
+ have spring tide at half moon, when our actual neaps occur;
+ and neap tide at full and change, when our actual springs occur.
+ The tides would also be of enormous height, because the depth
+ is nearly such as to make both tides quasi-infinite. If the depth
+ of the canal were very nearly $13\frac{3}{4}$~miles the solar tide might be
+ greater than the lunar. But these exceptional cases have only
+ a theoretical interest.}
+
+This discussion should have made it clear that
+any tidal theory, worthy of the name, must take
+account of motion, and it explains why the prediction
+of the equilibrium theory is so wide from
+the truth. Notwithstanding, however, this condemnation
+of the equilibrium theory, it is of the
+utmost service in the discussion of the tides,
+because by far the most convenient and complete
+way of specifying the forces which act on the
+ocean at each instant is to determine the figure
+which the ocean would assume, if the forces had
+abundant time to act.
+
+\TB
+
+When the sea is confined to an equatorial
+canal, the tidal problem is much simpler than
+\PageSep{174}
+when the ocean covers the whole planet, and
+this is much simpler than when the sea is interrupted
+by continents. Then again, we have
+thus far supposed the sun and moon to be always
+exactly over the equator, whereas they actually
+range a long way both to the north and to the
+south of the equator; and so here also the true
+problem is more complicated than the one under
+consideration. Let us next consider a case, still
+far simpler than actuality, and suppose that
+whilst the moon or sun still always move over
+the equator, the ocean is confined to several
+canals which run round the globe, following parallels
+of latitude.
+\index{Latitude!tidal wave in canal in high|(}%
+
+The circumference of a canal in latitude~$60°$
+\index{Canal!canal in high latitude|(}%
+is only $12,500$~miles, instead of~$25,000$. If a
+free wave were generated in such a canal with
+two crests at opposite sides of the globe, the distance
+from crest to crest would be $6,250$~miles.
+Now if an equatorial canal and one in latitude~$60°$
+have equal depths, a free wave will travel at
+the same rate along each; and if in each canal
+there be a wave with two antipodal crests, the
+time occupied by the wave in latitude~$60°$ in
+traveling through a space equal to its length will
+be only half of the similar period for the equatorial
+waves. The period of a free wave in latitude~$60°$
+\index{Waves!in canal in high latitude|(}%
+is therefore half what it is at the
+equator, for a pair of canals of equal depths.
+But there is only one sun, and it takes $12$~hours
+\PageSep{175}
+to go half round the planet, and therefore for
+both canals the forced tide wave has a period of
+$12$~hours. If, for example, both canals were
+$8$~miles deep, in the equatorial canal the
+\index{Canal!tides in ocean partitioned into canals}%
+period of the free wave would be greater than
+$12$~hours, whilst in the canal at $60°$~of latitude
+it would be less than $12$~hours. It follows then
+from the general principle as to forced and free
+oscillations, that whilst the tide in the equatorial
+canal would be inverted, that in latitude~$60°$
+would be direct. Therefore, whilst it would be
+low water under the moon at the equator, it
+would be high water under the moon in latitude~$60°$.
+Somewhere, between latitude~$60°$ and the
+equator, there must be a place at which the free
+period in a canal $8$~miles deep is the same as
+the forced period, and in a canal at this latitude
+the tide would be infinite in height, in the modified
+sense explained earlier. It follows therefore
+that there is for any given depth of canal, less
+than $14$~miles, a critical latitude, at which the
+tide tends to become infinite in height.
+
+We conclude, that if the whole planet were
+divided up into canals each partitioned off from
+its neighbor, and if the canals were shallower
+than $14$~miles, we should have inverted tides in
+the equatorial region, and direct tides in the
+polar regions, and, in one of the canals in some
+middle latitude, very great tides the nature of
+which cannot be specified exactly.
+\PageSep{176}
+
+The supposed partitions between neighboring
+canals have introduced a limitation which must
+be removed, if we are to approach actuality, but
+I am unable by general reasoning to do more
+than indicate what will be the effect of the removal
+of the partitions. It is clear that when
+the sea swells up to form the high water, the
+water comes not only from the east and the west
+of the place of high water, but also from the
+north and south. The earth, as it rotates, carries
+with it the ocean; the equatorial water is
+carried over a space of $25,000$~miles in $24$~hours,
+whereas the water in latitude~$60°$ is carried over
+only $12,500$~miles in the same time. When, in
+the northern hemisphere, water moves from north
+to south it passes from a place where the surface
+of the earth is moving slower, to where it is
+moving quicker. Then, as the water goes to the
+south, it carries with it only the velocity adapted
+to the northern latitude, and so it gets left behind
+by the earth. Since the earth spins from
+west to east, a southerly current acquires a westward
+trend. Conversely, when water is carried
+northward of its proper latitude, it leaves the
+\index{Latitude!tidal wave in canal in high|)}%
+earth behind and is carried eastward. Hence
+the water cannot oscillate northward and southward,
+without at the same time oscillating eastward
+and westward. Since in an ocean not
+partitioned into canals, the water must necessarily
+move not only east and west but also north
+\index{Canal!canal in high latitude|)}%
+\index{Waves!in canal in high latitude|)}%
+\PageSep{177}
+\index{Earth and moon!rotation of, effects on tides}%
+\index{Rotation!of earth involved in tidal problem}%
+\index{Vortical motion in oceanic tides}%
+and south, it follows that tidal movements in the
+ocean must result in eddies or vortices. The
+\index{Eddies, tidal oscillation involves}%
+eddying motion of the water must exist everywhere,
+but it would be impossible, without mathematical
+reasoning, to explain how all the eddies
+fit into one another in time and place. It must
+suffice for the present discussion for the reader
+to know that the full mathematical treatment of
+the problem shows this general conclusion to be
+correct.
+
+The very difficult mathematical problem of
+the tides of an ocean covering the globe to a
+uniform depth was first successfully attacked by
+Laplace. He showed that whilst the tides of a
+\index{Laplace!theory of tides|(}%
+shallow ocean are inverted at the equator, as
+proved by Newton, that they are direct towards
+the pole. We have just arrived at the same
+conclusion by considering the tide wave in a
+canal in latitude~$60°$. But our reasoning indicated
+that somewhere in between higher latitudes
+and the equator, the tide would be of an undefined
+character, with an enormous range of rise
+and fall. The complete solution of the problem
+shows, however, that this indication of the
+canal theory is wrong, and that the tidal variation
+of level absolutely vanishes in some latitude
+intermediate between the equator and the pole.
+The conclusion of the mathematician is that
+there is a certain circle of latitude, whose position
+depends on the depth of the sea, where
+there is neither rise nor fall of tide.
+\PageSep{178}
+\index{Vortical motion in oceanic tides}%
+
+At this circle the water flows northward and
+southward, and to and fro between east and
+west, but in such a way as never to raise or depress
+the level of the sea. It is not true to say
+that there is no tide at this circle, for there are
+tidal currents without rise and fall. When the
+ocean was supposed to be cut into canals, we
+thereby obliterated the northerly and southerly
+currents, and it is exactly these currents which
+prevent the tides becoming very great, as we
+were then led to suppose they would be.
+
+It may seem strange that, whereas the first
+rough solution of the problem indicates an oscillation
+of infinite magnitude at a certain parallel
+of latitude, the more accurate treatment of the
+case should show that there is no oscillation of
+level at all. Yet to the mathematician such a
+result is not a cause of surprise. But whether
+strange or not, it should be clear that if at the
+equator it is low water under the moon, and if
+near the pole it is high water under the moon,
+there must in some intermediate latitude be a
+place where the water is neither high nor low,
+that is to say, where there is neither rise nor fall.\footnote
+ {The mathematician knows that a quantity may change sign,
+ either by passing through infinity or through zero. Where a
+ change from positive to negative undoubtedly takes place, and
+ where a passage through infinity can have no physical meaning,
+ the change must take place by passage through zero.}
+
+\TB
+
+Now let us take one more step towards actuality,
+and suppose the earth's equator to be
+\PageSep{179}
+oblique to the orbits of the moon and sun, so
+that they may sometimes stand to the north and
+sometimes to the south of the equator. We
+have seen that in this case the equilibrium theory
+indicates that the two successive tides on any
+one day have unequal ranges. The mathematical
+solution of the problem shows that this conclusion
+is correct. It appears also that if the
+ocean is deeper at the poles than at the equator,
+that tide is the greater which is asserted to be
+so by the equilibrium theory. If, however, the
+ocean is shallower at the poles than at the equator,
+it is found that the high water which the
+equilibrium theory would make the larger is actually
+the smaller and vice versa.
+
+If the ocean is of the same depth everywhere,
+we have a case intermediate between the two,
+where it is shallower at the poles, and where it is
+deeper at the poles. Now in one of these cases
+it appears that the higher high water occurs
+where in the other we find the lower high water
+to occur; and so, when the depth is uniform,
+the higher high water and the lower high water
+must attain the same heights. We thus arrive
+at the remarkable conclusion that, in an ocean
+of uniform depth, the diurnal inequality of the
+\index{Diurnal inequality!in Laplace's solution}%
+tide is evanescent. There are, however, diurnal
+inequalities in the tidal currents, which are so
+adjusted as not to produce a rise or fall. This
+result was first arrived at by the great mathematician
+Laplace.
+\PageSep{180}
+
+According to the equilibrium theory, when the
+moon stands some distance north of the equator,
+the inequality between the successive tides on
+the coasts of Europe should be very great, but
+the difference is actually so small as to escape
+ordinary observation. In the days of Laplace,
+the knowledge of the tides in other parts of the
+world was very imperfect, and it was naturally
+thought that the European tides were fairly
+representative of the whole world. When, then,
+it was discovered that there would be no diurnal
+inequality in an ocean of uniform depth covering
+the whole globe, it was thought that a fair explanation
+had been found for the absence of
+that inequality in Europe. But since the days
+of Laplace much has been learnt about the tides
+\index{Laplace!theory of tides|)}%
+in the Pacific and Indian oceans, and we now
+know that a large diurnal inequality is almost
+\index{Diurnal inequality!in Atlantic, Pacific, and Indian Oceans}%
+universal, so that the tides of the North Atlantic
+are exceptional in their simplicity. In fact, the
+evanescence of the diurnal inequality is not much
+closer to the truth than the large inequality
+predicted by the equilibrium theory; and both
+theories must be abandoned as satisfactory explanations
+of the true condition of affairs. But
+notwithstanding their deficiencies both these
+theories are of importance in teaching us how
+the tides are to be predicted. In the next chapter
+I shall show how a further approximation to
+the truth is attainable.
+\PageSep{181}
+
+\begin{Authorities}
+The canal theory in its elementary form is treated in many
+works on Hydrodynamics, and in \Title{Tides}, ``Encyclopædia Britannica.''
+
+An elaborate treatment of the subject is contained in Airy's
+\index{Airy, Sir G. B.!attack on Laplace}%
+\Title{Tides and Waves}, ``Encyclopædia Metropolitana.'' Airy there
+attacks Laplace for his treatment of the wider tidal problem,
+but his strictures are now universally regarded as unsound.
+
+Laplace's theory is contained in the \Title{Mécanique Céleste}, but it
+is better studied in more recent works.
+
+A full presentment of this theory is contained in Professor
+Horace Lamb's \Title{Hydrodynamics}, Camb.\ Univ.\ Press, 1895, chapter~viii.
+\index{Lamb, H., presentation of Laplace's theory}%
+
+Important papers, extending Laplace's work, by Mr.~S.~S.
+Hough, are contained in the \Title{Philosophical Transactions of the
+\index{Hough, S. S.!dynamical solution of tidal problem}%
+Royal Society}, A.~1897, pp.~201--258, and A.~1898, pp.~139--185.
+\end{Authorities}
+\index{Dynamical theory of tide-wave|)}%
+\PageSep{182}
+
+
+\Chapter{X}
+{Tides in Lakes---Cotidal Chart}
+
+\First{If} the conditions of the tidal problem are to
+\index{Lakes!tides in|(}%
+agree with reality, an ocean must be considered
+which is interrupted by continental barriers of
+land. The case of a sea or lake entirely surrounded
+by land affords the simplest and most
+complete limitation to the continuity of the
+water. I shall therefore begin by considering
+the tides in a lake.
+
+The oscillations of a pendulum under the tidal
+attraction of the moon were considered in \Ref{Chapter}{VI}.,
+and we there saw that the pendulum
+would swing to and fro, although the scale of
+displacement would be too minute for actual
+observation. Now a pendulum always hangs
+perpendicularly to the surface of water, and
+must therefore be regarded as a sort of level.
+As it sways to and fro under the changing action
+of the tidal force, so also must the surface
+of water. If the water in question is a lake, the
+rocking of the level of the lake is a true tide.
+
+A lake of say a hundred miles in length is
+very small compared with the size of the earth,
+and its waters must respond almost instantaneously
+\PageSep{183}
+to the changes in the tidal force. Such
+a lake is not large enough to introduce, to a
+perceptible extent, those complications which
+make the complete theory of oceanic tides so
+difficult. The equilibrium theory is here actually
+true, because the currents due to the changes in
+the tidal force have not many yards to run before
+equilibrium is established, and the lake may
+be regarded as a level which responds almost
+instantaneously to the tidal deflections of gravity.
+The open ocean is a great level also, but sufficient
+time is not allowed it to respond to the changes
+in the direction of gravity, before that direction
+has itself changed.
+
+It was stated in \Ref{Chapter}{V}.\ that the maximum
+horizontal force due to the moon has an intensity
+equal to $\frac{1}{11,664,000}$~part of gravity, and that
+therefore a pendulum $10$~metres long is deflected
+through $\frac{1}{11,664,000}$~of $10$~metres, or through $\frac{1}{1,166}$~of
+a millimetre. Now suppose our lake, $200$~kilometres
+in length, runs east and west, and that
+our pendulum is hung up at the middle of the
+lake, $100$~kilometres from either end. In \fig{31}
+let $\Seg{C}{D}$ represent the level of the lake as
+undisturbed, and $\Seg{A}{B}$~an exaggerated pendulum.
+When the tide-generating force displaces the
+pendulum to~$\Seg{A}{B'}$, the surface of the lake must
+assume the position~$\Seg{C'}{D'}$. Now $\Seg{A}{B}$~being $10$~metres,
+$\Seg{B}{B'}$~may range as far as $\frac{1}{1,166}$~of a millimetre;
+and it is obvious that $\Seg{C}{C'}$~must bear the
+\PageSep{184}
+same relation to~$\Seg{C}{B}$ that $\Seg{B}{B'}$ does to~$\Seg{A}{B}$.
+Hence $\Seg{C}{C'}$ at its greatest may be $\frac{1}{11,664,000}$~of half
+the length of the lake. The lake is supposed
+to be twice $100$~kilometres in length, and $100$~kilometres
+is $10$~million centimetres; thus $\Seg{C}{C'}$~is
+$\frac{1}{1.1664}$~centimetre, or $\frac{9}{10}$~of a centimetre. When
+the pendulum is deflected in the other direction
+the lake rocks the other way, and $C'$~is just as
+much above~$C$ as it was below it before. It
+follows from this that the lunar tide at the ends
+of a lake, $200$~kilometres or $120$~miles in length,
+has a range of $1\frac{3}{4}$~centimetres or $\frac{2}{3}$~of an inch.
+The solar tidal force is a little less than half as
+strong as that due to the moon, and when the
+two forces conspire together at the times of
+spring tide, we should find a tide with a range
+of $2\frac{1}{2}$~centimetres.
+\Figure{31}{The Tide in a Lake}{png}
+
+If the same rule were to apply to a lake $2,000$~kilometres
+or $1,200$~miles in length, the range
+of lunar tide would be about $17$~centimetres or
+$7$~inches, and the addition of solar tides would
+bring the range up to $25$~centimetres or $10$~inches.
+\PageSep{185}
+I dare say that, for a lake of such a
+size, this rule would not be very largely in error.
+But as we make the lake longer, the currents
+set up by the tidal forces have not sufficient time
+to produce their full effects before the intensity
+and direction of the tidal forces change. Besides
+this, if the lake were broad from north to south,
+the earth's rotation would have an appreciable
+effect, so that the water which flows from the
+north to the south would be deflected westward,
+and that which flows from south to north would
+tend to flow eastward. The curvature of the
+earth's surface must also begin to affect the
+motion. For these reasons, such a simple rule
+would then no longer suffice for calculating the
+tide.
+
+Mathematicians have not yet succeeded in
+solving the tidal problem for a lake of large
+dimensions, and so it is impossible to describe
+the mode of oscillation. It may, however, be asserted
+that the shape, dimensions, and depth of
+the lake, and the latitudes of its boundaries will
+affect the result. The tides on the northern and
+southern shores will be different, and there will
+be nodal lines, along which there will be no rise
+and fall of the water.
+\index{Lakes!tides in|)}%
+
+The Straits of Gibraltar are so narrow, that
+\index{Mediterranean Sea, tides in}%
+the amount of water which can flow through
+them in the six hours which elapse between
+high and low water in the Atlantic is inconsiderable.
+\PageSep{186}
+\index{Waves!propagated northward in Atlantic|(}%
+Hence the Mediterranean Sea is virtually
+\index{Mediterranean Sea, tides in}%
+a closed lake. The tides of this sea are
+much complicated by the constriction formed
+by the Sicilian and Tunisian promontories. Its
+tides probably more nearly resemble those of two
+lakes than of a single sheet of water. The tides
+of the Mediterranean are, in most places, so inconspicuous
+that it is usually, but incorrectly,
+described as a tideless sea. Every visitor to
+Venice must, however, have seen, or may we say
+smelt, the tides, which at springs have a range of
+some four feet. The considerable range of tide
+at Venice appears to indicate that the Adriatic
+\index{Adriatic, tide in}%
+acts as a resonator for the tidal oscillation, in the
+same way that a hollow vessel, tuned to a particular
+note, picks out and resonates loudly when
+that note is sounded.
+
+We see, then, that whilst the tides of a small
+lake are calculable by the equilibrium theory,
+those of a large one, such as the Mediterranean,
+remain intractable. It is clear, then, that the
+tides of the ocean must present a problem yet
+more complex than those of a large lake.
+
+In the Pacific and Southern oceans the tidal
+\index{Pacific Ocean, tide in, affects Atlantic}%
+forces have almost uninterrupted sway, but the promontories
+of Africa and of South America must
+profoundly affect the progress of the tide wave
+from east to west. The Atlantic Ocean forms a
+\index{Atlantic, tide in|(}%
+great bay in this vaster tract of water. If this
+inlet were closed by a barrier from the Cape of
+\PageSep{187}
+\index{Pacific Ocean, tide in, affects Atlantic}%
+Good Hope to Cape Horn, it would form a lake
+large enough for the generation of much larger
+tides than those of the Mediterranean Sea, although
+probably much smaller than those which
+we actually observe on our coasts. Let us now
+suppose that the tides proper to the Atlantic are
+non-existent, and let us remove the barrier between
+the two capes. Then the great tide wave
+sweeps across the Southern ocean from east to
+west, and, on reaching the tract between Africa
+and South America, generates a wave which
+travels northward up the Atlantic inlet. This
+secondary wave travels ``freely,'' at a rate dependent
+only on the depth of the ocean. The
+energy of the wave motion is concentrated, where
+the channel narrows between North Africa and
+Brazil, and the height of the wave must be augmented
+in that region. Then the energy is
+weakened by spreading, where the sea broadens
+again, and it is again reconcentrated by the projection
+of the North American coast line towards
+Europe. Hence, even in this case, ideally simplified
+as it is by the omission of the direct action
+of the moon and sun, the range of tide would
+differ at every portion of the coasts on each side
+of the Atlantic.
+
+The time of high water at any place must also
+depend on the varying depth of the ocean, for it
+is governed by the time occupied by the ``free
+wave'' in traveling from the southern region to
+\PageSep{188}
+\index{Cotidal chart}%
+the north. But in the south, between the two
+capes of Africa and South America, the tidal
+oscillation is constrained to keep regular time
+with the moon, and so it will keep the same
+rhythm at every place to the northward, at whatever
+variable pace the wave may move. The
+time of high water will of course differ at every
+point, being later as we go northward. The
+wave may indeed occupy so long on its journey,
+that one high water may have only just arrived
+at the northern coast of Africa, when another is
+rounding the Cape of Good Hope.
+
+Under the true conditions of the case, this
+``free'' wave, generated in and propagated from
+the southern ocean, is fused with the true
+``forced'' tide wave generated in the Atlantic itself.
+\index{Atlantic, tide in|)}%
+It may be conjectured that on the coast of
+Europe the latter is of less importance than the
+\index{Europe, tides on coasts of}%
+former. It is interesting to reflect that our tides
+to-day depend even more on what occurred yesterday
+or the day before in the Southern Pacific
+and Indian oceans, than on the direct action of
+the moon to-day. But the relative importance
+of the two causes must remain a matter of conjecture,
+for the problem is one of insoluble complexity.
+
+Some sixty years ago Whewell, and after him
+\index{Whewell!on cotidal charts}%
+Airy, drew charts illustrative of what has just
+\index{Airy, Sir G. B.!cotidal chart}%
+been described. A map showing the march
+of the tide wave is reproduced from Airy's
+\index{Waves!propagated northward in Atlantic|)}%
+\PageSep{189}
+\index{Establishment of port!shown in cotidal chart}%
+``Tides and Waves,'' in~\fig{32}. It claims to
+show, from the observed times of high water at
+the various parts of the earth, how the tide wave
+travels over the oceans. Whewell and Airy were
+\index{Whewell!on cotidal charts}%
+well aware that their map could only be regarded
+as the roughest approximation to reality. Much
+has been learnt since their days, and the then
+incomplete state of knowledge hardly permitted
+them to fully realize how very rough was their
+approximation to the truth. No more recent attempt
+has been made to construct such a map,
+and we must rest satisfied with this one. Even
+if its lines may in places depart pretty widely
+from the truth, it presents features of much interest.
+I do not reproduce the Pacific Ocean,
+because it is left almost blank, from deficiency
+of data. Thus, in that part of the world where
+the tides are most normal, and where the knowledge
+of them would possess the greatest scientific
+interest, we are compelled to admit an almost
+total ignorance.
+
+The lines on the map, \fig{32}, give the Greenwich
+times of high water at full and change of
+moon. They thus purport to represent the successive
+positions of the crest of the tide wave.
+For example, at noon and midnight (XII~o'clock),
+at full and change of moon, the crest
+of the tide wave runs from North Australia to
+Sumatra, thence to Ceylon, whence it bends back
+to the Island of Bourbon, and, passing some hundreds
+\PageSep{190}
+\TallFig[0.875]{32}{Chart of Cotidal Lines}{jpg}
+\PageSep{191}
+\index{Cotidal chart!for diurnal tide hitherto undetermined}%
+of miles south of the Cape of Good Hope,
+trends away towards the Antarctic Ocean. At
+the same moment the previous tide crest has
+traveled up the Atlantic, and is found running
+across from Newfoundland to the Canary
+Islands. A yet earlier crest has reached the
+north of Norway. At this moment it is low
+water from Brazil to the Gold Coast, and again
+at Great Britain.
+
+The successive lines then exhibit the progress
+of the wave from hour to hour, and we see how
+the wave is propagated into the Atlantic. The
+crowding together of lines in places is the graphical
+representation of the retardation of the
+wave, as it runs into shallower water.
+
+But even if this chart were perfectly trustworthy,
+it would only tell us of the progress of
+the ordinary semidiurnal wave, which produces
+high water twice a day. We have, however, seen
+reason to believe that two successive tides should
+not rise to equal heights, and this figure does
+not even profess to give any suggestion as to
+how this inequality is propagated. In other
+words, it is impossible to say whether two successive
+tides of unequal heights tend to become
+more or less unequal, as they run into any of
+the great oceanic inlets. Thus the map affords
+no indication of the law of the propagation of
+the diurnal inequality.
+\index{Diurnal inequality!not shown in cotidal chart}%
+
+This sketch of the difficulties in the solution
+\PageSep{192}
+\index{Cotidal chart!for diurnal tide hitherto undetermined}%
+of the full tidal problem might well lead to despair
+of the possibility of tidal prediction on our
+coasts. I shall, however, show in the next chapter
+how such prediction is possible.
+
+\begin{Authorities}
+For cotidal charts see Whewell, \Title{Phil.\ Trans.\ Roy.\ Soc.}\ 1833,
+or Airy's \Title{Tides and Waves}, ``Encyclopædia Metropolitana.''
+\index{Airy, Sir G. B.!\Title{Tides and Waves}}%
+\end{Authorities}
+\PageSep{193}
+
+
+\Chapter{XI}
+{Harmonic Analysis of the Tide}
+
+\First{It} is not probable that it will ever be possible
+\index{Analysis, harmonic, of tide|(}%
+\index{Harmonic analysis!account of|(}%
+\index{Moon and earth!tide due to ideal, moving in equator}%
+to determine the nature of the oceanic oscillation
+as a whole with any accuracy. It is true that
+we have already some knowledge of the general
+march of the tide wave, and we shall doubtless
+learn more in the future, but this can never suffice
+for accurate prediction of the tide at any
+place.
+
+Although the equilibrium theory is totally
+false as regards its prediction of the time of passage
+and of the height of the tide wave, yet it
+furnishes the stepping-stone leading towards the
+truth, because it is in effect a compendious statement
+of the infinite variety of the tidal force in
+time and place.
+
+I will begin my explanation of the practical
+method of tidal prediction by obliterating the
+sun, and by supposing that the moon revolves in
+an equatorial circle round the earth. In this
+case the equilibrium theory indicates that each
+tide exactly resembles its predecessors and its
+successors for all time, and that the successive
+and simultaneous passages of the moon and of
+\PageSep{194}
+\index{Moon and earth!tide due to ideal, moving in equator}%
+the wave crests across any place follow one
+another at intervals of $12$~hours $25$~minutes. It
+would always be exactly high water under or
+opposite to the moon, and the height of high
+water would be exactly determinate. In actual
+oceans, even although only subject to the action
+of such a single satellite, the motion of the water
+would be so complex that it would be impossible
+to predict the exact height or time of high or
+of low water. But since the tidal forces operate
+in a stereotyped fashion day after day, there will
+be none of that variability which actually occurs
+on the real earth under the actions of the real
+sun and moon, and we may positively assert that
+whatever the water does to-day it will do to-morrow.
+Thus, if at a given place it is high water
+at a definite number of hours after the equatorial
+moon has crossed the meridian to-day, it will be
+so to-morrow at the same number of hours after
+the moon's passage, and the water will rise and
+fall every day to the same height above and below
+the mean sea level. If then we wanted to
+know how the tide would rise and fall in a given
+harbor, we need only watch the motion of the
+sea at that place, for however the water may
+move elsewhere its motion will always produce
+the same result at the port of observation.
+Thus, apart from the effects of wind, we should
+only have to note the tide on any one day
+to be able to predict it for all time. For by a
+\PageSep{195}
+\index{Satellites!tide due to single equatorial}%
+single day of observation it would be easy to
+note how many hours after the moon's passage
+high water occurs, and how many feet it rises
+and falls with reference to some fixed mark on
+the shore. The delay after the moon's passage
+and the amount of rise and fall would differ geographically,
+but at each place there would be two
+definite numbers giving the height of the tide
+and the interval after the moon's passage until
+high water. These two numbers are called the
+tidal constants for the port; they would virtually
+\index{Constants, tidal, explained}%
+contain tidal predictions for all time.
+
+Now if the moon were obliterated, leaving the
+sun alone, and if he also always moved over
+the equator, a similar rule would hold good,
+but exactly $12$~hours would elapse from one
+high water to the next, instead of $12$~hours $25$~minutes
+as in the case of the moon's isolated
+action. Thus two other tidal constants, expressive
+of height and interval, would virtually contain
+tidal prediction for the solar tide for all
+time.
+
+Theory here gives us some power of foreseeing
+the relative importance of the purely lunar
+and of the purely solar tide. The two waves
+due to the sun alone or to the moon alone have
+the same character, but the solar waves follow
+one another a little quicker than the lunar waves,
+and the sun's force is a little less than half the
+moon's force. The close similarity between the
+\PageSep{196}
+\index{Satellites!tide due to single equatorial}%
+actions of the sun and moon makes it safe to conclude
+that the delay of the isolated solar wave
+after the passage of the sun would not differ
+much from the delay of the isolated lunar wave
+after the passage of the moon, and that the
+height of the solar wave would be about half of
+that of the lunar wave. But theory can only be
+trusted far enough to predict a rough proportionality
+of the heights of the two tide waves to their
+respective generating forces, and the approximate
+equality of the intervals of retardation; but the
+height and retardation of the solar wave could
+not be accurately foretold from observation of
+the lunar wave.
+
+When the sun and moon coëxist, but still
+move in equatorial circles, the two waves, which
+we have considered separately, are combined.
+The four tidal constants, two for the moon and
+two for the sun, would contain the prediction of
+the height of water for all time, for it is easy at
+any future moment of time to discover the two
+intervals of time since the moon and since the
+sun have crossed the meridian of the place of
+observation; we should then calculate the height
+of the water above some mark on the shore on
+the supposition that the moon exists alone, and,
+again, on the supposition that the sun exists
+alone, and adding the two results together, should
+obtain the required height of the water at the
+moment in question.
+\PageSep{197}
+
+But the real moon and sun do not move in
+equatorial circles, but in planes which are oblique
+to the earth's equator, and they are therefore
+sometimes to the north and sometimes to the
+south of the equator; they are also sometimes
+nearer and sometimes further from the earth on
+account of the eccentricity of the orbits in which
+they move. Now the mathematician treats this
+complication in the following way: he first considers
+the moon alone and replaces it by a number
+of satellites of various masses, which move
+in various planes. It is a matter of indifference
+that such a system of satellites could not maintain
+the orbits assigned to them if they were allowed
+to go free, but a mysterious being may be
+postulated who compels the satellites to move in
+the assigned orbits. One, and this is the largest
+of these ideal satellites, has nearly the same mass
+as the real moon and moves in a circle over the
+equator; it is in fact the simple isolated moon
+whose action I first considered. Another small
+satellite stands still amongst the stars; others
+move in such orbits that they are always vertically
+overhead in latitude~$45°$; others repel instead
+of attracting; and others move backwards
+amongst the stars. Now all these satellites are
+so arranged as to their masses and their orbits,
+that the sum of their tidal forces is exactly the
+same as those due to the real moon moving in
+her actual orbit.
+\PageSep{198}
+\index{Interval from moon's transit to high water!in case of ideal satellite}%
+
+So far the problem seems to be complicated
+rather than simplified, for we have to consider a
+dozen moons instead of one. The simplification,
+however, arises from the fact that each satellite
+either moves uniformly in an orbit parallel to the
+equator, or else stands still amongst the stars.
+It follows that each of the ideal satellites creates
+a tide in the ocean which is of a simple character,
+and repeats itself day after day in the same
+way as the tide due to an isolated equatorial
+moon. If all but one of these ideal satellites
+were obliterated the observation of the tide for
+a single day would enable us to predict the tide
+for all time; because it would only be necessary
+to note the time of high water after the ideal
+satellite had crossed the meridian, and the height
+\index{Height of tide!due to ideal satellite}%
+of the high water, and then these two data would
+virtually contain a tidal prediction for that tide
+at the place of observation for all future time.
+The interval and height are together a pair of
+``tidal constants'' for the particular satellite in
+question, and refer only to the particular place
+at which the observation is made.
+
+In actuality all the ideal satellites coëxist, and
+the determination of the pair of tidal constants
+appropriate to any one of them has to be made
+by a complex method of analysis, of which I shall
+say more hereafter. For the present it will suffice
+to know that if we could at will annul all
+the ideal satellites except one, and observe its
+\PageSep{199}
+\index{Moon and earth!ideal satellites replacing actual}%
+\index{Partial tides in harmonic method}%
+tide even for a single day, its pair of constants
+could be easily determined. It would then only
+be necessary to choose in succession all the satellites
+\index{Satellites!ideal replacing sun and moon in harmonic analysis}%
+as subjects of observation, and the materials
+for a lunar tide table for all time would be obtained.
+
+The motion of the sun round the earth is analogous
+to that of the moon, and so the sun has
+also to be replaced by a similar series of ideal
+suns, and the partial tide due to each of them
+has to be found. Finally at any harbor some
+twenty pairs of numbers, corresponding to twenty
+ideal moons and suns, give the materials for tidal
+prediction for all time. Theoretically an infinite
+number of ideal bodies is necessary for an absolutely
+perfect representation of the tides, but
+after we have taken some twenty of them, the
+remainder are found to be excessively small in
+mass, and therefore the tides raised by them are
+so minute that they may be safely omitted. This
+method of separating the tide wave into a number
+of partial constituents is called ``harmonic
+analysis.'' It was first suggested, and put into
+practice as a practical treatment of the tidal
+problem, by Sir William Thomson, now Lord
+Kelvin, and it is in extensive use.
+\index{Kelvin, Lord!initiates harmonic analysis}%
+
+In this method the aggregate tide wave is considered
+as the sum of a number of simple waves
+following one another at exactly equal intervals
+of time, and always presenting a constant rise
+\PageSep{200}
+\index{Moon and earth!ideal satellites replacing actual}%
+\index{Prediction of tide!due to ideal satellite}%
+and fall at the place of observation. When the
+time of high water and the height of any one of
+these constituent waves is known on any one
+day, we can predict, with certainty, the height
+of the water, as due to it alone, at any future
+time however distant. The period of time which
+elapses between the passage of one crest and of
+the next is absolutely exact, for it is derived from
+a study of the motions of the moon or sun, and
+is determined to within a thousandth of a second.
+The instant at which any one of the satellites
+\index{Satellites!ideal replacing sun and moon in harmonic analysis}%
+passes the meridian of the place is also
+known with absolute accuracy, but the interval
+after the passage of the satellite up to the high
+water of any one of these constituent waves, and
+the height to which the water will rise are only
+derivable from observation at each port.
+
+Since there are about twenty coëxistent waves
+of sensible magnitude, a long series of observations
+is requisite for disentangling any particular
+wave from among the rest. The series must
+also be so long that the disturbing influence of
+the wind, both on height and time, may be eliminated
+by the taking of averages. It may be
+well to reiterate that each harbor has to be considered
+by itself, and that a separate set of tidal
+constants has to be found for each place. If it
+is only required to predict the tides with moderate
+accuracy some eight partial waves suffice, but
+if high accuracy is to be attained, we have to
+\PageSep{201}
+consider a number of the smaller ones, bringing
+the total up to $20$ or~$25$.
+
+When the observed tidal motions of the sea
+have been analyzed into partial tide waves, they
+are found to fall naturally into three groups,
+which correspond with the dissections of the sun
+\index{Sun!ideal, replacing real sun in harmonic analysis}%
+and moon into the ideal satellites. In the first
+and most important group the crests follow one
+another at intervals of somewhere about $12$~hours;
+these are called the semidiurnal tides.
+In the second group, the waves of which are in
+most places of somewhat less height than those
+of the semidiurnal group, the crests follow one
+another at intervals of somewhere about $24$~hours,
+and they are called diurnal. The tides
+of the third group have a very slow periodicity,
+for their periods are a fortnight, a month, half
+a year, and a year; they are commonly of very
+small height, and have scarcely any practical
+importance; I shall therefore make no further
+reference to them.
+
+Let us now consider the semidiurnal group.
+The most important of these is called ``the principal
+lunar semidiurnal tide.'' It is the tide
+\index{Lunar!tide, principal}%
+\index{Semidiurnal tide!in harmonic method|(}%
+raised by an ideal satellite, which moves in a circle
+round the earth's equator. I began my explanation
+of this method by a somewhat detailed
+consideration of this wave. In this case, the
+wave crests follow one another at intervals of
+$12$~hours $25$~minutes $14\frac{1}{6}$~seconds. The average
+\PageSep{202}
+interval of time between the successive visible
+transits of the moon over the meridian of the
+place of observation is $24$~hours $50$~minutes $28\frac{1}{3}$~seconds;
+and as the invisible transit corresponds
+to a tide as well as the visible one, the interval
+between the successive high waters is the time
+between the successive transits, of which only
+each alternate one is visible.
+
+The tide next in importance is ``the principal
+solar semidiurnal tide.'' This tide bears the
+\index{Solar!principal tide}%
+same relationship to the real sun that the principal
+lunar semidiurnal tide bears to the real moon.
+The crests follow one another at intervals of
+exactly $12$~hours, which is the time from noon to
+midnight and of midnight to noon. The height
+of this partial wave is, at most places, a little less
+than half of that of the principal lunar tide.
+
+The interval between successive lunar tides is
+$25\frac{1}{4}$~minutes longer than that between successive
+solar tides, and as there are two tides a day, the
+lunar tide falls behind the solar tide by $50\frac{1}{2}$~minutes
+a day. If we imagine the two tides to start
+together with simultaneous high waters, then in
+about $7$~days the lunar tide will have fallen about
+$6$~hours behind the solar tide, because $7$~times
+$50\frac{1}{2}$~minutes is $5$~hours $54$~minutes. The period
+from high water to low water of the principal
+solar semidiurnal tide is $6$~hours, being half the
+time between successive high waters. Accordingly,
+when the lunar tide has fallen $6$~hours
+\PageSep{203}
+\index{Spring and neap tides!represented by principal lunar and solar tides}%
+behind the solar tide, the low water of the solar
+tide falls in with the high water of the lunar
+tide. It may facilitate the comprehension of
+this matter to take a numerical example; suppose
+then that the lunar tide rises $4$~feet above
+and falls by the same amount below the mean
+level of the sea, and that the solar tide rises and
+falls $2$~feet above and below the same level;
+then if the two partial waves be started with their
+high waters simultaneous, the joint wave will at
+first rise and fall by $6$~feet. But after $7$~days it
+is low solar tide when it is high lunar tide, and
+so the solar tide is subtracted from the lunar
+tide, and the compound wave has a height of
+$4$~feet less $2$~feet, that is to say, of $2$~feet.
+After nearly another $7$~days, or more exactly
+after $14\frac{1}{2}$~days from the start, the lunar tide has
+lost another $6$~hours, so that it has fallen back
+$12$~hours in all, and the two high waters agree
+together again, and the joint wave has again a
+rise and fall of $6$~feet. When the two high
+waters conspire it is called spring tide, and when
+the low water of the solar tide conspires with the
+high water of the lunar tide, it is called neap
+tide. It thus appears that the principal lunar
+and principal solar semidiurnal tides together
+represent the most prominent feature of the tidal
+oscillation.
+
+The next in importance of the semidiurnal
+waves is called the ``lunar elliptic tide,'' and here
+\PageSep{204}
+\index{Neap and spring tides!represented by principal lunar and solar tides}%
+the crests follow one another at intervals of $12$~hours
+$39$~minutes $30$~seconds. Now the interval
+between the successive principal lunar tides was
+\index{Lunar!elliptic tide}%
+$12$~hours $25$~minutes $14$~seconds; hence, this
+new tide falls behind the principal lunar tide by
+$14\frac{1}{2}$~minutes in each half day. If this tide starts
+so that its high water agrees with that of the
+principal lunar tide, then after $13\frac{3}{4}$~days from
+the start, its hollow falls in with the crest of the
+former, and in $27\frac{1}{2}$~days from the start the two
+crests agree again.
+
+The moon moves round the earth in an ellipse,
+and if to-day it is nearest to the earth, in $13\frac{3}{4}$~days
+it will be furthest, and in $27\frac{1}{2}$~days it will
+be nearest again. The moon must clearly exercise
+a stronger tidal force and create higher
+tides when she is near than when she is far;
+hence every $27\frac{1}{2}$~days the tides must be larger,
+and halfway between they must be smaller.
+But the tide under consideration conspires with
+the principal lunar tide every $27\frac{1}{2}$~days, and,
+accordingly, the joint wave is larger every $27\frac{1}{2}$~days
+and smaller in between. Thus this lunar
+elliptic tide represents the principal effect of the
+\index{Elliptic tide, lunar}%
+elliptic motion of the moon round the earth.
+There are other semidiurnal waves besides the
+three which I have mentioned, but it would
+hardly be in place to consider them further
+here.
+\index{Semidiurnal tide!in harmonic method|)}%
+
+Now turning to the waves of the second kind,
+\PageSep{205}
+\index{Diurnal inequality!in harmonic method}%
+which are diurnal in character, we find three, all
+of great importance. In one of them the high
+waters succeed one another at intervals of $25$~hours
+$49$~minutes $9\frac{1}{2}$~seconds, and of the second
+and third, one has a period of about $4$~minutes
+less than $24$~hours and the other of about $4$~minutes
+greater than the $24$~hours. It would
+hardly be possible to show by general reasoning
+how these three waves arise from the attraction
+of three ideal satellites, and how these satellites
+together are a substitute for the actions of the
+true moon and sun. It must, however, be obvious
+that the oscillation resulting from three coëxistent
+waves will be very complicated.
+
+All the semidiurnal tides result from waves of
+essentially similar character, although some follow
+one another a little more rapidly than others,
+and some are higher and some are lower. An
+accurate cotidal map, illustrating the progress of
+any one of these semidiurnal waves over the
+ocean, would certainly tell all that we care to
+know about the progress of all the other waves
+of the group.
+
+Again, all the diurnal tides arise from waves
+of the same character, but they are quite diverse
+in origin from the semidiurnal waves, and have
+only one high water a day instead of two. A
+complete knowledge of the behavior of semidiurnal
+waves would afford but little insight into
+the behavior of the diurnal waves. At some
+\PageSep{206}
+time in the future the endeavor ought to be
+made to draw a diurnal cotidal chart distinct
+from the semidiurnal one, but our knowledge is
+not yet sufficiently advanced to make the construction
+of such a chart feasible.
+
+\TB
+
+All the waves of which I have spoken thus
+far are generated by the attractions of the sun
+and moon and are therefore called astronomical
+tides, but the sea level is also affected by other
+oscillations arising from other causes.
+
+Most of the places, at which a knowledge of
+the tides is practically important, are situated in
+estuaries and in rivers. Now rain is more prevalent
+\index{Rivers!annual meteorological tide in}%
+at one season than at another, and mountain
+snow melts in summer; hence rivers and
+estuaries are subject to seasonal variability of
+level. In many estuaries this kind of inequality
+may amount to one or two feet, and such a considerable
+change cannot be disregarded in tidal
+prediction. It is represented by inequalities with
+periods of a year and of half a year, which are
+called the annual and semiannual meteorological
+\index{Annual and semi-annual tides}%
+\index{Meteorological!tides}%
+tides.
+
+Then again, at many places, especially in the
+Tropics, there is a regular alternation of day and
+night breezes, the effect of which is to heap up
+% [** TN: "inland", "off-shore" on line breaks in the original; sole instances]
+the water in-shore as long as the wind blows in-land,
+\index{Wind!a cause of meteorological tides}%
+and to lower it when the wind blows off-shore.
+Hence there results a diurnal inequality
+\PageSep{207}
+\index{Estuary, annual meteorological tide in}%
+of sea-level, which is taken into account in tidal
+prediction by means of a ``solar diurnal meteorological
+\index{Meteorological!tides}%
+tide.'' Although these inequalities depend
+entirely on meteorological influences and
+have no astronomical counterpart, yet it is necessary
+to take them into account in tidal prediction.
+
+\TB
+
+But besides their direct astronomical action,
+the sun and moon exercise an influence on the
+sea in a way of which I have not yet spoken.
+We have seen how waves gradually change their
+shape as they progress in a shallow river, so that
+the crests become sharper and the hollows flatter,
+while the advancing slope becomes steeper and
+the receding one less steep. An extreme exaggeration
+of this sort of change of shape was
+found in the bore. Now it is an absolute rule,
+in the harmonic analysis of the tide, that the
+partial waves shall be of the simplest character,
+and shall have a certain standard law of slope
+on each side of their crests. If then any wave
+ceases to present this standard simple form, it is
+necessary to conceive of it as compound, and to
+build it up out of several simple waves. By the
+composition of a simple wave with other simple
+waves of a half, a third, a quarter of the wave
+length, a resultant wave can be built up which
+shall assume any desired form. For a given
+compound wave, there is no alternative of choice,
+\PageSep{208}
+for it can only be built up in one way. The
+analogy with musical notes is here complete, for
+a musical note of any quality is built up from
+a fundamental, together with its octave and
+twelfth, which are called overtones. So also the
+distorted tide wave in a river is regarded as consisting
+of simple fundamental tide, with over-tides
+of half and third length. The periods of
+these over-tides are also one half and one third
+of that of the fundamental wave.
+
+Out in the open ocean, the principal lunar
+semidiurnal tide is a simple wave, but when it
+runs into shallow water at the coast line, and
+still more so in an estuary, it changes its shape.
+\index{Estuary, annual meteorological tide in}%
+The low water lasts longer than the high water,
+and the time which elapses from low water to
+high water is usually shorter than that from
+high water to low water. The wave is in fact
+no longer simple, and this is taken into account
+by considering it to consist of a fundamental
+lunar semidiurnal wave with a period of $12$~hours
+$50$~minutes, of the first over-tide or octave
+with a period of $6$~hours $25$~minutes, of the second
+over-tide or twelfth with a period of $4$~hours
+$17$~minutes, and of the third over-tide or
+double octave with a period of $3$~hours $13$~minutes.
+In estuaries, the first over-tide of the
+lunar semidiurnal tide is often of great importance,
+and even the second is considerable; the
+third is usually very small, and the fourth and
+\PageSep{209}
+higher over-tides are imperceptible. In the same
+way over-tides must be introduced, to represent
+the change of form of the principal solar semidiurnal
+tide. But it is not usually found necessary
+to consider them in the cases of the less
+important partial tides. The octave, the twelfth,
+and the upper octave may be legitimately described
+as tides, because they are due to the
+attractions of the moon and of the sun, although
+they arise indirectly through the distorting influence
+of the shallowness of the water.
+
+\TB
+
+I have said above that about twenty different
+simple waves afford a good representation of the
+tides at any port. Out of these twenty waves,
+some represent the seasonal change of level in
+the water due to unequal rainfall and evaporation
+at different times of the year, and others
+represent the change of shape of the wave due
+to shallowing of the water. Deducting these
+quasi-tides, we are left with about twelve to
+represent the true astronomical tide. It is not
+possible to give an exact estimate of the number
+of partial tides necessary to insure a good representation
+of the aggregate tide wave, because
+the characteristics of the motion are so different
+at various places that partial waves, important
+at one place, are insignificant at others. For
+example, at an oceanic island the tides may be
+more accurately represented as the sum of a
+\PageSep{210}
+dozen simple waves than by two dozen in a tidal
+river.
+
+The method of analyzing a tide into its constituent
+parts, of which I have now given an
+account, is not the only method by which the
+tides may be treated, but as it is the most recent
+and the best way, I shall not consider the older
+methods in detail. The nature of the procedure
+adopted formerly will, however, be indicated in
+\Ref{Chapter}{XIII}.
+
+\begin{Authorities}[Authority]
+G.~H. Darwin, \Title{Harmonic Analysis of Tidal Observations}:
+\index{Analysis, harmonic, of tide|)}%
+\index{Darwin, G. H.!harmonic analysis}%
+\index{Harmonic analysis!account of|)}%
+``Report to British Association.'' Southport,~1883.
+
+An outline of the method is also contained in \Title{Tides}, ``Encyclopædia
+Britannica.''
+\end{Authorities}
+\PageSep{211}
+
+
+\Chapter{XII}
+{Reduction of Tidal Observations}
+
+\First{I have} now to explain the process by which
+\index{Reduction of tidal observations|(}%
+the several partial tides may be disentangled
+from one another.
+
+The tide gauge furnishes a complete tidal record,
+so that measurement of the tide curve gives
+the height of the water at every instant of time
+during the whole period of observation. The
+\index{Observation!reduction of tidal|(}%
+record may be supposed to begin at noon of a
+given day, say of the first of January. The
+longitude of the port of observation is of course
+known, and the Nautical Almanack gives the
+positions of the sun and moon on the day and
+at the hour in question, with perfect accuracy.
+The real moon has now to be replaced by a
+series of ideal satellites, and the rules for the
+substitution are absolutely precise. Accordingly,
+the position in the heavens of each of
+the ideal satellites is known at the moment of
+time at which the observations begin. The
+same is true of the ideal suns which replace the
+actual sun.
+
+I shall now refer to only a single one of the
+ideal moons or suns, for, \textit{mutatis mutandis},
+\PageSep{212}
+what is true of one is true of all. It is easy to
+calculate at what hour of the clock, measured in
+the time of the place of observation, the satellite
+in question will be due south. If the ideal
+satellite under consideration were the one which
+generates the principal lunar semidiurnal tide, it
+would be due south very nearly when the real
+moon is south, and the ideal sun which generates
+the principal solar tide is south exactly at noon.
+But there is no such obvious celestial phenomenon
+associated with the transit of any other of
+the satellites, although it is easy to calculate the
+time of the southing of each of them. We have
+then to discover how many hours elapse after
+the passage of the particular satellite up to the
+high water of its tide wave. The height of
+the wave crest above, and the depression of the
+wave hollow below the mean water mark must
+also be determined. When this problem has
+been solved for all the ideal satellites and suns,
+the tides are said to be reduced, and the reduction
+furnishes the materials for a tide table for
+the place of observation.
+
+The difficulty of finding the time of passage
+and the height of the wave due to any one of
+the satellites arises from the fact that all the
+waves really coëxist, and are not separately
+manifest. The nature of the disentanglement
+may be most easily explained from a special
+case, say for example that of the principal lunar
+\PageSep{213}
+semidiurnal tide, of which the crests follow one
+another at intervals of $12$~hours $25$~minutes $14\frac{1}{6}$~seconds.
+
+Since the waves follow one another at intervals
+of approximately, but not exactly, a half-day, it
+is convenient to manipulate the time scale so as
+\index{Time!lunar}%
+to make them exactly semidiurnal. Accordingly
+we describe $24$~hours $50$~minutes $28\frac{1}{3}$~seconds as
+a lunar day, so that there are exactly two waves
+\index{Lunar!time}%
+following one another in the lunar day.
+
+The tide curve furnishes the height of the
+\index{Curve, tide!partitioned into lunar time}%
+water at every moment of time, but the time
+having been registered by the clock of the tide
+gauge is partitioned into ordinary days and
+hours. It may, however, be partitioned at intervals
+of $24$~hours $50$~minutes $28\frac{1}{3}$~seconds, and
+into the twenty-fourth parts of that period, and
+it will then be divided into lunar days and hours.
+On each lunar day the tide for which we are
+searching presents itself in the same way, so
+that it is always high and low water at the same
+hour of the lunar clock, with exactly two high
+waters and two low waters in the lunar day.
+
+Now the other simple tides are governed by
+other scales of time, so that in a long succession
+of days their high waters and low waters occur
+at every hour of the lunar clock. If then we
+find the average curve of rise and fall of the
+water, when the time is divided into lunar days
+and hours, and if we use for the average a long
+\PageSep{214}
+succession of days, all the other tide waves will
+disappear, and we shall be left with only the
+lunar semidiurnal tide, purified from all the
+others which really coëxist with it.
+
+The numerical process of averaging thus leads
+to the obliteration of all but one of the ideal
+satellites, and this is the foundation of the
+method of harmonic analysis. The average
+lunar tide curve may be looked on as the outcome
+of a single day of observation, when all
+but the selected satellite have been obliterated.
+The height of the average wave, and the interval
+after lunar noon up to high water, are the
+two tidal constants for the lunar semidiurnal
+tide, and they enable us to foretell that tide for
+all future time.
+
+If the tide curve were partitioned into other
+days and hours of appropriate lengths, it would
+be possible by a similar process of averaging to
+single out another of the constituent tide waves,
+and to determine its two tidal constants, which
+contain the elements of prediction with respect
+to it. By continued repetition of operations of
+this kind, all the constituents of practical importance
+can be determined, and recorded numerically
+by means of their pairs of tidal constants.
+
+The possibility of the disentanglement has
+now been demonstrated, but the work of carrying
+out these numerical operations would be
+\PageSep{215}
+\index{Schedule for reducing tidal observations}%
+fearfully laborious. The tide curve would have
+to be partitioned into about a dozen kinds of
+days of various lengths, and the process would
+entail measurements at each of the $24$~hours of
+each sort of day throughout the whole series.
+There are about nine thousand hours in a year,
+and it would need about a hundred thousand
+measurements of the curve to evaluate twelve
+different partial tides; each set of measured
+heights would then have to be treated separately
+to find the several sorts of averages. Work of
+this kind has usually to be done by paid computers,
+and the magnitude of the operation
+would make it financially prohibitive. It is,
+however, fortunately possible to devise abridged
+methods, which bring the work within manageable
+limits.
+
+In order to minimize the number of measurements,
+the tide curve is only measured at each
+of the $24$~exact hours of ordinary time, the
+height at noon being numbered $0$~hr., and that
+at midnight $12$~hrs., and so on up to $24$~hrs.
+After obtaining a set of $24$~measurements for
+each day, the original tide curve is of no further
+use. The number of measurements involved is
+still large, but not prohibitive. It would be
+somewhat too technical, in a book of this kind,
+to explain in detail how the measured heights of
+the water at the exact hours of ordinary time
+may be made to give, with fair approximation,
+\PageSep{216}
+\index{Schedule for reducing tidal observations}%
+the heights at the exact hours of other time
+scales. It may, however, be well to explain that
+this approximate method is based on the fact,
+that each exact hour of any one of the special
+time scales must of necessity fall within half an
+hour of one of the exact hours of ordinary time.
+The height of the water at the nearest ordinary
+hour is then accepted as giving the height at the
+exact hour of the special time. The results, as
+computed in this way, are subjected to a certain
+small correction, which renders the convention
+accurate enough for all practical purposes.
+
+A schedule, serviceable for all time and for
+all places, is prepared which shows the hour of
+ordinary time lying nearest to each successive
+hour of any one of the special times. The successive
+$24$~hourly heights, as measured on the tide
+curve, are entered in this schedule, and when
+the entry is completed the heights are found to
+be arranged in columns, which follow the special
+time scale with a sufficiently good approximation
+to accuracy. A different form of schedule is
+required for each partial tide, and the entry of
+the numbers therein is still enormously laborious,
+although far less so than the re-partitions and
+re-measurements of the tide curve would be.
+
+The operation of sorting the numbers into
+schedules has been carried out in various ways.
+In the work of the Indian Survey, the numbers
+\index{Indian Survey!method of reducing tidal observations}%
+have been re-copied over and over again. In
+\PageSep{217}
+\index{Abacus for reducing tidal observations|(}%
+\index{Darwin, G. H.!tidal abacus|(}%
+\index{Indian Survey!method of reducing tidal observations}%
+the office of the United States Coast Survey use
+\index{United States Coast Survey!method of reducing tidal observations}%
+is made of certain card templates pierced with
+holes. These templates are laid upon the tabulation
+of the measurements of the tide curve,
+and the numbers themselves are visible through
+the holes. On the surface of the template lines
+are drawn from hole to hole, and these lines
+indicate the same grouping of the numbers as
+that given by the Indian schedules. Dr.~Börgen,
+\index{Borgen@Börgen, method of reducing tidal observations}%
+of the Imperial German Marine Observatory
+\index{German method of reducing tidal observations}%
+at Wilhelmshaven, has used sheets of tracing
+paper to attain the same end. The Indian procedure
+is unnecessarily laborious, and the American
+and German plans appear to have some
+disadvantage in the fact that the numbers to be
+added together lie diagonally across the page.
+I am assured by some professional computers
+that diagonal addition is easy to perform correctly;
+nevertheless this appeared to me to be
+so serious a drawback, that I devised another
+plan by which the numbers should be brought
+into vertical columns, without the necessity of
+re-copying them. In my plan each day is treated
+as a unit and is shifted appropriately. It might
+be thought that the results of the grouping
+would be considerably less accurate than in the
+former methods, but in fact there is found to be
+no appreciable loss of accuracy.
+
+I have $74$~narrow writing-tablets of xylonite,
+divided by lines into $24$~compartments; the
+\PageSep{218}
+tablets are furnished with spikes on the under
+side, so that they can be fixed temporarily in any
+position on an ordinary drawing-board. The
+compartments on each strip are provided for the
+entry of the $24$~tidal measurements appertaining
+to each day. Each strip is stamped at its end
+with a number specifying the number of the day
+to which it is appropriated.
+
+The arrangement of these little tablets, so that
+the numbers written on them may fall into columns,
+is indicated by a sheet of paper marked
+with a sort of staircase, which shows where each
+tablet is to be set down, with its spikes piercing
+the guide sheet. When the strips are in place,
+as shown in~\fig{33}, the numbers fall into $48$~columns,
+numbered $0$,~$1$,~\dots~$23$, $0$,~$1$,~\dots~$23$
+twice over. The guide sheet shown in the figure~\figref{33}
+is the one appropriate for the lunar semidiurnal
+tide for the fourth set of $74$~days of
+a year of observation. The upper half of the
+tablets are in position, but the lower ones are
+left unmounted, so as the better to show the
+staircase of marks.
+
+Then I say that the average of all the $74$~numbers
+standing under the two~$0$'s combined
+will give the average height of water at $0$~hr.\
+of lunar time, and the average of the numbers
+under~$1$, that at $1$~hr.\ of lunar time, and so forth.
+Thus, after the strips are pegged out, the computer
+has only to add the numbers in columns in
+%% Plate 2
+\TallFig{33}{Tidal Abacus}{jpg}
+%% Facing page
+%[Blank Page]
+\PageSep{219}
+order to find the averages. There are other
+sheets of paper marked for such other rearrangements
+of the strips that each new setting gives
+one of the required results; thus a single writing
+of the numbers serves for the whole computation.
+It is usual to treat a whole year of
+observations at one time, but the board being
+adapted for taking only $74$~successive days, five
+series of writings are required for $370$~days,
+which is just over a year. The number~$74$ was
+chosen for simultaneous treatment, because $74$~days
+is almost exactly five semilunations, and
+accordingly there will always be five spring tides
+on record at once.
+
+In order to guard the computer against the
+use of the wrong paper with any set of strips,
+the guide sheets for the first set of $74$~days are
+red; for the second they are yellow; for the
+third green; for the fourth blue; for the fifth
+violet, the colors being those of the rainbow.
+
+The preparation of these papers entailed a
+great deal of calculation in the first instance, but
+the tidal computer has merely to peg out the
+tablets in their right places, verifying that the
+numbers stamped on the ends of the strips agree
+with the numbers on the paper. The addition
+of the long columns of figures is certainly laborious,
+but it is a necessary incident of every
+method of reducing tidal observations.
+
+The result of all the methods is that for each
+\PageSep{220}
+partial tide we have a set of $24$~numbers, which
+represent the oscillations of the sea due to the
+isolated action of one of the ideal satellites, during
+the period embraced between two successive
+passages of that satellite to the south of the
+place of observation. The examination of each
+partial tide wave gives its height, and the interval
+of time which elapses after its satellite has
+passed the meridian until it is high water for
+that particular tide. The height and interval
+are the tidal constants for that particular tide, at
+the port of observation.
+\index{Observation!reduction of tidal|)}%
+
+The results of this ``reduction of the observations''
+are contained in some fifteen or twenty
+pairs of tidal constants, and these numbers contain
+a complete record of the behavior of the sea
+at the place in question.
+
+\begin{Authorities}
+G.~H. Darwin, \Title{Harmonic Analysis, \&c.}: ``Report to British Association,''
+1883.
+
+G.~H. Darwin, \Title{On an apparatus for facilitating the reduction of
+\index{Darwin, G. H.!tidal abacus|)}%
+tidal observations}: ``Proceedings of the Royal Society,'' vol.~lii.\
+1892.
+\end{Authorities}
+\index{Abacus for reducing tidal observations|)}%
+\index{Reduction of tidal observations|)}%
+\PageSep{221}
+
+
+\Chapter{XIII}
+{Tide Tables}
+
+\First{A tide table} professes to tell, at a given
+\index{Tables, tide|(}%
+place and on a given day, the time of high and
+low water, together with the height of the rise
+and the depth of the fall of the water, with
+reference to some standard mark on the shore.
+A perfect tide table would tell the height of the
+water at every moment of the day, but such a
+table would be so bulky that it is usual to predict
+only the high and low waters.
+
+There are two kinds of tide table, namely,
+those which give the heights and times of high
+and low water for each successive day of each
+year, and those which predict the high and low
+water only by reference to some conspicuous
+celestial phenomenon. Both sorts of tide table
+refer only to the particular harbor for which they
+are prepared.
+
+The first kind contains definite forecasts for
+each day, and may be called a special tide table.
+Such a table is expensive to calculate, and must
+be published a full year beforehand. Special
+tide tables are published by all civilized countries
+for their most important harbors. I believe that
+\PageSep{222}
+\index{Indian Survey!tide tables}%
+\index{United States Coast Survey!tide tables of}%
+the most extensive publications are those of the
+Indian Government for the Indian Ocean, and
+of the United States Government for the coasts
+of North America. The Indian tables contain
+\index{America, North, tide tables for}%
+predictions for about thirty-seven ports.
+
+The second kind of table, where the tide is
+given by reference to a celestial phenomenon,
+may be described as a general one. It is here
+necessary to refer to the Nautical Almanack for
+the time of occurrence of the celestial phenomenon,
+and a little simple calculation must then be
+made to obtain the prediction. The phenomenon
+to which the tide is usually referred is the passage
+of the moon across the meridian of the place of
+observation, and the table states that high and
+low water will occur so many hours after the
+moon's passage, and that the water will stand at
+such and such a height.
+
+The moon, at her change, is close to the sun
+and crosses the meridian at noon; she would
+then be visible but for the sun's brightness, and
+if she did not turn her dark side towards us.
+She again crosses the meridian invisibly at midnight.
+At full moon she is on the meridian,
+visibly at midnight, and invisibly at noon. At
+waxing half moon she is visibly on the meridian
+at six at night, and at waning half moon at six
+in the morning. The hour of the clock at which
+the moon passes the meridian is therefore in effect
+a statement of her phase. Accordingly the
+\PageSep{223}
+relative position of the sun and moon is directly
+involved in a statement of the tide as corresponding
+to a definite hour of the moon's passage. A
+table founded on the time of the moon's passage
+must therefore involve the principal lunar and
+solar semidiurnal tides.
+
+At places where successive tides differ but little
+from one another, a simple table of this kind
+suffices for rough predictions. The curves marked
+Portsmouth in \fig{34} show graphically the interval
+after the moon's passage, and the height
+of high water at that port, for all the hours of
+the moon's passage. We have seen in \Ref{Chapter}{X}.\
+that the tide in the North Atlantic is principally
+due to a wave propagated from the Southern
+Ocean. Since this wave takes a considerable
+time to travel from the Cape of Good Hope to
+England, the tide here depends, in great measure,
+on that generated in the south at a considerable
+time earlier. It has therefore been found better
+to refer the high water to a transit of the moon
+which occurred before the immediately preceding
+one. The reader will observe that it is noted on
+the upper figure that $28$~hours have been subtracted
+from the Portsmouth intervals; that is
+to say, the intervals on the vertical scale marked
+$6$,~$7$,~$8$ hours are, for Portsmouth, to be interpreted
+as meaning $34$,~$35$,~$36$ hours. These are
+the hours which elapse after any transit of the
+moon up to high water. The horizontal scale is
+\PageSep{224}
+\index{Moon and earth!tidal prediction by reference to transit|(}%
+one of the times of moon's transit and of phases
+of the moon; the vertical scale in the lower figure
+is one of feet, and it shows the height to
+which the water will rise measured from a certain
+mark ashore. These Portsmouth curves do not
+extend beyond 12~o'clock of moon's transit; this
+is because there is hardly any diurnal inequality,
+\index{Diurnal inequality!complicates prediction}%
+and it is not necessary to differentiate the hours
+as either diurnal or nocturnal, the statement being
+equally true of either day or night. Thus
+if the Portsmouth curves had been extended onward
+from $12$~hours to $24$~hours of the clock time
+of the moon's passage, the second halves of the
+curves would have been merely the duplicates of
+the first halves.\footnote
+ {Before the introduction of the harmonic analysis of the tides
+ described in preceding chapters, tidal observations were ``reduced''
+ by the construction of such figures as these, directly from
+ the tidal observations. Every high water was tabulated as appertaining
+ to a particular phase of the moon, both as to its height
+ and as to the interval between the moon's transit and the occurrence
+ of high water. The average of a long series of observations
+ may be represented in the form of curves by such figures
+ as these.}
+
+But the time of the moon's passage leaves her
+angular distance from the equator and her linear
+distance from the earth indeterminate; and since
+the variability of both of these has its influence
+on the tide, corrections are needed which add
+something to or subtract something from the
+tabular values of the interval and height, as dependent
+solely on the time of the moon's passage.
+\PageSep{225}
+\index{Diurnal inequality!complicates prediction}%
+\index{Interval from moon's transit to high water!at Portsmouth and at Aden}%
+The sun also moves in a plane which is oblique
+to the equator, and so similar allowances must be
+made for the distance of the sun from the equator,
+and for the variability in his distance from the
+earth. In order to attain accuracy with a tide
+table of this sort, eight or ten corrections are
+needed, and the use of the table becomes complicated.
+
+It is, however, possible by increasing the number
+of such figures or tables to introduce into
+them many of the corrections referred to; and
+the use of a general tide table then becomes comparatively
+simple. The sun occupies a definite
+position with reference to the equator, and stands
+at a definite distance from the earth on each day
+of the year; also the moon's path amongst the
+stars does not differ very much from the sun's.
+Accordingly a tide table which states the interval
+after the moon's passage to high or low water
+and the height of the water on a given day of
+\index{Height of tide!at Portsmouth and at Aden}%
+the year will directly involve the principal inequalities
+in the tides. As the sun moves slowly
+amongst the stars, a table applicable to a given
+day of the year is nearly correct for a short time
+before and after that date. If, then, a tide table,
+stating the time and height of the water by reference
+to the moon's passage, be computed for
+say every ten days of the year, it will be very
+nearly correct for five days before and for five
+days after the date for which it is calculated.
+\PageSep{226}
+
+The curves marked Aden, March and June, in
+\index{Prediction of tide!example at Aden|(}%
+\fig{34}, show the intervals and heights of tide,
+on the 15th of those months at that port, for all
+the hours of the moon's passage. The curves are
+to be read in the same way as those for Portsmouth,
+but it is here necessary to distinguish the
+hours of the day from those of the night, and
+accordingly the clock times of moon's transit are
+numbered from $0$~hr.\ at noon up to $24$~hrs.\ at
+the next noon. The curves for March differ so
+much from those for June, that the corrections
+would be very large, if the tides were treated at
+Aden by a single pair of average curves as at
+Portsmouth.
+
+The law of the tides, as here shown graphically,
+may also be stated numerically, and the
+use of such a table is easy. The process will be
+best explained by an example, which happens to
+be retrospective instead of prophetic. It will involve
+that part of the complete table (or series of
+curves) for Aden which applies to the 15th of
+March of any year. Let it be required then to
+find the time and height of high water on March~17,
+1889. The Nautical Almanack for that year
+shows that on that day the moon passed the meridian
+of Aden at eleven minutes past noon of
+Aden time, or in astronomical language at $0$~hr.\
+$11$~mins. Now the table, or the figure of intervals,
+shows that if the moon had passed at $0$~hr.,
+or exactly at noon, the interval would have been
+\PageSep{227}
+$8$~hrs.\ $9$~mins., and that if she had passed at $0$~hr.\
+$20$~mins., or 12.20~\PM\ of the day, the interval
+would have been $7$~hrs.\ $59$~mins. But on
+March~17th the moon actually crossed at $0$~hr.\
+\Figure{34}{Curves of Intervals and Heights at Portsmouth
+and at Aden}{png}
+$11$~mins., very nearly halfway between noon and
+$20$~mins.\ past noon. Hence the interval was
+halfway between $8$~hrs.\ $9$~mins.\ and $7$~hrs.\ $59$~mins.,
+so that it was $8$~hrs.\ $4$~mins. Accordingly
+it was high water $8$~hrs.\ $4$~mins.\ after the moon
+\PageSep{228}
+crossed the meridian. But the moon crossed at
+$0$~hr.\ $11$~mins., therefore the high water occurred
+at 8.15~\PM.
+
+Again the table of heights, or the figure, shows
+that on March~15th, if the moon crossed at $0$~hr.\
+$0$~min.\ the high water would be $6.86$~ft.\ above
+a certain mark ashore, and if she crossed at $0$~hr.\
+$20$~mins.\ the height would be $6.92$~ft. But on
+March~17th the moon crossed halfway between
+$0$~hr.\ $0$~min.\ and $0$~hr.\ $20$~mins., and therefore
+the height was halfway between $6.86$~ft.\ and
+$6.92$~ft., that is to say, it was $6.89$~ft., or $6$~ft.\
+$11$~in. We therefore conclude that on March~17,
+1889, the sea at high water rose to $6$~ft.\
+$11$~in., at 8.15~\PM. I have no information as
+to the actual height and time of high water on
+that day, but from the known accuracy of other
+predictions at Aden we may be sure that this
+agrees pretty nearly with actuality. The predictions
+derived from this table are markedly improved
+when a correction, either additive or subtractive,
+is applied, to allow for the elliptic motion
+of the moon round the earth. On this particular
+occasion the moon stood rather nearer the earth
+than the average, and therefore the correction to
+the height is additive; the correction to the time
+also happens to be additive, although it could
+not be foreseen by general reasoning that this
+would be the case. The corrections for March~17,
+1889, are found to add about $2$~mins.\ to the
+\PageSep{229}
+time, bringing it to 8.17~\PM, and nearly two
+inches to the height, bringing it to $7$~ft.\ $1$~in.
+
+This sort of elaborate general tide table has
+been, as yet, but little used. It is expensive to
+calculate, in the first instance, and it would occupy
+two or three pages of a book. The expense
+is, however, incurred once for all, and the table
+is available for all time, provided that the tidal
+observations on which it is based have been good.
+A sea captain arriving off his port of destination
+would not take five minutes to calculate the two
+or three tides he might require to know, and the
+information would often be of the greatest value
+to him.
+
+As things stand at present, a ship sailing to
+most Chinese, Pacific, or Australian ports is only
+furnished with a statement, often subject to considerable
+error, that the high water will occur at
+so many hours after the moon's passage and will
+rise so many feet. The average rise at springs and
+neaps is generally stated, but the law of the variability
+according to the phases of the moon is wanting.
+But this is not the most serious defect in the
+information, for it is frequently noted that the
+tide is ``affected by diurnal inequality,'' and this
+note is really a warning to the navigator that he
+cannot foretell the time of high water within two
+or three hours of time, or the height within several
+feet.
+
+Tables of the kind I have described would
+\PageSep{230}
+\index{Prediction of tide!method of computing|(}%
+banish this extreme vagueness, but they are more
+likely to be of service at ports of second-rate importance
+than at the great centres of trade, because
+at the latter it is worth while to compute
+full special tide tables for each year.
+\index{Tables, tide!method of calculating|(}%
+
+It is unnecessary to comment on the use of
+tables containing predictions for definite days,
+since it merely entails reference to a book, as to
+a railway time table. Such special tables are undoubtedly
+the most convenient, but the number
+of ports which can ever be deemed worthy of the
+great expense incidental to their preparation
+must always be very limited.
+
+\TB
+
+We must now consider the manner in which
+tide tables are calculated. It is supposed that
+careful observations have been made, and that
+the tidal constants, which state the laws governing
+the several partial tides, have been accurately
+determined by harmonic analysis. The analysis
+of tidal observations consists in the dissection of
+the aggregate tide wave into its constituent partial
+waves, and prediction involves the recomposition
+or synthesis of those waves. In the synthetic
+\index{Synthesis of partial tides for prediction|(}%
+process care must be taken that the partial
+waves shall be recompounded in their proper
+relative positions, which are determined by the
+places of the moon and sun at the moment of
+time chosen for the commencement of prediction.
+
+The synthesis of partial waves may be best
+\index{Moon and earth!tidal prediction by reference to transit|)}%
+\index{Prediction of tide!example at Aden|)}%
+\PageSep{231}
+arranged in two stages. It has been shown in
+\Ref{Chapter}{XI}.\ that the partial waves fall naturally
+into three groups, of which the third is practically
+insignificant. The first and second are the
+semidiurnal and diurnal groups. The first process
+is to unite each group into a single wave.
+
+We will first consider the semidiurnal group.
+Let us now, for the moment, banish the tides
+from our minds, and imagine that there are two
+trains of waves traveling simultaneously along a
+straight canal. If either train existed by itself
+every wave would be exactly like all its brethren,
+both in height, length, and period. Now suppose
+that the lengths and periods of the waves
+of the two coëxistent trains do not differ much
+from one another, although their heights may
+differ widely. Then the resultant must be a single
+train of waves of lengths and periods intermediate
+between those of the constituent waves,
+but in one part of the canal the waves will be
+high, where the two sets of crests fall in the
+same place, whilst in another they will be low,
+where the hollow of the smaller constituent wave
+falls in with the crest of the larger. If only one
+part of the canal were visible to us, a train of
+waves would pass before us, whose heights would
+gradually vary, whilst their periods would change
+but little.
+
+In the same way two of the semidiurnal tide
+waves, when united by the addition of their separate
+\PageSep{232}
+displacements from the mean level, form a
+single wave of variable height, with a period still
+semidiurnal, although slightly variable. But
+there is nothing in this process which limits the
+synthesis to two waves, and we may add a third
+and a fourth, finally obtaining a single semidiurnal
+wave, the height of which varies according
+to a very complex law.
+
+A similar synthesis is then applied to the second
+group of waves, so that we have a single
+variable wave of approximately diurnal period.
+The final step consists in the union of the single
+semidiurnal wave with the single diurnal one into
+a resultant wave. When the diurnal wave is
+large, the resultant is found to undergo very
+great variability both in period and height. The
+principal variations in the relative positions of
+the partial tide waves are determined by the
+phases of the moon and by the time of year, and
+there is, corresponding to each arrangement of
+the partial waves, a definite form for the single
+resultant wave. The task of forming a general
+tide table therefore consists in the determination
+of all the possible periods and heights of the resultant
+wave and the tabulation of the heights
+and intervals after the moon's passage of its high
+and low waters.
+
+I supposed formerly that the captain would
+himself calculate the tide he required from the
+general tide table, but such calculation may be
+\PageSep{233}
+\index{Machine, tide-predicting}%
+\index{Predicting machine for tides|(}%
+done beforehand for every day of a specified
+year, and the result will be a special tide table.
+There are about $\DPchg{1400}{1,400}$ high and low waters in
+a year, so that the task is very laborious, and
+has to be repeated each year.
+
+\TB
+
+It is, however, possible to compute a special
+tide table by a different and far less laborious
+method. In this plan an ingenious mechanical
+device replaces the labor of the computer. The
+first suggestion for instrumental prediction of
+tides was made, I think, by Sir William Thomson,
+now Lord Kelvin, in~1872. Mr.~Edward
+\index{Kelvin, Lord!tide predicting machine}%
+Roberts bore an important part in the practical
+\index{Roberts, E., the tide-predicting machine}%
+realization of such a machine, and a tide predicter
+was constructed by Messrs.\ Légé for the
+\index{Lege@Légé, constructor of tide-predicting machine}%
+Indian Government under his supervision. This
+is, as yet, the only complete instrument in existence.
+But others are said to be now in course
+of construction for the Government of the
+United States and for that of France. The
+Indian machine cost so much and works so well,
+that it is a pity it should not be used to the full
+extent of its capacity. The Indian Government
+has, of course, the first claim on it, but the use
+of it is allowed to others on the payment of a
+small fee. I believe that, pending the construction
+of their own machine, the French authorities
+are obtaining the curves for certain tidal
+predictions from the instrument in London.
+\index{Prediction of tide!method of computing|)}%
+\index{Synthesis of partial tides for prediction|)}%
+\PageSep{234}
+
+Although the principle involved in the tide
+predicter is simple, yet the practical realization
+of it is so complex that a picture of the whole
+machine would convey no idea of how it works.
+I shall therefore only illustrate it diagrammatically,
+in~\fig{35}, without any pretension to scale
+or proportion. The reader must at first imagine
+that there are only two pulleys, namely, $A$~and~$B$,
+so that the cord passes from the fixed end~$F$
+under~$A$ and over~$B$, and so onward to the pencil.
+The pulley~$B$ is fixed, and the pulley~$A$ can slide
+vertically up and down in a slot, which is not
+shown in the diagram. If $A$~moves vertically
+through any distance, the pencil must clearly
+move through double that distance, so that
+when $A$~is highest the pencil is lowest, and vice
+versa.
+
+The pencil touches a uniformly revolving
+drum, covered with paper; thus if the pulley~$A$
+executes a simple vertical oscillation, the pencil
+draws a simple wave on the drum. Now the
+pulley is mounted on an inverted T-shaped
+frame, and a pin, fixed in a crank~$C$, engages in
+the slit in the horizontal arm of the T-piece.
+When the crank~$C$ revolves, the pulley~$A$ executes
+a simple vertical oscillation with a range depending
+on the throw of the crank.\footnote
+ {I now notice that the throw of the crank~$C$ is too small to
+ have allowed the pencil to draw so large a wave as that shown
+ on the drum. But as this is a mere diagram, I have not thought
+ it worth while to redraw the whole.}
+The position
+\PageSep{235}
+of the pin is susceptible of adjustment on the
+crank, so that its throw and the range of oscillation
+of the pulley can be set to any required
+\Figure[0.7]{35}{Diagram of Tide-predicting Instrument}{png}
+length---of course within definite limits determined
+by the size of the apparatus.
+
+The drum is connected to the crank~$C$ by a
+train of wheels, so that as the crank rotates the
+drum also turns at some definitely proportional
+rate. If, for example, the crank revolves twice
+for one turn of the drum, the pencil will draw a
+simple wave, with exactly two crests in one circumference
+of the drum. If one revolution of
+the drum represents a day, the graphical time
+scale is $24$~hours to the circumference of the
+\PageSep{236}
+drum. If the throw of the crank be one inch,
+the pulley will oscillate with a total range of two
+inches, and the pencil with a total range of four
+inches. Then taking two inches lengthwise on
+the drum to represent a foot of water, the curve
+drawn by the pencil might be taken to represent
+the principal solar semidiurnal tide, rising one
+foot above and falling one foot below the mean
+sea level.
+
+I will now show how the machine is to be
+adjusted so as to give predictions. We will
+suppose that it is known that, at noon of the
+first day for which prediction is required, the
+solar tide will stand at $1$~ft.\ $9$~in.\ above mean
+sea level and that the water will be rising. Then,
+the semi-range of this tide being one foot, the
+pin is adjusted in the crank at one inch from
+the centre, so as to make the pencil rock through
+a total range of $4$~inches, representing $2$~feet.
+The drum is now turned so as to bring the noon-line
+of its circumference under the pencil, and
+the crank is turned so that the pencil shall be
+$3\frac{1}{2}$~inches (representing $1$~ft.\ $9$~in.\ of water)
+below the middle of the drum, and so that when
+the machine starts, the pencil will begin to descend.
+The curve being drawn upside-down,
+the pencil is set below the middle line because
+the water is to be above mean level, and it must
+begin to descend because the water is to ascend.
+The train of wheels connecting the crank and
+\PageSep{237}
+drum is then thrown into gear, and the machine
+is started; it will then draw the solar tide curve,
+on the scale of $2$~inches to the foot, for all
+time.
+
+If the train of wheels connecting the crank to
+the drum were to make the drum revolve once
+whilst the crank revolves $1.93227$~times, the
+curve would represent a lunar semidiurnal tide.
+The reason of this is that $1.93227$~is the ratio
+of $24$~hours to $12$~h.\ $25$~m.\ $14$~s., that is to say,
+of a day to a lunar half day. We suppose the
+circumference of the drum still to represent an
+ordinary day of $24$~hours, and therefore the
+curve drawn by the pencil will have lunar semidiurnal
+periodicity. In order that these curves
+may give predictions of the future march of that
+tide, the throw of the crank must be set to give the
+correct range and its angular position must give
+the proper height at the moment of time chosen
+for beginning. When these adjustments are
+made the curve will represent that tide for all
+time.
+
+We have now shown that, by means of appropriate
+trains of wheels, the machine can be made
+to predict either the solar or the lunar tide; but
+we have to explain the arrangement for combining
+them. If, still supposing there to be
+only the two pulleys $A$,~$B$, the end~$F$ of the cord
+were moved up or down, its motion would be
+transmitted to the pencil, whether the crank~$C$
+\PageSep{238}
+and pulley~$A$ were in motion, or at rest; but if
+they were in motion, the pencil would add the
+motion of the end of the cord to that of the
+pulley. If then there be added another fixed
+pulley~$B'$, and another movable pulley~$A'$, driven
+by a crank and T-piece (not shown in the diagram),
+the pencil will add together the movements
+of the two pulleys $A$~and~$A'$. There must
+now be two trains of wheels, one connecting $A$
+with the drum and the other for~$A'$. If a single
+revolution of the drum causes the crank~$C$ to
+turn twice, whilst it makes the crank of~$A'$ rotate
+$1.93227$~times, the curve drawn will represent
+the union of the principal solar and lunar semidiurnal
+tides. The trains of wheels requisite for
+transmitting motion from the drum to the two
+cranks in the proper proportions are complicated,
+but it is obviously only a matter of calculation
+to determine the numbers of the teeth in the
+several wheels in the trains. It is true that rigorous
+accuracy is not attainable, but the mechanism
+is made so nearly exact that the error in the
+sum of the two tides would be barely sensible
+even after $\DPchg{3000}{3,000}$~revolutions of the drum. It is
+of course necessary to set the two cranks with
+their proper throws and at their proper angles
+so as to draw a curve which shall, from the noon
+of a given day, correspond to the tide at a given
+place.
+
+It must now be clear that we may add as
+\PageSep{239}
+many more movable pulleys as we like. When
+the motion of each pulley is governed by an
+appropriate train of wheels, the movement of
+the pencil, in as far as it is determined by that
+pulley, corresponds to the tide due to one of our
+ideal satellites. The resultant curve drawn on
+the drum is then the synthesis of all the partial
+tides, and corresponds with the motion of the
+sea.
+
+The instrument of the Indian Government
+unites twenty-four partial tides. In order to
+trace a tide curve, the throws of all the cranks
+are set so as to correspond with the known
+heights of the partial tides, and each crank is set
+at the proper angle to correspond with the moment
+of time chosen for the beginning of the tide
+table. It is not very difficult to set the cranks
+and pins correctly, although close attention is of
+course necessary. The apparatus is then driven
+by the fall of a weight, and the paper is fed
+automatically on to the drum and coiled off on
+to a second drum, with the tide curve drawn on
+it. It is only necessary to see that the paper
+runs on and off smoothly, and to write the date
+from time to time on the paper as it passes, in
+order to save future trouble in the identification
+of the days. It takes about four hours to run
+off the tides for a year.
+
+The Indian Government sends home annually
+the latest revision of the tidal constants for
+\PageSep{240}
+thirty-seven ports in the Indian Ocean. Mr.~Roberts
+sets the machine for each port, so as to
+correspond with noon of a future 1st~of January,
+and then lets it run off a complete tide
+curve for a whole year. The curve is subsequently
+measured for the time and height of
+each high and low water, and the printed tables
+are sold at the moderate price of four rupees.
+The publication is made sufficiently long beforehand
+to render the tables available for future
+voyages. These tide tables are certainly amongst
+the most admirable in the world.
+
+\TB
+
+It is characteristic of England that the machine
+is not, as I believe, used for any of the
+home ports, and only for a few of the colonies.
+The neglect of the English authorities is not,
+however, so unreasonable as it might appear to
+be. The tides at English ports are remarkably
+simple, because the diurnal inequality is practically
+absent. The applicability of the older
+methods of prediction, by means of such curves
+as that for Portsmouth in~\fig{34}, is accordingly
+easy, and the various corrections are well determined.
+The arithmetical processes are therefore
+not very complicated, and ordinary computers
+are capable of preparing the tables with but
+little skilled supervision. Still it is to be regretted
+that this beautiful instrument should not
+be more used for the home and colonial ports.
+\PageSep{241}
+
+The excellent tide tables of the Government
+of the United States have hitherto been prepared
+by the aid of a machine of quite a different
+character, the invention of the late Professor
+Ferrel. This apparatus virtually carries out
+\index{Ferrel, tide-predicting instrument}%
+\index{Predicting machine for tides!Ferrel's}%
+that process of compounding all the waves together
+into a single one, which I have described
+as being done by a computer for the formation
+of a general tide table. It only registers, however,
+the time and height of the maxima and
+minima---the high and low waters. I do not
+think it necessary to describe its principle in
+detail, because it will shortly be superseded by a
+machine like, but not identical with, that of the
+Indian Government.
+
+\begin{Authorities}
+G.~H. Darwin, \Title{On Tidal Prediction}. ``Philosophical Transactions
+of the Royal Society,'' A.~1891, pp.~159--229.
+
+In the example of the use of a general tide table at Aden,
+given in this chapter, the datum from which the height is measured
+is $0.37$~ft.\ higher than that used in the Indian Tide Tables;
+\index{Tables, tide|)}%
+\index{Tables, tide!method of calculating|)}%
+accordingly $4\frac{1}{2}$~inches must be added to the height, in order to
+bring it into accordance with the official table.
+
+Sir William Thomson, \Title{Tidal Instruments}, and the subsequent
+discussion. ``Institute of Civil Engineers,'' vol.~lxv.
+
+William Ferrel, \Title{Description of a Maxima and Minima Tide-predicting
+\index{Machine, tide-predicting}%
+Machine}. ``United States Coast Survey,'' 1883.
+\end{Authorities}
+\index{Predicting machine for tides|)}%
+\PageSep{242}
+
+
+\Chapter[Accuracy of Tidal Prediction]{XIV}
+{The Degree of Accuracy of Tidal Prediction}
+
+\First{The} success of tidal predictions varies much
+\index{Atmospheric pressure!influence on tidal prediction}%
+\index{Prediction of tide!errors in|(}%
+according to the place of observation. They are
+not unfrequently considerably in error in our
+latitude, and throughout those regions called by
+sailors ``the roaring forties.'' The utmost that
+can be expected of a tide table is that it shall
+be correct in calm weather and with a steady
+barometer. But such conditions are practically
+non-existent, and in the North Atlantic the great
+variability in the meteorological elements renders
+tidal prediction somewhat uncertain.
+
+The sea generally stands higher when the
+barometer is low, and lower when the barometer
+is high, an inch of mercury corresponding to
+rather more than a foot of water. The pressure
+of the air on the sea in fact depresses it in those
+places where the barometer is high, and allows it
+to rise where the opposite condition prevails.
+
+Then again a landward wind usually raises the
+\index{Wind!perturbation of, in tidal prediction}%
+sea level, and in estuaries the rise is sometimes
+very great. There is a known instance when the
+Thames at London was raised by five feet in a
+strong gale. Even on the open coast the effect
+\PageSep{243}
+\index{Atmospheric pressure!influence on tidal prediction}%
+of wind is sometimes great. A disastrous example
+\index{Wind!perturbation of, in tidal prediction}%
+of this was afforded on the east coast of England
+in the autumn of~1897, when the conjunction
+of a gale with springtide caused the sea to
+do an enormous amount of damage, by breaking
+embankments and flooding low-lying land.
+
+But sometimes the wind has no apparent effect,
+and we must then suppose that it had been blowing
+previously elsewhere in such a way as to depress
+the water at the point at which we watch it.
+The gale might then only restore the water to its
+normal level, and the two effects might mask one
+another. The length of time during which the
+wind has lasted is clearly an important factor,
+because the currents generated by the wind must
+be more effective in raising or depressing the sea
+level the longer they have lasted.
+
+It does not then seem possible to formulate
+any certain system of allowance for barometric
+pressure and wind. There are, at each harbor,
+certain rules of probability, the application of
+which will generally lead to improvement in the
+prediction; but occasionally such empirical corrections
+will be found to augment the error.
+
+But notwithstanding these perturbations, good
+tide tables are usually of surprising accuracy
+even in northern latitudes; this may be seen
+from the following table showing the results of
+comparisons between prediction and actuality at
+Portsmouth. The importance of the errors in
+\index{Errors in tidal prediction|(}%
+\PageSep{244}
+height depends of course on the range of tide;
+it is therefore well to note that the average ranges
+of tide at springs and neaps are $13$~ft.\ $9$~in.\ and
+$7$~ft.\ $9$~in.\ respectively.
+\begin{table}[hbtp!]
+\caption{Table of Errors in the Prediction of High Water at
+Portsmouth in the Months of January, May, And
+\index{Portsmouth, table of errors in tidal predictions}%
+September, 1897.}
+\[
+\begin{array}{|c|c||c|c|}
+\hline
+\multicolumn{2}{|c||}{\ColHead{Time}} &
+\multicolumn{2}{c|}{\ColHead{Height}} \\
+\hline
+\ColHead[1.2in]{Magnitude of error} &
+\ColHead[0.5in]{Number of cases} &
+\ColHead[1.2in]{Magnitude of error} &
+\ColHead[0.5in]{Number of cases} \\
+\hline
+&& \TEntry{Inches} & \\
+\Z0\mm\text{ to }\Z5\mm& 69 & \Z0\text{ to }\Z6 & 89 \\
+\Z6\mm\text{ to }10\mm & 50 & \Z7\text{ to }12 & 58 \\
+ 11\mm\text{ to }15\mm & 25 & 13\text{ to }18 & 24 \\
+ 16\mm\text{ to }20\mm & 10 & 19\text{ to }24 & \Z6 \\
+ 21\mm\text{ to }25\mm & 11 & \Dash & \Dash \\
+ 26\mm\text{ to }30\mm & \Z7 & \Dash & \Dash \\
+ 31\mm\text{ to }35\mm & \Z4 & \Dash & \Dash \\
+ 52\mm & \Z1 & \Dash & \Dash \\
+\hline
+\Strut\Dash & \llap{$1$}77&\Dash & \llap{$1$}77 \\
+\hline
+\end{array}
+\]
+\end{table}
+\begin{table}[hbtp!]
+\caption{Errors in Height for the Year 1892,
+Excepting Part of July}
+\[
+\begin{array}{|c|c|}
+\hline
+\ColHead{Magnitude of error} & \ColHead{Number of cases} \\
+\hline
+\TEntry{Inches} & \\
+\Z0\text{ to }\Z6& 381 \\
+\Z7\text{ to }12 & 228 \\
+ 13\text{ to }18 & \Z52 \\
+ 19\text{ to }24 & \Z\Z8 \\
+ 31 & \Z\Z1 \\
+\hline
+\Strut\Dash & 670 \\
+\hline
+\end{array}
+\]
+\end{table}
+\PageSep{245}
+
+\begin{Remark}
+\NB---The comparison seems to indicate that these predictions
+might be much improved, because the predicted height is
+nearly always above the observed height, and because the diurnal
+inequality has not been taken into account sufficiently, if at
+all.
+\end{Remark}
+
+In tropical regions the weather is very uniform,
+and in many places the ``meteorological
+tides'' produced by the regularly periodic variations
+of wind and barometric pressure are taken
+into account in tidal predictions.
+
+The apparent irregularity of the tides at Aden
+is so great, that an officer of the Royal Engineers
+has told me that, when he was stationed there
+many years ago, it was commonly believed that
+the strange inequalities of water level were due
+to the wind at distant places. We now know
+that the tide at Aden is in fact marvelously
+regular, although the rule according to which it
+proceeds is very complex. In almost every month
+in the year there are a few successive days when
+there is only one high water and one low water
+in the $24$~hours; and the water often remains
+almost stagnant for three or four hours at a
+time. This apparent irregularity is due to the
+diurnal inequality, which is very great at Aden,
+whereas on the coasts of Europe it is insignificant.
+
+I happen to have a comparison with actuality
+of a few predictions of high water at Aden,
+where the maximum range of the tide is about
+$8$~ft.\ $6$~in. They embrace the periods from March~10
+\index{Errors in tidal prediction|)}%
+\PageSep{246}
+\index{Aden, errors of tidal prediction at}%
+\index{Tables, tide!amount of error in}%
+to April~9, and again from November~12 to
+December~12, 1884. In these two periods there
+were $118$~high waters, but through an accident
+to the tide gauge one high water was not registered.
+On one occasion, when the regular semidiurnal
+sequence of the tide would lead us to
+expect high water, there occurred one of those
+periods of stagnation to which I have referred.
+Thus we are left with $116$~cases of comparison
+between the predicted and actual high waters.
+
+The results are exhibited in the following
+table:---
+\[
+\begin{array}{|c|c||c|c|}
+\hline
+\multicolumn{2}{|c||}{\ColHead{Time}} &
+\multicolumn{2}{c|}{\ColHead{Height}} \\
+\hline
+\ColHead{Magnitude of errors} &
+\ColHead[0.5in]{Number of high waters} &
+\ColHead{Magnitude of errors} &
+\ColHead[0.5in]{Number of high waters} \\
+\hline
+&& \TEntry{Inches} & \\
+\Z0\mm\PadTxt{ and }{to}\Z5\mm & 35 & 0 & 15 \\
+\Z5\mm\PadTxt{ and }{to}10\mm & 32 & 1 & 48 \\
+ 10\mm\PadTxt{ and }{to}15\mm & 19 & 2 & 28 \\
+ 15\mm\PadTxt{ and }{to}20\mm & 19 & 3 & 14 \\
+ 20\mm\PadTxt{ and }{to}25\mm & \Z5 & 4 & 11 \\
+ 26\mm\text{ and }28\mm & \Z2 &
+\multicolumn{1}{c|}{\TEntry{No high water}} & \Z1 \\
+ 33\mm\text{ and }36\mm & \Z2 & \Dash & \Dash \\
+ 56\mm\text{ and }57\mm & \Z2 & \Dash & \Dash \\
+\multicolumn{1}{|c|}{\TEntry{No high water}} & \Z1 & \Dash & \Dash \\
+\hline
+\Strut & 117 && 117 \\
+\hline
+\end{array}
+\]
+
+It would be natural to think that when the
+prediction is erroneous by as much as $57$~minutes,
+it is a very bad one; but I shall show that
+\PageSep{247}
+\index{Tables, tide!amount of error in}%
+this would be to do injustice to the table. On
+several of the occasions comprised in this list
+the water was very nearly stagnant. Now if the
+water only rises about a foot from low to high
+water in the course of four or five hours, it is
+almost impossible to say with accuracy when it
+was highest, and two observers might differ in
+their estimate by half an hour or even by an
+hour.
+
+In the table of comparison there are $11$~cases
+in which the error of time is equal to or greater
+than twenty minutes, and I have examined these
+cases in order to see whether the water was then
+nearly stagnant. A measure of the degree of
+stagnation is afforded by the amount of the rise
+from low water to high water, or of the fall from
+high water to low water. The following table
+gives a classification of the errors of time according
+to the rise or fall:---
+\begin{table}[hp!]
+\caption{Analysis of Errors in Time.}
+\centering
+\begin{tabular}{|c|c|}
+\hline
+\ColHead[1.5in]{Ranges from low water to high water} &
+\ColHead{Errors of time} \\
+\hline
+\Strut
+Nil & \Dash \\
+$6$ in.\ to $8$ in. & $22$, $26$, $28$, $56$, $57$~minutes \\
+$13$ in. & $36$~minutes \\
+$17$ in. & $22$ \Ditto{minutes} \\
+$19$ in. & $33$ \Ditto{minutes} \\
+$2$ ft.\ $10$ in. & $22$ \Ditto{minutes} \\
+$3$ ft.\ $\Z9$ in. & $23$ \Ditto{minutes} \\
+$3$ ft.\ $11$ in. & $20$ \Ditto{minutes} \\
+\hline
+\end{tabular}
+\end{table}
+\PageSep{248}
+
+There are then only three cases when the rise
+of water was considerable, and in the greatest of
+them it was only $3$~ft.\ $11$~in.
+
+If we deduct all the tides in which the range
+between low and high water was equal to or less
+than $19$~inches, we are left with $108$~predictions,
+and in these cases the greatest error in time is
+$23$~mins. In $86$~cases the error is equal to or less
+than a quarter of an hour. This leaves $22$~cases
+where the error was greater than $15$~mins.\ made
+up as follows: $18$~cases with error greater than
+$15$~mins.\ and less than $20$~mins.\ and $3$~cases with
+errors of $20$~mins., $22$~mins., $23$~mins. Thus in
+$106$ out of~$108$ predictions the error of time was
+equal to or less than $20$~minutes.
+
+Two independent measurements of a tide
+curve, for the determination of the time of high
+water, lead to results which frequently differ by
+five minutes, and sometimes by ten minutes. It
+may therefore be claimed that these predictions
+have a very high order of accuracy as regards
+time.
+
+Turning now to the heights, out of $116$~predictions
+the error in the predicted height was
+equal to or less than $2$~inches in $91$~cases, it
+amounted to $3$~inches in $14$~cases, and in the
+remaining $11$~cases it was $4$~inches. It thus appears
+that, as regards the height of the tide also,
+the predictions are of great accuracy. This
+short series of comparisons affords a not unduly
+\PageSep{249}
+favorable example of the remarkable success attainable,
+where tidal observation and prediction
+have been thoroughly carried out at a place
+subject to only slight meteorological disturbance.
+
+If our theory of tides were incorrect, so that
+we imagined that there was a partial tide wave
+of a certain period, whereas in fact such a wave
+has no true counterpart in physical causation,
+the reduction of a year of tidal observation would
+undoubtedly assign some definite small height,
+and some definite retardation of the high water
+after the passage of the corresponding, but
+erroneous, satellite. But when a second series
+of observations is reduced, the two tidal constants
+would show no relationship to their previous
+evaluations. If then reductions carried
+out year after year assign, as they do, fairly
+consistent values to the tidal constants, we may
+feel confident that true physical causation is involved,
+even when the heights of some of the
+constituent tide waves do not exceed an inch
+or two.
+
+Prediction must inevitably fail, unless we have
+lighted on the true causes of the phenomena;
+success is therefore a guarantee of the truth of
+the theory. When we consider that the incessant
+variability of the tidal forces, the complex
+outlines of our coasts, the depth of the sea and
+the earth's rotation are all involved, we should
+\PageSep{250}
+regard good tidal prediction as one of the
+greatest triumphs of the theory of universal
+gravitation.
+
+\begin{Authorities}
+The Portsmouth comparisons were given to the author by the
+Hydrographer of the Admiralty, Admiral Sir W.~J. Wharton.
+
+G.~H. Darwin, \Title{On Tidal Prediction}. ``Philosophical Transactions
+of the Royal Society,'' A.~1891.
+\end{Authorities}
+\index{Prediction of tide!errors in|)}%
+\PageSep{251}
+
+
+\Chapter[Rigidity of the Earth]{XV}
+{Chandler's Nutation---The Rigidity of the
+Earth}
+
+\First{In} the present chapter I have to explain the
+\index{Nutation!Chandler's|(}%
+\index{Variation of latitude|(}%
+origin of a tide of an entirely different character
+from any of those considered hitherto. It may
+fairly be described as a true tide, although it is
+not due to the attraction of either the sun or
+the moon.
+
+We have all spun a top, and have seen it, as
+boys say, go to sleep. At first it nods a little,
+but gradually it settles down to perfect steadiness.
+Now the earth may be likened to a top,
+and it also may either have a nutational or nodding
+motion, or it may spin steadily; it is only
+by observation that we can decide whether it is
+nodding or sound asleep.
+
+The equator must now be defined as a plane
+through the earth's centre at right angles to the
+axis of rotation, and not as a plane fixed with
+reference to the solid earth. The latitude of
+\index{Latitude!periodic variations of|(}%
+any place is the angle\footnote
+ {This angle is technically called the geocentric latitude; the
+ distinction between true and geocentric latitude is immaterial in
+ the present discussion.}
+between the equator and
+\PageSep{252}
+a line drawn from the centre of the earth to the
+place of observation. Now when the earth
+nutates, the axis of rotation shifts, and its
+extremity describes a small circle round the spot
+which is usually described as the pole. The
+equator, being perpendicular to the axis of rotation,
+of course shifts also, and therefore the
+latitude of a place fixed on the solid earth varies.
+During the whole course of the nutation, the
+earth's axis of rotation is always directed towards
+the same point in the heavens, and therefore the
+angle between the celestial pole and the vertical
+or plumb-line at the place of observation must
+oscillate about some mean value; the period of
+the oscillation is that of the earth's nutation.
+This movement is called a ``free'' nutation,
+because it is independent of the action of external
+forces.
+
+There are, besides, other nutations resulting
+from the attractions of the moon and sun on the
+protuberant matter at the equator, and from the
+same cause there is a slow shift in space of the
+earth's axis, called the precession. These movements
+are said to be ``forced,'' because they are
+due to external forces. The measurements of
+the forced nutations and of the precession afford
+the means of determining the period of the free
+nutation, if it should exist. It has thus been
+concluded that if there is any variation in the
+latitude, it should be periodic in $305$~days; but
+\PageSep{253}
+only observation can decide whether there is
+such a variation of latitude or not.
+
+Until recently astronomers were so convinced
+of the sufficiency of this reasoning, that, when
+they made systematic examination of the latitudes
+of many observatories, they always searched
+for an inequality with a period of $305$~days.
+Some thought that they had detected it, but
+when the observations extended over long periods,
+it always seemed to vanish, as though what
+they had observed were due to the inevitable
+errors of observation. At length it occurred
+to Mr.~Chandler to examine the observations
+\index{Chandler, free nutation of earth, and variation of latitude|(}%
+of latitude without any prepossession as to the
+period of the inequality. By the treatment of
+enormous masses of observation, he came to the
+conclusion that there is really such an inequality,
+but that the period is $427$~days instead of $305$~days.
+He also found other inequalities in the
+motion of the axis of rotation, of somewhat
+obscure origin, and of which I have no occasion
+to say more.\footnote
+ {They are perhaps due to the unequal melting of polar
+ ice and unequal rainfall in successive years. These irregular
+ variations in the latitude are such that some astronomers are
+ still skeptical as to the reality of Chandler's nutation, and think
+ that it will perhaps be found to lose its regularly rhythmical
+ character in the future.}
+
+The question then arises as to how the theory
+can be so amended as to justify the extension of
+the period of nutation. It was, I believe, Newcomb,
+\PageSep{254}
+\index{Newcomb, S., theoretical explanation of Chandler's nutation}%
+of the United States Naval Observatory,
+who first suggested that the explanation is to be
+sought in the fact that the axis of rotation is an
+axis of centrifugal repulsion, and that when it
+shifts, the distribution of centrifugal force is
+changed with reference to the solid earth, so
+that the earth is put into a state of stress, to
+which it must yield like any other elastic body.
+The strain or yielding consequent on this stress
+must be such as to produce a slight variability
+in the position of the equatorial protuberance
+with reference to places fixed on the earth.
+Now the period of $305$~days was computed on
+the hypothesis that the position of the equatorial
+protuberance is absolutely invariable, but
+periodic variations of the earth's figure would
+operate so as to lengthen the period of the free
+nutation, to an extent dependent on the average
+elasticity of the whole earth.
+\index{Elasticity of earth}%
+
+Mr.~Chandler's investigation demanded the
+utmost patience and skill in marshaling large
+masses of the most refined astronomical observations.
+His conclusions are not only of the
+greatest importance to astronomy, but they also
+give an indication of the amount by which the
+solid earth is capable of yielding to external
+forces. It would seem that the average stiffness
+of the whole earth must be such that it yields a
+little less than if it were made of steel.\footnote
+ {Mr.~S. S. Hough, p.~338 of the paper referred to in the list
+\index{Hough, S. S.!rigidity of earth}%
+ of authorities at the end of the chapter.}
+But
+\PageSep{255}
+\index{Elasticity of earth}%
+the amount by which the surface yields remains
+unknown, because we are unable to say what
+proportion of the aggregate change is superficial
+and what is deep-seated. It is, however, certain
+that the movements are excessively small, because
+the circle described by the extremity of
+the earth's axis of rotation, about the point on
+the earth which we call the pole, has a radius of
+only fifteen feet.
+
+It is easily intelligible that as the axis of
+rotation shifts in the earth, the oceans will tend
+to swash about, and that a sort of tide will be
+generated. If the displacement of the axis were
+considerable, whole continents would be drowned
+by a gigantic wave, but the movement is so
+small that the swaying of the ocean is very
+feeble. Two investigators have endeavored to
+detect an oceanic tide with a period of $427$~days;
+they are Dr.~Bakhuyzen of Leyden and
+\index{Bakhuyzen on tide due to variation of latitude}%
+Mr.~Christie of the United States Coast Survey.
+\index{Christie, A. S., tide due to variation of latitude}%
+The former considered observations of sea-level
+on the coasts of Holland, the latter those on the
+coasts of the United States; and they both conclude
+that the sea-level undergoes a minute
+variability with a period of about $430$~days. A
+similar investigation is now being prosecuted by
+the Tidal Survey of India, and as the Indian
+tidal observations are amongst the best in the
+world, we may hope for the detection of this
+minute tide in the Indian Ocean also.
+\PageSep{256}
+\index{Earth and moon!rigidity of|(}%
+\index{Rigidity of earth|(}%
+
+The inequality in water level is so slight and
+extends over so long a period that its measurement
+cannot yet be accepted as certain. The
+mean level of the sea is subject to slight irregular
+variations, which are probably due to unequal
+rainfall and unequal melting of polar ice in
+successive years. But whatever be the origin of
+these irregularities they exceed in magnitude the
+one to be measured. The arithmetical processes,
+employed to eliminate the ordinary tides and the
+irregular variability, will always leave behind
+some residual quantities, and therefore the examination
+of a tidal record will always apparently
+yield an inequality of any arbitrary period whatever.
+It is only when several independent determinations
+yield fairly consistent values of the
+magnitude of the rise and fall and of the moment
+of high water, that we can feel confidence
+in the result. Now although the reductions of
+Bakhuyzen and Christie are fairly consistent
+\index{Bakhuyzen on tide due to variation of latitude}%
+\index{Christie, A. S., tide due to variation of latitude}%
+with one another, and with the time and height
+suggested by Chandler's nutation, yet it is by no
+means impossible that accident may have led to
+this agreement. The whole calculation must
+therefore be repeated for several places and at
+several times, before confidence can be attained
+in the detection of this latitudinal tide.
+
+\TB
+
+The prolongation of the period of Chandler's
+nutation from $305$ to $427$~days seems to indicate
+\index{Latitude!periodic variations of|)}%
+\index{Nutation!Chandler's|)}%
+\index{Variation of latitude|)}%
+\PageSep{257}
+that our planet yields to external forces, and we
+naturally desire to learn more on so interesting
+a subject. Up to fifty years ago it was generally
+held that the earth was a globe of molten
+matter covered by a thin crust. The ejection of
+lava from volcanoes and the great increase of
+temperature in mines seemed to present evidence
+in favor of this belief. But the geologists and
+physicists of that time seemed not to have perceived
+that the inference might be false, if great
+pressure is capable of imparting rigidity to matter
+at a very high temperature, because the interior
+of the earth might then be solid although
+very hot. Now it has been proved experimentally
+that rock expands in melting, and a physical
+corollary from this is that when rock is under
+great pressure a higher temperature is needed to
+melt it than when the pressure is removed. The
+pressure inside the earth much exceeds any that
+can be produced in the laboratory, and it is uncertain
+up to what degree of increase of pressure
+the law of the rise of the temperature of
+melting would hold good; but there can be no
+doubt that, in so far as experiments in the laboratory
+can be deemed applicable to the conditions
+prevailing in the interior of the earth, they
+tend to show that the matter there is not improbably
+solid.
+
+But Lord Kelvin reinforces this argument
+\index{Kelvin, Lord!rigidity of earth|(}%
+from another point of view. Rock in the solid
+\index{Chandler, free nutation of earth, and variation of latitude|)}%
+\PageSep{258}
+condition is undoubtedly heavier than when it is
+molten. Now the solidified crust on the surface
+of a molten planet must have been fractured
+many times during the history of the planet,
+and the fragments would sink through the liquid,
+and thus build up a solid nucleus. It will
+be observed that this argument does not repose
+on the rise in the melting temperature of rock
+through pressure, although it is undoubtedly
+reinforced thereby.
+
+Hopkins was, I think, the first to adduce arguments
+\index{Hopkins on rigidity of earth}%
+of weight in favor of the earth's solidity.
+He examined the laws of the precession and
+nutation of a rigid shell inclosing liquid, and
+found that the motion of such a system would
+differ to a marked degree from that of the earth.
+From this he concluded that the interior of the
+earth was not liquid.
+
+Lord Kelvin has pointed out that although
+Hopkins's investigation is by no means complete,
+yet as he was the first to show that the
+motion of the earth as a whole affords indications
+of the condition of the interior, an important
+share in the discovery of the solidity of the
+earth should be assigned to him. Lord Kelvin
+then resumed Hopkins's work, and showed that
+if the liquid interior of the planet were inclosed
+in an unyielding crust, a very slight departure
+from perfect sphericity in the shell would render
+the motion of the system almost identical with
+\PageSep{259}
+\index{Height of tide!reduced by elastic yielding of earth}%
+that of a globe solid from centre to surface,
+although this would not be the case with the
+more rapid nutations. A yet more important
+deficiency in Hopkins's investigation is that he
+\index{Hopkins on rigidity of earth}%
+did not consider that, unless the crust were more
+rigid than the stiffest steel, it would yield to the
+surging of the imprisoned liquid as freely as
+india-rubber; and, besides, that if the crust
+yielded freely, the precession and nutations of
+the whole mass would hardly be distinguishable
+from those of a solid globe. Hopkins's argument,
+as thus amended by Lord Kelvin, leads
+to one of two alternatives: either the globe is
+solid throughout, or else the crust yields with
+nearly the same freedom to external forces as
+though it were liquid.
+
+We have now to show that the latter hypothesis
+is negatived by other considerations. The
+oceanic tides, as we perceive them, consist in a
+motion of the water relatively to the land. Now
+if the solid earth were to yield to the tidal forces
+with the same freedom as the superjacent sea,
+the cause for the relative movement of the sea
+would disappear. And if the solid yielded to
+some extent, the apparent oceanic tide would be
+proportionately diminished. The very existence
+of tides in the sea, therefore, proves at least that
+the land does not yield with perfect freedom.
+
+Lord Kelvin has shown that the oceanic tides,
+on a globe of the same rigidity as that of glass,
+\PageSep{260}
+would only have an apparent range of two fifths
+of those on a perfectly rigid globe; whilst, if
+the rigidity was equal to that of steel, the fraction
+of diminution would be two thirds. I have
+myself extended his argument to the hypothesis
+that the earth may be composed of a viscous
+material, which yields slowly under the application
+of continuous forces, and also to the hypothesis
+of a material which shares the properties
+of viscosity and rigidity, and have been led to
+\index{Rigidity of earth|)}%
+analogous conclusions.
+
+The difficulty of the problem of oceanic tides
+is so great that we cannot say how high the tides
+would be if the earth were absolutely rigid, but
+Lord Kelvin is of opinion that they certainly
+\index{Kelvin, Lord!rigidity of earth|)}%
+would not be twice as great as they are, and he
+concludes that the earth possesses a greater average
+stiffness than that of glass, although perhaps
+not greater than that of steel. It is proper to
+add that the validity of this argument depends
+principally on the observed height of an inequality
+of sea level with a period of a fortnight. This
+is one of the partial tides of the third kind, which
+I described in \Ref{Chapter}{XI}.\ as practically unimportant,
+and did not discuss in detail. The value
+of this inequality in the present argument is due
+to the fact that it is possible to form a much
+closer estimate of its magnitude on a rigid earth
+\index{Earth and moon!rigidity of|)}%
+than in the case of the semidiurnal and diurnal
+tides.
+\PageSep{261}
+\index{Darwin, G. H.!rigidity of earth}%
+\index{Earthquakes!shock perceptible at great distance}%
+
+It may ultimately be possible to derive further
+indications concerning the physical condition of
+the inside of the earth from the science of seismology.
+The tremor of an earthquake has frequently
+been observed instrumentally at an enormous
+distance from its origin; as, for example,
+when the shock of a Japanese earthquake is
+perceived in England.
+
+The vibrations which are transmitted through
+the earth are of two kinds. The first sort of wave
+is one in which the matter through which it passes
+is alternately compressed and dilated; it may be
+described as a wave of compression. In the
+second sort the shape of each minute portion of
+the solid is distorted, but the volume remains
+unchanged, and it may be called a wave of distortion.
+These two vibrations travel at different
+speeds, and the compressional wave outpaces
+the distortional one. Now the first sign of a
+distant earthquake is that the instrumental record
+shows a succession of minute tremors.
+These are supposed to be due to waves of compression,
+and they are succeeded by a much
+more strongly marked disturbance, which, however,
+lasts only a short time. This second phase
+in the instrumental record is supposed to be due
+to the wave of distortion.
+
+If the natures of these two disturbances are
+correctly ascribed to their respective sources, it
+is certain that the matter through which the vibration
+\PageSep{262}
+\index{Darwin, G. H.!rigidity of earth}%
+has passed was solid. For, although a
+compressional wave might be transmitted without
+much loss of intensity, from a solid to a
+liquid and back again to a solid, as would have
+to be the case if the interior of the earth is molten,
+yet this cannot be true of the distortional
+wave. It has been supposed that vibrations due
+to earthquakes pass in a straight line through
+the earth; if then this could be proved, we
+should know with certainty that the earth is
+solid, at least far down towards its centre.
+
+Although there are still some---principally
+amongst the geologists---who believe in the existence
+of liquid matter immediately under the
+solid crust of the earth,\footnote
+ {See the Rev.~Osmond Fisher's \Title{Physics of the Earth's Crust}.}
+\index{Fisher, Osmond, on molten interior of earth}%
+yet the arguments which
+I have sketched appear to most men of science
+conclusive against such belief.
+
+\begin{Authorities}
+Mr.~S.~C. Chandler's investigations are published in the ``Astronomical
+Journal,'' vol.~11 and following volumes. A summary
+is contained in ``Science,'' May~3, 1895.
+
+R.~S. Woodward, \Title{Mechanical Interpretation of the Variations of
+\index{Woodward on variation of latitude}%
+Latitude}, ``Ast.\ Journ.'' vol.~15, May,~1895.
+
+Simon Newcomb, \Title{On the Dynamics of the Earth's Rotation},
+``Monthly Notices of the R.~Astron.\ Soc.,'' vol.~52 (1892),
+p.~336.
+
+S.~S. Hough, \Title{The Rotation of an Elastic Spheroid}, ``Philosoph.\
+\index{Hough, S. S.!Chandler's nutation}%
+Trans.\ of the Royal Society,'' A.~1896, p.~319. He indicates a
+slight oversight on the part of Newcomb.
+
+H.~G. van~de Sande Bakhuyzen, \Title{Ueber die Aenderung der Polhoehe},
+``Astron.\ Nachrichten,'' No.~3261.
+\PageSep{263}
+
+A.~S. Christie, \Title{The Latitude-variation Tide}, ``Phil.\ Soc.\ of
+Washington, Bulletin,'' vol.~12 (1895), p.~103.
+
+Lord Kelvin, in Thomson and Tait's ``Natural Philosophy,''
+on the Rigidity of the Earth; and ``Popular Lectures,'' vol.~3.
+
+G.~H. Darwin, \Title{Bodily Tides of Viscous and Semi-elastic Spheroids,~\&c.},
+``Philosoph.\ Trans.\ of the Royal Society,'' Part.~I.
+1879.
+\end{Authorities}
+\PageSep{264}
+
+
+%[** TN: Footnote mark handled by \Chapter logic]
+\Chapter{XVI}
+{Tidal Friction}
+\footnotetext{A considerable portion of this and of the succeeding chapter
+ appeared as an article in \Title{The Atlantic Monthly} for April,~1898.}
+
+\First{The} fact that the earth, the moon, and the
+\index{Friction of tides|(}%
+planets are all nearly spherical proves that in
+early times they were molten and plastic, and
+assumed their present round shape under the
+influence of gravitation. When the material of
+which any planet is formed was semi-liquid
+through heat, its satellites, or at any rate the
+sun, must have produced tidal oscillations in the
+molten rock, just as the sun and moon now produce
+the tides in our oceans.
+
+Molten rock and molten iron are rather sticky
+or viscous substances, and any movement which
+agitates them must be subject to much friction.
+Even water, which is a very good lubricant, is
+not entirely free from friction, and so our present
+oceanic tides must be influenced by fluid
+friction, although to a far less extent than the
+molten solid just referred to. Now, all moving
+systems which are subject to friction gradually
+come to rest. A train will run a long way when
+the steam is turned off, but it stops at last, and
+\PageSep{265}
+a fly-wheel will continue to spin for only a limited
+time. This general law renders it certain that
+the friction of the tide, whether it consists in the
+swaying of molten lava or of an ocean, must be
+retarding the rotation of the planet, or at any
+rate retarding the motion of the system in some
+way.
+
+It is the friction upon its bearings which brings
+a fly-wheel to rest; but as the earth has no bearings,
+it is not easy to see how the friction of the
+tidal wave, whether corporeal or oceanic, can
+tend to stop its rate of rotation. The result
+must clearly be brought about, in some way, by
+the interaction between the moon and the earth.
+Action and reaction must be equal and opposite,
+and if we are correct in supposing that the friction
+of the tides is retarding the earth's rotation,
+there must be a reaction upon the moon which
+must tend to hurry her onwards. To give a
+homely illustration of the effects of reaction, I
+may recall to mind how a man riding a high
+bicycle, on applying the brake too suddenly, was
+thrown over the handles. The desired action
+was to stop the front wheel, but this could not
+be done without the reaction on the rider, which
+sometimes led to unpleasant consequences.
+
+The general conclusion as to the action and
+reaction due to tidal friction is of so vague a
+character that it is desirable to consider in detail
+how they operate.
+\PageSep{266}
+
+The circle in \fig{36} is supposed to represent
+the undisturbed shape of the planet, which rotates
+in the direction of the curved arrow. A portion
+of the orbit of the satellite is indicated by part
+\Figure{36}{Frictionally retarded Tide}{png}
+of a circle, and the direction of its motion is
+shown by an arrow. I will first suppose that the
+water lying on the planet, or the molten rock of
+which it is formed, is a perfect lubricant devoid
+of friction, and that at the moment represented
+in the figure the satellite is at~$M'$. The fluid will
+then be distorted by the tidal force until it assumes
+the egg-like shape marked by the ellipse,
+projecting on both sides beyond the circle. It
+will, however, be well to observe that if this figure
+represents an ocean, it must be a very deep
+one, far deeper than those which actually exist
+on the earth; for we have seen that it is only in
+deep oceans that the high water stands underneath
+and opposite to the moon; whereas in
+shallow water it is low water where we should
+\PageSep{267}
+naturally expect high water. Accepting the hypothesis
+that the high tide is opposite to the
+moon, and supposing that the liquid is devoid of
+friction, the long axis of the egg is always directed
+straight towards the satellite~$M'$, and the
+liquid maintains a continuous rhythmical movement,
+so that as the planet rotates and the satellite
+revolves, it always maintains the same shape
+and attitude towards the satellite.
+
+But when, as in reality, the liquid is subject to
+friction, it gets belated in its rhythmical rise and
+fall, and the protuberance is carried onward by
+the rotation of the planet beyond its proper
+place. In order to make the same figure serve
+for this condition, I set the satellite backward to~$M$;
+for this amounts to just the same thing, and
+is less confusing than redrawing the protuberance
+in its more advanced position. The planet
+then constantly maintains this shape and attitude
+with regard to the satellite, and the interaction
+between the two will be the same as though the
+planet were solid, but continually altering its
+shape.
+
+We have now to examine what effects must
+follow from the attraction of the satellite on an
+egg-shaped planet, when the two constantly
+maintain the same attitude relatively to each
+other. It will make the matter somewhat easier
+of comprehension if we replace the tidal protuberances
+by two particles of equal masses, one at~$P$,
+\PageSep{268}
+\index{Earth and moon!rotation retarded by tidal friction}%
+\index{Retardation of earth's rotation}%
+and the other at~$P'$. If the masses of these
+particles be properly chosen, so as to represent
+the amount of matter in the protuberances, the
+proposed change will make no material difference
+in the action.
+
+The gravitational attraction of the satellite is
+greater on bodies which are near than on those
+which are far, and accordingly it attracts the
+particle~$P$ more strongly than the particle~$P'$. It
+is obvious from the figure that the attraction on~$P$
+must tend to stop the planet's rotation, whilst
+\index{Rotation!retarded by tidal friction}%
+that on~$P'$ must tend to accelerate it. If a man
+pushes equally on the two pedals of a bicycle,
+the crank has no tendency to turn, and besides
+there are dead points in the revolution where
+pushing and pulling have no effect. So also in
+the astronomical problem, if the two attractions
+were exactly equal, or if the protuberances were
+at a dead point, there would be no resultant effect
+on the rotation of the planet. But it is
+obvious that here the retarding pull is stronger
+than the accelerating one, and that the set of the
+protuberances is such that we have passed the
+dead point. It follows from this that the primary
+effect of fluid friction is to throw the tidal
+protuberance forward, and the secondary effect
+is to retard the planet's rotation.
+
+It has been already remarked that this figure is
+drawn so as to apply only to the case of corporeal
+tides or to those of a very deep ocean. If
+\PageSep{269}
+\index{Moon and earth!retardation of motion by tidal friction}%
+the ocean were shallow and frictionless, it would
+be low water under and opposite to the satellite.
+If then the effect of friction were still to throw
+the protuberances forward, the rotation of the
+planet would be accelerated instead of retarded.
+But in fact the effect of fluid friction in a shallow
+ocean is to throw the protuberances backward,
+and a similar figure, drawn to illustrate such a
+displacement of the tide, would at once make it
+clear that here also tidal friction will lead to the
+retardation of the planet's rotation. Henceforth
+then I shall confine myself to the case illustrated
+by~\fig{36}.
+
+Action and reaction are equal and opposite,
+and if the satellite pulls at the protuberances,
+they pull in return on the satellite. The figure
+shows that the attraction of the protuberance~$P$
+tends in some measure to hurry the satellite onward
+in its orbit, whilst that of~$P'$ tends to retard
+it. But the attraction of~$P$ is stronger than that
+of~$P'$, and therefore the resultant of the two is a
+force tending to carry the satellite forward in the
+direction of the arrow.
+
+If a stone be whirled at the end of an elastic
+string, a retarding force, such as the friction of
+the air, will cause the string to shorten, and an
+accelerating force will make it lengthen. In the
+same way the satellite, being as it were tied to
+the planet by the attraction of gravitation, when
+subjected to an onward force, recedes from the
+\PageSep{270}
+\index{Moon and earth!retardation of motion by tidal friction}%
+planet, and moves in a spiral curve at ever increasing
+distances. The time occupied by the
+satellite in making a circuit round the planet is
+prolonged, and this lengthening of the periodic
+time is not merely due to the lengthening of the
+arc described by it, but also to an actual retardation
+of its velocity. It appears paradoxical that
+the effect of an accelerating force should be a
+retardation, but a consideration of the mode in
+which the force operates will remove the paradox.
+The effect of the tangential accelerating
+force on the satellite is to make it describe an
+increasing spiral curve. Now if the reader will
+draw an exaggerated figure to illustrate part of
+such a spiral orbit, he will perceive that the central
+force, acting directly towards the planet,
+must operate in some measure to retard the velocity
+of the satellite. The central force is very
+great compared with the tangential force due to
+the tidal friction, and therefore a very small
+fraction of the central force may be greater than
+the tangential force. Although in a very slowly
+increasing spiral the fraction of the central force
+productive of retardation is very small, yet it is
+found to be greater than the tangential accelerating
+force, and thus the resultant effect is a
+retardation of the satellite's velocity.
+
+The converse case where a retarding force results
+in increase of velocity will perhaps be more
+intelligible, as being more familiar. A meteorite,
+\PageSep{271}
+rushing through the earth's atmosphere, moves
+faster and faster, because it gains more speed
+from the attraction of gravity than it loses by the
+friction of the air.
+
+Now let us apply these ideas to the case of the
+earth and the moon. A man standing on the
+planet, as it rotates, is carried past places where
+the fluid is deeper and shallower alternately; at
+the deep places he says that it is high tide, and
+at the shallow places that it is low tide. In \fig{36}
+it is high tide when the observer is carried
+past~$P$. Now it was pointed out that when there
+is no fluid friction we must put the moon at~$M'$,
+but when there is friction she must be at~$M$.
+Accordingly, if there is no friction it is high tide
+when the moon is over the observer's head, but
+when there is friction the moon has passed his
+zenith before he reaches high tide. Hence he
+would remark that fluid friction retards the time
+of high tide.
+
+A day is the name for the time in which the
+earth rotates once, and a month for the time in
+which the moon revolves once. Then since tidal
+friction retards the earth's rotation and the
+moon's revolution, we may state that both the
+day and the month are being lengthened, and
+that these results follow from the retardation of
+the time of high tide.
+
+It must also be noted that the spiral in which
+the moon moves is an increasing one, so that her
+\PageSep{272}
+\index{Assyrian records of eclipses}%
+distance from the earth also increases. These
+are absolutely certain and inevitable results of
+the mechanical interaction of the two bodies.
+
+At the present time the rates of increase of
+the day and month are excessively small, so that
+it has not been found possible to determine them
+with any approach to accuracy. It may be well
+to notice in passing that if the rate of either increase
+of element were determinable, that of the
+other would be deducible by calculation.
+
+The extreme slowness of the changes within
+historical times is established by the early records
+in Greek and Assyrian history of eclipses of the
+\index{Eclipses, ancient, and earth's rotation}%
+\index{Greek!records of ancient eclipses}%
+sun, which occurred on certain days and in certain
+places. Notwithstanding the changes in the
+calendar, it is possible to identify the day according
+to our modern reckoning, and the identification
+of the place presents no difficulty.
+Astronomy affords the means of calculating the
+exact time and place of the occurrence of an
+eclipse even three thousand years ago, on the
+supposition that the earth spun at the same rate
+then as now, and that the complex laws governing
+the moon's motion are unchanged.
+
+The particular eclipse referred to in history is
+known, but any considerable change in the
+earth's rotation and in the moon's position would
+have shifted the position of visibility on the
+earth from the situation to which modern computation
+would assign it. Most astronomical
+\PageSep{273}
+observations would be worthless if the exact time
+of the occurrence were uncertain, but in the
+case of eclipses the place of observation affords
+\index{Eclipses, ancient, and earth's rotation}%
+just that element of precision which is otherwise
+wanting. As, then, the situations of the ancient
+eclipses agree fairly well with modern computations,
+we are sure that there has been no great
+change within the last three thousand years,
+either in the earth's rotation or in the moon's
+motion. There is, however, a small outstanding
+discrepancy which indicates that there has been
+some change. But the exact amount of change
+involves elements of uncertainty, because our
+knowledge of the laws of the moon's motion is
+not yet quite accurate enough for the absolutely
+perfect calculation of eclipses which occurred
+many centuries ago. In this way, it is known
+that within historical times the retardation of the
+earth's rotation and the recession of the moon
+have been at any rate very slow.
+
+It does not, however, follow from this that
+the changes have always been equally slow; indeed,
+it may be shown that the efficiency of tidal
+friction increases with great rapidity as we bring
+the tide-generating satellite nearer to the planet.
+
+It has been shown in \Ref{Chapter}{V}.\ that the intensity
+of tide-generating force varies as the inverse
+cube of the distance between the moon and
+the earth, so that if the moon's distance were
+reduced successively to $\frac{1}{2}$,~$\frac{1}{3}$,~$\frac{1}{4}$, of its original distance,
+\PageSep{274}
+the force and the tide generated by it
+would be multiplied $8$,~$27$,~$64$ times. But the
+efficiency of tidal friction increases far more rapidly
+than this, because not only is the tide itself
+augmented, but also the attraction of the moon.
+In order to see how these two factors will coöperate,
+let us begin by supposing that the
+height of the tide remains unaffected by the approach
+or retrogression of the moon. Then the
+same line of argument, which led to the conclusion
+that tide-generating force varies inversely as
+the cube of the distance, shows that the action
+of the moon on protuberances of definite magnitude
+must also vary inversely as the cube of the
+distance. But the height of the tide is not in
+fact a fixed quantity, but varies inversely as the
+cube of the distance, so that when account is
+taken both of the augmentation of the tide and
+of the increased attraction of the moon, it follows
+that the tidal retardation of the earth's rotation
+must vary as the inverse sixth power of
+the distance. Now since the sixth power of~$2$ is~$64$,
+the lunar tidal friction, with the moon at
+half her present distance, would be $64$~times as
+efficient as at present. Similarly, if her distance
+were diminished to a third and a quarter of what
+it is, the tidal friction would act with $729$ and
+$4,096$~times its present strength. Thus, although
+the action may be insensibly slow now, it must
+have gone on with much greater rapidity when
+the moon was nearer to us.
+\PageSep{275}
+
+There are many problems in which it would
+be very difficult to follow the changes according
+to the times of their occurrence, but where it is
+possible to banish time from consideration, and
+to trace the changes themselves, in due order,
+without reference to time. In the sphere of
+common life, we know the succession of stations
+which a train must pass between London and
+Edinburgh, although we may have no time-table.
+This is the case with our astronomical
+problem; for although we have no time-table,
+yet the sequence of the changes in the system
+can be traced accurately.
+
+Let us then banish time, and look forward to
+the ultimate outcome of the tidal interaction of
+the moon and earth. The day and the month
+\index{Day, change in length of, under tidal friction}%
+\index{Month, change in, under tidal friction|(}%
+are lengthening at relative rates which are calculable,
+although the absolute rates in time are
+unknown. It will suffice for a general comprehension
+of the problem to know that the present
+rate of increase of the day is much more rapid
+than that of the month, and that this will hold
+good in the future. Thus, the number of rotations
+of the earth in the interval comprised in
+one revolution of the moon diminishes; or, in
+other words, the number of days in the month
+diminishes, although the month itself is longer
+than at present. For example, when the day
+shall be equal in length to two of our actual
+days, the month may be as long as thirty-seven
+\PageSep{276}
+of our days, and then the earth will spin round
+only about eighteen times in the month.
+
+This gradual change in the day and month
+\index{Day, change in length of, under tidal friction}%
+proceeds continuously until the duration of a
+rotation of the earth is prolonged to fifty-five of
+our present days. At the same time the month,
+or the time of revolution of the moon round the
+earth, will also occupy fifty-five of our days.
+Since the month here means the period of the
+return of the moon to the same place among the
+stars, and since the day is to be estimated in
+the same way, the moon must then always face
+the same part of the earth's surface, and the
+two bodies must move as though they were
+united by a bar. The outcome of the lunar
+tidal friction will therefore be that the moon
+and the earth go round as though locked together,
+in a period of fifty-five of our present
+days, with the day and the month identical in
+length.
+
+Now looking backward in time, we find the
+day and the month shortening, but the day
+changing more rapidly than the month. The
+earth was therefore able to complete more revolutions
+in the month, although that month was
+itself shorter than it is now. We get back in
+fact to a time when there were $29$~rotations of
+the earth in a month instead of~$27\frac{1}{3}$, as at present.
+This epoch is a sort of crisis in the history
+of the moon and the earth, for it may be proved
+\PageSep{277}
+that there never could have been more than $29$~days
+in the month. Earlier than this epoch, the
+days were fewer than~$29$, and later fewer also.
+Although measured in years, this epoch in the
+earth's history must be very remote, yet when we
+contemplate the whole series of changes it must
+be considered as a comparatively recent event.
+In a sense, indeed, we may be said to have passed
+recently through the middle stage of our history.
+
+Now, pursuing the series of changes further
+back than the epoch when there was the maximum
+number of days in the month, we find the
+earth still rotating faster and faster, and the
+moon drawing nearer and nearer to the earth,
+and revolving in shorter and shorter periods.
+But a change has now supervened, so that the
+rate at which the month is shortening is more
+rapid than the rate of change in the day. Consequently,
+the moon now gains, as it were, on
+the earth, which cannot get round so frequently
+in the month as it did before. In other words,
+the number of days in the month declines from
+the maximum of~$29$, and is finally reduced to
+one. When there is only one day in the month,
+\index{Month, change in, under tidal friction|)}%
+the earth and the moon go round at the same
+rate, so that the moon always looks at the same
+side of the earth, and so far as concerns the
+motion they might be fastened together by a
+rigid bar.
+
+This is the same conclusion at which we arrived
+\PageSep{278}
+with respect to the remote future. But
+the two cases differ widely; for whereas in the
+future the period of the common rotation will
+be $55$~of our present days, in the past we find
+the two bodies going round each other in between
+three and five of our present hours. A
+satellite revolving round the earth in so short a
+period must almost touch the earth's surface.
+The system is therefore traced until the moon
+nearly touches the earth, and the two go round
+each other like a single solid body in about three
+to five hours.
+
+The series of changes has been traced forward
+and backward from the present time, but it will
+make the whole process more intelligible, and
+the opportunity will be afforded for certain further
+considerations, if I sketch the history again
+\index{History!of earth and moon|(}%
+in the form of a continuous narrative.
+
+Let us imagine a planet attended by a satellite
+which revolves so as nearly to touch its surface,
+and continuously to face the same side of the
+planet's surface. If now, for some reason, the
+satellite's month comes to differ very slightly
+from the planet's day, the satellite will no longer
+continuously face the same side of the planet,
+but will pass over every part of the planet's
+equator in turn. This is the condition necessary
+for the generation of tidal oscillations in the
+planet, and as the molten lava, of which we
+suppose it to be formed, is a sticky or viscous
+\PageSep{279}
+fluid, the tidal oscillations must be subject to
+friction. Tidal friction will then begin to do its
+work, but the result will be very different according
+as the satellite revolves a little faster or
+a little slower than the planet. If it revolves a
+little faster, so that the month is shorter than
+the day, we have a condition not contemplated
+in~\fig{36}; it is easy to see, however, that as
+the satellite is always leaving the planet behind
+it, the apex of the trial protuberance must be
+directed to a point behind the satellite in its
+orbit. In this case the rotation of the planet
+must be \DPtypo{acclerated}{accelerated} by the tidal friction, and the
+satellite will be drawn inward towards the planet,
+into which it must ultimately fall. In the application
+of this theory to the earth and moon, it
+is obvious that the very existence of the moon
+negatives the hypothesis that the initial month
+was even infinitesimally shorter than the day.
+We must then suppose that the moon revolved
+a little more slowly than the earth rotated. In
+this case the tidal friction would retard the
+earth's rotation, and force the moon to recede
+from the earth, and so perform her orbit more
+slowly. Accordingly, the primitive day and the
+primitive month lengthen, but the month increases
+much more rapidly than the day, so that
+the number of days in a month increases. This
+proceeds until that number reaches a maximum,
+which in the case of our planet is about~$29$.
+\PageSep{280}
+\index{Instability!nature of dynamical, and initial of moon's motion|(}%
+\index{Stability!nature of dynamical}%
+
+After the epoch of the maximum number of
+days in the month, the rate of change in the
+length of the day becomes less rapid than that
+in the length of the month; and although both
+periods increase, the number of days in the
+month begins to diminish. The series of
+changes then proceeds until the two periods
+come again to an identity, when we have the
+earth and the moon as they were at the beginning,
+revolving in the same period, with the
+moon always facing the same side of the earth.
+But in her final condition the moon will be a
+long way off the earth instead of being quite
+close to it.
+
+Although the initial and final states resemble
+each other, yet they differ in one respect which
+is of much importance, for in the initial condition
+the motion is unstable, whilst finally it is
+stable. The meaning of this is, that if the
+moon were even infinitesimally disturbed from
+the initial mode of motion, she would necessarily
+either fall into the planet, or recede therefrom,
+and it would be impossible for her to continue
+to move in that neighborhood. She is unstable
+in the same sense in which an egg when balanced
+on its point is unstable; the smallest mote
+of dust will upset it, and practically it cannot
+stay in that position. But the final condition
+resembles the case of the egg lying on its side,
+which only rocks a little when we disturb it.
+\PageSep{281}
+\index{Stability!nature of dynamical}%
+So if the moon were slightly disturbed from her
+final condition, she would continue to describe
+very nearly the same path round the earth, and
+would not assume some entirely new form of
+orbit.
+
+It is by methods of rigorous argument that
+the moon is traced back to the initial unstable
+condition when she revolved close to the earth.
+But the argument here breaks down, and calculation
+is incompetent to tell us what occurred
+before, and how she attained that unstable mode
+of motion. If we were to find a pendulum
+swinging in a room, where we knew that it had
+been undisturbed for a long time, we might, by
+observing its velocity and allowing for the resistance
+of the air, conclude that at some previous
+moment it had just been upside down, but
+calculation could never tell us how it had
+reached that position. We should of course
+feel confident that some one had started it.
+Now a similar hiatus must occur in the history
+of the moon, but it is not so easy to supply the
+missing episode. It is indeed only possible to
+speculate as to the preceding history.
+
+But there is some basis for our speculation;
+for I say that if a planet, such as the earth,
+made each rotation in three hours, it would very
+nearly fly to pieces. The attraction of gravity
+would be barely strong enough to hold it together,
+just as the cohesive strength of iron is
+\PageSep{282}
+\index{Forced oscillation!due to solar tide, possibly related to birth of moon|(}%
+\index{Moon and earth!origin of}%
+insufficient to hold a fly-wheel together if it is
+spun too fast. There is, of course, an important
+distinction between the case of the ruptured
+fly-wheel and the supposed break-up of the
+earth; for when a fly-wheel breaks, the pieces
+are hurled apart as soon as the force of cohesion
+fails, whereas when a planet breaks up through
+too rapid rotation, gravity must continue to
+hold the pieces together after they have ceased
+to form parts of a single body.
+
+Hence we have grounds for conjecturing that
+the moon is composed of fragments of the primitive
+planet which we now call the earth, which
+detached themselves when the planet spun very
+swiftly, and afterwards became consolidated. It
+surpasses the power of mathematical calculation
+to trace the details of the process of this rupture
+and subsequent consolidation, but we can hardly
+doubt that the system would pass through a
+period of turbulence, before order was reëstablished
+in the formation of a satellite.
+
+I have said above that rapid rotation was probably
+the cause of the birth of the moon, but it
+may perhaps not have been brought about by
+this cause alone. There are certain considerations
+which make it difficult to ascertain the
+initial common period of revolution of the moon
+and the earth with accuracy; it may lie between
+three and five hours. Now I think that such
+a speed might not quite suffice to cause the
+\index{Instability!nature of dynamical, and initial of moon's motion|)}%
+\PageSep{283}
+\index{Moon and earth!origin of}%
+primitive planet to break up. In \Ref{Chapter}{XVIII}.\
+we shall consider in greater detail the conditions
+under which a rotating mass of liquid would
+rupture, but for the present it may suffice to say
+that, where the rotating body is heterogeneous in
+density, like the earth, the exact determination
+of the limiting speed of rotation is not possible.
+Is there, then, any other cause which might coöperate
+with rapid rotation in producing rupture?
+I think there is such a cause, and, although
+we are here dealing with guesswork, I
+will hazard the suggestion.
+
+The primitive planet, before the birth of the
+moon, was rotating rapidly with reference to the
+sun, and it must therefore have been agitated by
+solar tides. In \Ref{Chapter}{IX}.\ it was pointed out
+that there is a general dynamical law which enables
+us to foresee the magnitude of the oscillations
+of a system under the action of external
+forces. That law depended on the natural or
+free period of the oscillation of the system when
+disturbed and left to itself, free from the intervention
+of external forces. We saw that the
+more nearly the periodic forces were timed to
+agree with the free period, the greater was the
+amplitude of the oscillations of the system. Now
+it is easy to calculate the natural or free period
+of the oscillation of a homogeneous liquid globe
+of the same density as the earth, namely, five
+and a half times as heavy as water; the period
+\PageSep{284}
+\index{Sun!possible influence of, in assisting birth of moon}%
+is found to be $1$~hour $34$~minutes. The heterogeneity
+of the earth introduces a complication of
+which we cannot take account, but it seems likely
+that the period would be from $1\frac{1}{2}$ to $2$~hours.
+The period of the solar semidiurnal tide is half a
+\index{Solar!possible effect of tide in assisting birth of moon}%
+day, and if the day were from $3$ to $4$ of our present
+hours the forced period of the tide would
+be in close agreement with the free period of
+oscillation.
+
+May we not then conjecture that as the rotation
+of the primitive earth was gradually reduced
+by solar tidal friction, the period of the solar tide
+was brought into closer and closer agreement
+with the free period, and that consequently the
+solar tide increased more and more in height?
+In this case the oscillation might at length become
+so violent that, in coöperation with the
+rapid rotation, it shook the planet to pieces, and
+that huge fragments were detached which ultimately
+became our moon.
+
+There is nothing to tell us whether this theory
+affords the true explanation of the birth of the
+moon, and I say that it is only a wild speculation,
+incapable of verification.
+
+But the truth or falsity of this speculation
+does not militate against the acceptance of the
+general theory of tidal friction, which, standing
+on the firm basis of mechanical necessity, throws
+much light on the history of the earth and the
+moon, and correlates the lengths of our present
+day and month.
+\index{Forced oscillation!due to solar tide, possibly related to birth of moon|)}%
+\PageSep{285}
+\index{Sun!possible influence of, in assisting birth of moon}%
+
+I have said above that the sequence of events
+has been stated without reference to the scale of
+time. It is, however, of the utmost importance
+\index{Time!requisite for evolution of moon}%
+to gain some idea of the time requisite for all the
+changes in the system. If millions of millions
+of years were necessary, the theory would have
+to be rejected, because it is known from other
+lines of argument that there is not an unlimited
+bank of time on which to draw. The uncertainty
+as to the duration of the solar system is
+\index{Solar!possible effect of tide in assisting birth of moon}%
+wide, yet we are sure that it has not existed for
+an almost infinite past.
+
+Now, although the actual time scale is indeterminate,
+it is possible to find the minimum time
+adequate for the transformation of the moon's
+orbit from its supposed initial condition to its
+present shape. It may be proved, in fact, that
+if tidal friction always operated under the conditions
+most favorable for producing rapid change,
+the sequence of events from the beginning until
+to-day would have occupied a period of between
+$50$ and $60$~millions of years. The actual period,
+of course, must have been much greater. Various
+lines of argument as to the age of the solar
+system have led to results which differ widely
+among themselves, yet I cannot think that the
+applicability of the theory is negatived by the
+magnitude of the period demanded. It may be
+that science will have to reject the theory in its
+full extent, but it seems unlikely that the ultimate
+\PageSep{286}
+\index{Moon and earth!rotation annulled by tidal friction and present libration}%
+verdict will be adverse to the preponderating
+influence of the tide in the evolution of our
+planet.
+
+\TB
+
+If this history be true of the earth and moon,
+\index{History!of earth and moon|)}%
+it should throw light on many peculiarities of the
+solar system. In the first place, a corresponding
+series of changes must have taken place in the
+moon herself. Once on a time the moon must
+have been molten, and the great extinct volcanoes
+revealed by the telescope are evidences of
+her primitive heat. The molten mass must have
+been semi-fluid, and the earth must have raised
+in it enormous tides of molten lava. Doubtless
+the moon once rotated rapidly on her axis, and
+the frictional resistance to her tides must have
+impeded her rotation. This cause must have
+\index{Rotation!of moon annulled by tidal friction}%
+added to the moon's recession from the earth,
+but as the moon's mass is only an eightieth part
+of that of the earth, the effect on the moon's
+orbit must have been small. The only point to
+which we need now pay attention is that the
+rate of her rotation was reduced. She rotated
+then more and more slowly until the tide solidified,
+and thenceforward and to the present day
+she has shown the same face to the earth. Kant
+\index{Kant!rotation of moon}%
+and Laplace in the last century, and Helmholtz
+\index{Helmholtz!on rotation of the moon}%
+\index{Laplace!on rotation of moon}%
+in recent times, have adduced this as the explanation
+of the fact that the moon always shows
+us the same face. Our theory, then, receives a
+\PageSep{287}
+striking confirmation from the moon; for, having
+ceased to rotate relatively to us, she has actually
+advanced to that condition which may be
+foreseen as the fate of the earth.
+
+The earth tide in the moon is now solidified
+so that the moon's equator is not quite circular,
+and the longer axis is directed towards the earth.
+Laplace has considered the action of the earth
+\index{Laplace!on rotation of moon}%
+on this solidified tide, and has shown that the
+moon must rock a little as she moves round the
+earth. In consequence of this rocking motion or
+libration of the moon, and also of the fact that
+her orbit is elliptic, we are able to see just a little
+more than half of the moon's surface.
+
+\TB
+
+Thus far I have referred in only one passage
+to the influence of solar tides, but these are of
+considerable importance, being large enough to
+cause the conspicuous phenomena of spring and
+neap tides. Now, whilst the moon is retarding
+the earth's rotation, the sun is doing so also.
+But these solar tides react only on the earth's
+motion round the sun, leaving the moon's motion
+round the earth unaffected. It might perhaps
+be expected that parallel changes in the
+earth's orbit would have proceeded step by step,
+and that the earth might be traced to an origin
+close to the sun. The earth's mass is less than $\frac{1}{300,000}$~part
+of the sun's, and the reactive effect on the
+earth's orbit round the sun is altogether negligible.
+\PageSep{288}
+It is improbable, in fact, that the year is,
+from this cause at any rate, longer by more than
+a few seconds than it was at the very birth of
+the solar system.
+
+Although the solar tides cannot have had any
+perceptible influence upon the earth's movement
+in its orbit, they will have affected the rotation
+of the earth to a considerable extent. Let us
+imagine ourselves transported to the indefinite
+future, when the moon's orbital period and the
+earth's diurnal period shall both be prolonged to
+$55$~of our present days. The lunar tide in the
+earth will then be unchanging, just as the earth
+tide in the moon is now fixed; but the earth will
+be rotating with reference to the sun, and, if
+there are still oceans on the earth, her rotation
+will be subject to retardation in consequence of
+the solar tidal friction. The day will then become
+longer than the month, whilst the moon
+will at first continue to revolve round the earth
+in $55$~days. Lunar tides will now be again generated,
+but as the motion of the earth will be
+very slow relatively to the moon, the oscillations
+will also be very slow, and subject to little friction.
+But that friction will act in opposition to
+the solar tides, and the earth's rotation will to
+some slight extent be assisted by the moon.
+The moon herself will slowly approach the earth,
+moving with a shorter period, and must ultimately
+fall back into the earth. We know that
+\PageSep{289}
+there are neither oceans nor atmosphere on the
+moon, but if there were such, the moon would
+have been subject to solar tidal friction, and
+would now be rotating slower than she revolves.
+
+%[** TN: Not hyperlinking chapter reference]
+\begin{Authorities}
+See the end of Chapter~XVII.
+\end{Authorities}
+\PageSep{290}
+
+
+\Chapter[Tidal Friction]{XVII}
+{Tidal Friction (Continued)}
+
+\First{It} has been shown in the last chapter that the
+prolongation of the day and of the month under
+the influence of tidal friction takes place in such
+a manner that the month will ultimately become
+longer than the day. Until recent times no case
+had been observed in the solar system in which
+a satellite revolved more rapidly than its planet
+rotated, and this might have been plausibly adduced
+as a reason for rejecting the actual efficiency
+of solar tidal friction in the process of
+celestial evolution. At length however, in~1877,
+Professor Asaph Hall discovered in the system
+\index{Hall, Asaph, discovery of Martian satellites|(}%
+of the planet Mars a case of the kind of motion
+\index{Mars!discovery of satellites|(}%
+which we foresee as the future fate of the moon
+and earth, for he found that the planet was attended
+by two satellites, the nearer of which has
+\index{Satellites!discovery of those of Mars|(}%
+a month shorter than the planet's day. He gives
+an interesting account of what had been conjectured,
+partly in jest and partly in earnest, as to
+the existence of satellites attending that planet.
+This foreshadowing of future discoveries is so
+curious that I quote the following passage from
+Professor Hall's paper. He writes:---
+\PageSep{291}
+
+``Since the discovery of the satellites of Mars,
+the remarkable statements of Dean Swift and
+Voltaire concerning the satellites of this planet,
+and the arguments of Dr.~Thomas Dick and
+others for the existence of such bodies, have attracted
+so much attention, that a brief account
+of the writings on this subject may be interesting.
+
+``The following letter of Kepler was written
+\index{Kepler!argument respecting Martian satellites}%
+to one of his friends soon after the discovery by
+Galileo in~1610 of the four satellites of Jupiter,
+\index{Galileo!discovery of Jupiter's satellites}%
+and when doubts had been expressed as to the
+reality of this discovery. The news of the discovery
+was communicated to him by his friend
+Wachenfels; and Kepler says:---
+
+``\,`Such a fit of wonder seized me at a report
+which seemed to be so very absurd, and I was
+thrown into such agitation at seeing an old dispute
+between us decided in this way, that between
+his joy, my coloring, and the laughter of
+both, confounded as we were by such a novelty,
+we were hardly capable, he of speaking, or I of
+listening. On our parting, I immediately began
+to think how there could be any addition to the
+number of the planets without overturning my
+``Cosmographic Mystery,'' according to which
+Euclid's five regular solids do not allow more
+than six planets round the sun\dots. I am so
+far from disbelieving the existence of the four
+circumjovial planets, that I long for a telescope,
+to anticipate you, if possible, in discovering \emph{two}
+\PageSep{292}
+\index{Kepler!argument respecting Martian satellites}%
+round Mars, as the proportion seems to require,
+\emph{six} or \emph{eight} round Saturn, and perhaps \emph{one} each
+round Mercury and Venus.'
+
+``Dean Swift's statement concerning the satellites
+\index{Swift, satire on mathematicians|(}%
+of Mars is in his famous satire, `The
+Travels of Mr.~Lemuel Gulliver.' After describing
+\index{Gulliver@\Title{Gulliver's Travels}, satire on mathematics|(}%
+his arrival in Laputa, and the devotion
+of the Laputians to mathematics and music,
+Gulliver says:---
+
+``\,`The knowledge I had in mathematics gave
+me great assistance in acquiring their phraseology,
+which depended much upon that science,
+and music; and in the latter I was not unskilled.
+Their ideas were perpetually conversant in lines
+and figures. If they would, for example, praise
+the beauty of a woman, or of any other animal,
+they describe it by rhombs, circles, parallelograms,
+ellipses, and other geometrical terms, or
+by words of art drawn from music, needless here
+to repeat\dots. And although they are dexterous
+enough upon a piece of paper, in the management
+of the rule, the pencil, and the divider,
+yet in the common actions and the behavior of
+life, I have not seen a more clumsy, awkward,
+and unhandy people, nor so slow and perplexed
+in their conceptions upon all subjects, except
+those of mathematics and music. They are very
+bad reasoners, and vehemently given to opposition,
+unless when they happen to be of the right
+opinion, which is seldom their case\dots. These
+\PageSep{293}
+people are under continual disquietudes, never
+enjoying a minute's peace of mind; and their
+disturbances proceed from causes which very
+little affect the rest of mortals. Their apprehensions
+arise from several changes they dread
+in the celestial bodies. For instance, that the
+earth, by the continual approaches of the sun
+towards it, must, in the course of time, be absorbed,
+or swallowed up. That the face of the
+sun will, by degrees, be encrusted with its own
+effluvia, and give no more light to the world.
+That the earth very narrowly escaped a brush
+from the tail of the last comet, which would
+have infallibly reduced it to ashes; and that the
+next, which they have calculated for one-and-thirty
+years hence, will probably destroy us.
+For if, in its perihelion, it should approach
+within a certain degree of the sun (as by their
+calculations they have reason to dread,) it will
+receive a degree of heat ten thousand times
+more intense than that of red-hot glowing iron;
+and, in its absence from the sun, carry a blazing
+tail ten hundred thousand and fourteen miles
+long; through which, if the earth should pass
+at the distance of one hundred thousand miles
+from the nucleus, or main body of the comet, it
+must, in its passage, be set on fire, and reduced
+to ashes. That the sun, daily spending its rays,
+without any nutriment to supply them, will at
+last be wholly consumed and annihilated; which
+\PageSep{294}
+must be attended with the destruction of this
+earth, and of all the planets that receive their
+light from it.
+
+``\,`They are so perpetually alarmed with the
+apprehension of these, and the like impending
+dangers, that they can neither sleep quietly in
+their beds, nor have any relish for the common
+pleasures and amusements of life. When they
+meet an acquaintance in the morning, the first
+question is about the sun's health, how he looked
+at his setting and rising, and what hopes they had
+to avoid the stroke of the approaching comet\dots.
+They spend the greatest part of their lives
+in observing the celestial bodies, which they do
+by the assistance of glasses, far excelling ours in
+goodness. For although their largest telescopes
+do not exceed three feet, they magnify much
+more than those of a hundred with us, and show
+the stars with greater clearness. This advantage
+has enabled them to extend their discoveries
+much further than our astronomers in Europe;
+for they have made a catalogue of ten thousand
+fixed stars, whereas the largest of ours do not
+contain above one-third of that number\dots.
+They have likewise discovered two lesser stars,
+or satellites, which revolve about Mars; whereof
+the innermost is distant from the centre of the
+primary planet exactly three of his diameters,
+and the outermost, five; the former revolves in
+the space of ten hours, and the latter in twenty-one
+\PageSep{295}
+and a half; so that the squares of their
+periodical times are very near in the same proportion
+with the cubes of their distance from
+the centre of Mars; which evidently shows them
+to be governed by the same law of gravitation
+that influences the other heavenly bodies.'
+
+``The reference which Voltaire makes to the
+\index{Voltaire, satire on mathematicians, and Martian satellites}%
+moons of Mars is in his `Micromegas, Histoire
+Philosophique.' Micromegas was an inhabitant
+of Sirius, who, having written a book which a
+suspicious old man thought smelt of heresy, left
+Sirius and visited our solar system. Voltaire
+says:---
+
+``\,`Mais revenons à nos voyageurs. En sortant
+de Jupiter, ils traversèrent un espace d'environ
+cent millions de lieues, et ils côtoyèrent
+la planète de Mars, qui, comme on sait, est cinq
+fois plus petite que noire petit globe; ils virent
+deux lunes qui servent à cette planète, et qui ont
+échappé aux regards de nos astronomes. Je sais
+bien que le père \emph{Castel} écrira, et même plaisamment,
+\index{Castel, Father, ridiculed by Voltaire}%
+contre l'existence de ces deux lunes; mais
+je m'en rapporte à ceux qui raisonnent par analogie.
+Ces bons philosophes-là savent combien il
+serait difficile que Mars, qui est si loin du soleil,
+se passât à moins de deux lunes.'
+
+``The argument by analogy for the existence
+of a satellite of Mars was revived by writers like
+Dr.~Thomas Dick, Dr.~Lardner, and others. In
+\index{Dick, argument as to Martian satellites}%
+\index{Lardner, possibility of Martian satellites}%
+addition to what may be called the analogies of
+\index{Gulliver@\Title{Gulliver's Travels}, satire on mathematics|)}%
+\index{Swift, satire on mathematicians|)}%
+\PageSep{296}
+astronomy, these writers appear to rest on the
+idea that a beneficent Creator would not place
+a planet so far from the sun as Mars without
+giving it a satellite. This kind of argument has
+passed into some of our handbooks of astronomy,
+and is stated as follows by Mr.~Chambers
+\index{Chambers on possible existence of Martian satellites}%
+in his excellent book on `Descriptive Astronomy,'
+2d~edition, p.~89, published in~1867:---
+
+``\,`As far as we know, Mars possesses no satellite,
+though analogy does not forbid, but rather,
+on the contrary, infers the existence of one; and
+its never having been seen, in this case at least,
+proves nothing. The second satellite of Jupiter
+is only $\frac{1}{43}$~of the diameter of the primary, and
+a satellite $\frac{1}{43}$~of the diameter of Mars would
+be less than $100$~miles in diameter, and therefore
+of a size barely within the reach of our largest
+telescopes, allowing nothing for its possibly close
+proximity to the planet. The fact that one of
+the satellites of Saturn was only discovered a
+few years ago renders the discovery of a satellite
+of Mars by no means so great an improbability
+as might be imagined.'
+
+``Swift seems to have had a hearty contempt
+for mathematicians and astronomers, which he
+has expressed in his description of the inhabitants
+of Laputa. Voltaire shared this contempt,
+\index{Voltaire, satire on mathematicians, and Martian satellites}%
+and delighted in making fun of the philosophers
+whom Frederick the Great collected at Berlin.
+The `père Castel' may have been le~père Louis
+\index{Castel, Father, ridiculed by Voltaire}%
+\PageSep{297}
+Castel, who published books on physics and
+mathematics at Paris in 1743 and~1758. The
+probable origin of these speculations about the
+moons of Mars was, I think, Kepler's analogies.
+Astronomers failing to verify these, an opportunity
+was afforded to satirists like Swift and
+Voltaire to ridicule such arguments.''\footnote
+ {\Title{Observations and Orbits of the Satellites of Mars}, by Asaph
+ Hall. Washington, Government Printing Office, 1878.}
+
+As I have already said, these prognostications
+were at length verified by Professor Asaph Hall
+in the discovery of two satellites, which he named
+Phobos and Deimos---Fear and Panic, the dogs
+\index{Deimos, a satellite of Mars}%
+\index{Phobos, a satellite of Mars}%
+of war. The period of Deimos is about $30$~hours,
+and that of Phobos somewhat less than $8$~hours,
+whilst the Martian day is of nearly the same
+length as our own. The month of the inner
+minute satellite is thus less than a third of the
+planet's day; it rises to the Martians in the west,
+and passes through all its phases in a few hours;
+sometimes it must even rise twice in a single
+Martian night. As we here find an illustration
+of the condition foreseen for the earth and moon,
+it seems legitimate to suppose that solar tidal
+friction has retarded the planet's rotation until it
+has become slower than the revolution of one of
+the satellites. It would seem as if the ultimate
+fate of Phobos will be absorption in the planet.
+
+Several of the satellites of Jupiter and of Saturn
+present faint inequalities of coloring, and
+\PageSep{298}
+\index{Jupiter!satellites constantly face planet}%
+\index{Saturn!satellites always face the planet}%
+telescopic examination has led astronomers to believe
+that they always present the same face to
+their planets. The theory of tidal friction would
+\index{Planets!rotation of some, annulled by tidal friction}%
+certainly lead us to expect that these enormous
+planets should work out the same result for their
+relatively small satellites that the earth has produced
+\index{Satellites!discovery of those of Mars|)}%
+\index{Satellites!rotation of those of Jupiter and Saturn annulled}%
+in the moon.
+
+The proximity of the planets Mercury and
+\index{Mercury, rotation of}%
+Venus to the sun should obviously render solar
+\index{Venus, rotation of}%
+tidal friction far more effective than with us.
+The determination of the periods of rotation of
+\index{Rotation!of Mercury, Venus, and satellites of Jupiter and Saturn annulled by tidal friction}%
+these planets thus becomes a matter of much interest.
+But the markings on their disks are so
+obscure that the rates of their rotations have remained
+under discussion for many years. Until
+recently the prevailing opinion was that in both
+cases the day was of nearly the same length as
+ours; but a few years ago Schiaparelli of Milan,
+\index{Schiaparelli on rotation of Venus and Mercury}%
+an observer endowed with extraordinary acuteness
+of vision, announced as the result of his observations
+that both Mercury and Venus rotate
+only once in their respective years, and that
+each of them constantly presents the same face
+to the sun. These conclusions have recently been
+confirmed by Mr.~Percival Lowell from observations
+\index{Lowell, P., on rotations of Venus and Mercury}%
+made in Arizona. Although on reading
+the papers of these astronomers it is not easy
+to see how they can be mistaken, yet it should
+be noted that others have failed to detect the
+markings on the planet's disks, although they
+\index{Hall, Asaph, discovery of Martian satellites|)}%
+\index{Mars!discovery of satellites|)}%
+\PageSep{299}
+\index{Lowell, P., on rotations of Venus and Mercury}%
+apparently enjoyed equal advantages for observation.\footnote
+ {Dr.~See, a member of the staff of the Flagstaff Observatory,
+ Arizona, tells me that he has occasionally looked at these planets
+ through the telescope, although he took no part in the systematic
+ observation. In his opinion it would be impossible for any one
+ at Flagstaff to doubt the reality of the markings. There are,
+ however, many astronomers of eminence who suspend their
+ judgment, and await confirmation by other observers at other
+ stations.}
+
+If, as I am disposed to do, we accept these observations
+as sound, we find that evidence favorable
+to the theory of tidal friction is furnished
+by the planets Mercury and Venus, and by the
+\index{Mercury, rotation of}%
+\index{Venus, rotation of}%
+satellites of the earth, Jupiter and Saturn, whilst
+\index{Earth and moon!figure of}%
+\index{Earth and moon!adjustment of figure to suit change of rotation|(}%
+the Martian system is yet more striking as an
+instance of an advanced stage in evolution.
+
+\TB
+
+It is well known that the figure of the earth
+is flattened by the diurnal rotation, so that the
+polar axis is shorter than any equatorial diameter.
+At the present time the excess of the equatorial
+radius over the polar radius is $\frac{1}{290}$~part of
+either of them. Now in tracing the history of
+the earth and moon, we found that the earth's
+rotation had been retarded, so that the day is
+now longer than it was. If then the solid earth
+has always been absolutely unyielding, and if an
+ocean formerly covered the planet to a uniform
+depth, the sea must have gradually retreated
+towards the poles, leaving the dry land exposed
+at the equator. If on the other hand the solid
+\PageSep{300}
+\index{Geological evidence of earth's plasticity}%
+\index{Plasticity of earth under change of rotation|(}%
+earth had formerly its present shape, there must
+then have been polar continents and a deep equatorial
+sea.
+
+But any considerable change in the speed of
+the earth's rotation would, through the action of
+gravity, bring enormous forces to bear on the
+solid earth. These forces are such as would, if
+they acted on a plastic material, tend to restore
+the planet's figure to the form appropriate to its
+changed rotation. It has been shown experimentally
+by M.~Tresca and others that even very
+\index{Tresca on flow of solids}%
+rigid and elastic substances lose their rigidity
+and their elasticity, and become plastic under the
+action of sufficiently great forces. It appears to
+me, therefore, legitimate to hold to the belief in
+the temporary rigidity of the earth's mass, as explained
+in \Ref{Chapter}{XV}., whilst contending that
+under a change of rotational velocity the earth
+may have become plastic, and so have maintained
+a figure adapted to its speed. Geological observation
+shows that rocks have been freely twisted
+and bent near the earth's surface, and it is impossible
+to doubt that under altered rotation the
+deeper portions of the earth would have been
+subjected to very great stress. I conjecture that
+the internal layers might adapt themselves by
+continuous flow, whilst the superficial portion
+might yield impulsively. Earthquakes are probably
+due to unequal shrinkage of the planet in
+cooling, and each shock would tend to bring the
+\PageSep{301}
+strata into their position of rest; thus the earth's
+surface would avail itself of the opportunity afforded
+by earthquakes of acquiring its proper
+shape. The deposit in the sea of sediment, derived
+from the denudation of continents, affords
+another means of adjustment of the figure of the
+planet. I believe then that the earth has always
+maintained a shape nearly appropriate to its rotation.
+The existence of the continents proves
+that the adjustment has not been perfect, and we
+shall see reason to believe that there has been
+also a similar absence of complete adjustment in
+the interior.
+
+But the opinion here maintained is not shared
+by the most eminent of living authorities, Lord
+Kelvin; for he holds that the fact that the average
+\index{Kelvin, Lord!denies adjustment of earth's figure to changed rotation}%
+figure of the earth corresponds with the
+actual length of the day proves that the planet
+was consolidated at a time when the rotation was
+but little more rapid than it is now. The difference
+between us is, however, only one of degree,
+for he considers that the power of adjustment is
+slight, whilst I hold that it would be sufficient
+to bring about a considerable change of shape
+within the period comprised in geological history.
+
+If the adjustment of the planet's figure were
+perfect, the continents would sink below the
+ocean, which would then be of uniform depth.
+But there is no superficial sign, other than the
+dry land, of absence of adaptation to the present
+\PageSep{302}
+\index{Moon and earth!inequality in motion indicates internal density of earth}%
+rotation---unless indeed the deep polar sea discovered
+by Nansen be such. Yet, as I have
+hinted above, some tokens still exist in the earth
+\index{Earth and moon!internal density}%
+of the shorter day of the past. The detection of
+this evidence depends however on arguments of
+so technical a character that I cannot hope in
+such a work as this to do more than indicate the
+nature of the proof.
+
+The earth is denser towards the centre than
+outside, and the layers of equal density are concentric.
+\index{Density!of earth, law of internal}%
+If then the materials were perfectly
+plastic throughout, not only the surface, but
+also each of these layers would be flattened to a
+definite extent, which depends on the rate of rotation
+and on the law governing the internal
+density of the earth. Although the rate at
+which the earth gets denser is unknown, yet it is
+possible to assign limits to the density at various
+depths. Thus it can be proved that at any internal
+point the density must lie between two
+values which depend on the position of the point
+in question. So also, the degree of flattening at
+any internal point is found to lie between two
+extreme limits, provided that all the internal layers
+are arranged as they would be if the whole
+mass were plastic.
+
+Now variations in the law of internal density
+and in the internal flattening would betray themselves
+to our observation in several ways. In
+the first place, gravity on the earth's surface
+\index{Earth and moon!adjustment of figure to suit change of rotation|)}%
+\index{Gravity, variation according to latitude}%
+\index{Plasticity of earth under change of rotation|)}%
+\PageSep{303}
+\index{Meteorological!conditions dependent on earth's rotation}%
+\index{Moon and earth!inequality in motion indicates internal density of earth}%
+\index{Nutation!value of, indicates internal density of earth}%
+would be changed. The force of gravity at the
+\index{Gravity, variation according to latitude}%
+poles is greater than at the equator, and the law
+of its variation according to latitude is known.
+In the second place the amount of the flattening
+of the earth's surface would be altered, and the
+present figure of the earth is known with considerable
+exactness. Thirdly the figure and law of
+density of the earth govern a certain irregularity
+or inequality in the moon's motion, which has
+been carefully evaluated by astronomers. Lastly
+the precessional and nutational motion of the
+earth is determined by the same causes, and these
+motions also are accurately known. These four
+facts of observation---gravity, the ellipticity of
+\index{Ellipticity of earth's strata in excess for present rotation}%
+the earth, the lunar inequality, and the precessional
+and nutational motion of the earth---are
+so intimately intertwined that one of them cannot
+be touched without affecting the others.
+
+Now Édouard Roche, a French mathematician,
+\index{Roche, E.!ellipticity of internal strata of earth}%
+has shown that if the earth is perfectly plastic,
+so that each layer is exactly of the proper shape
+for the existing rotation, it is not possible to adjust
+the unknown law of internal density so as
+\index{Precession, value of, indicates internal density of earth}%
+to make the values of all these elements accord
+with observation. If the density be assumed
+such as to fit one of the data, it will produce a
+disagreement with observation in others. If,
+however, the hypothesis be abandoned that the
+internal strata all have the proper shapes, and if
+it be granted that they are a little more flattened
+\PageSep{304}
+\index{Ellipticity of earth's strata in excess for present rotation}%
+\index{Geological evidence of earth's plasticity!as to retardation of earth's rotation|(}%
+than is due to the present rate of rotation, the
+data are harmonized together; and this is just
+what would be expected according to the theory
+of tidal friction. But it would not be right to
+attach great weight to this argument, for the
+absence of harmony is so minute that it might
+be plausibly explained by errors in the numerical
+data of observation. I notice, however, that the
+most competent judges of this intricate subject
+are disposed to regard the discrepancy as a
+reality.
+
+\DPchg{}{\TB}
+
+We have seen in the preceding chapter that
+the length of day has changed but little within
+historical times. But the period comprised in
+written history is almost as nothing compared
+with the whole geological history of the earth.
+We ought then to consider whether geology furnishes
+any evidence bearing on the theory of
+tidal friction. The meteorological conditions on
+the earth are dependent to a considerable extent
+on the diurnal rotation of the planet, and therefore
+those conditions must have differed in the
+past. Our storms are of the nature of aerial eddies,
+and they derive their rotation from that of
+the earth. Accordingly storms were probably
+more intense when the earth spun more rapidly.
+The trunks of trees should be stronger than they
+are now to withstand more violent storms. But
+I cannot learn that there is any direct geological
+evidence on this head, for deciduous trees with
+\PageSep{305}
+\index{Ripple mark in sand preserved in geological strata}%
+stiff trunks seem to have been a modern product
+of geological time, whilst the earlier trees more
+nearly resembled bamboos, which yield to the
+wind instead of standing up to it. It seems possible
+that trees and plants would not be exterminated,
+even if they suffered far more wreckage
+than they do now. If trees with stiff trunks
+could only withstand the struggle for existence
+when storms became moderate in intensity, their
+absence from earlier geological formations would
+be directly due to the greater rapidity of the
+earth's rotation in those times.
+
+According to our theory the tides on the seacoast
+must certainly have had a much wider
+range, and river floods must probably have been
+more severe. The question then arises whether
+these agencies should have produced sedimentary
+deposits of coarser grain than at present. Although
+I am no geologist, I venture to express a
+doubt whether it is possible to tell, within very
+wide limits, the speed of the current or the range
+of the tide that has brought down and distributed
+any sedimentary deposit. I doubt whether any
+geologist would assert that floods might not have
+been twice or thrice as frequent, or that the tide
+might not have had a very much greater range
+than at present.
+
+In some geological strata ripple-marks have
+been preserved which exactly resemble modern
+ones. This has, I believe, been adduced as an
+\PageSep{306}
+argument against the existence of tides of great
+range. Ripples are, however, never produced
+by a violent scour of water, but only by gentle
+currents or by moderate waves. The turn of
+the tide must be gentle to whatever height it
+rises, and so the formation of ripple-mark should
+have no relationship to the range of tide.
+
+It appears then that whilst geology affords no
+direct confirmation of the theory, yet it does not
+present any evidence inconsistent with it. Increased
+activity in the factors of change is important
+to geologists, since it renders intelligible
+a diminution in the time occupied by the history
+of the earth; and thus brings the views of the
+\index{Earth and moon!probably once molten}%
+geologist and of the physicist into better harmony.
+
+Although in this discussion I have maintained
+the possibility that a considerable portion of the
+changes due to tidal friction may have occurred
+within geological history, yet it seems to me
+probable that the greater part must be referred
+back to pre-geological times, when the planet
+was partially or entirely molten.
+
+\TB
+
+The action of the moon and sun on a plastic
+and viscous planet would have an effect of which
+some remains may perhaps still be traceable.
+The relative positions of the moon and of the
+frictionally retarded tide were illustrated in the
+last chapter by~\fig{36}. That figure shows that
+\index{Geological evidence of earth's plasticity!as to retardation of earth's rotation|)}%
+\PageSep{307}
+the earth's rotation is retarded by forces acting
+\index{Earth and moon!distortion under primeval tidal friction}%
+on the tidal protuberances in a direction adverse
+to the planet's rotation. As the plastic substance,
+of which we now suppose the planet to
+be formed, rises and falls rhythmically with the
+tide, the protuberant portions are continually
+subject to this retarding force. Meanwhile the
+internal portions are urged onward by the
+inertia due to their velocity. Accordingly there
+must be a slow motion of the more superficial
+portions with reference to the interior. From
+the same causes, under present conditions, the
+whole ocean must have a slow westerly drift, although
+it has not been detected by observation.
+
+Returning however to our plastic planet, the
+equatorial portion is subjected to greater force
+than the polar regions, and if meridians were
+painted on its surface, as on a map, they would
+gradually become distorted. In the equatorial
+belt the original meridional lines would still run
+north and south, but in the northern hemisphere
+they would trend towards the northeast, and in
+the southern hemisphere towards the southeast.
+This distortion of the surface would cause the
+surface to wrinkle, and the wrinkles should be
+warped in the directions just ascribed to the
+meridional lines. If the material yielded very
+easily I imagine that the wrinkles would be
+small, but if it were so stiff as only to yield with
+difficulty they might be large.
+\PageSep{308}
+
+There can be no doubt as to the correctness
+\index{History!of earth and moon|(}%
+of this conclusion as to a stiff yet viscous planet,
+but the application of these ideas to the earth is
+hazardous and highly speculative. We do, however,
+observe that the continents, in fact, run
+\index{Continents, trend of, possibly due to primeval tidal friction}%
+roughly north and south. It may appear fanciful
+to note, also, that the northeastern coast of
+America, the northern coast of China, and the
+southern extremity of South America have the
+proper theoretical trends. But the northwestern
+coast of America follows a line directly adverse
+to the theory, and the other features of the globe
+are by no means sufficiently regular to inspire
+much confidence in the justice of the conjecture.\footnote
+ {See, also, W. Prinz, \Title{Torsion apparente des planètes}, ``Annuaire
+ de l'Obs.~R. de~Bruxelles,'' 1891.}
+
+\TB
+
+We must now revert to the astronomical aspects
+of our problem. It is natural to inquire
+whether the theory of tidal friction is competent
+to explain any peculiarities of the motion of the
+moon and earth other than those already considered.
+It has been supposed thus far that the
+moon moves over the earth's equator in a circular
+orbit, and that the equator coincides with the
+plane in which the earth moves in its orbit. But
+the moon actually moves in a plane different
+from that in which the earth revolves round the
+sun, her orbit is not circular but elliptic, and the
+\PageSep{309}
+earth's equator is oblique to the orbit. We must
+consider, then, how tidal friction will affect these
+three factors.
+
+Let us begin by considering the obliquity of
+the equator to the ecliptic, which produces the
+seasonal changes of winter and summer. The
+problem involved in the disturbance of the motion
+of a rotating body by any external force is
+too complex for treatment by general reasoning,
+and I shall not attempt to explain in detail the
+interaction of the moon and earth in this respect.
+
+The attractions of the moon and sun on the
+equatorial protuberance of the earth causes the
+earth's axis to move slowly and continuously
+with reference to the fixed stars. At present,
+the axis points to the pole-star, but $13,000$~years
+hence the present pole-star will be $47°$~distant
+from the pole, and in another $13,000$~years it
+will again be the pole-star. Throughout this
+precessional movement the obliquity of the equator
+to the ecliptic remains constant, so that winter
+and summer remain as at present. There is
+also, superposed on the precession, the nutational
+or nodding motion of the pole to which I referred
+in \Ref{Chapter}{XV}. In the absence of tidal
+friction the attractions of the moon and sun on
+the tidal protuberance would slightly augment
+the precession due to the solid equatorial protuberance,
+and would add certain very minute
+nutations of the earth's axis; the amount of
+\PageSep{310}
+these tidal effects, is, however, quite insignificant.
+But under the influence of tidal friction,
+the matter assumes a different aspect, for the
+earth's axis will not return at the end of each
+nutation to exactly the same position it would
+have had in the absence of friction, and there is
+a minute residual effect which always tends in
+the same direction. A motion of the pole may
+be insignificant when it is perfectly periodic, but
+it becomes important in a very long period of
+time when the path described is not absolutely
+reëntrant. Now this is the case with regard to
+the motion of the earth's axis under the influence
+of frictionally retarded tides, for it is found
+to be subject to a gradual drift in one direction.
+
+In tracing the history of the earth and moon
+backwards in time we found the day and month
+growing shorter, but at such relative speeds that
+the number of days in the month diminished until
+the day and month became equal. This conclusion
+remains correct when the earth is oblique
+to its orbit, but the effect on the obliquity is
+\index{Ecliptic, obliquity of, due to tidal friction|(}%
+\index{Obliquity of ecliptic, effects of tidal friction on|(}%
+found to depend in a remarkable manner upon
+the number of days in the month. At present
+and for a long time in the past the obliquity
+is increasing, so that it was smaller long ago.
+But on going back to the time when the day
+was six and the month twelve of our present
+hours we find that the tendency for the obliquity
+to increase vanishes. In other words, if
+\PageSep{311}
+there are more than two days in a month the
+obliquity will increase, if less than two it will
+diminish.
+
+Whatever may be the number of days in the
+month, the rate of increase or diminution of
+obliquity varies as the obliquity which exists at
+the moment under consideration. If, then, a
+planet be spinning about an axis absolutely perpendicular
+to the plane of its satellite's orbit, the
+obliquity remains invariable. But if we impart
+infinitesimal obliquity to a planet whose day is
+less than half a month, that infinitesimal obliquity
+will increase; whilst, if the day is more
+than half a month, the infinitesimal obliquity
+will diminish. Accordingly, the motion of a
+planet spinning upright is stable, if there are
+less than two days in a month, and unstable if
+there are more than two.
+
+It is not legitimate to ascribe the whole of
+the present obliquity of~$23\frac{1}{2}°$ to the influence of
+tidal friction, because it appears that when there
+were only two days in the month, the obliquity
+was still as much as~$11°$. It is, moreover, impossible
+to explain the considerable obliquity of the
+other planets to their orbits by this cause. It
+must, therefore, be granted that there was some
+unknown cause which started the planets in rotation
+about axes oblique to their orbits. It remains,
+however, certain that a planet, rotating primitively
+without obliquity, would gradually become
+\PageSep{312}
+inclined to its orbit, although probably not to so
+great an extent as we find in the case of the
+earth.
+
+The next subject to be considered is the fact
+that the moon's orbit is not circular but eccentric.
+Here, again, it is found that if the tides
+were not subject to friction, there would be no
+sensible effect on the shape of the moon's path,
+but tidal friction produces a reaction on the
+moon tending to change the degree of eccentricity.
+In this case, it is possible to indicate by
+general reasoning the manner in which this reaction
+operates. We have seen that tidal reaction
+tends to increase the moon's distance from the
+earth. Now, when the moon is nearest, in perigee,
+the reaction is stronger than when she is
+furthest, in apogee. The effect of the forces in
+perigee is such that the moon's distance at the
+next succeeding apogee is greater than it was at
+the next preceding apogee; so, also, the effect
+of the forces in apogee is an increase in the perigeal
+distance. But the perigeal effect is stronger
+than the apogeal, and, therefore, the apogeal distances
+increase more rapidly than the perigeal
+ones. It follows, therefore, that, whilst the orbit
+as a whole expands, it becomes at the same time
+more eccentric.
+\index{Ecliptic, obliquity of, due to tidal friction|)}%
+\index{Obliquity of ecliptic, effects of tidal friction on|)}%
+
+The lunar orbit is then becoming more eccentric,
+and numerical calculation shows that in
+very early times it must have been nearly circular.
+\PageSep{313}
+\index{Eccentricity of orbit!due to tidal friction}%
+\index{Moon and earth!eccentricity of orbit increased by tidal friction}%
+\index{Saint@St.\ Vénant on flow of solids}%
+\index{See, T. J. J.!eccentricity of orbits of double stars}%
+But mathematical analysis indicates that in
+this case, as with the obliquity, the rate of
+increase depends in a remarkable manner upon
+the number of days in the month. I find in
+fact that if eighteen days are less than eleven
+months the eccentricity will increase, but in the
+converse case it will diminish; in other words
+the critical stage at which the eccentricity is
+stationary is when $1\frac{7}{11}$~days is equal to the
+month. It follows from this that the circular
+orbit of the satellite is dynamically stable or
+\index{Orbit!of double stars, very eccentric}%
+unstable according as $1\frac{7}{11}$~days is less or greater
+than the month.
+
+The effect of tidal friction on the eccentricity
+has been made the basis of extensive astronomical
+speculations by Dr.~See. I shall revert to
+this subject in \Ref{Chapter}{XIX}., and will here
+merely remark that systems of double stars are
+\index{Stars!double, eccentricity of orbits}%
+found to revolve about one another in orbits of
+great eccentricity, and that Dr.~See supposes
+that the eccentricity has arisen from the tidal
+action of each star on the other.
+
+The last effect of tidal friction to which I
+have to refer is that on the plane of the moon's
+orbit. The lunar orbit is inclined to that of the
+earth round the sun at an angle of~$5°$, and the
+problem to be solved is as to the nature of the
+effect of tidal friction on that inclination. The
+nature of the relation of the moon's orbit to the
+ecliptic is however so complex that it appears
+\index{History!of earth and moon|)}%
+\PageSep{314}
+\index{Eccentricity of orbit!due to tidal friction}%
+\index{Moon and earth!eccentricity of orbit increased by tidal friction}%
+hopeless to explain the effects of tidal action
+without the use of mathematical language, and
+I must frankly give up the attempt. I may,
+however, state that when the moon was near the
+earth she must have moved nearly in the plane
+of the earth's equator, but that the motion gradually
+changed so that she has ultimately come to
+move nearly in the plane of the ecliptic. These
+two extreme cases are easily intelligible, but the
+transition from one case to the other is very
+complicated. It may suffice for this general
+account of the subject to know that the effects
+of tidal friction are quite consistent with the
+present condition of the moon's motion, and
+with the rest of the history which has been
+traced.
+
+This discussion of the effects of tidal friction
+may be summed up thus:---
+
+If a planet consisted partly or wholly of molten
+lava or of other fluid, and rotated rapidly about
+an axis perpendicular to the plane of its orbit,
+and if that planet was attended by a single satellite,
+revolving with its month a little longer than
+the planet's day, then a system would necessarily
+be developed which would have a strong resemblance
+to that of the earth and moon.
+
+A theory reposing on \textit{veræ causæ} which brings
+into quantitative correlation the lengths of the
+present day and month, the obliquity of the
+ecliptic, the eccentricity and the inclination of
+\PageSep{315}
+the lunar orbit, should have strong claims to
+acceptance.
+
+\begin{Authorities}
+G.~H. Darwin. A series of papers in the ``Phil.\ Trans.\ Roy.\
+\index{Darwin, G. H.!papers on tidal friction}%
+Soc.'' pt.~i.\ 1879, pt.~ii.\ 1879, pt.~ii.\ 1880, pt.~ii.\ 1881, pt.~i.\ 1882,
+and abstracts (containing general reasoning) in the corresponding
+Proceedings; also ``Proc.\ Roy.\ Soc.''\ vol.~29, 1879, p.~168 (in
+part republished in Thomson and Tait's \Title{Natural Philosophy}),
+and vol.~30, 1880, p.~255.
+
+Lord Kelvin, \Title{On Geological Time}, ``Popular Lectures and
+\index{Kelvin, Lord!on geological time}%
+Addresses,'' vol.~iii. Macmillan, 1894.
+
+Roche. The investigations of Roche and of others are given
+in Tisserand's \Title{Mécanique Céleste}, vol.~ii. Gauthier-Villars, 1891.
+\index{Tisserand, Roche's investigations as to earth's figure}%
+
+Tresca and St.~Vénant, \Title{Sur l'écoulement des Corps Solides},
+``Mémoires des Savants Étrangers,'' Académie des Sciences de
+Paris, vols.\ 18~and~20.
+
+Schiaparelli, \Title{Considerazioni sul moto rotatorio del pianeta
+\index{Schiaparelli on rotation of Venus and Mercury}%
+Venere}. Five notes in the ``Rendiconti del R.~Istituto Lombardo,''
+vol.~23, and \Title{Sulla rotazione di Mercurio}, ``Ast.\ Nach.,''
+No.~2944. An abstract is given in ``Report of Council of R.
+Ast.\ Soc.,'' Feb.~1891.
+
+Lowell, Mercury, ``Ast.\ Nach.,'' No.~3417. \Title{Mercury and Determination
+\index{Lowell, P., on rotations of Venus and Mercury}%
+of Rotation Period~\dots\ of Venus}, ``Monthly Notices
+R. Ast.\ Soc.,'' vol.~57, 1897, p.~148. \Title{Further proof},~\&c., \textit{ibid}.\
+p.~402.
+
+Douglass, \Title{Jupiter's third Satellite}, ``Ast.\ Nach.,'' No.~3432.
+\index{Douglass, rotation of Jupiter's satellites}%
+\Title{Rotation des IV~Jupitersmondes}, ``Ast.\ Nach.,'' No.~3427, confirming
+Engelmann, \Title{Ueber~\dots\ Jupiterstrabanten}, Leipzig, 1871.
+
+Barnard, \Title{The third and fourth Satellites of Jupiter}, ``Ast.\
+\index{Barnard, rotation of Jupiter's satellites}%
+Nach.,'' No.~3453.
+\end{Authorities}
+\index{Friction of tides|)}%
+\PageSep{316}
+
+
+\Chapter[Figures of Equilibrium]{XVIII}
+{The Figures of Equilibrium of a Rotating
+Mass of Liquid}
+
+\First{The} theory of the tides involves the determination
+\index{Equilibrium, figures of, of rotating liquid|(}%
+\index{Figure of equilibrium!of rotating liquid|(}%
+\index{Rotating liquid, figures of equilibrium|(}%
+of the form assumed by the ocean under
+the attraction of a distant body, and it now
+remains to discuss the figure which a rotating
+mass of liquid may assume when it is removed
+from all external influences. The forces which
+act upon the liquid are the mutual gravitation
+of its particles, and the centrifugal force due to
+its rotation. If the mass be of the appropriate
+shape, these two opposing forces will balance
+one another, and the shape will be permanent.
+The problem in hand is, then, to determine
+what shapes of this kind are possible.
+
+In 1842 a distinguished Belgian physicist, M.~Plateau,\footnote
+ {He is justly celebrated not only for his discoveries, but also
+ for his splendid perseverance in continuing his researches after
+ he had become totally blind.}
+devised an experiment which affords
+\index{Capillarity of liquids, and Plateau's experiment|(}%
+\index{Plateau, experiment on figure of rotating globule|(}%
+a beautiful illustration of the present subject.
+The experiment needs very nice adjustment in
+several respects, but I refer the reader to
+Plateau's paper for an account of the necessary
+\PageSep{317}
+\index{Surface tension of liquids}%
+precautions. Alcohol and water may be so
+mixed as to have the same density as olive oil.
+If the adjustment of density is sufficiently exact,
+a mass of oil will float in the mixture, in the
+form of a spherical globule, without any tendency
+to rise or fall. The oil is thus virtually
+relieved from the effect of gravity. A straight
+wire, carrying a small circular disk at right
+angles to itself, is then introduced from the top
+of the vessel. When the disk reaches the
+globule, the oil automatically congregates itself
+round the disk in a spherical form, symmetrical
+with the wire.
+
+The disk is then rotated slowly and uniformly,
+and carries with it the oil, but leaves the surrounding
+mixture at rest. The globule is then
+seen to become flattened like an orange, and as
+the rotation quickens it dimples at the centre,
+and finally detaches itself from the disk in the
+form of a perfect ring. This latter form is only
+transient; for the oil usually closes in again
+round the disk, or sometimes, with slightly different
+manipulation, the ring may break into
+drops which revolve round the centre, rotating
+round their axes as they go.
+
+The force which holds a drop of water, or
+this globule of oil, together is called ``surface
+tension'' or ``capillarity.'' It is due to a certain
+molecular attraction, quite distinct from
+that of gravitation, and it produces the same
+\PageSep{318}
+effect as if the surface of the liquid were enclosed
+in an elastic skin. There is of course no
+actual skin, and yet when the liquid is stirred
+the superficial particles attract their temporary
+neighbors so as to restore the superficial elasticity,
+continuously and immediately. The intensity
+of surface tension depends on the nature
+\index{Surface tension of liquids}%
+of the material with which the liquid is in contact;
+thus there is a definite degree of tension
+in the skin of olive oil in contact with spirits
+and water.
+
+A globule at rest necessarily assumes the form
+of a sphere under the action of surface tension,
+but when it rotates it is distorted by centrifugal
+force. The polar regions become less curved,
+and the equatorial region becomes more curved,
+until the excess of the retaining power at the
+equator over that at the poles is sufficient to
+restrain the centrifugal force. Accordingly the
+struggle between surface tension and centrifugal
+force results in the assumption by the globule
+of an orange-like shape, or, with greater speed
+of rotation, of the other figures of equilibrium.
+\index{Capillarity of liquids, and Plateau's experiment|)}%
+
+In very nearly the same way a large mass of
+gravitating and rotating liquid will naturally
+assume certain definite forms. The simplest
+case of the kind is when the fluid is at rest in
+space, without any rotation. Then mutual gravitation
+is the only force which acts on the system.
+The water will obviously crowd together
+\PageSep{319}
+into the smallest possible space, so that every
+particle may get as near to the centre as its
+neighbors will let it. I suppose the water to be
+incompressible, so that the central portion, although
+pressed by that which lies outside of it,
+does not become more dense; and so the water
+does not weigh more per cubic foot near the
+centre than towards the outside. Since there
+is no upwards and downwards, or right and
+left about the system, it must be symmetrical in
+every direction; and the only figure which possesses
+this quality of universal symmetry is the
+sphere. A sphere is then said to be a figure of
+equilibrium of a mass of fluid at rest.
+
+If such a sphere of water were to be slightly
+deformed, and then released, it would oscillate
+to and fro, but would always maintain a nearly
+spherical shape. The speed of the oscillation
+depends on the nature of the deformation impressed
+upon it. If the water were flattened to
+the shape of an orange and released, it would
+spring back towards the spherical form, but
+would overshoot the mark, and pass on to a
+lemon shape, as much elongated as the orange
+was flattened. It would then return to the
+orange shape, and so on backwards and forwards,
+passing through the spherical form at
+each oscillation. This is the simplest kind of
+oscillation which the system can undergo, but
+there is an infinite number of other modes of
+\index{Plateau, experiment on figure of rotating globule|)}%
+\PageSep{320}
+any degree of complexity. The mathematician
+can easily prove that a liquid globe, of the same
+density as the earth, would take an hour and a
+half to pass from the orange shape to the lemon
+shape, and back to the orange shape. At present,
+the exact period of the oscillation is not
+the important point, but it is to be noted that if
+the body be set oscillating in any way whatever,
+it will continue to oscillate and will always remain
+nearly spherical. We say then that the
+sphere is a stable form of equilibrium of a mass
+of fluid. The distinction between stability and
+instability has been already illustrated in \Ref{Chapter}{XVI}.\
+by the cases of an egg lying on its
+side and balanced on its end, and there is a
+similar distinction between stable and unstable
+modes of motion.
+
+Let us now suppose the mass of water to rotate
+slowly, all in one piece as if it were solid.
+We may by analogy with the earth describe the
+axis of rotation as polar, and the central plane,
+at right angles to the axis, as equatorial. The
+equatorial region tends to move outwards in consequence
+of the centrifugal force of the rotation,
+and this tendency is resisted by gravitation which
+tends to draw the water together towards the
+centre. As the rotation is supposed to be very
+slow, centrifugal force is weak, and its effects are
+small thus the globe is very slightly flattened at
+the poles, like an orange or like the earth itself.
+\PageSep{321}
+Such a body resembles the sphere in its behavior
+when disturbed; it will oscillate, and its average
+figure in the course of its swing is the orange
+shape. It is therefore stable.
+
+But it has been discovered that the liquid may
+also assume two other alternative forms. One
+of these is extremely flattened and resembles a
+flat cheese with rounded edges. As the disk of
+liquid is very wide, the centrifugal force at the
+equator is very great, although the rotation is
+very slow. In the case of the orange-shaped figure,
+the slower the rotation the less is the equatorial
+centrifugal force, because it diminishes
+both with diminution of radius and fall of speed.
+But in the cheese shape the equatorial centrifugal
+force gains more by the increase of equatorial
+radius than it loses by diminution of rotation.
+Therefore the slower the rotation the broader the
+disk, and, if the rotation were infinitely slow, the
+liquid would be an infinitely thin, flat, circular
+disk.
+
+The cheese-like form differs in an important
+respect from the orange-like form. If it were
+slightly disturbed, it would break up, probably
+into a number of detached pieces. The nature
+of the break-up would depend on the disturbance
+from which it started, but it is impossible to trace
+the details of the rupture in any case. We say
+then that the cheese shape is an unstable figure
+of equilibrium of a rotating mass of liquid.
+\PageSep{322}
+\index{Stability!of figures of equilibrium}%
+
+The third form is strikingly different from
+either of the preceding ones. We must now imagine
+the liquid to be shaped like a long cigar,
+and to be rotating about a central axis perpendicular
+to its length. Here again the ends of
+the cigar are so distant from the axis of rotation
+that the centrifugal force is great, and with infinitely
+slow rotation the figure becomes infinitely
+long and thin. Now this form resembles the
+cheese in being unstable. It is remarkable that
+these three forms are independent of the scale on
+which they are constructed, for they are perfectly
+similar whether they contain a few pounds of
+water or millions of tons.\footnote
+ {It is supposed that they are more than a fraction of an inch
+ across, otherwise surface tension would be called into play.}
+If the period of rotation
+and the density of the liquid are given,
+the shapes are absolutely determinable.
+
+The first of the three figures resembles the
+earth and may be called the planetary figure, and
+\index{Planetary figure of equilibrium of rotating liquid}%
+I may continue to refer to the other two as the
+cheese shape and the cigar shape. The planetary
+and cheese shape are sometimes called the spheroids
+of Maclaurin, after their discoverer, and
+\index{Maclaurin!figure of equilibrium of rotating liquid|(}%
+the cigar shape is generally named after Jacobi,
+\index{Jacobi, figure of equilibrium of rotating liquid|(}%
+the great German mathematician. For slow rotations
+the planetary form is stable, and the
+cheese and cigar are unstable. There are probably
+other possible forms of equilibrium, such as
+a ring, or several rings, or two detached masses
+\PageSep{323}
+revolving about one another like a planet and
+satellite, but for the present I only consider these
+three forms.
+
+Now imagine three equal masses of liquid, infinitely
+distant from one another, and each rotating
+\Figure[0.85]{37}{}{png}
+at the same slow speed, and let one of them
+have the planetary shape, the second the cheese
+shape, and the third the cigar shape. When the
+rotations are simultaneously and equally augmented,
+we find the planetary form becoming
+flatter, the cheese form shrinking in diameter
+and thickening, and the cigar form shortening
+and becoming fatter. There is as yet no change
+in the stability, the first remaining stable and
+\index{Stability!of figures of equilibrium}%
+the second and third unstable. The three figures
+are illustrated in~\fig{37}, but the cigar shape
+is hardly recognizable by that name, since it has
+already become quite short and its girth is
+considerable.
+\PageSep{324}
+
+Now it has been proved that as the cigar shape
+shortens, its tendency to break up becomes less
+marked, or in other words its degree of instability
+diminishes. At a certain stage, not as yet
+exactly determined, but which probably occurs
+when the cigar is about twice as long as broad,
+the instability disappears and the cigar form just
+becomes stable. I shall have to return to the
+consideration of this phase later. The condition
+of the three figures is now as follows: The planetary
+form of Maclaurin has become much flattened,
+but is still stable; the cigar form of Jacobi
+has become short and thick, and is just stable;
+and the cheese form of Maclaurin is still unstable,
+\index{Maclaurin!figure of equilibrium of rotating liquid|)}%
+but its diameter has shrunk so much that the
+figure might be better described as a very flat
+orange.
+
+On further augmenting the rotation the form
+of Jacobi still shrinks in length and increases in
+girth, until its length becomes equal to its
+greater breadth. Throughout the transformation
+the axis of rotation has always remained the
+shortest of the three, so that when the length
+becomes equal to the shorter equatorial diameter,
+the shape is not spherical, but resembles that of
+a much flattened orange. In fact, at this stage
+Jacobi's figure of equilibrium has degenerated to
+\index{Jacobi, figure of equilibrium of rotating liquid|)}%
+identity with the planetary shape. One of the
+upper ovals in \fig{38} represents the section of
+the form in which the planetary figure and the
+\PageSep{325}
+cigar figure coalesce, the former by continuous
+flattening, the latter by continuous shortening.
+The other upper figure represents the form to
+which the cheese-like figure of Maclaurin has
+\Figure[0.85]{38}{}{png}
+\index{Poincaré!figure of rotating liquid}%
+been reduced; it will be observed that it presents
+some resemblance to the coalescent form.
+
+When the rotation is further augmented, there
+is no longer the possibility of an elongated Jacobian
+figure, and there remain only the two
+spheroids of Maclaurin. But an important change
+has now supervened, for both these are now unstable,
+and indeed no stable form consisting of a
+single mass of liquid has yet been discovered.
+
+Still quickening the rotation, the two remaining
+forms, both unstable, grow in resemblance to
+one another, until at length they become identical
+in shape. This limiting form of Maclaurin's
+spheroids is shown in the lower part of~\fig{38}.
+If the liquid were water, it must rotate in $2$~hours
+\PageSep{326}
+$25$~minutes to attain this figure, but it would be
+unstable.
+
+A figure for yet more rapid rotation has not
+been determined, but it seems probable that
+dimples would be formed on the axis, that the
+dimples would deepen until they met, and that
+the shape would then be annular. The actual
+existence of such figures in Plateau's experiment
+is confirmatory of this conjecture.
+
+We must now revert to the consideration of
+the cigar-shaped figure of Jacobi, at the stage
+when it has just become stable. The whole of
+this argument depends on the fact that any figure
+of equilibrium is a member of a continuous
+series of figures of the same class, which gradually
+transforms itself as the rotation varies. Now
+M.~Poincaré has proved that, when we follow a
+\index{Poincaré!law of interchange of stability}%
+given series of figures and find a change from instability
+to stability, we are, as it were, served with
+a notice that there exists another series of figures
+coalescent with the first at that stage. We have already
+seen an example of this law, for the planetary
+figure of Maclaurin changed from stability
+to instability at the moment of its coalescence
+with the figure of Jacobi. Now I said that when
+the cigar form of Jacobi was very long it was
+unstable, but that when its length had shrunk to
+about twice its breadth it became stable; hence
+we have notice that at the moment of change
+another series of forms was coalescent with the
+\PageSep{327}
+cigar. It follows also from Poincaré's investigation
+\index{Poincaré!law of interchange of stability}%
+\index{Poincaré!figure of rotating liquid}%
+that the other series of forms must have
+been stable before the coalescence.
+
+Let us imagine then a mass of liquid in the
+form of Jacobi's cigar-shaped body rotating at
+the speed which just admits of stability, and let
+us pursue the series of changes backwards by
+making it rotate a little slower. We know that
+this retardation of rotation lengthens Jacobi's
+figure, and induces instability, but Poincaré has
+not only proved the existence and stability of the
+other series, but has shown that the shape is
+something like a pear.
+
+Poincaré's figure is represented approximately
+in~\fig{38}, but the mathematical difficulty of the
+problem has been too great to admit of an absolutely
+exact drawing. The further development
+of the pear shape is unknown, when the rotation
+slackens still more. There can, however, be
+hardly any doubt that the pear becomes more
+constricted in the waist, and begins to resemble
+an hour-glass; that the neck of the hour-glass
+becomes thinner, and that ultimately the body
+separates into two parts. It is of course likewise
+unknown up to what stage in these changes
+Poincaré's figure retains its stability.
+
+I have myself attacked this problem from an
+entirely different point of view, and my conclusions
+throw an interesting light on the subject,
+although they are very imperfect in comparison
+\PageSep{328}
+\index{Darwin, G. H.!hour-glass figure of rotating liquid|(}%
+with Poincaré's masterly work. To understand
+this new point of view, we must consider a new
+series of figures, namely that of a liquid planet
+attended by a liquid satellite. The two bodies
+are supposed to move in a circle round one another,
+and each is also to revolve on its axis at
+such a speed as always to exhibit the same face
+to its neighbor. Such a system, although divided
+into two parts, may be described as a figure of
+equilibrium. If the earth were to turn round
+once in twenty-seven days, it would always show
+to the moon the same side, and the moon actually
+does present the same side to us. In this
+case the earth and the moon would form such a
+system as that I am describing. Both the planet
+and the satellite are slightly flattened by their
+rotations, and each of them exercises a tidal influence
+on the other, whereby they are elongated
+towards the other.
+
+The system then consists of a liquid planet
+and liquid satellite revolving round one another,
+so as always to exhibit the same face to one another,
+and each tidally distorting the other. It
+is certain that if the two bodies are sufficiently
+far apart the system is a stable one, for if any
+slight disturbance be given, the whole system will
+not break up. But little is known as yet as to
+the limiting proximity of the planet and satellite,
+which will insure stability.
+
+Now if the rotations and revolutions of the
+\PageSep{329}
+bodies be accelerated, the two masses must be
+brought nearer together in order that the greater
+attraction may counterbalance the centrifugal
+force. But as the two are brought nearer the
+tide-generating force increases in intensity with
+great rapidity, and accordingly the tidal elongation
+of the two bodies is much augmented.
+
+A time will at length come when the ends of
+the two bodies will just touch, and we then have
+a form shaped like an hour-glass with a very
+\Figure[0.7]{39}{Hour-glass Figure of Equilibrium}{png}
+\PageLabel[pg]{329}% [** TN: Used by reference on p. 356 of the original]
+thin neck. The form is clearly Poincaré's figure,
+at an advanced stage of its evolution.
+
+The figure~\figref{39} shows the form of one possible
+\PageSep{330}
+figure of this class; it arises from the coalescence
+of two equal masses of liquid, and the
+shape shown was determined by calculation.
+But there are any number of different sorts of
+hour-glass shapes, according to the relative sizes
+of the planet and satellite which coalesce; and
+in order to form a continuous series with Poincaré's
+pear, it would be necessary to start with
+a planet and satellite of some definitely proportionate
+sizes. Unfortunately I do not know
+what the proportion may be. There are, however,
+certain indications which may ultimately
+lead to a complete knowledge of the series of
+figures from Jacobi's cigar shape down to the
+planet and satellite. It may be shown---and I
+shall have in \Ref{Chapter}{XX}.\ to consider the point
+more in detail---that if our liquid satellite had
+only, say, a thousandth of the mass of the planet,
+and if the two bodies were brought nearer one
+another, at a certain calculable distance the tidal
+action of the big planet on the very small satellite
+would become so intense that it would tear
+it to pieces. Accordingly the contact and coalescence
+of a very small satellite with a large
+planet is impossible. It is, however, certain that
+a large enough satellite---say of half the mass
+of the planet---could be brought up to contact
+with the planet, without the tidal action of the
+planet on the satellite becoming too intense to
+admit of the existence of the latter. There
+\PageSep{331}
+must then be some mass of the satellite, which
+will just allow the two to touch at the same
+moment that the tidal action of the larger on
+the smaller body is on the point of disrupting
+it. Now I suspect, although I do not know,
+that the series of figures which we should find in
+this case is in fact Poincaré's series. This discussion
+shows that the subject still affords an
+interesting field for future mathematicians.
+
+These investigations as to the form of rotating
+masses of liquid are of a very abstract character,
+and seem at first sight remote from practical
+conclusions, yet they have some very interesting
+applications.
+
+The planetary body of Maclaurin is flattened
+at the poles like the actual planets, and the
+degree of its flattening is exactly appropriate to
+the rapidity of its rotation. Although the planets
+are, at least in large part, composed of solid
+matter, yet that matter is now, or was once,
+sufficiently plastic to permit it to yield to the
+enormous forces called into play by rotation and
+gravitation. Hence it follows that the theory
+of Maclaurin's figure is the foundation of that
+of the figures of planets, and of the variation of
+gravity at the various parts of their surfaces.
+In the liquid considered hitherto, every particle
+attracted every other particle, the fluid was
+equally dense throughout, and the figure assumed
+was the resultant of the battle between
+\PageSep{332}
+\index{Figure of planets and their density}%
+\index{Gravity, variation according to latitude}%
+\index{Saturn!law of density and figure}%
+the centrifugal force and gravitation. At every
+part of the liquid the resultant attraction was
+directed nearly, but not quite, towards the
+centre of the shape. But if the attraction had
+everywhere been directed exactly to the centre,
+the degree of flattening would have been
+diminished. We may see that this must be so,
+because if the rotation were annulled, the mass
+would be exactly spherical, and if the rotation
+were not annulled, yet the forces would be such
+as to make the fluid pack closer, and so assume
+a more nearly spherical form than when the
+forces were not absolutely directed to the centre.
+It may be shown in fact that the flattening is
+$2\frac{1}{2}$~times greater in the case of Maclaurin's
+body than it is when the seat of gravitation is
+exactly central.
+
+In the case of actual planets the denser matter
+\index{Planets!figures and internal densities}%
+must lie in the centre and the less dense outside.
+If the central matter were enormously
+denser than superficial rock, the attraction would
+be directed towards the centre. There are then
+two extreme cases in which the degree of flattening
+can be determined,---one in which the density
+\index{Density!of planets determinable from their figures}%
+of the planet is the same all through, giving
+Maclaurin's figure; the other when the density
+is enormously greater at the centre. The flattening
+in the former is $2\frac{1}{2}$~times as great as in
+the latter. The actual condition of a real planet
+must lie between these two extremes. The
+\index{Darwin, G. H.!hour-glass figure of rotating liquid|)}%
+\PageSep{333}
+\index{Figure of planets and their density}%
+knowledge of the rate of rotation of a planet
+and of the degree of its flattening furnishes us
+with some insight into the law of its internal
+density. If it is very much less flat than Maclaurin's
+\index{Density!of planets determinable from their figures}%
+figure, we conclude that it is very dense
+in its central portion. In this way it is known
+with certainty that the central portions of the
+planets Jupiter and Saturn are much denser,
+\index{Jupiter!figure and law of internal density}%
+\index{Planets!figures and internal densities}%
+compared with their superficial portions, than is
+the case with the earth.
+
+I do not propose to pursue this subject into
+the consideration of the law of the variation of
+gravity on the surface of a planet; but enough
+has been said to show that these abstract investigations
+have most important practical applications.
+
+\begin{Authorities}
+Plateau, ``Mémoires de l'Académie Royale de~Belgique,''
+vol.~xvi. 1843.
+
+Thomson and Tait's \Title{Natural Philosophy} or other works on
+hydrodynamics give an account of figures of equilibrium.
+
+Poincaré, \Title{Sur l'équilibre d'une masse fluide animée d'un mouvement
+de rotation}, ``Acta Mathematica,'' vol.~7, 1885.
+
+An easier and different presentation of the subject is contained
+in an inaugural dissertation by Schwarzschild (Annals of Munich
+\index{Schwarzschild!exposition of Poincaré's theory}%
+Observatory, vol.~iii. 1896). He considers that Poincaré's
+proof of the stability of his figure is not absolutely conclusive.
+
+G.~H. Darwin, \Title{Figures of Equilibrium of Rotating Masses of
+\index{Darwin, G. H.!Jacobi's ellipsoid}%
+Fluid}, ``Transactions of Royal Society,'' vol.~178, 1887.
+
+G.~H. Darwin, \Title{Jacobi's Figure of Equilibrium},~\&c., ``Proceedings
+\index{Equilibrium, figures of, of rotating liquid|)}%
+\index{Figure of equilibrium!of rotating liquid|)}%
+Roy.\ Soc.,'' vol.~41, 1886, p.~319.
+
+S.~Krüger, \Title{Ellipsoidale Evenwichtsvormen},~\&c., Leeuwen, Leiden,
+\index{Krüger, figures of equilibrium of liquid}%
+1896; \Title{Sur l'ellipsoïde de Jacobi}, ``Nieuw Archief voor Wiskunde,''
+2d~series, 3d~part, 1898. The author shows that G.~H.
+Darwin had been forestalled in much of his work on Jacobi's
+figure, and he corrects certain mistakes.
+\end{Authorities}
+\index{Rotating liquid, figures of equilibrium|)}%
+\PageSep{334}
+
+
+\Chapter{XIX}
+{The Evolution of Celestial Systems}
+
+\First{Men} will always aspire to peer into the remote
+\index{Evolution of celestial systems|(}%
+\index{Solar!system, nebular hypothesis as to origin of|(}%
+past to the utmost of their power, and the fact
+that their success or failure cannot appreciably
+influence their life on the earth will never deter
+them from such endeavors. From this point
+of view the investigations explained in the last
+chapter acquire much interest, since they form
+the basis of the theories of cosmogony which
+seem most probable by the light of our present
+knowledge.
+
+We have seen that an annular figure of equilibrium
+\index{Kant!nebular hypothesis|(}%
+\index{Nebular hypothesis|(}%
+actually exists in Plateau's experiment,
+and it is almost certainly a possible form amongst
+celestial bodies. Plateau's ring has however
+only a transient existence, and tends to break up
+into globules, spinning on their axes and revolving
+round the centre. In this result we saw a
+close analogy with the origin of the planets, and
+regarded his experiment as confirmatory of the
+Nebular Hypothesis, of which I shall now give a
+short account.\footnote
+ {My knowledge of the history of the Nebular Hypothesis is
+ entirely derived from an interesting paper by Mr.~G.~F. Becker,
+\index{Becker, G. F., on Nebular Hypothesis}%
+ on ``Kant as a Natural Philosopher,'' \Title{American Journal of Science},
+ vol.~v. Feb.~1898.}
+\PageSep{335}
+
+The first germs of this theory are to be found
+in Descartes' ``Principles of Philosophy,'' published
+\index{Descartes, vortical theory of cosmogony}%
+in~1644. According to him the sun and
+planets were represented by eddies or vortices in
+a primitive chaos of matter, which afterwards
+formed the centres for the accretion of matter.
+As the theory of universal gravitation was propounded
+for the first time half a century later
+than the date of Descartes' book, it does not
+seem worth while to follow his speculations
+further. Swedenborg formulated another vortical
+cosmogony in~1734, and Thomas Wright of
+\index{Wright, Thomas, on a theory of cosmogony}%
+Durham published in 1750 a book of preternatural
+dullness on the same subject. It might not
+have been worth while to mention Wright, but
+that Kant acknowledges his obligation to him.
+
+The Nebular Hypothesis has been commonly
+associated with the name of Laplace, and he undoubtedly
+\index{Laplace!nebular hypothesis|(}%
+avoided certain errors into which his
+precursors had fallen. I shall therefore explain
+Laplace's theory, and afterwards show how he
+was, in most respects, really forestalled by the
+great German philosopher Kant.
+
+Laplace supposed that the matter now forming
+the solar system once existed in the form of a
+lens-shaped nebula of highly rarefied gas, that it
+rotated slowly about an axis perpendicular to the
+present orbits of the planets, and that the nebula
+extended beyond the present orbit of the furthest
+planet. The gas was at first expanded by heat,
+\PageSep{336}
+and as the surface cooled the central portion
+condensed and its temperature rose. The speed
+of rotation increased in consequence of the contraction,
+according to a well known law of mechanics
+called ``the conservation of moment of
+momentum;''\footnote
+ {Kant fell into error through ignorance of the generality of
+ this law, for he imagined that rotation could be generated from
+ rest.}
+the edges of the lenticular mass
+of gas then ceased to be continuous with the
+more central portion, and a ring of matter was
+detached, in much the same way as in Plateau's
+experiment. Further cooling led to further contraction
+and consequently to increased rotation,
+until a second ring was shed, and so on successively.
+The rings then ruptured and aggregated
+themselves into planets whilst the central nucleus
+formed the sun.
+
+Virtually the same theory had been propounded
+by Kant many years previously, but I am not
+aware that there is any reason to suppose that
+Laplace had ever read Kant's works. In a paper,
+to which I have referred above, Mr.~G.~F.
+Becker makes the following excellent summary
+\index{Becker, G. F., on Nebular Hypothesis|(}%
+of the relative merits of Kant and Laplace; he
+writes:---
+
+``Kant seems to have anticipated Laplace almost
+completely in the more essential portions
+of the nebular hypothesis. The great Frenchman
+was a child when Kant's theory was issued,
+\PageSep{337}
+and the `Système du Monde,' which closes with
+the nebular hypothesis, did not appear until
+1796. Laplace, like Kant, infers unity of origin
+for the members of the solar system from the
+similarity of their movements, the small obliquity
+and small eccentricity of the orbits of either
+planets or satellites.\footnote
+ {``The retrograde satellites of Uranus were discovered by
+ Herschel in~1787, but Laplace in his hypothesis does not refer to
+ them.''}
+Only a fluid extending
+throughout the solar system could have produced
+such a result. He is led to conclude that the
+atmosphere of the sun, in virtue of excessive
+heat, originally extended beyond the solar system
+and gradually shrank to its present limits. This
+nebula was endowed with moment of momentum
+which Kant tried to develop by collisions. Planets
+formed from zones of vapor, which on breaking
+agglomerated\dots. The main points of
+comparison between Kant and Laplace seem to
+be these. Kant begins with a cold, stationary
+nebula which, however, becomes hot by compression
+and at its first regenesis would be in a state
+of rotation. It is with a hot, rotating nebula
+that Laplace starts, without any attempt to account
+for the heat. Kant supposes annular
+zones of freely revolving nebulous matter to
+gather together by attraction during condensation
+of the nebula. Laplace supposes rings left
+\index{Laplace!nebular hypothesis|)}%
+behind by the cooling of the nebula to agglomerate
+\PageSep{338}
+in the same way as Kant had done. While
+both appeal to the rings of Saturn as an example
+of the hypothesis, neither explains satisfactorily
+why the planetary rings are not as stable
+as those of Saturn. Both assert that the positive
+rotation of the planets is a necessary
+consequence of agglomeration, but neither is
+sufficiently explicit. The genesis of satellites is
+for each of them a repetition on a small scale of
+the formation of the system\dots. While Laplace
+assigns no cause for the heat which he ascribes
+to his nebula, Lord Kelvin goes further
+back and supposes a cold nebula consisting of
+separate atoms or of meteoric stones, initially
+possessed of a resultant moment of momentum
+equal or superior to that of the solar system.
+Collision at the centre will reduce them to a
+vapor which then expanding far beyond Neptune's
+orbit will give a nebula such as Laplace
+postulates.\footnote
+ {\Title{Popular Lectures}, vol.~i.\ p.~421.}
+Thus Kelvin goes back to the same
+initial condition as Kant, excepting that Kant
+endeavored (of course vainly) to develop a moment
+of momentum for his system from collisions.''\footnote
+ {Becker, \Title{Amer.\ Journ.\ Science}, vol.~v. 1898, pp.~107,~108.}
+\index{Becker, G. F., on Nebular Hypothesis|)}%
+
+There is good reason for believing that the
+Nebular Hypothesis presents a true statement in
+outline of the origin of the solar system, and of
+the planetary subsystems, because photographs
+%% Plate 3
+\TallFig[0.85]{40}{Nebula in Andromeda}{jpg}
+%% Facing page
+%[Blank Page]
+\PageSep{339}
+of nebulæ have been taken recently in which we
+can almost see the process in action. \Fig{40} is
+a reproduction of a remarkable photograph by
+Dr.~Isaac Roberts of the great nebula in the constellation
+\index{Roberts, I., photograph of nebula in Andromeda}%
+of Andromeda. In it we may see the
+\index{Andromeda, nebula in}%
+\index{Nebula in Andromeda}%
+lenticular nebula with its central condensation,
+the annulation of the outer portions, and even
+the condensations in the rings which will doubtless
+at some time form planets. This system is
+built on a colossal scale, compared with which
+our solar system is utterly insignificant. Other
+\index{Solar!system, nebular hypothesis as to origin of|)}%
+\index{Solar!system, distribution of satellites in|(}%
+nebulæ show the same thing, and although they
+are less striking we derive from them good
+grounds for accepting this theory of evolution
+as substantially true.
+
+\TB
+
+I explained in \Ref{Chapter}{XVI}.\ how the theory
+of tidal friction showed that the moon took her
+origin very near to the present surface of the
+earth. But it was also pointed out that the same
+theory cannot be invoked to explain an origin
+for the planets at a point close to the sun. They
+must in fact have always moved at nearly their
+present distances. In the same way the dimensions
+of the orbits of the satellites of Mars, Jupiter,
+\index{Satellites!distribution of, in solar system|(}%
+Saturn, and Neptune cannot have been
+largely augmented, whatever other effects tidal
+friction may have had. We must therefore still
+rely on the Nebular Hypothesis for the explanation
+\index{Nebular hypothesis|)}%
+of the main features of the system as a
+whole.
+\index{Kant!nebular hypothesis|)}%
+\PageSep{340}
+
+It may, at first sight, appear illogical to maintain
+that an action, predominant in its influence
+on our satellite, should have been insignificant
+in regulating the orbits of all the other bodies
+of the system. But this is not so, for whilst the
+earth is only $80$~times as heavy as the moon, Saturn
+weighs about $4,600$~times as much as its
+satellite Titan, which is by far the largest satellite
+in the solar system; and all the other satellites
+are almost infinitesimal in comparison with their
+primaries. Since, then, the relationship of the
+moon to the earth is unique, it may be fairly contended
+that a factor of evolution, which has been
+predominant in our own history, has been relatively
+insignificant elsewhere.
+
+There is indeed a reason explanatory of this
+singularity in the moon and earth; it lies in the
+fact that the earth is nearer to the sun than any
+other planet attended by a satellite. To explain
+the bearing of this fact on the origin of satellites
+and on their sizes, I must now show how tidal
+friction has probably operated as a perturbing
+influence in the sequence of events, which would
+be normal according to the Nebular Hypothesis.
+
+We have seen that rings should be shed from
+the central nucleus, when the contraction of the
+nebula has induced a certain degree of augmentation
+of rotation. Now if the rotation were
+retarded by some external cause, the genesis of
+a ring would be retarded, or might be entirely
+prevented.
+\PageSep{341}
+
+The friction of the solar tides in a planetary
+nebula furnishes such an external cause, and accordingly
+the rotation of a planetary nebula near
+to the sun might be so much retarded that a ring
+would never be detached from it, and no satellite
+would be generated. From this point of view
+it is noteworthy that Mercury and Venus have
+no satellites; that Mars has two, Jupiter five,
+and that all the exterior planets have several
+satellites. I suggest then that the solar tidal
+friction of the terrestrial nebula was sufficient to
+retard the birth of a satellite, but not to prevent
+it, and that the planetary mass had contracted
+to nearly the present dimensions of the earth
+and had partially condensed into the solid and
+liquid forms, before the rotation had augmented
+sufficiently to permit the birth of a satellite.
+When satellites arise under conditions which are
+widely different, it is reasonable to suppose that
+their masses will also differ much. Hence we can
+understand how it has come about that the relationship
+between the moon and the earth is so
+unlike that between other satellites and their
+planets. In \Ref{Chapter}{XVII}.\ I showed that there
+are reasons for believing that solar tidal friction
+\index{Solar!system, distribution of satellites in|)}%
+has really been an efficient cause of change, and
+this makes it legitimate to invoke its aid in explaining
+the birth and distribution of satellites.
+\index{Satellites!distribution of, in solar system|)}%
+
+\TB
+
+In speaking of the origin of the moon I have
+\PageSep{342}
+\index{Eccentricity of orbit!theory of, in case of double stars}%
+\index{See, T. J. J.!theory of evolution of double stars|(}%
+been careful not to imply that the matter of
+which she is formed was necessarily first arranged
+in the form of a ring. Indeed, the genesis of
+the hour-glass figure of equilibrium from Jacobi's
+form and its fission into two parts indicate the
+possibility of an entirely different sequence of
+events. It may perhaps be conjectured that the
+moon was detached from the primitive earth in
+this way, possibly with the help of tidal oscillations
+due to the solar action. Even if this suggestion
+is only a guess, it is interesting to make
+such speculations, when they have some basis of
+reason.
+
+In recent years astronomers have been trying,
+principally by aid of the spectroscope, to determine
+the orbits of pairs of double stars around
+\index{Stars!theory of evolution|(}%
+one another. It has been observed that, in the
+majority of these systems, the masses of the two
+component stars do not differ from one another
+extremely; and Dr.~See, who has specially devoted
+himself to this research, has drawn attention
+to the great contrast between these systems
+and that of the sun, attended by a retinue of
+infinitesimal planets. He maintains, with justice,
+that the paths of evolution pursued in the two
+cases have probably also been strikingly different.
+
+It is hardly credible that two stars should
+have gained their present companionship by an
+accidental approach from infinite space. They
+cannot always have moved as they do now, and
+\PageSep{343}
+so we are driven to reflect on the changes which
+might supervene in such a system under the
+action of known forces.
+
+The only efficient interaction between a pair
+of celestial bodies, which is known hitherto, is
+a tidal one, and the friction of the oscillations
+introduces a cause of change in the system.
+Tidal friction tends to increase the eccentricity
+of the orbit in which two bodies revolve about
+one another, and its efficiency is much increased
+when the pair are not very unequal in mass and
+when each is perturbed by the tides due to the
+other. The fact that the orbits of the majority
+of the known pairs are very eccentric affords a
+reason for accepting the tidal explanation. The
+only adverse reason, that I know of, is that the
+eccentricities are frequently so great that we
+may perhaps be putting too severe a strain on
+the supposed cause.
+
+But the principal effect of tidal friction is the
+repulsion of the two bodies from one another,
+so that when their history is traced backwards
+we ultimately find them close together. If then
+this cause has been as potent as Dr.~See believes
+it to have been, the two components of a binary
+system must once have been close together.
+From this stage it is but a step to picture to
+ourselves the rupture of a nebula, in the form
+of an hour-glass, into two detached masses.
+
+The theory embraces all the facts of the case,
+\PageSep{344}
+and as such is worthy of at least a provisional
+acceptance. But we must not disguise from
+ourselves that out of the thousands, and perhaps
+millions of double stars which may be visible
+from the earth, we only as yet know the orbits
+and masses of a dozen.
+
+Many years ago Sir John Herschel drew a
+\index{Herschel, observations of twin nebulæ}%
+number of twin nebulæ as they appear through
+a powerful telescope. The drawings probably
+possess the highest degree of accuracy attainable
+by this method of delineation, and the shapes
+present evidence confirmatory of the theory of
+the fission of nebulæ adopted by Dr.~See. But
+since Herschel's time it has been discovered that
+many details, to which our eyes must remain forever
+blind, are revealed by celestial photography.
+The photographic film is, in fact, sensitive to
+those ``actinic'' rays which we may call invisible
+light, and many nebulæ are now found to be
+hardly recognizable, when photographs of them
+are compared with drawings. A conspicuous
+example of this is furnished by the great nebula
+in Andromeda, illustrated above in~\fig{40}.
+
+Photographs, however, do not always aid interpretation,
+for there are some which serve only
+to increase the chaos visible with the telescope.
+We may suspect, indeed, that the complete system
+of a nebula often contains masses of cold
+and photographically invisible gas, and in such
+cases it would seem that the true nature of the
+whole will always be concealed from us.
+\PageSep{345}
+
+Another group of strange celestial objects is
+that of the spiral nebulæ, whose forms irresistibly
+\index{Nebulae@Nebulæ, description of various}%
+suggest violent whirlpools of incandescent
+gas. Although in all probability the motion of
+the gas is very rapid, yet no change of form has
+been detected. We are here reminded of a
+rapid stream rushing past a post, where the form
+of the surface remains constant whilst the water
+itself is in rapid movement; and it seems reasonable
+to suppose that in these nebulæ it is
+only the lines of the flow of the gas which are
+visible. Again, there are other cases in which
+the telescopic view may be almost deceptive in
+its physical suggestions. Thus the Dumb-Bell
+\index{Dumb-bell nebula, description of photograph of}%
+nebula (27~Messier Vulpeculæ), as seen telescopically,
+might be taken as a good illustration of a
+nebula almost ready to split into two stars. If
+this were so, the rotation would be about an
+axis at right angles to the length of the nebula.
+But a photograph of this object shows that the
+system really consists of a luminous globe surrounded
+by a thick and less luminous ring, and
+that the opacity of the sides of the ring takes a
+bite, as it were, out of each side of the disk, and
+so gives it the apparent form of a dumb-bell.
+In this case the rotation must be about an axis
+at right angles to the ring, and therefore along
+the length of the dumb-bell. It is proper to
+add that Dr.~See is well aware of this, and does
+not refer to this nebula as a case of incipient
+fission.
+\PageSep{346}
+
+I have made these remarks in order to show
+that every theory of stellar evolution must be
+full of difficulty and uncertainty. According to
+our present knowledge Dr.~See's theory appears
+to have much in its favor, but we must await its
+confirmation or refutation from the results of
+future researches with the photographic plate,
+the spectroscope, and the telescope.
+
+\begin{Authorities}
+Mr.~G.~F. Becker (\Title{Amer.\ Jour.\ Science}, vol.~v. 1898, art.~xv.)\
+gives the following references to Kant's work: \Title{Sämmtliche
+Werke}, ed.~Hartenstein, 1868 (Tidal Friction and the Aging of
+the Earth), vol.~i.\ pp.~179--206; (Nebular Hypothesis), vol.~i.\
+pp.~207--345.
+
+Laplace, \Title{Système du Monde}, last appendix; the tidal retardation
+of the moon's rotation is only mentioned in the later
+editions.
+
+T.~J.~J. See, \Title{Die Entwickelung der Doppelstern-systeme}, ``Inaugural
+Dissertation,'' 1892. Schade, Berlin.
+
+T.~J.~J. See, \Title{Evolution of the Stellar Systems}, vol.~i.\ 1896.
+Nichols Press, Lynn, Massachusetts. Also a popular article,
+\Title{The Atlantic Monthly}, October, 1897.
+
+G.~H. Darwin, \Title{Tidal Friction~\dots\ and Evolution}, ``Phil.\ Trans.\
+\index{Darwin, G. H.!evolution of satellites}%
+Roy.\ Soc.,'' part~ii.\ 1881, p.~525.
+\end{Authorities}
+\index{Evolution of celestial systems|)}%
+\index{See, T. J. J.!theory of evolution of double stars|)}%
+\index{Stars!theory of evolution|)}%
+\PageSep{347}
+
+
+\Chapter[Saturn's Rings]{XX}
+{Saturn's Rings\protect\footnotemark}
+
+\footnotetext{Part of this chapter appeared as an article in \Title{Harper's
+ Magazine} for June,~1889.}
+
+\First{To} the naked eye Saturn appears as a brilliant
+\index{Saturn!description and picture|(}%
+star, which shines, without twinkling, with a
+yellowish light. It is always to be found very
+nearly in the ecliptic, moving slowly amongst
+the fixed stars at the rate of only thirteen degrees
+per annum. It is the second largest
+planet of the solar system, being only exceeded
+in size by the giant Jupiter. It weighs $91$~times
+as much as our earth, but, being as light as cork,
+occupies $690$~times the volume, and is nine times
+as great in circumference. Notwithstanding its
+great size it rotates around its axis far more
+rapidly than does the earth, its day being only
+$10\frac{1}{2}$~of our hours. It is ten times as far from
+the sun as we are, and its year, or time of revolution
+round the sun, is equal to thirty of our
+years. It was deemed by the early astronomers
+to be the planet furthest from the sun, but that
+was before the discovery by Herschel, at the
+end of the last century, of the further planet
+Uranus, and that of the still more distant Neptune
+by Adams and Leverrier in the year~1846.
+\PageSep{348}
+
+The telescope has shown that Saturn is attended
+by a retinue of satellites almost as numerous
+as, and closely analogous to, the planets
+circling round the sun. These moons are eight
+in number, are of the most various sizes, the
+largest as great as the planet Mars, and the
+smallest very small, and are equally diverse in
+respect of their distances from the planet. But
+besides its eight moons Saturn has another attendant
+absolutely unique in the heavens; it is
+girdled with a flat ring, which, like the planet
+itself, is only rendered visible to us by the
+illumination of sunlight. \Fig{41}, to which
+further reference is made below, shows the general
+appearance of the planet and of its ring.
+The theory of the physical constitution of that
+ring forms the subject of the present chapter.
+
+A system so rich in details, so diversified and
+so extraordinary, would afford, and doubtless
+has afforded, the subject for many descriptive
+essays; but description is not my present object.
+
+The existence of the ring of Saturn seems
+now a very commonplace piece of knowledge,
+and yet it is not $300$~years since the moons of
+Jupiter and Saturn were first detected, and since
+suspicion was first aroused that there was something
+altogether peculiar about the Saturnian
+system. These discoveries, indeed, depended
+entirely on the invention of the telescope. It
+may assist the reader to realize how necessary
+\PageSep{349}
+%[** TN: Oriented vertically in the original]
+\Figure{41}{The Planet Saturn}{jpg}
+\PageSep{350}
+the aid of that instrument was when I say that
+Saturn, when at his nearest to us, is the same in
+size as a sixpenny piece held up at a distance of
+$210$~yards.
+
+It was the celebrated Galileo who first invented
+\index{Galileo!Saturn's ring}%
+a combination of lenses such as is still
+used in our present opera-glasses, for the purpose
+of magnifying distant objects.
+
+In July of~1610 he began to examine Saturn
+with his telescope. His most powerful instrument
+only magnified $32$~times, and although
+such an enlargement should have amply sufficed
+to enable him to make out the ring, yet he persuaded
+himself that what he saw was a large
+bright disk, with two smaller ones touching it,
+one on each side. His lenses were doubtless
+imperfect, but the principal cause of his error
+must have been the extreme improbability of the
+existence of a ring girdling the planet. He
+wrote an account of what he had seen to the
+Grand Duke of Tuscany, Giuliano de'~Medici,
+and to others; he also published to the world an
+anagram which, when the letters were properly
+arranged, read as follows: ``Altissimum planetam
+tergeminum observavi'' (I have seen the
+furthest planet as triple), for it must be remembered
+that Saturn was then the furthest known
+planet.
+
+In 1612 Galileo again examined Saturn, and
+was utterly perplexed and discouraged to find
+\PageSep{351}
+his triple star replaced by a single disk. He
+writes, ``Is it possible that some mocking demon
+has deceived me?'' And here it may be well to
+remark that there are several positions in which
+Saturn's rings vanish from sight, or so nearly
+vanish as to be only visible with the most powerful
+modern telescopes. When the plane of the
+ring passes through the sun, only its very thin
+edge is illuminated; this was the case in~1612,
+when Galileo lost it; secondly, if the plane of
+the ring passes through the earth, we have only
+a very thin edge to look at; and thirdly, when
+the sun and the earth are on opposite sides of
+the ring, the face of the ring which is presented
+to us is in shadow, and therefore invisible.
+
+Some time afterwards Galileo's perplexity was
+increased by seeing that the planet had then a
+pair of arms, but he never succeeded in unraveling
+the mystery, and blindness closed his career
+as an astronomer in~1626.
+
+About thirty years after this, the great Dutch
+astronomer Huyghens, having invented a new
+\index{Huyghens, discovery of Saturn's ring}%
+sort of telescope (on the principle of our present
+powerful refractors), began to examine the planet
+and saw that it was furnished with two loops or
+handles. Soon after the ring disappeared; but
+when, in~1659, it came into view again, he at
+last recognized its true character, and announced
+that the planet was attended by a broad, flat
+ring.
+\PageSep{352}
+
+A few years later it was perceived that there
+were two rings, concentric with one another.
+The division, which may be easily seen in drawings
+of the planet, is still named after Cassini,
+\index{Cassini, discovery of division in Saturn's rings}%
+one of its discoverers. Subsequent observers
+have detected other less marked divisions.
+
+Nearly two centuries later, namely, in~1850,
+Bond in America and Dawes in England, independently
+\index{Bond, discovery of inner ring of Saturn}%
+\index{Dawes, discovery of inner ring of Saturn}%
+and within a fortnight of the same
+time, observed that inside of the well-known
+bright rings there is another very faint dark
+ring, which is so transparent that the edge of
+the planet is visible through it. There is some
+reason to believe that this ring has really become
+more conspicuous within the last $200$~years,
+so that it would not be right to attribute the
+lateness of its detection entirely to the imperfection
+of earlier observations.
+
+It was already discovered in the last century
+that the ring is not quite of the same thickness
+at all points of its circumference, that it is not
+strictly concentric with the planet, and that it
+revolves round its centre. Herschel, with his
+magnificent reflecting telescope, detected little
+beads on the outer ring, and by watching these
+he concluded that the ring completes its revolution
+in $10\frac{1}{2}$~hours.
+
+This sketch of the discovery and observation
+of Saturn's rings has been necessarily very incomplete,
+but we have perhaps already occupied
+too much space with it.
+\PageSep{353}
+
+\Fig{41} exhibits the appearance of Saturn and
+his ring. The drawing is by Bond of Harvard
+University, and is considered an excellent one.
+
+It is usual to represent the planets as they are
+seen through an astronomical telescope, that is
+\Figure{42}{Diagram of Saturn and his Rings}{png}
+to say, reversed. Thus in \fig{41} the south
+pole of the planet is at the top of the plate, and
+unless the telescope were being driven by clockwork,
+the planet would appear to move across
+the field of view from right to left.
+
+The plane of the ring is coincident with the
+equator of the planet, and both ring and equator
+are inclined to the plane of the planet's orbit at
+an angle of $27$~degrees.
+
+A whole essay might be devoted to the discussion
+of this and of other pictures, but we must
+confine ourselves to drawing attention to the
+well-marked split, called Cassini's division, and
+\PageSep{354}
+to the faint internal ring, through which the
+edge of the planet is visible.
+
+The scale on which the whole system is constructed
+is best seen in a diagram of concentric
+circles, showing the limits of the planet's body
+and of the successive rings. Such a diagram,
+with explanatory notes, is given in~\fig{42}.
+
+An explanation of the outermost circle, called
+\emph{Roche's limit}, will be given later. The following
+are the dimensions of the system:---
+\begin{center}
+\begin{tabular}{l>{\qquad}r}
+Equatorial diameter of planet & $73,000$ miles \\
+Interior diameter of dark ring & $93,000$ \Ditto{miles} \\
+Interior diameter of bright rings & $111,000$ \Ditto{miles} \\
+Exterior diameter of bright rings & $169,000$ \Ditto{miles}
+\end{tabular}
+\end{center}
+We may also remark that the radius of the
+limit of the rings is $2.38$~times the mean radius
+of the planet, whilst Roche's limit is $2.44$~such
+radii. The greatest thickness of the ring is uncertain,
+but it seems probable that it does not
+exceed $200$~or $300$~miles.
+
+The pictorial interest, as we may call it, of all
+this wonderful combination is obvious, but our
+curiosity is further stimulated when we reflect on
+the difficulty of reconciling the existence of this
+strange satellite with what we know of our own
+planet and of other celestial bodies.
+
+It may be admitted that no disturbance to our
+ordinary way of life would take place if Saturn's
+\index{Saturn!description and picture|)}%
+rings were annihilated, but, as Clerk-Maxwell
+has remarked, ``from a purely scientific point of
+\PageSep{355}
+view, they become the most remarkable bodies in
+the heavens, except, perhaps, those still less \emph{useful}
+bodies---the spiral nebulæ. When we have
+actually seen that great arch swung over the
+equator of the planet without any visible connection,
+we cannot bring our minds to rest. We
+cannot simply admit that such is the case, and
+describe it as one of the observed facts of nature,
+not admitting or requiring explanation. We
+must either explain its motion on the principles
+of mechanics, or admit that, in Saturnian realms,
+there can be motion regulated by laws which we
+are unable to explain.''
+
+I must now revert to the subject of \Ref{Chapter}{XVIII}.\
+and show how the investigations, there
+explained, bear on the system of the planet. We
+then imagined a liquid satellite revolving in a
+circular orbit about a liquid planet, and supposed
+that each of these two masses moved so as always
+to present the same face to the other. It was
+pointed out that each body must be somewhat
+flattened by its rotation round an axis at right
+angles to the plane of the orbit, and that the
+tidal attraction of each must deform the other.
+In the application of this theory to the system of
+Saturn it is not necessary to consider further the
+tidal action of the satellite on the planet, and we
+must concentrate our attention on the action of
+the planet on the satellite. We have found reason
+to suppose that the earth once raised enormous
+\PageSep{356}
+\index{Saturn!theory of ring|(}%
+tides in the moon, when her body was
+molten, and any planet must act in the same way
+on its satellite. When, as we now suppose, the
+satellite moves so as always to present the same
+face to the planet, the tide is fixed and degenerates
+into a permanent distortion of the equator
+of the satellite into an elliptic shape. If the
+satellite is very small compared with its planet,
+and if it is gradually brought closer and closer
+to the planet, the tide-generating force, which
+varies inversely as the cube of the distance, increases
+with great rapidity, and we shall find the
+satellite to assume a more and more elongated
+shape. When the satellite is not excessively
+small, the two bodies may be brought together
+until they actually touch, and form the hour-glass
+figure exhibited in \fig{39}, \PageRef{p.}{329}.
+
+The general question of the limiting proximity
+of a liquid planet and satellite which just insures
+stability is as yet unsolved. But it has been
+proved that there is one case in which instability
+sets in. Édouard Roche has shown that this approach
+\index{Roche, E.!theory of limit and Saturn's ring|(}%
+up to contact is not possible when the
+satellite is very small, for at a certain distance
+the tidal distortion of a small satellite becomes
+so extreme that it can no longer subsist as a
+single mass of fluid. He also calculated the
+form of the satellite when it is elongated as much
+as possible. \Fig{43} represents the satellite in
+its limiting form. We must suppose the planet
+\PageSep{357}
+about which it revolves to be a large globe, with
+its centre lying on the prolongation of the longest
+axis of the egg-like body in the direction
+of~$E$. As it revolves, the longest axis of the satellite
+always points straight towards its planet.
+The egg, though not strictly circular in girth, is
+\Figure[0.7]{43}{Roche's Figure of a Satellite when elongated
+to the utmost}{png}
+very nearly so. Thus another section at right
+angles to this one would be of nearly the same
+shape. One diameter of the girth is in fact only
+longer than the other by a seventeenth part.
+The shortest of the three axes of the slightly flattened
+egg is at right angles to the plane of the
+orbit in which the satellite revolves. The longest
+axis of the body is nearly twice as long as
+either of the two shorter ones; for if we take
+the longest as~$\DPchg{1000}{1,000}$, the other two would be $496$
+and~$469$. \Fig{43} represents a section through
+the two axes equal respectively to~$\DPchg{1000}{1,000}$ and to~$469$,
+so that we are here supposed to be looking
+at the satellite's orbit edgewise.
+\PageSep{358}
+
+But, as I have said, Roche determined not
+\index{Earth and moon!Roche's limit for}%
+only the shape of the satellite when thus elongated
+to the utmost possible extent, but also in
+its nearness to the planet, and he proved that if
+the planet and satellite be formed of matter of
+the same density, the centre of such a satellite
+must be at a distance from the planet's centre of
+$2\frac{11}{25}$~of the planet's radius. This distance of $2\frac{11}{25}$
+or $2.44$~of a planet's radius I call Roche's limit
+for that planet. The meaning of this is that inside
+of a circle drawn around a planet at a distance
+so proportionate to its radius no small
+satellite can circulate; the reason being that if
+a lump of matter were started to revolve about
+the planet inside of that circle, it would be torn
+to pieces under the action of the forces we have
+been considering. It is true that if the lump of
+matter were so small as to be more properly described
+as a stone than as a satellite, then the
+cohesive force of stone might be strong enough
+to resist the disruptive force. But the size for
+which cohesion is sufficient to hold a mass of
+matter together is small compared with the
+smallest satellite.
+
+I have said that Roche's limit as evaluated at
+$2.44$~radii is dependent on the assumption of
+equal densities in the satellite and planet. If
+the planet be denser than the satellite, Roche's
+limit is a larger multiple of the planet's radius,
+and if it be less dense the multiple is smaller.
+\PageSep{359}
+But the variation of distance is not great for
+considerable variations in the relative densities
+of the two bodies, the law being that the~$2.44$
+must be multiplied by the cube root of the ratio
+of the density of the planet to that of the satellite.
+If for example the planet be on the average
+of its whole volume twice as dense as the
+satellite, the limit is only augmented from $2.44$
+to $3$~times the planet's radius; and if it be half
+as dense, the $2.44$ is depressed to~$1.94$. Thus
+the variation of density of the planet from a
+half to twice that of the planet---that is to
+say, the multiplication of the smaller density by
+four---only changes Roche's limit from $2$ to $3$~radii.
+It follows from this that, within pretty
+wide limits of variation of relative densities,
+Roche's limit changes but little.
+
+The only relative density of planet and satellite
+that we know with accuracy is that of the
+earth and moon. Now the earth is more dense
+than the moon in the proportion of $8$~to~$5$; hence
+Roche's limit for the earth is the cube root of~$\frac{8}{5}$
+multiplied by~$2.44$, that is to say, it is $2.86$~times
+the earth's radius. It follows that if the moon
+were to revolve at a distance of less than $2.86$~radii,
+or $11,000$~miles, she would be torn to pieces
+by the earth's tidal force.
+
+If this result be compared with the conclusions
+drawn from the theory of tidal friction, it follows
+that at the earliest stage to which the moon was
+\PageSep{360}
+\index{Saturn!Roche's limit for}%
+traced, she could not have existed in her present
+form, but the matter which is now consolidated
+in the form of a satellite must then have been a
+mere swarm of loose fragments. Such fragments,
+if concentrated in one part of the orbit, would
+be nearly as efficient in generating tides in the
+planet as though they were agglomerated in the
+form of a satellite. Accordingly the action of
+tidal friction does not necessitate the agglomeration
+of the satellite. The origin and earliest history
+of the moon must always remain highly
+speculative, and it seems fruitless to formulate
+exact theories on the subject.\footnote
+ {Mr.~Nolan has criticised the theory of tidal friction from
+\index{Nolan, criticism of tidal theory of moon's origin}%
+ this point of view (\Title{Genesis of the Moon}, Melbourne, 1885; also
+ \Title{Nature}, Feb.~18 and July~29, 1886).}
+
+When we apply this reasoning to the other
+planets, exact data are wanting. The planet
+Mars resembles the earth in so many respects
+\index{Mars!Roche's limit}%
+that it is reasonable to suppose that there is much
+the same relationship between the densities of
+the planet and satellites as with us. As with the
+case of the earth and moon, this would bring
+Roche's limit to $2.86$~times the planet's radius.
+The satellite Phobos, however, revolves at a
+distance of $2.75$~radii of Mars; hence we are
+bound to suppose that the density of Phobos is
+a very little more nearly equal to that of Mars
+than in the case of the moon and earth; if
+it were not so, Phobos would be disrupted by
+\PageSep{361}
+tidal action. How interesting it will be if future
+generations shall cease to see the satellite Phobos,
+for they will then conclude that Phobos has been
+drawn within the charmed circle, and has been
+broken to pieces.
+
+In considering the planets Jupiter and Saturn,
+\index{Jupiter!Roche's limit for}%
+we are deprived of the indications which are useful
+in the case of Mars. The satellites are probably
+solid, and these planets are known to have
+a low mean density. Hence it is probable that
+Roche's limit is a somewhat smaller multiple than
+$2.44$~of the radii of Jupiter and Saturn. The
+only satellite which is in danger is the innermost
+and recently discovered satellite of Jupiter, which
+revolves at $2.6$~times the planet's mean radius,
+for with the same ratio of densities as obtains
+here the satellite would be broken up. This confirms
+the conclusion that the mean density of
+Jupiter is at least not greater than that of the
+satellite.
+
+We are also ignorant of the relative densities
+of Saturn and its satellites, and so in the figure
+Roche's limit is placed at $2.44$~times the planet's
+radius, corresponding to equal densities. But
+the density of the planet is very small, and therefore
+the limit is almost certainly slightly nearer
+to the planet than is shown.
+
+This system affords the only known instance
+where matter is clearly visible circulating round
+an attractive centre at a distance certainly less
+\PageSep{362}
+than the theoretical limit, and the belief seems
+justified that Saturn's rings consist of dust and
+fragments.
+
+Although Roche himself dismissed this matter
+in one or two sentences, he saw the full bearing
+of his remarks, and to do him justice we should
+date from~1848 the proof that Saturn's rings
+consist of meteoric stones.
+
+The theoretical limit lies just outside the limit
+of the rings, but we may suspect that the relative
+densities of the planet and satellite are such that
+the limit should be displaced to a distance just
+inside of the outer edge of the ring, because any
+solid satellite would almost necessarily have a
+mean density greater than that of the planet.
+
+Although Roche's paper was published about
+fifty years ago, it has only recently been mentioned
+in text-books and general treatises. Indeed,
+it has been stated that Bond was the first
+in modern times to suggest the meteoric constitution
+of the rings. His suggestion, based on
+telescopic evidence, was however made in~1851.
+
+\TB
+
+And now to explain how a Cambridge mathematician
+to whom reference was made above, in
+ignorance of Roche's work of nine years before,
+\index{Roche, E.!theory of limit and Saturn's ring|)}%
+arrived at the same conclusion. In~1857, Clerk-Maxwell,
+one of the most brilliant men of science
+who have taught in the University of Cambridge,
+and whose early death we still deplore, attacked
+\PageSep{363}
+\index{Instability!of Saturn's ring}%
+the problem of Saturn's rings in a celebrated
+essay, which gained for him what is called the
+Adams prize. Laplace had early in the century
+considered the theory that the ring is solid, and
+Maxwell first took up the question of the motion
+\index{Maxwell on Saturn's ring|(}%
+of such a solid ring at the point where it had
+been left. He determined what amount of
+weighting at one point of a solid uniform ring is
+necessary to insure its steady motion round the
+planet. He found that there must be a mass
+attached to the circumference of the ring weighing
+$4\frac{1}{2}$~times as much as the ring itself. In fact,
+the system becomes a satellite with a light ring
+attached to it.
+
+``As there is no appearance,'' he says, ``about
+the rings justifying a belief in so great an irregularity,
+the theory of the solidity of the rings
+becomes very improbable. When we come to
+consider the additional difficulty of the tendency
+of the fluid or loose parts of the ring to accumulate
+at the thicker parts, and thus to destroy that
+nice adjustment of the load on which the stability
+depends, we have another powerful argument
+against solidity. And when we consider the immense
+size of the rings and their comparative
+thinness, the absurdity of treating them as rigid
+bodies becomes self-evident. An iron ring of
+such a size would be not only plastic, but semi-fluid,
+under the forces which it would experience,
+and we have no reason to believe these rings to
+\PageSep{364}
+\index{Instability!of Saturn's ring}%
+be artificially strengthened with any material
+unknown on this earth.''
+
+The hypothesis of solidity being condemned,
+Maxwell proceeds to suppose that the ring is
+composed of a number of equal small satellites.
+This is a step towards the hypothesis of an indefinite
+number of meteorites of all sizes. The
+consideration of the motion of these equal satellites
+affords a problem of immense difficulty, for
+each satellite is attracted by all the others and
+by the planet, and they are all in motion.
+
+If they were arranged in a circle round the
+planet at equal distances, they might continue to
+revolve round the planet, provided that each
+satellite remained in its place with mathematical
+exactness. Let us consider that the proper place
+of each satellite is at the ends of the spokes of
+a revolving wheel, and then let us suppose that
+none of them is exactly in its place, some being
+a little too far advanced, some a little behind,
+some too near and some too far from the centre
+of the wheel---that is to say, from the planet---then
+we want to know whether they will swing
+to and fro in the neighborhood of their places,
+or will get further and further from their places,
+and whether the ring will end in confusion.
+
+Maxwell treated this problem with consummate
+skill, and showed that if the satellites were
+not too large, confusion would not ensue, but
+each satellite would oscillate about its proper
+place.
+\PageSep{365}
+\index{Stability!of Saturn's ring}%
+
+At any moment there are places where the
+satellites are crowded and others where they are
+spaced out, and he showed that the places of
+crowding and of spacing out will travel round
+the ring at a different speed from that with
+which the ring as a whole revolves. In other
+words, waves of condensation and of rarefaction
+are propagated round the ring as it rotates.
+
+He constructed a model, now in the laboratory
+at Cambridge, to exhibit these movements; it is
+pretty to observe the changes of the shape of the
+ring and of the crowding of the model satellites
+as they revolve.
+
+I cannot sum up the general conclusions at
+which Maxwell arrived better than by quoting
+his own words.
+
+In the summary of his paper he says:---
+
+``If the satellites are unequal, the propagation
+of waves will no longer be regular, but the disturbances
+of the ring will in this, as in the
+former case, produce only waves, and not growing
+confusion. Supposing the ring to consist,
+not of a single row of large-satellites, but of a
+cloud of evenly distributed unconnected particles,
+we found that such a cloud must have a
+very small density in order to be permanent, and
+that this is inconsistent with its outer and inner
+parts moving with the same angular velocity.
+Supposing the ring to be fluid and continuous,
+we found that it will necessarily be broken up
+into small portions.
+\PageSep{366}
+\index{Stability!of Saturn's ring}%
+
+``We conclude, therefore, that the rings must
+consist of disconnected particles; these may be
+either solid or liquid, but they must be independent.
+The entire system of rings must therefore
+consist either of a series of many concentric
+rings, each moving with its own velocity, and
+having its own system of waves, or else of a confused
+multitude of revolving particles, not arranged
+in rings, and continually coming into
+collision with each other.
+
+``Taking the first case, we found that in an
+indefinite number of possible cases the mutual
+perturbation of two rings, stable in themselves,
+might mount up in time to a destructive magnitude,
+and that such cases must continually occur
+in an extensive system like that of Saturn, the
+only retarding cause being the possible irregularity
+of the rings.
+
+``The result of long-continued disturbance
+was found to be the spreading out of the rings
+in breadth, the outer rings pressing outward,
+while the inner rings press inward.
+
+``The final result, therefore, of the mechanical
+theory is, that the only system of rings which
+can exist is one composed of an indefinite number
+of unconnected particles, revolving round the
+planet with different velocities according to their
+respective distances. These particles may be
+arranged in a series of narrow rings, or they may
+move through each other irregularly. In the
+\PageSep{367}
+\index{Keeler, spectroscopic examination of Saturn's ring|(}%
+first case the destruction of the system will be
+very slow, in the second case it will be more
+rapid, but there may be a tendency towards an
+arrangement in narrow rings, which may retard
+the process.
+
+``We are not able to ascertain by observation
+the constitution of the two outer divisions of the
+system of rings, but the inner ring is certainly
+transparent, for the limb (\ie~edge) of Saturn
+has been observed through it. It is also certain,
+that though the space occupied by the ring is
+transparent, it is not through the material particles
+of it that Saturn was seen, for his limb was
+observed without distortion; which shows that
+there was no refraction, and therefore that the
+rays did not pass through a medium at all, but
+between the solid or liquid particles of which the
+ring is composed. Here then we have an optical
+argument in favor of the theory of independent
+particles as the material of the rings. The
+two outer rings may be of the same nature, but
+not so exceedingly rare that a ray of light can
+pass through their whole thickness without encountering
+one of the particles.''
+\index{Maxwell on Saturn's ring|)}%
+
+\TB
+
+The last link in the chain of evidence has been
+furnished by recent observations made in America.
+If it can be proved that every part of the
+apparently solid ring moves round the planet's
+centre at a different rate, and that the speed at
+\PageSep{368}
+\index{Meteoric constitution of Saturn's ring}%
+\index{Spectroscopic proof of rotation of Saturn's ring}%
+each part is appropriate at its distance from the
+centre, the conclusion is inevitable that the ring
+consists of scattered fragments.
+
+Every one must have noticed that when a
+train passes at full speed with the whistle blowing,
+there is an abrupt fall in the pitch of the
+note. This change of note is only apparent to
+the stationary listener, and is caused by the
+crowding together of the waves of sound as the
+train approaches, and by their spacing out as it
+recedes. The same thing is true of light-waves,
+and if we could imagine a colored light to pass
+us at an almost inconceivable velocity it would
+change in tint as it passed.\footnote
+ {This statement is strictly correct only of monochromatic
+ light. I might, in the subsequent argument, have introduced
+ the limitation that the moving body shall emit only monochromatic
+ light. The qualification would, however, only complicate
+ the statement, and thus render the displacement of the lines of
+ the spectrum less easily intelligible.}
+Now there are certain
+lines in the spectrum of sunlight, and the
+shifting of their positions affords an excessively
+delicate measure of a change which, when magnified
+enormously, would produce a change of
+tint. For example, the sun is a rotating body,
+and when we look at its disk one edge is approaching
+us and the other is receding. The
+two edges are infinitesimally of different colors,
+and the change of tint is measurable by the displacement
+of the lines I have mentioned. In
+the same way Saturn's ring is illuminated by
+sunlight, and if different portions are moving at
+\PageSep{369}
+\index{Spectroscopic proof of rotation of Saturn's ring}%
+different velocities, those portions are infinitesimally
+of different colors. Now Professor Keeler,
+the present director of the Lick Observatory, has
+actually observed the reflected sunlight from the
+several parts of Saturn's ring, and he finds that
+the lines in the spectrum of the several parts
+are differently displaced. From measurement of
+these displacements he has concluded that every
+part of the ring moves at the same pace as if it
+were an independent satellite. The proof of the
+meteoric constitution of the ring is therefore
+\index{Meteoric constitution of Saturn's ring}%
+complete.
+
+It would be hard to find in science a more
+beautiful instance of arguments of the most
+diverse natures concentrating themselves on a
+definite and final conclusion.
+
+\begin{Authorities}
+Édouard Roche, \Title{La figure d'une masse fluide soumise à l'attraction
+\index{Roche, E.!stability of ellipsoid of}%
+d'un point éloigné}, ``Mém.\ Acad.\ de~Montpelier,'' vol.~i.\
+(Sciences), 1847--50.
+
+Maxwell, \Title{Stability of Saturn's Rings}, Macmillan, 1859.
+
+Keeler, \Title{Spectroscopic Proof of the Meteoric Constitution of
+Saturn's Rings}, ``Astrophysical Journal,'' May, 1895; see also
+\index{Keeler, spectroscopic examination of Saturn's ring|)}%
+\index{Saturn!theory of ring|)}%
+the same for June, 1895.
+
+Schwarzschild, \Title{Die Poincarésche Theorie des Gleichgewichts},
+\index{Schwarzschild!stability of Roche's ellipsoid}%
+``Annals of Munich Observatory,'' vol.~iii.\ 1896. He considers
+the stability of Roche's ellipsoid.
+\end{Authorities}
+\PageSep{370}
+%[Blank Page]
+\PageSep{371}
+\BackMatter
+\printindex
+\iffalse
+INDEX
+
+Abacus for reducing tidal observations#abacus, 217-220.
+
+Abbadie, tidal deflection of vertical#Abbadie, 143, 144.
+
+Aden, errors of tidal prediction at#Aden, 246.
+
+Adriatic, tide in#Adriatic, 186.
+
+Airy, Sir G. B.#Airy,
+ tides in rivers, 75;
+ attack on Laplace, 181;
+ cotidal chart, 188;
+ \Title{Tides and Waves}, 192.
+
+America, North, tide tables for#America, 222.
+
+Analysis, harmonic, of tide#analysis, 193-210.
+
+Andromeda, nebula in#Andromeda, 339.
+
+Annual and semi-annual tides#annual, 206.
+
+Arabian theories of tide#Arab, 77-79.
+
+Aristotle on tides#Aristotle, 81.
+
+Assyrian records of eclipses, 272.
+
+Atlantic, tide in#Atlantic, 186-188.
+
+Atmospheric pressure,
+ cause of seiches, 40;
+ distortion of soil by, 145, 146;
+ influence on tidal prediction, 242, 243.
+
+Atmospheric waves, Helmholtz on#Helmholtz, 48-51.
+
+Attraction,
+ of weight resting on elastic slab proportional to slope, 136, 137;
+ of tide calculated, 143.
+
+Baird, \Title{Manual for Tidal Observation}#Baird, 16.
+
+Bakhuyzen on tide due to variation of latitude#Bakhuyzen, 255, 256.
+
+Barnard, rotation of Jupiter's satellites#Barnard, 315.
+
+Barometric pressure. |see{Atmospheric pressure}. 0
+
+Becker, G. F., on Nebular Hypothesis#Becker, 334, 336-338.
+
+Bernoulli, Daniel, essay on tides#Bernoulli, 86, 88.
+
+Bertelli on Italian seismology#Bertelli, 126, 127.
+
+Bifilar. |see{Pendulum}. 0
+
+Borgen@Börgen, method of reducing tidal observations#Börgen, 217.
+
+Bond, discovery of inner ring of Saturn#Bond, 352.
+
+Bore,
+ definition, 59;
+ bore-shelter, 63;
+ diagram of rise in Tsien-Tang, 66;
+ pictures, 67;
+ rivers where found, 71;
+ causes, 72;
+ Chinese superstition, 68-70.
+
+Browne, E. G., Arabian theories of tide#Browne, 77-79.
+
+Cambridge, experiments with bifilar pendulum at#bifilar, 115-125.
+
+Canal,
+ theory of tide wave in, 165-167;
+ critical depth, 163-165;
+ tides in ocean partitioned into canals, 175;
+ canal in high latitude, 174-176.
+
+Capillarity of liquids, and Plateau's experiment#Plateau, 316-318.
+
+Cassini, discovery of division in Saturn's rings#Cassini, 352.
+
+Castel, Father, ridiculed by Voltaire#Castel, 295, 296.
+
+Cavalleri, essay on tides#Cavalleri, 86.
+
+Centripetal and centrifugal forces#Centripetal, 91-93.
+
+Chambers on possible existence of Martian satellites#Chambers, 296.
+\PageSep{372}
+
+Chandler, free nutation of earth, and variation of latitude#Chandler, 253-257.
+
+Chinese
+ superstition as to bore, 68-70;
+ theories of tide, 76, 77.
+
+Christie, A. S., tide due to variation of latitude#Christie, 255, 256.
+
+Constants, tidal, explained#constants, 195.
+
+Continents, trend of, possibly due to primeval tidal friction#continents, 308.
+
+Cotidal chart, 188;
+ for diurnal tide hitherto undetermined, 191, 192.
+
+Currents, tidal, in rivers#tidal current, 56.
+
+Curve, tide#tide curve,
+ irregularities in, 10-16;
+ at Bombay, 12;
+ partitioned into lunar time, 213.
+
+D'Abbadie. |see{Abbadie}. 0
+
+Darwin, G. H.#Darwin,
+ bifilar pendulum, 115-125;
+ harmonic analysis, 210;
+ tidal abacus, 217-220;
+ distortion of earth's surface by varying loads, 134-148;
+ rigidity of earth, 261, 262;
+ papers on tidal friction, 315;
+ hour-glass figure of rotating liquid, 328-332;
+ Jacobi's ellipsoid, 333;
+ evolution of satellites, 346.
+
+Darwin, Horace, bifilar pendulum#Horace, 115-125.
+
+Davis, method of presenting tide-generating force#Davis, 96, 97.
+
+Davison, history of bifilar and horizontal pendulums#Davison, 133.
+
+Dawes, discovery of inner ring of Saturn#Dawes, 352.
+
+Dawson coöperates in investigation of seiches#Dawson, 48.
+
+Day, change in length of, under tidal friction#day, 275, 276.
+
+Deflection of the vertical, 109-133;
+ experiments to measure, 115-125;
+ due to tide, 134-143.
+
+Deimos, a satellite of Mars#Deimos, 297.
+
+Denison, F. Napier, vibrations and seiches on lakes#Denison, 48-53.
+
+Density
+ of earth, law of internal, 302;
+ of planets determinable from their figures, 332, 333.
+
+Descartes, vortical theory of cosmogony#Descartes, 335.
+
+Dick, argument as to Martian satellites#Dick, 295.
+
+Dimple,
+ in soil, due to weight, 123;
+ form of, in elastic slab, 135.
+
+Distortion of soil
+ by weight, 123;
+ by varying loads, 134-148.
+
+Diurnal inequality
+ observed by Seleucus, 84, 85;
+ according to equilibrium theory, 156;
+ in Laplace's solution, 179;
+ in Atlantic, Pacific, and Indian Oceans, 180;
+ not shown in cotidal chart, 191;
+ in harmonic method, 205;
+ complicates prediction, 224, 225.
+
+Douglass, rotation of Jupiter's satellites#Douglass, 315.
+
+Dumb-bell nebula, description of photograph of#dumb-bell, 345.
+
+Dynamical theory of tide-wave, 163-181.
+
+Earth and moon#Earth,
+ diagram, 93;
+ rotation of, effects on tides, 177;
+ rigidity of, 256-260;
+ rotation retarded by tidal friction, 268;
+ figure of, 299;
+ adjustment of figure to suit change of rotation, 299-302;
+ internal density, 302;
+ probably once molten, 306;
+ distortion under primeval tidal friction, 307;
+ Roche's limit for, 358.
+
+Earthquakes,
+ a cause of seiches, 39;
+ microsisms and earth tremors, 125-127;
+ shock perceptible at great distance, 261.
+
+Ebb and flow defined#ebb, 56.
+
+Eccentricity of orbit
+ due to tidal friction, 313, 314;
+ theory of, in case of double stars, 342.
+\PageSep{373}
+
+Eclipses, ancient, and earth's rotation#eclipses, 272, 273.
+
+Ecliptic, obliquity of, due to tidal friction#ecliptic, 308-312.
+
+Eddies, tidal oscillation involves#eddies, 177.
+
+Ehlert, observation with horizontal pendulum#Ehlert, 132.
+
+Elastic distortion#elastic
+ of soil by weight, 123;
+ of earth by varying loads, 134-148;
+ calculation and illustration, 138-140;
+ by atmospheric pressure, 145-147.
+
+Elasticity of earth#elasticity, 254, 255.
+
+Elliptic tide, lunar#elliptic tide, 204.
+
+Ellipticity of earth's strata in excess for present rotation#ellipticity, 303, 304.
+
+Energy, tidal, utilization of#tidal energy, 73, 74.
+
+Equatorial canal, tide wave in#canal, 173.
+
+Equilibrium, figures of, of rotating liquid#equilibrium, 316-333.
+
+Equilibrium theory of tides#equilibrium, 149-162;
+ chart and law of tide, 151-153;
+ defects of, 160.
+
+Errors in tidal prediction#errors, 243-245.
+
+Establishment of port,
+ definition, 161, 162;
+ zero in equilibrium theory, 161;
+ shown in cotidal chart, 189.
+
+Estuary, annual meteorological tide in#estuary, 207, 208.
+
+Euler, essay on tides#Euler, 86.
+
+Europe, tides on coasts of#Europe, 188.
+
+Evolution of celestial systems, 334-346.
+
+Ferrel, tide-predicting instrument#Ferrel, 241.
+
+Figure of equilibrium
+ of ocean under tidal forces, 151-153;
+ of rotating liquid, 316-333.
+
+Figure of planets and their density, 332, 333.
+
+Fisher, Osmond, on molten interior of earth#Fisher, 262.
+
+Flow and ebb defined#ebb, 56.
+
+Forced oscillation,
+ principle of, 169, 170;
+ due to solar tide, possibly related to birth of moon, 282-284.
+
+Forced wave, explanation and contrast with free wave#forced wave, 164.
+
+Forces,
+ centripetal and centrifugal, 91-93;
+ tide-generating, 93-108;
+ numerical estimate, 109-111;
+ deflection of vertical by, 109-133;
+ figure of equilibrium under tidal, 151-153;
+ those of sun and moon compared, 156-158.
+
+Forel
+ on seiches, 17-38;
+ list of papers, 53, 54.
+
+Free oscillation contrasted with forced#free oscillation, 169, 170.
+
+Free wave, explanation and contrast with forced#free wave, 164.
+
+Friction of tides, 264-315.
+
+Galileo,
+ blames Kepler for his tidal theory, 85;
+ discovery of Jupiter's satellites, 291;
+ Saturn's ring, 350.
+
+Gauge, tide#tide gauge,
+ description of, 6-11;
+ site for, 14.
+
+Geneva,
+ seiches in lake, 17-28;
+ model of lake, 28.
+
+Geological evidence of earth's plasticity#plasticity, 300;
+ as to retardation of earth's rotation, 304-306.
+
+German method of reducing tidal observations#German, 217.
+
+Giles on Chinese theories of the tide#Giles, 76, 77.
+
+Gravity, variation according to latitude#gravity, 302, 303, 332.
+
+Greek
+ theory and description of tides, 81-85;
+ records of ancient eclipses, 272.
+
+Gulliver@\Title{Gulliver's Travels}, satire on mathematics#Gulliver, 292-295.
+
+Hall, Asaph, discovery of Martian satellites#Hall, 290-298.
+
+Hangchow, the bore at#Hangchow, 60-70.
+\PageSep{374}
+
+Harmonic analysis
+ initiated by Lord Kelvin, 87;
+ account of, 193-210.
+
+Height of tide#height
+ due to ideal satellite, 198;
+ at Portsmouth and at Aden, 225;
+ reduced by elastic yielding of earth, 259.
+
+Helmholtz
+ on atmospheric waves, 48-51;
+ on rotation of the moon, 286.
+
+Herschel, observations of twin nebulæ#Herschel, 344.
+
+High water
+ under moon in equilibrium theory, 160;
+ position in shallow and deep canals in dynamical theory, 171, 172.
+
+History
+ of tidal theories, 76-88;
+ of earth and moon, 278-286, 308-313.
+
+Hopkins on rigidity of earth#Hopkins, 258, 259.
+
+Horizontal tide-generating force, 107.
+
+Horizontal tide-generating force |see{also Pendulum}. 0
+
+Hough, S. S.#Hough,
+ frictional extinction of waves, 47;
+ dynamical solution of tidal problem, 181;
+ rigidity of earth, 254;
+ Chandler's nutation, 262.
+
+Hugli, bore on the#Hugli, 71.
+
+Huyghens, discovery of Saturn's ring#Huyghens, 351.
+
+Icelandic theory of tides, 79, 80.
+
+Indian Survey,
+ method of reducing tidal observations, 216, 217;
+ tide tables, 222.
+
+Instability,
+ nature of dynamical, and initial of moon's motion, 280-282;
+ of Saturn's ring, 363, 364.
+
+Interval from moon's transit to high water
+ in case of ideal satellite, 198;
+ at Portsmouth and at Aden, 225.
+
+Italian investigations in seismology, 125-130.
+
+Jacobi, figure of equilibrium of rotating liquid#Jacobi, 322-324.
+
+Japan, frequency of earthquakes#Japan, 130, 131.
+
+Jupiter,
+ satellites constantly face planet, 298;
+ figure and law of internal density, 333;
+ Roche's limit for, 361.
+
+Kant,
+ rotation of moon, 286;
+ nebular hypothesis, 334-339.
+
+Keeler, spectroscopic examination of Saturn's ring#Keeler, 367-369.
+
+Kelvin, Lord#Kelvin,
+ initiates harmonic analysis, 87, 199;
+ calculation of tidal attraction, 143;
+ tide predicting machine, 233;
+ rigidity of earth, 257-260;
+ denies adjustment of earth's figure to changed rotation, 301;
+ on geological time, 315.
+
+Kepler,
+ ideas concerning tides, 85, 86;
+ argument respecting Martian satellites, 291, 292.
+
+Krüger, figures of equilibrium of liquid#Krüger, 333.
+
+Lakes,
+ seiches in, 17-54;
+ mode of rocking in seiches, 24, 25;
+ vibrations, 41-53;
+ tides in, 182-185.
+
+Lamb, H., presentation of Laplace's theory#Lamb, 181.
+
+Laplace,
+ theory of tides, 86-88, 177-180;
+ on rotation of moon, 286, 287;
+ nebular hypothesis, 335-337.
+
+Lardner, possibility of Martian satellites#Lardner, 295.
+
+Latitude,
+ tidal wave in canal in high, 174-176;
+ periodic variations of, 251-256.
+
+Lege@Légé, constructor of tide-predicting machine#Légé, 233.
+
+Level of sea affected by atmospheric pressure#atmospheric pressure, 146.
+
+Limnimeter, a form of tide gauge#limnimeter, 24.
+
+Lowell, P., on rotations of Venus and Mercury#Lowell, 298, 299, 315.
+
+Low water. |see{High water}.
+\PageSep{375}
+
+Lubbock, Sir J., senior, on tides#Lubbock, 87.
+
+Lunar
+ tide-generating force compared with solar, 156-158;
+ tide, principal, 201;
+ elliptic tide, 204;
+ time, 213.
+
+Machine, tide-predicting#tide-predicting, 233, 241.
+
+Mackerel sky, evidence of air-waves#mackerel, 49.
+
+Maclaurin,
+ essay on tides, 86;
+ figure of equilibrium of rotating liquid, 322-324.
+
+Magnússon on Icelandic theories of tides#Magnússon, 79, 80.
+
+Marco Polo, resident of Hangchow#Marco, 70.
+
+Mars,
+ discovery of satellites, 290-298;
+ Roche's limit, 360.
+
+Maxwell on Saturn's ring#Maxwell, 363-367.
+
+Mediterranean Sea, tides in#Mediterranean, 185, 186.
+
+Mercury, rotation of#Mercury, 298, 299.
+
+Meteoric constitution of Saturn's ring#meteoric, 368, 369.
+
+Meteorological
+ tides, 206, 207;
+ conditions dependent on earth's rotation, 303.
+
+Microphone as a seismological instrument#microphone, 128-130.
+
+Microsisms, minute earthquakes#microsisms, 125-127.
+
+Mills worked by the tide#mills, 74, 75.
+
+Milne on seismology#Milne, 125, 130.
+
+Month, change in, under tidal friction#month, 275-277.
+
+Moon and earth,
+ diagram, 93;
+ tide-generating force compared with sun's, 156-158;
+ tide due to ideal, moving in equator, 193, 194;
+ ideal satellites replacing actual, 199, 200;
+ tidal prediction by reference to transit, 224-230;
+ retardation of motion by tidal friction, 269, 270;
+ origin of, 282, 283;
+ rotation annulled by tidal friction and present libration, 286;
+ inequality in motion indicates internal density of earth, 302, 303;
+ eccentricity of orbit increased by tidal friction, 313, 314.
+
+Moore, Captain#Moore,
+ illustrations of bore, 67;
+ survey of Tsien-Tang-Kiang, 60-70.
+
+Neap and spring tides#neap
+ in equilibrium theory, 159;
+ represented by principal lunar and solar tides, 204.
+
+Nebula in Andromeda#Andromeda, 339.
+
+Nebulae@Nebulæ, description of various#Nebulæ, 345.
+
+Nebular hypothesis, 334-339.
+
+Newcomb, S., theoretical explanation of Chandler's nutation#Newcomb, 254.
+
+Newton,
+ founder of tidal theory, 86;
+ theory of tide in equatorial canal, 172.
+
+Nolan, criticism of tidal theory of moon's origin#Nolan, 360.
+
+Nutation,
+ value of, indicates internal density of earth, 303;
+ Chandler's, 251-256.
+
+Obliquity of ecliptic, effects of tidal friction on#obliquity, 310-312.
+
+Observation,
+ methods of tidal, 6-14;
+ reduction of tidal, 211-220.
+
+Orbit
+ of moon and earth, 93-95;
+ of double stars, very eccentric, 313.
+
+Pacific Ocean, tide in, affects Atlantic#Pacific, 186, 187.
+
+Partial tides in harmonic method#partial tides, 199.
+
+Paschwitz, von Rebeur#Paschwitz,
+ on horizontal pendulum, 130-132;
+ tidal deflection of vertical at Wilhelmshaven, 144.
+
+Pendulum,
+ curves traced by, under tidal force, 111, 112;
+ bifilar, 115-125;
+ as seismological instrument, 126, 127;
+ horizontal, 130-132.
+\PageSep{376}
+
+Petitcodiac, bore in the#Petitcodiac, 71.
+
+Phobos, a satellite of Mars#Phobos, 297.
+
+Planetary figure of equilibrium of rotating liquid#planetary figure, 322.
+
+Planets,
+ rotation of some, annulled by tidal friction, 298;
+ figures and internal densities, 332, 333.
+
+Plasticity of earth under change of rotation#plasticity, 300-302.
+
+Plateau, experiment on figure of rotating globule#Plateau, 316-319.
+
+Plemyrameter, observation of seiches with#plemyrameter, 19-22.
+
+Poincaré,
+ law of interchange of stability, 326, 327;
+ figure of rotating liquid, 325, 327.
+
+Polibius on tides at Cadiz#Polibus, 83.
+
+Portsmouth, table of errors in tidal predictions#Portsmouth, 244.
+
+Posidonius on tides#Posidonius, 81-84.
+
+Precession, value of, indicates internal density of earth#internal density, 303.
+
+Predicting machine for tides, 233-241;
+ Ferrel's, 241.
+
+Prediction of tide,
+ due to ideal satellite, 200;
+ example at Aden, 226-230;
+ method of computing, 230-233;
+ errors in, 242-250.
+
+Pressure of atmosphere, elastic distortion of soil by#distortion, 145, 146.
+
+Principle of forced oscillations#forced oscillations, 169, 170.
+
+Rebeur. |see{Paschwitz}. 0
+
+Reduction of tidal observations, 211-220.
+
+Retardation of earth's rotation, 268.
+
+Rigidity of earth#rigidity, 256-260.
+
+Ripple mark in sand preserved in geological strata#ripple mark, 305.
+
+Rivers,
+ tide wave in, 55-59;
+ Airy on tide in, 75;
+ annual meteorological tide in, 206.
+
+Roberts, E., the tide-predicting machine#Roberts, 233.
+
+Roberts, I., photograph of nebula in Andromeda#Roberts, 339.
+
+Roche, E.#Roche,
+ ellipticity of internal strata of earth, 303;
+ theory of limit and Saturn's ring, 356-362;
+ stability of ellipsoid of, 369.
+
+Roman description of tides#Roman, 81-85.
+
+Rossi on Italian seismology#Rossi, 128-130.
+
+Rotating liquid, figures of equilibrium#rotating liquid, 316-333.
+
+Rotation
+ of earth involved in tidal problem, 177;
+ retarded by tidal friction, 268;
+ of moon annulled by tidal friction, 286;
+ of Mercury, Venus, and satellites of Jupiter and Saturn annulled by tidal friction, 298.
+
+Russell, observation of seiches in New South Wales#Russell, 47.
+
+Saint@St.\ Vénant on flow of solids#Vénant, 313.
+
+Satellites,
+ tide due to single equatorial, 195, 196;
+ ideal replacing sun and moon in harmonic analysis, 199, 200;
+ discovery of those of Mars, 290-298;
+ rotation of those of Jupiter and Saturn annulled, 298;
+ distribution of, in solar system, 339-341.
+
+Saturn,
+ satellites always face the planet, 298;
+ law of density and figure, 332;
+ description and picture, 347-354;
+ theory of ring, 356-369;
+ Roche's limit for, 360.
+
+Schedule for reducing tidal observations, 215, 216.
+
+Schiaparelli on rotation of Venus and Mercury#Schiaparelli, 298, 315.
+
+Schwarzschild,
+ exposition of Poincaré's theory, 333;
+ stability of Roche's ellipsoid, 369.
+
+Sea,
+ vibrations of, 44, 45;
+ level affected by atmospheric pressure, 146.
+\PageSep{377}
+
+See, T. J. J.,
+ eccentricity of orbits of double stars, 313;
+ theory of evolution of double stars, 342-346.
+
+Seiches,
+ definition, 17;
+ records of, 21;
+ longitudinal and transverse, 25-27;
+ periods of, 27;
+ causes of, 39, 40.
+
+Seine, bore in the#Seine, 71.
+
+Seismology, 133.
+
+Seleucus, observation of tides of Indian Ocean#Seleucus, 84, 85.
+
+Semidiurnal tide
+ in equilibrium theory, 153-156;
+ in harmonic method, 201-204.
+
+Severn, bore in the#Severn, 71.
+
+Slope of soil
+ due to elastic distortion, 136;
+ calculation and illustration of, 138-140.
+
+Solar
+ tide-generating force compared with lunar, 156-158;
+ principal tide, 202;
+ possible effect of tide in assisting birth of moon, 284, 285;
+ system, nebular hypothesis as to origin of, 334-339;
+ system, distribution of satellites in, 339-341.
+
+Spectroscopic proof of rotation of Saturn's ring#spectroscop, 368, 369.
+
+Spring and neap tides
+ in equilibrium theory, 159;
+ represented by principal lunar and solar tides, 203.
+
+Stability,
+ nature of dynamical, 280, 281;
+ of figures of equilibrium, 322, 323;
+ of Saturn's ring, 365, 366.
+
+Stars,
+ double, eccentricity of orbits, 313;
+ theory of evolution, 342-346.
+
+Storms a cause of seiches#storms, 39, 40.
+
+Strabo on tides#Strabo, 81-85.
+
+Stupart coöperates in investigation of seiches#Stupart, 48.
+
+Sun,
+ tide-generating force of, compared with that of moon, 156-158;
+ ideal, replacing real sun in harmonic analysis, 201;
+ possible influence of, in assisting birth of moon, 284, 285.
+
+Surface tension of liquids#surface tension, 317, 318.
+
+Swift, satire on mathematicians#Swift, 292-295.
+
+Synthesis of partial tides for prediction#synthesis, 230-233.
+
+Tables, tide#tide tables, 221-241;
+ method of calculating, 230-241;
+ amount of error in, 246, 247.
+
+Thomson, Sir W. |see{Kelvin}. 0
+
+Tidal problem. |see{Laplace, Harmonic Analysis, etc.} 0
+
+Tide,
+ definition, 1-3;
+ general description, 4-6.
+
+Tide, |see{also other headings; \eg\ for tide-generating force, |see{Force}}. 0
+
+Time,
+ lunar, 213;
+ requisite for evolution of moon, 285.
+
+Tisserand, Roche's investigations as to earth's figure#Tisserand, 315.
+
+Tremors, earth#tremors, 125.
+
+Tresca on flow of solids#Tresca, 300.
+
+Tromometer, a seismological instrument#tromometer, 126, 127.
+
+Tsien-Tang-Kiang, the bore in#Tsien, 60-70.
+
+United States Coast Survey,
+ method of reducing tidal observations, 217;
+ tide tables of, 222.
+
+Variation of latitude, 251-256.
+
+Vaucher, record of a great seiche at Geneva#Vaucher, 17.
+
+Venus, rotation of#Venus, 298, 299.
+
+Vertical. |see{Deflection}. 0
+
+Vibration of lakes, 41-53.
+
+Voltaire, satire on mathematicians, and Martian satellites#Voltaire, 295, 296.
+
+Vortical motion in oceanic tides#vorticial, 177, 178.
+\PageSep{378}
+
+Waves
+ in deep and shallow water, 29;
+ speed of, 31;
+ composition of, 33-37;
+ in atmosphere, 48-50;
+ forced and free, 164;
+ of tide in equatorial canal, 173;
+ in canal in high latitude, 174-176;
+ propagated northward in Atlantic, 186-188.
+
+Wharton, Sir W. J., illustration of bore#Wharton, 69.
+
+Whewell
+ on tides, 87;
+ empirical construction of tide tables, 87-90;
+ on cotidal charts, 188, 189.
+
+Wind,
+ a cause of seiches, 39;
+ vibrations of lakes due to, 41, 42;
+ a cause of meteorological tides, 206;
+ perturbation of, in tidal prediction, 242, 243.
+
+Woodward on variation of latitude#Woodward, 262.
+
+Wright, Thomas, on a theory of cosmogony#Wright, 335.
+
+Wye, bore in the#Wye, 71.
+\fi
+\PageSep{379}
+\clearpage
+\null\vfill
+\begin{center}
+\textgoth{The Riverside Press}
+
+\footnotesize
+CAMBRIDGE, MASSACHUSETTS, U. S. A. \\
+ELECTROTYPED AND PRINTED BY \\
+H. O. HOUGHTON AND CO.
+\end{center}
+\vfill
+%%%%%%%%%%%%%%%%%%%%%%%%% GUTENBERG LICENSE %%%%%%%%%%%%%%%%%%%%%%%%%%
+\FlushRunningHeads
+\PGLicense
+\begin{PGtext}
+End of the Project Gutenberg EBook of The Tides and Kindred Phenomena in the
+Solar System, by Sir George Howard Darwin
+
+*** END OF THIS PROJECT GUTENBERG EBOOK TIDES AND KINDRED PHENOMENA ***
+
+***** This file should be named 38722-t.tex or 38722-t.zip *****
+This and all associated files of various formats will be found in:
+ http://www.gutenberg.org/3/8/7/2/38722/
+
+Produced by Andrew D. Hwang, Bryan Ness and the Online
+Distributed Proofreading Team at http://www.pgdp.net (This
+file was produced from images generously made available
+by The Internet Archive/Canadian Libraries)
+
+
+Updated editions will replace the previous one--the old editions
+will be renamed.
+
+Creating the works from public domain print editions means that no
+one owns a United States copyright in these works, so the Foundation
+(and you!) can copy and distribute it in the United States without
+permission and without paying copyright royalties. Special rules,
+set forth in the General Terms of Use part of this license, apply to
+copying and distributing Project Gutenberg-tm electronic works to
+protect the PROJECT GUTENBERG-tm concept and trademark. Project
+Gutenberg is a registered trademark, and may not be used if you
+charge for the eBooks, unless you receive specific permission. If you
+do not charge anything for copies of this eBook, complying with the
+rules is very easy. You may use this eBook for nearly any purpose
+such as creation of derivative works, reports, performances and
+research. They may be modified and printed and given away--you may do
+practically ANYTHING with public domain eBooks. Redistribution is
+subject to the trademark license, especially commercial
+redistribution.
+
+
+
+*** START: FULL LICENSE ***
+
+THE FULL PROJECT GUTENBERG LICENSE
+PLEASE READ THIS BEFORE YOU DISTRIBUTE OR USE THIS WORK
+
+To protect the Project Gutenberg-tm mission of promoting the free
+distribution of electronic works, by using or distributing this work
+(or any other work associated in any way with the phrase "Project
+Gutenberg"), you agree to comply with all the terms of the Full Project
+Gutenberg-tm License (available with this file or online at
+http://gutenberg.net/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.net
+
+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.net),
+you must, at no additional cost, fee or expense to the user, provide a
+copy, a means of exporting a copy, or a means of obtaining a copy upon
+request, of the work in its original "Plain Vanilla ASCII" or other
+form. Any alternate format must include the full Project Gutenberg-tm
+License as specified in paragraph 1.E.1.
+
+1.E.7. Do not charge a fee for access to, viewing, displaying,
+performing, copying or distributing any Project Gutenberg-tm works
+unless you comply with paragraph 1.E.8 or 1.E.9.
+
+1.E.8. You may charge a reasonable fee for copies of or providing
+access to or distributing Project Gutenberg-tm electronic works provided
+that
+
+- You pay a royalty fee of 20% of the gross profits you derive from
+ the use of Project Gutenberg-tm works calculated using the method
+ you already use to calculate your applicable taxes. The fee is
+ owed to the owner of the Project Gutenberg-tm trademark, but he
+ has agreed to donate royalties under this paragraph to the
+ Project Gutenberg Literary Archive Foundation. Royalty payments
+ must be paid within 60 days following each date on which you
+ prepare (or are legally required to prepare) your periodic tax
+ returns. Royalty payments should be clearly marked as such and
+ sent to the Project Gutenberg Literary Archive Foundation at the
+ address specified in Section 4, "Information about donations to
+ the Project Gutenberg Literary Archive Foundation."
+
+- You provide a full refund of any money paid by a user who notifies
+ you in writing (or by e-mail) within 30 days of receipt that s/he
+ does not agree to the terms of the full Project Gutenberg-tm
+ License. You must require such a user to return or
+ destroy all copies of the works possessed in a physical medium
+ and discontinue all use of and all access to other copies of
+ Project Gutenberg-tm works.
+
+- You provide, in accordance with paragraph 1.F.3, a full refund of any
+ money paid for a work or a replacement copy, if a defect in the
+ electronic work is discovered and reported to you within 90 days
+ of receipt of the work.
+
+- You comply with all other terms of this agreement for free
+ distribution of Project Gutenberg-tm works.
+
+1.E.9. If you wish to charge a fee or distribute a Project Gutenberg-tm
+electronic work or group of works on different terms than are set
+forth in this agreement, you must obtain permission in writing from
+both the Project Gutenberg Literary Archive Foundation and Michael
+Hart, the owner of the Project Gutenberg-tm trademark. Contact the
+Foundation as set forth in Section 3 below.
+
+1.F.
+
+1.F.1. Project Gutenberg volunteers and employees expend considerable
+effort to identify, do copyright research on, transcribe and proofread
+public domain works in creating the Project Gutenberg-tm
+collection. Despite these efforts, Project Gutenberg-tm electronic
+works, and the medium on which they may be stored, may contain
+"Defects," such as, but not limited to, incomplete, inaccurate or
+corrupt data, transcription errors, a copyright or other intellectual
+property infringement, a defective or damaged disk or other medium, a
+computer virus, or computer codes that damage or cannot be read by
+your equipment.
+
+1.F.2. LIMITED WARRANTY, DISCLAIMER OF DAMAGES - Except for the "Right
+of Replacement or Refund" described in paragraph 1.F.3, the Project
+Gutenberg Literary Archive Foundation, the owner of the Project
+Gutenberg-tm trademark, and any other party distributing a Project
+Gutenberg-tm electronic work under this agreement, disclaim all
+liability to you for damages, costs and expenses, including legal
+fees. YOU AGREE THAT YOU HAVE NO REMEDIES FOR NEGLIGENCE, STRICT
+LIABILITY, BREACH OF WARRANTY OR BREACH OF CONTRACT EXCEPT THOSE
+PROVIDED IN PARAGRAPH 1.F.3. YOU AGREE THAT THE FOUNDATION, THE
+TRADEMARK OWNER, AND ANY DISTRIBUTOR UNDER THIS AGREEMENT WILL NOT BE
+LIABLE TO YOU FOR ACTUAL, DIRECT, INDIRECT, CONSEQUENTIAL, PUNITIVE OR
+INCIDENTAL DAMAGES EVEN IF YOU GIVE NOTICE OF THE POSSIBILITY OF SUCH
+DAMAGE.
+
+1.F.3. LIMITED RIGHT OF REPLACEMENT OR REFUND - If you discover a
+defect in this electronic work within 90 days of receiving it, you can
+receive a refund of the money (if any) you paid for it by sending a
+written explanation to the person you received the work from. If you
+received the work on a physical medium, you must return the medium with
+your written explanation. The person or entity that provided you with
+the defective work may elect to provide a replacement copy in lieu of a
+refund. If you received the work electronically, the person or entity
+providing it to you may choose to give you a second opportunity to
+receive the work electronically in lieu of a refund. If the second copy
+is also defective, you may demand a refund in writing without further
+opportunities to fix the problem.
+
+1.F.4. Except for the limited right of replacement or refund set forth
+in paragraph 1.F.3, this work is provided to you 'AS-IS' WITH NO OTHER
+WARRANTIES OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO
+WARRANTIES OF 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 http://www.pglaf.org.
+
+
+Section 3. Information about the Project Gutenberg Literary Archive
+Foundation
+
+The Project Gutenberg Literary Archive Foundation is a non profit
+501(c)(3) educational corporation organized under the laws of the
+state of Mississippi and granted tax exempt status by the Internal
+Revenue Service. The Foundation's EIN or federal tax identification
+number is 64-6221541. Its 501(c)(3) letter is posted at
+http://pglaf.org/fundraising. Contributions to the Project Gutenberg
+Literary Archive Foundation are tax deductible to the full extent
+permitted by U.S. federal laws and your state's laws.
+
+The Foundation's principal office is located at 4557 Melan Dr. S.
+Fairbanks, AK, 99712., but its volunteers and employees are scattered
+throughout numerous locations. Its business office is located at
+809 North 1500 West, Salt Lake City, UT 84116, (801) 596-1887, email
+business@pglaf.org. Email contact links and up to date contact
+information can be found at the Foundation's web site and official
+page at http://pglaf.org
+
+For additional contact information:
+ Dr. Gregory B. Newby
+ Chief Executive and Director
+ gbnewby@pglaf.org
+
+
+Section 4. Information about Donations to the Project Gutenberg
+Literary Archive Foundation
+
+Project Gutenberg-tm depends upon and cannot survive without wide
+spread public support and donations to carry out its mission of
+increasing the number of public domain and licensed works that can be
+freely distributed in machine readable form accessible by the widest
+array of equipment including outdated equipment. Many small donations
+($1 to $5,000) are particularly important to maintaining tax exempt
+status with the IRS.
+
+The Foundation is committed to complying with the laws regulating
+charities and charitable donations in all 50 states of the United
+States. Compliance requirements are not uniform and it takes a
+considerable effort, much paperwork and many fees to meet and keep up
+with these requirements. We do not solicit donations in locations
+where we have not received written confirmation of compliance. To
+SEND DONATIONS or determine the status of compliance for any
+particular state visit http://pglaf.org
+
+While we cannot and do not solicit contributions from states where we
+have not met the solicitation requirements, we know of no prohibition
+against accepting unsolicited donations from donors in such states who
+approach us with offers to donate.
+
+International donations are gratefully accepted, but we cannot make
+any statements concerning tax treatment of donations received from
+outside the United States. U.S. laws alone swamp our small staff.
+
+Please check the Project Gutenberg Web pages for current donation
+methods and addresses. Donations are accepted in a number of other
+ways including including checks, online payments and credit card
+donations. To donate, please visit: http://pglaf.org/donate
+
+
+Section 5. General Information About Project Gutenberg-tm electronic
+works.
+
+Professor Michael S. Hart is the originator of the Project Gutenberg-tm
+concept of a library of electronic works that could be freely shared
+with anyone. For thirty years, he produced and distributed Project
+Gutenberg-tm eBooks with only a loose network of volunteer support.
+
+
+Project Gutenberg-tm eBooks are often created from several printed
+editions, all of which are confirmed as Public Domain in the U.S.
+unless a copyright notice is included. Thus, we do not necessarily
+keep eBooks in compliance with any particular paper edition.
+
+
+Most people start at our Web site which has the main PG search facility:
+
+ http://www.gutenberg.net
+
+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.
+\end{PGtext}
+
+% %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% %
+% %
+% End of the Project Gutenberg EBook of The Tides and Kindred Phenomena in the
+% Solar System, by Sir George Howard Darwin %
+% %
+% *** END OF THIS PROJECT GUTENBERG EBOOK TIDES AND KINDRED PHENOMENA *** %
+% %
+% ***** This file should be named 38722-t.tex or 38722-t.zip ***** %
+% This and all associated files of various formats will be found in: %
+% http://www.gutenberg.org/3/8/7/2/38722/ %
+% %
+% %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% %
+
+\end{document}
+###
+@ControlwordReplace = (
+ ['\\ie', 'i.e.'],
+ ['\\eg', 'e.g.'],
+ ['\\NB', 'N.B.'],
+ ['\\ieme', '^{e}'],
+ ['\\iere', '^{re}'],
+ ['\\AD', 'A.D.'],
+ ['\\AM', 'A.M.'],
+ ['\\PM', 'P.M.'],
+ ['\\Contents', 'CONTENTS'],
+ ['\\Illustrations', 'LIST OF ILLUSTRATIONS'],
+ ['\\Preface', 'PREFACE'],
+ ['\\printindex', 'INDEX']
+ );
+
+@ControlwordArguments = (
+ ['\\Chapter', 0, 0, '', '', 1, 1, 'Chapter ', '. ', 1, 1, '', ''],
+ ['\\TranscribersNote', 1, 0, '', ''],
+ ['\\Signature', 1, 1, '', ' ', 1, 1, '', ''],
+ ['\\ToCChap', 1, 1, '', ' ', 1, 1, '', ''],
+ ['\\ToCSect', 1, 1, '', ' ', 1, 0, '', ''],
+ ['\\LoFFig', 1, 1, '', ' ', 1, 1, '', '', 1, 0, '', ''],
+ ['\\Graphic', 0, 0, '', '', 1, 0, '', '', 1, 0, '', ''],
+ ['\\Figure', 0, 0, '', '', 1, 1, '<Figure ', '>', 1, 0, '', '', 1, 0, '', ''],
+ ['\\TallFig', 0, 0, '', '', 1, 1, '<Figure ', '>', 1, 0, '', '', 1, 0, '', ''],
+ ['\\begin{Authorities}', 0, 0, 'Authorities. ', ''],
+ ['\\Paper', 1, 1, '', ''],
+ ['\\Ref', 1, 1, '', ' ', 1, 1, '', ''],
+ ['\\PageLabel', 0, 0, '', '', 1, 0, '', ''],
+ ['\\PageRef', 1, 1, '', ' ', 1, 1, '', ''],
+ ['\\First', 1, 1, '', ''],
+ ['\\bar', 1, 1, '', ''],
+ ['\\figref', 1, 1, '', ''],
+ ['\\fig', 1, 1, 'fig. ', ''],
+ ['\\Fig', 1, 1, 'Fig. ', ''],
+ ['\\Ref', 1, 1, '', ' ', 1, 1, '', ''],
+ ['\\DPtypo', 1, 0, '', '', 1, 1, '', ''],
+ ['\\DPchg', 1, 0, '', '', 1, 1, '', ''],
+ ['\\DPnote', 1, 0, '', '']
+ );
+
+$PageSeparator = qr/^\\PageSep/;
+$CustomClean = 'print "\\nCustom cleaning in progress...";
+my $cline = 0;
+ while ($cline <= $#file) {
+ $file[$cline] =~ s/--------[^\n]*\n//; # strip page separators
+ $cline++
+ }
+ print "done\\n";';
+###
+This is pdfTeX, Version 3.1415926-1.40.10 (TeX Live 2009/Debian) (format=pdflatex 2011.9.6) 30 JAN 2012 21:37
+entering extended mode
+ %&-line parsing enabled.
+**38722-t.tex
+(./38722-t.tex
+LaTeX2e <2009/09/24>
+Babel <v3.8l> and hyphenation patterns for english, usenglishmax, dumylang, noh
+yphenation, farsi, arabic, croatian, bulgarian, ukrainian, russian, czech, slov
+ak, danish, dutch, finnish, french, basque, ngerman, german, german-x-2009-06-1
+9, ngerman-x-2009-06-19, ibycus, monogreek, greek, ancientgreek, hungarian, san
+skrit, italian, latin, latvian, lithuanian, mongolian2a, mongolian, bokmal, nyn
+orsk, romanian, irish, coptic, serbian, turkish, welsh, esperanto, uppersorbian
+, estonian, indonesian, interlingua, icelandic, kurmanji, slovenian, polish, po
+rtuguese, spanish, galician, catalan, swedish, ukenglish, pinyin, loaded.
+(/usr/share/texmf-texlive/tex/latex/base/book.cls
+Document Class: book 2007/10/19 v1.4h Standard LaTeX document class
+(/usr/share/texmf-texlive/tex/latex/base/bk12.clo
+File: bk12.clo 2007/10/19 v1.4h Standard LaTeX file (size option)
+)
+\c@part=\count79
+\c@chapter=\count80
+\c@section=\count81
+\c@subsection=\count82
+\c@subsubsection=\count83
+\c@paragraph=\count84
+\c@subparagraph=\count85
+\c@figure=\count86
+\c@table=\count87
+\abovecaptionskip=\skip41
+\belowcaptionskip=\skip42
+\bibindent=\dimen102
+) (/usr/share/texmf-texlive/tex/latex/base/inputenc.sty
+Package: inputenc 2008/03/30 v1.1d Input encoding file
+\inpenc@prehook=\toks14
+\inpenc@posthook=\toks15
+(/usr/share/texmf-texlive/tex/latex/base/latin1.def
+File: latin1.def 2008/03/30 v1.1d Input encoding file
+)) (/usr/share/texmf-texlive/tex/latex/base/ifthen.sty
+Package: ifthen 2001/05/26 v1.1c Standard LaTeX ifthen package (DPC)
+) (/usr/share/texmf-texlive/tex/latex/amsmath/amsmath.sty
+Package: amsmath 2000/07/18 v2.13 AMS math features
+\@mathmargin=\skip43
+For additional information on amsmath, use the `?' option.
+(/usr/share/texmf-texlive/tex/latex/amsmath/amstext.sty
+Package: amstext 2000/06/29 v2.01
+(/usr/share/texmf-texlive/tex/latex/amsmath/amsgen.sty
+File: amsgen.sty 1999/11/30 v2.0
+\@emptytoks=\toks16
+\ex@=\dimen103
+)) (/usr/share/texmf-texlive/tex/latex/amsmath/amsbsy.sty
+Package: amsbsy 1999/11/29 v1.2d
+\pmbraise@=\dimen104
+) (/usr/share/texmf-texlive/tex/latex/amsmath/amsopn.sty
+Package: amsopn 1999/12/14 v2.01 operator names
+)
+\inf@bad=\count88
+LaTeX Info: Redefining \frac on input line 211.
+\uproot@=\count89
+\leftroot@=\count90
+LaTeX Info: Redefining \overline on input line 307.
+\classnum@=\count91
+\DOTSCASE@=\count92
+LaTeX Info: Redefining \ldots on input line 379.
+LaTeX Info: Redefining \dots on input line 382.
+LaTeX Info: Redefining \cdots on input line 467.
+\Mathstrutbox@=\box26
+\strutbox@=\box27
+\big@size=\dimen105
+LaTeX Font Info: Redeclaring font encoding OML on input line 567.
+LaTeX Font Info: Redeclaring font encoding OMS on input line 568.
+\macc@depth=\count93
+\c@MaxMatrixCols=\count94
+\dotsspace@=\muskip10
+\c@parentequation=\count95
+\dspbrk@lvl=\count96
+\tag@help=\toks17
+\row@=\count97
+\column@=\count98
+\maxfields@=\count99
+\andhelp@=\toks18
+\eqnshift@=\dimen106
+\alignsep@=\dimen107
+\tagshift@=\dimen108
+\tagwidth@=\dimen109
+\totwidth@=\dimen110
+\lineht@=\dimen111
+\@envbody=\toks19
+\multlinegap=\skip44
+\multlinetaggap=\skip45
+\mathdisplay@stack=\toks20
+LaTeX Info: Redefining \[ on input line 2666.
+LaTeX Info: Redefining \] on input line 2667.
+) (/usr/share/texmf-texlive/tex/latex/amsfonts/amssymb.sty
+Package: amssymb 2009/06/22 v3.00
+(/usr/share/texmf-texlive/tex/latex/amsfonts/amsfonts.sty
+Package: amsfonts 2009/06/22 v3.00 Basic AMSFonts support
+\symAMSa=\mathgroup4
+\symAMSb=\mathgroup5
+LaTeX Font Info: Overwriting math alphabet `\mathfrak' in version `bold'
+(Font) U/euf/m/n --> U/euf/b/n on input line 96.
+)) (/usr/share/texmf-texlive/tex/latex/base/alltt.sty
+Package: alltt 1997/06/16 v2.0g defines alltt environment
+) (/usr/share/texmf-texlive/tex/latex/footmisc/footmisc.sty
+Package: footmisc 2009/09/15 v5.5a a miscellany of footnote facilities
+\FN@temptoken=\toks21
+\footnotemargin=\dimen112
+\c@pp@next@reset=\count100
+\c@@fnserial=\count101
+Package footmisc Info: Declaring symbol style bringhurst on input line 855.
+Package footmisc Info: Declaring symbol style chicago on input line 863.
+Package footmisc Info: Declaring symbol style wiley on input line 872.
+Package footmisc Info: Declaring symbol style lamport-robust on input line 883.
+
+Package footmisc Info: Declaring symbol style lamport* on input line 903.
+Package footmisc Info: Declaring symbol style lamport*-robust on input line 924
+.
+) (/usr/share/texmf-texlive/tex/latex/tools/multicol.sty
+Package: multicol 2008/12/05 v1.6h multicolumn formatting (FMi)
+\c@tracingmulticols=\count102
+\mult@box=\box28
+\multicol@leftmargin=\dimen113
+\c@unbalance=\count103
+\c@collectmore=\count104
+\doublecol@number=\count105
+\multicoltolerance=\count106
+\multicolpretolerance=\count107
+\full@width=\dimen114
+\page@free=\dimen115
+\premulticols=\dimen116
+\postmulticols=\dimen117
+\multicolsep=\skip46
+\multicolbaselineskip=\skip47
+\partial@page=\box29
+\last@line=\box30
+\mult@rightbox=\box31
+\mult@grightbox=\box32
+\mult@gfirstbox=\box33
+\mult@firstbox=\box34
+\@tempa=\box35
+\@tempa=\box36
+\@tempa=\box37
+\@tempa=\box38
+\@tempa=\box39
+\@tempa=\box40
+\@tempa=\box41
+\@tempa=\box42
+\@tempa=\box43
+\@tempa=\box44
+\@tempa=\box45
+\@tempa=\box46
+\@tempa=\box47
+\@tempa=\box48
+\@tempa=\box49
+\@tempa=\box50
+\@tempa=\box51
+\c@columnbadness=\count108
+\c@finalcolumnbadness=\count109
+\last@try=\dimen118
+\multicolovershoot=\dimen119
+\multicolundershoot=\dimen120
+\mult@nat@firstbox=\box52
+\colbreak@box=\box53
+) (/usr/share/texmf-texlive/tex/latex/base/makeidx.sty
+Package: makeidx 2000/03/29 v1.0m Standard LaTeX package
+) (/usr/share/texmf-texlive/tex/latex/tools/indentfirst.sty
+Package: indentfirst 1995/11/23 v1.03 Indent first paragraph (DPC)
+) (/usr/share/texmf-texlive/tex/latex/was/icomma.sty
+Package: icomma 2002/03/10 v2.0 (WaS)
+) (/usr/share/texmf-texlive/tex/latex/tools/calc.sty
+Package: calc 2007/08/22 v4.3 Infix arithmetic (KKT,FJ)
+\calc@Acount=\count110
+\calc@Bcount=\count111
+\calc@Adimen=\dimen121
+\calc@Bdimen=\dimen122
+\calc@Askip=\skip48
+\calc@Bskip=\skip49
+LaTeX Info: Redefining \setlength on input line 76.
+LaTeX Info: Redefining \addtolength on input line 77.
+\calc@Ccount=\count112
+\calc@Cskip=\skip50
+) (/usr/share/texmf-texlive/tex/latex/tools/array.sty
+Package: array 2008/09/09 v2.4c Tabular extension package (FMi)
+\col@sep=\dimen123
+\extrarowheight=\dimen124
+\NC@list=\toks22
+\extratabsurround=\skip51
+\backup@length=\skip52
+) (/usr/share/texmf-texlive/tex/latex/yfonts/yfonts.sty
+Package: yfonts 2003/01/08 v1.3 (WaS)
+) (/usr/share/texmf-texlive/tex/latex/fancyhdr/fancyhdr.sty
+\fancy@headwidth=\skip53
+\f@ncyO@elh=\skip54
+\f@ncyO@erh=\skip55
+\f@ncyO@olh=\skip56
+\f@ncyO@orh=\skip57
+\f@ncyO@elf=\skip58
+\f@ncyO@erf=\skip59
+\f@ncyO@olf=\skip60
+\f@ncyO@orf=\skip61
+) (/usr/share/texmf-texlive/tex/latex/graphics/graphicx.sty
+Package: graphicx 1999/02/16 v1.0f Enhanced LaTeX Graphics (DPC,SPQR)
+(/usr/share/texmf-texlive/tex/latex/graphics/keyval.sty
+Package: keyval 1999/03/16 v1.13 key=value parser (DPC)
+\KV@toks@=\toks23
+) (/usr/share/texmf-texlive/tex/latex/graphics/graphics.sty
+Package: graphics 2009/02/05 v1.0o Standard LaTeX Graphics (DPC,SPQR)
+(/usr/share/texmf-texlive/tex/latex/graphics/trig.sty
+Package: trig 1999/03/16 v1.09 sin cos tan (DPC)
+) (/etc/texmf/tex/latex/config/graphics.cfg
+File: graphics.cfg 2009/08/28 v1.8 graphics configuration of TeX Live
+)
+Package graphics Info: Driver file: pdftex.def on input line 91.
+(/usr/share/texmf-texlive/tex/latex/pdftex-def/pdftex.def
+File: pdftex.def 2009/08/25 v0.04m Graphics/color for pdfTeX
+\Gread@gobject=\count113
+))
+\Gin@req@height=\dimen125
+\Gin@req@width=\dimen126
+) (/usr/share/texmf-texlive/tex/latex/caption/caption.sty
+Package: caption 2009/10/09 v3.1k Customizing captions (AR)
+(/usr/share/texmf-texlive/tex/latex/caption/caption3.sty
+Package: caption3 2009/10/09 v3.1k caption3 kernel (AR)
+\captionmargin=\dimen127
+\captionmargin@=\dimen128
+\captionwidth=\dimen129
+\caption@indent=\dimen130
+\caption@parindent=\dimen131
+\caption@hangindent=\dimen132
+)
+\c@ContinuedFloat=\count114
+) (/usr/share/texmf-texlive/tex/latex/geometry/geometry.sty
+Package: geometry 2008/12/21 v4.2 Page Geometry
+(/usr/share/texmf-texlive/tex/generic/oberdiek/ifpdf.sty
+Package: ifpdf 2009/04/10 v2.0 Provides the ifpdf switch (HO)
+Package ifpdf Info: pdfTeX in pdf mode detected.
+) (/usr/share/texmf-texlive/tex/generic/oberdiek/ifvtex.sty
+Package: ifvtex 2008/11/04 v1.4 Switches for detecting VTeX and its modes (HO)
+Package ifvtex Info: VTeX not detected.
+)
+\Gm@cnth=\count115
+\Gm@cntv=\count116
+\c@Gm@tempcnt=\count117
+\Gm@bindingoffset=\dimen133
+\Gm@wd@mp=\dimen134
+\Gm@odd@mp=\dimen135
+\Gm@even@mp=\dimen136
+\Gm@dimlist=\toks24
+(/usr/share/texmf-texlive/tex/xelatex/xetexconfig/geometry.cfg)) (/usr/share/te
+xmf-texlive/tex/latex/hyperref/hyperref.sty
+Package: hyperref 2009/10/09 v6.79a Hypertext links for LaTeX
+(/usr/share/texmf-texlive/tex/generic/ifxetex/ifxetex.sty
+Package: ifxetex 2009/01/23 v0.5 Provides ifxetex conditional
+) (/usr/share/texmf-texlive/tex/latex/oberdiek/hycolor.sty
+Package: hycolor 2009/10/02 v1.5 Code for color options of hyperref/bookmark (H
+O)
+(/usr/share/texmf-texlive/tex/latex/oberdiek/xcolor-patch.sty
+Package: xcolor-patch 2009/10/02 xcolor patch
+))
+\@linkdim=\dimen137
+\Hy@linkcounter=\count118
+\Hy@pagecounter=\count119
+(/usr/share/texmf-texlive/tex/latex/hyperref/pd1enc.def
+File: pd1enc.def 2009/10/09 v6.79a Hyperref: PDFDocEncoding definition (HO)
+) (/usr/share/texmf-texlive/tex/generic/oberdiek/etexcmds.sty
+Package: etexcmds 2007/12/12 v1.2 Prefix for e-TeX command names (HO)
+(/usr/share/texmf-texlive/tex/generic/oberdiek/infwarerr.sty
+Package: infwarerr 2007/09/09 v1.2 Providing info/warning/message (HO)
+)
+Package etexcmds Info: Could not find \expanded.
+(etexcmds) That can mean that you are not using pdfTeX 1.50 or
+(etexcmds) that some package has redefined \expanded.
+(etexcmds) In the latter case, load this package earlier.
+) (/etc/texmf/tex/latex/config/hyperref.cfg
+File: hyperref.cfg 2002/06/06 v1.2 hyperref configuration of TeXLive
+) (/usr/share/texmf-texlive/tex/latex/oberdiek/kvoptions.sty
+Package: kvoptions 2009/08/13 v3.4 Keyval support for LaTeX options (HO)
+(/usr/share/texmf-texlive/tex/generic/oberdiek/kvsetkeys.sty
+Package: kvsetkeys 2009/07/30 v1.5 Key value parser with default handler suppor
+t (HO)
+))
+Package hyperref Info: Option `hyperfootnotes' set `false' on input line 2864.
+Package hyperref Info: Option `bookmarks' set `true' on input line 2864.
+Package hyperref Info: Option `linktocpage' set `false' on input line 2864.
+Package hyperref Info: Option `pdfdisplaydoctitle' set `true' on input line 286
+4.
+Package hyperref Info: Option `pdfpagelabels' set `true' on input line 2864.
+Package hyperref Info: Option `bookmarksopen' set `true' on input line 2864.
+Package hyperref Info: Option `colorlinks' set `true' on input line 2864.
+Package hyperref Info: Hyper figures OFF on input line 2975.
+Package hyperref Info: Link nesting OFF on input line 2980.
+Package hyperref Info: Hyper index ON on input line 2983.
+Package hyperref Info: Plain pages OFF on input line 2990.
+Package hyperref Info: Backreferencing OFF on input line 2995.
+Implicit mode ON; LaTeX internals redefined
+Package hyperref Info: Bookmarks ON on input line 3191.
+(/usr/share/texmf-texlive/tex/latex/ltxmisc/url.sty
+\Urlmuskip=\muskip11
+Package: url 2006/04/12 ver 3.3 Verb mode for urls, etc.
+)
+LaTeX Info: Redefining \url on input line 3428.
+(/usr/share/texmf-texlive/tex/generic/oberdiek/bitset.sty
+Package: bitset 2007/09/28 v1.0 Data type bit set (HO)
+(/usr/share/texmf-texlive/tex/generic/oberdiek/intcalc.sty
+Package: intcalc 2007/09/27 v1.1 Expandable integer calculations (HO)
+) (/usr/share/texmf-texlive/tex/generic/oberdiek/bigintcalc.sty
+Package: bigintcalc 2007/11/11 v1.1 Expandable big integer calculations (HO)
+(/usr/share/texmf-texlive/tex/generic/oberdiek/pdftexcmds.sty
+Package: pdftexcmds 2009/09/23 v0.6 LuaTeX support for pdfTeX utility functions
+ (HO)
+(/usr/share/texmf-texlive/tex/generic/oberdiek/ifluatex.sty
+Package: ifluatex 2009/04/17 v1.2 Provides the ifluatex switch (HO)
+Package ifluatex Info: LuaTeX not detected.
+) (/usr/share/texmf-texlive/tex/generic/oberdiek/ltxcmds.sty
+Package: ltxcmds 2009/08/05 v1.0 Some LaTeX kernel commands for general use (HO
+)
+)
+Package pdftexcmds Info: LuaTeX not detected.
+Package pdftexcmds Info: \pdf@primitive is available.
+Package pdftexcmds Info: \pdf@ifprimitive is available.
+)))
+\Fld@menulength=\count120
+\Field@Width=\dimen138
+\Fld@charsize=\dimen139
+\Field@toks=\toks25
+Package hyperref Info: Hyper figures OFF on input line 4377.
+Package hyperref Info: Link nesting OFF on input line 4382.
+Package hyperref Info: Hyper index ON on input line 4385.
+Package hyperref Info: backreferencing OFF on input line 4392.
+Package hyperref Info: Link coloring ON on input line 4395.
+Package hyperref Info: Link coloring with OCG OFF on input line 4402.
+Package hyperref Info: PDF/A mode OFF on input line 4407.
+(/usr/share/texmf-texlive/tex/generic/oberdiek/atbegshi.sty
+Package: atbegshi 2008/07/31 v1.9 At begin shipout hook (HO)
+)
+\Hy@abspage=\count121
+\c@Item=\count122
+)
+*hyperref using driver hpdftex*
+(/usr/share/texmf-texlive/tex/latex/hyperref/hpdftex.def
+File: hpdftex.def 2009/10/09 v6.79a Hyperref driver for pdfTeX
+\Fld@listcount=\count123
+)
+\@indexfile=\write3
+\openout3 = `38722-t.idx'.
+
+Writing index file 38722-t.idx
+\TmpLen=\skip62
+LaTeX Font Info: Encoding `OT1' has changed to `T1' for symbol font
+(Font) `operators' in the math version `normal' on input line 594.
+
+LaTeX Font Info: Overwriting symbol font `operators' in version `normal'
+(Font) OT1/cmr/m/n --> T1/cmsc/m/n on input line 594.
+(./38722-t.aux)
+\openout1 = `38722-t.aux'.
+
+LaTeX Font Info: Checking defaults for OML/cmm/m/it on input line 613.
+LaTeX Font Info: ... okay on input line 613.
+LaTeX Font Info: Checking defaults for T1/cmr/m/n on input line 613.
+LaTeX Font Info: ... okay on input line 613.
+LaTeX Font Info: Checking defaults for OT1/cmr/m/n on input line 613.
+LaTeX Font Info: ... okay on input line 613.
+LaTeX Font Info: Checking defaults for OMS/cmsy/m/n on input line 613.
+LaTeX Font Info: ... okay on input line 613.
+LaTeX Font Info: Checking defaults for OMX/cmex/m/n on input line 613.
+LaTeX Font Info: ... okay on input line 613.
+LaTeX Font Info: Checking defaults for U/cmr/m/n on input line 613.
+LaTeX Font Info: ... okay on input line 613.
+LaTeX Font Info: Checking defaults for LY/yfrak/m/n on input line 613.
+LaTeX Font Info: ... okay on input line 613.
+LaTeX Font Info: Checking defaults for LYG/ygoth/m/n on input line 613.
+LaTeX Font Info: ... okay on input line 613.
+LaTeX Font Info: Checking defaults for PD1/pdf/m/n on input line 613.
+LaTeX Font Info: ... okay on input line 613.
+(/usr/share/texmf/tex/context/base/supp-pdf.mkii
+[Loading MPS to PDF converter (version 2006.09.02).]
+\scratchcounter=\count124
+\scratchdimen=\dimen140
+\scratchbox=\box54
+\nofMPsegments=\count125
+\nofMParguments=\count126
+\everyMPshowfont=\toks26
+\MPscratchCnt=\count127
+\MPscratchDim=\dimen141
+\MPnumerator=\count128
+\everyMPtoPDFconversion=\toks27
+)
+Package caption Info: Begin \AtBeginDocument code.
+Package caption Info: hyperref package is loaded.
+Package caption Info: End \AtBeginDocument code.
+*geometry auto-detecting driver*
+*geometry detected driver: pdftex*
+-------------------- Geometry parameters
+paper: class default
+landscape: --
+twocolumn: --
+twoside: true
+asymmetric: --
+h-parts: 9.03375pt, 307.14749pt, 9.03375pt
+v-parts: 4.15848pt, 423.22379pt, 6.23773pt
+hmarginratio: 1:1
+vmarginratio: 2:3
+lines: --
+heightrounded: --
+bindingoffset: 0.0pt
+truedimen: --
+includehead: true
+includefoot: true
+includemp: --
+driver: pdftex
+-------------------- Page layout dimensions and switches
+\paperwidth 325.215pt
+\paperheight 433.62pt
+\textwidth 307.14749pt
+\textheight 361.34999pt
+\oddsidemargin -63.23624pt
+\evensidemargin -63.23624pt
+\topmargin -68.11151pt
+\headheight 15.0pt
+\headsep 19.8738pt
+\footskip 30.0pt
+\marginparwidth 98.0pt
+\marginparsep 7.0pt
+\columnsep 10.0pt
+\skip\footins 10.8pt plus 4.0pt minus 2.0pt
+\hoffset 0.0pt
+\voffset 0.0pt
+\mag 1000
+\@twosidetrue \@mparswitchtrue
+(1in=72.27pt, 1cm=28.45pt)
+-----------------------
+(/usr/share/texmf-texlive/tex/latex/graphics/color.sty
+Package: color 2005/11/14 v1.0j Standard LaTeX Color (DPC)
+(/etc/texmf/tex/latex/config/color.cfg
+File: color.cfg 2007/01/18 v1.5 color configuration of teTeX/TeXLive
+)
+Package color Info: Driver file: pdftex.def on input line 130.
+)
+Package hyperref Info: Link coloring ON on input line 613.
+(/usr/share/texmf-texlive/tex/latex/hyperref/nameref.sty
+Package: nameref 2007/05/29 v2.31 Cross-referencing by name of section
+(/usr/share/texmf-texlive/tex/latex/oberdiek/refcount.sty
+Package: refcount 2008/08/11 v3.1 Data extraction from references (HO)
+)
+\c@section@level=\count129
+)
+LaTeX Info: Redefining \ref on input line 613.
+LaTeX Info: Redefining \pageref on input line 613.
+(./38722-t.out) (./38722-t.out)
+\@outlinefile=\write4
+\openout4 = `38722-t.out'.
+
+\AtBeginShipoutBox=\box55
+
+Overfull \hbox (3.10696pt too wide) in paragraph at lines 642--642
+[]\OT1/cmtt/m/n/8 *** START OF THIS PROJECT GUTENBERG EBOOK TIDES AND KINDRED P
+HENOMENA ***[]
+ []
+
+LaTeX Font Info: Try loading font information for T1+cmsc on input line 644.
+
+(/usr/share/texmf-texlive/tex/latex/taupin/t1cmsc.fd
+File: T1cmsc.fd 1995/12/20 v2.4f Standard LaTeX font definitions
+)
+LaTeX Font Info: Try loading font information for U+msa on input line 644.
+(/usr/share/texmf-texlive/tex/latex/amsfonts/umsa.fd
+File: umsa.fd 2009/06/22 v3.00 AMS symbols A
+)
+LaTeX Font Info: Try loading font information for U+msb on input line 644.
+(/usr/share/texmf-texlive/tex/latex/amsfonts/umsb.fd
+File: umsb.fd 2009/06/22 v3.00 AMS symbols B
+) [1
+
+
+
+{/var/lib/texmf/fonts/map/pdftex/updmap/pdftex.map}] [2
+
+] <./images/riverside.png, id=137, 78.0516pt x 104.0688pt>
+File: ./images/riverside.png Graphic file (type png)
+<use ./images/riverside.png> [1
+
+ <./images/riverside.png (PNG copy)>] [2
+
+] [3
+
+
+] [4] [5] [6
+
+
+] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17
+
+
+] [18] [19] [1
+
+
+
+
+
+] [2] [3] [4] [5] <./images/fig1.png, id=1253, 236.3229pt x 230.5413pt>
+File: ./images/fig1.png Graphic file (type png)
+<use ./images/fig1.png> [6] [7 <./images/fig1.png (PNG copy)>] <./images/fig2.p
+ng, id=1264, 235.6002pt x 164.7756pt>
+File: ./images/fig2.png Graphic file (type png)
+<use ./images/fig2.png> [8] [9 <./images/fig2.png (PNG copy)>] <./images/fig3.j
+pg, id=1275, 154.176pt x 74.679pt>
+File: ./images/fig3.jpg Graphic file (type jpg)
+<use ./images/fig3.jpg> <./images/fig4.jpg, id=1276, 154.176pt x 241.1409pt>
+File: ./images/fig4.jpg Graphic file (type jpg)
+<use ./images/fig4.jpg> [10] [11 <./images/fig3.jpg>] [12 <./images/fig4.jpg>]
+[13] <./images/fig5.jpg, id=1300, 501.875pt x 500.87125pt>
+File: ./images/fig5.jpg Graphic file (type jpg)
+<use ./images/fig5.jpg> <./images/fig6.png, id=1301, 238.491pt x 108.405pt>
+File: ./images/fig6.png Graphic file (type png)
+<use ./images/fig6.png> [14] [15 <./images/fig5.jpg>] [16 <./images/fig6.png (P
+NG copy)>] [17] [18
+
+
+] [19] <./images/fig7.png, id=1340, 248.6088pt x 106.2369pt>
+File: ./images/fig7.png Graphic file (type png)
+<use ./images/fig7.png> [20] [21 <./images/fig7.png (PNG copy)>] [22] <./images
+/fig8.png, id=1360, 244.9953pt x 104.7915pt>
+File: ./images/fig8.png Graphic file (type png)
+<use ./images/fig8.png> [23] [24 <./images/fig8.png (PNG copy)>] [25] <./images
+/fig10.png, id=1374, 233.4321pt x 132.9768pt>
+File: ./images/fig10.png Graphic file (type png)
+<use ./images/fig10.png> [26] [27 <./images/fig10.png (PNG copy)>] [28] [29] <.
+/images/fig11.png, id=1397, 239.2137pt x 135.1449pt>
+File: ./images/fig11.png Graphic file (type png)
+<use ./images/fig11.png> [30] [31 <./images/fig11.png (PNG copy)>] <./images/fi
+g12.png, id=1408, 242.1045pt x 66.4884pt>
+File: ./images/fig12.png Graphic file (type png)
+<use ./images/fig12.png> [32 <./images/fig12.png (PNG copy)>] [33] <./images/fi
+g13.png, id=1421, 238.491pt x 68.6565pt>
+File: ./images/fig13.png Graphic file (type png)
+<use ./images/fig13.png> <./images/fig13b.pdf, id=1422, 293.095pt x 431.6125pt>
+File: ./images/fig13b.pdf Graphic file (type pdf)
+<use ./images/fig13b.pdf> [34 <./images/fig13.png (PNG copy)>] [35] [36 <./imag
+es/fig13b.pdf>] [37] [38] [39] [40] [41] [42] [43] [44] <./images/fig14.png, id
+=1492, 71.5473pt x 389.5353pt>
+File: ./images/fig14.png Graphic file (type png)
+<use ./images/fig14.png>
+
+LaTeX Warning: Float too large for page by 3.9314pt on input line 2525.
+
+[45] [46] [47 <./images/fig14.png (PNG copy)>] [48] [49] [50] [51] [52] [53] [5
+4] [55] [56] [57
+
+
+] [58] [59] <./images/fig15.png, id=1572, 242.1045pt x 36.135pt>
+File: ./images/fig15.png Graphic file (type png)
+<use ./images/fig15.png> [60] [61 <./images/fig15.png (PNG copy)>] <./images/fi
+g16.png, id=1581, 255.1131pt x 114.9093pt>
+File: ./images/fig16.png Graphic file (type png)
+<use ./images/fig16.png> [62] [63 <./images/fig16.png (PNG copy)>] [64] <./imag
+es/fig17.jpg, id=1599, 524.96124pt x 310.15875pt>
+File: ./images/fig17.jpg Graphic file (type jpg)
+<use ./images/fig17.jpg> <./images/fig18.jpg, id=1601, 154.176pt x 91.542pt>
+File: ./images/fig18.jpg Graphic file (type jpg)
+<use ./images/fig18.jpg> [65] [66 <./images/fig17.jpg>] [67] [68 <./images/fig1
+8.jpg>] [69] <./images/fig19.jpg, id=1625, 422.57875pt x 642.4pt>
+File: ./images/fig19.jpg Graphic file (type jpg)
+<use ./images/fig19.jpg>
+
+LaTeX Warning: Float too large for page by 10.85399pt on input line 3278.
+
+[70] [71 <./images/fig19.jpg>] [72] [73] [74] [75] [76] [77] [78
+
+
+] [79] [80] [81] [82] [83] [84] [85] [86] [87] [88] [89] [90] [91] [92] [93
+
+
+] <./images/fig20.png, id=1745, 242.8272pt x 120.6909pt>
+File: ./images/fig20.png Graphic file (type png)
+<use ./images/fig20.png> [94] [95 <./images/fig20.png (PNG copy)>] [96] [97] [9
+8] <./images/fig21.png, id=1771, 233.4321pt x 128.6406pt>
+File: ./images/fig21.png Graphic file (type png)
+<use ./images/fig21.png> [99] [100 <./images/fig21.png (PNG copy)>] <./images/f
+ig22.png, id=1784, 250.7769pt x 146.7081pt>
+File: ./images/fig22.png Graphic file (type png)
+<use ./images/fig22.png> [101] [102 <./images/fig22.png (PNG copy)>] [103] [104
+] [105] [106] [107] <./images/fig23.jpg, id=1827, 154.176pt x 170.3163pt>
+File: ./images/fig23.jpg Graphic file (type jpg)
+<use ./images/fig23.jpg>
+Underfull \hbox (badness 2469) in paragraph at lines 4551--4563
+[]\OT1/cmr/m/n/12 Taking then the di-a-gram of tide-generating forces
+ []
+
+[108] [109 <./images/fig23.jpg>] [110] [111] <./images/fig24.png, id=1847, 238.
+491pt x 111.2958pt>
+File: ./images/fig24.png Graphic file (type png)
+<use ./images/fig24.png> [112
+
+
+] [113] [114 <./images/fig24.png (PNG copy)>] <./images/fig25.png, id=1866, 232
+.7094pt x 125.0271pt>
+File: ./images/fig25.png Graphic file (type png)
+<use ./images/fig25.png> [115] [116 <./images/fig25.png (PNG copy)>] [117] <./i
+mages/fig26.png, id=1882, 52.0344pt x 255.1131pt>
+File: ./images/fig26.png Graphic file (type png)
+<use ./images/fig26.png> [118] [119 <./images/fig26.png (PNG copy)>] [120] [121
+] [122] [123] [124] [125] [126] [127] [128] [129] [130] [131] [132] [133] [134]
+[135] [136] [137] <./images/fig27.png, id=1983, 247.8861pt x 117.8001pt>
+File: ./images/fig27.png Graphic file (type png)
+<use ./images/fig27.png> [138
+
+
+] [139 <./images/fig27.png (PNG copy)>] [140] [141] <./images/fig28.png, id=200
+8, 235.6002pt x 66.4884pt>
+File: ./images/fig28.png Graphic file (type png)
+<use ./images/fig28.png> [142] [143 <./images/fig28.png (PNG copy)>] [144] [145
+] [146] [147] [148] [149] [150] [151] [152] [153
+
+
+] [154] <./images/fig29.png, id=2081, 198.7425pt x 401.8212pt>
+File: ./images/fig29.png Graphic file (type png)
+<use ./images/fig29.png>
+
+LaTeX Warning: Float too large for page by 3.9314pt on input line 6094.
+
+[155] [156 <./images/fig29.png (PNG copy)>] [157] [158] [159] [160] [161] [162]
+[163] [164] [165] [166] [167
+
+
+] [168] [169] [170] [171] [172] [173] <./images/fig30.png, id=2187, 237.7683pt
+x 94.6737pt>
+File: ./images/fig30.png Graphic file (type png)
+<use ./images/fig30.png> [174] [175 <./images/fig30.png (PNG copy)>] [176] [177
+] [178] [179] [180] [181] [182] [183] [184] [185] [186
+
+
+] <./images/fig31.png, id=2253, 241.3818pt x 80.9424pt>
+File: ./images/fig31.png Graphic file (type png)
+<use ./images/fig31.png> [187] [188 <./images/fig31.png (PNG copy)>] [189] [190
+] [191] [192] <./images/fig32.jpg, id=2284, 410.53375pt x 642.4pt>
+File: ./images/fig32.jpg Graphic file (type jpg)
+<use ./images/fig32.jpg> [193] [194 <./images/fig32.jpg>] [195] [196
+
+
+] [197] [198] [199] [200] [201] [202] [203] [204] [205] [206] [207] [208] [209]
+[210] [211] [212] [213
+
+
+] [214] [215] [216] [217] [218] [219] <./images/fig33.jpg, id=2427, 386.44376pt
+ x 642.4pt>
+File: ./images/fig33.jpg Graphic file (type jpg)
+<use ./images/fig33.jpg>
+
+LaTeX Warning: Float too large for page by 3.9314pt on input line 8282.
+
+[220] [221 <./images/fig33.jpg>] [222] [223] [224
+
+
+] [225] [226] [227] [228] <./images/fig34.png, id=2477, 244.2726pt x 247.8861pt
+>
+File: ./images/fig34.png Graphic file (type png)
+<use ./images/fig34.png> [229] [230 <./images/fig34.png (PNG copy)>] [231] [232
+] [233] [234] [235] [236] <./images/fig35.png, id=2517, 236.3229pt x 190.7928pt
+>
+File: ./images/fig35.png Graphic file (type png)
+<use ./images/fig35.png> [237] [238 <./images/fig35.png (PNG copy)>] [239] [240
+] [241] [242] [243] [244] [245
+
+
+] [246] [247] [248] [249] [250] [251] [252] [253] [254] [255
+
+
+] [256] [257] [258] [259] [260] [261] [262] [263] [264] [265] [266] [267
+
+
+] <./images/fig36.png, id=2682, 235.6002pt x 104.7915pt>
+File: ./images/fig36.png Graphic file (type png)
+<use ./images/fig36.png> [268] [269 <./images/fig36.png (PNG copy)>] [270] [271
+] [272] [273] [274] [275] [276] [277] [278] [279] [280] [281] [282] [283] [284]
+[285] [286] [287] [288] [289] [290] [291] [292
+
+
+] [293] [294] [295] [296] [297] [298] [299] [300] [301] [302] [303] [304] [305]
+[306] [307] [308] [309] [310] [311] [312] [313] [314] [315] [316] [317] [318
+
+
+] [319] [320] [321] [322] [323] <./images/fig37.png, id=2971, 235.6002pt x 140.
+9265pt>
+File: ./images/fig37.png Graphic file (type png)
+<use ./images/fig37.png> [324] [325 <./images/fig37.png (PNG copy)>] <./images/
+fig38.png, id=2982, 239.9364pt x 130.8087pt>
+File: ./images/fig38.png Graphic file (type png)
+<use ./images/fig38.png> [326] [327 <./images/fig38.png (PNG copy)>] [328] [329
+] [330] <./images/fig39.png, id=3008, 247.1634pt x 203.8014pt>
+File: ./images/fig39.png Graphic file (type png)
+<use ./images/fig39.png> [331 <./images/fig39.png (PNG copy)>] [332] [333] [334
+] [335] [336] [337
+
+
+] [338] [339] [340] <./images/fig40.jpg, id=3063, 431.6125pt x 642.4pt>
+File: ./images/fig40.jpg Graphic file (type jpg)
+<use ./images/fig40.jpg> [341] [342 <./images/fig40.jpg>] [343] [344] [345] [34
+6] [347] [348] [349] [350] [351
+
+
+] <./images/fig41.jpg, id=3121, 154.176pt x 78.5334pt>
+File: ./images/fig41.jpg Graphic file (type jpg)
+<use ./images/fig41.jpg> [352] [353 <./images/fig41.jpg>] [354] [355] <./images
+/fig42.png, id=3143, 238.491pt x 123.5817pt>
+File: ./images/fig42.png Graphic file (type png)
+<use ./images/fig42.png> [356] [357 <./images/fig42.png (PNG copy)>] [358] [359
+] <./images/fig43.png, id=3169, 236.3229pt x 83.8332pt>
+File: ./images/fig43.png Graphic file (type png)
+<use ./images/fig43.png> [360] [361 <./images/fig43.png (PNG copy)>] [362] [363
+] [364] [365] [366] [367]
+Underfull \hbox (badness 1087) in paragraph at lines 13273--13276
+[]\OT1/cmr/m/n/12 I can-not sum up the gen-eral con-clu-sions at which
+ []
+
+[368] [369] [370] [371] [372] [373] (./38722-t.ind [374
+
+
+
+
+
+] [375] [376] [377] [378] [379] [380] [381] [382] [383] [384] [385] [386] [387]
+[388] [389] [390]) [391
+
+
+
+]
+Overfull \hbox (15.85715pt too wide) in paragraph at lines 14256--14256
+[]\OT1/cmtt/m/n/8 End of the Project Gutenberg EBook of The Tides and Kindred P
+henomena in the[]
+ []
+
+[1
+
+
+
+
+]
+Overfull \hbox (3.10696pt too wide) in paragraph at lines 14330--14330
+[]\OT1/cmtt/m/n/8 1.C. The Project Gutenberg Literary Archive Foundation ("the
+ Foundation"[]
+ []
+
+[2]
+Overfull \hbox (3.10696pt too wide) in paragraph at lines 14335--14335
+[]\OT1/cmtt/m/n/8 located in the United States, we do not claim a right to prev
+ent you from[]
+ []
+
+
+Overfull \hbox (3.10696pt too wide) in paragraph at lines 14340--14340
+[]\OT1/cmtt/m/n/8 freely sharing Project Gutenberg-tm works in compliance with
+the terms of[]
+ []
+
+[3]
+Overfull \hbox (3.10696pt too wide) in paragraph at lines 14403--14403
+[]\OT1/cmtt/m/n/8 posted on the official Project Gutenberg-tm web site (www.gut
+enberg.net),[]
+ []
+
+[4] [5] [6] [7] [8] [9] [10] (./38722-t.aux)
+
+ *File List*
+ book.cls 2007/10/19 v1.4h Standard LaTeX document class
+ bk12.clo 2007/10/19 v1.4h Standard LaTeX file (size option)
+inputenc.sty 2008/03/30 v1.1d Input encoding file
+ latin1.def 2008/03/30 v1.1d Input encoding file
+ ifthen.sty 2001/05/26 v1.1c Standard LaTeX ifthen package (DPC)
+ amsmath.sty 2000/07/18 v2.13 AMS math features
+ amstext.sty 2000/06/29 v2.01
+ amsgen.sty 1999/11/30 v2.0
+ amsbsy.sty 1999/11/29 v1.2d
+ amsopn.sty 1999/12/14 v2.01 operator names
+ amssymb.sty 2009/06/22 v3.00
+amsfonts.sty 2009/06/22 v3.00 Basic AMSFonts support
+ alltt.sty 1997/06/16 v2.0g defines alltt environment
+footmisc.sty 2009/09/15 v5.5a a miscellany of footnote facilities
+multicol.sty 2008/12/05 v1.6h multicolumn formatting (FMi)
+ makeidx.sty 2000/03/29 v1.0m Standard LaTeX package
+indentfirst.sty 1995/11/23 v1.03 Indent first paragraph (DPC)
+ icomma.sty 2002/03/10 v2.0 (WaS)
+ calc.sty 2007/08/22 v4.3 Infix arithmetic (KKT,FJ)
+ array.sty 2008/09/09 v2.4c Tabular extension package (FMi)
+ yfonts.sty 2003/01/08 v1.3 (WaS)
+fancyhdr.sty
+graphicx.sty 1999/02/16 v1.0f Enhanced LaTeX Graphics (DPC,SPQR)
+ keyval.sty 1999/03/16 v1.13 key=value parser (DPC)
+graphics.sty 2009/02/05 v1.0o Standard LaTeX Graphics (DPC,SPQR)
+ trig.sty 1999/03/16 v1.09 sin cos tan (DPC)
+graphics.cfg 2009/08/28 v1.8 graphics configuration of TeX Live
+ pdftex.def 2009/08/25 v0.04m Graphics/color for pdfTeX
+ caption.sty 2009/10/09 v3.1k Customizing captions (AR)
+caption3.sty 2009/10/09 v3.1k caption3 kernel (AR)
+geometry.sty 2008/12/21 v4.2 Page Geometry
+ ifpdf.sty 2009/04/10 v2.0 Provides the ifpdf switch (HO)
+ ifvtex.sty 2008/11/04 v1.4 Switches for detecting VTeX and its modes (HO)
+geometry.cfg
+hyperref.sty 2009/10/09 v6.79a Hypertext links for LaTeX
+ ifxetex.sty 2009/01/23 v0.5 Provides ifxetex conditional
+ hycolor.sty 2009/10/02 v1.5 Code for color options of hyperref/bookmark (HO
+)
+xcolor-patch.sty 2009/10/02 xcolor patch
+ pd1enc.def 2009/10/09 v6.79a Hyperref: PDFDocEncoding definition (HO)
+etexcmds.sty 2007/12/12 v1.2 Prefix for e-TeX command names (HO)
+infwarerr.sty 2007/09/09 v1.2 Providing info/warning/message (HO)
+hyperref.cfg 2002/06/06 v1.2 hyperref configuration of TeXLive
+kvoptions.sty 2009/08/13 v3.4 Keyval support for LaTeX options (HO)
+kvsetkeys.sty 2009/07/30 v1.5 Key value parser with default handler support
+(HO)
+ url.sty 2006/04/12 ver 3.3 Verb mode for urls, etc.
+ bitset.sty 2007/09/28 v1.0 Data type bit set (HO)
+ intcalc.sty 2007/09/27 v1.1 Expandable integer calculations (HO)
+bigintcalc.sty 2007/11/11 v1.1 Expandable big integer calculations (HO)
+pdftexcmds.sty 2009/09/23 v0.6 LuaTeX support for pdfTeX utility functions (
+HO)
+ifluatex.sty 2009/04/17 v1.2 Provides the ifluatex switch (HO)
+ ltxcmds.sty 2009/08/05 v1.0 Some LaTeX kernel commands for general use (HO)
+
+atbegshi.sty 2008/07/31 v1.9 At begin shipout hook (HO)
+ hpdftex.def 2009/10/09 v6.79a Hyperref driver for pdfTeX
+supp-pdf.mkii
+ color.sty 2005/11/14 v1.0j Standard LaTeX Color (DPC)
+ color.cfg 2007/01/18 v1.5 color configuration of teTeX/TeXLive
+ nameref.sty 2007/05/29 v2.31 Cross-referencing by name of section
+refcount.sty 2008/08/11 v3.1 Data extraction from references (HO)
+ 38722-t.out
+ 38722-t.out
+ t1cmsc.fd
+ umsa.fd 2009/06/22 v3.00 AMS symbols A
+ umsb.fd 2009/06/22 v3.00 AMS symbols B
+./images/riverside.png
+./images/fig1.png
+./images/fig2.png
+./images/fig3.jpg
+./images/fig4.jpg
+./images/fig5.jpg
+./images/fig6.png
+./images/fig7.png
+./images/fig8.png
+./images/fig10.png
+./images/fig11.png
+./images/fig12.png
+./images/fig13.png
+./images/fig13b.pdf
+./images/fig14.png
+./images/fig15.png
+./images/fig16.png
+./images/fig17.jpg
+./images/fig18.jpg
+./images/fig19.jpg
+./images/fig20.png
+./images/fig21.png
+./images/fig22.png
+./images/fig23.jpg
+./images/fig24.png
+./images/fig25.png
+./images/fig26.png
+./images/fig27.png
+./images/fig28.png
+./images/fig29.png
+./images/fig30.png
+./images/fig31.png
+./images/fig32.jpg
+./images/fig33.jpg
+./images/fig34.png
+./images/fig35.png
+./images/fig36.png
+./images/fig37.png
+./images/fig38.png
+./images/fig39.png
+./images/fig40.jpg
+./images/fig41.jpg
+./images/fig42.png
+./images/fig43.png
+ 38722-t.ind
+ ***********
+
+ )
+Here is how much of TeX's memory you used:
+ 8940 strings out of 493848
+ 125292 string characters out of 1152824
+ 218123 words of memory out of 3000000
+ 10981 multiletter control sequences out of 15000+50000
+ 23054 words of font info for 76 fonts, out of 3000000 for 9000
+ 714 hyphenation exceptions out of 8191
+ 37i,14n,44p,330b,1043s stack positions out of 5000i,500n,10000p,200000b,50000s
+{/usr/share/texmf/fonts/enc/dvips/cm-super/cm-super-t1.enc}</usr/share/texmf-
+texlive/fonts/type1/public/amsfonts/cm/cmbx12.pfb></usr/share/texmf-texlive/fon
+ts/type1/public/amsfonts/cm/cmcsc10.pfb></usr/share/texmf-texlive/fonts/type1/p
+ublic/amsfonts/cm/cmmi10.pfb></usr/share/texmf-texlive/fonts/type1/public/amsfo
+nts/cm/cmmi12.pfb></usr/share/texmf-texlive/fonts/type1/public/amsfonts/cm/cmmi
+5.pfb></usr/share/texmf-texlive/fonts/type1/public/amsfonts/cm/cmmi7.pfb></usr/
+share/texmf-texlive/fonts/type1/public/amsfonts/cm/cmmi8.pfb></usr/share/texmf-
+texlive/fonts/type1/public/amsfonts/cm/cmr10.pfb></usr/share/texmf-texlive/font
+s/type1/public/amsfonts/cm/cmr12.pfb></usr/share/texmf-texlive/fonts/type1/publ
+ic/amsfonts/cm/cmr17.pfb></usr/share/texmf-texlive/fonts/type1/public/amsfonts/
+cm/cmr7.pfb></usr/share/texmf-texlive/fonts/type1/public/amsfonts/cm/cmr8.pfb><
+/usr/share/texmf-texlive/fonts/type1/public/amsfonts/cm/cmsy10.pfb></usr/share/
+texmf-texlive/fonts/type1/public/amsfonts/cm/cmsy7.pfb></usr/share/texmf-texliv
+e/fonts/type1/public/amsfonts/cm/cmsy8.pfb></usr/share/texmf-texlive/fonts/type
+1/public/amsfonts/cm/cmti10.pfb></usr/share/texmf-texlive/fonts/type1/public/am
+sfonts/cm/cmti12.pfb></usr/share/texmf-texlive/fonts/type1/public/amsfonts/cm/c
+mtt10.pfb></usr/share/texmf-texlive/fonts/type1/public/amsfonts/cm/cmtt8.pfb></
+usr/share/texmf/fonts/type1/public/cm-super/sfcc0500.pfb></usr/share/texmf/font
+s/type1/public/cm-super/sfcc0600.pfb></usr/share/texmf/fonts/type1/public/cm-su
+per/sfcc0700.pfb></usr/share/texmf/fonts/type1/public/cm-super/sfcc0800.pfb></u
+sr/share/texmf/fonts/type1/public/cm-super/sfcc1000.pfb></usr/share/texmf/fonts
+/type1/public/cm-super/sfcc1095.pfb></usr/share/texmf/fonts/type1/public/cm-sup
+er/sfcc1200.pfb></usr/share/texmf-texlive/fonts/type1/public/gothic/ygoth.pfb>
+Output written on 38722-t.pdf (422 pages, 3217612 bytes).
+PDF statistics:
+ 4299 PDF objects out of 5155 (max. 8388607)
+ 1042 named destinations out of 1200 (max. 500000)
+ 453 words of extra memory for PDF output out of 10000 (max. 10000000)
+
diff --git a/38722-t/old/38722-t.zip b/38722-t/old/38722-t.zip
new file mode 100644
index 0000000..b45ec52
--- /dev/null
+++ b/38722-t/old/38722-t.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..c34386f
--- /dev/null
+++ b/README.md
@@ -0,0 +1,2 @@
+Project Gutenberg (https://www.gutenberg.org) public repository for
+eBook #38722 (https://www.gutenberg.org/ebooks/38722)