path: root/Master/texmf-dist/doc/latex/latex2e-help-texinfo/latex2e.dbk

<?xml version="1.0"?>
<!DOCTYPE book PUBLIC "-//OASIS//DTD DocBook XML V4.2//EN" "http://www.oasis-open.org/docbook/xml/4.2/docbookx.dtd" [
  <!ENTITY tex "TeX">
  <!ENTITY latex "LaTeX">
]>
<book id="latex2e.dbk" lang="en">
<!-- $Id$ -->
<!-- Public domain. -->
        

<title>&latex;2e unofficial reference manual (July 2018)</title>
<!-- %**end of header (This is for running Texinfo on a region.) -->

<!-- latex 2.09 commands should all be present now, -->
<!-- xx but latex2e stuff is missing. -->
<!-- xx random list of a few of the missing items is at the end of this file -->
<!-- -->
<!-- xx ctan, distributions, components of TeX -->
<!-- xx classes and packages - required, additional, useful; oberdiek; fonts -->
<!-- -->
<!-- xx merge permuted-index -->
<!-- xx merge latex-manual from savannah -->
<!-- xx merge display style math -->
<!-- xx JH explain nfss somewhere -->
<!-- xx JH expand BiBTeX -->
<!-- xx JH expand theorem, AMS math -->
<!-- xx JH something on code listings -->
<!-- xx JH ligatures -->
<!-- xx JH \xspace -->
<!-- xx JH \stretch -->
<!-- xx JH \mathstrut -->
<!-- xx JH \phantom https://tex.stackexchange.com/questions/4519/how-do-i-create-an-invisible-character -->
<!-- xx JH \baselineskip https://texfaq.org/FAQ-baselinepar -->
<!-- xx JH \contentsline, \@@dottedtocline? -->
<!-- xx JH more on \write18, beyond what's mentioned in Command line. -->
<!-- -->
<!-- xx The typeset source2e has an index with all kernel -->
<!-- xx commands, though some are internal and shouldn't be included. -->
<!-- xx classes.dtx et al. define additional commands. -->
<!-- xx See also http://ctan.org/pkg/macros2e. -->

<bookinfo><legalnotice><para>This document is an unofficial reference manual for &latex;, a
document preparation system, version of July 2018.
</para>
<para>This manual was originally translated from <filename>LATEX.HLP</filename> v1.0a in
the VMS Help Library.  The pre-translation version was written by
George&#160;D. Greenwade of Sam Houston State University.  The
&latex;&#160;2.09 version was written by Stephen Gilmore.  The
&latex;2e version was adapted from this by Torsten Martinsen.  Karl
Berry made further updates and additions, and gratefully acknowledges
using <citetitle>Hypertext Help with &latex;</citetitle>, by Sheldon Green, and
<citetitle>&latex; Command Summary</citetitle> (for &latex;&#160;2.09) by
L.&#160;Botway and C.&#160;Biemesderfer (published by the &tex; Users
Group as <citetitle>&tex;niques</citetitle> number 10), as reference material (no
text was directly copied).
</para>
<para>Copyright 2007, 2008, 2009, 2010, 2011, 2012, 2013,
2014, 2015, 2016, 2017, 2018 Karl Berry.

Copyright 1988, 1994, 2007 Stephen Gilmore.

Copyright 1994, 1995, 1996 Torsten Martinsen.
</para>
<!-- start of License -->
<para>Permission is granted to make and distribute verbatim copies of
this manual provided the copyright notice and this permission notice
are preserved on all copies.
</para>

<para>Permission is granted to copy and distribute modified versions of this
manual under the conditions for verbatim copying, provided that the entire
resulting derived work is distributed under the terms of a permission
notice identical to this one.
</para>
<para>Permission is granted to copy and distribute translations of this manual
into another language, under the above conditions for modified versions.
<!-- end of License -->
</para></legalnotice></bookinfo>
<!-- Merge into one index (arbitrarily chosen to be the concept index). -->



<para>This document is an unofficial reference manual for &latex;, a
document preparation system, version of July 2018.
</para>
<para>This manual was originally translated from <filename>LATEX.HLP</filename> v1.0a in
the VMS Help Library.  The pre-translation version was written by
George&#160;D. Greenwade of Sam Houston State University.  The
&latex;&#160;2.09 version was written by Stephen Gilmore.  The
&latex;2e version was adapted from this by Torsten Martinsen.  Karl
Berry made further updates and additions, and gratefully acknowledges
using <citetitle>Hypertext Help with &latex;</citetitle>, by Sheldon Green, and
<citetitle>&latex; Command Summary</citetitle> (for &latex;&#160;2.09) by
L.&#160;Botway and C.&#160;Biemesderfer (published by the &tex; Users
Group as <citetitle>&tex;niques</citetitle> number 10), as reference material (no
text was directly copied).
</para>
<para>Copyright 2007, 2008, 2009, 2010, 2011, 2012, 2013,
2014, 2015, 2016, 2017, 2018 Karl Berry.

Copyright 1988, 1994, 2007 Stephen Gilmore.

Copyright 1994, 1995, 1996 Torsten Martinsen.
</para>
<!-- start of License -->
<para>Permission is granted to make and distribute verbatim copies of
this manual provided the copyright notice and this permission notice
are preserved on all copies.
</para>

<para>Permission is granted to copy and distribute modified versions of this
manual under the conditions for verbatim copying, provided that the entire
resulting derived work is distributed under the terms of a permission
notice identical to this one.
</para>
<para>Permission is granted to copy and distribute translations of this manual
into another language, under the above conditions for modified versions.
<!-- end of License -->
</para>


<!-- Best Effort Symbol -->


<chapter label="" id="Top">
<title>&latex;2e: An unofficial reference manual</title>

<para>This document is an unofficial reference manual (version of
July 2018) for &latex;2e, a document preparation system.
</para>


</chapter>
<chapter label="1" id="About-this-document">
<title>About this document</title>

<indexterm role="cp"><primary>home page for manual</primary></indexterm>
<para>This is an unofficial reference manual for the &latex;2e document
preparation system, which is a macro package for the &tex;
typesetting program (see <link linkend="Overview">Overview</link>).  This document&#8217;s home page is
<ulink url="puszcza.gnu.org.ua/software/latexrefman">puszcza.gnu.org.ua/software/latexrefman</ulink>.  That page has links to the
current output in various formats, sources, mailing list archives and
subscriptions, and other infrastructure.
</para>
<indexterm role="cp"><primary>&latex; vs. &latex;2e</primary></indexterm>
<para>In this document, we will mostly just use &#8216;&latex;&#8217; rather than
&#8216;&latex;2e&#8217;, since the previous version of &latex;&#160;(2.09) was
frozen decades ago.
</para>
<indexterm role="cp"><primary>unofficial nature of this manual</primary></indexterm>
<indexterm role="cp"><primary>&latex; Project team</primary></indexterm>
<indexterm role="fn"><primary><email>latexrefman@tug.org</email> email address</primary></indexterm>
<para>&latex; is currently maintained by a group of volunteers
(<ulink url="http://latex-project.org">http://latex-project.org</ulink>).  The official documentation written by
the &latex; project is available from their web site.  This document is
completely unofficial and has not been reviewed by the &latex;
maintainers.  
<indexterm role="cp"><primary>bug reporting</primary></indexterm>
<indexterm role="cp"><primary>reporting bugs</primary></indexterm>
Do not send bug reports or anything else about this document to them.
Instead, please send all comments to <email>latexrefman@tug.org</email>.
</para>
<para>This document is a reference.  There is a vast array of other sources
of information about &latex;, at all levels.  Here are a few
introductions.
</para>
<variablelist><varlistentry><term><ulink url="http://ctan.org/pkg/latex-doc-ptr">http://ctan.org/pkg/latex-doc-ptr</ulink>
</term><listitem><indexterm role="fn"><primary>latex-doc-ptr document</primary></indexterm>
<para>Two pages of recommended references to &latex; documentation.
</para>
</listitem></varlistentry><varlistentry><term><ulink url="http://ctan.org/pkg/first-latex-doc">http://ctan.org/pkg/first-latex-doc</ulink>
</term><listitem><indexterm role="fn"><primary>first-latex-doc document</primary></indexterm>
<para>Writing your first document, with a bit of both text and math.
</para>
</listitem></varlistentry><varlistentry><term><ulink url="http://ctan.org/pkg/usrguide">http://ctan.org/pkg/usrguide</ulink>
</term><listitem><indexterm role="fn"><primary>usrguide official documentation</primary></indexterm>
<para>The guide for document authors that is maintained as part of &latex;.
Many other guides by many other people are also available, independent
of &latex; itself; one such is the next item:
</para>
</listitem></varlistentry><varlistentry><term><ulink url="http://ctan.org/pkg/lshort">http://ctan.org/pkg/lshort</ulink>
</term><listitem><indexterm role="fn"><primary>lshort document</primary></indexterm>
<para>A short introduction to &latex;, translated to many languages.
</para>
</listitem></varlistentry><varlistentry><term><ulink url="http://tug.org/begin.html">http://tug.org/begin.html</ulink>
</term><listitem><para>Introduction to the &tex; system, including &latex;, with further
references.
</para>
</listitem></varlistentry></variablelist>

</chapter>
<chapter label="2" id="Overview">
<title>Overview of &latex;</title>

<indexterm role="cp"><primary>overview of &latex;</primary></indexterm>
<indexterm role="cp"><primary>basics of &latex;</primary></indexterm>
<indexterm role="cp"><primary>Knuth, Donald E.</primary></indexterm>
<indexterm role="cp"><primary>Lamport, Leslie</primary></indexterm>
<indexterm role="cp"><primary>&latex; overview</primary></indexterm>

<para>&latex; is a system for typesetting documents.  It was originally
created by Leslie Lamport and is now maintained by a group of volunteers
(<ulink url="http://latex-project.org">http://latex-project.org</ulink>).  It is widely used, particularly for
complex and technical documents, such as those involving mathematics.
</para>
<indexterm role="cp"><primary>macro package, &latex; as</primary></indexterm>
<para>A &latex; user writes an input file containing text along with
interspersed commands, for instance commands describing how the text
should be formatted.  It is implemented as a set of related commands
that interface with Donald&#160;E. Knuth&#8217;s &tex; typesetting program
(the technical term is that &latex; is a <firstterm>macro package</firstterm> for the
&tex; engine).  The user produces the output document by giving that
input file to the &tex; engine.
</para>
<para>The term &latex; is also sometimes used to mean the language in which
the document is marked up, that is, to mean the set of commands
available to a &latex; user.
</para>
<indexterm role="cp"><primary>Lamport &tex;</primary></indexterm>
<indexterm role="cp"><primary>pronunciation</primary></indexterm>
<para>The name &latex; is short for &#8220;Lamport &tex;&#8221;.  It is pronounced
LAH-teck or LAY-teck, or sometimes LAY-tecks.  Inside a document,
produce the logo with <literal>\LaTeX</literal>.  Where use of the logo is not
sensible, such as in plain text, write it as &#8216;<literal>LaTeX</literal>&#8217;.
</para>


<sect1 label="2.1" id="Starting-and-ending">
<title>Starting and ending</title>

<anchor id="Starting-_0026-ending"/><!-- old name -->
<indexterm role="cp"><primary>starting and ending</primary></indexterm>
<indexterm role="cp"><primary>ending and starting</primary></indexterm>
<indexterm role="cp"><primary>hello, world</primary></indexterm>

<para>&latex; files have a simple global structure, with a standard beginning
and ending.  This is a small example.
</para>
<screen>\documentclass{article}
\begin{document}
Hello, \LaTeX\ world.
\end{document}
</screen>
<para>Every &latex; document has a <literal>\begin{document}</literal> line and an
<literal>\end{document}</literal> line.
</para>
<indexterm role="cp"><primary>document class, defined</primary></indexterm>
<para>Here, the &#8216;<literal>article</literal>&#8217; is the <firstterm>document class</firstterm>.  It is implemented
in a file <filename>article.cls</filename>.  You can use any document class on your
system.  A few document classes are defined by &latex; itself, and vast
array of others are widely available.  See <link linkend="Document-classes">Document classes</link>.
</para>
<indexterm role="cp"><primary>preamble, defined</primary></indexterm>
<para>You can include other &latex; commands between the
<literal>\documentclass</literal> and the <literal>\begin{document}</literal> commands.
This area is called the <firstterm>preamble</firstterm>.
</para>
<para>The <literal>\begin{document}</literal>, <literal>\end{document}</literal> pair defines an
<indexterm role="cp"><primary>environment</primary></indexterm>
<firstterm>environment</firstterm>; the &#8216;<literal>document</literal>&#8217; environment (and no others) is
required in all &latex; documents (see <link linkend="document">document</link>).  &latex; make
available to you many environments that are documented here
(see <link linkend="Environments">Environments</link>).  Many more are available to you from external
packages, most importantly those available at CTAN (see <link linkend="CTAN">CTAN</link>).
</para>
<para>The following sections discuss how to produce PDF or other output from
a &latex; input file.
</para>

</sect1>
<sect1 label="2.2" id="Output-files">
<title>Output files</title>

<para>&latex; produces a main output file and at least two auxiliary files.
The main output file&#8217;s name ends in either <filename>.dvi</filename> or <filename>.pdf</filename>.
</para>
<variablelist><varlistentry><term><literal>.dvi</literal>
</term><listitem><indexterm role="fn"><primary>.dvi file</primary></indexterm>
<indexterm role="fn"><primary>latex command</primary></indexterm>
<indexterm role="fn"><primary>xdvi command</primary></indexterm>
<indexterm role="fn"><primary>dvips command</primary></indexterm>
<indexterm role="fn"><primary>dvipdfmx command</primary></indexterm>
<indexterm role="fn"><primary>dvitype command</primary></indexterm>
<anchor id="output-files-dvi"/><para>If &latex; is invoked with the system command <command>latex</command> then it
produces a DeVice Independent file, with extension <filename>.dvi</filename>.  You
can view this file with a command such as <command>xdvi</command>, or convert
it to a PostScript <literal>.ps</literal> file with <command>dvips</command> or to a
Portable Document Format <literal>.pdf</literal> file with <command>dvipdfmx</command>.
The contents of the file can be dumped in human-readable form with
<command>dvitype</command>.  A vast array of other DVI utility programs are
available (<ulink url="http://mirror.ctan.org/dviware">http://mirror.ctan.org/dviware</ulink>).
</para>
</listitem></varlistentry><varlistentry><term><literal>.pdf</literal>
</term><listitem><indexterm role="fn"><primary>.pdf file</primary></indexterm>
<indexterm role="cp"><primary>pdf&tex;</primary></indexterm>
<indexterm role="fn"><primary>pdflatex command</primary></indexterm>
<anchor id="output-files-pdf"/><para>If &latex; is invoked via the system command <command>pdflatex</command>,
among other commands (see <link linkend="TeX-engines">&tex; engines</link>), then the main output is
a Portable Document Format (PDF) file.  Typically this is a
self-contained file, with all fonts and images included.
</para>
</listitem></varlistentry></variablelist>
<para>&latex; also produces at least two additional files.
</para>
<variablelist><varlistentry><term><literal>.log</literal>
</term><listitem><indexterm role="cp"><primary>transcript file</primary></indexterm>
<indexterm role="cp"><primary>log file</primary></indexterm>
<indexterm role="fn"><primary>.log file</primary></indexterm>
<anchor id="output-files-log"/><para>This transcript file contains summary information such as a list of
loaded packages.  It also includes diagnostic messages and perhaps
additional information for any errors.
</para>
</listitem></varlistentry><varlistentry><term><literal>.aux</literal>
</term><listitem><indexterm role="cp"><primary>auxiliary file</primary></indexterm>
<indexterm role="fn"><primary>.aux file</primary></indexterm>
<indexterm role="cp"><primary>cross references, resolving</primary></indexterm>
<indexterm role="cp"><primary>forward references, resolving</primary></indexterm>
<indexterm role="cp"><primary>references, resolving forward</primary></indexterm>
<anchor id="output-files-aux"/><para>Auxiliary information is used by &latex; for things such as
cross references.  For example, the first time that &latex; finds a
forward reference&#8212;a cross reference to something that has not yet
appeared in the source&#8212;it will appear in the output as a doubled
question mark <literal>??</literal>.  When the referred-to spot does eventually
appear in the source then &latex; writes its location information to
this <literal>.aux</literal> file.  On the next invocation, &latex; reads the
location information from this file and uses it to resolve the
reference, replacing the double question mark with the remembered
location.
</para>
</listitem></varlistentry></variablelist>
<indexterm role="fn"><primary>.lof file</primary></indexterm>
<indexterm role="cp"><primary>list of figures file</primary></indexterm>
<indexterm role="fn"><primary>.lot file</primary></indexterm>
<indexterm role="cp"><primary>list of tables file</primary></indexterm>
<indexterm role="fn"><primary>.toc file</primary></indexterm>
<indexterm role="cp"><primary>table of contents file</primary></indexterm>
<indexterm role="cp"><primary>contents file</primary></indexterm>
<para>&latex; may produce yet more files, characterized by the filename
ending.  These include a <literal>.lof</literal> file that is used to make a list of
figures, a <literal>.lot</literal> file used to make a list of tables, and a
<literal>.toc</literal> file used to make a table of contents (see <link linkend="Table-of-contents-etc_002e">Table of
contents etc.</link>).  A particular class may create others; the list is
open-ended.
</para>

</sect1>
<sect1 label="2.3" id="TeX-engines">
<title>&tex; engines</title>

<indexterm role="cp"><primary>engines, &tex;</primary></indexterm>
<indexterm role="cp"><primary>implementations of &tex;</primary></indexterm>
<indexterm role="cp"><primary>UTF-8</primary></indexterm>
<indexterm role="cp"><primary>Unicode input, native</primary></indexterm>
<indexterm role="cp"><primary>TrueType fonts</primary></indexterm>
<indexterm role="cp"><primary>OpenType fonts</primary></indexterm>

<para>&latex; is defined to be a set of commands that are run by a &tex;
implementation (see <link linkend="Overview">Overview</link>).  This section gives a terse
overview of the main programs (see also&#160;<link linkend="Command-line">Command line</link>).
</para>
<variablelist><varlistentry><term><literal>latex</literal>
</term><term><literal>pdflatex</literal>
</term><listitem><indexterm role="cp"><primary>pdf&tex; engine</primary></indexterm>
<indexterm role="fn"><primary>etex command</primary></indexterm>
<indexterm role="cp"><primary>e-&tex;</primary></indexterm>
<anchor id="tex-engines-latex"/><para>In &tex; Live (<ulink url="http://tug.org/texlive">http://tug.org/texlive</ulink>), if &latex; is invoked
via either the system command <command>latex</command> or <command>pdflatex</command>,
then the pdf&tex; engine is run (<ulink url="http://ctan.org/pkg/pdftex">http://ctan.org/pkg/pdftex</ulink>).
When invoked as <command>latex</command>, the main output is a <filename>.dvi</filename>
file; as <command>pdflatex</command>, the main output is a <filename>.pdf</filename> file.
</para>
<para>pdf&tex; incorporates the e-&tex; extensions to Knuth&#8217;s original
program (<ulink url="http://ctan.org/pkg/etex">http://ctan.org/pkg/etex</ulink>), including additional
programming features and bi-directional typesetting, and has plenty of
extensions of its own.  e-&tex; is available on its own as the system
command <command>etex</command>, but this is plain &tex; (and produces
<filename>.dvi</filename>).
</para>
<para>In other &tex; distributions, <command>latex</command> may invoke e-&tex;
rather than pdf&tex;.  In any case, the e-&tex; extensions can be
assumed to be available in &latex;.
</para>
</listitem></varlistentry><varlistentry><term><literal>lualatex</literal>
</term><listitem><indexterm role="fn"><primary>lualatex command</primary></indexterm>
<indexterm role="cp"><primary>Lua&tex;</primary></indexterm>
<anchor id="tex-engines-lualatex"/><para>If &latex; is invoked via the system command <command>lualatex</command>, the
Lua&tex; engine is run (<ulink url="http://ctan.org/pkg/luatex">http://ctan.org/pkg/luatex</ulink>).  This
program allows code written in the scripting language Lua
(<ulink url="http://luatex.org">http://luatex.org</ulink>) to interact with &tex;&#8217;s typesetting.
Lua&tex; handles UTF-8 Unicode input natively, can handle OpenType
and TrueType fonts, and produces a <filename>.pdf</filename> file by default.
There is also <command>dvilualatex</command> to produce a <filename>.dvi</filename> file,
but this is rarely used.
</para>
</listitem></varlistentry><varlistentry><term><literal>xelatex</literal>
</term><listitem><indexterm role="fn"><primary>xelatex command</primary></indexterm>
<indexterm role="cp"><primary>Xe&tex;</primary></indexterm>
<indexterm role="fn"><primary>.xdv file</primary></indexterm>
<indexterm role="fn"><primary>xdvipdfmx</primary></indexterm>
<anchor id="tex-engines-xelatex"/><para>If &latex; is invoked with the system command <command>xelatex</command>, the
Xe&tex; engine is run (<ulink url="http://tug.org/xetex">http://tug.org/xetex</ulink>).  Like Lua&tex;,
Xe&tex; natively supports UTF-8 Unicode and TrueType and OpenType
fonts, though the implementation is completely different, mainly using
external libraries instead of internal code.  Xe&tex; produces a
<filename>.pdf</filename> file as output; it does not support DVI output.
</para>
<para>Internally, Xe&tex; creates an <literal>.xdv</literal> file, a variant of DVI,
and translates that to PDF using the (<literal>x</literal>)<literal>dvipdfmx</literal>
program, but this process is automatic.  The <literal>.xdv</literal> file is only
useful for debugging.
</para>
</listitem></varlistentry></variablelist>
<para>Other variants of &latex; and &tex; exist, e.g., to provide
additional support for Japanese and other languages ([u]p&tex;,
<ulink url="http://ctan.org/pkg/ptex">http://ctan.org/pkg/ptex</ulink>, <ulink url="http://ctan.org/pkg/uptex">http://ctan.org/pkg/uptex</ulink>).
</para>

</sect1>
<sect1 label="2.4" id="LaTeX-command-syntax">
<title>&latex; command syntax</title>

<indexterm role="cp"><primary>command syntax</primary></indexterm>
<indexterm role="fn"><primary>\ character starting commands</primary></indexterm>
<indexterm role="fn"><primary>[...] for optional arguments</primary></indexterm>
<indexterm role="fn"><primary>{...} for required arguments</primary></indexterm>
<para>In the &latex; input file, a command name starts with a backslash
character, <literal>\</literal>.  The name itself then consists of either
(a)&#160;a string of letters or (b)&#160;a single non-letter.
</para>
<para>&latex; commands names are case sensitive so that <literal>\pagebreak</literal>
differs from <literal>\Pagebreak</literal> (the latter is not a standard command).
Most commands are lowercase, but in any event you must enter all
commands in the same case as they are defined.
</para>
<para>A command may be followed by zero, one, or more arguments. These
arguments may be either required or optional.  Required arguments are
contained in curly braces, <literal>{...}</literal>.  Optional arguments are
contained in square brackets, <literal>[...]</literal>.  Generally, but not
universally, if the command accepts an optional argument, it comes
first, before any required arguments.
</para>
<para>Inside of an optional argument, to use the character close square
bracket&#160;(<literal>]</literal>) hide it inside curly braces, as
in&#160;<literal>\item[closing bracket {]}]</literal>.  Similarly, if an optional
argument comes last, with no required argument after it, then to make
the first character of the following text be an open square bracket,
hide it inside curly braces.
</para>
<para>&latex; has the convention that some commands have a <literal>*</literal> form that
is related to the form without a <literal>*</literal>, such as <literal>\chapter</literal> and
<literal>\chapter*</literal>.  The exact difference in behavior varies from command
to command.
</para>
<para>This manual describes all accepted options and <literal>*</literal>-forms for the
commands it covers (barring unintentional omissions, a.k.a. bugs).
</para>


<sect2 label="2.4.1" id="Environment">
<title>Environments</title>

<para>Synopsis:
</para>
<screen>\begin{<replaceable>environment name</replaceable>}
  ...
\end{<replaceable>environment name</replaceable>}
</screen>
<para>An area of &latex; source, inside of which there is a distinct
behavior.  For instance, for poetry in &latex; put the lines between
<literal>\begin{verse}</literal> and <literal>\end{verse}</literal>.
</para>
<screen>\begin{verse}
  There once was a man from Nantucket \\
  ...
\end{verse}
</screen>
<para>See <link linkend="Environments">Environments</link> for a list of environments. 
</para>
<para>The <replaceable>environment name</replaceable> at the beginning must exactly match that at
the end.  This includes the case where <replaceable>environment name</replaceable> ends in a
star&#160;(<literal>*</literal>); both the <literal>\begin</literal> and <literal>\end</literal> texts must
include the star.
</para>
<para>Environments may have arguments, including optional arguments.  This
example produces a table.  The first argument is optional (and causes
the table to be aligned on its top row) while the second argument is
required (it specifies the formatting of columns).
</para>
<screen>\begin{tabular}[t]{r|l}
  ... rows of table ...
\end{tabular}
</screen>

</sect2>
<sect2 label="2.4.2" id="Declaration">
<title>Command declarations</title>

<para>A command that changes the value, or changes the meaning, of some other
command or parameter.  For instance, the <literal>\mainmatter</literal> command
changes the setting of page numbers from roman numerals to arabic.
</para>

</sect2>
<sect2 label="2.4.3" id="_005cmakeatletter-_0026-_005cmakeatother">
<title><literal>\makeatletter</literal> &amp; <literal>\makeatother</literal></title>

<para>Synopsis:
</para>
<screen>\makeatletter
  ... definition of commands with @ in their name ..
\makeatother
</screen>
<para>Used to redefine internal &latex; commands.  <literal>\makeatletter</literal> makes
the at-sign character&#160;<literal>@</literal> have the category code of a letter,
11.  <literal>\makeatother</literal> sets the category code of&#160;<literal>@</literal> to 12,
its original value.
</para>
<para>As each character is read by &tex; for &latex;, it is assigned a
character category code, or 
<indexterm role="cp"><primary>catcode</primary></indexterm>
<indexterm role="cp"><primary>character category code</primary></indexterm>
<indexterm role="cp"><primary>category code, character</primary></indexterm>
<firstterm>catcode</firstterm> for short. For instance, the backslash&#160;<literal>\</literal> is
assigned the catcode 0, for characters that start a command.  These two
commands alter the catcode assigned to <literal>@</literal>.
</para>
<para>The alteration is needed because many of &latex;&#8217;s commands use
<literal>@</literal> in their name, to prevent users from accidentally defining a
command that replaces one of &latex;&#8217;s own.  Command names consist of a
category&#160;0 character, ordinarily backslash, followed by letters,
category&#160;11 characters (except that a command name can also consist
of a category&#160;0 character followed by a single non-letter symbol).
So under the default category codes, user-defined commands cannot
contain an <literal>@</literal>.  But <literal>\makeatletter</literal> and <literal>\makeatother</literal>
allow users to define or redefine commands named with <literal>@</literal>.
</para>
<para>Use these two commands inside a <filename>.tex</filename> file, in the preamble, when
defining or redefining a command with <literal>@</literal> in its name.  Don&#8217;t use
them inside <filename>.sty</filename> or <filename>.cls</filename> files since the
<literal>\usepackage</literal> and <literal>\documentclass</literal> commands set the at sign to
have the character code of a letter.
</para>
<indexterm role="cp"><primary>package, <literal>macros2e</literal></primary></indexterm>
<indexterm role="cp"><primary><literal>macros2e</literal> package</primary></indexterm>

<para>For a comprehensive list of macros with an at-sign
in their names see <ulink url="http://ctan.org/pkg/macros2e">http://ctan.org/pkg/macros2e</ulink>. These macros are
mainly intended to package or class authors.
</para>
<para>In this example the class file has a command
<literal>\thesis@universityname</literal> that the user wants to change.  These
three lines should go in the preamble, before the
<literal>\begin{document}</literal>.
</para>
<screen>\makeatletter
\renewcommand{\thesis@universityname}{Saint Michael's College}
\makeatother
</screen>


<sect3 label="2.4.3.1" id="_005c_0040ifstar">
<title><literal>\@ifstar</literal></title>

<indexterm role="fn"><primary>\@ifstar</primary></indexterm>
<indexterm role="cp"><primary>commands, star-variants</primary></indexterm>
<indexterm role="cp"><primary>star-variants, commands</primary></indexterm>

<para>Synopsis:
</para>
<screen>\newcommand{\mycmd}{\@ifstar{\mycmd@star}{\mycmd@nostar}}
\newcommand{\mycmd@nostar}[<replaceable>nostar-num-args</replaceable>]{<replaceable>nostar-body</replaceable>} 
\newcommand{\mycmd@star}[<replaceable>star-num-args</replaceable>]{<replaceable>star-body</replaceable>}
</screen>
<para>Many standard &latex; environments or commands have a variant with the
same name but ending with a star character&#160;<literal>*</literal>, an asterisk.
Examples are the <literal>table</literal> and <literal>table*</literal> environments and the
<literal>\section</literal> and <literal>\section*</literal> commands.
</para>
<para>When defining environments, following this pattern is straightforward
because <literal>\newenvironment</literal> and <literal>\renewenvironment</literal> allow the
environment name to contain a star.  For commands the situation is more
complex.  As in the synopsis above, there will be a user-called command,
given above as <literal>\mycmd</literal>, which peeks ahead to see if it is followed
by a star.  For instance, &latex; does not really have a
<literal>\section*</literal> command; instead, the <literal>\section</literal> command peeks
ahead.  This command does not accept arguments but instead expands to
one of two commands that do accept arguments.  In the synopsis these two
are <literal>\mycmd@nostar</literal> and <literal>\mycmd@star</literal>.  They could take the
same number of arguments or a different number, or no arguments at all.
As always, in a &latex; document a command using at-sign&#160;<literal>@</literal>
must be enclosed inside a <literal>\makeatletter ... \makeatother</literal> block
(see <link linkend="_005cmakeatletter-_0026-_005cmakeatother">\makeatletter &amp; \makeatother</link>).
</para>
<para>This example of <literal>\@ifstar</literal> defines the command <literal>\ciel</literal> and a
variant <literal>\ciel*</literal>.  Both have one required argument.  A call to
<literal>\ciel{night}</literal> will return &quot;starry night sky&quot; while
<literal>\ciel*{blue}</literal> will return &quot;starry not blue sky&quot;.
</para>
<screen>\newcommand*{\ciel@unstarred}[1]{starry #1 sky}
\newcommand*{\ciel@starred}[1]{starry not #1 sky}
\newcommand*{\ciel}{\@ifstar{\ciel@starred}{\ciel@unstarred}}
</screen>
<para>In the next example, the starred variant takes a different number of
arguments than the unstarred one.  With this definition, Agent 007&#8217;s
<literal>``My name is \agentsecret*{Bond},
\agentsecret{James}{Bond}.''</literal> is equivalent to entering the commands
<literal>``My name is \textsc{Bond}, \textit{James} textsc{Bond}.''</literal>
</para>
<screen>\newcommand*{\agentsecret@unstarred}[2]{\textit{#1} \textsc{#2}}
\newcommand*{\agentsecret@starred}[1]{\textsc{#1}}
\newcommand*{\agentsecret}{%
  \@ifstar{\agentsecret@starred}{\agentsecret@unstarred}}
</screen>
<para>There are two sometimes more convenient ways to accomplish the work of
<literal>\@ifstar</literal>.  The <filename>suffix</filename> package allows the construct
<literal>\newcommand\mycommand{<replaceable>unstarred version</replaceable>}</literal> followed by
<literal>\WithSuffix\newcommand\mycommand*{<replaceable>starred version</replaceable>}</literal>.  And
&latex;3 has the <filename>xparse</filename> package that allows this code.
</para>
<screen>\NewDocumentCommand\foo{s}{\IfBooleanTF#1
  {<replaceable>starred version</replaceable>}%
  {<replaceable>unstarred version</replaceable>}% 
  }
</screen>

</sect3>
</sect2>
</sect1>
<sect1 label="2.5" id="CTAN">
<title>CTAN: Comprehensive &tex; Archive Network</title>

<indexterm role="cp"><primary>CTAN</primary></indexterm>

<para>The Comprehensive &tex; Archive Network, CTAN, is the &tex; and
&latex; community&#8217;s repository of free material.  It is a set of
Internet sites around the world that offer material related to &latex;
for download.  Visit CTAN on the web at <ulink url="https://ctan.org">https://ctan.org</ulink>.
</para>
<para>This material is organized into packages, discrete bundles that
typically offer some coherent functionality and are maintained by one
person or a small number of people.  For instance, many publishers have
a package that allows authors to format papers to that publisher&#8217;s
specifications.
</para>
<para>In addition to the massive holdings, the web site offers features such
as search by name or by functionality.
</para>
<indexterm role="cp"><primary>DANTE e.V.</primary></indexterm>
<indexterm role="cp"><primary>mirrors of CTAN</primary></indexterm>
<para>CTAN is not a single site, but instead is a set of sites. One of the
sites is the core. This site actively manages the material, for
instance, by accepting uploads of new or updated packages. It is
hosted by the German &tex; group DANTE e.V. Other sites around the
world help out by mirroring, that is, automatically syncing their
collections with the core site and then in turn making their copies
publicly available. This gives users close to their location better
access and relieves the load on the core site. The list of mirrors is
at <ulink url="https://ctan.org/mirrors">https://ctan.org/mirrors</ulink>.
</para>

</sect1>
</chapter>
<chapter label="3" id="Document-classes">
<title>Document classes</title>

<indexterm role="cp"><primary>document classes</primary></indexterm>
<indexterm role="cp"><primary>classes of documents</primary></indexterm>
<indexterm role="fn"><primary>\documentclass</primary></indexterm>

<para>The document&#8217;s overall class is defined with this command, which is
normally the first command in a &latex; source file.
</para>
<screen>\documentclass[<replaceable>options</replaceable>]{<replaceable>class</replaceable>}
</screen>
<indexterm role="fn"><primary>article class</primary></indexterm>
<indexterm role="fn"><primary>report class</primary></indexterm>
<indexterm role="fn"><primary>book class</primary></indexterm>
<indexterm role="fn"><primary>letter class</primary></indexterm>
<indexterm role="fn"><primary>slides class</primary></indexterm>
<para>The following document <replaceable>class</replaceable> names are built into &latex;.
(Many other document classes are available as separate packages;
see <link linkend="Overview">Overview</link>.)
</para>
<variablelist><varlistentry><term><literal>article</literal>
</term><listitem><anchor id="document-classes-article"/><para>For a journal article, a presentation, and miscellaneous general use.
</para>
</listitem></varlistentry><varlistentry><term><literal>book</literal>
</term><listitem><anchor id="document-classes-book"/><para>Full-length books, including chapters and possibly including front
matter, such as a preface, and back matter, such as an appendix
(see <link linkend="Front_002fback-matter">Front/back matter</link>).
</para>
</listitem></varlistentry><varlistentry><term><literal>letter</literal>
</term><listitem><anchor id="document-classes-letter"/><para>Mail, optionally including mailing labels 
(see <link linkend="Letters">Letters</link>).
</para>
</listitem></varlistentry><varlistentry><term><literal>report</literal>
</term><listitem><anchor id="document-classes-report"/><para>For documents of length between an <literal>article</literal> and a <literal>book</literal>,
such as technical reports or theses, which may contain several chapters.
</para>
</listitem></varlistentry><varlistentry><term><literal>slides</literal>
</term><listitem><anchor id="document-classes-slides"/><para>For slide presentations&#8212;rarely used today.  In its place the
<literal>beamer</literal> package is perhaps the most prevalent (see <link linkend="beamer-template">beamer
template</link>).
</para>
</listitem></varlistentry></variablelist>
<para>Standard <replaceable>options</replaceable> are described in the next section.
</para>


<sect1 label="3.1" id="Document-class-options">
<title>Document class options</title>

<indexterm role="cp"><primary>document class options</primary></indexterm>
<indexterm role="cp"><primary>options, document class</primary></indexterm>
<indexterm role="cp"><primary>class options</primary></indexterm>
<indexterm role="cp"><primary>global options</primary></indexterm>

<para>You can specify <firstterm>global options</firstterm> or <firstterm>class options</firstterm> to the
<literal>\documentclass</literal> command by enclosing them in square brackets.  To
specify more than one <replaceable>option</replaceable>, separate them with a comma.
</para>
<screen>\documentclass[<replaceable>option1</replaceable>,<replaceable>option2</replaceable>,...]{<replaceable>class</replaceable>}
</screen>
<para>Here is the list of the standard class options.
</para>
<indexterm role="fn"><primary>10pt option</primary></indexterm>
<indexterm role="fn"><primary>11pt option</primary></indexterm>
<indexterm role="fn"><primary>12pt option</primary></indexterm>
<para>All of the standard classes except <literal>slides</literal> accept the following
options for selecting the typeface size (default is <literal>10pt</literal>):
</para>
<screen>10pt  11pt  12pt
</screen>
<indexterm role="fn"><primary>a4paper option</primary></indexterm>
<indexterm role="fn"><primary>a5paper option</primary></indexterm>
<indexterm role="fn"><primary>b5paper option</primary></indexterm>
<indexterm role="fn"><primary>executivepaper option</primary></indexterm>
<indexterm role="fn"><primary>legalpaper option</primary></indexterm>
<indexterm role="fn"><primary>letterpaper option</primary></indexterm>
<para>All of the standard classes accept these options for selecting the paper
size (these show height by width):
</para>
<variablelist><varlistentry><term><literal>a4paper</literal>
</term><listitem><para>210 by 297mm (about 8.25 by 11.75&#160;inches)
</para>
</listitem></varlistentry><varlistentry><term><literal>a5paper</literal>
</term><listitem><para>148 by 210mm (about 5.8 by 8.3&#160;inches)
</para>
</listitem></varlistentry><varlistentry><term><literal>b5paper</literal>
</term><listitem><para>176 by 250mm (about 6.9 by 9.8&#160;inches)
</para> 
</listitem></varlistentry><varlistentry><term><literal>executivepaper</literal>
</term><listitem><para>7.25 by 10.5&#160;inches
</para> 
</listitem></varlistentry><varlistentry><term><literal>legalpaper</literal>
</term><listitem><para>8.5 by 14&#160;inches
</para> 
</listitem></varlistentry><varlistentry><term><literal>letterpaper</literal>
</term><listitem><para>8.5 by 11&#160;inches (the default)
</para></listitem></varlistentry></variablelist>
<indexterm role="fn"><primary>\pdfpagewidth</primary></indexterm>
<indexterm role="fn"><primary>\pdfpageheight</primary></indexterm>
<indexterm role="cp"><primary>package, <literal>geometry</literal></primary></indexterm>
<indexterm role="cp"><primary><literal>geometry</literal> package</primary></indexterm>

<para>When using one of the engines pdf&latex;, Lua&latex;, or Xe&latex;
(see <link linkend="TeX-engines">&tex; engines</link>), options other than <literal>letterpaper</literal> set
the print area but you must also set the physical paper size.  One way
to do that is to put <literal>\pdfpagewidth=\paperwidth</literal> and
<literal>\pdfpageheight=\paperheight</literal> in your document&#8217;s preamble.
<indexterm role="cp"><primary>package, <literal>geometry</literal></primary></indexterm>
<indexterm role="cp"><primary><literal>geometry</literal> package</primary></indexterm>
</para>
<para>The <literal>geometry</literal> package provides flexible ways of setting the print
area and physical page size.
</para>
<indexterm role="fn"><primary>draft option</primary></indexterm>
<indexterm role="fn"><primary>final option</primary></indexterm>
<indexterm role="fn"><primary>fleqn option</primary></indexterm>
<indexterm role="fn"><primary>landscape option</primary></indexterm>
<indexterm role="fn"><primary>leqno option</primary></indexterm>
<indexterm role="fn"><primary>openbib option</primary></indexterm>
<indexterm role="fn"><primary>titlepage option</primary></indexterm>
<indexterm role="fn"><primary>notitlepage option</primary></indexterm>
<para>Miscellaneous other options:
</para>
<variablelist><varlistentry><term><literal>draft</literal>
</term><term><literal>final</literal>
</term><listitem><indexterm role="cp"><primary>black boxes, omitting</primary></indexterm>
<para>Mark (<literal>draft</literal>) or do not mark (<literal>final</literal>) overfull boxes with a
black box in the margin; default is <literal>final</literal>.
</para>
</listitem></varlistentry><varlistentry><term><literal>fleqn</literal>
</term><listitem><indexterm role="cp"><primary>flush left equations</primary></indexterm>
<indexterm role="cp"><primary>centered equations</primary></indexterm>
<indexterm role="cp"><primary>equations, flush left vs. centered</primary></indexterm>
<para>Put displayed formulas flush left; default is centered.
</para>
</listitem></varlistentry><varlistentry><term><literal>landscape</literal>
</term><listitem><indexterm role="cp"><primary>landscape orientation</primary></indexterm>
<indexterm role="cp"><primary>portrait orientation</primary></indexterm>
<para>Selects landscape format; default is portrait.
</para>
</listitem></varlistentry><varlistentry><term><literal>leqno</literal>
</term><listitem><indexterm role="cp"><primary>left-hand equation numbers</primary></indexterm>
<indexterm role="cp"><primary>right-hand equation numbers</primary></indexterm>
<indexterm role="cp"><primary>equation numbers, left vs. right</primary></indexterm>
<para>Put equation numbers on the left side of equations; default is the right side.
</para>
</listitem></varlistentry><varlistentry><term><literal>openbib</literal>
</term><listitem><indexterm role="cp"><primary>bibliography format, open</primary></indexterm>
<para>Use &#8220;open&#8221; bibliography format.
</para>
</listitem></varlistentry><varlistentry><term><literal>titlepage</literal>
</term><term><literal>notitlepage</literal>
</term><listitem><indexterm role="cp"><primary>title page, separate or run-in</primary></indexterm>
<para>Specifies whether there is a separate page for the title information and
for the abstract also, if there is one.  The default for the
<literal>report</literal> class is <literal>titlepage</literal>, for the other classes it is
<literal>notitlepage</literal>.
</para></listitem></varlistentry></variablelist>
<para>The following options are not available with the <literal>slides</literal> class.
</para>
<indexterm role="fn"><primary>onecolumn option</primary></indexterm>
<indexterm role="fn"><primary>twocolumn option</primary></indexterm>
<indexterm role="fn"><primary>oneside option</primary></indexterm>
<indexterm role="fn"><primary>twoside option</primary></indexterm>
<indexterm role="fn"><primary>openright option</primary></indexterm>
<indexterm role="fn"><primary>openany option</primary></indexterm>
<variablelist><varlistentry><term><literal>onecolumn</literal>
</term><term><literal>twocolumn</literal>
</term><listitem><para>Typeset in one or two columns; default is <literal>onecolumn</literal>.
</para>
</listitem></varlistentry><varlistentry><term><literal>oneside</literal>
</term><term><literal>twoside</literal>
</term><listitem><indexterm role="fn"><primary>\evensidemargin</primary></indexterm>
<indexterm role="fn"><primary>\oddsidemargin</primary></indexterm>
<para>Selects one- or two-sided layout; default is <literal>oneside</literal>, except
that in the <literal>book</literal> class the default is <literal>twoside</literal>.
</para>
<para>For one-sided printing, the text is centered on the page.  For two-sided
printing, the <literal>\evensidemargin</literal> (<literal>\oddsidemargin</literal>) parameter
determines the distance on even (odd) numbered pages between the left
side of the page and the text&#8217;s left margin, with <literal>\oddsidemargin</literal>
being 40% of the difference between <literal>\paperwidth</literal> and
<literal>\textwidth</literal>, and <literal>\evensidemargin</literal> is the remainder.
</para>
</listitem></varlistentry><varlistentry><term><literal>openright</literal>
</term><term><literal>openany</literal>
</term><listitem><para>Determines if a chapter should start on a right-hand page; default is
<literal>openright</literal> for <literal>book</literal>, and <literal>openany</literal> for <literal>report</literal>.
</para></listitem></varlistentry></variablelist>
<indexterm role="fn"><primary>clock option to <literal>slides</literal> class</primary></indexterm>
<para>The <literal>slides</literal> class offers the option <literal>clock</literal> for printing
the time at the bottom of each note.
</para>

</sect1>
<sect1 label="3.2" id="Additional-packages">
<title>Additional packages</title>

<indexterm role="cp"><primary>loading additional packages</primary></indexterm>
<indexterm role="cp"><primary>packages, loading additional</primary></indexterm>
<indexterm role="cp"><primary>additional packages, loading</primary></indexterm>
<indexterm role="fn"><primary>\usepackage</primary></indexterm>
<para>Load a package <replaceable>pkg</replaceable>, with the package options given in the comma-separated
list <replaceable>options</replaceable>, as here.
</para>
<screen>\usepackage[<replaceable>options</replaceable>]{<replaceable>pkg</replaceable>}.
</screen>
<para>To specify more than one package you can separate them with a comma,
as in <literal>\usepackage{<replaceable>pkg1</replaceable>,<replaceable>pkg2</replaceable>,...}</literal>, or use multiple
<literal>\usepackage</literal> commands.
</para>
<indexterm role="cp"><primary>global options</primary></indexterm>
<indexterm role="cp"><primary>options, global</primary></indexterm>
<para>Any options given in the <literal>\documentclass</literal> command that are unknown
to the selected document class are passed on to the packages loaded with
<literal>\usepackage</literal>.
</para>

</sect1>
<sect1 label="3.3" id="Class-and-package-construction">
<title>Class and package construction</title>

<indexterm role="cp"><primary>document class commands</primary></indexterm>
<indexterm role="cp"><primary>commands, document class</primary></indexterm>
<indexterm role="cp"><primary>new class commands</primary></indexterm>

<para>You can create new document classes and new packages.  For instance, if
your memos must satisfy some local requirements, such as a
standard header for each page, then you could create a new class
<literal>smcmemo.cls</literal> and begin your documents with
<literal>\documentclass{smcmemo}</literal>.
</para>
<para>What separates a package from a document class is that the commands in a
package are useful across classes while those in a document class are
specific to that class.  Thus, a command to set page headers is for a
package while a command to make the page headers say <literal>Memo from the
SMC Math Department</literal> is for a class.
<indexterm role="cp"><primary>class and package difference</primary></indexterm>
<indexterm role="cp"><primary>difference between class and package</primary></indexterm>
</para>
<para>Inside of a class or package file you can use the at-sign <literal>@</literal> as a
character in command names without having to surround the code
containing that command with <literal>\makeatletter</literal> and
<literal>\makeatother</literal>.  See <link linkend="_005cmakeatletter-_0026-_005cmakeatother">\makeatletter &amp; \makeatother</link>. This allow
you to create commands that users will not accidentally redefine.
Another technique is to preface class- or package-specific commands with
some string to prevent your class or package from interfering with
others. For instance, the class <literal>smcmemo</literal> might have commands
<literal>\smc@tolist</literal>, <literal>\smc@fromlist</literal>, etc.
</para>



<sect2 label="3.3.1" id="Class-and-package-structure">
<title>Class and package structure</title>

<indexterm role="cp"><primary>class and package structure</primary></indexterm>
<indexterm role="cp"><primary>class file layout</primary></indexterm>
<indexterm role="cp"><primary>package file layout</primary></indexterm>
<indexterm role="cp"><primary>options, document class</primary></indexterm>
<indexterm role="cp"><primary>options, package</primary></indexterm>
<indexterm role="cp"><primary>class options</primary></indexterm>
<indexterm role="cp"><primary>package options</primary></indexterm>

<para>A class file or package file typically has four parts.  
</para><orderedlist numeration="arabic"><para>In the <firstterm>identification part</firstterm>, the file says that it is a &latex;
package or class and describes itself, using the <literal>\NeedsTeXFormat</literal>
and <literal>\ProvidesClass</literal> or <literal>\ProvidesPackage</literal> commands.
</para><listitem><para>The <firstterm>preliminary declarations part</firstterm> declares some commands and
can also load other files. Usually these commands will be those needed
for the code used in the next part.  For example, an <literal>smcmemo</literal>
class might be called with an option to read in a file with a list of
people for the to-head, as <literal>\documentclass[mathto]{smcmemo}</literal>, and
therefore needs to define a command
<literal>\newcommand{\setto}[1]{\def\@tolist{#1}}</literal> used in that
file.
</para></listitem><listitem><para>In the <firstterm>handle options part</firstterm> the class or package declares
and processes its options.  Class options allow a user to start their
document as <literal>\documentclass[<replaceable>option list</replaceable>]{<replaceable>class
name</replaceable>}</literal>, to modify the behavior of the class.  An example is when you
declare <literal>\documentclass[11pt]{article}</literal> to set the default
document font size.  
</para></listitem><listitem><para>Finally, in the <firstterm>more declarations part</firstterm> the class or package usually does
most of its work: declaring new variables, commands and fonts, and
loading other files.
</para></listitem></orderedlist>
<para>Here is a starting class file, which should be saved as <filename>stub.cls</filename>
where &latex; can find it, for example in the same directory as the
<filename>.tex</filename> file.
</para>
<screen>\NeedsTeXFormat{LaTeX2e}
\ProvidesClass{stub}[2017/07/06 stub to start building classes from]
\DeclareOption*{\PassOptionsToClass{\CurrentOption}{article}}
\ProcessOptions\relax
\LoadClass{article}
</screen><indexterm role="cp"><primary>class file example</primary></indexterm>

<para>It identifies itself, handles the class options via the default of
passing them all to the <literal>article</literal> class, and then loads the
<literal>article</literal> class to provide the basis for this class&#8217;s code.
</para>
<para>For more, see the official guide for class and package writers, the
Class Guide, at
<ulink url="http://www.latex-project.org/help/documentation/clsguide.pdf">http://www.latex-project.org/help/documentation/clsguide.pdf</ulink> (much
of the descriptions here derive from this document), or the tutorial
<ulink url="https://www.tug.org/TUGboat/tb26-3/tb84heff.pdf">https://www.tug.org/TUGboat/tb26-3/tb84heff.pdf</ulink>.
</para>

</sect2>
<sect2 label="3.3.2" id="Class-and-package-commands">
<title>Class and package commands</title>
<indexterm role="cp"><primary>class and package commands</primary></indexterm>
<indexterm role="cp"><primary>commands, class and package</primary></indexterm>

<para>These are the commands designed to help writers of classes or packages.
</para>
<variablelist><varlistentry><term><literal>\AtBeginDvi{specials}</literal>
</term><listitem><indexterm role="fn"><primary>\AtBeginDvi</primary></indexterm>
<para>Save in a box register things that are written to the <filename>.dvi</filename> file
at the beginning of the shipout of the first page of the document.
</para>
</listitem></varlistentry><varlistentry><term><literal>\AtEndOfClass{<replaceable>code</replaceable>}</literal>
</term></varlistentry><varlistentry><term><literal>\AtEndOfPackage{<replaceable>code</replaceable>}</literal>
</term><listitem><indexterm role="fn"><primary>\AtEndOfClass</primary></indexterm>
<indexterm role="fn"><primary>\AtEndOfPackage</primary></indexterm>
<para>Hook to insert <replaceable>code</replaceable> to be executed when &latex; finishes
processing the current class or package.  You can use these hooks
multiple times; the <literal>code</literal> will be executed in the order that you
called it.  See also <link linkend="_005cAtBeginDocument">\AtBeginDocument</link>.
</para>
</listitem></varlistentry><varlistentry><term><literal>\CheckCommand{<replaceable>cmd</replaceable>}[<replaceable>num</replaceable>][<replaceable>default</replaceable>]{<replaceable>definition</replaceable>}</literal>
</term></varlistentry><varlistentry><term><literal>\CheckCommand*{<replaceable>cmd</replaceable>}[<replaceable>num</replaceable>][<replaceable>default</replaceable>]{<replaceable>definition</replaceable>}</literal>
</term><listitem><indexterm role="fn"><primary>\CheckCommand</primary></indexterm>
<indexterm role="fn"><primary>\CheckCommand*</primary></indexterm>
<indexterm role="cp"><primary>new command, check</primary></indexterm>
<para>Like <literal>\newcommand</literal> (see <link linkend="_005cnewcommand-_0026-_005crenewcommand">\newcommand &amp; \renewcommand</link>) but does
not define <replaceable>cmd</replaceable>; instead it checks that the current definition of
<replaceable>cmd</replaceable> is exactly as given by <replaceable>definition</replaceable> and is or is not 
<indexterm role="cp"><primary>long command</primary></indexterm>
<firstterm>long</firstterm> as expected.  A long command is a command that accepts
<literal>\par</literal> within an argument.  The <replaceable>cmd</replaceable> command is expected to be
long with the unstarred version of <literal>\CheckCommand</literal>.  Raises an
error when the check fails.  This allows you to check before you start
redefining <literal>cmd</literal> yourself that no other package has already
redefined this command.
</para>
</listitem></varlistentry><varlistentry><term><literal>\ClassError{<replaceable>class name</replaceable>}{<replaceable>error text</replaceable>}{<replaceable>help text</replaceable>}</literal>
</term></varlistentry><varlistentry><term><literal>\PackageError{<replaceable>package name</replaceable>}{<replaceable>error text</replaceable>}{<replaceable>help text</replaceable>}</literal>
</term></varlistentry><varlistentry><term><literal>\ClassWarning{<replaceable>class name</replaceable>}{<replaceable>warning text</replaceable>}</literal>
</term></varlistentry><varlistentry><term><literal>\PackageWarning{<replaceable>package name</replaceable>}{<replaceable>warning text</replaceable>}</literal>
</term></varlistentry><varlistentry><term><literal>\ClassWarningNoLine{<replaceable>class name</replaceable>}{<replaceable>warning text</replaceable>}</literal>
</term></varlistentry><varlistentry><term><literal>\PackageWarningNoLine{<replaceable>package name</replaceable>}{<replaceable>warning text</replaceable>}</literal>
</term></varlistentry><varlistentry><term><literal>\ClassInfo{<replaceable>class name</replaceable>}{<replaceable>info text</replaceable>}</literal>
</term></varlistentry><varlistentry><term><literal>\PackageInfo{<replaceable>package name</replaceable>}{<replaceable>info text</replaceable>}</literal>
</term></varlistentry><varlistentry><term><literal>\ClassInfoNoLine{<replaceable>class name</replaceable>}{<replaceable>info text</replaceable>}</literal>
</term></varlistentry><varlistentry><term><literal>\PackageInfoNoLine{<replaceable>package name</replaceable>}{<replaceable>info text</replaceable>}</literal>
</term><listitem><indexterm role="fn"><primary>\ClassError</primary></indexterm>
<indexterm role="fn"><primary>\PackageError</primary></indexterm>
<indexterm role="fn"><primary>\ClassWarning</primary></indexterm>
<indexterm role="fn"><primary>\PackageWarning</primary></indexterm>
<indexterm role="fn"><primary>\ClassWarningNoLine</primary></indexterm>
<indexterm role="fn"><primary>\PackageWarningNoLine</primary></indexterm>
<indexterm role="fn"><primary>\ClassInfo</primary></indexterm>
<indexterm role="fn"><primary>\PackageInfo</primary></indexterm>
<indexterm role="fn"><primary>\ClassInfoNoLine</primary></indexterm>
<indexterm role="fn"><primary>\PackageInfoNoLine</primary></indexterm>
<para>Produce an error message, or warning or informational messages.
</para>
<para>For <literal>\ClassError</literal> and <literal>\PackageError</literal> the message is
<replaceable>error text</replaceable>, followed by &tex;&#8217;s <literal>?</literal> error prompt. If the
user then asks for help by typing <literal>h</literal>, they see the <replaceable>help
text</replaceable>.
</para>
<para>The four warning commands are similar except that they write
<replaceable>warning text</replaceable> on the screen with no error prompt.  The four info
commands write <replaceable>info text</replaceable> only in the transcript file.  The
<literal>NoLine</literal> versions do not show the number of the line generating the
message, while the other versions do show that number.
</para>
<para>To format the messages, including the <replaceable>help text</replaceable>: use
<literal>\protect</literal> to stop a command from expanding, get a line break with
<literal>\MessageBreak</literal>, and get a space with <literal>\space</literal> when a space
character does not allow it, like after a command.  Note that &latex;
appends a period to the messages.
</para>
</listitem></varlistentry><varlistentry><term><literal>\CurrentOption</literal>
</term><listitem><indexterm role="fn"><primary>\CurrentOption</primary></indexterm>
<para>Expands to the name of the currently-being-processed option.  Can only
be used within the <replaceable>code</replaceable> argument of either <literal>\DeclareOption</literal>
or <literal>\DeclareOption*</literal>.
</para>
</listitem></varlistentry><varlistentry><term><literal>\DeclareOption{<replaceable>option</replaceable>}{<replaceable>code</replaceable>}</literal>
</term></varlistentry><varlistentry><term><literal>\DeclareOption*{<replaceable>code</replaceable>}</literal>
</term><listitem><indexterm role="fn"><primary>\DeclareOption</primary></indexterm>
<indexterm role="fn"><primary>\DeclareOption*</primary></indexterm>
<indexterm role="cp"><primary>class options</primary></indexterm>
<indexterm role="cp"><primary>package options</primary></indexterm>
<indexterm role="cp"><primary>options, class</primary></indexterm>
<indexterm role="cp"><primary>options, package</primary></indexterm>
<para>Make an option available to a user to invoke in their
<literal>\documentclass</literal> command.  For example, the <literal>smcmemo</literal> class
could have an option <literal>\documentclass[logo]{smcmemo}</literal> allowing
users to put the institutional logo on the first page.  The class file
must contain <literal>\DeclareOption{logo}{<replaceable>code</replaceable>}</literal> (and later,
<literal>\ProcessOptions</literal>).
</para>
<para>If you request an option that has not been declared, by default this
will produce a warning like <literal>Unused global option(s): [badoption].</literal>
Change this behaviour with the starred version
<literal>\DeclareOption*{<replaceable>code</replaceable>}</literal>.  For example, many classes extend
an existing class, using a declaration such as
<literal>\LoadClass{article}</literal>, and for passing extra options to the
underlying class use code such as this.
</para>
<screen>\DeclareOption*{%
\PassOptionsToClass{\CurrentOption}{article}%
}
</screen>
<para>Another example is that the class <literal>smcmemo</literal> may allow users to keep
lists of memo recipients in external files.  Then the user could invoke
<literal>\documentclass[math]{smcmemo}</literal> and it will read the file
<literal>math.memo</literal>.  This code handles the file if it exists and otherwise
passes the option to the <literal>article</literal> class.
</para>
<screen>\DeclareOption*{\InputIfFileExists{\CurrentOption.memo}{}{%
    \PassOptionsToClass{\CurrentOption}{article}}}
</screen>
</listitem></varlistentry><varlistentry><term><literal>\DeclareRobustCommand{<replaceable>cmd</replaceable>}[<replaceable>num</replaceable>][<replaceable>default</replaceable>]{<replaceable>definition</replaceable>}</literal>
</term></varlistentry><varlistentry><term><literal>\DeclareRobustCommand*{<replaceable>cmd</replaceable>}[<replaceable>num</replaceable>][<replaceable>default</replaceable>]{<replaceable>definition</replaceable>}</literal>
</term><listitem><indexterm role="fn"><primary>\DeclareRobustCommand</primary></indexterm>
<indexterm role="fn"><primary>\DeclareRobustCommand*</primary></indexterm>
<indexterm role="cp"><primary>new command, definition</primary></indexterm>
<para>Like <literal>\newcommand</literal> and <literal>\newcommand*</literal> (see <link linkend="_005cnewcommand-_0026-_005crenewcommand">\newcommand &amp;
\renewcommand</link>) but these declare a robust command, even if some code
within the <replaceable>definition</replaceable> is fragile.  (For a discussion of robust and
fragile commands see <link linkend="_005cprotect">\protect</link>.) Use this command to define new
robust commands or to redefine existing commands and make them
robust. Unlike <literal>\newcommand</literal> these do not give an error if macro
<replaceable>cmd</replaceable> already exists; instead, a log message is put into the
transcript file if a command is redefined.
</para>
<para>Commands defined this way are a bit less efficient than those defined
using <literal>\newcommand</literal> so unless the command&#8217;s data is fragile and the
command is used within a moving argument, use <literal>\newcommand</literal>.
</para>
<indexterm role="cp"><primary>package, <literal>etoolbox</literal></primary></indexterm>
<indexterm role="cp"><primary><literal>etoolbox</literal> package</primary></indexterm>
<para>The <filename>etoolbox</filename> package offers the commands
<literal>\newrobustcmd</literal>, <literal>\newrobustcmd*</literal>, as well as the commands
<literal>\renewrobustcmd</literal>, <literal>\renewrobustcmd*</literal>, and the commands
<literal>\providerobustcmd</literal>, and <literal>\providerobustcmd*</literal>.  These are
similar to <literal>\newcommand</literal>, <literal>\newcommand*</literal>,
<literal>\renewcommand</literal>, <literal>\renewcommand*</literal>, <literal>\providecommand</literal>, and
<literal>\providecommand*</literal>, but define a robust <replaceable>cmd</replaceable> with two
advantages as compared to <literal>\DeclareRobustCommand</literal>:
</para><orderedlist numeration="arabic"><listitem><para>They use the low-level e-&tex; protection mechanism rather than the
higher level &latex; <literal>\protect</literal> mechanism, so they do not incur
the slight loss of performance mentioned above, and
</para></listitem><listitem><para>They make the same distinction between <literal>\new&#8230;</literal>,
<literal>\renew&#8230;</literal>, and <literal>\provide&#8230;</literal>, as the standard
commands, so they do not just make a log message when you redefine
<replaceable>cmd</replaceable> that already exists, in that case you need to use either
<literal>\renew&#8230;</literal> or <literal>\provide&#8230;</literal> or you get an error.
</para></listitem></orderedlist>

</listitem></varlistentry><varlistentry><term><literal>\IfFileExists{<replaceable>file name</replaceable>}{<replaceable>true code</replaceable>}{<replaceable>false code</replaceable>}</literal>
</term></varlistentry><varlistentry><term><literal>\InputIfFileExists{<replaceable>file name</replaceable>}{<replaceable>true code</replaceable>}{<replaceable>false code</replaceable>}</literal>
</term><listitem><indexterm role="fn"><primary>\IfFileExists</primary></indexterm>
<indexterm role="fn"><primary>\InputIfFileExists</primary></indexterm>
<para>Execute <replaceable>true code</replaceable> if &latex; finds the file <filename><replaceable>file
name</replaceable></filename> or <replaceable>false code</replaceable> otherwise.  In the first case it executing
<replaceable>true code</replaceable> and then inputs the file.  Thus the command
</para>
<screen>\IfFileExists{img.pdf}{%
  \includegraphics{img.pdf}}{\typeout{!! img.pdf not found}
</screen>
<para>will include the graphic <filename>img.pdf</filename> if it is found and otherwise
give a warning.
</para>
<para>This command looks for the file in all search paths that &latex; uses,
not only in the current directory.  To look only in the current
directory do something like <literal>\IfFileExists{./filename}{<replaceable>true
code</replaceable>}{<replaceable>false code</replaceable>}</literal>.  If you ask for a filename without a
<literal>.tex</literal> extension then &latex; will first look for the file by
appending the <literal>.tex</literal>; for more on how &latex; handles file
extensions see <link linkend="_005cinput">\input</link>.
</para>
</listitem></varlistentry><varlistentry><term><literal>\LoadClass[<replaceable>options list</replaceable>]{<replaceable>class name</replaceable>}[<replaceable>release date</replaceable>]</literal>
</term></varlistentry><varlistentry><term><literal>\LoadClassWithOptions{<replaceable>class name</replaceable>}[<replaceable>release date</replaceable>]</literal>
</term><listitem><indexterm role="fn"><primary>\LoadClass</primary></indexterm>
<indexterm role="fn"><primary>\LoadClassWithOptions</primary></indexterm>
<para>Load a class, as with <literal>\documentclass[<replaceable>options
list</replaceable>]{<replaceable>class name</replaceable>}[<replaceable>release info</replaceable>]</literal>.  An example is
<literal>\LoadClass[twoside]{article}</literal>.
</para>
<para>The <replaceable>options list</replaceable>, if present, is a comma-separated list.  The
<replaceable>release date</replaceable> is optional.  If present it must have the form
<replaceable>YYYY/MM/DD</replaceable>.
<!-- BTW, there are at-macros documented in macros2e.pdf to check the version -->
<!-- and do some actions conditionnally on version later or not to some -->
<!-- date. -->
</para>
<para>If you request a <replaceable>release date</replaceable> and the date of the package
installed on your system is earlier, then you get a warning on the
screen and in the log like this.
</para>
<screen>You have requested, on input line 4, version `2038/01/19' of
document class article, but only version `2014/09/29 v1.4h
Standard LaTeX document class' is available.
</screen>
<para>The command version <literal>\LoadClassWithOptions</literal> uses the list of
options for the current class.  This means it ignores any options passed
to it via <literal>\PassOptionsToClass</literal>.  This is a convenience command
that lets you build classes on existing ones, such as the standard
<literal>article</literal> class, without having to track which options were passed.
</para>
</listitem></varlistentry><varlistentry><term><literal>\ExecuteOptions{<replaceable>options-list</replaceable>}</literal>
</term><listitem><indexterm role="fn"><primary>\ExecuteOptions</primary></indexterm>
<para>For each option <replaceable>option</replaceable> in the <replaceable>options-list</replaceable>, in order, this command
executes the command <literal>\ds@<replaceable>option</replaceable></literal>.  If this command is not
defined then that option is silently ignored.
</para>
<para>It can be used to provide a default option list before
<literal>\ProcessOptions</literal>.  For example, if in a class file you want the
default to be 11pt fonts then you could specify
<literal>\ExecuteOptions{11pt}\ProcessOptions\relax</literal>.
</para>
</listitem></varlistentry><varlistentry><term><literal>\NeedsTeXFormat{<replaceable>format</replaceable>}[<replaceable>format date</replaceable>]</literal>
</term><listitem><indexterm role="fn"><primary>\NeedsTeXFormat</primary></indexterm>
<para>Specifies the format that this class must be run under.  Often issued
as the first line of a class file, and most often used as:
<literal>\NeedsTeXFormat{LaTeX2e}</literal>.  When a document using that class is
processed, the format name given here must match the format that is
actually being run (including that the <replaceable>format</replaceable> string is case
sensitive).  If it does not match then execution stops with an error
like &#8216;<literal>This file needs format `LaTeX2e' but this is `xxx'.</literal>&#8217;
</para>
<para>To specify a version of the format that you know to have certain
features, include the optional <replaceable>format date</replaceable> on which those features
were implemented.  If present it must be in the form <literal>YYYY/MM/DD</literal>.
If the format version installed on your system is earlier than
<replaceable>format date</replaceable> then you get a warning like this.
</para>
<screen>You have requested release `2038/01/20' of LaTeX, but only
release `2016/02/01' is available.
</screen>
</listitem></varlistentry><varlistentry><term><literal>\OptionNotUsed</literal>
</term><listitem><indexterm role="fn"><primary>\OptionNotUsed</primary></indexterm>
<para>Adds the current option to the list of unused options.  Can only be used
within the <replaceable>code</replaceable> argument of either <literal>\DeclareOption</literal> or
<literal>\DeclareOption*</literal>.
</para>
<!-- I cannot reproduce this behavior as it is documented in clsguide. -->
<!-- In the absence of a @code{\DeclareOption*} declaration, @LaTeX{} issues -->
<!-- on the console a warning like @code{LaTeX Warning: Unused global -->
<!-- option(s): [unusedoption].} with the list of not-used options when it -->
<!-- reaches @code{\begin@{document@}}. -->

</listitem></varlistentry><varlistentry><term><literal>\PassOptionsToClass{<replaceable>option list</replaceable>}{<replaceable>class name</replaceable>}</literal>
</term></varlistentry><varlistentry><term><literal>\PassOptionsToPackage{<replaceable>option list</replaceable>}{<replaceable>package name</replaceable>}</literal>
</term><listitem><indexterm role="fn"><primary>\PassOptionsToClass</primary></indexterm>
<indexterm role="fn"><primary>\PassOptionsToPackage</primary></indexterm>
<para>Adds the options in the comma-separated list <replaceable>option list</replaceable> to the
options used by any future <literal>\RequirePackage</literal> or <literal>\usepackage</literal>
command for package <replaceable>package name</replaceable> or the class <replaceable>class name</replaceable>.
</para>
<para>The reason for these commands is: you may load a package any number of
times with no options but if you want options then you may only supply
them when you first load the package.  Loading a package with options
more than once will get you an error like <literal>Option clash for package
foo.</literal> (&latex; throws an error even if there is no conflict between the
options.)
</para>
<para>If your own code is bringing in a package twice then you can collapse
that to once, for example replacing the two
<literal>\RequirePackage[landscape]{geometry}</literal> and
<literal>\RequirePackage[margins=1in]{geometry}</literal> with the single command
<literal>\RequirePackage[landscape,margins=1in]{geometry}</literal>.
</para>
<para>However, imagine that you are loading <filename>firstpkg</filename> and inside that
package it loads <filename>secondpkg</filename>, and you need the second package to be
loaded with option <literal>draft</literal>.  Then before doing the first package
you must queue up the options for the second package, like this.
</para>
<screen>\PassOptionsToPackage{draft}{secondpkg}
\RequirePackage{firstpkg}
</screen>
<para>(If <literal>firstpkg.sty</literal> loads an option in conflict with what you want
then you may have to alter its source.)
</para>
<para>These commands are useful for general users as well as class and package
writers.  For instance, suppose a user wants to load the <literal>graphicx</literal>
package with the option <literal>draft</literal> and also wants to use a class
<literal>foo</literal> that loads the <literal>graphicx</literal> package, but without that
option. The user could start their &latex; file with
<literal>\PassOptionsToPackage{draft}{graphicx}\documentclass{foo}</literal>.
</para>
</listitem></varlistentry><varlistentry><term><literal>\ProcessOptions</literal>
</term></varlistentry><varlistentry><term><literal>\ProcessOptions*<replaceable>\@options</replaceable></literal>
</term><listitem><indexterm role="fn"><primary>\ProcessOptions</primary></indexterm>
<indexterm role="fn"><primary>\ProcessOptions*</primary></indexterm>
<para>Execute the code for each option that the user has invoked.  Include it
in the class file as <literal>\ProcessOptions\relax</literal> (because of the
existence of the starred command).
</para>
<para>Options come in two types.  <firstterm>Local options</firstterm> have been specified for this
particular package in the <replaceable>options</replaceable> argument of
<literal>\PassOptionsToPackage{<replaceable>options</replaceable>}</literal>,
<literal>\usepackage[<replaceable>options</replaceable>]</literal>, or
<literal>\RequirePackage[<replaceable>options</replaceable>]</literal>.  <firstterm>Global options</firstterm> are those given
by the class user in <literal>\documentclass[<replaceable>options</replaceable>]</literal> (If an option
is specified both locally and globally then it is local.)
</para>
<para>When <literal>\ProcessOptions</literal> is called for a package <filename>pkg.sty</filename>, the
following happens:
</para><orderedlist numeration="arabic"><listitem><para>For each option <replaceable>option</replaceable> so far declared
with <literal>\DeclareOption</literal>, it looks to see if that option is either a
global or a local option for <literal>pkg</literal>. If so then it executes the
declared code.  This is done in the order in which these options were
given in <filename>pkg.sty</filename>.
</para></listitem><listitem><para>For each remaining local option, it executes the command
<literal>\ds@</literal><replaceable>option</replaceable> if it has been defined somewhere (other than by
a <literal>\DeclareOption</literal>); otherwise, it executes the default option code
given in <literal>\DeclareOption*</literal>. If no default option code has been
declared then it gives an error message.  This is done in the order in
which these options were specified.
</para></listitem></orderedlist>
<para>When <literal>\ProcessOptions</literal> is called for a class it works in the same
way except that all options are local, and the default <replaceable>code</replaceable> for
<literal>\DeclareOption*</literal> is <literal>\OptionNotUsed</literal> rather than an error.
</para>
<para>The starred version <literal>\ProcessOptions*</literal> executes the
options in the order specified in the calling commands, rather than in
the order of declaration in the class or package. For a package this
means that the global options are processed first.
</para>
</listitem></varlistentry><varlistentry><term><literal>\ProvidesClass{<replaceable>class name</replaceable>}[<replaceable>release date</replaceable> <replaceable>brief additional information</replaceable>]</literal>
</term></varlistentry><varlistentry><term><literal>\ProvidesClass{<replaceable>class name</replaceable>}[<replaceable>release date</replaceable>]</literal>
</term></varlistentry><varlistentry><term><literal>\ProvidesPackage{<replaceable>package name</replaceable>}[<replaceable>release date</replaceable> <replaceable>brief additional information</replaceable>]</literal>
</term></varlistentry><varlistentry><term><literal>\ProvidesPackage{<replaceable>package name</replaceable>}[<replaceable>release date</replaceable>]</literal>
</term><listitem><indexterm role="fn"><primary>\ProvidesClass</primary></indexterm>
<indexterm role="fn"><primary>\ProvidesPackage</primary></indexterm>
<para>Identifies the class or package, printing a message to the screen and
the log file.
</para>
<para>When you load a class or package, for example with
<literal>\documentclass{smcmemo}</literal> or <literal>\usepackage{test}</literal>, &latex;
inputs a file.  If the name of the file does not match the class or
package name declared in it then you get a warning.  Thus, if you invoke
<literal>\documentclass{smcmemo}</literal>, and the file <filename>smcmemo.cls</filename> has
the statement <literal>\ProvidesClass{xxx}</literal> then you get a warning like
<literal>You have requested document class `smcmemo', but the document
class provides 'xxx'.</literal>  This warning does not prevent &latex; from
processing the rest of the class file normally.
</para>
<para>If you include the optional argument then you must include a date,
before any spaces, of the form <literal>YYYY/MM/DD</literal>. The rest of the
optional argument is free-form, although it traditionally identifies the
class, and is written to the screen during compilation and to the log
file.  Thus, if your file <filename>smcmemo.cls</filename> contains the line
<literal>\ProvidesClass{smcmemo}[2008/06/01 v1.0 SMC memo class]</literal> and
your document&#8217;s first line is <literal>\documentclass{smcmemo}</literal> then you
will see <literal>Document Class: smcmemo 2008/06/01 v1.0 SMC memo class</literal>.
</para>
<para>The date in the optional argument allows class and package users to ask
to be warned if the version of the class or package is earlier than
<replaceable>release date</replaceable>.  For instance, a user could enter
<literal>\documentclass{smcmemo}[2018/10/12]</literal> or
<literal>\usepackage{foo}[[2017/07/07]]</literal> to require a class or package
with certain features by specifying that it must be released no earlier
than the given date.  (Although, in practice package users only rarely
include a date, and class users almost never do.)
</para>
</listitem></varlistentry><varlistentry><term><literal>\ProvidesFile{<replaceable>file name</replaceable>}[<replaceable>additional information</replaceable>]</literal>
</term><listitem><indexterm role="fn"><primary>\ProvidesFile</primary></indexterm>
<para>Declare a file other than the main class and package files, such as
configuration files or font definition files.  Put this command in that
file and you get in the log a string like <literal>File: test.config
2017/10/12 config file for test.cls</literal> for <replaceable>file name</replaceable> equal to
&#8216;<literal>test.config</literal>&#8217; and <replaceable>additional information</replaceable> equal to
&#8216;<literal>2017/10/12 config file for test.cls</literal>&#8217;.
</para>
</listitem></varlistentry><varlistentry><term><literal>\RequirePackage[<replaceable>option list</replaceable>]{<replaceable>package name</replaceable>}[<replaceable>release date</replaceable>]</literal>
</term></varlistentry><varlistentry><term><literal>\RequirePackageWithOptions{<replaceable>package name</replaceable>}[<replaceable>release date</replaceable>]</literal>
</term><listitem><indexterm role="fn"><primary>\RequirePackage</primary></indexterm>
<indexterm role="fn"><primary>\RequirePackageWithOptions</primary></indexterm>
<para>Load a package, like the command <literal>\usepackage</literal> (see <link linkend="Additional-packages">Additional
packages</link>). The &latex; development team strongly recommends use of
these commands over Plain&#160;&tex;&#8217;s <literal>\input</literal>; see the Class
Guide.  An example is
<literal>\RequirePackage[landscape,margin=1in]{geometry}</literal>.
</para>
<para>The <replaceable>option list</replaceable>, if present, is a comma-separated list.  The
<replaceable>release date</replaceable>, if present, must have the form <replaceable>YYYY/MM/DD</replaceable>.  If
the release date of the package as installed on your system is earlier
than <replaceable>release date</replaceable> then you get a warning like <literal>You have
requested, on input line 9, version `2017/07/03' of package jhtest, but
only version `2000/01/01' is available</literal>.
</para>
<para>The <literal>\RequirePackageWithOptions</literal> version uses the list of options
for the current class.  This means it ignores any options passed to it
via <literal>\PassOptionsToClass</literal>.  This is a convenience command to allow
easily building classes on existing ones without having to track which
options were passed.
</para>
<para>The difference between <literal>\usepackage</literal> and <literal>\RequirePackage</literal> is
small.  The <literal>\usepackage</literal> command is intended for the document file
while <literal>\RequirePackage</literal> is intended for package and class files.
Thus, using <literal>\usepackage</literal> before the <literal>\documentclass</literal> command
causes &latex; to give error like <literal>\usepackage before
\documentclass</literal>, but you can use <literal>\RequirePackage</literal> there.
</para></listitem></varlistentry></variablelist>

</sect2>
</sect1>
</chapter>
<chapter label="4" id="Fonts">
<title>Fonts</title>
<anchor id="Typefaces"/><!-- old name -->

<indexterm role="cp"><primary>typefaces</primary></indexterm>
<indexterm role="cp"><primary>fonts</primary></indexterm>

<para>Two important aspects of selecting a <firstterm>font</firstterm> are specifying a size
and a style.  The &latex; commands for doing this are described here.
</para>


<sect1 label="4.1" id="Font-styles">
<title>Font styles</title>

<indexterm role="cp"><primary>font styles</primary></indexterm>
<indexterm role="cp"><primary>type styles</primary></indexterm>
<indexterm role="cp"><primary>styles of text</primary></indexterm>

<para>The following type style commands are supported by &latex;.
</para>
<para>In the table below the listed commands, the <literal>\text...</literal>  commands,
is used with an argument, as in <literal>\textit{<replaceable>text</replaceable>}</literal>.  This is
the preferred form.  But shown after it, in parenthesis, is the
corresponding declaration form, which is sometimes useful.  This form
takes no arguments, as in <literal>{\itshape <replaceable>text</replaceable>}</literal>.  The scope of
the declaration form lasts until the next type style command or the end
of the current group.  In addition, each has an environment form such as
<literal>\begin{itshape}...\end{itshape}</literal>.
</para>
<para>These commands, in both the argument form and the declaration form, are
cumulative; for instance you can get bold sans serif by saying either of
<literal>\sffamily\bfseries</literal> or <literal>\bfseries\sffamily</literal>.
</para>
<indexterm role="fn"><primary>\nocorrlist</primary></indexterm>
<indexterm role="fn"><primary>\nocorr</primary></indexterm>
<para>One advantage of these commands is that they automatically insert italic
corrections if needed (see <link linkend="_005c_002f">\/</link>).  Specifically, they insert the
italic correction unless the following character is in the list
<literal>\nocorrlist</literal>, which by default consists of a period and a comma.
To suppress the automatic insertion of italic correction, use
<literal>\nocorr</literal> at the start or end of the command argument, such as
<literal>\textit{\nocorr text}</literal> or <literal>\textsc{text \nocorr}</literal>.
</para>
<variablelist><varlistentry><term><literal>\textrm (\rmfamily)</literal>
</term><listitem><indexterm role="fn"><primary>\textrm</primary></indexterm>
<indexterm role="fn"><primary>\rmfamily</primary></indexterm>
<para>Roman.
</para>
</listitem></varlistentry><varlistentry><term><literal>\textit (\itshape)</literal>
</term><listitem><indexterm role="fn"><primary>\textit</primary></indexterm>
<indexterm role="fn"><primary>\itshape</primary></indexterm>
<para>Italics.
</para>
</listitem></varlistentry><varlistentry><term><literal>\textmd (\mdseries)</literal>
</term><listitem><indexterm role="fn"><primary>\textmd</primary></indexterm>
<indexterm role="fn"><primary>\mdseries</primary></indexterm>
<para>Medium weight (default).
</para>
</listitem></varlistentry><varlistentry><term><literal>\textbf (\bfseries)</literal>
</term><listitem><indexterm role="fn"><primary>\textbf</primary></indexterm>
<indexterm role="fn"><primary>\bfseries</primary></indexterm>
<para>Boldface.
</para>
</listitem></varlistentry><varlistentry><term><literal>\textup (\upshape)</literal>
</term><listitem><indexterm role="fn"><primary>\textup</primary></indexterm>
<indexterm role="fn"><primary>\upshape</primary></indexterm>
<para>Upright (default).
</para>
</listitem></varlistentry><varlistentry><term><literal>\textsl (\slshape)</literal>
</term><listitem><indexterm role="fn"><primary>\textsl</primary></indexterm>
<indexterm role="fn"><primary>\slshape</primary></indexterm>
<para>Slanted.
</para>
</listitem></varlistentry><varlistentry><term><literal>\textsf (\sffamily)</literal>
</term><listitem><indexterm role="fn"><primary>\textsf</primary></indexterm>
<indexterm role="fn"><primary>\sffamily</primary></indexterm>
<para>Sans serif.
</para>
</listitem></varlistentry><varlistentry><term><literal>\textsc (\scshape)</literal>
</term><listitem><indexterm role="fn"><primary>\textsc</primary></indexterm>
<indexterm role="fn"><primary>\scshape</primary></indexterm>
<para>Small caps.
</para>
</listitem></varlistentry><varlistentry><term><literal>\texttt (\ttfamily)</literal>
</term><listitem><indexterm role="fn"><primary>\texttt</primary></indexterm>
<indexterm role="fn"><primary>\ttfamily</primary></indexterm>
<para>Typewriter.
</para>
</listitem></varlistentry><varlistentry><term><literal>\textnormal (\normalfont)</literal>
</term><listitem><indexterm role="fn"><primary>\textnormal</primary></indexterm>
<indexterm role="fn"><primary>\normalfont</primary></indexterm>
<para>Main document font.
</para>
</listitem></varlistentry></variablelist>
<indexterm role="cp"><primary>emphasis</primary></indexterm>
<indexterm role="fn"><primary>\emph</primary></indexterm>
<para>Although it also changes fonts, the <literal>\emph{<replaceable>text</replaceable>}</literal> command
is semantic, for text to be emphasized, and should not be used as a
substitute for <literal>\textit</literal>.  For example, <literal>\emph{<replaceable>start
text</replaceable> \emph{<replaceable>middle text</replaceable>} <replaceable>end text</replaceable>}</literal> will result in the
<replaceable>start text</replaceable> and <replaceable>end text</replaceable> in italics, but <replaceable>middle text</replaceable>
will be in roman.
</para>
<para>&latex; also provides the following commands, which unconditionally
switch to the given style, that is, are <emphasis>not</emphasis> cumulative.  They are
used as declarations: <literal>{\<replaceable>cmd</replaceable>...}</literal> instead of
<literal>\<replaceable>cmd</replaceable>{...}</literal>.
</para>
<para>(The unconditional commands below are an older version of font
switching. The earlier commands are an improvement in most
circumstances. But sometimes an unconditional font switch is precisely
what you want.)
</para>
<variablelist><varlistentry><term><indexterm role="fn"><primary>\bf</primary></indexterm><literal>\bf</literal>
</term><listitem><indexterm role="cp"><primary>bold font</primary></indexterm>
<para>Switch to bold face.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\cal</primary></indexterm><literal>\cal</literal>
</term><listitem><indexterm role="cp"><primary>script letters for math</primary></indexterm>
<indexterm role="cp"><primary>calligraphic letters for math</primary></indexterm>
<para>Switch to calligraphic letters for math.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\it</primary></indexterm><literal>\it</literal>
</term><listitem><indexterm role="cp"><primary>italic font</primary></indexterm>
<para>Italics.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\rm</primary></indexterm><literal>\rm</literal>
</term><listitem><indexterm role="cp"><primary>roman font</primary></indexterm>
<para>Roman.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\sc</primary></indexterm><literal>\sc</literal>
</term><listitem><indexterm role="cp"><primary>small caps font</primary></indexterm>
<para>Small caps.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\sf</primary></indexterm><literal>\sf</literal>
</term><listitem><indexterm role="cp"><primary>sans serif font</primary></indexterm>
<para>Sans serif.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\sl</primary></indexterm><literal>\sl</literal>
</term><listitem><indexterm role="cp"><primary>slanted font</primary></indexterm>
<indexterm role="cp"><primary>oblique font</primary></indexterm>
<para>Slanted (oblique).
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\tt</primary></indexterm><literal>\tt</literal>
</term><listitem><indexterm role="cp"><primary>typewriter font</primary></indexterm>
<indexterm role="cp"><primary>monospace font</primary></indexterm>
<indexterm role="cp"><primary>fixed-width font</primary></indexterm>
<para>Typewriter (monospace, fixed-width).
</para>
</listitem></varlistentry></variablelist>
<para>The <literal>\em</literal> command is the unconditional version of <literal>\emph</literal>.
</para>
<para>The following commands are for use in math mode.  They are not
cumulative, so <literal>\mathbf{\mathit{<replaceable>symbol</replaceable>}}</literal> does not
create a boldface and italic <replaceable>symbol</replaceable>; instead, it will just be in
italics.  This is because typically math symbols need consistent
typographic treatment, regardless of the surrounding environment.
</para>
<variablelist><varlistentry><term><literal>\mathrm</literal>
</term><listitem><indexterm role="fn"><primary>\mathrm</primary></indexterm>
<para>Roman, for use in math mode.
</para>
</listitem></varlistentry><varlistentry><term><literal>\mathbf</literal>
</term><listitem><indexterm role="fn"><primary>\mathbf</primary></indexterm>
<para>Boldface, for use in math mode.
</para>
</listitem></varlistentry><varlistentry><term><literal>\mathsf</literal>
</term><listitem><indexterm role="fn"><primary>\mathsf</primary></indexterm>
<para>Sans serif, for use in math mode.
</para>
</listitem></varlistentry><varlistentry><term><literal>\mathtt</literal>
</term><listitem><indexterm role="fn"><primary>\mathtt</primary></indexterm>
<para>Typewriter, for use in math mode.
</para>
</listitem></varlistentry><varlistentry><term><literal>\mathit</literal>
</term><term><literal>(\mit)</literal>
</term><listitem><para>Italics, for use in math mode.
</para>
</listitem></varlistentry><varlistentry><term><literal>\mathnormal</literal>
</term><listitem><indexterm role="fn"><primary>\mathnormal</primary></indexterm>
<para>For use in math mode, e.g., inside another type style declaration.
</para>
</listitem></varlistentry><varlistentry><term><literal>\mathcal</literal>
</term><listitem><indexterm role="fn"><primary>\mathcal</primary></indexterm>
<para>Calligraphic letters, for use in math mode.
</para>
</listitem></varlistentry></variablelist>
<indexterm role="fn"><primary>\mathversion</primary></indexterm>
<indexterm role="cp"><primary>math, bold</primary></indexterm>
<indexterm role="cp"><primary>bold math</primary></indexterm>
<para>In addition, the command <literal>\mathversion{bold}</literal> can be used for
switching to bold letters and symbols in
formulas. <literal>\mathversion{normal}</literal> restores the default.
</para>
<indexterm role="fn"><primary>\oldstylenums</primary></indexterm>
<indexterm role="cp"><primary>numerals, old-style</primary></indexterm>
<indexterm role="cp"><primary>old-style numerals</primary></indexterm>
<indexterm role="cp"><primary>lining numerals</primary></indexterm>
<indexterm role="cp"><primary>package, <literal>textcomp</literal></primary></indexterm>
<indexterm role="cp"><primary><literal>textcomp</literal> package</primary></indexterm>

<para>Finally, the command <literal>\oldstylenums{<replaceable>numerals</replaceable>}</literal> will typeset
so-called &#8220;old-style&#8221; numerals, which have differing heights and
depths (and sometimes widths) from the standard &#8220;lining&#8221; numerals,
which all have the same height as uppercase letters.  &latex;&#8217;s
default fonts support this, and will respect <literal>\textbf</literal> (but not
other styles; there are no italic old-style numerals in Computer
Modern).  Many other fonts have old-style numerals also; sometimes the
<literal>textcomp</literal> package must be loaded, and sometimes package options
are provided to make them the default.  FAQ entry:
<ulink url="http://www.tex.ac.uk/cgi-bin/texfaq2html?label=osf">http://www.tex.ac.uk/cgi-bin/texfaq2html?label=osf</ulink>.
</para>

</sect1>
<sect1 label="4.2" id="Font-sizes">
<title>Font sizes</title>

<indexterm role="cp"><primary>font sizes</primary></indexterm>
<indexterm role="cp"><primary>typeface sizes</primary></indexterm>
<indexterm role="cp"><primary>sizes of text</primary></indexterm>

<para>The following standard type size commands are supported by &latex;.
The table shows the command name and the corresponding actual font
size used (in points) with the &#8216;<literal>10pt</literal>&#8217;, &#8216;<literal>11pt</literal>&#8217;, and
&#8216;<literal>12pt</literal>&#8217; document size options, respectively (see <link linkend="Document-class-options">Document class
options</link>).
</para>
<indexterm role="fn"><primary>\tiny</primary></indexterm>
<indexterm role="fn"><primary>\scriptsize</primary></indexterm>
<indexterm role="fn"><primary>\footnotesize</primary></indexterm>
<indexterm role="fn"><primary>\small</primary></indexterm>
<indexterm role="fn"><primary>\normalsize</primary></indexterm>
<indexterm role="fn"><primary>\large</primary></indexterm>
<indexterm role="fn"><primary>\Large</primary></indexterm>
<indexterm role="fn"><primary>\LARGE</primary></indexterm>
<indexterm role="fn"><primary>\huge</primary></indexterm>
<indexterm role="fn"><primary>\Huge</primary></indexterm>

<informaltable><tgroup cols="4"><colspec colwidth="23*"></colspec><colspec colwidth="7*"></colspec><colspec colwidth="7*"></colspec><colspec colwidth="5*"></colspec><thead><row><entry><para>Command  </para></entry><entry><para><literal>10pt</literal>  </para></entry><entry><para><literal>11pt</literal>  </para></entry><entry><para><literal>12pt</literal>
</para></entry></row></thead><tbody><row><entry><para><literal>\tiny</literal>
</para></entry><entry><para>5          </para></entry><entry><para>6          </para></entry><entry><para>6
</para></entry></row><row><entry><para><literal>\scriptsize</literal>
</para></entry><entry><para>7          </para></entry><entry><para>8          </para></entry><entry><para>8
</para></entry></row><row><entry><para><literal>\footnotesize</literal>
</para></entry><entry><para>8          </para></entry><entry><para>9          </para></entry><entry><para>10
</para></entry></row><row><entry><para><literal>\small</literal>
</para></entry><entry><para>9          </para></entry><entry><para>10         </para></entry><entry><para>10.95
</para></entry></row><row><entry><para><literal>\normalsize</literal> (default)
</para></entry><entry><para>10         </para></entry><entry><para>10.95      </para></entry><entry><para>12
</para></entry></row><row><entry><para><literal>\large</literal>
</para></entry><entry><para>12         </para></entry><entry><para>12         </para></entry><entry><para>14.4
</para></entry></row><row><entry><para><literal>\Large</literal>
</para></entry><entry><para>14.4       </para></entry><entry><para>14.4       </para></entry><entry><para>17.28
</para></entry></row><row><entry><para><literal>\LARGE</literal>
</para></entry><entry><para>17.28      </para></entry><entry><para>17.28      </para></entry><entry><para>20.74
</para></entry></row><row><entry><para><literal>\huge</literal>
</para></entry><entry><para>20.74      </para></entry><entry><para>20.74      </para></entry><entry><para>24.88
</para></entry></row><row><entry><para><literal>\Huge</literal>
</para></entry><entry><para>24.88      </para></entry><entry><para>24.88      </para></entry><entry><para>24.88
</para></entry></row></tbody></tgroup></informaltable>
<para>The commands are listed here in declaration forms. You use them by
declaring them, as with this example.
</para>
<screen>\begin{quotation} \small
  The Tao that can be named is not the eternal Tao.
\end{quotation}
</screen>
<para>The scope of the <literal>\small</literal> lasts until the end of the
<literal>quotation</literal> environment.  It would also end at the next type style
command or the end of the current group, so you could enclose it in
extra curly braces <literal>{\small We are here, we are here, we are
here!}</literal>.  You can instead use the environment form of these commands;
for instance, <literal>\begin{tiny}...\end{tiny}</literal>.
</para>

</sect1>
<sect1 label="4.3" id="Low_002dlevel-font-commands">
<title>Low-level font commands</title>

<indexterm role="cp"><primary>low-level font commands</primary></indexterm>
<indexterm role="cp"><primary>font commands, low-level</primary></indexterm>

<para>These commands are primarily intended for writers of macros and
packages.  The commands listed here are only a subset of the available
ones.
<!-- xx but it should be complete -->
<!-- xx something about ultimately reading ENCFAM.fd? -->
</para>
<variablelist><anchor id="low-level-font-commands-fontencoding"/><varlistentry><term><literal>\fontencoding{<replaceable>encoding</replaceable>}</literal>
</term><listitem><indexterm role="fn"><primary>\fontencoding</primary></indexterm>
<para>Select the font encoding, the encoding of the output font. There are a
large number of valid encodings.  The most common are <literal>OT1</literal>,
Knuth&#8217;s original encoding for Computer Modern (the default), and
<literal>T1</literal>, also known as the Cork encoding, which has support for the
accented characters used by the most widespread European languages
(German, French, Italian, Polish and others), which allows &tex; to
hyphenate words containing accented letters. For more, see
<ulink url="https://ctan.org/pkg/encguide">https://ctan.org/pkg/encguide</ulink>.
</para>
<anchor id="low-level-font-commands-fontfamily"/></listitem></varlistentry><varlistentry><term><literal>\fontfamily{<replaceable>family</replaceable>}</literal>
</term><listitem><indexterm role="fn"><primary>\fontfamily</primary></indexterm>
<indexterm role="cp"><primary>families, of fonts</primary></indexterm>
<indexterm role="cp"><primary>font catalogue</primary></indexterm>
<para>Select the font family.  The web page
<ulink url="http://www.tug.dk/FontCatalogue/">http://www.tug.dk/FontCatalogue/</ulink> provides one way to browse
through many of the fonts easily used with &latex;.  Here are
examples of some common families.
</para>
<informaltable><tgroup cols="2"><colspec colwidth="4*"></colspec><colspec colwidth="37*"></colspec><tbody><row><entry><para><literal>pag</literal>
</para></entry><entry><para>Avant Garde
</para></entry></row><row><entry><para><literal>fvs</literal>
</para></entry><entry><para>Bitstream Vera Sans
</para></entry></row><row><entry><para><literal>pbk</literal>
</para></entry><entry><para>Bookman
</para></entry></row><row><entry><para><literal>bch</literal>
</para></entry><entry><para>Charter
</para></entry></row><row><entry><para><literal>ccr</literal>
</para></entry><entry><para>Computer Concrete
</para></entry></row><row><entry><para><literal>cmr</literal>
</para></entry><entry><para>Computer Modern
</para></entry></row><row><entry><para><literal>cmss</literal>
</para></entry><entry><para>Computer Modern Sans Serif
</para></entry></row><row><entry><para><literal>cmtt</literal>
</para></entry><entry><para>Computer Modern Typewriter
</para></entry></row><row><entry><para><literal>pcr</literal>
</para></entry><entry><para>Courier
</para></entry></row><row><entry><para><literal>phv</literal>
</para></entry><entry><para>Helvetica
</para></entry></row><row><entry><para><literal>fi4</literal>
</para></entry><entry><para>Inconsolata
</para></entry></row><row><entry><para><literal>lmr</literal>
</para></entry><entry><para>Latin Modern
</para></entry></row><row><entry><para><literal>lmss</literal>
</para></entry><entry><para>Latin Modern Sans
</para></entry></row><row><entry><para><literal>lmtt</literal>
</para></entry><entry><para>Latin Modern Typewriter
</para></entry></row><row><entry><para><literal>pnc</literal>
</para></entry><entry><para>New Century Schoolbook
</para></entry></row><row><entry><para><literal>ppl</literal>
</para></entry><entry><para>Palatino
</para></entry></row><row><entry><para><literal>ptm</literal>
</para></entry><entry><para>Times
</para></entry></row><row><entry><para><literal>uncl</literal>
</para></entry><entry><para>Uncial
</para></entry></row><row><entry><para><literal>put</literal>
</para></entry><entry><para>Utopia
</para></entry></row><row><entry><para><literal>pzc</literal>
</para></entry><entry><para>Zapf Chancery
</para></entry></row></tbody></tgroup></informaltable>

<anchor id="low-level-font-commands-fontseries"/></listitem></varlistentry><varlistentry><term><literal>\fontseries{<replaceable>series</replaceable>}</literal>
</term><listitem><indexterm role="fn"><primary>\fontseries</primary></indexterm>
<indexterm role="cp"><primary>series, of fonts</primary></indexterm>
<para>Select the font series.  A <firstterm>series</firstterm> combines a <firstterm>weight</firstterm> and a
<firstterm>width</firstterm>.  Typically, a font supports only a few of the possible
combinations.  Some common combined series values include:
</para>
<informaltable><tgroup cols="2"><colspec colwidth="2*"></colspec><colspec colwidth="17*"></colspec><tbody><row><entry><para><literal>m</literal>
</para></entry><entry><para>Medium (normal)
</para></entry></row><row><entry><para><literal>b</literal>
</para></entry><entry><para>Bold
</para></entry></row><row><entry><para><literal>c</literal>
</para></entry><entry><para>Condensed
</para></entry></row><row><entry><para><literal>bc</literal>
</para></entry><entry><para>Bold condensed
</para></entry></row><row><entry><para><literal>bx</literal>
</para></entry><entry><para>Bold extended
</para></entry></row></tbody></tgroup></informaltable>
<indexterm role="cp"><primary>weights, of fonts</primary></indexterm>
<para>The possible values for weight, individually, are:
</para>
<informaltable><tgroup cols="2"><colspec colwidth="2*"></colspec><colspec colwidth="18*"></colspec><tbody><row><entry><para><literal>ul</literal>
</para></entry><entry><para>Ultra light
</para></entry></row><row><entry><para><literal>el</literal>
</para></entry><entry><para>Extra light
</para></entry></row><row><entry><para><literal>l</literal>
</para></entry><entry><para>Light
</para></entry></row><row><entry><para><literal>sl</literal>
</para></entry><entry><para>Semi light
</para></entry></row><row><entry><para><literal>m</literal>
</para></entry><entry><para>Medium (normal)
</para></entry></row><row><entry><para><literal>sb</literal>
</para></entry><entry><para>Semi bold
</para></entry></row><row><entry><para><literal>b</literal>
</para></entry><entry><para>Bold
</para></entry></row><row><entry><para><literal>eb</literal>
</para></entry><entry><para>Extra bold
</para></entry></row><row><entry><para><literal>ub</literal>
</para></entry><entry><para>Ultra bold
</para></entry></row></tbody></tgroup></informaltable> 
<indexterm role="cp"><primary>widths, of fonts</primary></indexterm>
<para>The possible values for width, individually, are (the meaning and
relationship of these terms varies with individual typefaces):
</para>
<informaltable><tgroup cols="2"><colspec colwidth="2*"></colspec><colspec colwidth="15*"></colspec><tbody><row><entry><para><literal>uc</literal>
</para></entry><entry><para>Ultra condensed
</para></entry></row><row><entry><para><literal>ec</literal>
</para></entry><entry><para>Extra condensed
</para></entry></row><row><entry><para><literal>c</literal>
</para></entry><entry><para>Condensed
</para></entry></row><row><entry><para><literal>sc</literal>
</para></entry><entry><para>Semi condensed
</para></entry></row><row><entry><para><literal>m</literal>
</para></entry><entry><para>Medium
</para></entry></row><row><entry><para><literal>sx</literal>
</para></entry><entry><para>Semi expanded
</para></entry></row><row><entry><para><literal>x</literal>
</para></entry><entry><para>Expanded
</para></entry></row><row><entry><para><literal>ex</literal>
</para></entry><entry><para>Extra expanded
</para></entry></row><row><entry><para><literal>ux</literal>
</para></entry><entry><para>Ultra expanded
</para></entry></row></tbody></tgroup></informaltable>
<para>When forming the <replaceable>series</replaceable> string from the weight and width, drop the
<literal>m</literal> that stands for medium weight or medium width, unless both
weight and width are <literal>m</literal>, in which case use just one
(&#8216;<literal><literal>m</literal></literal>&#8217;).
</para> 
</listitem></varlistentry><varlistentry><term><literal>\fontshape{<replaceable>shape</replaceable>}</literal>
</term><listitem><indexterm role="fn"><primary>\fontshape</primary></indexterm>
<indexterm role="cp"><primary>shapes, of fonts</primary></indexterm>
<anchor id="low-level-font-commands-fontshape"/><para>Select font shape. Valid shapes are:
</para>
<informaltable><tgroup cols="2"><colspec colwidth="2*"></colspec><colspec colwidth="19*"></colspec><tbody><row><entry><para><literal>n</literal>
</para></entry><entry><para>Upright (normal)
</para></entry></row><row><entry><para><literal>it</literal>
</para></entry><entry><para>Italic
</para></entry></row><row><entry><para><literal>sl</literal>
</para></entry><entry><para>Slanted (oblique)
</para></entry></row><row><entry><para><literal>sc</literal>
</para></entry><entry><para>Small caps
</para></entry></row><row><entry><para><literal>ui</literal>
</para></entry><entry><para>Upright italics
</para></entry></row><row><entry><para><literal>ol</literal>
</para></entry><entry><para>Outline
</para></entry></row></tbody></tgroup></informaltable>
<para>The two last shapes are not available for most font families, and
small caps are often missing as well.
</para>
<anchor id="low-level-font-commands-fontsize"/></listitem></varlistentry><varlistentry><term><literal>\fontsize{<replaceable>size</replaceable>}{<replaceable>skip</replaceable>}</literal>
</term><listitem><indexterm role="fn"><primary>\fontsize</primary></indexterm>
<indexterm role="cp"><primary>font size</primary></indexterm>
<indexterm role="fn"><primary>\baselineskip</primary></indexterm>
<para>Set the font size and the line spacing.  The unit of both parameters
defaults to points (<literal>pt</literal>).  The line spacing is the nominal
vertical space between lines, baseline to baseline.  It is stored in the
parameter <literal>\baselineskip</literal>.  The default <literal>\baselineskip</literal> for
the Computer Modern typeface is 1.2 times the <literal>\fontsize</literal>.
Changing <literal>\baselineskip</literal> directly is inadvisable since its value is
reset every time a size change happens; see <literal>\baselinestretch</literal>, next.
</para>
<anchor id="low-level-font-commands-baselinestretch"/></listitem></varlistentry><varlistentry><term><literal>\baselinestretch</literal>
</term><listitem><indexterm role="fn"><primary>\baselinestretch</primary></indexterm>
<para>&latex; multiplies the line spacing by the value of the
<literal>\baselinestretch</literal> parameter; the default factor is 1.  A change
takes effect when <literal>\selectfont</literal> (see below) is called.  You can
make a line skip change happen for the entire document, for instance
doubling it, by doing <literal>\renewcommand{\baselinestretch}{2.0}</literal> in
the preamble.
</para>
<indexterm role="cp"><primary>package, <literal>setspace</literal></primary></indexterm>
<indexterm role="cp"><primary><literal>setspace</literal> package</primary></indexterm>

<indexterm role="cp"><primary>double spacing</primary></indexterm>
<para>However, the best way to double-space a document is to use the
<filename>setspace</filename> package.  In addition to offering a number of spacing
options, this package keeps the line spacing single-spaced in places
where that is typically desirable, such as footnotes and figure
captions.  See the package documentation.
</para>
<anchor id="low-level-font-commands-linespread"/></listitem></varlistentry><varlistentry><term><literal>\linespread{<replaceable>factor</replaceable>}</literal>
</term><listitem><indexterm role="fn"><primary>\linespread</primary></indexterm>
<para>Equivalent to
<literal>\renewcommand{\baselinestretch}{<replaceable>factor</replaceable>}</literal>, and
therefore must be followed by <literal>\selectfont</literal> to have any effect.
Best specified in the preamble, or use the <literal>setspace</literal> package, as
just described.
</para>
<anchor id="low-level-font-commands-selectfont"/></listitem></varlistentry><varlistentry><term><literal>\selectfont</literal>
</term><listitem><indexterm role="fn"><primary>\selectfont</primary></indexterm>
<para>The effects of the font commands described above do not happen until
<literal>\selectfont</literal> is called, as in
<literal>\fontfamily{<replaceable>familyname</replaceable>}\selectfont</literal>.  It is often useful
to put this in a macro:

<literal>\newcommand*{\myfont}{\fontfamily{<replaceable>familyname</replaceable>}\selectfont}</literal>

(see <link linkend="_005cnewcommand-_0026-_005crenewcommand">\newcommand &amp; \renewcommand</link>).
</para>
<anchor id="low-level-font-commands-usefont"/></listitem></varlistentry><varlistentry><term><literal>\usefont{<replaceable>enc</replaceable>}{<replaceable>family</replaceable>}{<replaceable>series</replaceable>}{<replaceable>shape</replaceable>}</literal>
</term><listitem><indexterm role="fn"><primary>\usefont</primary></indexterm>
<para>The same as invoking <literal>\fontencoding</literal>, <literal>\fontfamily</literal>,
<literal>\fontseries</literal> and <literal>\fontshape</literal> with the given parameters,
followed by <literal>\selectfont</literal>.  For example:
</para>
<screen>\usefont{ot1}{cmr}{m}{n}
</screen>
</listitem></varlistentry></variablelist>

</sect1>
</chapter>
<chapter label="5" id="Layout">
<title>Layout</title>

<indexterm role="cp"><primary>layout commands</primary></indexterm>

<para>Commands for controlling the general page layout.
</para>


<sect1 label="5.1" id="_005conecolumn">
<title><literal>\onecolumn</literal></title>

<indexterm role="fn"><primary>\onecolumn</primary></indexterm>
<indexterm role="cp"><primary>one-column output</primary></indexterm>

<para>Synopsis:
</para>
<screen>\onecolumn
</screen>
<para>Start a new page and produce single-column output.  If the document is
given the class option <literal>onecolumn</literal> then this is the default
behavior (see <link linkend="Document-class-options">Document class options</link>).  This command is fragile
(see <link linkend="_005cprotect">\protect</link>).
</para>

</sect1>
<sect1 label="5.2" id="_005ctwocolumn">
<title><literal>\twocolumn</literal></title>

<indexterm role="fn"><primary>\twocolumn</primary></indexterm>
<indexterm role="cp"><primary>multicolumn text</primary></indexterm>
<indexterm role="cp"><primary>two-column output</primary></indexterm>

<para>Synopses:
</para>
<screen>\twocolumn
\twocolumn[<replaceable>prelim one column text</replaceable>]
</screen>
<para>Start a new page and produce two-column output. If the document is given
the class option <literal>twocolumn</literal> then this is the default
(see <link linkend="Document-class-options">Document class options</link>).  This command is fragile
(see <link linkend="_005cprotect">\protect</link>).
</para>
<para>If the optional <replaceable>prelim one column text</replaceable> argument
is present, it is typeset in one-column mode before the two-column
typesetting starts.
</para>
<para>These parameters control typesetting in two-column output:
</para>
<variablelist><varlistentry><term><indexterm role="fn"><primary>\columnsep</primary></indexterm><literal>\columnsep</literal>
</term><listitem><anchor id="twocolumn-columnsep"/><para>The distance between columns. The default is 35pt.  Change it with a
command such as <literal>\setlength{\columnsep}{40pt}</literal> You must change
it before the two column environment starts; in the preamble is a good
place.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\columnseprule</primary></indexterm><literal>\columnseprule</literal>
</term><listitem><anchor id="twocolumn-columnseprule"/><para>The width of the rule between columns. The rule appears halfway between
the two columns.  The default is 0pt, meaning that there is no rule.
Change it with a command such as
<literal>\setlength{\columnseprule}{0.4pt}</literal>, before the two-column
environment starts.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\columnwidth</primary></indexterm><literal>\columnwidth</literal>
</term><listitem><anchor id="twocolumn-columnwidth"/><para>The width of a single column.  In one-column mode this is equal to
<literal>\textwidth</literal>.  In two-column mode by default &latex; sets the
width of each of the two columns to be half of <literal>\textwidth</literal> minus
<literal>\columnsep</literal>.
</para>
</listitem></varlistentry></variablelist>
<para>In a two-column document, the starred environments <literal>table*</literal> and
<literal>figure*</literal> are two columns wide, whereas the unstarred environments
<literal>table</literal> and <literal>figure</literal> take up only one column (see <link linkend="figure">figure</link>
and see <link linkend="table">table</link>). &latex; places starred floats at the top of a page.
The following parameters control float behavior of two-column output.
</para>
<variablelist><anchor id="twocolumn-dbltopfraction"/><varlistentry><term><indexterm role="fn"><primary>\dbltopfraction</primary></indexterm><literal>\dbltopfraction</literal>
</term><listitem><para>The maximum fraction at the top of a two-column page that may be
occupied by two-column wide floats.  The default is 0.7, meaning that
the height of a <literal>table*</literal> or <literal>figure*</literal> environment must not
exceed <literal>0.7\textheight</literal> .  If the height of your starred float
environment exceeds this then you can take one of the following actions
to prevent it from floating all the way to the back of the document:
</para>
<itemizedlist><listitem><para>Use the <literal>[tp]</literal> location specifier to tell LaTeX to try to put
the bulky float on a page by itself, as well as at the top of a page.
</para>
</listitem><listitem><para>Use the <literal>[t!]</literal> location specifier to override the effect of
<literal>\dbltopfraction</literal> for this particular float.
</para>
</listitem><listitem><para>Increase the value of <literal>\dbltopfraction</literal> to a suitably large number,
to avoid going to float pages so soon.
</para></listitem></itemizedlist>
<para>You can redefine it, as with
<literal>\renewcommand{\dbltopfraction}{0.9}</literal>.
</para>
<anchor id="twocolumn-dblfloatpagefraction"/></listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\dblfloatpagefraction</primary></indexterm><literal>\dblfloatpagefraction</literal>
</term><listitem><para>For a float page of two-column wide floats, this is the minimum fraction
that must be occupied by floats, limiting the amount of blank space.
&latex;&#8217;s default is <literal>0.5</literal>.  Change it with <literal>\renewcommand</literal>.
</para>
<anchor id="twocolumn-dblfloatsep"/></listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\dblfloatsep</primary></indexterm><literal>\dblfloatsep</literal>
</term><listitem><para>On a float page of two-column wide floats, this length is the distance
between floats, at both the top and bottom of the page.  The default is
<literal>12pt plus2pt minus2pt</literal> for a document set at <literal>10pt</literal> or
<literal>11pt</literal>, and <literal>14pt plus2pt minus4pt</literal> for a document set at
<literal>12pt</literal>.
</para>
<anchor id="twocolumn-dbltextfloatsep"/></listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\dbltextfloatsep</primary></indexterm><literal>\dbltextfloatsep</literal>
</term><listitem><para>This length is the distance between a multi-column float at the top or
bottom of a page and the main text.  The default is <literal>20pt plus2pt
minus4pt</literal>.
</para>
<anchor id="twocolumn-dbltopnumber"/></listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\dbltopnumber</primary></indexterm><literal>\dbltopnumber</literal>
</term><listitem><para>On a float page of two-column wide floats, this counter gives the
maximum number of floats allowed at the top of the page.  The &latex;
default is <literal>2</literal>.
</para>
</listitem></varlistentry></variablelist>
<!-- From egreg at http://tex.stackexchange.com/a/142232/339 -->
<para>This example uses <literal>\twocolumn</literal>&#8217;s optional argument of to create a
title that spans the two-column article:
</para>
<screen>\documentclass[twocolumn]{article}
\newcommand{\authormark}[1]{\textsuperscript{#1}}
\begin{document}
\twocolumn[{% inside this optional argument goes one-column text
  \centering
  \LARGE The Title \\[1.5em]
  \large Author One\authormark{1},
         Author Two\authormark{2},
         Author Three\authormark{1} \\[1em]
  \normalsize
  \begin{tabular}{p{.2\textwidth}@{\hspace{2em}}p{.2\textwidth}}
    \authormark{1}Department one  &amp;\authormark{2}Department two \\ 
     School one                   &amp;School two 
  \end{tabular}\\[3em] % space below title part
  }]

Two column text here.
</screen>

</sect1>
<sect1 label="5.3" id="_005cflushbottom">
<title><literal>\flushbottom</literal></title>

<indexterm role="fn"><primary>\flushbottom</primary></indexterm>

<para>Make all pages in the documents after this declaration have the same
height, by stretching the vertical space where necessary to fill out the
page.  This is most often used when making two-sided documents since the
differences in facing pages can be glaring.
</para>
<para>If &tex; cannot satisfactorily stretch the vertical space in a page
then you get a message like &#8216;<literal>Underfull \vbox (badness 10000) has
occurred while \output is active</literal>&#8217;.  If you get that, one option is to
change to <literal>\raggedbottom</literal> (see <link linkend="_005craggedbottom">\raggedbottom</link>).  Alternatively,
you can adjust the <literal>textheight</literal> to make compatible pages, or you
can add some vertical stretch glue between lines or between paragraphs,
as in <literal>\setlength{\parskip}{0ex plus0.1ex}</literal>.  Your last option
is to, in a final editing stage, adjust the height of individual pages
(see <link linkend="_005cenlargethispage">\enlargethispage</link>).
</para>
<para>The <literal>\flushbottom</literal> state is the default only if you select the
<literal>twoside</literal> document class option (see <link linkend="Document-class-options">Document class options</link>).
</para>

</sect1>
<sect1 label="5.4" id="_005craggedbottom">
<title><literal>\raggedbottom</literal></title>

<indexterm role="fn"><primary>\raggedbottom</primary></indexterm>
<indexterm role="cp"><primary>stretch, omitting vertical</primary></indexterm>

<para>Make all later pages the natural height of the material on that page; no
rubber vertical lengths will be stretched.  Thus, in a two-sided
document the facing pages may be different heights.  This command can go
at any point in the document body.  See <link linkend="_005cflushbottom">\flushbottom</link>.
</para>
<para>This is the default unless you select the <literal>twoside</literal> document class
option (see <link linkend="Document-class-options">Document class options</link>).
</para>

</sect1>
<sect1 label="5.5" id="Page-layout-parameters">
<title>Page layout parameters</title>

<indexterm role="cp"><primary>page layout parameters</primary></indexterm>
<indexterm role="cp"><primary>parameters, page layout</primary></indexterm>
<indexterm role="cp"><primary>layout, page parameters for</primary></indexterm>
<indexterm role="cp"><primary>header, parameters for</primary></indexterm>
<indexterm role="cp"><primary>footer, parameters for</primary></indexterm>
<indexterm role="cp"><primary>running header and footer</primary></indexterm>

<variablelist><varlistentry><term><indexterm role="fn"><primary>\columnsep</primary></indexterm><literal>\columnsep</literal>
</term><term><indexterm role="fn"><primary>\columnseprule</primary></indexterm><literal>\columnseprule</literal>
</term><term><indexterm role="fn"><primary>\columnwidth</primary></indexterm><literal>\columnwidth</literal>
</term><listitem><indexterm role="fn"><primary>\columnsep</primary></indexterm>
<indexterm role="fn"><primary>\columnseprule</primary></indexterm>
<indexterm role="fn"><primary>\columnwidth</primary></indexterm>
<anchor id="page-layout-parameters-columnsep"/><anchor id="page-layout-parameters-columnseprule"/><anchor id="page-layout-parameters-columnwidth"/><para>The distance between the two columns, the width of a rule between the
columns, and the width of the columns, when the document class option
<literal>twocolumn</literal> is in effect (see <link linkend="Document-class-options">Document class options</link>).
See <link linkend="_005ctwocolumn">\twocolumn</link>.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\headheight</primary></indexterm><literal>\headheight</literal>
</term><listitem><indexterm role="fn"><primary>\headheight</primary></indexterm>
<anchor id="page-layout-parameters-headheight"/><para>Height of the box that contains the running head.  The default in the
<literal>article</literal>, <literal>report</literal>, and <literal>book</literal> classes is &#8216;<literal>12pt</literal>&#8217;,
at all type sizes.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\headsep</primary></indexterm><literal>\headsep</literal>
</term><listitem><indexterm role="fn"><primary>\headsep</primary></indexterm>
<anchor id="page-layout-parameters-headsep"/><para>Vertical distance between the bottom of the header line and the top of
the main text.  The default in the <literal>article</literal> and <literal>report</literal>
classes is &#8216;<literal>25pt</literal>&#8217;.  In the <literal>book</literal> class the default is: if the
document is set at 10pt then it is &#8216;<literal>0.25in</literal>&#8217;, and at 11pt and 12pt
it is &#8216;<literal>0.275in</literal>&#8217;.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\footskip</primary></indexterm><literal>\footskip</literal>
</term><listitem><indexterm role="fn"><primary>\footskip</primary></indexterm>
<anchor id="page-layout-parameters-footskip"/><para>Distance from the baseline of the last line of text to the baseline of
the page footer.  The default in the <literal>article</literal> and <literal>report</literal>
classes is &#8216;<literal>30pt</literal>&#8217;.  In the <literal>book</literal> class the default is: when
the type size is 10pt the default is &#8216;<literal>0.35in</literal>&#8217;, while at 11pt it is
&#8216;<literal>0.38in</literal>&#8217;, and at 12pt it is &#8216;<literal>30pt</literal>&#8217;.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\linewidth</primary></indexterm><literal>\linewidth</literal>
</term><listitem><indexterm role="fn"><primary>\linewidth</primary></indexterm>
<anchor id="page-layout-parameters-linewidth"/><para>Width of the current line, decreased for each nested <literal>list</literal>
(see <link linkend="list">list</link>).  That is, the nominal value for <literal>\linewidth</literal> is to
equal <literal>\textwidth</literal> but for each nested list the <literal>\linewidth</literal>
is decreased by the sum of that list&#8217;s <literal>\leftmargin</literal> and
<literal>\rightmargin</literal> (see <link linkend="itemize">itemize</link>).
<!-- The default varies with the font size, paper width, two-column mode, -->
<!-- etc.  For an @code{article} document set in 10pt, the default is -->
<!-- @samp{345pt}, while in two-column mode that becomes @samp{229.5pt}. -->
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\marginparpush</primary></indexterm><literal>\marginparpush</literal>
</term><term><indexterm role="fn"><primary>\marginsep</primary></indexterm><literal>\marginsep</literal>
</term><term><indexterm role="fn"><primary>\marginparwidth</primary></indexterm><literal>\marginparwidth</literal>
</term><listitem><indexterm role="fn"><primary>\marginparpush</primary></indexterm>
<indexterm role="fn"><primary>\marginsep</primary></indexterm>
<indexterm role="fn"><primary>\marginparwidth</primary></indexterm>
<anchor id="page-layout-parameters-marginparpush"/><anchor id="page-layout-parameters-marginsep"/><anchor id="page-layout-parameters-marginparwidth"/><para>The minimum vertical space between two marginal notes, the horizontal
space between the text body and the marginal notes, and the horizontal
width of the notes.
</para>
<para>Normally marginal notes appear on the outside of the page, but the
declaration <literal>\reversemarginpar</literal> changes that (and
<literal>\normalmarginpar</literal> changes it back).
</para>
<para>The defaults for <literal>\marginparpush</literal> in both <literal>book</literal> and
<literal>article</literal> classes are: &#8216;<literal>7pt</literal>&#8217; if the document is set at 12pt,
and &#8216;<literal>5pt</literal>&#8217; if the document is set at 11pt or 10pt.
</para>
<para>For <literal>\marginsep</literal>, in <literal>article</literal> class the default is
&#8216;<literal>10pt</literal>&#8217; except if the document is set at 10pt and in two-column mode
where the default is &#8216;<literal>11pt</literal>&#8217;.
</para>
<para>For <literal>\marginsep</literal> in <literal>book</literal> class the default is &#8216;<literal>10pt</literal>&#8217; in
two-column mode and &#8216;<literal>7pt</literal>&#8217; in one-column mode.  
</para>
<para>For <literal>\marginparwidth</literal> in both <literal>book</literal> and <literal>article</literal>
classes, in two-column mode the default is 60% of <literal>\paperwidth
&#8722; \textwidth</literal>, while in one-column mode it is 50% of that
distance.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\oddsidemargin</primary></indexterm><literal>\oddsidemargin</literal>
</term><term><indexterm role="fn"><primary>\evensidemargin</primary></indexterm><literal>\evensidemargin</literal>
</term><listitem><indexterm role="fn"><primary>\oddsidemargin</primary></indexterm>
<indexterm role="fn"><primary>\evensidemargin</primary></indexterm>
<anchor id="page-layout-parameters-oddsidemargin"/><anchor id="page-layout-parameters-evensidemargin"/><para>The <literal>\oddsidemargin</literal> is the extra distance between the left side of
the page and the text&#8217;s left margin, on odd-numbered pages when the
document class option <literal>twoside</literal> is chosen and on all pages when
<literal>oneside</literal> is in effect.  When <literal>twoside</literal> is in effect, on
even-numbered pages the extra distance on the left is
<literal>evensidemargin</literal>.
</para>
<para>&latex;&#8217;s default is that <literal>\oddsidemargin</literal> is 40% of the
difference between <literal>\paperwidth</literal> and <literal>\textwidth</literal>, and
<literal>\evensidemargin</literal> is the remainder.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\paperheight</primary></indexterm><literal>\paperheight</literal>
</term><listitem><indexterm role="fn"><primary>\paperheight</primary></indexterm>
<anchor id="page-layout-parameters-paperheight"/><para>The height of the paper, as distinct from the height of the print area.
Normally set with a document class option, as in
<literal>\documentclass[a4paper]{article}</literal> (see <link linkend="Document-class-options">Document class
options</link>).
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\paperwidth</primary></indexterm><literal>\paperwidth</literal>
</term><listitem><indexterm role="fn"><primary>\paperwidth</primary></indexterm>
<anchor id="page-layout-parameters-paperwidth"/><para>The width of the paper, as distinct from the width of the print area.
Normally set with a document class option, as in
<literal>\documentclass[a4paper]{article}</literal> (see <link linkend="Document-class-options">Document class
options</link>).
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\textheight</primary></indexterm><literal>\textheight</literal>
</term><listitem><indexterm role="fn"><primary>\textheight</primary></indexterm>
<anchor id="page-layout-parameters-textheight"/><para>The normal vertical height of the page body.  If the document is set at
a nominal type size of 10pt then for an <literal>article</literal> or <literal>report</literal>
the default is &#8216;<literal>43\baselineskip</literal>&#8217;, while for a <literal>book</literal> it is
&#8216;<literal>41\baselineskip</literal>&#8217;.  At a type size of 11pt the default is
&#8216;<literal>38\baselineskip</literal>&#8217; for all document classes.  At 12pt it is
&#8216;<literal>36\baselineskip</literal>&#8217; for all classes.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\textwidth</primary></indexterm><literal>\textwidth</literal>
</term><listitem><indexterm role="fn"><primary>\textwidth</primary></indexterm>
<anchor id="page-layout-parameters-textwidth"/><para>The full horizontal width of the entire page body.  For an
<literal>article</literal> or <literal>report</literal> document, the default is &#8216;<literal>345pt</literal>&#8217;
when the chosen type size is 10pt, the default is &#8216;<literal>360pt</literal>&#8217; at 11pt,
and it is &#8216;<literal>390pt</literal>&#8217; at 12pt.  For a <literal>book</literal> document, the default
is &#8216;<literal>4.5in</literal>&#8217; at a type size of 10pt, and &#8216;<literal>5in</literal>&#8217; at 11pt or 12pt.
</para>
<para>In multi-column output, <literal>\textwidth</literal> remains the width of the
entire page body, while <literal>\columnwidth</literal> is the width of one column
(see <link linkend="_005ctwocolumn">\twocolumn</link>).
</para>
<para>In lists (see <link linkend="list">list</link>), <literal>\textwidth</literal> remains the width of the
entire page body (and <literal>\columnwidth</literal> the width of the entire
column), while <literal>\linewidth</literal> may decrease for nested lists.
</para>
<para>Inside a minipage (see <link linkend="minipage">minipage</link>) or <literal>\parbox</literal>
(see <link linkend="_005cparbox">\parbox</link>), all the width-related parameters are set to the
specified width, and revert to their normal values at the end of the
<literal>minipage</literal> or <literal>\parbox</literal>.
</para>
<indexterm role="fn"><primary>\hsize</primary></indexterm>
<anchor id="page-layout-parameters-hsize"/></listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\hsize</primary></indexterm><literal>\hsize</literal>
</term><listitem><para>This entry is included for completeness: <literal>\hsize</literal> is the &tex;
primitive parameter used when text is broken into lines.  It should not
be used in normal &latex; documents.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\topmargin</primary></indexterm><literal>\topmargin</literal>
</term><listitem><indexterm role="fn"><primary>topmargin</primary></indexterm>
<anchor id="page-layout-parameters-topmargin"/><para>Space between the top of the &tex; page (one inch from the top of the
paper, by default) and the top of the header.  The value is computed
based on many other parameters: <literal>\paperheight &#8722; 2in &#8722;
\headheight &#8722; \headsep &#8722; \textheight &#8722; \footskip</literal>,
and then divided by two.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\topskip</primary></indexterm><literal>\topskip</literal>
</term><listitem><indexterm role="fn"><primary>\topskip</primary></indexterm>
<anchor id="page-layout-parameters-topskip"/><para>Minimum distance between the top of the page body and the baseline of
the first line of text.  For the standard classes, the default is the
same as the font size, e.g., &#8216;<literal>10pt</literal>&#8217; at a type size of 10pt.
</para>
</listitem></varlistentry></variablelist>

</sect1>
<sect1 label="5.6" id="Floats">
<title>Floats</title>

<para>Some typographic elements, such as figures and tables, cannot be broken
across pages.  They must be typeset outside of the normal flow of text,
for instance floating to the top of a later page.
</para>
<para>&latex; can have a number of different classes of floating material.
The default is the two classes, <literal>figure</literal> (see <link linkend="figure">figure</link>) and
<literal>table</literal> (see <link linkend="table">table</link>), but you can create a new class with the
package <filename>float</filename>.
</para>
<para>Within any one float class &latex; always respects the order, so that
the first figure in a document source must be typeset before the second
figure.  However, &latex; may mix the classes, so it can happen that
while the first table appears in the source before the first figure, it
appears in the output after it.
</para>
<para>The placement of floats is subject to parameters, given below, that
limit the number of floats that can appear at the top of a page, and
the bottom, etc. If so many floats are queued that the limits prevent
them all from fitting on a page then &latex; places what it can and
defers the rest to the next page.  In this way, floats may end up
being typeset far from their place in the source.  In particular, a
float that is big may migrate to the end of the document. In which
event, because all floats in a class must appear in sequential order,
every following float in that class also appears at the end.
</para>
<indexterm role="cp"><primary>placement of floats</primary></indexterm>
<indexterm role="cp"><primary>specifier, float placement</primary></indexterm>
<para>In addition to changing the parameters, for each float you can tweak
where the float placement algorithm tries to place it by using its
<replaceable>placement</replaceable> argument.  The possible values are a sequence of the
letters below. The default for both <literal>figure</literal> and <literal>table</literal>, in
both <literal>article</literal> and <literal>book</literal> classes, is <literal>tbp</literal>.
</para>
<variablelist><varlistentry><term><literal>t</literal>
</term><listitem><para>(Top)&#8212;at the top of a text page.
</para>
</listitem></varlistentry><varlistentry><term><literal>b</literal>
</term><listitem><para>(Bottom)&#8212;at the bottom of a text page.  (However, <literal>b</literal> is not
allowed for full-width floats (<literal>figure*</literal>) with double-column
output.  To ameliorate this, use the <filename>stfloats</filename> or
<filename>dblfloatfix</filename> package, but see the discussion at caveats in the
FAQ: <ulink url="http://www.tex.ac.uk/cgi-bin/texfaq2html?label=2colfloat">http://www.tex.ac.uk/cgi-bin/texfaq2html?label=2colfloat</ulink>.
</para>
</listitem></varlistentry><varlistentry><term><literal>h</literal>
</term><listitem><para>(Here)&#8212;at the position in the text where the <literal>figure</literal> environment
appears.  However, <literal>h</literal> is not allowed by itself; <literal>t</literal> is
automatically added.
</para>
<indexterm role="cp"><primary>here, putting floats</primary></indexterm>
<indexterm role="cp"><primary>package, <literal>float</literal></primary></indexterm>
<indexterm role="cp"><primary><literal>float</literal> package</primary></indexterm>

<para>To absolutely force a float to appear &#8220;here&#8221;, you can
<literal>\usepackage{float}</literal> and use the <literal>H</literal> specifier which it
defines.  For further discussion, see the FAQ entry at
<ulink url="http://www.tex.ac.uk/cgi-bin/texfaq2html?label=figurehere">http://www.tex.ac.uk/cgi-bin/texfaq2html?label=figurehere</ulink>.
</para>
</listitem></varlistentry><varlistentry><term><literal>p</literal>
</term><listitem><indexterm role="cp"><primary>float page</primary></indexterm>
<para>(Page of floats)&#8212;on a separate <firstterm>float page</firstterm>, which is a page
containing no text, only floats.
</para>
</listitem></varlistentry><varlistentry><term><literal>!</literal>
</term><listitem><para>Used in addition to one of the above; for this float only, &latex;
ignores the restrictions on both the number of floats that can appear
and the relative amounts of float and non-float text on the page.
The <literal>!</literal> specifier does <emphasis>not</emphasis> mean &#8220;put the float here&#8221;;
see above.
</para>
</listitem></varlistentry></variablelist>
<para>Note: the order in which letters appear in the <replaceable>placement</replaceable> argument
does not change the order in which &latex; tries to place the float;
for instance, <literal>btp</literal> has the same effect as <literal>tbp</literal>.  All that
<replaceable>placement</replaceable> does is that if a letter is not present then the
algorithm does not try that location.  Thus, &latex;&#8217;s default of
<literal>tbp</literal> is to try every location except placing the float where it
occurs in the source.
</para>
<para>To prevent &latex; from moving floats to the end of the document or a
chapter you can use a <literal>\clearpage</literal> command to start a new page and
insert all pending floats. If a pagebreak is undesirable then you can
use the <filename>afterpage</filename> package and issue
<literal>\afterpage{\clearpage}</literal>.  This will wait until the current page
is finished and then flush all outstanding floats.
</para>
<indexterm role="cp"><primary>package, <literal>flafter</literal></primary></indexterm>
<indexterm role="cp"><primary><literal>flafter</literal> package</primary></indexterm>

<para>&latex; can typeset a float before where it appears in the source
(although on the same output page) if there is a <literal>t</literal> specifier in
the <replaceable>placement</replaceable> parameter.  If this is not desired, and deleting
the <literal>t</literal> is not acceptable as it keeps the float from being placed
at the top of the next page, then you can prevent it by either using
the <filename>flafter</filename> package or using the command
<indexterm role="fn"><primary>\suppressfloats</primary></indexterm>
<literal>\suppressfloats[t]</literal>, which causes floats for the top position on
this page to moved to the next page.
</para>
<para>Parameters relating to fractions of pages occupied by float and
non-float text (change them with
<literal>\renewcommand{<replaceable>parameter</replaceable>}{<replaceable>decimal between 0 and 1</replaceable>}</literal>):
</para>
<variablelist><varlistentry><term><indexterm role="fn"><primary>\bottomfraction</primary></indexterm><literal>\bottomfraction</literal>
</term><listitem><indexterm role="fn"><primary>\bottomfraction</primary></indexterm>
<anchor id="floats-bottomfraction"/><para>The maximum fraction of the page allowed to be occupied by floats at
the bottom; default &#8216;<literal>.3</literal>&#8217;.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\floatpagefraction</primary></indexterm><literal>\floatpagefraction</literal>
</term><listitem><indexterm role="fn"><primary>\floatpagefraction</primary></indexterm>
<anchor id="floats-floatpagefraction"/><para>The minimum fraction of a float page that must be occupied by floats;
default &#8216;<literal>.5</literal>&#8217;.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\textfraction</primary></indexterm><literal>\textfraction</literal>
</term><listitem><indexterm role="fn"><primary>\textfraction</primary></indexterm>
<anchor id="floats-textfraction"/><para>Minimum fraction of a page that must be text; if floats take up too
much space to preserve this much text, floats will be moved to a
different page.  The default is &#8216;<literal>.2</literal>&#8217;.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\topfraction</primary></indexterm><literal>\topfraction</literal>
</term><listitem><indexterm role="fn"><primary>\topfraction</primary></indexterm>
<anchor id="floats-topfraction"/><para>Maximum fraction at the top of a page that may be occupied before
floats; default &#8216;<literal>.7</literal>&#8217;.
</para></listitem></varlistentry></variablelist>
<para>Parameters relating to vertical space around floats (change them with a
command of the form <literal>\setlength{<replaceable>parameter</replaceable>}{<replaceable>length
expression</replaceable>}</literal>):
</para>
<variablelist><varlistentry><term><indexterm role="fn"><primary>\floatsep</primary></indexterm><literal>\floatsep</literal>
</term><listitem><indexterm role="fn"><primary>\floatsep</primary></indexterm>
<anchor id="floats-floatsep"/><para>Space between floats at the top or bottom of a page; default
&#8216;<literal>12pt plus2pt minus2pt</literal>&#8217;.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\intextsep</primary></indexterm><literal>\intextsep</literal>
</term><listitem><indexterm role="fn"><primary>\intextsep</primary></indexterm>
<anchor id="floats-intextsep"/><para>Space above and below a float in the middle of the main text; default
&#8216;<literal>12pt plus2pt minus2pt</literal>&#8217; for 10 point and 11 point documents,
and &#8216;<literal>14pt plus4pt minus4pt</literal>&#8217; for 12 point documents.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\textfloatsep</primary></indexterm><literal>\textfloatsep</literal>
</term><listitem><indexterm role="fn"><primary>\textfloatsep</primary></indexterm>
<anchor id="floats-textfloatsep"/><para>Space between the last (first) float at the top (bottom) of a page;
default &#8216;<literal>20pt plus2pt minus4pt</literal>&#8217;.
</para></listitem></varlistentry></variablelist>
<para>Counters relating to the number of floats on a page (change them with a
command of the form <literal>\setcounter{<replaceable>ctrname</replaceable>}{<replaceable>natural
number</replaceable>}</literal>):
</para>
<variablelist><varlistentry><term><indexterm role="fn"><primary>bottomnumber</primary></indexterm><literal>bottomnumber</literal>
</term><listitem><indexterm role="fn"><primary>bottomnumber</primary></indexterm>
<anchor id="floats-bottomnumber"/><para>Maximum number of floats that can appear at the bottom of a text page;
default 1.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>dbltopnumber</primary></indexterm><literal>dbltopnumber</literal>
</term><listitem><indexterm role="fn"><primary>dbltopnumber</primary></indexterm>
<anchor id="floats-dbltopnumber"/><para>Maximum number of full-sized floats that can appear at the top of a
two-column page; default 2.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>topnumber</primary></indexterm><literal>topnumber</literal>
</term><listitem><indexterm role="fn"><primary>topnumber</primary></indexterm>
<anchor id="floats-topnumber"/><para>Maximum number of floats that can appear at the top of a text page;
default 2.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>totalnumber</primary></indexterm><literal>totalnumber</literal>
</term><listitem><indexterm role="fn"><primary>totalnumber</primary></indexterm>
<anchor id="floats-totalnumber"/><para>Maximum number of floats that can appear on a text page; default 3.
</para></listitem></varlistentry></variablelist>
<para>The principal &tex;&#160;FAQ entry relating to floats
<ulink url="http://www.tex.ac.uk/cgi-bin/texfaq2html?label=floats">http://www.tex.ac.uk/cgi-bin/texfaq2html?label=floats</ulink> contains
suggestions for relaxing &latex;&#8217;s default parameters to reduce the
problem of floats being pushed to the end.  A full explanation of the
float placement algorithm is in Frank Mittelbach&#8217;s article &#8220;How to
influence the position of float environments like figure and table in
&latex;?&#8221;
(<ulink url="http://latex-project.org/papers/tb111mitt-float.pdf">http://latex-project.org/papers/tb111mitt-float.pdf</ulink>).
</para>

</sect1>
</chapter>
<chapter label="6" id="Sectioning">
<title>Sectioning</title>

<indexterm role="cp"><primary>sectioning commands</primary></indexterm>
<indexterm role="cp"><primary>part</primary></indexterm>
<indexterm role="cp"><primary>chapter</primary></indexterm>
<indexterm role="cp"><primary>section</primary></indexterm>
<indexterm role="cp"><primary>subsection</primary></indexterm>
<indexterm role="cp"><primary>paragraph</primary></indexterm>
<indexterm role="cp"><primary>subparagraph</primary></indexterm>
<indexterm role="fn"><primary>\part</primary></indexterm>
<indexterm role="fn"><primary>\chapter</primary></indexterm>
<indexterm role="fn"><primary>\section</primary></indexterm>
<indexterm role="fn"><primary>\subsection</primary></indexterm>
<indexterm role="fn"><primary>\paragraph</primary></indexterm>
<indexterm role="fn"><primary>\subparagraph</primary></indexterm>

<para>Structure your text into divisions: parts, chapters, sections, etc.  All
sectioning commands have the same form, one of:
</para>
<screen><replaceable>sectioning-command</replaceable>{<replaceable>title</replaceable>}
<replaceable>sectioning-command</replaceable>*{<replaceable>title</replaceable>}
<replaceable>sectioning-command</replaceable>[<replaceable>toc-title</replaceable>]{<replaceable>title</replaceable>}
</screen>
<para>For instance, declare the start of a subsection as with
<literal>\subsection{Motivation}</literal>.
</para>
<para>The table has each <replaceable>sectioning-command</replaceable> in &latex;.  All are
available in all of &latex;&#8217;s standard document classes <literal>book</literal>,
<literal>report</literal>, and&#160;<literal>article</literal>, except that <literal>\chapter</literal> is
not available in <literal>article</literal>.
</para>
<informaltable><tgroup cols="3"><colspec colwidth="25*"></colspec><colspec colwidth="25*"></colspec><colspec colwidth="40*"></colspec><thead><row><entry><para>Sectioning unit  </para></entry><entry><para>Command  </para></entry><entry><para>Level 
</para></entry></row></thead><tbody><row><entry><para>Part
</para></entry><entry><para><literal>\part</literal>  </para></entry><entry><para>-1 (<literal>book</literal>, <literal>report</literal>), 0 (<literal>article</literal>)
</para></entry></row><row><entry><para>Chapter
</para></entry><entry><para><literal>\chapter</literal>   </para></entry><entry><para>0
</para></entry></row><row><entry><para>Section
</para></entry><entry><para><literal>\section</literal>   </para></entry><entry><para>1 
</para></entry></row><row><entry><para>Subsection
</para></entry><entry><para><literal>\subsection</literal>   </para></entry><entry><para>2 
</para></entry></row><row><entry><para>Subsubsection
</para></entry><entry><para><literal>\subsubsection</literal>   </para></entry><entry><para>3  
</para></entry></row><row><entry><para>Paragraph
</para></entry><entry><para><literal>\paragraph</literal>   </para></entry><entry><para>4  
</para></entry></row><row><entry><para>Subparagraph
</para></entry><entry><para><literal>\subparagraph</literal>   </para></entry><entry><para>5  
</para></entry></row></tbody></tgroup></informaltable>
<indexterm role="cp"><primary><literal>*</literal>-form of sectioning commands</primary></indexterm>
<para>All these commands have a <literal>*</literal>-form that prints <replaceable>title</replaceable> as usual
but is not numbered and does not make an entry in the table of contents.
An example of using this is for an appendix in an <literal>article</literal> .  The
input <literal>\appendix\section{Appendix}</literal> gives the output &#8216;<literal>A
Appendix</literal>&#8217; (see <link linkend="_005cappendix">\appendix</link>).  You can lose the numbering&#160;&#8216;<literal>A</literal>&#8217;
by instead entering <literal>\section*{Appendix}</literal> (articles often omit a
table of contents and have simple page headers so the other differences
from the <literal>\section</literal> command may not matter).
</para>
<para>The section title <replaceable>title</replaceable> provides the heading in the main text, but
it may also appear in the table of contents and in the running head or
foot (see <link linkend="Page-styles">Page styles</link>).  You may not want the same text in these
places as in the main text.  All of these commands have an optional
argument <replaceable>toc-title</replaceable> for these other places.
</para>
<para>The level number in the table above determines which sectional units are
numbered, and which appear in the table of contents.  If the sectioning
command&#8217;s <replaceable>level</replaceable> is less than or equal to the value of the counter
<literal>secnumdepth</literal> then the titles for this sectioning command will be
numbered (see <link linkend="Sectioning_002fsecnumdepth">Sectioning/secnumdepth</link>).  And, if <replaceable>level</replaceable> is less
than or equal to the value of the counter <literal>tocdepth</literal> then the table
of contents will have an entry for this sectioning unit
(see <link linkend="Sectioning_002ftocdepth">Sectioning/tocdepth</link>).
</para>
<para>&latex; expects that before you have a <literal>\subsection</literal> you will have
a <literal>\section</literal> and, in a book, that before a <literal>\section</literal> you will
have a <literal>\chapter</literal>.  Otherwise you can get a something like a
subsection numbered &#8216;<literal>3.0.1</literal>&#8217;.
</para>
<para>Two counters relate to the appearance of sectioning commands.
</para>
<variablelist><varlistentry><term><indexterm role="fn"><primary>secnumdepth</primary></indexterm><literal>secnumdepth</literal>
</term><listitem><indexterm role="fn"><primary>secnumdepth counter</primary></indexterm>
<indexterm role="cp"><primary>section numbers, printing</primary></indexterm>
<anchor id="sectioning-secnumdepth"/><anchor id="Sectioning_002fsecnumdepth"/><para>Controls which sectioning commands are
numbered.  Suppress numbering of sectioning at any depth greater than
<replaceable>level</replaceable> <literal>\setcounter{secnumdepth}{<replaceable>level</replaceable>}</literal>
(see <link linkend="_005csetcounter">\setcounter</link>).  See the above table for the level numbers.  For
instance, if the <literal>secnumdepth</literal> is 1 in an <literal>article</literal> then a
<literal>\section{Introduction}</literal> command will produce output like &#8216;<literal>1
Introduction</literal>&#8217; while <literal>\subsection{Discussion}</literal> will produce output
like &#8216;<literal>Discussion</literal>&#8217;, without the number. &latex;&#8217;s default
<literal>secnumdepth</literal> is&#160;3 in <filename>article</filename> class and &#160;2 in the
<filename>book</filename> and <filename>report</filename> classes.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>tocdepth</primary></indexterm><literal>tocdepth</literal>
</term><listitem><indexterm role="fn"><primary>tocdepth counter</primary></indexterm>
<indexterm role="cp"><primary>table of contents, sectioning numbers printed</primary></indexterm>
<anchor id="sectioning-tocdepth"/><anchor id="Sectioning_002ftocdepth"/><para>Controls which sectioning units are listed in the table of contents.
The setting <literal>\setcounter{tocdepth}{<replaceable>level</replaceable>}</literal> makes the
sectioning units at <replaceable>level</replaceable> be the smallest ones listed
(see <link linkend="_005csetcounter">\setcounter</link>).  See the above table for the level numbers.  For
instance, if <literal>tocdepth</literal> is&#160;1 then the table of contents will
list sections but not subsections.  &latex;&#8217;s default
<literal>secnumdepth</literal> is&#160;3 in <filename>article</filename> class and &#160;2 in the
<filename>book</filename> and <filename>report</filename> classes.
</para>
</listitem></varlistentry></variablelist>


<sect1 label="6.1" id="_005cpart">
<title><literal>\part</literal></title>

<indexterm role="fn"><primary>\part</primary></indexterm>
<indexterm role="cp"><primary>part</primary></indexterm>
<indexterm role="cp"><primary>sectioning, part</primary></indexterm>

<para>Synopsis, one of:
</para>
<screen>\part{<replaceable>title</replaceable>}
\part*{<replaceable>title</replaceable>}
\part[<replaceable>toc-title</replaceable>]{<replaceable>title</replaceable>}
</screen>
<para>Start a document part.  The standard &latex; classes <literal>book</literal>,
<literal>report</literal>, and <literal>article</literal>, all have this command.
</para>
<para>This produces a document part, in a book.
</para>
<screen>\part{VOLUME I \\
       PERSONAL MEMOIRS OF  U.\ S.\ GRANT}
\chapter{ANCESTRY--BIRTH--BOYHOOD.}
My family is American, and has been for generations,
in all its branches, direct and collateral.
</screen>
<para>In each standard class the <literal>\part</literal> command outputs a part number
such as &#8216;<literal>Part I</literal>&#8217;, alone on its line, in boldface, and in large
type.  Then &latex; outputs <replaceable>title</replaceable>, also alone on its line, in
bold and in even larger type.  In class <literal>book</literal>, the &latex;
default puts each part alone on its own page. If the book is two-sided
then &latex; will skip a page if needed to have the new part on an
odd-numbered page.  In <literal>report</literal> it is again alone on a page, but
&latex; won&#8217;t force it onto an odd-numbered page.  In an <literal>article</literal>
&latex; does not put it on a fresh page, but instead outputs the part
number and part title onto the main document page.
</para>
<para>The <literal>*</literal>&#160;form shows <replaceable>title</replaceable> 
but it does not show the part number, does not increment the
<literal>part</literal> counter, and produces no table of contents entry.
</para>
<para>The optional argument <replaceable>toc-title</replaceable> will appear as the part title in
the table of contents (see <link linkend="Table-of-contents-etc_002e">Table of contents etc.</link>) and in running
headers (see <link linkend="Page-styles">Page styles</link>).  If it is not present then <replaceable>title</replaceable>
will be there.  This example puts a line break in <replaceable>title</replaceable> but leaves
out the break in the table of contents.
</para>
<screen>\part[Up from the bottom; my life]{Up from the bottom\\ my life}
</screen>
<para>For determining which sectional units are numbered and which appear in
the table of contents, the level number of a part is&#160;-1
(see <link linkend="Sectioning_002fsecnumdepth">Sectioning/secnumdepth</link> and see <link linkend="Sectioning_002ftocdepth">Sectioning/tocdepth</link>).
</para>
<indexterm role="cp"><primary>package, <literal>indentfirst</literal></primary></indexterm>
<indexterm role="cp"><primary><literal>indentfirst</literal> package</primary></indexterm>

<para>In the class <literal>article</literal>, if a paragraph immediately follows the part
title then it is not indented.  To get an indent you can use the package
<filename>indentfirst</filename>.
</para>
<indexterm role="cp"><primary>package, <literal>titlesec</literal></primary></indexterm>
<indexterm role="cp"><primary><literal>titlesec</literal> package</primary></indexterm>

<para>One package to change the behavior of <literal>\part</literal> is <filename>titlesec</filename>.
See its documentation on CTAN.
</para>

</sect1>
<sect1 label="6.2" id="_005cchapter">
<title><literal>\chapter</literal></title>

<indexterm role="fn"><primary>\chapter</primary></indexterm>
<indexterm role="cp"><primary>chapter</primary></indexterm>

<para>Synopsis, one of:
</para>
<screen>\chapter{<replaceable>title</replaceable>}
\chapter*{<replaceable>title</replaceable>}
\chapter[<replaceable>toc-title</replaceable>]{<replaceable>title</replaceable>}
</screen>
<para>Start a chapter.  The standard &latex; classes <literal>book</literal> and
<literal>report</literal> have this command but <literal>article</literal> does not.
</para>
<para>This produces a chapter.
</para>
<screen>\chapter{Loomings}
Call me Ishmael.
Some years ago---never mind how long precisely---having little or no
money in my purse, and nothing particular to interest me on shore, I
thought I would sail about a little and see the watery part of
the world.
</screen>
<para>The &latex; default starts each chapter on a fresh page, an
odd-numbered page if the document is two-sided.  It produces a chapter
number such as &#8216;<literal>Chapter 1</literal>&#8217; in large boldface type (the size is
<literal>\huge</literal>).  It then puts <replaceable>title</replaceable> on a fresh line, in boldface
type that is still larger (size <literal>\Huge</literal>).  It also increments the
<literal>chapter</literal> counter, adds an entry to the table of contents
(see <link linkend="Table-of-contents-etc_002e">Table of contents etc.</link>), and sets the running header
information (see <link linkend="Page-styles">Page styles</link>).
</para>
<para>The <literal>*</literal>&#160;form shows <replaceable>title</replaceable> on a fresh line, in boldface.
But it does not show the chapter number, does not increment the
<literal>chapter</literal> counter, produces no table of contents entry, and does
not affect the running header.  (If you use the page style
<literal>headings</literal> in a two-sided document then the header will be from the
prior chapter.)  This example illustrates.
</para>
<screen>\chapter*{Preamble}
</screen>
<para>The optional argument <replaceable>toc-title</replaceable> will appear as the chapter title
in the table of contents (see <link linkend="Table-of-contents-etc_002e">Table of contents etc.</link>) and in
running headers (see <link linkend="Page-styles">Page styles</link>).  If it is not present then
<replaceable>title</replaceable> will be there.  This shows the full name in the chapter
title,
</para>
<screen>\chapter[Weyl]{Hermann Klaus Hugo (Peter) Weyl (1885--1955)}
</screen>
<para>but only &#8216;<literal>Weyl</literal>&#8217; on the contents page.  This puts a line break in
the title but that doesn&#8217;t work well with running headers so it omits
the break in the contents
</para>
<screen>\chapter[Given it all\\ my story]{Given it all\\ my story}
</screen>
<para>For determining which sectional units are numbered and which appear in
the table of contents, the level number of a chapter is&#160;0
(see <link linkend="Sectioning_002fsecnumdepth">Sectioning/secnumdepth</link> and see <link linkend="Sectioning_002ftocdepth">Sectioning/tocdepth</link>).
</para>
<indexterm role="cp"><primary>package, <literal>indentfirst</literal></primary></indexterm>
<indexterm role="cp"><primary><literal>indentfirst</literal> package</primary></indexterm>

<para>The paragraph that follows the chapter title is not indented, as is a
standard typographical practice.  To get an indent use the package
<filename>indentfirst</filename>.
</para>
<para>You can change what is shown for the chapter number.  To change it to
something like &#8216;<literal>Lecture 1</literal>&#8217;, put in the preamble either
<literal>\renewcommand{\chaptername}{Lecture}</literal> or this
(see <link linkend="_005cmakeatletter-_0026-_005cmakeatother">\makeatletter &amp; \makeatother</link>).
</para>
<screen>\makeatletter
\renewcommand{\@chapapp}{Lecture}
\makeatother
</screen>
<indexterm role="cp"><primary>package, <literal>babel</literal></primary></indexterm>
<indexterm role="cp"><primary><literal>babel</literal> package</primary></indexterm>

<para>To make this change because of the primary language for
the document, see the package <filename>babel</filename>.
</para>
<para>In a two-sided document &latex; puts a chapter on odd-numbered page, if
necessary leaving an even-numbered page that is blank except for any
running headers.  To make that page completely blank,
see&#160;<link linkend="_005cclearpage-_0026-_005ccleardoublepage">\clearpage &amp; \cleardoublepage</link>.
</para>
<indexterm role="cp"><primary>package, <literal>titlesec</literal></primary></indexterm>
<indexterm role="cp"><primary><literal>titlesec</literal> package</primary></indexterm>

<para>To change the behavior of the <literal>\chapter</literal> command, you can copy its
definition from the &latex; format file and make adjustments.  But
there are also many packages on CTAN that address this.  One is
<filename>titlesec</filename>.  See its documentation, but the example below gives a
sense of what it can do.
</para>
<screen>\usepackage{titlesec}   % in preamble
\titleformat{\chapter}
  {\Huge\bfseries}  % format of title
  {}                % label, such as 1.2 for a subsection
  {0pt}             % length of separation between label and title
  {}                % before-code hook
</screen>
<para>This omits the chapter number &#8216;<literal>Chapter 1</literal>&#8217; from the page but unlike
<literal>\chapter*</literal> it keeps the chapter in the table of contents and the
running headers.
</para>

</sect1>
<sect1 label="6.3" id="_005csection">
<title><literal>\section</literal></title>

<indexterm role="fn"><primary>\section</primary></indexterm>
<indexterm role="cp"><primary>section</primary></indexterm>

<para>Synopsis, one of:
</para>
<screen>\section{<replaceable>title</replaceable>}
\section*{<replaceable>title</replaceable>}
\section[<replaceable>toc-title</replaceable>]{<replaceable>title</replaceable>}
</screen>
<para>Start a section.  The standard &latex; classes <literal>article</literal>,
<literal>book</literal>, and <literal>report</literal> all have this command.
</para>
<para>This produces a section.
</para>
<screen>In this Part we tend to be more interested in the function,
in the input-output behavior,
than in the details of implementing that behavior.

\section{Turing machines}
Despite this desire to downplay implementation,
we follow the approach of A~Turing that the
first step toward defining the set of computable functions
is to reflect on the details of what mechanisms can do.
</screen>
<para>For the standard &latex; classes <literal>book</literal> and <literal>report</literal> the
default output is like &#8216;<literal>1.2 <replaceable>title</replaceable></literal>&#8217; (for chapter&#160;1,
section&#160;2), alone on its line and flush left, in boldface and a
larger type (the type size is <literal>\Large</literal>).  The same holds in
<literal>article</literal> except that there are no chapters in that class so it
looks like &#8216;<literal>2 <replaceable>title</replaceable></literal>&#8217;.
</para>
<para>The <literal>*</literal>&#160;form shows <replaceable>title</replaceable>.
But it does not show the section number, does not increment the
<literal>section</literal> counter, produces no table of contents entry, and does
not affect the running header.  (If you use the page style
<literal>headings</literal> in a two-sided document then the header will be from the
prior section.)
</para>
<para>The optional argument <replaceable>toc-title</replaceable> will appear as the section title
in the table of contents (see <link linkend="Table-of-contents-etc_002e">Table of contents etc.</link>) and in
running headers (see <link linkend="Page-styles">Page styles</link>).  If it is not present then
<replaceable>title</replaceable> will be there.  This shows the full name in the title of the
section,
</para>
<screen>\section[Elizabeth~II]{Elizabeth the Second,
  by the Grace of God of the United Kingdom,
  Canada and Her other Realms and Territories Queen,
  Head of the Commonwealth, Defender of the Faith.}
</screen>
<para>but only &#8216;<literal>Elizabeth II</literal>&#8217; on the contents page and in the headers.
This has a line break in <replaceable>title</replaceable> but that does not work with headers
so it is omitted from the contents and headers.
</para>
<screen>\section[Truth is, I cheated; my life story]{Truth is,
  I cheated\\my life story}
</screen>
<para>For determining which sectional units are numbered and which appear in
the table of contents, the level number of a section is&#160;1
(see <link linkend="Sectioning_002fsecnumdepth">Sectioning/secnumdepth</link> and see <link linkend="Sectioning_002ftocdepth">Sectioning/tocdepth</link>).
</para>
<indexterm role="cp"><primary>package, <literal>indentfirst</literal></primary></indexterm>
<indexterm role="cp"><primary><literal>indentfirst</literal> package</primary></indexterm>

<para>The paragraph that follows the section title is not indented, as is a
standard typographical practice.  One way to get an indent is to use the
package <filename>indentfirst</filename>.
</para>
<indexterm role="cp"><primary>package, <literal>titlesec</literal></primary></indexterm>
<indexterm role="cp"><primary><literal>titlesec</literal> package</primary></indexterm>

<para>In general, to change the behavior of the <literal>\section</literal> command, there
are a number of options.  One is the <literal>\@startsection</literal> command
(see <link linkend="_005c_0040startsection">\@startsection</link>).  There are also many packages on CTAN that
address this, including <filename>titlesec</filename>.  See the documentation but the
example below gives a sense of what they can do.
</para>
<!-- credit: egreg https://groups.google.com/forum/#!topic/comp.text.tex/tvc8oM5P4y4 -->
<screen>\usepackage{titlesec}   % in preamble
\titleformat{\section}
  {\normalfont\Large\bfseries}  % format of title
  {\makebox[1pc][r]{\thesection\hspace{1pc}}} % label
  {0pt}                   % length of separation between label and title
  {}                      % before-code hook
\titlespacing*{\section}
  {-1pc}{18pt}{10pt}[10pc]
</screen>
<para>That puts the section number in the margin.
</para>

</sect1>
<sect1 label="6.4" id="_005csubsection">
<title><literal>\subsection</literal></title>

<indexterm role="fn"><primary>\subsection</primary></indexterm>
<indexterm role="cp"><primary>subsection</primary></indexterm>

<para>Synopsis, one of:
</para>
<screen>\subsection{<replaceable>title</replaceable>}
\subsection*{<replaceable>title</replaceable>}
\subsection[<replaceable>toc-title</replaceable>]{<replaceable>title</replaceable>}
</screen>
<para>Start a subsection.  The standard &latex; classes <literal>article</literal>,
<literal>book</literal>, and <literal>report</literal> all have this command.
</para>
<para>This produces a subsection.
</para>
<screen>We will show that there are more functions than Turing machines and that
therefore some functions have no associated machine.

\subsection{Cardinality} We will begin with two paradoxes that
dramatize the challenge to our intuition posed by comparing the sizes of
infinite sets.
</screen>
<para>For the standard &latex; classes <literal>book</literal> and <literal>report</literal> the
default output is like &#8216;<literal>1.2.3 <replaceable>title</replaceable></literal>&#8217; (for chapter&#160;1,
section&#160;2, subsection&#160;3), alone on its line and flush left, in
boldface and a larger type (the type size is <literal>\large</literal>).  The same
holds in <literal>article</literal> except that there are no chapters in that class
so it looks like &#8216;<literal>2.3 <replaceable>title</replaceable></literal>&#8217;.
</para>
<para>The <literal>*</literal>&#160;form shows <replaceable>title</replaceable>.
But it does not show the section number, does not increment the
<literal>section</literal> counter, and produces no table of contents entry.
</para>
<para>The optional argument <replaceable>toc-title</replaceable> will appear as the section title
in the table of contents (see <link linkend="Table-of-contents-etc_002e">Table of contents etc.</link>).  If it is
not present then <replaceable>title</replaceable> will be there.  This shows the full name in
the title of the section,
</para>
<screen>\subsection[$\alpha,\beta,\gamma$ paper]{\textit{The Origin of
  Chemical Elements} by R.A.~Alpher, H.~Bethe, and G.~Gamow}
</screen>
<para>but only &#8216;<literal>&#x03B1;,&#x03B2;,&#x03B3;
paper</literal>&#8217; on the contents page.
</para>
<para>For determining which sectional units are numbered and which appear in
the table of contents, the level number of a subsection is&#160;2
(see <link linkend="Sectioning_002fsecnumdepth">Sectioning/secnumdepth</link> and see <link linkend="Sectioning_002ftocdepth">Sectioning/tocdepth</link>).
</para>
<indexterm role="cp"><primary>package, <literal>indentfirst</literal></primary></indexterm>
<indexterm role="cp"><primary><literal>indentfirst</literal> package</primary></indexterm>

<para>The paragraph that follows the subsection title is not indented, as is a
standard typographical practice.  One way to get an indent is to use the
package <filename>indentfirst</filename>.
</para>
<indexterm role="cp"><primary>package, <literal>titlesec</literal></primary></indexterm>
<indexterm role="cp"><primary><literal>titlesec</literal> package</primary></indexterm>

<para>There are a number of ways to change the behavior of the
<literal>\subsection</literal> command.  One is the <literal>\@startsection</literal> command
(see <link linkend="_005c_0040startsection">\@startsection</link>).  There are also many packages on CTAN that
address this, including <filename>titlesec</filename>.  See the documentation but the
example below gives a sense of what they can do.
</para>
<screen>\usepackage{titlesec}   % in preamble
\titleformat{\subsection}[runin]
  {\normalfont\normalsize\bfseries}  % format of the title
  {\thesubsection}                   % label
  {0.6em}                            % space between label and title
  {}                                 % before-code hook
</screen>
<para>That puts the subsection number and <replaceable>title</replaceable> in the first line of
text.
</para>

</sect1>

<sect1 label="6.5" id="_005csubsubsection-_0026-_005cparagraph-_0026-_005csubparagraph">
<title><literal>\subsubsection</literal>, <literal>\paragraph</literal>, <literal>\subparagraph</literal></title>

<indexterm role="fn"><primary>\subsubsection</primary></indexterm>
<indexterm role="cp"><primary>subsubsection</primary></indexterm>
<indexterm role="fn"><primary>\paragraph</primary></indexterm>
<indexterm role="cp"><primary>paragraph</primary></indexterm>
<indexterm role="fn"><primary>\subparagraph</primary></indexterm>
<indexterm role="cp"><primary>subparagraph</primary></indexterm>

<para>Synopsis, one of:
</para>
<screen>\subsubsection{<replaceable>title</replaceable>}
\subsubsection*{<replaceable>title</replaceable>}
\subsubsection[<replaceable>toc-title</replaceable>]{<replaceable>title</replaceable>}
</screen>
<para>or one of:
</para>
<screen>\paragraph{<replaceable>title</replaceable>}
\paragraph*{<replaceable>title</replaceable>}
\paragraph[<replaceable>toc-title</replaceable>]{<replaceable>title</replaceable>}
</screen>
<para>or one of:
</para>
<screen>\subparagraph{<replaceable>title</replaceable>}
\subparagraph*{<replaceable>title</replaceable>}
\subparagraph[<replaceable>toc-title</replaceable>]{<replaceable>title</replaceable>}
</screen>
<para>Start a subsubsection, paragraph, or subparagraph.  The standard
&latex; classes <literal>article</literal>, <literal>book</literal>, and <literal>report</literal> all have
these commands, although they are not commonly used.
</para>
<para>This produces a subsubsection.
</para>
<screen>\subsubsection{Piston ring compressors: structural performance}
Provide exterior/interior wall cladding assemblies
capable of withstanding the effects of load and stresses from 
consumer-grade gasoline engine piston rings.
</screen>
<para>The default output of each of the three does not change over the
standard &latex; classes <literal>article</literal>, <literal>book</literal>, and
<literal>report</literal>.  For <literal>\subsubsection</literal> the <replaceable>title</replaceable> is alone on
its line, in boldface and normal size type.  For <literal>\paragraph</literal> the
<replaceable>title</replaceable> is inline with the text, not indented, in boldface and
normal size type.  For <literal>\subparagraph</literal> the <replaceable>title</replaceable> is inline
with the text, with a paragraph indent, in boldface and normal size type
(Because an <literal>article</literal> has no chapters its subsubsections are
numbered and so it looks like &#8216;<literal>1.2.3 <replaceable>title</replaceable></literal>&#8217;, for
section&#160;1, subsection&#160;2, and subsubsection&#160;3.  The other
two divisions are not numbered.)
</para>
<para>The <literal>*</literal>&#160;form shows <replaceable>title</replaceable>.  But it does not increment the
associated counter and produces no table of contents entry (and does not
show the number for <literal>\subsubsection</literal>).
</para>
<para>The optional argument <replaceable>toc-title</replaceable> will appear as the division title
in the table of contents (see <link linkend="Table-of-contents-etc_002e">Table of contents etc.</link>).  If it is
not present then <replaceable>title</replaceable> will be there.
</para>
<para>For determining which sectional units are numbered and which appear in
the table of contents, the level number of a subsubsection is&#160;3, of
a paragraph is&#160;4, and of a subparagraph is&#160;5
(see <link linkend="Sectioning_002fsecnumdepth">Sectioning/secnumdepth</link> and see <link linkend="Sectioning_002ftocdepth">Sectioning/tocdepth</link>).
</para>
<indexterm role="cp"><primary>package, <literal>indentfirst</literal></primary></indexterm>
<indexterm role="cp"><primary><literal>indentfirst</literal> package</primary></indexterm>

<para>The paragraph that follows the subsubsection title is not indented, as is a
standard typographical practice.  One way to get an indent is to use the
package <filename>indentfirst</filename>.
</para>
<indexterm role="cp"><primary>package, <literal>titlesec</literal></primary></indexterm>
<indexterm role="cp"><primary><literal>titlesec</literal> package</primary></indexterm>

<para>There are a number of ways to change the behavior of the these commands.
One is the <literal>\@startsection</literal> command (see <link linkend="_005c_0040startsection">\@startsection</link>).
There are also many packages on CTAN that address this, including
<filename>titlesec</filename>.  See the documentation on CTAN.
</para>

</sect1>

<sect1 label="6.6" id="_005cappendix">
<title><literal>\appendix</literal></title>

<indexterm role="fn"><primary>\appendix</primary></indexterm>
<indexterm role="cp"><primary>appendix</primary></indexterm>
<indexterm role="cp"><primary>appendices</primary></indexterm>

<para>Synopsis:
</para>
<screen>\appendix
</screen>
<para>This does not directly produce any output.  But in a book or report it
declares that subsequent <literal>\chapter</literal> commands start an appendix.  In
an article it does the same, for <literal>\section</literal> commands.  It also
resets the <literal>chapter</literal> and <literal>section</literal> counters to&#160;0 in a
book or report, and in an article resets the <literal>section</literal> and
<literal>subsection</literal> counters.
</para>
<para>In this book
</para>
<screen>\chapter{One}  ...
\chapter{Two}  ...
 ...
\appendix
\chapter{Three}  ...
\chapter{Four}  ...
</screen>
<para>the first two will generate output numbered &#8216;<literal>Chapter 1</literal>&#8217; and
&#8216;<literal>Chapter 2</literal>&#8217;.  After the <literal>\appendix</literal> the numbering will be
&#8216;<literal>Appendix A</literal>&#8217; and &#8216;<literal>Appendix B</literal>&#8217;.  See <link linkend="Larger-book-template">Larger book template</link>
for another example.
</para>
<indexterm role="cp"><primary>package, <literal>appendix</literal></primary></indexterm>
<indexterm role="cp"><primary><literal>appendix</literal> package</primary></indexterm>
<para>The <filename>appendix</filename> package adds the command
<literal>\appendixpage</literal> to put a separate &#8216;<literal>Appendices</literal>&#8217; in the document
body before the first appendix, and the command <literal>\addappheadtotoc</literal>
to do the same in the table of contents.  You can reset the name
&#8216;<literal>Appendix</literal>&#8217; with a command like
<literal>\renewcommand{\appendixname}{Specification}</literal>, as well as a
number of other features.  See the documentation on CTAN.
</para>

</sect1>

<sect1 label="6.7" id="_005cfrontmatter-_0026-_005cmainmatter-_0026-_005cbackmatter">
<title><literal>\frontmatter</literal>, <literal>\mainmatter</literal>, <literal>\backmatter</literal></title>

<indexterm role="fn"><primary>\frontmatter</primary></indexterm>
<indexterm role="cp"><primary>book, front matter</primary></indexterm>
<indexterm role="fn"><primary>\mainmatter</primary></indexterm>
<indexterm role="cp"><primary>book, main matter</primary></indexterm>
<indexterm role="fn"><primary>\backmatter</primary></indexterm>
<indexterm role="cp"><primary>book, back matter</primary></indexterm>
<indexterm role="cp"><primary>book, end matter</primary></indexterm>

<para>Synopsis, one of:
</para>
<screen>\frontmatter
\mainmatter
\backmatter
</screen>
<para>Format a <literal>book</literal> class document differently according to which part
of the document is being produced.  All three commands are optional.
</para>
<para>Traditionally, a book&#8217;s front matter contains such things as the title
page, an abstract, a table of contents, a preface, a list of notations,
a list of figures, and a list of tables.  (Some of these front matter
pages, such as the title page, are traditionally not numbered.)  The
back matter may contain such things as a glossary, notes, a
bibliography, and an index.
</para>
<para>The <literal>\frontmatter</literal> declaration makes the pages numbered in
lowercase roman, and makes chapters not numbered, although each
chapter&#8217;s title appears in the table of contents; if you use other
sectioning commands here, use the <literal>*</literal>-version (see <link linkend="Sectioning">Sectioning</link>).
The <literal>\mainmatter</literal> changes the behavior back to the expected
version, and resets the page number.  The <literal>\backmatter</literal> leaves the
page numbering alone but switches the chapters back to being not
numbered.  See <link linkend="Larger-book-template">Larger book template</link> for an example using the three.
</para>

</sect1>
<sect1 label="6.8" id="_005c_0040startsection">
<title><literal>\@startsection</literal></title>

<indexterm role="fn"><primary>\@startsection</primary></indexterm>
<indexterm role="cp"><primary>section, redefining</primary></indexterm>

<para>Synopsis:
</para>
<screen>\@startsection{<replaceable>name</replaceable>}{<replaceable>level</replaceable>}{<replaceable>indent</replaceable>}{<replaceable>beforeskip</replaceable>}{<replaceable>afterskip</replaceable>}{<replaceable>style</replaceable>}
</screen>
<para>Used to help redefine the behavior of commands that start sectioning
divisions such as <literal>\section</literal> or <literal>\subsection</literal>.
</para>
<para>Note that the <filename>titlesec</filename> package makes manipulation of sectioning
easier.  Further, while most requirements for sectioning commands can be
satisfied with <literal>\@startsection</literal>, some cannot.  For instance, in
the standard &latex; <literal>book</literal> and <literal>report</literal> classes the commands
<literal>\chapter</literal> and <literal>\report</literal> are not constructed in this way.  To
make such a command you may want to use the <literal>\secdef</literal> command.
<!-- xx define, and make a cross reference to, secdef. -->
</para>
<para>Technically, <literal>\@startsection</literal> has the form
</para>
<screen>\@startsection{<replaceable>name</replaceable>}
  {<replaceable>level</replaceable>}
  {<replaceable>indent</replaceable>}
  {<replaceable>beforeskip</replaceable>}
  {<replaceable>afterskip</replaceable>}
  {<replaceable>style</replaceable>}*[<replaceable>toctitle</replaceable>]{<replaceable>title</replaceable>}
</screen>
<para>so that issuing
</para>
<screen>\renewcommand{\section}{\@startsection{<replaceable>name</replaceable>}
  {<replaceable>level</replaceable>}
  {<replaceable>indent</replaceable>}
  {<replaceable>beforeskip</replaceable>}
  {<replaceable>afterskip</replaceable>}
  {<replaceable>style</replaceable>}}
</screen>
<para>redefines <literal>\section</literal> to have the form
<literal>\section*[<replaceable>toctitle</replaceable>]{<replaceable>title</replaceable>}</literal> (here too, the
star&#160;<literal>*</literal> is optional).  See <link linkend="Sectioning">Sectioning</link>.  This implies that
when you write a command like <literal>\renewcommand{section}{...}</literal>,
the <literal>\@startsection{...}</literal> must come last in the definition.  See
the examples below.
</para>
<variablelist>
<varlistentry><term><replaceable>name</replaceable>
</term><listitem><anchor id="startsection-name"/><anchor id="_005c_0040startsection_002fname"/><para>Name of the counter used to number the sectioning header.  This counter
must be defined separately.  Most commonly this is either
<literal>section</literal>, <literal>subsection</literal>, or <literal>paragraph</literal>.  Although in
those cases the counter name is the same as the sectioning command
itself, you don&#8217;t have to use the same name.
</para>
<para>Then <literal>\the</literal><replaceable>name</replaceable> displays the title number and
<literal>\</literal><replaceable>name</replaceable><literal>mark</literal> is for the page headers.  See the third
example below.
</para>
</listitem></varlistentry><varlistentry><term><replaceable>level</replaceable>
</term><listitem><anchor id="startsection-level"/><anchor id="_005c_0040startsection_002flevel"/><para>An integer giving the depth of the sectioning command.
See <link linkend="Sectioning">Sectioning</link> for the list of standard level numbers.
</para>
<para>If <replaceable>level</replaceable> is less than or equal to the value of the counter
<literal>secnumdepth</literal> then titles for this sectioning command will be
numbered (see <link linkend="Sectioning_002fsecnumdepth">Sectioning/secnumdepth</link>).  For instance, if
<literal>secnumdepth</literal> is 1 in an <literal>article</literal> then the command
<literal>\section{Introduction}</literal> will produce output like &#8220;1
Introduction&#8221; while <literal>\subsection{Discussion}</literal> will produce
output like &#8220;Discussion&#8221;, without the number prefix.
</para>
<para>If <replaceable>level</replaceable> is less than or equal to the value of the counter
<replaceable>tocdepth</replaceable> then the table of contents will have an entry for this
sectioning unit (see <link linkend="Sectioning_002ftocdepth">Sectioning/tocdepth</link>).  For instance, in an
<literal>article</literal>, if <replaceable>tocdepth</replaceable> is 1 then the table of contents will
list sections but not subsections.
</para>
</listitem></varlistentry><varlistentry><term><replaceable>indent</replaceable>
</term><listitem><anchor id="startsection-indent"/><anchor id="_005c_0040startsection_002findent"/><para>A length giving the indentation of all of the title lines with respect
to the left margin.  To have the title flush with the margin use
<literal>0pt</literal>.  A negative indentation such as <literal>-\parindent</literal> will move
the title into the left margin.
</para>
</listitem></varlistentry><varlistentry><term><replaceable>beforeskip</replaceable>
</term><listitem><anchor id="startsection-beforeskip"/><anchor id="_005c_0040startsection_002fbeforeskip"/><para>The absolute value of this length is the amount of vertical space that
is inserted before this sectioning unit&#8217;s title.  This space will be
discarded if the sectioning unit happens to start at the top of a fresh
page.  If this number is negative then the first paragraph following the
header is not indented, if it is non-negative then the first paragraph
is indented.  (Note that the negative of <literal>1pt plus 2pt minus 3pt</literal>
is <literal>-1pt plus -2pt minus -3pt</literal>.)
</para>
<para>For example, if <replaceable>beforeskip</replaceable> is <literal>-3.5ex plus -1ex minus -0.2ex</literal>
then to start the new sectioning unit, &latex; will add about 3.5 times
the height of a letter x in vertical space, and the first paragraph in
the section will not be indented.  Using a rubber length, with
<literal>plus</literal> and <literal>minus</literal>, is good practice here since it gives
&latex; more flexibility in making up the page (see <link linkend="Lengths">Lengths</link>).
</para>
<para>The full accounting of the vertical space between the baseline of the
line prior to this sectioning unit&#8217;s header and the baseline of the
header is that it is the sum of the <literal>\parskip</literal> of the text font,
the <literal>\baselineskip</literal> of the title font, and the absolute value of
the <replaceable>beforeskip</replaceable>.  This space is typically rubber so it may stretch
or shrink.  (If the sectioning unit starts on a fresh page so that the
vertical space is discarded then the baseline of the header text will be
where &latex; would put the baseline of the first text line on that
page.)
</para>
</listitem></varlistentry><varlistentry><term><replaceable>afterskip</replaceable>
</term><listitem><anchor id="startsection-afterskip"/><anchor id="_005c_0040startsection_002fafterskip"/><para>This is a length.  If <replaceable>afterskip</replaceable> is non-negative then this is the
vertical space inserted after the sectioning unit&#8217;s title header.  If it
is negative then the title header becomes a run-in header, so that it
becomes part of the next paragraph.  In this case the absolute value of
the length gives the horizontal space between the end of the title and
the beginning of the following paragraph.  (Note that the negative of
<literal>1pt plus 2pt minus 3pt</literal> is <literal>-1pt plus -2pt minus -3pt</literal>.)
</para>
<para>As with <replaceable>beforeskip</replaceable>, using a rubber length, with <literal>plus</literal> and
<literal>minus</literal> components, is good practice here since it gives &latex;
more flexibility in putting together the page.
</para>
<para>If <literal>afterskip</literal> is non-negative then the full accounting of the
vertical space between the baseline of the sectioning unit&#8217;s header and
the baseline of the first line of the following paragraph is that it is
the sum of the <literal>\parskip</literal> of the title font, the
<literal>\baselineskip</literal> of the text font, and the value of <replaceable>after</replaceable>.
That space is typically rubber so it may stretch or shrink.  (Note that
because the sign of <literal>afterskip</literal> changes the sectioning unit
header&#8217;s from standalone to run-in, you cannot use a negative
<literal>afterskip</literal> to cancel part of the <literal>\parskip</literal>.)
</para>
</listitem></varlistentry><varlistentry><term><replaceable>style</replaceable>
</term><listitem><anchor id="startsection-style"/><anchor id="_005c_0040startsection_002fstyle"/><para>Controls the styling of the title.  See the examples below.  Typical
commands to use here are <literal>\centering</literal>, <literal>\raggedright</literal>,
<literal>\normalfont</literal>, <literal>\hrule</literal>, or <literal>\newpage</literal>.  The last command
in <replaceable>style</replaceable> may be one that takes one argument, such as
<literal>\MakeUppercase</literal> or <literal>\fbox</literal> that takes one argument.  The
section title will be supplied as the argument to this command. For
instance, setting <replaceable>style</replaceable> to <literal>\bfseries\MakeUppercase</literal> would
produce titles that are bold and uppercase.
</para></listitem></varlistentry></variablelist>
<para>These are &latex;&#8217;s defaults for the first three sectioning units that
are defined with <literal>\@startsection</literal>, for the <filename>article</filename>,
<filename>book</filename>, and <filename>report</filename> classes.  For section, the <replaceable>level</replaceable> is
1, the <replaceable>indent</replaceable> is 0pt, the <replaceable>beforeskip</replaceable> is <literal>-3.5ex
plus -1ex minus -0.2ex</literal>, the <replaceable>afterskip</replaceable> is <literal>2.3ex plus 0.2ex</literal>,
and the <replaceable>style</replaceable> is <literal>\normalfont\Large\bfseries</literal>.  For
subsection, the <replaceable>level</replaceable> is 2, the <replaceable>indent</replaceable> is 0pt, the
<replaceable>beforeskip</replaceable> is <literal>-3.25ex plus -1ex minus -0.2ex</literal>, the
<replaceable>afterskip</replaceable> is <literal>1.5ex plus 0.2ex</literal>, and the <replaceable>style</replaceable> is
<literal>\normalfont\large\bfseries</literal>.  For subsubsection, the <replaceable>level</replaceable>
is 3, the <replaceable>indent</replaceable> is 0pt, the <replaceable>beforeskip</replaceable> is
<literal>-3.25ex plus -1ex minus -0.2ex</literal>, the <replaceable>afterskip</replaceable> is
<literal>1.5ex plus 0.2ex</literal>, and the <replaceable>style</replaceable> is
<literal>\normalfont\normalsize\bfseries</literal>.
</para>
<para>Here are examples.  They go either in a package or class file or in the
preamble of a &latex; document.  If you put them in the preamble they
must go between a <literal>\makeatletter</literal> command and a
<literal>\makeatother</literal>.  (Probably the error message <literal>You can't use
`\spacefactor' in vertical mode.</literal>  means that you forgot this.)
See <link linkend="_005cmakeatletter-_0026-_005cmakeatother">\makeatletter &amp; \makeatother</link>.
</para>
<para>This will put section titles in large boldface type, centered.  It says
<literal>\renewcommand</literal> because &latex;&#8217;s standard classes have already
defined a <literal>\section</literal>.  For the same reason it does not define a
<literal>section</literal> counter, or the commands <literal>\thesection</literal> and
<literal>\l@section</literal>.
</para>
<screen>\renewcommand\section{%
  \@startsection{section}% <link linkend="_005c_0040startsection_002fname"><replaceable>name</replaceable></link>
    {1}% <link linkend="_005c_0040startsection_002flevel"><replaceable>level</replaceable></link>
    {0pt}% <link linkend="_005c_0040startsection_002findent"><replaceable>indent</replaceable></link>
    {-3.5ex plus -1ex minus -.2ex}% <link linkend="_005c_0040startsection_002fbeforeskip"><replaceable>beforeskip</replaceable></link>
    {2.3ex plus.2ex}% <link linkend="_005c_0040startsection_002fafterskip"><replaceable>afterskip</replaceable></link>
    {\centering\normalfont\Large\bfseries}% <link linkend="_005c_0040startsection_002fstyle"><replaceable>style</replaceable></link>
  }
</screen>
<para>This will put <literal>subsection</literal> titles in small caps type, inline with the paragraph.
</para>
<screen>\renewcommand\subsection{%
  \@startsection{subsection}%  <link linkend="_005c_0040startsection_002fname"><replaceable>name</replaceable></link>
    {2}% <link linkend="_005c_0040startsection_002flevel"><replaceable>level</replaceable></link>
    {0em}% <link linkend="_005c_0040startsection_002findent"><replaceable>indent</replaceable></link>
    {-1ex plus 0.1ex minus -0.05ex}% <link linkend="_005c_0040startsection_002fbeforeskip"><replaceable>beforeskip</replaceable></link>
    {-1em plus 0.2em}% <link linkend="_005c_0040startsection_002fafterskip"><replaceable>afterskip</replaceable></link>
    {\scshape}% <link linkend="_005c_0040startsection_002fstyle"><replaceable>style</replaceable></link>
  }
</screen>
<para>The prior examples redefined existing sectional unit title commands.
This defines a new one, illustrating the needed counter and macros to
display that counter.
</para>
<!-- From https://groups.google.com/forum/#!searchin/comp.text.tex/startsection%7Csort:relevance/comp.text.tex/sB-nTS-oL08/ZZeKYdG0llMJ -->
<screen>\setcounter{secnumdepth}{6}% show counters this far down
\newcounter{subsubparagraph}[subparagraph]% counter for numbering
\renewcommand{\thesubsubparagraph}%               how to display 
  {\thesubparagraph.\@arabic\c@subsubparagraph}%  numbering
\newcommand{\subsubparagraph}{\@startsection
                         {subsubparagraph}%
                         {6}%
                         {0em}%
                         {\baselineskip}%
                         {0.5\baselineskip}%
                         {\normalfont\normalsize}}
\newcommand*\l@subsubparagraph{\@dottedtocline{6}{10em}{5em}}% for toc
\newcommand{\subsubparagraphmark}[1]{}% for page headers
</screen>

</sect1>
</chapter>
<chapter label="7" id="Cross-references">
<title>Cross references</title>

<indexterm role="cp"><primary>cross references</primary></indexterm>

<indexterm role="cp"><primary>label</primary></indexterm>
<para>We often want something like &#8216;<literal>See Theorem~31</literal>&#8217;.  But by-hand typing
the 31 is poor practice.  Instead you should write a <firstterm>label</firstterm> such as
<literal>\label{eq:GreensThm}</literal> and then <firstterm>reference</firstterm> it, as with
<literal>See equation~\ref{eq:GreensThm}</literal>.  &latex; will automatically
work out the number, put it into the output, and will change that number
later if needed.
</para>
<screen>We will see this with Theorem~\ref{th:GreensThm}. % forward reference
...
\begin{theorem} \label{th:GreensThm}
  ...
\end{theorem}
...
See Theorem~\ref{th:GreensThm} on page~\pageref{th:GreensThm}.
</screen>
<para>&latex; tracks cross reference information in a file having the
extension <filename>.aux</filename> and with the same base name as the file containing
the <literal>\label</literal>.  So if <literal>\label</literal> is in <filename>calculus.tex</filename> then
the information is in <filename>calculus.aux</filename>.  &latex; puts the
information in that file every time it runs across a <literal>\label</literal>.
</para>
<indexterm role="cp"><primary>forward reference</primary></indexterm>
<indexterm role="cp"><primary>reference, forward</primary></indexterm>
<para>The behavior described in the prior paragraph results in a quirk that
happens when your document has a <firstterm>forward reference</firstterm>, a <literal>\ref</literal>
that appears before the associated <literal>\label</literal>.  If this is the first
time that you are compiling the document then you will get &#8216;<literal>LaTeX
Warning: Label(s) may have changed. Rerun to get cross references right</literal>&#8217;
and in the output the forward reference will appear as two question
marks&#160;&#8216;<literal>??</literal>&#8217;, in boldface.  A similar thing happens if you
change some things so the references changes; you get the same warning
and the output contains the old reference information.  In both cases,
resolve this by compiling the document a second time.
</para>
<indexterm role="cp"><primary>package, <literal>cleveref</literal></primary></indexterm>
<indexterm role="cp"><primary><literal>cleveref</literal> package</primary></indexterm>
<para>The <literal>cleveref</literal> package enhances &latex;&#8217;s
cross referencing features.  You can arrange that if you enter
<literal>\begin{thm}\label{th:Nerode}...\end{thm}</literal> then
<literal>\cref{th:Nerode}</literal> will output &#8216;<literal>Theorem 3.21</literal>&#8217;, without you
having to enter the &#8220;Theorem.&#8221;
</para>


<sect1 label="7.1" id="_005clabel">
<title><literal>\label</literal></title>

<indexterm role="fn"><primary>\label</primary></indexterm>

<para>Synopsis:
</para>
<screen>\label{<replaceable>key</replaceable>}
</screen>
<para>Assign a reference number to <replaceable>key</replaceable>.  In ordinary text
<literal>\label{<replaceable>key</replaceable>}</literal> assigns to <replaceable>key</replaceable> the number of the
current sectional unit. Inside an environment with numbering, such as a
<literal>table</literal> or <literal>theorem</literal> environment, <literal>\label{<replaceable>key</replaceable>}</literal>
assigns to <replaceable>key</replaceable> the number of that environment.  Retrieve the
assigned number with the <literal>\ref{<replaceable>key</replaceable>}</literal> command
(see <link linkend="_005cref">\ref</link>).
</para>
<para>A key name can consist of any sequence of letters, digits, or common
punctuation characters.  Upper and lowercase letters are
distinguished, as usual.
</para>
<para>A common convention is to use labels consisting of a prefix and a suffix
separated by a colon or period.  Thus, <literal>\label{fig:Post}</literal> is a
label for a figure with a portrait of Emil Post.  This helps to avoid
accidentally creating two labels with the same name, and makes your
source more readable.  Some commonly-used prefixes:
</para>
<variablelist><varlistentry><term><literal>ch</literal>
</term><listitem><para>for chapters
</para>
</listitem></varlistentry><varlistentry><term><literal>sec</literal>
</term><term><literal>subsec</literal>
</term><listitem><para>for lower-level sectioning commands
</para>
</listitem></varlistentry><varlistentry><term><literal>fig</literal>
</term><listitem><para>for figures
</para>
</listitem></varlistentry><varlistentry><term><literal>tab</literal>
</term><listitem><para>for tables
</para>
</listitem></varlistentry><varlistentry><term><literal>eq</literal>
</term><listitem><para>for equations
</para></listitem></varlistentry></variablelist>
<para>In the auxiliary file the reference information is kept as the text of
a command of the form
<literal>\newlabel{<replaceable>label</replaceable>}{{<replaceable>currentlabel</replaceable>}{<replaceable>pagenumber</replaceable>}}</literal>.
Here <replaceable>currentlabel</replaceable> is the current value of the macro
<literal>\@currentlabel</literal> that is usually updated whenever you call
<literal>\refstepcounter{<replaceable>counter</replaceable>}</literal>.
</para>
<para>Below, the key <literal>sec:test</literal> will get the number of the current
section and the key <literal>fig:test</literal> will get the number of the figure.
(Incidentally, put labels after captions in figures and tables.)
</para>
<screen>\section{section name}
\label{sec:test}
This is Section~\ref{sec:test}.
\begin{figure}
  ...
  \caption{caption text}
  \label{fig:test}
\end{figure}
See Figure~\ref{fig:test}.
</screen>

</sect1>
<sect1 label="7.2" id="_005cpageref">
<title><literal>\pageref</literal></title>

<indexterm role="fn"><primary>\pageref</primary></indexterm>
<indexterm role="cp"><primary>cross referencing with page number</primary></indexterm>
<indexterm role="cp"><primary>page number, cross referencing</primary></indexterm>

<para>Synopsis:
</para>
<screen>\pageref{<replaceable>key</replaceable>}
</screen>
<para>Produce the page number of the place in the text where the corresponding
<literal>\label</literal>{<replaceable>key</replaceable>} command appears.
</para>
<para>If there is no <literal>\label{<replaceable>key</replaceable>}</literal> then you get something like
&#8216;<literal>LaTeX Warning: Reference `th:GrensThm' on page 1 undefined on
input line 11.</literal>&#8217;
</para>
<para>Below, the <literal>\label{eq:main}</literal> is used both for the formula number
and for the page number.  (Note that the two references are forward
references so this document would need to be compiled twice to resolve
those.)
</para>
<screen>The main result is formula~\ref{eq:main} on page~\pageref{eq:main}.
  ...
\begin{equation} \label{eq:main}
   \mathbf{P}=\mathbf{NP}
\end{equation}
</screen>

</sect1>
<sect1 label="7.3" id="_005cref">
<title><literal>\ref</literal></title>

<indexterm role="fn"><primary>\ref</primary></indexterm>
<indexterm role="cp"><primary>cross referencing, symbolic</primary></indexterm>
<indexterm role="cp"><primary>section number, cross referencing</primary></indexterm>
<indexterm role="cp"><primary>equation number, cross referencing</primary></indexterm>
<indexterm role="cp"><primary>figure number, cross referencing</primary></indexterm>
<indexterm role="cp"><primary>footnote number, cross referencing</primary></indexterm>

<para>Synopsis:
</para>
<screen>\ref{<replaceable>key</replaceable>}
</screen>
<para>Produces the number of the sectional unit,
equation, footnote, figure, &#8230;, of the corresponding
<literal>\label</literal> command (see <link linkend="_005clabel">\label</link>).  It does not produce any text,
such as the word &#8216;Section&#8217; or &#8216;Figure&#8217;, just the bare number itself.
</para>
<para>If there is no <literal>\label{<replaceable>key</replaceable>}</literal> then you get something like
&#8216;<literal>LaTeX Warning: Reference `th:GrensThm' on page 1 undefined on
input line 11.</literal>&#8217;
</para>
<para>In this example the <literal>\ref{popular}</literal> produces &#8216;<literal>2</literal>&#8217;.  Note that
it is a forward reference since it comes before <literal>\label{popular}</literal>
so this document would have to be compiled twice.
</para>
<screen>The most widely-used format is item number~\ref{popular}.
\begin{enumerate}
\item Plain \TeX
\item \label{popular} \LaTeX
\item Con\TeX t
\end{enumerate}
</screen>
<indexterm role="cp"><primary>package, <literal>cleveref</literal></primary></indexterm>
<indexterm role="cp"><primary><literal>cleveref</literal> package</primary></indexterm>

<para>The <filename>cleveref</filename> package includes text such as &#8216;<literal>Theorem</literal>&#8217; in the
reference.  See the documentation on CTAN.
</para>

</sect1>
</chapter>
<chapter label="8" id="Environments">
<title>Environments</title>

<indexterm role="cp"><primary>environments</primary></indexterm>
<indexterm role="fn"><primary>\begin</primary></indexterm>
<indexterm role="fn"><primary>\end</primary></indexterm>

<para>&latex; provides many environments for delimiting certain behavior.
An environment begins with <literal>\begin</literal> and ends with <literal>\end</literal>,
like this:
</para>
<screen>\begin{<replaceable>environment-name</replaceable>}
  ...
\end{<replaceable>environment-name</replaceable>}
</screen>
<para>The <replaceable>environment-name</replaceable> at the beginning must exactly match that at
the end. For instance, the input
<literal>\begin{table*}...\end{table}</literal> will cause an error like:
&#8216;<literal>! LaTeX Error: \begin{table*} on input line 5 ended by
\end{table}.</literal>&#8217;
</para>
<indexterm role="cp"><primary>group, and environments</primary></indexterm>
<para>Environments are executed within a group.
</para>


<sect1 label="8.1" id="abstract">
<title><literal>abstract</literal></title>

<indexterm role="fn"><primary>environment, <literal>abstract</literal></primary></indexterm>
<indexterm role="fn"><primary><literal>abstract</literal> environment</primary></indexterm>
        
<indexterm role="cp"><primary>abstracts</primary></indexterm>

<para>Synopsis:
</para>
<screen>\begin{abstract}
...
\end{abstract}
</screen>
<para>Produce an abstract, possibly of multiple paragraphs.  This environment
is only defined in the <literal>article</literal> and <literal>report</literal> document classes
(see <link linkend="Document-classes">Document classes</link>).
</para>
<para>Using the example below in the <literal>article</literal> class produces a displayed
paragraph.  Document class option <literal>titlepage</literal> causes the abstract
to be on a separate page (see <link linkend="Document-class-options">Document class options</link>); this is the
default only in the <literal>report</literal> class.
</para>
<screen>\begin{abstract}
  We compare all known accounts of the proposal made by Porter Alexander 
  to Robert E Lee at the Appomattox Court House that the army continue
  in a guerrilla war, which Lee refused.
\end{abstract}
</screen>
<para>The next example produces a one column abstract in a two column document (for
a more flexible solution, use the package <filename>abstract</filename>).
</para>
<!-- Adopted from http://www.tex.ac.uk/FAQ-onecolabs.html -->
<screen>\documentclass[twocolumn]{article}
  ...
\begin{document}
\title{Babe Ruth as Cultural Progenitor: a Atavistic Approach}
\author{Smith \\ Jones \\ Robinson\thanks{Railroad tracking grant.}}
\twocolumn[
  \begin{@twocolumnfalse}
     \maketitle
     \begin{abstract}
       Ruth was not just the Sultan of Swat, he was the entire swat
       team.   
     \end{abstract}
   \end{@twocolumnfalse}
   ]
{   % by-hand insert a footnote at page bottom
 \renewcommand{\thefootnote}{\fnsymbol{footnote}}
 \footnotetext[1]{Thanks for all the fish.}
}
</screen>

</sect1>
<sect1 label="8.2" id="array">
<title><literal>array</literal></title>

<indexterm role="fn"><primary>environment, <literal>array</literal></primary></indexterm>
<indexterm role="fn"><primary><literal>array</literal> environment</primary></indexterm>
        
<indexterm role="cp"><primary>arrays, math</primary></indexterm>

<para>Synopsis:
</para>
<screen>\begin{array}{<replaceable>cols</replaceable>}
  <replaceable>column 1 entry</replaceable> &amp;<replaceable>column 2 entry</replaceable> ... &amp;<replaceable>column n entry</replaceable> \\
  ...
\end{array}
</screen>
<para>or:
</para>
<screen>\begin{array}[<replaceable>pos</replaceable>]{<replaceable>cols</replaceable>}
  <replaceable>column 1 entry</replaceable> &amp;<replaceable>column 2 entry</replaceable> ... &amp;<replaceable>column n entry</replaceable> \\
  ...
\end{array}
</screen>
<para>Produce a mathematical array.  This environment can only be used in math
mode, and normally appears within a displayed mathematics environment
such as <literal>equation</literal> (see <link linkend="equation">equation</link>).  Inside of each row the
column entries are separated by an ampersand, (<literal>&amp;</literal>).  Rows are
terminated with double-backslashes (see <link linkend="_005c_005c">\\</link>).
</para>
<para>This example shows a three by three array.
</para>
<screen>\begin{equation*}
  \chi(x) =
  \left|              % vertical bar fence
    \begin{array}{ccc}
      x-a  &amp;-b  &amp;-c  \\
      -d   &amp;x-e &amp;-f  \\
      -g   &amp;-h  &amp;x-i
    \end{array}
 \right|
\end{equation*}
</screen>
<para>The required argument <replaceable>cols</replaceable> describes the number of columns, their
alignment, and the formatting of the intercolumn regions.  For instance,
<literal>\begin{array}{rcl}...\end{array}</literal> gives three columns: the
first flush right, the second centered, and the third flush left.  See
<link linkend="tabular">tabular</link> for the complete description of <replaceable>cols</replaceable> and of the
other common features of the two environments, including the optional
<replaceable>pos</replaceable> argument.
</para>
<para>There are two ways that <literal>array</literal> diverges from <literal>tabular</literal>.  The
first is that <literal>array</literal> entries are typeset in math mode, in
textstyle (see <link linkend="Modes">Modes</link>) except if the <replaceable>cols</replaceable> definition specifies
the column with <literal>p{...}</literal>, which causes the entry to be typeset in
text mode.  The second is that, instead of <literal>tabular</literal>&#8217;s parameter
<literal>\tabcolsep</literal>, &latex;&#8217;s intercolumn space in an <literal>array</literal> is
governed by
<indexterm role="fn"><primary>\arraycolsep</primary></indexterm>
<literal>\arraycolsep</literal>, which gives half the width between columns. The
default for this is &#8216;<literal>5pt</literal>&#8217; so that between two columns comes
10pt of space.
</para>
<indexterm role="cp"><primary>package, <literal>amsmath</literal></primary></indexterm>
<indexterm role="cp"><primary><literal>amsmath</literal> package</primary></indexterm>

<para>To obtain arrays with braces the standard is to use the <filename>amsmath</filename>
package.  It comes with environments <literal>pmatrix</literal> for an array
surrounded by parentheses&#160;<literal>(...)</literal>, <literal>bmatrix</literal> for an array
surrounded by square brackets&#160;<literal>[...]</literal>, <literal>Bmatrix</literal> for an
array surrounded by curly braces&#160;<literal>{...}</literal>, <literal>vmatrix</literal> for
an array surrounded by vertical bars&#160;<literal>|...|</literal>, and
<literal>Vmatrix</literal> for an array surrounded by double vertical
bars&#160;<literal>||...||</literal>, along with a number of other array constructs.
</para>
<indexterm role="cp"><primary>package, <literal>amsmath</literal></primary></indexterm>
<indexterm role="cp"><primary><literal>amsmath</literal> package</primary></indexterm>

<para>The next example uses the <filename>amsmath</filename> package.
</para>
<screen>\usepackage{amsmath}  % in preamble

\begin{equation}    
  \begin{vmatrix}{cc}  % array with vert lines
    a  &amp;b \\
    c  &amp;d       
  \end{vmatrix}=ad-bc
\end{equation}
</screen>
<indexterm role="cp"><primary>package, <literal>array (package)</literal></primary></indexterm>
<indexterm role="cp"><primary><literal>array (package)</literal> package</primary></indexterm>

<indexterm role="cp"><primary>package, <literal>dcolumn</literal></primary></indexterm>
<indexterm role="cp"><primary><literal>dcolumn</literal> package</primary></indexterm>

<para>There are many packages concerning arrays.  The <filename>array</filename> package has
many useful extensions, including more column types.  The <filename>dcolumn</filename>
package adds a column type to center on a decimal point.  For both see
the documentation on CTAN.
</para>

</sect1>
<sect1 label="8.3" id="center">
<title><literal>center</literal></title>

<indexterm role="fn"><primary>environment, <literal>center</literal></primary></indexterm>
<indexterm role="fn"><primary><literal>center</literal> environment</primary></indexterm>
        
<indexterm role="cp"><primary>centering text, environment for</primary></indexterm>

<para>Synopsis:
</para>
<screen>\begin{center}
  <replaceable>line1</replaceable> \\
  <replaceable>line2</replaceable> \\
  ...
\end{center}
</screen>
<para>Create a new paragraph consisting of a sequence of lines that are
centered within the left and right margins.  Use
double-backslash, <literal>\\</literal>, to get a line break (see <link linkend="_005c_005c">\\</link>).
<indexterm role="fn"><primary>\\ (for <literal>center</literal>)</primary></indexterm>
If some text is too long to fit on a line then &latex; will insert line
breaks that avoid hyphenation and avoid stretching or shrinking any
interword space.
</para>
<para>This environment inserts space above and below the text body.  See
<link linkend="_005ccentering">\centering</link> to avoid such space, for example inside a <literal>figure</literal>
environment.
</para>
<para>This example produces three centered lines.  There is extra vertical
space between the last two lines.
</para>
<screen>\begin{center}
  A Thesis Submitted in Partial Fufillment \\
  of the Requirements of \\[0.5ex]
  the School of Environmental Engineering
\end{center}
</screen>
<para>In this example, depending on the page&#8217;s line width, &latex; may choose
a line break for the part before the double backslash.  If so, it will
center each of the two lines and if not it will center the single line.
Then &latex; will break at the double backslash, and will center the
ending.
</para>
<screen>\begin{center}
  My father considered that anyone who went to chapel and didn't drink 
  alcohol was not to be tolerated.\\ 
  I grew up in that belief.  --Richard Burton 
\end{center}
</screen>
<para>A double backslash after the final line is optional.  If present it
doesn&#8217;t add any vertical space.
</para>
<para>In a two-column document the text is centered in a column, not in the
entire page.
</para>


<sect2 label="8.3.1" id="_005ccentering">
<title><literal>\centering</literal></title>

<indexterm role="fn"><primary>\centering</primary></indexterm>
<indexterm role="cp"><primary>centering text, declaration for</primary></indexterm>


<para>Synopsis:
</para>
<screen>{\centering ... }
</screen>
<para>or
</para>
<screen>\begin{group}
  \centering ...
\end{group}
</screen>
<para>Center the material in its scope.  It is most often used inside an
environment such as <literal>figure</literal>, or in a <literal>parbox</literal>.
</para>
<para>This example&#8217;s <literal>\centering</literal> declaration causes the graphic to be
horizontally centered.
</para>
<screen>\begin{figure}
  \centering
  \includegraphics[width=0.6\textwidth]{ctan_lion.png}
  \caption{CTAN Lion}  \label{fig:CTANLion}
\end{figure}
</screen>
<para>The scope of this <literal>\centering</literal> ends with the <literal>\end{figure}</literal>.
</para>
<para>Unlike the <literal>center</literal> environment, the <literal>\centering</literal> command does
not add vertical space above and below the text.  That&#8217;s its advantage
in the above example; there is not an excess of space.
</para>
<para>It also does not start a new paragraph; it simply changes how &latex;
formats paragraph units.  If <literal>ww {\centering xx \\ yy} zz</literal> is
surrounded by blank lines then &latex; will create a paragraph whose
first line &#8216;<literal>ww xx</literal>&#8217; is centered and whose second line, not centered,
contains &#8216;<literal>yy zz</literal>&#8217;.  Usually what is desired is for the scope of the
declaration to contain a blank line or the <literal>\end</literal> command of an
environment such as <literal>figure</literal> or <literal>table</literal> that ends the
paragraph unit.  Thus, if <literal>{\centering xx \\ yy\par} zz</literal> is
surrounded by blank lines then it makes a new paragraph with two
centered lines &#8216;<literal>xx</literal>&#8217; and &#8216;<literal>yy</literal>&#8217;, followed by a new paragraph with
&#8216;<literal>zz</literal>&#8217; that is formatted as usual.
</para>

</sect2>
</sect1>
<sect1 label="8.4" id="description">
<title><literal>description</literal></title>

<indexterm role="fn"><primary>environment, <literal>description</literal></primary></indexterm>
<indexterm role="fn"><primary><literal>description</literal> environment</primary></indexterm>
        
<indexterm role="cp"><primary>labelled lists, creating</primary></indexterm>
<indexterm role="cp"><primary>description lists, creating</primary></indexterm>

<para>Synopsis:
</para>
<screen>\begin{description}
  \item[<replaceable>label of first item</replaceable>] <replaceable>text of first item</replaceable>
  \item[<replaceable>label of second item</replaceable>] <replaceable>text of second item</replaceable>
   ...
\end{description}
</screen>
<para>Environment to make a list of labeled items.  Each item&#8217;s <replaceable>label</replaceable> is
typeset in bold and is flush left, so that long labels continue into the
first line of the item text.  There must be at least one item; having
none causes the &latex; error &#8216;<literal>Something's wrong--perhaps a
missing \item</literal>&#8217;.
</para>
<para>This example shows the environment used for a sequence of definitions.
</para>
<screen>\begin{definition}
  \item[lama] A priest.
  \item[llama] A beast.
\end{definition}
</screen>
<para>The labels &#8216;<literal>lama</literal>&#8217; and &#8216;<literal>llama</literal>&#8217; are output in boldface, with the
left edge on the left margin.
</para>
<indexterm role="fn"><primary>\item</primary></indexterm>
<para>Start list items with the <literal>\item</literal> command (see <link linkend="_005citem">\item</link>).  Use the
optional labels, as in <literal>\item[Main point]</literal>, because there is
no sensible default.  Following the <literal>\item</literal> is optional text, which
may contain multiple paragraphs.
</para>
<indexterm role="cp"><primary>bold typewriter, avoiding</primary></indexterm>
<indexterm role="cp"><primary>typewriter labels in lists</primary></indexterm>
<para>Since the labels are in bold style, if the label text calls for a font
change given in argument style (see <link linkend="Font-styles">Font styles</link>) then it will come
out bold.  For instance, if the label text calls for typewriter with
<literal>\item[\texttt{label text}]</literal> then it will appear in bold
typewriter, if that is available. The simplest way around this, in this
example to get non-bold typewriter, is to use declarative style:
<literal>\item[{\tt label text}]</literal>.  Similarly, get the standard roman
font with <literal>\item[{\rm label text}]</literal>.
</para>
<para>For other major &latex; labelled list environments, see <link linkend="itemize">itemize</link>
and <link linkend="enumerate">enumerate</link>.  Unlike those environments, nesting
<literal>description</literal> environments does not change the default label; it is
boldface and flush left at all levels.
</para>
<para>For information about list layout parameters, including the default
values, and for information about customizing list layout, see
<link linkend="list">list</link>.  The package <filename>enumitem</filename> is useful for customizing
lists.
</para>
<para>This example changes the description labels to small caps.
</para>
<screen>\renewcommand{\descriptionlabel}[1]{%
  {\hspace{\labelsep}\textsc{#1}}}
</screen>

</sect1>
<sect1 label="8.5" id="displaymath">
<title><literal>displaymath</literal></title>
<!-- http://tex.stackexchange.com/questions/40492/what-are-the-differences-between-align-equation-and-displaymath -->

<indexterm role="fn"><primary>environment, <literal>displaymath</literal></primary></indexterm>
<indexterm role="fn"><primary><literal>displaymath</literal> environment</primary></indexterm>
        

<para>Synopsis:
</para>
<screen>\begin{displaymath}
  <replaceable>mathematical text</replaceable>
\end{displaymath}
</screen>
<para>Environment to typeset the math text on its own line, in display style
and centered.  To make the text be flush-left use the global option
<literal>fleqn</literal>; see <link linkend="Document-class-options">Document class options</link>.
</para>
<para>In the <literal>displaymath</literal> environment no equation number is added to the
math text. One way to get an equation number is to use the
<literal>equation</literal> environment (see <link linkend="equation">equation</link>).
</para>
<para>&latex; will not break the <replaceable>math text</replaceable> across lines.
</para>
<indexterm role="cp"><primary>package, <literal>amsmath</literal></primary></indexterm>
<indexterm role="cp"><primary><literal>amsmath</literal> package</primary></indexterm>

<para>Note that the <filename>amsmath</filename> package has significantly more extensive
displayed equation facilities.  For example, there are a number of
ways in that package for having math text broken across lines.
</para>
<para>The construct <literal>\[<replaceable>math text</replaceable>\]</literal> is essentially a synonym for
<literal>\begin{displaymath}<replaceable>math text</replaceable>\end{displaymath}</literal> but the
latter is easier to work with in the source file; for instance,
searching for a square bracket may get false positives but the word
<literal>displaymath</literal> will likely be unique.  (The construct
<literal>$$<replaceable>math text</replaceable>$$</literal> from Plain&#160;&tex; is sometimes
mistakenly used as a synonym for <literal>displaymath</literal>.  It is not a
synonym, because the <literal>displaymath</literal> environment checks that it isn&#8217;t
started in math mode and that it ends in math mode begun by the matching
environment start, because the <literal>displaymath</literal> environment has
different vertical spacing, and because the <literal>displaymath</literal>
environment honors the <literal>fleqn</literal> option.)
</para>
<para>The output from this example is centered and alone on its line. 
</para>
<screen>\begin{displaymath}
  \int_1^2 x^2\,dx=7/3
\end{displaymath}
</screen>
<para>Also, the integral sign is larger than the inline version  
<literal>\( \int_1^2 x^2\,dx=7/3 \)</literal> produces.
</para>

</sect1>
<sect1 label="8.6" id="document">
<title><literal>document</literal></title>

<indexterm role="fn"><primary>environment, <literal>document</literal></primary></indexterm>
<indexterm role="fn"><primary><literal>document</literal> environment</primary></indexterm>
        

<para>The <literal>document</literal> environment encloses the entire body of a document.
It is required in every &latex; document.  See <link linkend="Starting-and-ending">Starting and ending</link>.
</para>


<sect2 label="8.6.1" id="_005cAtBeginDocument">
<title><literal>\AtBeginDocument</literal></title>

<indexterm role="fn"><primary>\AtBeginDocument</primary></indexterm>
<indexterm role="cp"><primary>beginning of document hook</primary></indexterm>

<para>Synopsis:
</para>
<screen>\AtBeginDocument{<replaceable>code</replaceable>}
</screen>
<para>Save <replaceable>code</replaceable> and execute it when <literal>\begin{document}</literal> is
executed, at the very end of the preamble.  The code is executed after
the font selection tables have been set up, so the normal font for the
document is the current font.  However, the code is executed as part of
the preamble so you cannot do any typesetting with it.
</para>
<para>You can issue this command more than once; the successive code lines
will be executed in the order that you gave them.
</para>

</sect2>
<sect2 label="8.6.2" id="_005cAtEndDocument">
<title><literal>\AtEndDocument</literal></title>

<indexterm role="fn"><primary>\AtEndDocument</primary></indexterm>
<indexterm role="cp"><primary>end of document hook</primary></indexterm>

<para>Synopsis:
</para>
<screen>\AtEndDocument{<replaceable>code</replaceable>}
</screen>
<para>Save <replaceable>code</replaceable> and execute it near the end of the document.
Specifically, it is executed when <literal>\end{document}</literal> is executed,
before the final page is finished and before any leftover floating
environments are processed. If you want some of the code to be executed
after these two processes then include a <literal>\clearpage</literal> at the
appropriate point in <replaceable>code</replaceable>.
</para>
<para>You can issue this command more than once; the successive code lines
will be executed in the order that you gave them.
</para>

</sect2>
</sect1>
<sect1 label="8.7" id="enumerate">
<title><literal>enumerate</literal></title>

<indexterm role="fn"><primary>environment, <literal>enumerate</literal></primary></indexterm>
<indexterm role="fn"><primary><literal>enumerate</literal> environment</primary></indexterm>
        
<indexterm role="cp"><primary>lists of items, numbered</primary></indexterm>

<para>Synopsis:
</para>
<screen>\begin{enumerate}
  \item[<replaceable>optional label of first item</replaceable>] <replaceable>text of first item</replaceable>
  \item[<replaceable>optional label of second item</replaceable>] <replaceable>text of second item</replaceable>
  ...
\end{enumerate}
</screen>
<para>Environment to produce a numbered list of items.  The format of the
label numbering depends on the nesting level of this environment; see
below. The default top-level numbering is &#8216;<literal>1.</literal>&#8217;, &#8216;<literal>2.</literal>&#8217;,
etc. Each <literal>enumerate</literal> list environment must have at least one item;
having none causes the &latex; error &#8216;<literal>Something's wrong--perhaps a
missing \item</literal>&#8217;.
</para>
<para>This example gives the first two finishers in the 1908 Olympic marathon.
As a top-level list the labels would come out as &#8216;<literal>1.</literal>&#8217; and
&#8216;<literal>2.</literal>&#8217;.
</para>
<screen>\begin{enumerate}
 \item Johnny Hayes (USA)
 \item Charles Hefferon (RSA) 
\end{enumerate}
</screen>
<indexterm role="fn"><primary>\item</primary></indexterm>
<para>Start list items with the <literal>\item</literal> command (see <link linkend="_005citem">\item</link>).  If you
give <literal>\item</literal> an optional argument by following it with square
brackets, as in <literal>\item[Interstitial label]</literal>, then the next item
will continue the interrupted sequence (see <link linkend="_005citem">\item</link>).  That is, you
will get labels like &#8216;<literal>1.</literal>&#8217;, then &#8216;<literal>Interstitial label</literal>&#8217;, then
&#8216;<literal>2.</literal>&#8217;.  Following the <literal>\item</literal> is optional text, which may
contain multiple paragraphs.
</para>
<para>Enumerations may be nested within other <literal>enumerate</literal> environments,
or within any paragraph-making environment such as <literal>itemize</literal>
(see <link linkend="itemize">itemize</link>), up to four levels deep.  This gives &latex;&#8217;s
default for the format at each nesting level, where 1 is the top level,
the outermost level.
</para>
<orderedlist numeration="arabic"><listitem><para>arabic number followed by a period: &#8216;<literal>1.</literal>&#8217;, &#8216;<literal>2.</literal>&#8217;,&#160;&#8230;
</para></listitem><listitem><para>lowercase letter inside parentheses: &#8216;<literal>(a)</literal>&#8217;, &#8216;<literal>(b)</literal>&#8217;&#160;&#8230;
</para></listitem><listitem><para>lowercase roman numeral followed by a period: &#8216;<literal>i.</literal>&#8217;, &#8216;<literal>ii.</literal>&#8217;,&#160;&#8230;
</para></listitem><listitem><para>uppercase letter followed by a period: &#8216;<literal>A.</literal>&#8217;, &#8216;<literal>B.</literal>&#8217;,&#160;&#8230;
</para></listitem></orderedlist>
<indexterm role="fn"><primary>\enumi</primary></indexterm>
<indexterm role="fn"><primary>\enumii</primary></indexterm>
<indexterm role="fn"><primary>\enumiii</primary></indexterm>
<indexterm role="fn"><primary>\enumiv</primary></indexterm>
<anchor id="enumerate-enumi"/><anchor id="enumerate-enumii"/><anchor id="enumerate-enumiii"/><anchor id="enumerate-enumiv"/><para>The <literal>enumerate</literal> environment uses the counters <literal>\enumi</literal> through
<literal>\enumiv</literal> (see <link linkend="Counters">Counters</link>).
</para>
<para>For other major &latex; labeled list environments, see
<link linkend="description">description</link> and <link linkend="itemize">itemize</link>.  For information about list layout
parameters, including the default values, and for information about
customizing list layout, see <link linkend="list">list</link>.  The package <filename>enumitem</filename> is
useful for customizing lists.
</para>
<indexterm role="fn"><primary>\labelenumi</primary></indexterm>
<indexterm role="fn"><primary>\labelenumii</primary></indexterm>
<indexterm role="fn"><primary>\labelenumiii</primary></indexterm>
<indexterm role="fn"><primary>\labelenumiv</primary></indexterm>
<anchor id="enumerate-labelenumi"/><anchor id="enumerate-labelenumii"/><anchor id="enumerate-labelenumiii"/><anchor id="enumerate-labelenumiv"/><para>To change the format of the label use <literal>\renewcommand</literal>
(see <link linkend="_005cnewcommand-_0026-_005crenewcommand">\newcommand &amp; \renewcommand</link>) on the commands <literal>\labelenumi</literal>
through <literal>\labelenumiv</literal>. For instance, this first level list will be
labelled with uppercase letters, in boldface, and without a trailing
period.
</para>
<indexterm role="fn"><primary>\Alph example</primary></indexterm>
<screen>\renewcommand{\labelenumi}{\textbf{\Alph{enumi}}}
\begin{enumerate}
  \item Shows as boldface A
  \item Shows as boldface B
\end{enumerate}
</screen>
<para>For a list of counter-labeling commands see <link linkend="_005calph-_005cAlph-_005carabic-_005croman-_005cRoman-_005cfnsymbol">\alph \Alph \arabic
\roman \Roman \fnsymbol</link>.
</para>

</sect1>
<sect1 label="8.8" id="eqnarray">
<title><literal>eqnarray</literal></title>

<indexterm role="fn"><primary>environment, <literal>eqnarray</literal></primary></indexterm>
<indexterm role="fn"><primary><literal>eqnarray</literal> environment</primary></indexterm>
        
<indexterm role="cp"><primary>equations, aligning</primary></indexterm>
<indexterm role="cp"><primary>aligning equations</primary></indexterm>

<indexterm role="cp"><primary>align environment, from <literal>amsmath</literal></primary></indexterm>
<indexterm role="cp"><primary>amsmath package, replacing <literal>eqnarray</literal></primary></indexterm>
<indexterm role="cp"><primary>Madsen, Lars</primary></indexterm>
<para>The <literal>eqnarray</literal> environment is obsolete.  It has infelicities,
including spacing that is inconsistent with other mathematics elements.
(See &#8220;Avoid eqnarray!&#8221; by Lars Madsen
<ulink url="http://tug.org/TUGboat/tb33-1/tb103madsen.pdf">http://tug.org/TUGboat/tb33-1/tb103madsen.pdf</ulink>).  New documents
should include the <filename>amsmath</filename> package and use the displayed
mathematics environments provided there, such as the <literal>align</literal>
environment.  We include a description only for completeness and for
working with old documents.
</para>
<para>Synopsis:
</para>
<screen>\begin{eqnarray} 
  <replaceable>first formula left</replaceable>  &amp;<replaceable>first formula middle</replaceable>  &amp;<replaceable>first formula right</replaceable> \\
  ...
\end{eqnarray}
</screen>
<para>or 
</para>
<screen>\begin{eqnarray*} 
  <replaceable>first formula left</replaceable>  &amp;<replaceable>first formula middle</replaceable>  &amp;<replaceable>first formula right</replaceable> \\
  ...
\end{eqnarray*}
</screen>
<indexterm role="fn"><primary>\\ (for <literal>eqnarray</literal>)</primary></indexterm>
<para>Display a sequence of equations or inequalities.  The left and right
sides are typeset in display mode, while the middle is typeset in text
mode.
</para>
<para>It is similar to a three-column <literal>array</literal> environment, with items
within a row separated by an ampersand&#160;(<literal>&amp;</literal>), and with rows
separated by double backslash&#160; <literal>\\</literal>).
<indexterm role="fn"><primary>\\* (for <literal>eqnarray</literal>)</primary></indexterm>
The starred form of line break (<literal>\\*</literal>) can also be used to separate
equations, and will disallow a page break there (see <link linkend="_005c_005c">\\</link>).
</para>
<indexterm role="fn"><primary>\nonumber</primary></indexterm>
<indexterm role="cp"><primary>equation numbers, omitting</primary></indexterm>
<para>The unstarred form <literal>eqnarray</literal> places an equation number on every
line (using the <literal>equation</literal> counter), unless that line contains a
<literal>\nonumber</literal> command.  The starred form <literal>eqnarray*</literal> omits
equation numbering, while otherwise being the same.
</para>
<indexterm role="fn"><primary>\lefteqn</primary></indexterm>
<para>The command <literal>\lefteqn</literal> is used for splitting long formulas across
lines. It typesets its argument in display style flush left in a box of
zero width.
</para>
<para>This example shows three lines.  The first two lines make an inequality,
while the third line has not entry on the left side.
</para>
<screen>\begin{eqnarray*}
  \lefteqn{x_1+x_2+\cdots+x_n}     \\
    &amp;\leq &amp;y_1+y_2+\cdots+y_n      \\
    &amp;=    &amp;z+y_3+\cdots+y_n
\end{eqnarray*}
</screen>

</sect1>
<sect1 label="8.9" id="equation">
<title><literal>equation</literal></title>

<indexterm role="fn"><primary>environment, <literal>equation</literal></primary></indexterm>
<indexterm role="fn"><primary><literal>equation</literal> environment</primary></indexterm>
        
<indexterm role="cp"><primary>equations, environment for</primary></indexterm>
<indexterm role="cp"><primary>formulas, environment for</primary></indexterm>

<para>Synopsis:
</para>
<screen>\begin{equation}
  <replaceable>mathematical text</replaceable>
\end{equation}
</screen>
<para>The same as a <literal>displaymath</literal> environment (see <link linkend="displaymath">displaymath</link>)
except that &latex; puts an equation number flush to the right margin.
The equation number is generated using the <literal>equation</literal> counter.
</para>
<para>You should have no blank lines between <literal>\begin{equation}</literal> and
<literal>\begin{equation}</literal>, or &latex; will tell you that there is a
missing dollar sign.
</para>
<indexterm role="cp"><primary>package, <literal>amsmath</literal></primary></indexterm>
<indexterm role="cp"><primary><literal>amsmath</literal> package</primary></indexterm>

<para>The package <filename>amsmath</filename> package has extensive displayed equation
facilities.  New documents should include this package.
</para>

</sect1>
<sect1 label="8.10" id="figure">
<title><literal>figure</literal></title>

<indexterm role="fn"><primary>environment, <literal>figure</literal></primary></indexterm>
<indexterm role="fn"><primary><literal>figure</literal> environment</primary></indexterm>
        
<indexterm role="cp"><primary>inserting figures</primary></indexterm>
<indexterm role="cp"><primary>figures, inserting</primary></indexterm>

<para>Synopsis:
</para>
<screen>\begin{figure}[<replaceable>placement</replaceable>]
  <replaceable>figure body</replaceable>
  \caption[<replaceable>loftitle</replaceable>]{<replaceable>title</replaceable>}  % optional
  \label{<replaceable>label}</replaceable>              % optional
\end{figure}
</screen>
<para>or:
</para>
<screen>\begin{figure*}[<replaceable>placement</replaceable>]
  <replaceable>figure body</replaceable>
  \caption[<replaceable>loftitle</replaceable>]{<replaceable>title</replaceable>}  % optional
  \label{<replaceable>label}</replaceable>              % optional  
\end{figure*}
</screen>
<para>Figures are for material that is not part of the normal text.  An
example is material that you cannot have split between two pages, such
as a graphic.  Because of this, &latex; does not typeset figures in
sequence with normal text but instead &#8220;floats&#8221; them to a convenient
place, such as the top of a following page (see <link linkend="Floats">Floats</link>).
</para>
<para>The <replaceable>figure body</replaceable> can consist of imported graphics
(see <link linkend="Graphics">Graphics</link>), or text, &latex; commands, etc.  It is typeset in a
<literal>parbox</literal> of width <literal>\textwidth</literal>.
</para>
<para>The possible values of <replaceable>placement</replaceable> are <literal>h</literal> for &#8216;<literal>here</literal>&#8217;,
<literal>t</literal> for &#8216;<literal>top</literal>&#8217;, <literal>b</literal> for &#8216;<literal>bottom</literal>&#8217;, and <literal>p</literal> for
&#8216;<literal>on a separate page of floats</literal>&#8217;.  For the effect of these options on
the float placement algorithm, see <link linkend="Floats">Floats</link>.
</para>
<para>The starred form <literal>figure*</literal> is used when a document is in
double-column mode (see <link linkend="_005ctwocolumn">\twocolumn</link>).  It produces a figure that
spans both columns, at the top of the page.  To add the possibility of
placing at a page bottom see the discussion of <replaceable>placement</replaceable> <literal>b</literal>
in <link linkend="Floats">Floats</link>.
</para>
<para>The label is optional; it is used for cross references (see <link linkend="Cross-references">Cross
references</link>).
<indexterm role="fn"><primary>\caption</primary></indexterm>
The optional <literal>\caption</literal> command specifies caption text for the
figure.  By default it is numbered.  If <replaceable>loftitle</replaceable> is present, it is
used in the list of figures instead of <replaceable>title</replaceable> (see <link linkend="Table-of-contents-etc_002e">Table of
contents etc.</link>).
</para>
<para>This example makes a figure out of a graphic.  &latex; will place that
graphic and its caption at the top of a page or, if it is pushed to the
end of the document, on a page of floats.
</para>
<screen>\usepackage{graphicx}  % in preamble
  ...
\begin{figure}[t]
  \centering
  \includegraphics[width=0.5\textwidth]{CTANlion.png}
  \caption{The CTAN lion, by Duane Bibby}
\end{figure}
</screen>

</sect1>
<sect1 label="8.11" id="filecontents">
<title><literal>filecontents</literal>: Write an external file</title>

<indexterm role="fn"><primary>environment, <literal>filecontents</literal></primary></indexterm>
<indexterm role="fn"><primary><literal>filecontents</literal> environment</primary></indexterm>
        
<indexterm role="fn"><primary>environment, <literal>filecontents*</literal></primary></indexterm>
<indexterm role="fn"><primary><literal>filecontents*</literal> environment</primary></indexterm>
        
<indexterm role="cp"><primary>external files, writing</primary></indexterm>
<indexterm role="cp"><primary>writing external files</primary></indexterm>

<para>Synopsis:
</para>
<screen>\begin{filecontents}{<replaceable>filename</replaceable>}
  <replaceable>text</replaceable>
\end{filecontents}
</screen>
<para>or
</para>
<screen>\begin{filecontents*}{<replaceable>filename</replaceable>}
  <replaceable>text</replaceable>
\end{filecontents*}
</screen>
<para>Create a file named <replaceable>filename</replaceable> and fill it with <replaceable>text</replaceable>.  The
unstarred version of the environment <literal>filecontents</literal> prefixes the
content of the created file with a header; see the example below.  The
starred version <literal>filecontents*</literal> does not include the header.
</para>
<para>This environment can be used anywhere in the preamble, although it often
appears before the <literal>\documentclass</literal> command.  It is typically used
when a source file requires a nonstandard style or class file.  The
environment will write that file to the directory containing the source
and thus make the source file self-contained.  Another use is to include
<literal>bib</literal> references in the file, again to make it self-contained.
</para>
<para>The environment checks whether a file of that name already exists and if
so, does not do anything.  There is a <filename>filecontents</filename> package that
redefines the <literal>filecontents</literal> environment so that instead of doing
nothing in that case, it will overwrite the existing file.
</para>
<para>For example, this document
</para>
<screen>\documentclass{article}
\begin{filecontents}{JH.sty}
\newcommand{\myname}{Jim Hef{}feron}
\end{filecontents}
\usepackage{JH}
\begin{document}
Article by \myname.
\end{document}
</screen>
<para>produces this file <filename>JH.sty</filename>.
</para>
<screen>%% LaTeX2e file `JH.sty'
%% generated by the `filecontents' environment
%% from source `test' on 2015/10/12.
%%
\newcommand{\myname}{Jim Hef{}feron}
</screen>

</sect1>
<sect1 label="8.12" id="flushleft">
<title><literal>flushleft</literal></title>

<indexterm role="fn"><primary>environment, <literal>flushleft</literal></primary></indexterm>
<indexterm role="fn"><primary><literal>flushleft</literal> environment</primary></indexterm>
        
<indexterm role="cp"><primary>left-justifying text, environment for</primary></indexterm>
<indexterm role="cp"><primary>ragged right text, environment for</primary></indexterm>

<para>Synopsis:
</para>
<screen>\begin{flushleft}
  <replaceable>line1</replaceable> \\
  <replaceable>line2</replaceable> \\
  ...
\end{flushleft}
</screen>
<indexterm role="fn"><primary>\\ for <literal>flushleft</literal></primary></indexterm>
<para>An environment that creates a paragraph whose lines are flush to the
left-hand margin, and ragged right. If you have lines that are too long
then &latex; will linebreak them in a way that avoids hyphenation and
stretching or shrinking spaces.  To force a new line use a double
backslash, <literal>\\</literal>.  For the declaration form
see&#160;<link linkend="_005craggedright">\raggedright</link>.
</para>
<para>This creates a box of text that is at most 3 inches wide, with the text
flush left and ragged right.
</para>
<screen>\noindent\begin{minipage}{3in}
\begin{flushleft}
  A long sentence that will be broken by \LaTeX{}
    at a convenient spot. \\
  And, a fresh line forced by the double backslash.
\end{flushleft}
\end{minipage}
</screen>



<sect2 label="8.12.1" id="_005craggedright">
<title><literal>\raggedright</literal></title>

<indexterm role="fn"><primary>\raggedright</primary></indexterm>
<indexterm role="cp"><primary>ragged right text</primary></indexterm>
<indexterm role="cp"><primary>left-justifying text</primary></indexterm>
<indexterm role="cp"><primary>justification, ragged right</primary></indexterm>

<para>Synopses:
</para>
<screen>{\raggedright  ... }
</screen>
<para>or
</para>
<screen>\begin{<replaceable>environment</replaceable>} \raggedright
  ...
\end{<replaceable>environment</replaceable>}
</screen>
<para>A declaration which causes lines to be flush to the left margin and
ragged right. It can be used inside an environment such as <literal>quote</literal>
or in a <literal>parbox</literal>.  For the environment form
see&#160;<link linkend="flushleft">flushleft</link>.
</para>
<para>Unlike the <literal>flushleft</literal> environment, the <literal>\raggedright</literal>
command does not start a new paragraph; it only changes how &latex;
formats paragraph units.  To affect a paragraph unit&#8217;s format, the
scope of the declaration must contain the blank line or <literal>\end</literal>
command that ends the paragraph unit.
</para>
<para>Here <literal>\raggedright</literal> in each second column keeps &latex; from doing
very awkward typesetting to fit the text into the narrow column.  Note
that <literal>\raggedright</literal> is inside the curly braces <literal>{...}</literal> to
delimit its effect.
</para>
<screen>\begin{tabular}{rp{2in}}
  Team alpha  &amp;{\raggedright This team does all the real work.} \\
  Team beta   &amp;{\raggedright This team ensures that the water
                cooler is never empty.}                         \\
\end{tabular}
</screen>

</sect2>
</sect1>
<sect1 label="8.13" id="flushright">
<title><literal>flushright</literal></title>

<indexterm role="fn"><primary>environment, <literal>flushright</literal></primary></indexterm>
<indexterm role="fn"><primary><literal>flushright</literal> environment</primary></indexterm>
        
<indexterm role="cp"><primary>ragged left text, environment for</primary></indexterm>
<indexterm role="cp"><primary>right-justifying text, environment for</primary></indexterm>

<screen>\begin{flushright}
  <replaceable>line1</replaceable> \\
  <replaceable>line2</replaceable> \\
  ...
\end{flushright}
</screen>
<indexterm role="fn"><primary>\\ (for <literal>flushright</literal>)</primary></indexterm>
<para>An environment that creates a paragraph whose lines are flush to the
right-hand margin and ragged left. If you have lines that are too long
to fit the margins then &latex; will linebreak them in a way that
avoids hyphenation and stretching or shrinking spaces.  To force a new
line use a double backslash, <literal>\\</literal>.  For the declaration form
see&#160;<link linkend="_005craggedleft">\raggedleft</link>.
</para>
<para>For an example related to this environment, see&#160;<link linkend="flushleft">flushleft</link>.
</para>


<sect2 label="8.13.1" id="_005craggedleft">
<title><literal>\raggedleft</literal></title>

<indexterm role="fn"><primary>\raggedleft</primary></indexterm>
<indexterm role="cp"><primary>ragged left text</primary></indexterm>
<indexterm role="cp"><primary>justification, ragged left</primary></indexterm>
<indexterm role="cp"><primary>right-justifying text</primary></indexterm>

<para>Synopses:
</para>
<screen>{\raggedleft  ... }
</screen>
<para>or
</para>
<screen>\begin{<replaceable>environment</replaceable>} \raggedleft
  ...
\end{<replaceable>environment</replaceable>}
</screen>
<para>A declaration which causes lines to be flush to the right margin and
ragged left. It can be used inside an environment such as <literal>quote</literal>
or in a <literal>parbox</literal>.  For the environment form
see&#160;<link linkend="flushright">flushright</link>.
</para>
<para>Unlike the <literal>flushright</literal> environment, the <literal>\raggedleft</literal>
command does not start a new paragraph; it only changes how &latex;
formats paragraph units.  To affect a paragraph unit&#8217;s format, the
scope of the declaration must contain the blank line or <literal>\end</literal>
command that ends the paragraph unit.
</para>
<para>For an example related to this environment, see&#160;<link linkend="_005craggedright">\raggedright</link>.
</para>

</sect2>
</sect1>
<sect1 label="8.14" id="itemize">
<title><literal>itemize</literal></title>

<indexterm role="fn"><primary>environment, <literal>itemize</literal></primary></indexterm>
<indexterm role="fn"><primary><literal>itemize</literal> environment</primary></indexterm>
        
<indexterm role="fn"><primary>\item</primary></indexterm>
<indexterm role="cp"><primary>lists of items</primary></indexterm>
<indexterm role="cp"><primary>unordered lists</primary></indexterm>
<indexterm role="cp"><primary>bulleted lists</primary></indexterm>
<indexterm role="cp"><primary>bullet lists</primary></indexterm>

<para>Synopsis:
</para>
<screen>\begin{itemize}
  \item[<replaceable>optional label of first item</replaceable>] <replaceable>text of first item</replaceable>
  \item[<replaceable>optional label of second item</replaceable>] <replaceable>text of second item</replaceable>
  ...
\end{itemize}
</screen>
<para>Produce a list that is unordered, sometimes called a bullet list.  The
environment must have at least one <literal>\item</literal>; having none causes the
&latex; error &#8216;<literal>Something's wrong--perhaps a missing \item</literal>&#8217;.
</para>
<para>This gives a two-item list.
</para>
<screen>\begin{itemize}
 \item Pencil and watercolor sketch by Cassandra
 \item Rice portrait 
\end{itemize}
</screen>
<para>As a top-level list each label would come out as a bullet, &#8226;.
The format of the labeling depends on the nesting level; see below.
</para>
<indexterm role="fn"><primary>\item</primary></indexterm>
<para>Start list items with the <literal>\item</literal> command (see <link linkend="_005citem">\item</link>).  If you
give <literal>\item</literal> an optional argument by following it with square
brackets, as in <literal>\item[Optional label]</literal>, then by default it will
appear in bold and be flush right, so it could extend into the left
margin.  For labels that are flush left see the <link linkend="description">description</link>
environment. Following the <literal>\item</literal> is optional text, which may
contain multiple paragraphs.
</para>
<indexterm role="fn"><primary>\labelitemi</primary></indexterm>
<indexterm role="fn"><primary>\labelitemii</primary></indexterm>
<indexterm role="fn"><primary>\labelitemiii</primary></indexterm>
<indexterm role="fn"><primary>\labelitemiv</primary></indexterm>
<anchor id="itemize-labelitemi"/><anchor id="itemize-labelitemii"/><anchor id="itemize-labelitemiii"/><anchor id="itemize-labelitemiv"/><para>Itemized lists can be nested within one another, up to four levels deep.
They can also be nested within other paragraph-making environments, such
as <literal>enumerate</literal> (see <link linkend="enumerate">enumerate</link>).  The <literal>itemize</literal> environment
uses the commands <literal>\labelitemi</literal> through <literal>\labelitemiv</literal> to
produce the default label (this also uses the convention of lowercase
roman numerals at the end of the command names that signify the nesting
level).  These are the default marks at each level.
</para>
<orderedlist numeration="arabic"><listitem><para>&#8226; (bullet, from <literal>\textbullet</literal>)
</para></listitem><listitem><para><emphasis role="bold">--<!-- /@w --></emphasis> (bold en-dash, from <literal>\normalfont\bfseries\textendash</literal>)
</para></listitem><listitem><para>* (asterisk, from <literal>\textasteriskcentered</literal>)
</para></listitem><listitem><para>. (centered dot, rendered here as a period, from <literal>\textperiodcentered</literal>)
</para></listitem></orderedlist>
<para>Change the labels with <literal>\renewcommand</literal>.  For instance, this makes
the first level use diamonds.
</para>
<screen>\renewcommand{\labelitemi}{$\diamond$}
</screen>
<indexterm role="fn"><primary>\leftmargin</primary></indexterm>
<indexterm role="fn"><primary>\leftmargini</primary></indexterm>
<indexterm role="fn"><primary>\leftmarginii</primary></indexterm>
<indexterm role="fn"><primary>\leftmarginiii</primary></indexterm>
<indexterm role="fn"><primary>\leftmarginiv</primary></indexterm>
<indexterm role="fn"><primary>\leftmarginv</primary></indexterm>
<indexterm role="fn"><primary>\leftmarginvi</primary></indexterm>
<anchor id="itemize-leftmargin"/><anchor id="itemize-leftmargini"/><anchor id="itemize-leftmarginii"/><anchor id="itemize-leftmarginiii"/><anchor id="itemize-leftmarginiv"/><anchor id="itemize-leftmarginv"/><anchor id="itemize-leftmarginvi"/><para>The distance between the left margin of the enclosing environment and
the left margin of the <literal>itemize</literal> list is determined by the
parameters <literal>\leftmargini</literal> through <literal>\leftmarginvi</literal>.  (Note the
convention of using lowercase roman numerals a the end of the command
name to denote the nesting level.)  The defaults are: <literal>2.5em</literal> in
level 1 (<literal>2em</literal> in two-column mode), <literal>2.2em</literal> in level 2,
<literal>1.87em</literal> in level 3, and <literal>1.7em</literal> in level 4, with smaller
values for more deeply nested levels.
</para>
<para>For other major &latex; labeled list environments, see
<link linkend="description">description</link> and <link linkend="enumerate">enumerate</link>.  For information about list
layout parameters, including the default values, and for information
about customizing list layout, see <link linkend="list">list</link>.  The package
<filename>enumitem</filename> is useful for customizing lists.
</para>
<para>This example greatly reduces the margin space for outermost itemized
lists.
</para>
<screen>\setlength{\leftmargini}{1.25em} % default 2.5em
</screen>
<indexterm role="fn"><primary>\parskip example</primary></indexterm>
<para>Especially for lists with short items, it may be desirable to elide
space between items.  Here is an example defining an <literal>itemize*</literal>
environment with no extra spacing between items, or between paragraphs
within a single item (<literal>\parskip</literal> is not list-specific,
see <link linkend="_005cparindent-_0026-_005cparskip">\parindent &amp; \parskip</link>):
</para>
<screen>\newenvironment{itemize*}%
  {\begin{itemize}%
    \setlength{\itemsep}{0pt}%
    \setlength{\parsep}{0pt}}%
    \setlength{\parskip}{0pt}}%
  {\end{itemize}}
</screen>

</sect1>
<sect1 label="8.15" id="letter">
<title><literal>letter</literal> environment: writing letters</title>

<indexterm role="fn"><primary>environment, <literal>letter</literal></primary></indexterm>
<indexterm role="fn"><primary><literal>letter</literal> environment</primary></indexterm>
        

<para>This environment is used for creating letters.  See <link linkend="Letters">Letters</link>.
</para>

</sect1>
<sect1 label="8.16" id="list">
<title><literal>list</literal></title>

<indexterm role="fn"><primary>environment, <literal>list</literal></primary></indexterm>
<indexterm role="fn"><primary><literal>list</literal> environment</primary></indexterm>
        
<indexterm role="cp"><primary>lists of items, generic</primary></indexterm>

<para>Synopsis:
</para>
<screen>\begin{list}{<replaceable>labeling</replaceable>}{<replaceable>spacing</replaceable>}
  \item[<replaceable>optional label of first item</replaceable>] <replaceable>text of first item</replaceable>
  \item[<replaceable>optional label of second item</replaceable>] <replaceable>text of second item</replaceable>
  ...
\end{list}
</screen>
<para>An environment for constructing lists.
</para>
<para>Note that this environment does not typically appear in the document
body.  Most lists created by &latex; authors are the ones that come
standard: the <literal>description</literal>, <literal>enumerate</literal>, and <literal>itemize</literal>
environments (see <link linkend="description">description</link>, <link linkend="enumerate">enumerate</link>, and <link linkend="itemize">itemize</link>).
</para>
<para>Instead, the <literal>list</literal> environment is most often used in macros.  For
example, many standard &latex; environments that do not immediately
appear to be lists are in fact constructed using <literal>list</literal>, including
<literal>quotation</literal>, <literal>quote</literal>, and <literal>center</literal> (see <link linkend="quotation-_0026-quote">quotation &amp;
quote</link>, see <link linkend="center">center</link>).
</para>
<para>This uses the <literal>list</literal> environment to define a new custom
environment.
</para>
<screen>\newcounter{namedlistcounter}  % number the items
\newenvironment{named}
  {\begin{list}
     {Item~\Roman{namedlistcounter}.} % labeling 
     {\usecounter{namedlistcounter}   % set counter
      \setlength{\leftmargin}{3.5em}} % set spacing 
  }
  {\end{list}}

\begin{named}
  \item Shows as ``Item~I.''
  \item[Special label.] Shows as ``Special label.''
  \item Shows as ``Item~II.''
\end{named}
</screen>
<para>The mandatory first argument <replaceable>labeling</replaceable> specifies the default
labeling of list items.  It can contain text and &latex; commands, as
above where it contains both &#8216;<literal>Item</literal>&#8217; and &#8216;<literal>\Roman{...}</literal>&#8217;.
&latex; forms the label by putting the <replaceable>labeling</replaceable> argument in a box
of width <literal>\labelwidth</literal>.  If the label is wider than that, the
additional material extends to the right.  When making an instance of a
list you can override the default labeling by giving <literal>\item</literal> an
optional argument by including square braces and the text, as in the
above <literal>\item[Special label.]</literal>; see <link linkend="_005citem">\item</link>.
</para>
<para>The mandatory second argument <replaceable>spacing</replaceable> has a list of commands.
This list can be empty.  A command that can go in here is
<literal>\usecounter{<replaceable>countername</replaceable>}</literal> (see <link linkend="_005cusecounter">\usecounter</link>).  Use this
to tell &latex; to number the items using the given counter.  The
counter will be reset to zero each time &latex; enters the environment,
and the counter is incremented by one each time &latex; encounters an
<literal>\item</literal> that does not have an optional argument.
</para>
<indexterm role="fn"><primary>\makelabel</primary></indexterm>
<anchor id="list-makelabel"/><para>Another command that can go in <replaceable>spacing</replaceable> is
<literal>\makelabel</literal>, which constructs the label box.  By default it puts
the contents flush right.  Its only argument is the label, which it
typesets in LR mode (see <link linkend="Modes">Modes</link>). One example of changing its
definition is that to the above <literal>named</literal> example, before the
definition of the environment add
<literal>\newcommand{\namedmakelabel}[1]{\textsc{#1}}</literal>, and between
the <literal>\setlength</literal> command and the parenthesis that closes the
<replaceable>spacing</replaceable> argument also add <literal>\let\makelabel\namedmakelabel</literal>.
Then the items will be typeset in small caps.  Similarly, changing the
second code line to <literal>\let\makelabel\fbox</literal> puts the labels inside a
framed box.  Another example of the <literal>\makelabel</literal> command is below,
in the definition of the <literal>redlabel</literal> environment.
</para>
<para>Also often in <replaceable>spacing</replaceable> are commands to redefine the spacing for the
list.  Below are the spacing parameters with their default values.
(Default values for derived environments such as <literal>itemize</literal> can be
different than the values shown here.) See also the figure that follows
the list.  Each is a length (see <link linkend="Lengths">Lengths</link>).  The vertical spaces are
normally rubber lengths, with <literal>plus</literal> and <literal>minus</literal> components,
to give &tex; flexibility in setting the page.  Change each with a
command such as <literal>\setlength{itemsep}{2pt plus1pt minus1pt}</literal>.
For some effects these lengths should be zero or negative.
</para>
<variablelist><varlistentry><term><indexterm role="fn"><primary>\itemindent</primary></indexterm><literal>\itemindent</literal>
</term><listitem><anchor id="list-itemindent"/><para>Extra horizontal space indentation, beyond <literal>leftmargin</literal>, of the
first line each item.  Its default value is <literal>0pt</literal>.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\itemsep</primary></indexterm><literal>\itemsep</literal>
</term><listitem><anchor id="list-itemsep"/><para>Vertical space between items, beyond the <literal>\parsep</literal>.  The defaults
for the first three levels in &latex;&#8217;s &#8216;<literal>article</literal>&#8217;, &#8216;<literal>book</literal>&#8217;,
and &#8216;<literal>report</literal>&#8217; classes at 10 point size are: <literal>4pt plus2pt
minus1pt</literal>, <literal>\parsep</literal> (that is, <literal>2pt plus1pt minus1pt</literal>), and
<literal>\topsep</literal> (that is, <literal>2pt plus1pt minus1pt</literal>).  The defaults at
11 point are: <literal>4.5pt plus2pt minus1pt</literal>, <literal>\parsep</literal> (that is,
<literal>2pt plus1pt minus1pt</literal>), and <literal>topsep</literal> (that is, <literal>2pt
plus1pt minus1pt</literal>).  The defaults at 12 point are: <literal>5pt plus2.5pt
minus1pt</literal>, <literal>\parsep</literal> (that is, <literal>2.5pt plus1pt minus1pt</literal>), and
<literal>\topsep</literal> (that is, <literal>2.5pt plus1pt minus1pt</literal>).
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\labelsep</primary></indexterm><literal>\labelsep</literal>
</term><listitem><anchor id="list-labelsep"/><para>Horizontal space between the label and text of an item.
The default for &latex;&#8217;s &#8216;<literal>article</literal>&#8217;, &#8216;<literal>book</literal>&#8217;,
and &#8216;<literal>report</literal>&#8217; classes is <literal>0.5em</literal>.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\labelwidth</primary></indexterm><literal>\labelwidth</literal>
</term><listitem><anchor id="list-labelwidth"/><para>Horizontal width.  The box containing the label is nominally this wide.
If <literal>\makelabel</literal> returns text that is wider than this then the first
line of the item will be indented to make room for this extra material.
If <literal>\makelabel</literal> returns text of width less than or equal to
<literal>\labelwidth</literal> then &latex;&#8217;s default is that the label is typeset
flush right in a box of this width.
</para>
<para>The left edge of the label box is
<literal>\leftmargin</literal>+<literal>\itemindent</literal>-<literal>\labelsep</literal>-<literal>\labelwidth</literal>
from the left margin of the enclosing environment.
</para>
<para>The default for &latex;&#8217;s &#8216;<literal>article</literal>&#8217;, &#8216;<literal>book</literal>&#8217;, and
&#8216;<literal>report</literal>&#8217; classes at the top level is
<literal>\leftmargini</literal>-<literal>\labelsep</literal>, (which is <literal>2em</literal> in one column
mode and <literal>1.5em</literal> in two column mode).  At the second level it is
<literal>\leftmarginii</literal>-<literal>\labelsep</literal>, and at the third level it is
<literal>\leftmarginiii</literal>-<literal>\labelsep</literal>.  These definitions make the
label&#8217;s left edge coincide with the left margin of the enclosing
environment.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\leftmargin</primary></indexterm><literal>\leftmargin</literal>
</term><listitem><anchor id="list-leftmargin"/><para>Horizontal space between the left margin of the enclosing environment
(or the left margin of the page if this is a top-level list), and the
left margin of this list.  It must be non-negative.
</para>
<para>In the standard &latex; document classes, a top-level list has this set
to the value of <literal>\leftmargini</literal>, while a list that is nested inside
a top-level list has this margin set to <literal>\leftmarginii</literal>.  More
deeply nested lists get the values of <literal>\leftmarginiii</literal> through
<literal>\leftmarginvi</literal>.  (Nesting greater than level five generates the
error message &#8216;<literal>Too deeply nested</literal>&#8217;.)
</para>
<para>The defaults for the first three levels in &latex;&#8217;s &#8216;<literal>article</literal>&#8217;,
&#8216;<literal>book</literal>&#8217;, and &#8216;<literal>report</literal>&#8217; classes are: <literal>\leftmargini</literal> is
<literal>2.5em</literal> (in two column mode, <literal>2em</literal>), <literal>\leftmarginii</literal> is
<literal>2.2em</literal>, and <literal>\leftmarginiii</literal> is <literal>1.87em</literal>.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\listparindent</primary></indexterm><literal>\listparindent</literal>
</term><listitem><anchor id="list-listparindent"/><para>Horizontal space of additional line indentation, beyond
<literal>\leftmargin</literal>, for second and subsequent paragraphs within a list
item.  A negative value makes this an &#8220;outdent&#8221;.  Its default value
is <literal>0pt</literal>.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\parsep</primary></indexterm><literal>\parsep</literal>
</term><listitem><anchor id="list-parsep"/><para>Vertical space between paragraphs within an item.  In the &#8216;<literal>book</literal>&#8217;
and &#8216;<literal>article</literal>&#8217; classes The defaults for the first three levels in
&latex;&#8217;s &#8216;<literal>article</literal>&#8217;, &#8216;<literal>book</literal>&#8217;, and &#8216;<literal>report</literal>&#8217; classes at 10
point size are: <literal>4pt plus2pt minus1pt</literal>, <literal>2pt plus1pt
minus1pt</literal>, and <literal>0pt</literal>.  The defaults at 11 point size are:
<literal>4.5pt plus2pt minus1pt</literal>, <literal>2pt plus1pt minus1pt</literal>, and
<literal>0pt</literal>.  The defaults at 12 point size are: <literal>5pt plus2.5pt
minus1pt</literal>, <literal>2.5pt plus1pt minus1pt</literal>, and <literal>0pt</literal>.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\partopsep</primary></indexterm><literal>\partopsep</literal>
</term><listitem><anchor id="list-partopsep"/><para>Vertical space added, beyond <literal>\topsep</literal>+<literal>\parskip</literal>, to the top
and bottom of the entire environment if the list instance is preceded by
a blank line.  (A blank line in the &latex; source before the list
changes spacing at both the top and bottom of the list; whether the line
following the list is blank does not matter.)
</para>
<para>The defaults for the first three levels in &latex;&#8217;s &#8216;<literal>article</literal>&#8217;,
&#8216;<literal>book</literal>&#8217;, and &#8216;<literal>report</literal>&#8217; classes at 10 point size are: <literal>2pt
plus1 minus1pt</literal>, <literal>2pt plus1pt minus1pt</literal>, and <literal>1pt plus0pt
minus1pt</literal>.  The defaults at 11 point are: <literal>3pt plus1pt minus1pt</literal>,
<literal>3pt plus1pt minus1pt</literal>, and <literal>1pt plus0pt minus1pt</literal>).  The
defaults at 12 point are: <literal>3pt plus2pt minus3pt</literal>, <literal>3pt plus2pt
minus2pt</literal>, and <literal>1pt plus0pt minus1pt</literal>.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\rightmargin</primary></indexterm><literal>\rightmargin</literal>
</term><listitem><anchor id="list-rightmargin"/><para>Horizontal space between the right margin of the list and the right
margin of the enclosing environment.  Its default value is <literal>0pt</literal>.
It must be non-negative.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\topsep</primary></indexterm><literal>\topsep</literal>
</term><listitem><anchor id="list-topsep"/><para>Vertical space added to both the top and bottom of the list, in addition
to <literal>\parskip</literal> (see <link linkend="_005cparindent-_0026-_005cparskip">\parindent &amp; \parskip</link>).  The defaults for
the first three levels in &latex;&#8217;s &#8216;<literal>article</literal>&#8217;, &#8216;<literal>book</literal>&#8217;, and
&#8216;<literal>report</literal>&#8217; classes at 10 point size are: <literal>8pt plus2pt minus4pt</literal>,
<literal>4pt plus2pt minus1pt</literal>, and <literal>2pt plus1pt minus1pt</literal>.  The
defaults at 11 point are: <literal>9pt plus3pt minus5pt</literal>, <literal>4.5pt
plus2pt minus1pt</literal>, and <literal>2pt plus1pt minus1pt</literal>.  The defaults at 12
point are: <literal>10pt plus4pt minus6pt</literal>, <literal>5pt plus2.5pt minus1pt</literal>,
and <literal>2.5pt plus1pt minus1pt</literal>.
</para>
</listitem></varlistentry></variablelist>
<para>This shows the horizontal and vertical distances.
</para>
<informalfigure><mediaobject><imageobject><imagedata fileref="latex2e-figures/list.eps" format="EPS"></imagedata></imageobject><imageobject><imagedata fileref="latex2e-figures/list.pdf" format="PDF"></imagedata></imageobject><imageobject><imagedata fileref="latex2e-figures/list.png" format="PNG"></imagedata></imageobject><textobject><literallayout>.......................................................................
..............Surrounding environment, preceding material.............. -
                                                                        | v0 
 |---h0--|-h1-|                                                         |
 ..Label..    .....First item.....................................      -
      ............................................................
      ............................................................      -
                                                                        | v1
           ........First item, second paragraph ..................      -
      |-h2-|                                                            | v2
                                                                        |
 ..Label..    .....Second item....................................      -
      ............................................................
      ............................................................      -
|-h3--|-h4-|                                                     |-h5-| |v3
                                                                        |
..............Surrounding environment, following material.............. -
.......................................................................</literallayout></textobject></mediaobject></informalfigure>

<para>The lengths shown are listed below.  The key relationship is that the
right edge of the bracket for <replaceable>h1</replaceable> equals the right edge of the
bracket for <replaceable>h4</replaceable>, so that the left edge of the label box is at
<replaceable>h3</replaceable>+<replaceable>h4</replaceable>-(<replaceable>h0</replaceable>+<replaceable>h1</replaceable>).
</para>
<variablelist><varlistentry><term><replaceable>v0</replaceable>
</term><listitem><para><inlineequation><mathphrase><literal>\topsep</literal> + <literal>\parskip</literal></mathphrase></inlineequation> if
the list environment does not start a new paragraph, and
<literal>\topsep</literal>+<literal>\parskip</literal>+<literal>\partopsep</literal> if it does
</para>
</listitem></varlistentry><varlistentry><term><replaceable>v1</replaceable>
</term><listitem><para><literal>\parsep</literal>
</para>
</listitem></varlistentry><varlistentry><term><replaceable>v2</replaceable>
</term><listitem><para><literal>\itemsep</literal>+<literal>\parsep</literal>
</para>
</listitem></varlistentry><varlistentry><term><replaceable>v3</replaceable>
</term><listitem><para>Same as <replaceable>v0</replaceable>.  (This space is affected by whether a blank line
appears in the source above the environment; whether a blank line
appears in the source below the environment does not matter.)
</para>
</listitem></varlistentry><varlistentry><term><replaceable>h0</replaceable>
</term><listitem><para><literal>\labelwidth</literal>
</para>
</listitem></varlistentry><varlistentry><term><replaceable>h1</replaceable>
</term><listitem><para><literal>\labelsep</literal>
</para>
</listitem></varlistentry><varlistentry><term><replaceable>h2</replaceable>
</term><listitem><para><literal>\listparindent</literal>
</para>
</listitem></varlistentry><varlistentry><term><replaceable>h3</replaceable>
</term><listitem><para><literal>\leftmargin</literal>
</para>
</listitem></varlistentry><varlistentry><term><replaceable>h4</replaceable>
</term><listitem><para><literal>\itemindent</literal>
</para>
</listitem></varlistentry><varlistentry><term><replaceable>h5</replaceable>
</term><listitem><para><literal>\rightmargin</literal>
</para>
</listitem></varlistentry></variablelist>
<para>The list&#8217;s left and right margins, shown above as <replaceable>h3</replaceable> and <replaceable>h5</replaceable>,
are with respect to the ones provided by the surrounding environment, or
with respect to the page margins for a top-level list.  The line width
used for typesetting the list items is <literal>\linewidth</literal> (see <link linkend="Page-layout-parameters">Page
layout parameters</link>).  For instance, set the list&#8217;s left margin to be one
quarter of the distance between the left and right margins of the
enclosing environment with
<literal>\setlength{\leftmargin}{0.25\linewidth}</literal>.
</para>
<para>Page breaking in a list structure is controlled by the three
parameters below. For each, the &latex; default is
<literal>-\@lowpenalty</literal>, that is, <literal>-51</literal>.  Because it is negative,
it somewhat encourages a page break at each spot.  Change it with,
e.g., <literal>\@beginparpenalty=9999</literal>; a value of 10000 prohibits a
page break.
</para>
<variablelist><varlistentry><term><indexterm role="fn"><primary>\@beginparpenalty</primary></indexterm><literal>\@beginparpenalty</literal>
</term><listitem><anchor id="list-beginparpenalty"/><para>The page breaking penalty for breaking before the list (default <literal>-51</literal>).
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\@itempenalty</primary></indexterm><literal>\@itempenalty</literal>
</term><listitem><anchor id="list-itempenalty"/><para>The page breaking penalty for breaking before a list item (default <literal>-51</literal>).
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\@endparpenalty</primary></indexterm><literal>\@endparpenalty</literal>
</term><listitem><anchor id="list-endparpenalty"/><para>The page breaking penalty for breaking after a list (default <literal>-51</literal>).
</para>
</listitem></varlistentry></variablelist>
<indexterm role="cp"><primary>package, <literal>enumitem</literal></primary></indexterm>
<indexterm role="cp"><primary><literal>enumitem</literal> package</primary></indexterm>

<para>The package <literal>enumitem</literal> is useful for customizing lists.
</para>
<para>This example has the labels in red.  They are numbered, and the left
edge of the label lines up with the left edge of the item text.
See <link linkend="_005cusecounter">\usecounter</link>.
</para>
<screen>\usepackage{color}
\newcounter{cnt}  
\newcommand{\makeredlabel}[1]{\textcolor{red}{#1.}}
\newenvironment{redlabel}
  {\begin{list}
    {\arabic{cnt}}
    {\usecounter{cnt}
     \setlength{\labelwidth}{0em}
     \setlength{\labelsep}{0.5em}
     \setlength{\leftmargin}{1.5em}
     \setlength{\itemindent}{0.5em} % equals \labelwidth+\labelsep
     \let\makelabel=\makeredlabel
    }
  }
{\end{list} }
</screen>


<sect2 label="8.16.1" id="_005citem">
<title><literal>\item</literal>: An entry in a list</title>

<para>Synopsis:
</para>
<screen>\item text of item
</screen>
<para>or
</para>
<screen>\item[<replaceable>optional-label</replaceable>] text of item
</screen>
<para>An entry in a list.  The entries are prefixed by a label, whose default
depends on the list type.
</para>
<para>Because the optional label is surrounded by square brackets
&#8216;<literal>[...]</literal>&#8217;, if you have an item whose text starts with &#8216;<literal>[</literal>&#8217;, you
have to to hide the bracket inside curly braces, as in: <literal>\item
{[} is an open square bracket</literal>; otherwise, &latex; will think it
marks the start of an optional label.
</para>
<para>Similarly, if the item does have the optional label and you need a
close square bracket inside that label, you must hide it in the same
way: <literal>\item[Close square bracket, {]}]</literal>.  See <link linkend="LaTeX-command-syntax">&latex;
command syntax</link>.
</para>
<para>In this example the enumerate list has two items that use the default
label and one that uses the optional label.
</para>
<screen>\begin{enumerate}
  \item Moe
  \item[sometimes] Shemp
  \item Larry
\end{enumerate}
</screen>
<para>The first item is labelled &#8216;<literal>1.</literal>&#8217;, the second item is labelled
&#8216;<literal>sometimes</literal>&#8217;, and the third item is labelled &#8216;<literal>2.</literal>&#8217;.  Because
of the optional label in the second item, the third item is not
labelled&#160;&#8216;<literal>3.</literal>&#8217;.
</para>   

</sect2>
<sect2 label="8.16.2" id="trivlist">
<title><literal>trivlist</literal>: A restricted form of <literal>list</literal></title>

<para>Synopsis:
</para>
<screen>\begin{trivlist}
  ...
\end{trivlist}
</screen>
<para>A restricted version of the list environment, in which margins are not
indented and an <literal>\item</literal> without an optional argument produces no
text.  It is most often used in macros, to define an environment where
the <literal>\item</literal> command as part of the environment&#8217;s definition.  For
instance, the <literal>center</literal> environment is defined essentially like
this:
</para>
<screen>\newenvironment{center}
  {\begin{trivlist}\centering\item\relax}
  {\end{trivlist}}
</screen>
<para>Using <literal>trivlist</literal> in this way allows the macro to inherit some
common code: combining vertical space of two adjacent environments;
detecting whether the text following the environment should be
considered a new paragraph or a continuation of the previous one;
adjusting the left and right margins for possible nested list
environments.
</para>
<para>Specifically, <literal>trivlist</literal> uses the current values of the list
parameters (see <link linkend="list">list</link>), except that <literal>\parsep</literal> is set to the
value of <literal>\parskip</literal>, and <literal>\leftmargin</literal>, <literal>\labelwidth</literal>,
and <literal>\itemindent</literal> are set to zero.
</para>
<para>This example outputs the items as two paragraphs, except that (by
default) they have no paragraph indent and are vertically separated.
</para>
<screen>\begin{trivlist}
\item The \textit{Surprise} is not old; no one would call her old.
\item She has a bluff bow, lovely lines.
\end{trivlist}
</screen>

</sect2>
</sect1>
<sect1 label="8.17" id="math">
<title><literal>math</literal></title>

<indexterm role="fn"><primary>environment, <literal>math</literal></primary></indexterm>
<indexterm role="fn"><primary><literal>math</literal> environment</primary></indexterm>
        
<indexterm role="cp"><primary>in-line formulas</primary></indexterm>

<para>Synopsis:
</para>
<screen>\begin{math}
<replaceable>math</replaceable>
\end{math}
</screen>
<para>The <literal>math</literal> environment inserts given <replaceable>math</replaceable> material within
the running text.  <literal>\(...\)</literal> and <literal>$...$</literal> are synonyms.
See <link linkend="Math-formulas">Math formulas</link>.
</para>

</sect1>
<sect1 label="8.18" id="minipage">
<title><literal>minipage</literal></title>

<indexterm role="fn"><primary>environment, <literal>minipage</literal></primary></indexterm>
<indexterm role="fn"><primary><literal>minipage</literal> environment</primary></indexterm>
        
<indexterm role="cp"><primary>minipage, creating a</primary></indexterm>

<para>Synopses:
</para>
<screen>\begin{minipage}{<replaceable>width</replaceable>}
  <replaceable>contents</replaceable>
\end{minipage}
</screen>
<para>or
</para>
<screen>\begin{minipage}[<replaceable>position</replaceable>][<replaceable>height</replaceable>][<replaceable>inner-pos</replaceable>]{<replaceable>width</replaceable>}
  <replaceable>contents</replaceable>
\end{minipage}
</screen>
<para>Put <replaceable>contents</replaceable> into a box that is <replaceable>width</replaceable> wide.  This is like a
small version of a page; it can contain its own footnotes, itemized
lists, etc. (There are some restrictions, including that it cannot have
floats.)  This box will not be broken across pages.  So <literal>minipage</literal>
is similar to <literal>\parbox</literal> (see <link linkend="_005cparbox">\parbox</link>) but here you can have
paragraphs.
</para>
<para>This example will be 3&#160;inches wide, and has two paragraphs.
</para>
<screen>\begin{minipage}{3in}
  Stephen Kleene was a founder of the Theory of Computation.

  He was a student of Church, wrote three influential texts,
  was President of the Association for Symbolic Logic,
  and won the National Medal of Science.
\end{minipage}
</screen>
<para>See below for a discussion of the paragraph indent inside a
<literal>minipage</literal>.
</para>
<para>The required argument <replaceable>width</replaceable> is a rigid length (see <link linkend="Lengths">Lengths</link>).
It gives the width of the box into which <replaceable>contents</replaceable> are typeset.
</para>
<para>There are three optional arguments, <replaceable>position</replaceable>, <replaceable>height</replaceable>, and
<replaceable>inner-pos</replaceable>.  You need not include all three. For example, get the
default <replaceable>position</replaceable> and set the <replaceable>height</replaceable> with
<literal>\begin{minipage}[c][2.54cm] <replaceable>contents</replaceable> \end{minipage}</literal>.
(Get the natural height with an empty argument, <literal>[]</literal>.)
</para>
<para>The optional argument <replaceable>position</replaceable> governs how the <literal>minipage</literal>
vertically aligns with the surrounding material.
</para>
<variablelist><varlistentry><term><literal>c</literal>
</term><listitem><para>(synonym <literal>m</literal>) Default.  Positions the <literal>minipage</literal> so its
vertical center lines up with the center of the adjacent text line (what
Plain &tex; calls <literal>\vcenter</literal>).
</para>
</listitem></varlistentry><varlistentry><term><literal>t</literal>
</term><listitem><para>Match the top line in the <literal>minipage</literal> with the baseline of the
surrounding text (Plain &tex;&#8217;s <literal>\vtop</literal>.
</para>
</listitem></varlistentry><varlistentry><term><literal>b</literal>
</term><listitem><para>Match the bottom line in the <literal>minipage</literal> with the baseline of the
surrounding text (Plain &tex;&#8217;s <literal>\vbox</literal>.
</para></listitem></varlistentry></variablelist>
<para>To see the effects of these, contrast running this
</para>
<screen>---\begin{minipage}[c]{0.25in}
  first\\ second\\ third
\end{minipage}
</screen>
<para>with the results of changing <literal>c</literal> to <literal>b</literal> or&#160;<literal>t</literal>.
</para>
<para>The optional argument <replaceable>height</replaceable> is a rigid length (see <link linkend="Lengths">Lengths</link>).
It sets the height of the <literal>minipage</literal>.  You can enter any value
larger than, or equal to, or smaller than the <literal>minipage</literal>&#8217;s natural
height and &latex; will not give an error or warning.  You can also set
it to a height of zero or a negative value.
</para>
<para>The final optional argument <replaceable>inner-pos</replaceable> controls the placement of
<replaceable>content</replaceable> inside the box. These are the possible values are (the
default is the value of <replaceable>position</replaceable>).
</para>
<variablelist><varlistentry><term><literal>t</literal>
</term><listitem><para>Place <replaceable>content</replaceable> at the top of the box.
</para>
</listitem></varlistentry><varlistentry><term><literal>c</literal>
</term><listitem><para>Place it in the vertical center.
</para>
</listitem></varlistentry><varlistentry><term><literal>b</literal>
</term><listitem><para>Place it at the box bottom.
</para>
</listitem></varlistentry><varlistentry><term><literal>s</literal>
</term><listitem><para>Stretch <replaceable>contents</replaceable> out vertically; it must contain vertically
stretchable space.
</para>
</listitem></varlistentry></variablelist>
<para>The <replaceable>inner-pos</replaceable> argument makes sense when the <replaceable>height</replaceable> options
is set to a value larger than the <literal>minipage</literal>&#8217;s natural height.  To
see the effect of the options, run this example with the various choices
in place of <literal>b</literal>.
</para>
<screen>Text before
\begin{center}
  ---\begin{minipage}[c][3in][b]{0.25\textwidth}
       first\\ second\\ third
  \end{minipage}              
\end{center}
Text after  
</screen>
<indexterm role="cp"><primary>indentation of paragraphs, in minipage</primary></indexterm>
<indexterm role="cp"><primary>paragraph indentation, in minipage</primary></indexterm>
<indexterm role="fn"><primary>\parindent</primary></indexterm>
<para>By default paragraphs are not indented in a <literal>minipage</literal>.  Change
that with a command such as <literal>\setlength{\parindent}{1pc}</literal> at
the start of <replaceable>contents</replaceable>.
</para>
<indexterm role="cp"><primary>footnotes in figures</primary></indexterm>
<indexterm role="cp"><primary>figures, footnotes in</primary></indexterm>
<para>Footnotes in a <literal>minipage</literal> environment are handled in a way that is
particularly useful for putting footnotes in figures or tables.  A
<literal>\footnote</literal> or <literal>\footnotetext</literal> command puts the footnote at
the bottom of the minipage instead of at the bottom of the page, and it
uses the <literal>\mpfootnote</literal> counter instead of the ordinary
<literal>footnote</literal> counter (see <link linkend="Counters">Counters</link>).
</para>
<para>This puts the footnote at the bottom of the table, not the bottom of the
page.
</para>
<screen>\begin{center}           % center the minipage on the line
\begin{minipage}{2.5in}
  \begin{center}         % center the table inside the minipage
    \begin{tabular}{ll}
      \textsc{Monarch}  &amp;\textsc{Reign}             \\ \hline
      Elizabeth II      &amp;63 years\footnote{to date} \\
      Victoria          &amp;63 years                   \\
      George III        &amp;59 years
    \end{tabular}
  \end{center}  
\end{minipage}
\end{center}
</screen>
<para>If you nest minipages then there is an oddness when using footnotes.
Footnotes appear at the bottom of the text ended by the next
<literal>\end{minipage}</literal> which may not be their logical place.
</para>
<para>This puts a table containing data side by side with a map graphic.  They
are vertically centered.
</para>
<screen>\newcommand*{\vcenteredhbox}[1]{\begin{tabular}{@{}c@{}}#1\end{tabular}}
  ...
\begin{center}
  \vcenteredhbox{\includegraphics[width=0.3\textwidth]{nyc.png}}
  \hspace{0.1\textwidth}
  \begin{minipage}{0.5\textwidth}
    \begin{tabular}{r|l}
      \multicolumn{1}{r}{Borough} &amp;Pop (million)  \\ \hline
      The Bronx      &amp;$1.5$  \\
      Brooklyn       &amp;$2.6$  \\
      Manhattan      &amp;$1.6$  \\
      Queens         &amp;$2.3$  \\
      Staten Island  &amp;$0.5$  
    \end{tabular}
  \end{minipage}              
\end{center}
</screen>

</sect1>
<sect1 label="8.19" id="picture">
<title><literal>picture</literal></title>

<indexterm role="fn"><primary>environment, <literal>picture</literal></primary></indexterm>
<indexterm role="fn"><primary><literal>picture</literal> environment</primary></indexterm>
        
<indexterm role="cp"><primary>creating pictures</primary></indexterm>
<indexterm role="cp"><primary>pictures, creating</primary></indexterm>

<para>Synopses: 
</para><screen>\begin{picture}(<replaceable>width</replaceable>,<replaceable>height</replaceable>)
   <replaceable>picture commands</replaceable> 
\end{picture}
</screen>
<para>or
</para>
<screen>\begin{picture}(<replaceable>width</replaceable>,<replaceable>height</replaceable>)(<replaceable>xoffset</replaceable>,<replaceable>yoffset</replaceable>)
  <replaceable>picture commands</replaceable>
\end{picture}
</screen>
<para>An environment to create simple pictures containing lines, arrows,
boxes, circles, and text.  Note that while this environment is not
obsolete, new documents typically use much more powerful graphics
creation systems, such as <literal>TikZ</literal>, <literal>PSTricks</literal>, <literal>MetaPost</literal>,
or <literal>Asymptote</literal>.  These are not covered in this document; see CTAN.
</para>
<para>This shows the parallelogram law for adding vectors.
</para>
<indexterm role="fn"><primary>\unitlength</primary></indexterm>
<screen>\setlength{\unitlength}{1cm}
\begin{picture}(6,6)      % picture box will be 6cm wide by 6cm tall
  \put(0,0){\vector(2,1){4}}  % for every 2 over this vector goes 1 up
    \put(2,1){\makebox(0,0)[l]{\ first leg}}
  \put(4,2){\vector(1,2){2}}
    \put(5,4){\makebox(0,0)[l]{\ second leg}}  
  \put(0,0){\line(1,1){6}}
    \put(3,3){\makebox(0,0)[r]{sum\ }}  
\end{picture}
</screen>
<para>You can also use this environment to place arbitrary material at an
exact location.
</para>
<screen>\usepackage{color,graphicx}  % in preamble
  ...
\begin{center}
\setlength{\unitlength}{\textwidth} 
\begin{picture}(1,1)      % leave space, \textwidth wide and tall
  \put(0,0){\includegraphics[width=\textwidth]{desertedisland.jpg}}
  \put(0.25,0.35){\textcolor{red}{X Treasure here}}
\end{picture}
\end{center}
</screen>
<para>The red&#160;X will be precisely a quarter of the <literal>\linewidth</literal> from
the left margin, and <literal>0.35\linewidth</literal> up from the bottom.  Another
example of this usage is to put similar code in the page header to get
repeat material on each of a document&#8217;s pages.
</para>
<para>The <literal>picture</literal> environment has one required argument, a pair of
numbers (<replaceable>width</replaceable>,<replaceable>height</replaceable>).  Multiply these by the value
<literal>\unitlength</literal> to get the nominal size of the output, the space that
&latex; reserves on the output page.  This nominal size need not be how
large the picture really is; &latex; will draw things from the picture
outside the picture&#8217;s box.
</para>
<para>This environment also has an optional argument
(<replaceable>xoffset</replaceable>,<replaceable>yoffset</replaceable>).  It is used to shift the origin.  Unlike
most optional arguments, this one is not contained in square brackets.
As with the required argument, it consists of two real numbers.
Multiply these by <literal>\unitlength</literal> to get the point at the lower-left
corner of the picture.
</para>
<para>For example, if <literal>\unitlength</literal> has been set to <literal>1mm</literal>, the
command
</para>
<screen>\begin{picture}(100,200)(10,20)
</screen>
<para>produces a box of width 100 millimeters and height 200 millimeters.  The
picture&#8217;s origin is the point (10mm,20mm) and so the lower-left corner
is there, and the upper-right corner is at (110mm,220mm).  When you
first draw a picture you typically omit the optional argument, leaving
the origin at the lower-left corner.  If you then want to modify your
picture by shifting everything, you can just add the appropriate
optional argument.
</para>
<indexterm role="cp"><primary>position, in picture</primary></indexterm>
<para>Each <replaceable>picture command</replaceable> tells &latex; where to put something by
naming its position.  A <firstterm>position</firstterm> is a pair such as <literal>(2.4,-5)</literal>
giving the x- and y-coordinates.  A <firstterm>coordinate</firstterm> is a not a length,
it is a real number (it may have a decimal point or a minus sign).  It
specifies a length in multiples of the unit length <literal>\unitlength</literal>,
so if <literal>\unitlength</literal> has been set to <literal>1cm</literal>, then the coordinate
2.54 specifies a length of 2.54 centimeters.
</para>
<para>&latex;&#8217;s default for <literal>\unitlength</literal> is <literal>1pt</literal>.  it is a rigid
length (see <link linkend="Lengths">Lengths</link>).  Change it with the <literal>\setlength</literal> command
(see <link linkend="_005csetlength">\setlength</link>).  Make this change only outside of a <literal>picture</literal>
environment.
</para>
<para>Coordinates are given with respect to an origin, which is normally at
the lower-left corner of the picture.  Note that when a position appears
as an argument, as with <literal>\put(1,2){...}</literal>, it is not enclosed in
braces since the parentheses serve to delimit the argument.  Also,
unlike in some computer graphics systems, larger y-coordinates are
further up the page.
</para>
<para>There are four ways to put things in a picture: <literal>\put</literal>,
<literal>\multiput</literal>, <literal>\qbezier</literal>, and <literal>\graphpaper</literal>.  The most
often used is <literal>\put</literal>. This
</para>
<screen>\put(11.3,-0.3){...}
</screen>
<para>places the object with its reference point at coordinates
<inlineequation><mathphrase>(11.3,-0.3)</mathphrase></inlineequation>.  The reference points for various objects will be
described below.
<indexterm role="fn"><primary>LR box</primary></indexterm>
The <literal>\put</literal> command creates an <firstterm>LR box</firstterm> (see <link linkend="Modes">Modes</link>).
Anything that can go in an <literal>\mbox</literal> (see <link linkend="_005cmbox-_0026-_005cmakebox">\mbox &amp; \makebox</link>) can
go in the text argument of the <literal>\put</literal> command.  The reference point
will be the lower left corner of the box.  In this picture
</para>
<screen>\setlength{\unitlength}{1cm}
...\begin{picture}(1,1)
  \put(0,0){\line(1,0){1}}
  \put(0,0){\line(1,1){1}}
\end{picture}
</screen>
<para>the three dots are just slightly left of the point of the angle formed
by the two lines.  (Also, <literal>\line(1,1){1}</literal> does not call for a
line of length one; rather the line has a change in the x coordinate of
1.)
</para>
<para>The <literal>\multiput</literal>, <literal>qbezier</literal>, and <literal>graphpaper</literal> commands are
described below.
</para>
<para>This draws a rectangle with a wavy top, using <literal>\qbezier</literal> for
that curve.
</para>
<screen>\begin{picture}(3,1.5)
  \put(0,0){\vector(1,0){8}}  % x axis
  \put(0,0){\vector(0,1){4}}  % y axis
  \put(2,0){\line(0,1){3}}       % left side rectangle
  \put(4,0){\line(0,1){3.5}}     % right side
  \qbezier(2,3)(2.5,2.9)(3,3.25)
    \qbezier(3,3.25)(3.5,3.6)(4,3.5)
  \thicklines                 % below here, lines are twice as thick
  \put(2,3){\line(4,1){2}}
  \put(4.5,2.5){\framebox{Trapezoidal Rule}}
\end{picture}
</screen>


<sect2 label="8.19.1" id="_005cput">
<title><literal>\put</literal></title>

<indexterm role="fn"><primary>\put</primary></indexterm>

<para>Synopsis:
</para>
<screen>\put(<replaceable>xcoord</replaceable>,<replaceable>ycoord</replaceable>){<replaceable>content</replaceable>}
</screen>
<para>Place <replaceable>content</replaceable> at the coordinate (<replaceable>xcoord</replaceable>,<replaceable>ycoord</replaceable>).  See
the discussion of coordinates and <literal>\unitlength</literal> in <link linkend="picture">picture</link>.
The <replaceable>content</replaceable> is processed in LR mode (see <link linkend="Modes">Modes</link>) so it cannot
contain line breaks.
</para>
<para>This includes the text into the <literal>picture</literal>.
</para>
<screen>\put(4.5,2.5){Apply the \textit{unpoke} move}
</screen>
<para>The reference point, the location (4.5,2.5), is the lower left of the
text, at the bottom left of the &#8216;<literal>A</literal>&#8217;.
</para>

</sect2>
<sect2 label="8.19.2" id="_005cmultiput">
<title><literal>\multiput</literal></title>

<indexterm role="fn"><primary>\multiput</primary></indexterm>

<para>Synopsis:
</para>
<screen>\multiput(<replaceable>x</replaceable>,<replaceable>y</replaceable>)(<replaceable>delta_x</replaceable>,<replaceable>delta_y</replaceable>){<replaceable>num-copies</replaceable>}{<replaceable>obj</replaceable>}
</screen>
<para>Copy <replaceable>obj</replaceable> a total of <replaceable>num-copies</replaceable> times, with an increment of
<replaceable>delta_x,delta_y</replaceable>.  The <replaceable>obj</replaceable> first appears at position
<inlineequation><mathphrase>(x,y)</mathphrase></inlineequation>, then at <inlineequation><mathphrase>(x+\delta_x,y+\delta_y)</mathphrase></inlineequation>, and so on.
</para>
<para>This draws a simple grid with every fifth line in bold (see also
<link linkend="_005cgraphpaper">\graphpaper</link>).
</para>
<screen>\begin{picture}(10,10)
  \linethickness{0.05mm}
  \multiput(0,0)(1,0){10}{\line(0,1){10}} 
  \multiput(0,0)(0,1){10}{\line(1,0){10}}
  \linethickness{0.5mm}
  \multiput(0,0)(5,0){3}{\line(0,1){10}}
  \multiput(0,0)(0,5){3}{\line(1,0){10}}
\end{picture}
</screen>

</sect2>
<sect2 label="8.19.3" id="_005cqbezier">
<title><literal>\qbezier</literal></title>

<indexterm role="fn"><primary>\qbezier</primary></indexterm>

<para>Synopsis:
</para>
<screen>\qbezier(<replaceable>x1</replaceable>,<replaceable>y1</replaceable>)(<replaceable>x2</replaceable>,<replaceable>y2</replaceable>)(<replaceable>x3</replaceable>,<replaceable>y3</replaceable>)
\qbezier[<replaceable>num</replaceable>](<replaceable>x1</replaceable>,<replaceable>y1</replaceable>)(<replaceable>x2</replaceable>,<replaceable>y2</replaceable>)(<replaceable>x3</replaceable>,<replaceable>y3</replaceable>)
</screen>
<para>Draw a quadratic Bezier curve whose control points are given by the
three required arguments <literal>(<replaceable>x1</replaceable>,<replaceable>y1</replaceable>)</literal>,
<literal>(<replaceable>x2</replaceable>,<replaceable>y2</replaceable>)</literal>, and <literal>(<replaceable>x3</replaceable>,<replaceable>y3</replaceable>)</literal>.  That is,
the curve runs from <replaceable>(x1,y1)</replaceable> to <replaceable>(x3,y3)</replaceable>, is quadratic, and is
such that the tangent line at <replaceable>(x1,y1)</replaceable> passes through
<replaceable>(x2,y2)</replaceable>, as does the tangent line at <replaceable>(x3,y3)</replaceable>.
</para>
<para>This draws a curve from the coordinate (1,1) to (1,0).
</para>
<screen>\qbezier(1,1)(1.25,0.75)(1,0)  
</screen>
<para>The curve&#8217;s tangent line at (1,1) contains (1.25,0.75), as does the
curve&#8217;s tangent line at (1,0).
</para>
<para>The optional argument <replaceable>num</replaceable> gives the number of calculated
intermediate points.  The default is to draw a smooth curve whose
maximum number of points is <literal>\qbeziermax</literal> (change this value with
<literal>\renewcommand</literal>).
</para>

</sect2>
<sect2 label="8.19.4" id="_005cgraphpaper">
<title><literal>\graphpaper</literal></title>

<indexterm role="fn"><primary>\graphpaper</primary></indexterm>
<para>Synopsis:
</para>
<screen>\graphpaper(<replaceable>x_init</replaceable>,<replaceable>y_init</replaceable>)(<replaceable>x_dimen</replaceable>,<replaceable>y_dimen</replaceable>)
\graphpaper[<replaceable>spacing</replaceable>](<replaceable>x_init</replaceable>,<replaceable>y_init</replaceable>)(<replaceable>x_dimen</replaceable>,<replaceable>y_dimen</replaceable>)
</screen>
<para>Draw a coordinate grid.  Requires the <literal>graphpap</literal> package.
The grid&#8217;s origin is <literal>(<replaceable>x_init</replaceable>,<replaceable>y_init</replaceable>)</literal>.
Grid lines come every <replaceable>spacing</replaceable> units (the default is 10).
The grid extends <replaceable>x_dimen</replaceable> units to the right and <replaceable>y_dimen</replaceable> units up.
All arguments must be positive integers.
</para>
<para>This make a grid with seven vertical lines and eleven horizontal lines.
</para>
<screen>\usepackage{graphpap}    % in preamble
  ...
\begin{picture}(6,20)    % in document body
  \graphpaper[2](0,0)(12,20)
\end{picture}
</screen>
<para>The lines are numbered every ten units.
</para>

</sect2>
<sect2 label="8.19.5" id="_005cline">
<title><literal>\line</literal></title>

<indexterm role="fn"><primary>\line</primary></indexterm>

<para>Synopsis:
</para>
<screen>\line(<replaceable>x_run</replaceable>,<replaceable>y_rise</replaceable>){<replaceable>travel</replaceable>}
</screen>
<para>Draw a line.  It slopes such that it vertically rises <replaceable>y_rise</replaceable> for
every horizontal <replaceable>x_run</replaceable>.  The <replaceable>travel</replaceable> is the total horizontal
change&#160;&#8212; it is not the length of the vector, it is the change in
<inlineequation><mathphrase>x</mathphrase></inlineequation>.  In the special case of vertical lines, where
(<replaceable>x_run</replaceable>,<replaceable>y_rise</replaceable>)=(0,1), the <replaceable>travel</replaceable> gives the change in
<inlineequation><mathphrase>y</mathphrase></inlineequation>.
</para>
<para>This draws a line starting at coordinates (1,3).
</para>
<screen>\put(1,3){\line(2,5){4}}
</screen>
<para>For every over 2, this line will go up 5.  Because <replaceable>travel</replaceable>
specifies that this goes over 4, it must go up&#160;10.  Thus its
endpoint is <inlineequation><mathphrase>(1,3)+(4,10)=(5,13)</mathphrase></inlineequation>.  In particular, note that
<inlineequation><mathphrase><replaceable>travel</replaceable>=4</mathphrase></inlineequation> is not the length of the line, it is the change in
<inlineequation><mathphrase>x</mathphrase></inlineequation>.
</para>
<para>The arguments <replaceable>x_run</replaceable> and <replaceable>y_rise</replaceable> are integers that can be
positive, negative, or zero. (If both are 0 then &latex; treats the
second as 1.)  With
<literal>\put(<replaceable>x_init</replaceable>,<replaceable>y_init</replaceable>){\line(<replaceable>x_run</replaceable>,<replaceable>y_rise</replaceable>){<replaceable>travel</replaceable>}}</literal>,
if <replaceable>x_run</replaceable> is negative then the line&#8217;s ending point has a first
coordinate that is less than <replaceable>x_init</replaceable>.  If <replaceable>y_rise</replaceable> is negative
then the line&#8217;s ending point has a second coordinate that is less than
<replaceable>y_init</replaceable>.
</para>
<para>If <replaceable>travel</replaceable> is negative then you get <literal>LaTeX Error: Bad \line or
\vector argument.</literal>
</para>
<indexterm role="cp"><primary><literal>pict2e</literal> package</primary></indexterm>
<indexterm role="cp"><primary>graphics packages</primary></indexterm>
<indexterm role="cp"><primary>package, <literal>pict2e</literal></primary></indexterm>
<indexterm role="cp"><primary><literal>pict2e</literal> package</primary></indexterm>

<indexterm role="cp"><primary>package, <literal>TikZ</literal></primary></indexterm>
<indexterm role="cp"><primary><literal>TikZ</literal> package</primary></indexterm>

<indexterm role="cp"><primary>package, <literal>PSTricks</literal></primary></indexterm>
<indexterm role="cp"><primary><literal>PSTricks</literal> package</primary></indexterm>

<indexterm role="cp"><primary>package, <literal>MetaPost</literal></primary></indexterm>
<indexterm role="cp"><primary><literal>MetaPost</literal> package</primary></indexterm>

<indexterm role="cp"><primary>package, <literal>Asymptote</literal></primary></indexterm>
<indexterm role="cp"><primary><literal>Asymptote</literal> package</primary></indexterm>

<para>Standard &latex; can only draw lines with a limited range of slopes
because these lines are made by putting together line segments from
pre-made fonts.  The two numbers <replaceable>x_run</replaceable> and <replaceable>y_rise</replaceable> must have
integer values from &#8722;6 through&#160;6.  Also, they must be
relatively prime, so that <replaceable>(x_run,y_rise)</replaceable> can be (2,1) but not
(4,2) (if you choose the latter then instead of lines you get sequences
of arrowheads; the solution is to switch to the former).  To get lines
of arbitrary slope and plenty of other shapes in a system like
<literal>picture</literal>, see the package <filename>pict2e</filename> on CTAN.  Another solution
is to use a full-featured graphics system such as <filename>TikZ</filename>, or
<filename>PSTricks</filename>, or <filename>MetaPost</filename>, or <filename>Asymptote</filename>
</para>

</sect2>
<sect2 label="8.19.6" id="_005clinethickness">
<title><literal>\linethickness</literal></title>

<indexterm role="fn"><primary>\linethickness</primary></indexterm>

<para>Synopsis:
</para>
<screen>\linethickness{<replaceable>dim</replaceable>}
</screen>
<para>Declares the thickness of subsequent horizontal and vertical lines in a
picture to be <replaceable>dim</replaceable>, which must be a positive length
(see <link linkend="Lengths">Lengths</link>).  It differs from <literal>\thinlines</literal> and
<literal>\thicklines</literal> in that it does not affect the thickness of slanted
lines, circles, or ovals.
</para>

</sect2>
<sect2 label="8.19.7" id="_005cthinlines">
<title><literal>\thinlines</literal></title>

<indexterm role="fn"><primary>\thinlines</primary></indexterm>

<para>Declaration to set the thickness of subsequent lines, circles, and ovals
in a picture environment to be 0.4pt.  This is the default
thickness, so this command is unnecessary unless the thickness has been
changed with either <link linkend="_005clinethickness">\linethickness</link> or <link linkend="_005cthicklines">\thicklines</link>.
</para>

</sect2>
<sect2 label="8.19.8" id="_005cthicklines">
<title><literal>\thicklines</literal></title>

<indexterm role="fn"><primary>\thicklines</primary></indexterm>

<para>Declaration to set the thickness of subsequent lines, circles, and ovals
in a picture environment to be 0.8pt.  See also
<link linkend="_005clinethickness">\linethickness</link> and <link linkend="_005cthinlines">\thinlines</link>.  This command is illustrated
in the Trapezoidal Rule example of <link linkend="picture">picture</link>.
</para>

</sect2>
<sect2 label="8.19.9" id="_005ccircle">
<title><literal>\circle</literal></title>

<indexterm role="fn"><primary>\circle</primary></indexterm>

<para>Synopsis:
</para>
<screen>\circle{<replaceable>diameter</replaceable>}
\circle*{<replaceable>diameter</replaceable>}
</screen>
<para>Produces a circle with a diameter as close as possible to the specified
one.  The <literal>*</literal>&#160;form produces a filled-in circle.
</para>
<para>This draws a circle of radius 6, centered at <literal>(5,7)</literal>.
</para>
<screen>\put(5,7){\circle{6}}
</screen>
<para>The available radii for <literal>circle</literal> are, in points, the even
numbers from 2 to 20, inclusive.  For <literal>circle*</literal> they are all the
integers from 1 to 15.
</para>

</sect2>
<sect2 label="8.19.10" id="_005coval">
<title><literal>\oval</literal></title>

<indexterm role="fn"><primary>\oval</primary></indexterm>

<para>Synopsis:
</para>
<screen>\oval(<replaceable>width</replaceable>,<replaceable>height</replaceable>)
\oval(<replaceable>width</replaceable>,<replaceable>height</replaceable>)[<replaceable>portion</replaceable>]
</screen>
<para>Produce a rectangle with rounded corners.  The optional argument
<replaceable>portion</replaceable> allows you to produce only half or a quarter of the oval.
For half an oval take <replaceable>portion</replaceable> to be one of these.
</para>
<variablelist><varlistentry><term><literal>t</literal>
</term><listitem><para>top half
</para></listitem></varlistentry><varlistentry><term><literal>b</literal>
</term><listitem><para>bottom half
</para></listitem></varlistentry><varlistentry><term><literal>r</literal>
</term><listitem><para>right half
</para></listitem></varlistentry><varlistentry><term><literal>l</literal>
</term><listitem><para>left half
</para></listitem></varlistentry></variablelist>
<para>Produce only one quarter of the oval by setting <replaceable>portion</replaceable> to
<literal>tr</literal>, <literal>br</literal>, <literal>bl</literal>, or <literal>tl</literal>.
</para>
<para>This draws the top half of an oval that is 3 wide and 7 tall.  
</para>
<screen>\put(5,7){\oval(3,7)[t]}
</screen>
<para>The (5,7) is the center of the entire oval, not just the center of the
top half.
</para>
<para>These shapes are not ellipses.  They are rectangles whose corners are
made with quarter circles.  These circles have a maximum radius of
20pt (see <link linkend="_005ccircle">\circle</link> for the sizes).  Thus large ovals are just
boxes with a small amount of corner rounding.
</para>

</sect2>
<sect2 label="8.19.11" id="_005cshortstack">
<title><literal>\shortstack</literal></title>

<indexterm role="fn"><primary>\shortstack</primary></indexterm>

<para>Synopsis:
</para>
<screen>\shortstack[<replaceable>position</replaceable>]{<replaceable>line 1</replaceable> \\ ... }
</screen>
<para>Produce a vertical stack of objects.
</para>
<para>This labels the <inlineequation><mathphrase>y</mathphrase></inlineequation> axis.
</para>
<screen>\put(0,0){\vector(1,0){4}}   % x axis
\put(0,0){\vector(0,1){2}}   % y
\put(-0.25,2){\makebox[0][r]{\shortstack[r]{$y$\\ axis}}}
</screen>
<para>For a short stack, the reference point is the lower left of the stack.
In the above example the <link linkend="_005cmbox-_0026-_005cmakebox">\mbox &amp; \makebox</link> puts the stack flush
right in a zero width box so in total the short stack sits slightly to
the left of the <inlineequation><mathphrase>y</mathphrase></inlineequation>&#160;axis.
</para>
<para>The valid positions are:
</para>
<variablelist><varlistentry><term><literal>r</literal>
</term><listitem><para>Make objects flush right
</para></listitem></varlistentry><varlistentry><term><literal>l</literal>
</term><listitem><para>Make objects flush left
</para></listitem></varlistentry><varlistentry><term><literal>c</literal>
</term><listitem><para>Center objects (default)
</para></listitem></varlistentry></variablelist>
<indexterm role="fn"><primary>\\ (for <literal>\shortstack</literal> objects)</primary></indexterm>
<para>Separate objects into lines with <literal>\\</literal>.  These stacks are short in
that, unlike in a <literal>tabular</literal> or <literal>array</literal> environment, here the
rows are not spaced out to be of even heights.  Thus, in
<literal>\shortstack{X\\o\\o\\X}</literal> the first and last rows are taller than
the middle two.  You can adjust row heights either by putting in the
usual interline spacing with <literal>\shortstack{X\\ \strut o\\o\\X}</literal>,
or by hand, via an explicit zero-width box <literal>\shortstack{X \\
\rule{0pt}{12pt} o\\o\\X}</literal> or by using <literal>\\</literal>&#8217;s optional
argument <literal>\shortstack{X\\[2pt] o\\o\\X}</literal>.
</para>
<para>The <literal>\shortstack</literal> command is also available outside the
<literal>picture</literal> environment.
</para>

</sect2>
<sect2 label="8.19.12" id="_005cvector">
<title><literal>\vector</literal></title>

<indexterm role="fn"><primary>\vector</primary></indexterm>

<para>Synopsis:
</para>
<screen>\vector(<replaceable>x_run</replaceable>,<replaceable>y_rise</replaceable>){<replaceable>travel</replaceable>}
</screen>
<para>Draw a line ending in an arrow.  The slope of that line is: it
vertically rises <replaceable>y_rise</replaceable> for every horizontal <replaceable>x_run</replaceable>.  The
<replaceable>travel</replaceable> is the total horizontal change&#160;&#8212; it is not the
length of the vector, it is the change in <inlineequation><mathphrase>x</mathphrase></inlineequation>.  In the special case
of vertical vectors, if (<replaceable>x_run</replaceable>,<replaceable>y_rise</replaceable>)=(0,1), then
<replaceable>travel</replaceable> gives the change in <inlineequation><mathphrase>y</mathphrase></inlineequation>.
</para>
<para>For an example see <link linkend="picture">picture</link>.
</para>
<para>For elaboration on <replaceable>x_run</replaceable> and <replaceable>y_rise</replaceable> see <link linkend="_005cline">\line</link>.  As
there, the values of <replaceable>x_run</replaceable> and <replaceable>y_rise</replaceable> are limited.  For
<literal>\vector</literal> you must chooses integers between &#8722;4 and 4,
inclusive.  Also, the two you choose must be relatively prime.  Thus,
<literal>\vector(2,1){4}</literal> is acceptable but <literal>\vector(4,2){4}</literal> is
not (if you use the latter then you get a sequence of arrowheads).
</para>

</sect2>
<sect2 label="8.19.13" id="_005cmakebox-_0028picture_0029">
<title><literal>\makebox</literal> (picture)</title>

<indexterm role="fn"><primary>\makebox (for <literal>picture</literal>)</primary></indexterm>

<para>Synopsis:
</para>
<screen>\makebox(<replaceable>rec-width</replaceable>,<replaceable>rec-height</replaceable>){<replaceable>text</replaceable>}
\makebox(<replaceable>rec-width</replaceable>,<replaceable>rec-height</replaceable>)[<replaceable>position</replaceable>]{<replaceable>text</replaceable>}
</screen>
<para>Make a box to hold <replaceable>text</replaceable>.  This command fits with the
<literal>picture</literal> environment, although you can use it outside of there,
because <replaceable>rec-width</replaceable> and <replaceable>rec-height</replaceable> are numbers specifying
distances in terms of the <literal>\unitlength</literal> (see <link linkend="picture">picture</link>).  This
command is similar to the normal <literal>\makebox</literal> command (see <link linkend="_005cmbox-_0026-_005cmakebox">\mbox &amp;
\makebox</link>) except here that you must specify the width and height.  This
command is fragile (see <link linkend="_005cprotect">\protect</link>).
</para>
<para>This makes a box of length 3.5 times <literal>\unitlength</literal> and height 4
times <literal>\unitlength</literal>.
</para>
<screen>\put(1,2){\makebox(3.5,4){...}}
</screen>
<para>The optional argument <literal><replaceable>position</replaceable></literal> specifies where in the box
the <replaceable>text</replaceable> appears.  The default is to center it, both horizontally
and vertically.  To place it somewhere else, use a string with one or
two of these letters.
</para>
<variablelist><varlistentry><term><literal>t</literal>
</term><listitem><para>Puts <replaceable>text</replaceable> the top of the box.
</para>
</listitem></varlistentry><varlistentry><term><literal>b</literal>
</term><listitem><para>Put <replaceable>text</replaceable> at the bottom.
</para>
</listitem></varlistentry><varlistentry><term><literal>l</literal>
</term><listitem><para>Put <replaceable>text</replaceable> on the left.
</para>
</listitem></varlistentry><varlistentry><term><literal>r</literal>
</term><listitem><para>Put <replaceable>text</replaceable> on the right.
</para>
</listitem></varlistentry></variablelist>

</sect2>
<sect2 label="8.19.14" id="_005cframebox-_0028picture_0029">
<title><literal>\framebox</literal> (picture)</title>

<indexterm role="fn"><primary>\framebox</primary></indexterm>

<para>Synopsis:
</para>
<screen>\framebox(<replaceable>rec-width</replaceable>,<replaceable>rec-height</replaceable>){<replaceable>text</replaceable>}
\framebox(<replaceable>rec-width</replaceable>,<replaceable>rec-height</replaceable>)[<replaceable>position</replaceable>]{<replaceable>text</replaceable>}
</screen>
<para>This is the same as <link linkend="_005cmakebox-_0028picture_0029">\makebox (picture)</link> except that it puts a frame
around the outside of the box that it creates.  The reference point is
the bottom left corner of the frame.  This command fits with the
<literal>picture</literal> environment, although you can use it outside of there,
because lengths are numbers specifying the distance in terms of the
<literal>\unitlength</literal> (see <link linkend="picture">picture</link>).  This command is fragile
(see <link linkend="_005cprotect">\protect</link>).
</para>
<para>This example creates a frame 2.5&#160;inches by 3&#160;inches and puts
the text in the center.
</para>
<screen>\setlength{\unitlength}{1in}
\framebox(2.5,3){test text}
</screen>
<para>The required arguments are that the rectangle has overall width
<replaceable>rect-width</replaceable> units and height <replaceable>rect-height</replaceable> units.
</para>
<para>The optional argument <replaceable>position</replaceable> specifies the position of
<replaceable>text</replaceable>; see <link linkend="_005cmakebox-_0028picture_0029">\makebox (picture)</link> for the values that it can
take.
</para>
<indexterm role="fn"><primary>\fboxrule</primary></indexterm>
<indexterm role="fn"><primary>\fboxsep</primary></indexterm>
<para>The rule has thickness <literal>\fboxrule</literal> and there is a blank space
<literal>\fboxsep</literal> between the frame and the contents of the box.
</para>
<para>For this command, you must specify the <replaceable>width</replaceable> and <replaceable>height</replaceable>.  If
you want to just put a frame around some contents whose dimension is
determined in some other way then either use <literal>\fbox</literal> (see <link linkend="_005cfbox-_0026-_005cframebox">\fbox
&amp; \framebox</link>) or <literal>\frame</literal> (see <link linkend="_005cframe">\frame</link>).
</para>

</sect2>
<sect2 label="8.19.15" id="_005cframe">
<title><literal>\frame</literal></title>

<indexterm role="fn"><primary>\frame</primary></indexterm>

<para>Synopsis:
</para>
<screen>\frame{<replaceable>contents</replaceable>}
</screen>
<para>Puts a rectangular frame around <replaceable>contents</replaceable>.  The reference point is
the bottom left corner of the frame.  In contrast to <literal>\framebox</literal>
(see <link linkend="_005cframebox-_0028picture_0029">\framebox (picture)</link>), this command puts no extra space is put
between the frame and the object.  It is fragile (see <link linkend="_005cprotect">\protect</link>).
</para>

</sect2>
<sect2 label="8.19.16" id="_005cdashbox">
<title><literal>\dashbox</literal></title>

<indexterm role="fn"><primary>\dashbox</primary></indexterm>

<para>Synopsis:
</para>
<screen>\dashbox{<replaceable>dash-len</replaceable>}(<replaceable>rect-width</replaceable>,<replaceable>rect-height</replaceable>){<replaceable>text</replaceable>}
\dashbox{<replaceable>dash-len</replaceable>}(<replaceable>rect-width</replaceable>,<replaceable>rect-height</replaceable>)[<replaceable>position</replaceable>]{<replaceable>text</replaceable>}
</screen>
<para>Create a dashed rectangle around <replaceable>text</replaceable>.  This command fits with the
<literal>picture</literal> environment, although you can use it outside of there,
because lengths are numbers specifying the distance in terms of the
<literal>\unitlength</literal> (see <link linkend="picture">picture</link>).
</para>
<para>The required arguments are: dashes are <replaceable>dash-len</replaceable> units long, with
the same length gap, and the rectangle has overall width
<replaceable>rect-width</replaceable> units and height <replaceable>rect-height</replaceable> units.
</para>
<para>The optional argument <replaceable>position</replaceable> specifies the position of
<replaceable>text</replaceable>; see <link linkend="_005cmakebox-_0028picture_0029">\makebox (picture)</link> for the values that it can
take.
</para>
<para>This shows that you can use non-integer value for <replaceable>dash-len</replaceable>.
</para>
<screen>\put(0,0){\dashbox{0.1}(5,0.5){My hovercraft is full of eels.}}
</screen>
<para>Each dash will be <literal>0.1\unitlength</literal> long, the box&#8217;s width is
<literal>5\unitlength</literal> and its height is <literal>0.5\unitlength</literal>.
</para>
<para>As in that example, a dashed box looks best when <replaceable>rect-width</replaceable> and
<replaceable>rect-height</replaceable> are multiples of the <replaceable>dash-len</replaceable>.
</para>

</sect2>
</sect1>
<sect1 label="8.20" id="quotation-_0026-quote">
<title><literal>quotation</literal> &amp; <literal>quote</literal></title>

<indexterm role="fn"><primary>environment, <literal>quotation</literal></primary></indexterm>
<indexterm role="fn"><primary><literal>quotation</literal> environment</primary></indexterm>
        
<indexterm role="cp"><primary>quoted text with paragraph indentation, displaying</primary></indexterm>
<indexterm role="cp"><primary>displaying quoted text with paragraph indentation</primary></indexterm>
<indexterm role="cp"><primary>paragraph indentations in quoted text</primary></indexterm>
<indexterm role="fn"><primary>environment, <literal>quote</literal></primary></indexterm>
<indexterm role="fn"><primary><literal>quote</literal> environment</primary></indexterm>
        
<indexterm role="cp"><primary>quoted text without paragraph indentation, displaying</primary></indexterm>
<indexterm role="cp"><primary>displaying quoted text without paragraph indentation</primary></indexterm>
<indexterm role="cp"><primary>paragraph indentations in quoted text, omitting</primary></indexterm>

<para>Synopsis:
</para>
<screen>\begin{quotation}
  <replaceable>text</replaceable>
\end{quotation}
</screen>
<para>or 
</para>
<screen>\begin{quote}
  <replaceable>text</replaceable>
\end{quote}
</screen>
<para>Include a quotation. Both environments indent margins on both sides by
<literal>\leftmargin</literal> and the text is right-justified.
</para>
<para>They differ in how they treat paragraphs. In the <literal>quotation</literal>
environment, paragraphs are indented by 1.5em and the space
between paragraphs is small, <literal>0pt plus 1pt</literal>.  In the <literal>quote</literal>
environment, paragraphs are not indented and there is vertical space
between paragraphs (it is the rubber length <literal>\parsep</literal>).
</para>
<screen>\begin{quotation} \small\it
  Four score and seven years ago
  ... shall not perish from the earth.
  \hspace{1em plus 1fill}---Abraham Lincoln
\end{quotation}
</screen>

</sect1>
<sect1 label="8.21" id="tabbing">
<title><literal>tabbing</literal></title>

<indexterm role="fn"><primary>environment, <literal>tabbing</literal></primary></indexterm>
<indexterm role="fn"><primary><literal>tabbing</literal> environment</primary></indexterm>
        
<indexterm role="cp"><primary>tab stops, using</primary></indexterm>
<indexterm role="cp"><primary>lining text up using tab stops</primary></indexterm>
<indexterm role="cp"><primary>alignment via tabbing</primary></indexterm>

<para>Synopsis:
</para>
<screen>\begin{tabbing}
<replaceable>row1col1</replaceable> \= <replaceable>row1col2</replaceable> ...  \\
<replaceable>row2col1</replaceable> \&gt; <replaceable>row2col2</replaceable> ...  \\
...
\end{tabbing}
</screen>
<para>Align text in columns, by setting tab stops and tabbing to them much as
was done on a typewriter.  This is less often used than the environments
<literal>tabular</literal> (see <link linkend="tabular">tabular</link>) or <literal>array</literal> (see <link linkend="array">array</link>) because
in those the width of each column need not be constant and need not be
known in advance.
</para>
<para>This example has a first line where the tab stops are set to explicit
widths, ended by a <literal>\kill</literal> command (which is described below):
</para>
<screen>\begin{tabbing}
\hspace{0.75in}     \= \hspace{0.40in}  \= \hspace{0.40in}    \kill
Ship                \&gt; Guns             \&gt; Year    \\ 
\textit{Sophie}     \&gt; 14               \&gt; 1800    \\
\textit{Polychrest} \&gt; 24               \&gt; 1803    \\
\textit{Lively}     \&gt; 38               \&gt; 1804    \\
\textit{Surprise}   \&gt; 28               \&gt; 1805    \\
\end{tabbing}
</screen>
<para>Both the <literal>tabbing</literal> environment and the more widely-used
<literal>tabular</literal> environment put text in columns.  The most important
distinction is that in <literal>tabular</literal> the width of columns is
determined automatically by &latex;, while in <literal>tabbing</literal> the user
sets the tab stops.  Another distinction is that <literal>tabular</literal>
generates a box, but <literal>tabbing</literal> can be broken across pages.
Finally, while <literal>tabular</literal> can be used in any mode, <literal>tabbing</literal>
can be used only in paragraph mode and it starts a new paragraph.
</para>
<para>A <literal>tabbing</literal> environment always starts a new paragraph, without
indentation. Moreover, as shown in the example above, there is no need
to use the starred form of the <literal>\hspace</literal> command at the beginning
of a tabbed row.  The right margin of the <literal>tabbing</literal> environment is
the end of line, so that the width of the environment is
<literal>\linewidth</literal>.
</para>
<indexterm role="cp"><primary>row, tabbing</primary></indexterm>
<para>The <literal>tabbing</literal> environment contains a sequence of <firstterm>tabbed
rows</firstterm>.  The first tabbed row begins immediately after
<literal>\begin{tabbing}</literal> and each row ends with <literal>\\</literal> or
<literal>\kill</literal>. The last row may omit the <literal>\\</literal> and end with just
<literal>\end{tabbing}</literal>.
</para>
<para>At any point the <literal>tabbing</literal> environment has a current tab stop
pattern, a sequence of <inlineequation><mathphrase><replaceable>n</replaceable> &gt; 0</mathphrase></inlineequation> tab stops, numbered 0, 1,
etc.  These create <replaceable>n</replaceable> corresponding columns.  Tab stop&#160;0 is
always the left margin, defined by the enclosing environment.  Tab
stop number&#160;<replaceable>i</replaceable> is set if it is assigned a horizontal
position on the page.  Tab stop number&#160;<replaceable>i</replaceable> can only be set if
all the stops 0, &#8230;, <inlineequation><mathphrase>i-1</mathphrase></inlineequation> have already been set; normally
later stops are to the right of earlier ones.
</para>
<para>By default any text typeset in a <literal>tabbing</literal> environment is typeset
ragged right and left-aligned on the current tab stop. Typesetting is
done in LR mode (see <link linkend="Modes">Modes</link>).
</para>
<para>The following commands can be used inside a <literal>tabbing</literal> environment.
They are all fragile (see <link linkend="_005cprotect">\protect</link>).
</para>
<variablelist><varlistentry><term><indexterm role="fn"><primary>\\ (tabbing)</primary></indexterm><literal>\\ (tabbing)</literal>
</term><listitem><para>End a tabbed line and typeset it.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\= (tabbing)</primary></indexterm><literal>\= (tabbing)</literal>
</term><listitem><para>Sets a tab stop at the current position.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\&gt; (tabbing)</primary></indexterm><literal>\&gt; (tabbing)</literal>
</term><listitem><indexterm role="fn"><primary>\&gt;</primary></indexterm>
<para>Advances to the next tab stop.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\&lt;</primary></indexterm><literal>\&lt;</literal>
</term><listitem><para>Put following text to the left of the local margin (without changing
the margin).  Can only be used at the start of the line.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\+</primary></indexterm><literal>\+</literal>
</term><listitem><para>Moves the left margin of the next and all the
following commands one tab stop to the right, beginning tabbed line if
necessary.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\-</primary></indexterm><literal>\-</literal>
</term><listitem><para>Moves the left margin of the next and all the
following commands one tab stop to the left, beginning tabbed line if
necessary.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\' (tabbing)</primary></indexterm><literal>\' (tabbing)</literal>
</term><listitem><para>Moves everything that you have typed so far in the current column, i.e.,
everything from the most recent <literal>\&gt;</literal>, <literal>\&lt;</literal>, <literal>\'</literal>,
<literal>\\</literal>, or <literal>\kill</literal> command, to the previous column and aligned
to the right, flush against the current column&#8217;s tab stop.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\` (tabbing)</primary></indexterm><literal>\` (tabbing)</literal>
</term><listitem><para>Allows you to put text flush right against any tab stop, including tab
stop&#160;0.  However, it can&#8217;t move text to the right of the last
column because there&#8217;s no tab stop there.  The <literal>\`</literal> command moves
all the text that follows it, up to the <literal>\\</literal> or
<literal>\end{tabbing}</literal> command that ends the line, to the right margin
of the <literal>tabbing</literal> environment.  There must be no <literal>\&gt;</literal> or
<literal>\'</literal> command between the <literal>\`</literal> and the <literal>\\</literal> or
<literal>\end{tabbing}</literal> command that ends the line.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\a (tabbing)</primary></indexterm><literal>\a (tabbing)</literal>
</term><listitem><indexterm role="fn"><primary>\a' (acute accent in tabbing)</primary></indexterm>
<indexterm role="fn"><primary>\a` (grave accent in tabbing)</primary></indexterm>
<indexterm role="fn"><primary>\a= (macron accent in tabbing)</primary></indexterm>
<para>In a <literal>tabbing</literal> environment, the commands <literal>\=</literal>, <literal>\'</literal> and
<literal>\`</literal> do not produce accents as usual (see <link linkend="Accents">Accents</link>).  Instead,
use the commands <literal>\a=</literal>, <literal>\a'</literal> and <literal>\a`</literal>.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\kill</primary></indexterm><literal>\kill</literal>
</term><listitem><para>Sets tab stops without producing text.  Works just like <literal>\\</literal> except
that it throws away the current line instead of producing output for it.
Any <literal>\=</literal>, <literal>\+</literal> or <literal>\-</literal> commands in that line remain in
effect.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\poptabs</primary></indexterm><literal>\poptabs</literal>
</term><listitem><indexterm role="fn"><primary>\poptabs</primary></indexterm>
<para>Restores the tab stop positions saved by the last <literal>\pushtabs</literal>.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\pushtabs</primary></indexterm><literal>\pushtabs</literal>
</term><listitem><para>Saves all current tab stop positions. Useful for temporarily changing
tab stop positions in the middle of a <literal>tabbing</literal> environment.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\tabbingsep</primary></indexterm><literal>\tabbingsep</literal>
</term><listitem><para>Distance of the text moved by <literal>\'</literal> to left of current tab stop.
</para>
</listitem></varlistentry></variablelist>
<para>This example typesets a Pascal function:
</para>
<screen>\begin{tabbing}
function \= fact(n : integer) : integer;\\
         \&gt; begin \= \+ \\
               \&gt; if \= n &gt; 1 then \+ \\
                        fact := n * fact(n-1) \- \\
                  else \+ \\
                        fact := 1; \-\- \\
            end;\\
\end{tabbing}
</screen>
<para>The output looks like this.
</para>
<screen>function fact(n : integer) : integer;
         begin
               if n &gt; 1 then
                  fact := n * fact(n-1);
               else
                  fact := 1;
         end;
</screen>
<indexterm role="cp"><primary>package, <literal>algorithm2e</literal></primary></indexterm>
<indexterm role="cp"><primary><literal>algorithm2e</literal> package</primary></indexterm>
<indexterm role="cp"><primary>package, <literal>listings</literal></primary></indexterm>
<indexterm role="cp"><primary><literal>listings</literal> package</primary></indexterm>
<indexterm role="cp"><primary>package, <literal>minted</literal></primary></indexterm>
<indexterm role="cp"><primary><literal>minted</literal> package</primary></indexterm>

<indexterm role="cp"><primary>package, <literal>fancyvrb</literal></primary></indexterm>
<indexterm role="cp"><primary><literal>fancyvrb</literal> package</primary></indexterm>

<para>This example is just for illustration of the environment.  To actually
typeset computer code in typewriter like this, a verbatim environment
(see <link linkend="verbatim">verbatim</link>) would normally be best.  For pretty-printed code,
there are quite a few packages, including <literal>algorithm2e</literal>,
<literal>fancyvrb</literal>, <literal>listings</literal>, and <literal>minted</literal>.
</para>
        
</sect1>
<sect1 label="8.22" id="table">
<title><literal>table</literal></title>

<indexterm role="fn"><primary>environment, <literal>table</literal></primary></indexterm>
<indexterm role="fn"><primary><literal>table</literal> environment</primary></indexterm>
        
<indexterm role="cp"><primary>tables, creating</primary></indexterm>
<indexterm role="cp"><primary>creating tables</primary></indexterm>

<para>Synopsis:
</para>
<screen>\begin{table}[<replaceable>placement</replaceable>]
  <replaceable>table body</replaceable>
  \caption[<replaceable>loftitle</replaceable>]{<replaceable>title</replaceable>}  % optional
  \label{<replaceable>label}</replaceable>              % also optional
\end{table}
</screen>
<para>A class of floats (see <link linkend="Floats">Floats</link>).  They cannot be split across pages
and so they are not typeset in sequence with the normal text but instead
are floated to a convenient place, such as the top of a following page.
</para>
<para>This example <literal>table</literal> environment contains a <literal>tabular</literal>
</para>
<screen>\begin{table}
  \centering\small
  \begin{tabular}{ll}
    \multicolumn{1}{c}{\textit{Author}}
      &amp;\multicolumn{1}{c}{\textit{Piece}}  \\ \hline
    Bach            &amp;Cello Suite Number 1  \\
    Beethoven       &amp;Cello Sonata Number 3 \\
    Brahms          &amp;Cello Sonata Number 1
  \end{tabular}
  \caption{Top cello pieces}
  \label{tab:cello}
\end{table}
</screen>
<para>but you can put many different kinds of content in a <literal>table</literal>,
including text, &latex; commands, etc.
</para>
<para>For the possible values of <replaceable>placement</replaceable> and their effect on the 
float placement algorithm, see <link linkend="Floats">Floats</link>.
</para>
<para>The table body is typeset in a <literal>parbox</literal> of width <literal>\textwidth</literal>.
It can contain text, commands, graphics, etc.
</para>
<para>The label is optional; it is used for cross references (see <link linkend="Cross-references">Cross
references</link>).  
<indexterm role="fn"><primary>\caption</primary></indexterm>
The <literal>\caption</literal> command is also optional.  It specifies caption text
for the table.  By default it is numbered.  If its optional
<replaceable>lottitle</replaceable> is present then that text is used in the list of tables
instead of <replaceable>title</replaceable> (see <link linkend="Table-of-contents-etc_002e">Table of contents etc.</link>).
</para>
<para>In this example the table and caption will float to the bottom of a page,
unless it is pushed to a float page at the end.
</para>
<screen>\begin{table}[b]
  \centering
  \begin{tabular}{r|p{2in}} \hline
    One &amp;The loneliest number \\
    Two &amp;Can be as sad as one.
         It's the loneliest number since the number one.
  \end{tabular}
  \caption{Cardinal virtues}
  \label{tab:CardinalVirtues}
\end{table}
</screen>

</sect1>
<sect1 label="8.23" id="tabular">
<title><literal>tabular</literal></title>

<indexterm role="fn"><primary>environment, <literal>tabular</literal></primary></indexterm>
<indexterm role="fn"><primary><literal>tabular</literal> environment</primary></indexterm>
        
<indexterm role="cp"><primary>lines in tables</primary></indexterm>
<indexterm role="cp"><primary>lining text up in tables</primary></indexterm>

<para>Synopsis:
</para>
<screen>\begin{tabular}[<replaceable>pos</replaceable>]{<replaceable>cols</replaceable>}
  <replaceable>column 1 entry</replaceable>  &amp;<replaceable>column 2 entry</replaceable>  ...  &amp;<replaceable>column n entry</replaceable> \\
  ...
\end{tabular}
</screen>
<para>or
</para>
<screen>\begin{tabular*}{<replaceable>width</replaceable>}[<replaceable>pos</replaceable>]{<replaceable>cols</replaceable>}
  <replaceable>column 1 entry</replaceable>  &amp;<replaceable>column 2 entry</replaceable>  ...  &amp;<replaceable>column n entry</replaceable> \\
  ...
\end{tabular*}
</screen>
<para>Produce a table, a box consisting of a sequence of horizontal rows.
Each row consists of items that are aligned vertically in columns.  This
illustrates many of the features.
</para>
<screen>\begin{tabular}{l|l}
  \textit{Player name}  &amp;\textit{Career home runs}  \\ 
  \hline
  Hank Aaron  &amp;755 \\
  Babe Ruth   &amp;714
\end{tabular}
</screen>
<para>The output will have two left-aligned columns with a vertical bar
between them.  This is specified in <literal>tabular</literal>&#8217;s argument
<literal>{l|l}</literal>.
<indexterm role="fn"><primary>&amp;</primary></indexterm>
Put the entries into different columns by separating them with an
ampersand, <literal>&amp;</literal>.  The end of each row is marked with a double
backslash, <literal>\\</literal>.  Put a horizontal rule below a row, after a double
backslash, with <literal>\hline</literal>.
<indexterm role="fn"><primary>\\ for <literal>tabular</literal></primary></indexterm>
This <literal>\\</literal> is optional after the last row unless an <literal>\hline</literal>
command follows, to put a rule below the table.
</para>
<para>The required and optional arguments to <literal>tabular</literal> consist of:
</para>
<variablelist><varlistentry><term><replaceable>pos</replaceable>
</term><listitem><para>Optional.  Specifies the table&#8217;s vertical position.  The default is to
align the table so its vertical center matches the baseline of the
surrounding text.  There are two other possible alignments: <literal>t</literal>
aligns the table so its top row matches the baseline of the surrounding
text, and <literal>b</literal> aligns on the bottom row.
</para>
<para>This only has an effect if there is other text.  In the common case of a
<literal>tabular</literal> alone in a <literal>center</literal> environment this option makes
no difference.
</para>
</listitem></varlistentry><varlistentry><term><replaceable>cols</replaceable>
</term><listitem><para>Required.  Specifies the formatting of columns.  It consists of a
sequence of the following specifiers, corresponding to the types of
column and intercolumn material.
</para>
<variablelist><varlistentry><term><literal>l</literal>
</term><listitem><para>A column of left-aligned items.
</para>
</listitem></varlistentry><varlistentry><term><literal>r</literal>
</term><listitem><para>A column of right-aligned items.
</para>
</listitem></varlistentry><varlistentry><term><literal>c</literal>
</term><listitem><para>A column of centered items.
</para>
</listitem></varlistentry><varlistentry><term><literal>|</literal>
</term><listitem><para>A vertical line the full height and depth of the environment.
</para>
</listitem></varlistentry><varlistentry><term><literal>@{<replaceable>text or space</replaceable>}</literal>
</term><listitem><para>Insert <replaceable>text or space</replaceable> at this location in every row.  The <replaceable>text
or space</replaceable> material is typeset in LR mode.  This text is fragile
(see <link linkend="_005cprotect">\protect</link>).
</para>
<para>If between two columns there is no @-expression then &latex;&#8217;s
<literal>book</literal>, <literal>article</literal>, and <literal>report</literal> classes will put on
either side of each column a space of length <literal>\tabcolsep</literal>, which by
default is 6pt.  That is, by default adjacent columns are
separated by 12pt (so <literal>\tabcolsep</literal> is misleadingly named
since it is only half of the separation between tabular columns).  In
addition, a space of 6pt also comes before the first column and
after the final column, unless you put a <literal>@{...}</literal> or <literal>|</literal>
there.
</para>
<para>If you override the default and use an @-expression then &latex; does
not insert <literal>\tabcolsep</literal> so you must insert any desired space
yourself, as in <literal>@{\hspace{1em}}</literal>.
</para>
<para>An empty expression <literal>@{}</literal> will eliminate the space.  In
particular, sometimes you want to eliminate the the space before the
first column or after the last one, as in the example below where the
tabular lines need to lie on the left margin.
</para>
<screen>\begin{flushleft}
  \begin{tabular}{@{}l}
    ...
  \end{tabular}
\end{flushleft}
</screen>
<para>The next example shows text, a decimal point between the columns,
arranged so the numbers in the table are aligned on it.
</para>
<screen>\begin{tabular}{r@{$.$}l}
  $3$ &amp;$14$  \\
  $9$ &amp;$80665$
\end{tabular}
</screen>
<indexterm role="fn"><primary>\extracolsep</primary></indexterm>
<para>An <literal>\extracolsep{<replaceable>wd</replaceable>}</literal> command in an @-expression causes an
extra space of width <replaceable>wd</replaceable> to appear to the left of all subsequent
columns, until countermanded by another <literal>\extracolsep</literal>.  Unlike
ordinary intercolumn space, this extra space is not suppressed by an
@-expression.  An <literal>\extracolsep</literal> command can be used only in an
@-expression in the <literal>cols</literal> argument.  Below, &latex; inserts the
right amount of intercolumn space to make the entire table 4 inches
wide.
</para>
<screen>\begin{tabular*}{4in}{l@{\extracolsep{\fill}}l}
  Seven times down, eight times up \ldots 
  &amp;such is life!
\end{tabular*}
</screen>
<para>To insert commands that are automatically executed before a given
column, load the <literal>array</literal> package and use the <literal>&gt;{...}</literal>
specifier.
<!-- xx should fully explain array, tabularx, and all other base packages... -->
</para>
</listitem></varlistentry><varlistentry><term><literal>p{<replaceable>wd</replaceable>}</literal>
</term><listitem><para>Each item in the column is typeset in a parbox of width <replaceable>wd</replaceable>, as if
it were the argument of a <literal>\parbox[t]{wd}{...}</literal> command.
</para>
<para>A line break double backslash <literal>\\</literal> may not appear in the item,
except inside an environment like <literal>minipage</literal>, <literal>array</literal>, or
<literal>tabular</literal>, or inside an explicit <literal>\parbox</literal>, or in the scope of
a <literal>\centering</literal>, <literal>\raggedright</literal>, or <literal>\raggedleft</literal>
declaration (when used in a <literal>p</literal>-column element these declarations
must appear inside braces, as with <literal>{\centering .. \\
..}</literal>). Otherwise &latex; will misinterpret the double backslash as
ending the row.  Instead, to get a line break in there use
<literal>\newline</literal> (see <link linkend="_005cnewline">\newline</link>).
</para>
</listitem></varlistentry><varlistentry><term><literal>*{<replaceable>num</replaceable>}{<replaceable>cols</replaceable>}</literal>
</term><listitem><para>Equivalent to <replaceable>num</replaceable> copies of <replaceable>cols</replaceable>, where <replaceable>num</replaceable> is a
positive integer and <replaceable>cols</replaceable> is a list of specifiers.  Thus the
specifier <literal>\begin{tabular}{|*{3}{l|r}|}</literal> is equivalent to
the specifier <literal>\begin{tabular}{|l|rl|rl|r|}</literal>.  Note that
<replaceable>cols</replaceable> may contain another <literal>*</literal>-expression.
</para> 
</listitem></varlistentry></variablelist>
</listitem></varlistentry><varlistentry><term><replaceable>width</replaceable>
</term><listitem><para>Required for <literal>tabular*</literal>, not allowed for <literal>tabular</literal>. Specifies
the width of the <literal>tabular*</literal> environment.  The space between columns
should be rubber, as with <literal>@{\extracolsep{\fill}}</literal>, to allow
the table to stretch or shrink to make the specified width, or else you
are likely to get the <literal>Underfull \hbox (badness 10000) in alignment
...</literal> warning.
</para>
</listitem></varlistentry></variablelist>
<para>Parameters that control formatting:
<!-- xx defaults, own node (xref from array)? -->
</para>
<variablelist><varlistentry><term><indexterm role="fn"><primary>\arrayrulewidth</primary></indexterm><literal>\arrayrulewidth</literal>
</term><listitem><anchor id="tabular-arrayrulewidth"/><para>A length that is the thickness of the rule created by <literal>|</literal>,
<literal>\hline</literal>, and <literal>\vline</literal> in the <literal>tabular</literal> and <literal>array</literal>
environments.  The default is &#8216;<literal>.4pt</literal>&#8217;. Change it as in
<literal>\setlength{\arrayrulewidth}{0.8pt}</literal>.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\arraystretch</primary></indexterm><literal>\arraystretch</literal>
</term><listitem><anchor id="tabular-arraystrech"/><para>A factor by which the spacing between rows in the <literal>tabular</literal> and
<literal>array</literal> environments is multiplied.  The default is &#8216;<literal>1</literal>&#8217;, for
no scaling.  Change it as <literal>\renewcommand{\arraystretch}{1.2}</literal>.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\doublerulesep</primary></indexterm><literal>\doublerulesep</literal>
</term><listitem><anchor id="tabular-doublerulesep"/><para>A length that is the distance between the vertical rules produced by the
<literal>||</literal> specifier.  The default is &#8216;<literal>2pt</literal>&#8217;.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\tabcolsep</primary></indexterm><literal>\tabcolsep</literal>
</term><listitem><anchor id="tabular-tabcolsep"/><para>A length that is half of the space between columns. The default is
&#8216;<literal>6pt</literal>&#8217;.  Change it with <literal>\setlength</literal>.
</para>
</listitem></varlistentry></variablelist>
<para>The following commands can be used inside the body of a <literal>tabular</literal>
environment, the first two inside an entry and the second two between
lines:
</para>


<sect2 label="8.23.1" id="_005cmulticolumn">
<title><literal>\multicolumn</literal></title>

<indexterm role="fn"><primary>\multicolumn</primary></indexterm>

<para>Synopsis:
</para>
<screen>\multicolumn{<replaceable>numcols</replaceable>}{<replaceable>cols</replaceable>}{<replaceable>text</replaceable>}
</screen>
<para>Make an <literal>array</literal> or <literal>tabular</literal> entry that spans several columns.
The first argument <replaceable>numcols</replaceable> gives the number of columns to span.
The second argument <replaceable>cols</replaceable> specifies the formatting of the entry,
with <literal>c</literal> for centered, <literal>l</literal> for flush left, or <literal>r</literal> for
flush right.  The third argument <replaceable>text</replaceable> gives the contents of that
entry.
</para>
<para>In this example, in the first row, the second and third columns are
spanned by the single heading &#8216;<literal>Name</literal>&#8217;.
</para>
<screen>\begin{tabular}{lccl} 
  \textit{ID}       &amp;\multicolumn{2}{c}{\textit{Name}} &amp;\textit{Age} \\
  \hline  
  978-0-393-03701-2 &amp;O'Brian &amp;Patrick                  &amp;55           \\
    ...
\end{tabular}
</screen>
<para>What counts as a column is:&#160;the column format specifier for the
<literal>array</literal> or <literal>tabular</literal> environment is broken into parts, where
each part (except the first) begins with <literal>l</literal>, <literal>c</literal>, <literal>r</literal>,
or&#160;<literal>p</literal>.  So from <literal>\begin{tabular}{|r|ccp{1.5in}|}</literal>
the parts are <literal>|r|</literal>, <literal>c</literal>, <literal>c</literal>,
and&#160;<literal>p{1.5in}|</literal>.
</para>
<para>The <replaceable>cols</replaceable> argument overrides the <literal>array</literal> or <literal>tabular</literal>
environment&#8217;s intercolumn area default adjoining this multicolumn
entry. To affect that area, this argument can contain vertical bars
<literal>|</literal> indicating the placement of vertical rules, and <literal>@{...}</literal>
expressions.  Thus if <replaceable>cols</replaceable> is &#8216;<literal>|c|</literal>&#8217; then this multicolumn
entry will be centered and a vertical rule will come in the intercolumn
area before it and after it.  This table details the exact behavior.
</para>
<screen>\begin{tabular}{|cc|c|c|}
  \multicolumn{1}{r}{w}       % entry one
    &amp;\multicolumn{1}{|r|}{x}  % entry two 
    &amp;\multicolumn{1}{|r}{y}   % entry three
    &amp;z                        % entry four
\end{tabular}
</screen>
<para>Before the first entry the output will not have a vertical rule because
the <literal>\multicolumn</literal> has the <replaceable>cols</replaceable> specifier &#8216;<literal>r</literal>&#8217; with no
initial vertical bar.  Between entry one and entry two there will be a
vertical rule; although the first <replaceable>cols</replaceable> does not have an ending
vertical bar, the second <replaceable>cols</replaceable> does have a starting one.  Between
entry two and entry three there is a single vertical rule; despite that
the <replaceable>cols</replaceable> in both of the surrounding <literal>multicolumn</literal>&#8217;s call for
a vertical rule, you only get one rule.  Between entry three and entry
four there is no vertical rule; the default calls for one but the
<replaceable>cols</replaceable> in the entry three <literal>\multicolumn</literal> leaves it out, and
that takes precedence.  Finally, following entry four there is a
vertical rule because of the default.
</para>
<para>The number of spanned columns <replaceable>numcols</replaceable> can be 1.  Besides giving
the ability to change the horizontal alignment, this also is useful to
override for one row the <literal>tabular</literal> definition&#8217;s default intercolumn
area specification, including the placement of vertical rules.
</para>
<para>In the example below, in the <literal>tabular</literal> definition the first column
is specified to default to left justified but in the first row the entry
is centered with <literal>\multicolumn{1}{c}{\textsc{Period}}</literal>.
Also in the first row, the second and third columns are spanned by a
single entry with <literal>\multicolumn{2}{c}{\textsc{Span}}</literal>,
overriding the specification to center those two columns on the page
range en-dash.
</para>
<screen>\begin{tabular}{l|r@{--}l} 
  \multicolumn{1}{c}{\textsc{Period}}  
    &amp;multicolumn{2}{c}{\textsc{Span}} \\ \hline
  Baroque          &amp;1600           &amp;1760         \\
  Classical        &amp;1730           &amp;1820         \\
  Romantic         &amp;1780           &amp;1910         \\
  Impressionistic  &amp;1875           &amp;1925
\end{tabular}
</screen>
<para>Note that although the <literal>tabular</literal> specification by default puts a
vertical rule between the first and second columns, because there is no
vertical bar in the <replaceable>cols</replaceable> of either of the first row&#8217;s
<literal>\multicolumn</literal> commands, no rule appears in the first row.
</para>

</sect2>
<sect2 label="8.23.2" id="_005cvline">
<title><literal>\vline</literal></title>

<indexterm role="fn"><primary>\vline</primary></indexterm>
<para>Draw a vertical line in a <literal>tabular</literal> or <literal>array</literal> environment
extending the full height and depth of an entry&#8217;s row.  Can also be
used in an @-expression, although its synonym vertical
bar&#160;<literal>|</literal> is more common.  This command is rarely used in the
body of a table; typically a table&#8217;s vertical lines are specified in
<literal>tabular</literal>&#8217;s <replaceable>cols</replaceable> argument and overridden as needed with
<literal>\multicolumn</literal> (see <link linkend="tabular">tabular</link>).
</para>
<para>The example below illustrates some pitfalls.  In the first row&#8217;s second
entry the <literal>\hfill</literal> moves the <literal>\vline</literal> to the left edge of the
cell.  But that is different than putting it halfway between the two
columns, so between the first and second columns there are two vertical
rules, with the one from the <literal>{c|cc}</literal> specifier coming before the
one produced by the <literal>\vline\hfill</literal>.  In contrast, the first row&#8217;s
third entry shows the usual way to put a vertical bar between two
columns.  In the second row, the <literal>ghi</literal> is the widest entry in its
column so in the <literal>\vline\hfill</literal> the <literal>\hfill</literal> has no effect and
the vertical line in that entry appears immediately next to the
<literal>g</literal>, with no whitespace.
</para>
<screen>\begin{tabular}{c|cc}
  x   &amp;\vline\hfill y   &amp;\multicolumn{1}{|r}{z} \\ % row 1  
  abc &amp;def &amp;\vline\hfill ghi                       % row 2  
\end{tabular}
</screen>

</sect2>
<sect2 label="8.23.3" id="_005ccline">
<title><literal>\cline</literal></title>

<indexterm role="fn"><primary>\cline</primary></indexterm>

<para>Synopsis:
</para>
<screen>\cline{<replaceable>i</replaceable>-<replaceable>j</replaceable>}
</screen>
<para>In an <literal>array</literal> or <literal>tabular</literal> environment, draw a horizontal rule
beginning in column&#160;<replaceable>i</replaceable> and ending in column&#160;<replaceable>j</replaceable>. The
dash, <literal>-</literal>, must appear in the mandatory argument. To span a single
column use the number twice, as with <literal>\cline{2-2}</literal>.
</para>
<para>This example puts two horizontal lines between the first and second
rows, one line in the first column only, and the other spanning the
third and fourth columns.  The two lines are side-by-side, at the same
height.
</para>
<screen>\begin{tabular}{llrr} 
  a &amp;b &amp;c &amp;d \\ \cline{1-1} \cline{3-4} 
  e &amp;f &amp;g &amp;h 
\end{tabular}
</screen>

</sect2>
<sect2 label="8.23.4" id="_005chline">
<title><literal>\hline</literal></title>

<indexterm role="fn"><primary>\hline</primary></indexterm>

<para>Draw a horizontal line the width of the enclosing <literal>tabular</literal> or
<literal>array</literal> environment.  It&#8217;s most commonly used to draw a line at the
top, bottom, and between the rows of a table.
</para>
<para>In this example the top of the table has two horizontal rules, one above
the other, that span both columns.  The bottom of the table has a single
rule spanning both columns.  Because of the <literal>\hline</literal>, the
<literal>tabular</literal> second row&#8217;s line ending double backslash&#160;<literal>\\</literal>
is required.
</para>
<screen>\begin{tabular}{ll} \hline\hline
  Baseball   &amp;Red Sox  \\
  Basketball &amp;Celtics  \\ \hline
\end{tabular}
</screen>

</sect2>
</sect1>
<sect1 label="8.24" id="thebibliography">
<title><literal>thebibliography</literal></title>

<indexterm role="fn"><primary>environment, <literal>thebibliography</literal></primary></indexterm>
<indexterm role="fn"><primary><literal>thebibliography</literal> environment</primary></indexterm>
        
<indexterm role="cp"><primary>bibliography, creating (manually)</primary></indexterm>

<para>Synopsis:
</para>
<screen>\begin{thebibliography}{<replaceable>widest-label</replaceable>}
  \bibitem[<replaceable>label</replaceable>]{<replaceable>cite_key}</replaceable>
  ...
\end{thebibliography}
</screen>
<para>Produce a bibliography or reference list.  There are two ways to produce
bibliographic lists.  This environment is suitable when you have only a
few references and can maintain the list by hand.  See <link linkend="Using-BibTeX">Using BibTeX</link>
for a more sophisticated approach.
</para>
<para>This shows the environment with two entries.
</para>
<screen>This work is based on \cite{latexdps}.
Together they are \cite{latexdps, texbook}.
  ...
\begin{thebibliography}{9}
\bibitem{latexdps} 
  Leslie Lamport. 
  \textit{\LaTeX{}: a document preparation system}. 
  Addison-Wesley, Reading, Massachusetts, 1993.
\bibitem{texbook} 
  Donald Ervin Knuth. 
  \textit{The \TeX book}. 
  Addison-Wesley, Reading, Massachusetts, 1983.
\end{thebibliography}
</screen>
<para>This styles the first reference as &#8216;<literal>[1] Leslie ...</literal>&#8217;, and so that
<literal>\cite{latexdps}</literal> produces the matching &#8216;<literal>... based on [1]</literal>&#8217;.
The second <literal>\cite</literal> produces &#8216;<literal>[1, 2]</literal>&#8217;.  You must compile the
document twice to resolve these references.
</para>
<para>The mandatory argument <replaceable>widest-label</replaceable> is text that, when typeset, is
as wide as the widest item label produced by the <literal>\bibitem</literal>
commands.  The tradition is to use <literal>9</literal> for bibliographies with less
than 10 references, <literal>99</literal> for ones with less than 100, etc.
</para>
<para>The bibliographic list is headed by a title such as &#8216;<literal>Bibliography</literal>&#8217;.
To change it there are two cases.  In the <filename>book</filename> and <filename>report</filename>
classes, where the top level sectioning is <literal>\chapter</literal> and the
default title is &#8216;<literal>Bibliography</literal>&#8217;, that title is in the macro
<literal>\bibname</literal>.  For <filename>article</filename>, where the class&#8217;s top level
sectioning is <literal>\section</literal> and the default is &#8216;<literal>References</literal>&#8217;, the
title is in macro <literal>\refname</literal>.  Change it by redefining the command,
as with <literal>\renewcommand{\refname}{Cited references}</literal> after
<literal>\begin{document}</literal>.
</para>
<indexterm role="cp"><primary>package, <literal>babel</literal></primary></indexterm>
<indexterm role="cp"><primary><literal>babel</literal> package</primary></indexterm>

<para>Language support packages such as <filename>babel</filename> will automatically
redefine <literal>\refname</literal> or <literal>\bibname</literal> to fit the selected
language.
</para>


<sect2 label="8.24.1" id="_005cbibitem">
<title><literal>\bibitem</literal></title>

<indexterm role="fn"><primary>\bibitem</primary></indexterm>

<para>Synopsis:
</para>
<screen>\bibitem{<replaceable>cite_key</replaceable>}
</screen>
<para>or
</para>
<screen>\bibitem[<replaceable>label</replaceable>]{<replaceable>cite_key</replaceable>}
</screen>
<para>Generate an entry labeled by <replaceable>label</replaceable>.  The default is for &latex; to
generates a number using the <literal>enumi</literal> counter.  The <firstterm>citation key</firstterm> 
<indexterm role="cp"><primary>citation key</primary></indexterm>
<replaceable>cite_key</replaceable> is a string of
letters, numbers, and punctuation symbols (but not comma).
</para>
<para>See <link linkend="thebibliography">thebibliography</link> for an example.
</para>
<para>The optional <replaceable>label</replaceable> changes the default label from an integer to the
given string.  With this
</para>
<screen>\begin{thebibliography}
\bibitem[Lamport 1993]{latexdps} 
  Leslie Lamport. 
  \textit{\LaTeX{}: a document preparation system}. 
  Addison-Wesley, Reading, Massachusetts, 1993.
\bibitem{texbook} 
  Donald Ervin Knuth. 
  \textit{The \TeX book}. 
  Addison-Wesley, Reading, Massachusetts, 1983.
\end{thebibliography}
</screen>
<para>the first entry will be styled as &#8216;<literal>[Lamport 1993] Leslie ...</literal>&#8217;  (The
amount of horizontal space that &latex; leaves for the label depends on
the <replaceable>widest-label</replaceable> argument of the <literal>thebibliography</literal>
environment; see <link linkend="thebibliography">thebibliography</link>.)  Similarly, <literal>... based on
\cite{latexdps}</literal> will produce &#8216;<literal>... based on [Lamport 1994]</literal>&#8217;.
</para>
<para>If you mix <literal>\bibitem</literal> entries having a <replaceable>label</replaceable> with those that
do not then &latex; will number the unlabelled ones sequentially.  In
the example above the <literal>texbook</literal> entry will appear as &#8216;<literal>[1]
Donald ...</literal>&#8217;, despite that it is the second entry.
</para>
<para>If you use the same <replaceable>cite_key</replaceable> twice then you get &#8216;<literal>LaTeX
Warning: There were multiply-defined labels</literal>&#8217;.
</para>
<para>Under the hood, &latex; remembers the <replaceable>cite_key</replaceable> and <replaceable>label</replaceable>
information because <literal>\bibitem</literal> writes it to the auxiliary file
<filename><replaceable>filename</replaceable>.aux</filename>.  For instance, the above example causes
<literal>\bibcite{latexdps}{Lamport, 1993}</literal> and
<literal>\bibcite{texbook}{1}</literal> to appear in that file.  The <filename>.aux</filename>
file is read by the <literal>\begin{document}</literal> command and then the
information is available for <literal>\cite</literal> commands.  This explains why
you need to run &latex; twice to resolve references: once to write it
out and once to read it in.
</para>
<para>Because of this two-pass algorithm, when you add a <literal>\bibitem</literal> or
change its <replaceable>cite_key</replaceable> you may get &#8216;<literal>LaTeX Warning: Label(s) may
have changed. Rerun to get cross-references right</literal>&#8217;.  Fix it by
recompiling.
</para>

</sect2>
<sect2 label="8.24.2" id="_005ccite">
<title><literal>\cite</literal></title>

<indexterm role="fn"><primary>\cite</primary></indexterm>

<para>Synopsis:
</para>
<screen>\cite{<replaceable>keys</replaceable>}
</screen>
<para>or
</para>
<screen>\cite[<replaceable>subcite</replaceable>]{<replaceable>keys</replaceable>}
</screen>
<para>Generate as output a citation to the references associated with
<replaceable>keys</replaceable>.  The mandatory <replaceable>keys</replaceable> is a citation key, or a
comma-separated list of citation keys (see <link linkend="_005cbibitem">\bibitem</link>).
</para>
<para>This 
</para>
<screen>The ultimate source is \cite{texbook}.
  ...
\begin{thebibliography}
\bibitem{texbook} 
  Donald Ervin Knuth. 
  \textit{The \TeX book}. 
  Addison-Wesley, Reading, Massachusetts, 1983.
\end{thebibliography}
</screen>
<para>produces the output &#8216;<literal>... source is [1]</literal>&#8217;.
</para>
<para>The optional argument <replaceable>subcite</replaceable> is appended to the citation.  For
example, <literal>See 14.3 in \cite[p.~314]{texbook}</literal> might produce
&#8216;<literal>See 14.3 in [1, p.&#160;314]</literal>&#8217;.
</para>
<para>If <replaceable>keys</replaceable> is not in your bibliography information then you get
&#8216;<literal>LaTeX Warning: There were undefined references</literal>&#8217;, and in the output
the citation shows as a boldface question mark between square brackets.
There are two possible causes.  If you have mistyped something, as in
<literal>\cite{texbok}</literal> then you need to correct the spelling.  On the
other hand, if you have just added or modified the bibliographic
information and so changed the <filename>.aux</filename> file (see <link linkend="_005cbibitem">\bibitem</link>) then
the fix may be to just run &latex; again.
</para>
<para>In addition to what appears in the output, <literal>\cite</literal> writes
information to the auxiliary file <filename><replaceable>filename</replaceable>.aux</filename>.  For
instance, <literal>\cite{latexdps}</literal> writes &#8216;<literal>\citation{latexdps}</literal>&#8217;
to that file.  This information is used by Bib&tex; to include in your
reference list only those works that you have actually cited; see
<link linkend="_005cnocite">\nocite</link> also.
</para>

</sect2>
<sect2 label="8.24.3" id="_005cnocite">
<title><literal>\nocite</literal></title>

<indexterm role="fn"><primary>\nocite</primary></indexterm>

<para>Synopsis:
</para>
<screen><literal>\nocite{<replaceable>keys</replaceable>}</literal>
</screen>
<para>Produces no output but writes <replaceable>keys</replaceable> to the auxiliary file
<filename><replaceable>doc-filename</replaceable>.aux</filename>.
</para>
<para>The mandatory argument <replaceable>keys</replaceable> is a comma-separated list of one or
more citation keys (see <link linkend="_005cbibitem">\bibitem</link>).  This information is used by
Bib&tex; to include these works in your reference list even though you
have not cited them (see <link linkend="_005ccite">\cite</link>).
</para>

</sect2>
<sect2 label="8.24.4" id="Using-BibTeX">
<title>Using Bib&tex;</title>

<indexterm role="cp"><primary>using Bib&tex;</primary></indexterm>
<indexterm role="cp"><primary>bib&tex;, using</primary></indexterm>
<indexterm role="cp"><primary>bibliography, creating (automatically)</primary></indexterm>
<indexterm role="fn"><primary>\bibliographystyle</primary></indexterm>
<indexterm role="fn"><primary>\bibliography</primary></indexterm>

<para>As described in <literal>thebibliography</literal> (see <link linkend="thebibliography">thebibliography</link>), a
sophisticated approach to managing bibliographies is provided by the
Bib&tex; program.  This is only an introduction; see the full
documentation on CTAN.
</para>
<para>With Bib&tex;, you don&#8217;t use <literal>thebibliography</literal>
(see <link linkend="thebibliography">thebibliography</link>). Instead, include these lines.
</para>
<screen>\bibliographystyle{<replaceable>bibstyle</replaceable>}
\bibliography{<replaceable>bibfile1</replaceable>, <replaceable>bibfile2</replaceable>, ...}
</screen>
<para>The <replaceable>bibstyle</replaceable> refers to a file <filename><replaceable>bibstyle</replaceable>.bst</filename>, which
defines how your citations will look.  The standard <replaceable>bibstyle</replaceable>&#8217;s
distributed with Bib&tex; are:
</para>
<variablelist><varlistentry><term><literal>alpha</literal>
</term><listitem><para>Labels are formed from name of author and year of publication.
The bibliographic items are sorted alphabetically.
</para></listitem></varlistentry><varlistentry><term><literal>plain</literal>
</term><listitem><para>Labels are integers.
Sort the  bibliographic items alphabetically.
</para></listitem></varlistentry><varlistentry><term><literal>unsrt</literal>
</term><listitem><para>Like <literal>plain</literal>, but entries are in order of citation.
</para></listitem></varlistentry><varlistentry><term><literal>abbrv</literal>
</term><listitem><para>Like <literal>plain</literal>, but more compact labels.
</para></listitem></varlistentry></variablelist>
<para>Many, many other Bib&tex; style files exist,
tailored to the demands of various publications.  See CTAN&#8217;s listing
<ulink url="http://mirror.ctan.org/biblio/bibtex/contrib">http://mirror.ctan.org/biblio/bibtex/contrib</ulink>.
</para>
<para>The <literal>\bibliography</literal> command is what actually produces the
bibliography.  Its argument is a comma-separated list, referring to
files named <filename><replaceable>bibfile1</replaceable>.bib</filename>, <filename><replaceable>bibfile2</replaceable>.bib</filename>,
&#8230; These contain your database in Bib&tex; format.  This shows a
typical couple of entries in that format.
</para>
<screen>@book{texbook,
  title     = {The {{\TeX}}book},
  author    = {D.E. Knuth},
  isbn      = {0201134489},
  series    = {Computers \&amp; typesetting},
  year      = {1983},
  publisher = {Addison-Wesley}
}
@book{sexbook,
    author    = {W.H. Masters and V.E. Johnson},
    title     = {Human Sexual Response},
    year      = {1966},
    publisher = {Bantam Books}
}
</screen>
<para>Only the bibliographic entries referred to via <literal>\cite</literal> and
<literal>\nocite</literal> will be listed in the document&#8217;s bibliography.  Thus you
can keep all your sources together in one file, or a small number of
files, and rely on Bib&tex; to include in this document only those that
you used.
</para>

</sect2>
</sect1>
<sect1 label="8.25" id="theorem">
<title><literal>theorem</literal></title>

<indexterm role="fn"><primary>environment, <literal>theorem</literal></primary></indexterm>
<indexterm role="fn"><primary><literal>theorem</literal> environment</primary></indexterm>
        
<indexterm role="cp"><primary>theorems, typesetting</primary></indexterm>

<para>Synopsis:
</para>
<screen>\begin{theorem}
  <replaceable>theorem body</replaceable>
\end{theorem}
</screen>
<para>Produces &#8216;<literal>Theorem <replaceable>n</replaceable></literal>&#8217; in boldface followed by <replaceable>theorem
body</replaceable> in italics.  The numbering possibilities for <replaceable>n</replaceable> are described under
<literal>\newtheorem</literal> (see <link linkend="_005cnewtheorem">\newtheorem</link>).
</para>
<screen>\newtheorem{lem}{Lemma}      % in preamble
\newtheorem{thm}{Theorem}    
  ...
\begin{lem}                  % in  document body
  text of lemma
\end{lem}

The next result follows immediately.
\begin{thm}[Gauss]   %  put `Gauss' in parens after theorem head
  text of theorem
\end{thm}
</screen>
<indexterm role="cp"><primary>package, <literal>amsmath</literal></primary></indexterm>
<indexterm role="cp"><primary><literal>amsmath</literal> package</primary></indexterm>

<indexterm role="cp"><primary>package, <literal>amsthm</literal></primary></indexterm>
<indexterm role="cp"><primary><literal>amsthm</literal> package</primary></indexterm>

<para>Most new documents use the packages <literal>amsthm</literal> and <literal>amsmath</literal>
from the American Mathematical Society.  Among other things these
packages include a large number of options for theorem environments,
such as styling options.
</para>

</sect1>
<sect1 label="8.26" id="titlepage">
<title><literal>titlepage</literal></title>

<indexterm role="fn"><primary>environment, <literal>titlepage</literal></primary></indexterm>
<indexterm role="fn"><primary><literal>titlepage</literal> environment</primary></indexterm>
        
<indexterm role="cp"><primary>making a title page</primary></indexterm>
<indexterm role="cp"><primary>title pages, creating</primary></indexterm>

<para>Synopsis:
</para>
<screen>\begin{titlepage}
  ... text and spacing ...
\end{titlepage}
</screen>
<para>Create a title page, a page with no printed page number or heading and
with succeeding pages numbered starting with page one.
</para>
<para>In this example all formatting, including vertical spacing, is left to
the author.
</para>
<screen>\begin{titlepage}
\vspace*{\stretch{1}}
\begin{center}
  {\huge\bfseries Thesis \\[1ex] 
                  title}                  \\[6.5ex]
  {\large\bfseries Author name}           \\
  \vspace{4ex}
  Thesis  submitted to                    \\[5pt]
  \textit{University name}                \\[2cm]
  in partial fulfilment for the award of the degree of \\[2cm]
  \textsc{\Large Doctor of Philosophy}    \\[2ex]
  \textsc{\large Mathematics}             \\[12ex]
  \vfill
  Department of Mathematics               \\
  Address                                 \\
  \vfill
  \today
\end{center}
\vspace{\stretch{2}}
\end{titlepage}
</screen>
<para>To instead produce a standard title page without a <literal>titlepage</literal>
environment, use <literal>\maketitle</literal> (see <link linkend="_005cmaketitle">\maketitle</link>).
</para>

</sect1>
<sect1 label="8.27" id="verbatim">
<title><literal>verbatim</literal></title>

<indexterm role="fn"><primary>environment, <literal>verbatim</literal></primary></indexterm>
<indexterm role="fn"><primary><literal>verbatim</literal> environment</primary></indexterm>
        
<indexterm role="cp"><primary>verbatim text</primary></indexterm>
<indexterm role="cp"><primary>simulating typed text</primary></indexterm>
<indexterm role="cp"><primary>typed text, simulating</primary></indexterm>
<indexterm role="cp"><primary>code, typesetting</primary></indexterm>
<indexterm role="cp"><primary>computer programs, typesetting</primary></indexterm>

<para>Synopsis:
</para>
<screen>\begin{verbatim}
<replaceable>literal-text</replaceable>
\end{verbatim}
</screen>
<para>A paragraph-making environment in which &latex; produces as output
exactly what you type as input.  For instance inside <replaceable>literal-text</replaceable>
the backslash&#160;<literal>\</literal> character does not start commands, it
produces a printed &#8216;<literal>\</literal>&#8217;, and carriage returns and blanks are taken
literally.  The output appears in a monospaced typewriter-like font
(<literal>\tt</literal>).
</para>
<screen>\begin{verbatim}
Symbol swearing: %&amp;$#?!.
\end{verbatim}
</screen>
<para>The only restriction on <literal>literal-text</literal> is that it cannot include
the string <literal>\end{verbatim}</literal>.
</para>
<indexterm role="cp"><primary>package, <literal>cprotect</literal></primary></indexterm>
<indexterm role="cp"><primary><literal>cprotect</literal> package</primary></indexterm>

<para>You cannot use the verbatim environment in the argument to macros, for
instance in the argument to a <literal>\section</literal>.  This is not the same as
commands being fragile (see <link linkend="_005cprotect">\protect</link>), instead it just cannot appear
there. (But the <literal>cprotect</literal> package can help with this.)
</para>
<indexterm role="cp"><primary>package, <literal>listings</literal></primary></indexterm>
<indexterm role="cp"><primary><literal>listings</literal> package</primary></indexterm>

<indexterm role="cp"><primary>package, <literal>minted</literal></primary></indexterm>
<indexterm role="cp"><primary><literal>minted</literal> package</primary></indexterm>

<para>One common use of verbatim input is to typeset computer code.  There are
packages that are an improvement the <literal>verbatim</literal> environment.  For
instance, one improvement is to allow the verbatim inclusion of external
files, or parts of those files.  Such packages include <literal>listings</literal>,
and <literal>minted</literal>.
</para>
<indexterm role="cp"><primary>package, <literal>fancyvrb</literal></primary></indexterm>
<indexterm role="cp"><primary><literal>fancyvrb</literal> package</primary></indexterm>

<indexterm role="cp"><primary>package, <literal>verbatimbox</literal></primary></indexterm>
<indexterm role="cp"><primary><literal>verbatimbox</literal> package</primary></indexterm>

<para>A package that provides many more options for verbatim environments is
<literal>fancyvrb</literal>.  Another is <literal>verbatimbox</literal>.
</para>
<para>For a list of all the relevant packages, see CTAN.
</para>


<sect2 label="8.27.1" id="_005cverb">
<title><literal>\verb</literal></title>

<indexterm role="fn"><primary>\verb</primary></indexterm>
<indexterm role="cp"><primary>verbatim text, inline</primary></indexterm>

<para>Synopsis:
</para>
<screen>\verb <replaceable>char</replaceable> <replaceable>literal-text</replaceable> <replaceable>char</replaceable>
\verb* <replaceable>char</replaceable> <replaceable>literal-text</replaceable> <replaceable>char</replaceable>
</screen>
<para>Typeset <replaceable>literal-text</replaceable> as it is input, including special characters
and spaces, using the typewriter (<literal>\tt</literal>) font.
</para>
<para>This example shows two different invocations of <literal>\verb</literal>.
</para>
<screen>This is \verb!literally! the biggest pumpkin ever.
And this is the best squash, \verb+literally!+
</screen>
<para>The first <literal>\verb</literal> has its <replaceable>literal-text</replaceable> surrounded with
exclamation point, <literal>!</literal>.  The second instead uses plus, <literal>+</literal>,
because the exclamation point is part of <literal>literal-text</literal>.
</para>
<para>The single-character delimiter <replaceable>char</replaceable> surrounds
<replaceable>literal-text</replaceable>&#160;&#8212; it must be the same character before and
after.  No spaces come between <literal>\verb</literal> or <literal>\verb*</literal> and
<replaceable>char</replaceable>, or between <replaceable>char</replaceable> and <replaceable>literal-text</replaceable>, or between
<replaceable>literal-text</replaceable> and the second occurrence of <replaceable>char</replaceable> (the synopsis
shows a space only to distinguish one component from the other).  The
delimiter must not appear in <replaceable>literal-text</replaceable>. The <replaceable>literal-text</replaceable>
cannot include a line break.
</para>
<indexterm role="cp"><primary>visible space</primary></indexterm>
<para>The <literal>*</literal>-form differs only in that spaces are printed with a visible
space character.
</para>
<para>The output from this will include a character showing the spaces.
</para>
<screen>The commands's first argument is \verb*!filename with extension! and ...
</screen>
<indexterm role="cp"><primary>package, <literal>url</literal></primary></indexterm>
<indexterm role="cp"><primary><literal>url</literal> package</primary></indexterm>

<para>For typesetting Internet addresses, urls, the package <literal>url</literal>
provides an option that is better than the <literal>\verb</literal> command, since
it allows line breaks.
</para>
<indexterm role="cp"><primary>package, <literal>listings</literal></primary></indexterm>
<indexterm role="cp"><primary><literal>listings</literal> package</primary></indexterm>

<indexterm role="cp"><primary>package, <literal>minted</literal></primary></indexterm>
<indexterm role="cp"><primary><literal>minted</literal> package</primary></indexterm>

<para>For computer code there are many packages with advantages over
<literal>\verb</literal>.  One is <filename>listings</filename>, another is <filename>minted</filename>.
</para>
<indexterm role="cp"><primary>package, <literal>cprotect</literal></primary></indexterm>
<indexterm role="cp"><primary><literal>cprotect</literal> package</primary></indexterm>

<para>You cannot use <literal>\verb</literal> in the argument to a macro, for instance in
the argument to a <literal>\section</literal>.  It is not a question of <literal>\verb</literal>
being fragile (see <link linkend="_005cprotect">\protect</link>), instead it just cannot appear there.
(But the <literal>cprotect</literal> package can help with this.)
</para>

</sect2>
</sect1>
<sect1 label="8.28" id="verse">
<title><literal>verse</literal></title>

<indexterm role="fn"><primary>environment, <literal>verse</literal></primary></indexterm>
<indexterm role="fn"><primary><literal>verse</literal> environment</primary></indexterm>
        
<indexterm role="cp"><primary>poetry, an environment for</primary></indexterm>

<para>Synopsis:
</para>
<screen>\begin{verse}
  <replaceable>line1</replaceable> \\
  <replaceable>line2</replaceable> \\
  ...
\end{verse}
</screen>
<para>An environment for poetry.
</para>
<para>Here are two lines from Shakespeare&#8217;s Romeo and Juliet.
</para>
<screen>Then plainly know my heart's dear love is set \\
On the fair daughter of rich Capulet.
</screen>
<indexterm role="fn"><primary>\\ for <literal>verse</literal></primary></indexterm>
<para>Separate the lines of each stanza with <literal>\\</literal>, and use one or more
blank lines to separate the stanzas.
</para>
<screen>\begin{verse}
\makebox[\linewidth][c]{\textit{Shut Not Your Doors} ---Walt Whitman}
  \\[1\baselineskip]
Shut not your doors to me proud libraries,                  \\
For that which was lacking on all your well-fill'd shelves, \\
\qquad yet needed most, I bring,                             \\
Forth from the war emerging, a book I have made,            \\
The words of my book nothing, the drift of it every thing,  \\
A book separate, not link'd with the rest nor felt by the intellect, \\
But you ye untold latencies will thrill to every page.
\end{verse}
</screen>
<para>The output has margins indented on the left and the right, paragraphs
are not indented, and the text is not right-justified.
</para>

</sect1>
</chapter>
<chapter label="9" id="Line-breaking">
<title>Line breaking</title>

<indexterm role="cp"><primary>line breaking</primary></indexterm>
<indexterm role="cp"><primary>breaking lines</primary></indexterm>

<para>The first thing &latex; does when processing ordinary text is to
translate your input file into a sequence of glyphs and spaces.  To
produce a printed document, this sequence must be broken into lines
(and these lines must be broken into pages).
</para>
<para>&latex; usually does the line (and page) breaking in the text body for
you but in some environments you manually force line breaks.
</para>
<para>A common workflow is to get a final version of the document content
before taking a final pass through and considering line breaks (and page
breaks).  This differs from word processing, where you are formatting
text as you input it.  Putting these off until the end prevents a lot of
fiddling with breaks that will change anyway.
</para>


<sect1 label="9.1" id="_005c_005c">
<title><literal>\\</literal></title>

<indexterm role="fn"><primary>\\ force line break</primary></indexterm>
<indexterm role="cp"><primary>new line, starting</primary></indexterm>
<indexterm role="cp"><primary>line break, forcing</primary></indexterm>

<para>Synopsis, one of:
</para>
<screen>\\
\\[<replaceable>morespace</replaceable>]
</screen>
<para>or one of: 
</para>
<screen>\\*
\\*[<replaceable>morespace</replaceable>]
</screen>
<para>End the current line.  The optional argument <replaceable>morespace</replaceable> specifies
extra vertical space to be inserted before the next line.  This is a
rubber length (see <link linkend="Lengths">Lengths</link>) and can be negative.  The text before
the line break is set at its normal length, that is, it is not stretched
to fill out the line width.  This command is fragile (see <link linkend="_005cprotect">\protect</link>).
</para>
<para>The starred form, <literal>\\*</literal>, tells &latex; not to start a new page
between the two lines, by issuing a <literal>\nobreak</literal>.
</para>
<screen>\title{My story: \\[0.25in]
       a tale of woe}
</screen>
<para>Explicit line breaks in the main text body are unusual in &latex;.  In
particular, don&#8217;t start new paragraphs with <literal>\\</literal>.  Instead leave a
blank line between the two paragraphs.  And don&#8217;t put in a sequence of
<literal>\\</literal>&#8217;s to make vertical space. Instead use
<literal>\vspace{<replaceable>length</replaceable>}</literal>, or
<literal>\leavevmode\vspace{<replaceable>length</replaceable>}</literal>, or
<literal>\vspace*{<replaceable>length</replaceable>}</literal> if you want the space to not be thrown
out at the top of a new page (see <link linkend="_005cvspace">\vspace</link>).
</para>
<para>The <literal>\\</literal> command is mostly used outside of the main flow of text
such as in a <literal>tabular</literal> or <literal>array</literal> environment or in an
equation environment.
</para>
<para>The <literal>\\</literal> command is a synonym for <literal>\newline</literal>
(see <link linkend="_005cnewline">\newline</link>) under ordinary circumstances (an example of an
exception is the <literal>p{...}</literal> column in a <literal>tabular</literal>
environment; see <link linkend="tabular">tabular</link>).
</para>
<!-- credit: David Carlisle https://tex.stackexchange.com/a/82666 -->
<para>The <literal>\\</literal> command is a macro, and its definition changes by context
so that its definition in normal text, a <literal>center</literal> environment, a
<literal>flushleft</literal> environment, and a <literal>tabular</literal> are all different.
In normal text when it forces a linebreak it is essentially a shorthand
for <literal>\newline</literal>.  It does not end horizontal mode or end the
paragraph, it just inserts some glue and penalties so that when the
paragraph does end a linebreak will occur at that point, with the short
line padded with white space.
</para>
<para>You get &#8216;<literal>LaTeX Error: There's no line here to end</literal>&#8217; if you use
<literal>\\</literal> to ask for a new line, rather than to end the current line.
An example is if you have <literal>\begin{document}\\</literal> or, more likely,
something like this.
</para>
<screen>\begin{center}
  \begin{minipage}{0.5\textwidth}
  \\
  In that vertical space put your mark.
  \end{minipage}
\end{center}
</screen>
<para>Fix it by replacing the double backslash with something like
<literal>\vspace{\baselineskip}</literal>.
</para>

</sect1>
<sect1 label="9.2" id="_005cobeycr-_0026-_005crestorecr">
<title><literal>\obeycr</literal> &amp; <literal>\restorecr</literal></title>

<indexterm role="fn"><primary>\obeycr</primary></indexterm>
<indexterm role="fn"><primary>\restorecr</primary></indexterm>
<indexterm role="cp"><primary>new line, output as input</primary></indexterm>

<para>The <literal>\obeycr</literal> command makes a return in the input file (&#8216;<literal>^^M</literal>&#8217;,
internally) the same as <literal>\\</literal>, followed by <literal>\relax</literal>.  So each
new line in the input will also be a new line in the output.  The
<literal>\restorecr</literal> command restores normal line-breaking behavior.
</para>
<para>This is not the way to show verbatim text or computer code.
See <link linkend="verbatim">verbatim</link> instead.
</para>
<para>With &latex;&#8217;s usual defaults, this
</para>
<screen>aaa
bbb

\obeycr
ccc
ddd
   eee

\restorecr
fff
ggg

hhh
iii
</screen>
<para>produces output like this.
</para>
<screen>  aaa bbb
  ccc
ddd
eee

fff ggg
  hhh iii
</screen>
<para>The indents are paragraph indents.
</para>

</sect1>
<sect1 label="9.3" id="_005cnewline">
<title><literal>\newline</literal></title>

<indexterm role="fn"><primary>\newline</primary></indexterm>
<indexterm role="cp"><primary>new line, starting (paragraph mode)</primary></indexterm>

<para>In ordinary text, this ends a line in a way that does not right-justify
the line, so the prior text is not stretched. That is, in paragraph mode
(see <link linkend="Modes">Modes</link>), the <literal>\newline</literal> command is equivalent to
double-backslash (see <link linkend="_005c_005c">\\</link>).  This command is fragile
(see <link linkend="_005cprotect">\protect</link>).
</para>
<para>However, the two commands are different inside a <literal>tabular</literal> or
<literal>array</literal> environment. In a column with a specifier producing a
paragraph box such as typically <literal>p{...}</literal>, <literal>\newline</literal> will
insert a line end inside of the column; that is, it does not break the
entire tabular row.  To break the entire row use <literal>\\</literal> or its
equivalent <literal>\tabularnewline</literal>.
</para>
<para>This will print &#8216;<literal>Name:</literal>&#8217; and &#8216;<literal>Address:</literal>&#8217; as two lines in a
single cell of the table.
</para>
<screen>\begin{tabular}{p{1in}{\hspace{2in}}p{1in}}
  Name: \newline Address: &amp;Date: \\ \hline
\end{tabular}
</screen>
<para>The &#8216;<literal>Date:</literal>&#8217; will be baseline-aligned with &#8216;<literal>Name:</literal>&#8217;.
</para>

</sect1>
<sect1 label="9.4" id="_005c_002d-_0028hyphenation_0029">
<title><literal>\-</literal> (discretionary hyphen)</title>

<indexterm role="fn"><primary>\- (hyphenation)</primary></indexterm>
<indexterm role="cp"><primary>hyphenation, forcing</primary></indexterm>

<para>Tell &latex; that it may hyphenate the word at that point.  When you
insert <literal>\-</literal> commands in a word, the word will only be hyphenated at
those points and not at any of the hyphenation points that &latex;
might otherwise have chosen.  This command is robust (see <link linkend="_005cprotect">\protect</link>).
</para>
<para>&latex; is good at hyphenating and usually finds most of the correct
hyphenation points, while almost never using an incorrect one.  The
<literal>\-</literal> command is for exceptional cases.
</para>
<para>For example, &latex; does not ordinarily hyphenate words containing a
hyphen.  Below, the long and hyphenated word means &latex; has to put
in unacceptably large spaces to set the narrow column.
</para>
<screen>\begin{tabular}{rp{1.75in}}
  Isaac Asimov &amp;The strain of
               anti-intellectualism
               % an\-ti-in\-tel\-lec\-tu\-al\-ism
               has been a constant thread winding its way through our
               political and cultural life, nurtured by
               the false notion that democracy means that
               `my ignorance is just as good as your knowledge'.
\end{tabular}
</screen>
<para>Commenting out the third line and uncommenting the fourth makes a much
better fit.
</para>
<para>The <literal>\-</literal> command only allows &latex; to break there, it does not
require that it break there.  You can insist on a split with something
like <literal>Hef-\linebreak feron</literal>.  Of course, if you later change the
text then this forced break may look very odd, so this approach requires
care.
</para>

</sect1>
<sect1 label="9.5" id="_005cdiscretionary">
<title><literal>\discretionary</literal> (generalized hyphenation point)</title>
<indexterm role="cp"><primary>hyphenation, discretionary</primary></indexterm>
<indexterm role="cp"><primary>discretionary hyphenation</primary></indexterm>

<para>Synopsis:
</para>
<screen>\discretionary{<replaceable>pre-break</replaceable>}{<replaceable>post-break</replaceable>}{<replaceable>no-break</replaceable>}
</screen>
<para>Handle word changes around hyphens.  This command is not often used in
&latex; documents.
</para>
<para>If a line break occurs at the point where <literal>\discretionary</literal> appears
then &tex; puts <replaceable>pre-break</replaceable> at the end of the current line and puts
<replaceable>post-break</replaceable> at the start of the next line.  If there is no line
break here then &tex; puts <replaceable>no-break</replaceable>
</para>
<para>In &#8216;<literal>difficult</literal>&#8217; the three letters <literal>ffi</literal> form a ligature.  But
&tex; can nonetheless break between the two f&#8217;s with this.
</para>
<screen>di\discretionary{f-}{fi}{ffi}cult
</screen>
<para>Note that users do not have to do this.  It is typically handled
automatically by &tex;&#8217;s hyphenation algorithm.
</para>

</sect1>
<sect1 label="9.6" id="_005cfussy-_0026-_005csloppy">
<title><literal>\fussy</literal> &amp; <literal>\sloppy</literal></title>

<indexterm role="fn"><primary>\fussy</primary></indexterm>
<indexterm role="fn"><primary>\sloppy</primary></indexterm>
<indexterm role="cp"><primary>line breaks, changing</primary></indexterm>

<para>Declarations to make &tex; more picky or less picky about line
breaking.  Declaring <literal>\fussy</literal> usually avoids too much space between
words, at the cost of an occasional overfull box.  Conversely,
<literal>\sloppy</literal> avoids overfull boxes while suffering loose interword
spacing.
</para>
<para>The default is <literal>\fussy</literal>.  Line breaking in a paragraph is
controlled by whichever declaration is current at the blank line, or
<literal>\par</literal>, or displayed equation ending that paragraph.  So to affect
the line breaks, include that paragraph-ending material in the scope of
the command.
</para>


<sect2 label="9.6.1" id="sloppypar">
<title><literal>sloppypar</literal></title>

<indexterm role="fn"><primary>sloppypar</primary></indexterm>
<indexterm role="cp"><primary>sloppypar environment</primary></indexterm>

<para>Synopsis:
</para>
<screen>\begin{sloppypar}
  ... paragraphs ...
\end{sloppypar}
</screen>
<para>Typeset the paragraphs with <literal>\sloppy</literal> in effect (see <link linkend="_005cfussy-_0026-_005csloppy">\fussy &amp;
\sloppy</link>).  Use this to locally adjust line breaking, to avoid
&#8216;<literal>Overfull box</literal>&#8217; or &#8216;<literal>Underfull box</literal>&#8217; errors.
</para>
<para>The example is simple.
</para>
<screen>\begin{sloppypar}
  Her plan for the morning thus settled, she sat quietly down to her
  book after breakfast, resolving to remain in the same place and the
  same employment till the clock struck one; and from habitude very
  little incommoded by the remarks and ejaculations of Mrs.\ Allen,
  whose vacancy of mind and incapacity for thinking were such, that
  as she never talked a great deal, so she could never be entirely
  silent; and, therefore, while she sat at her work, if she lost her
  needle or broke her thread, if she heard a carriage in the street,
  or saw a speck upon her gown, she must observe it aloud, whether
  there were anyone at leisure to answer her or not.
\end{sloppypar}
</screen>

</sect2>
</sect1>
<sect1 label="9.7" id="_005chyphenation">
<title><literal>\hyphenation</literal></title>

<indexterm role="fn"><primary>\hyphenation</primary></indexterm>
<indexterm role="cp"><primary>hyphenation, defining</primary></indexterm>

<para>Synopsis:
</para>
<screen>\hyphenation{<replaceable>word1</replaceable> ...}
</screen>
<para>Declares allowed hyphenation points within the words in the list.  The
words in that list are separated by spaces.  Show permitted points for
hyphenation with a dash character, <literal>-</literal>.
</para>
<para>Here is an example:
</para>
<screen>\hyphenation{hat-er il-lit-e-ra-ti tru-th-i-ness}
</screen>
<para>Use lowercase letters.  &tex; will only hyphenate if the word matches
exactly.  Multiple <literal>\hyphenation</literal> commands accumulate.
</para>

</sect1>
<sect1 label="9.8" id="_005clinebreak-_0026-_005cnolinebreak">
<title><literal>\linebreak</literal> &amp; <literal>\nolinebreak</literal></title>

<indexterm role="fn"><primary>\linebreak</primary></indexterm>
<indexterm role="fn"><primary>\nolinebreak</primary></indexterm>
<indexterm role="cp"><primary>line breaks, forcing</primary></indexterm>
<indexterm role="cp"><primary>line breaks, preventing</primary></indexterm>

<para>Synopses, one of:
</para>
<screen>\linebreak
\linebreak[<replaceable>zero-to-four</replaceable>]
</screen>
<para>or one of these.
</para>
<screen>\nolinebreak
\nolinebreak[<replaceable>zero-to-four</replaceable>]
</screen>
<para>Encourage or discourage a line break.  The optional <replaceable>zero-to-four</replaceable>
is an integer that allows you to soften the instruction.  The default is
4, so that without the optional argument these commands entirely force
or prevent the break.  But for instance, <literal>\nolinebreak[1]</literal> is a
suggestion that another place may be better.  The higher the number, the
more insistent the request.  Both commands are fragile
(see <link linkend="_005cprotect">\protect</link>).
</para>
<para>Here we tell &latex; that a good place to put a linebreak is after the
standard legal text.
</para>
<screen>\boilerplatelegal{} \linebreak[2]
We especially encourage applications from members of traditionally
underrepresented groups.
</screen>
<para>When you issue <literal>\linebreak</literal>, the spaces in the line are stretched
out so that it extends to the right margin.  See <link linkend="_005c_005c">\\</link>
and&#160;<link linkend="_005cnewline">\newline</link> to have the spaces not stretched out.
</para>


</sect1>
</chapter>
<chapter label="10" id="Page-breaking">
<title>Page breaking</title>

<indexterm role="cp"><primary>page breaking</primary></indexterm>
<indexterm role="cp"><primary>breaking pages</primary></indexterm>

<para>Ordinarily &latex; automatically takes care of breaking output into
pages with its usual aplomb.  But if you are writing commands, or
tweaking the final version of a document, then you may need to
understand how to influence its actions.
</para>
<!-- credit: H Vogt https://tex.stackexchange.com/a/115563 -->
<para>&latex;&#8217;s algorithm for splitting a document into pages is more complex
than just waiting until there is enough material to fill a page and
outputting the result.  Instead, &latex; typesets more material than
would fit on the page and then chooses a break that is optimal in some
way (it has the smallest badness). An example of the advantage of this
approach is that if the page has some vertical space that can be
stretched or shrunk, such as with rubber lengths between paragraphs,
then &latex; can use that to avoid widow lines (where a new page starts
with the last line of a paragraph; &latex; can squeeze the extra line
onto the first page) and orphans (where the first line of paragraph is
at the end of a page; &latex; can stretch the material of the first
page so the extra line falls on the second page). Another example is
where &latex; uses available vertical shrinkage to fit on a page not
just the header for a new section but also the first two lines of that
section.
</para>
<para>But &latex; does not optimize over the entire document&#8217;s set of page
breaks. So it can happen that the first page break is great but the
second one is lousy; to break the current page &latex; doesn&#8217;t look as
far ahead as the next page break.  So occasionally you may want to
influence page breaks while preparing a final version of a document.
</para>
<para>See <link linkend="Layout">Layout</link> for more material that is relevant to page breaking.
</para>


<sect1 label="10.1" id="_005cclearpage-_0026-_005ccleardoublepage">
<title><literal>\clearpage</literal> &amp; <literal>\cleardoublepage</literal></title>

<indexterm role="fn"><primary>\clearpage</primary></indexterm>
<indexterm role="cp"><primary>flushing floats and starting a page</primary></indexterm>
<indexterm role="cp"><primary>starting a new page and clearing floats</primary></indexterm>
<indexterm role="fn"><primary>\cleardoublepage</primary></indexterm>
<indexterm role="cp"><primary>starting on a right-hand page</primary></indexterm>

<para>Synopsis:
</para>
<screen>\clearpage
</screen>
<para>or
</para>
<screen>\cleardoublepage
</screen>
<para>End the current page and output all of the pending floating figures and
tables (see <link linkend="Floats">Floats</link>).  If there are too many floats to fit on the
page then &latex; will put in extra pages containing only floats.  In
two-sided printing, <literal>\cleardoublepage</literal> also makes the next page of
content a right-hand page, an odd-numbered page, if necessary inserting
a blank page.  The <literal>\clearpage</literal> command is robust while
<literal>\cleardoublepage</literal> is fragile (see <link linkend="_005cprotect">\protect</link>).
</para>
<para>&latex;&#8217;s page breaks are optimized so ordinarily you only use this
command in a document body to polish the final version, or inside
commands.
</para>
<!-- credit: https://www.tex.ac.uk/FAQ-reallyblank.html -->
<para>The <literal>\cleardoublepage</literal> command will put in a blank page, but it
will have the running headers and footers.  To get a really blank
page, use this command.
</para>
<screen>\let\origdoublepage\cleardoublepage
\newcommand{\clearemptydoublepage}{%
  \clearpage
  {\pagestyle{empty}\origdoublepage}%
}
</screen>
<para>If you want &latex;&#8217;s standard <literal>\chapter</literal> command to do this then
add the line <literal>\let\cleardoublepage\clearemptydoublepage</literal>.
</para>
<para>The command <literal>\newpage</literal> (see <link linkend="_005cnewpage">\newpage</link>) also ends the current
page, but without clearing pending floats.  And, if &latex; is in
two-column mode then <literal>\newpage</literal> ends the current column while
<literal>\clearpage</literal> and <literal>\cleardoublepage</literal> end the current page.
</para>


</sect1>
<sect1 label="10.2" id="_005cnewpage">
<title><literal>\newpage</literal></title>

<indexterm role="fn"><primary>\newpage</primary></indexterm>
<indexterm role="cp"><primary>new page, starting</primary></indexterm>
<indexterm role="cp"><primary>starting a new page</primary></indexterm>

<para>Synopsis:
</para>
<screen>\newpage
</screen>
<para>End the current page.  This command is robust (see <link linkend="_005cprotect">\protect</link>).
</para>
<para>&latex;&#8217;s page breaks are optimized so ordinarily you only use this
command in a document body to polish the final version, or inside
commands.
</para>
<para>While the commands <literal>\clearpage</literal> and <literal>\cleardoublepage</literal> also
end the current page, in addition they clear pending floats
(see <link linkend="_005cclearpage-_0026-_005ccleardoublepage">\clearpage &amp; \cleardoublepage</link>).  And, if &latex; is in
two-column mode then <literal>\clearpage</literal> and <literal>\cleardoublepage</literal> end
the current page, possibly leaving an empty column, while
<literal>\newpage</literal> only ends the current column.
</para>
<para>In contrast with <literal>\pagebreak</literal> (see <link linkend="_005cpagebreak-_0026-_005cnopagebreak">\pagebreak &amp; \nopagebreak</link>),
the <literal>\newpage</literal> command will cause the new page to start right where
requested.  This
</para>
<screen>Four score and seven years ago our fathers brought forth on this
continent,
\newpage
\noindent a new nation, conceived in Liberty, and dedicated to the
proposition that all men are created equal.
</screen>
<para>makes a new page start after &#8216;<literal>continent,</literal>&#8217; and the cut-off line is
not right justified.  In addition, <literal>\newpage</literal> does not vertically
stretch out the page, as <literal>\pagebreak</literal> does.
</para>

</sect1>
<sect1 label="10.3" id="_005cenlargethispage">
<title><literal>\enlargethispage</literal></title>

<indexterm role="fn"><primary>\enlargethispage</primary></indexterm>
<indexterm role="cp"><primary>enlarge current page</primary></indexterm>

<para>Synopsis, one of:
</para>
<screen>\enlargethispage{size}
\enlargethispage*{size}
</screen>
<para>Enlarge the <literal>\textheight</literal> for the current page.  The required
argument <replaceable>size</replaceable> must be a rigid length (see <link linkend="Lengths">Lengths</link>).  It may be
positive or negative.  This command is fragile (see <link linkend="_005cprotect">\protect</link>).
</para>
<para>A common strategy is to wait until you have the final text of a
document, and then pass through it tweaking line and page breaks.  This
command allows you some page size leeway.
</para>
<para>This will allow one extra line on the current page.
</para>
<screen>\enlargethispage{\baselineskip}
</screen>
<para>The starred form <literal>\enlargesthispage*</literal> tries to squeeze the material
together on the page as much as possible, for the common use case of
getting one more line on the page. This is often used together with an
explicit <literal>\pagebreak</literal>.
</para>

</sect1>
<sect1 label="10.4" id="_005cpagebreak-_0026-_005cnopagebreak">
<title><literal>\pagebreak</literal> &amp; <literal>\nopagebreak</literal></title>

<indexterm role="fn"><primary>\pagebreak</primary></indexterm>
<indexterm role="fn"><primary>\nopagebreak</primary></indexterm>
<indexterm role="cp"><primary>page break, forcing</primary></indexterm>
<indexterm role="cp"><primary>page break, preventing</primary></indexterm>

<para>Synopses:
</para>
<screen>\pagebreak
\pagebreak[<replaceable>zero-to-four</replaceable>]
</screen>
<para>or
</para>
<screen>\nopagebreak
\nopagebreak[<replaceable>zero-to-four</replaceable>]
</screen>
<para>Encourage or discourage a page break.  The optional <replaceable>zero-to-four</replaceable>
is an integer that allows you to soften the request. The default is 4,
so that without the optional argument these commands entirely force or
prevent the break.  But for instance <literal>\nopagebreak[1]</literal> suggests to
&latex; that another spot might be preferable.  The higher the number,
the more insistent the request.  Both commands are fragile
(see <link linkend="_005cprotect">\protect</link>).
</para>
<para>&latex;&#8217;s page endings are optimized so ordinarily you only use this
command in a document body to polish the final version, or inside
commands.
</para>
<para>If you use these inside a paragraph, they apply to the point following
the line in which they appear.  So this
</para>
<screen>Four score and seven years ago our fathers brought forth on this
continent,
\pagebreak
a new nation, conceived in Liberty, and dedicated to the proposition
that all men are created equal.
</screen>
<para>does not give a page break at &#8216;<literal>continent,</literal>&#8217; but instead at
&#8216;<literal>nation,</literal>&#8217; since that is where &latex; breaks that line.  In
addition, with <literal>\pagebreak</literal> the vertical space on the page is
stretched out where possible so that it extends to the normal bottom
margin.  This can look strange, and if <literal>\flushbottom</literal> is in effect
this can cause you to get &#8216;<literal>Underfull \vbox (badness 10000) has
occurred while \output is active</literal>&#8217;.  See <link linkend="_005cnewpage">\newpage</link> for a command that
does not have these effects.
</para>

</sect1>
</chapter>
<chapter label="11" id="Footnotes">
<title>Footnotes</title>

<indexterm role="cp"><primary>footnotes, creating</primary></indexterm>

<para>Place a footnote at the bottom of the current page, as here.
</para>
<screen>No&#235;l Coward quipped that having to read a footnote is like having
to go downstairs to answer the door, while in the midst of making
love.\footnote{%
  I wouldn't know, I don't read footnotes.}
</screen>
<para>You can put multiple footnotes on a page. If the footnote text becomes
too long then it will flow to the next page.
</para>
<para>You can also produce footnotes by combining the <literal>\footnotemark</literal> and
the <literal>\footnotetext</literal> commands, which is useful in special
circumstances.
</para>
<para>To make bibliographic references come out as footnotes you need to
include a bibliographic style with that behavior (see <link linkend="Using-BibTeX">Using BibTeX</link>).
</para>


<sect1 label="11.1" id="_005cfootnote">
<title><literal>\footnote</literal></title>

<indexterm role="fn"><primary>\footnote</primary></indexterm>

<para>Synopsis, one of:
</para>
<screen>\footnote{<replaceable>text</replaceable>}
\footnote[<replaceable>number</replaceable>]{<replaceable>text</replaceable>}
</screen>
<para>Place a footnote <replaceable>text</replaceable> at the bottom of the current page.
</para>
<screen>There are over a thousand footnotes in Gibbon's 
\textit{Decline and Fall of the Roman Empire}.\footnote{%
  After reading an early version with endnotes David Hume complained, 
  ``One is also plagued with his Notes, according to the present Method 
  of printing the Book'' and suggested that they ``only to be printed 
  at the Margin or the Bottom of the Page.''}
</screen>
<para>The optional argument <replaceable>number</replaceable> allows you to specify the number of
the footnote.  If you use this then &latex; does not increment the
<literal>footnote</literal> counter.
</para>
<indexterm role="cp"><primary>footnotes, symbols instead of numbers</primary></indexterm>
<indexterm role="fn"><primary>\fnsymbol, and footnotes</primary></indexterm>
<indexterm role="fn"><primary>\@fnsymbol</primary></indexterm>
<para>By default, &latex; uses arabic numbers as footnote markers.  Change
this with something like
<literal>\renewcommand{\thefootnote}{\fnsymbol{footnote}}</literal>, which
uses a sequence of symbols (see <link linkend="_005calph-_005cAlph-_005carabic-_005croman-_005cRoman-_005cfnsymbol">\alph \Alph \arabic \roman \Roman
\fnsymbol</link>).  To make this change global put that in the preamble.  If
you make the change local then you may want to reset the counter with
<literal>\setcounter{footnote}{0}</literal>.
</para>
<para>&latex; determines the spacing of footnotes with two parameters.
</para>
<indexterm role="cp"><primary>footnote parameters</primary></indexterm>
<indexterm role="cp"><primary>parameters, for footnotes</primary></indexterm>

<variablelist><varlistentry><term><indexterm role="fn"><primary>\footnoterule</primary></indexterm><literal>\footnoterule</literal>
</term><listitem><anchor id="footnote-footnoterule"/><para>Produces the rule separating the main text on a page from the page&#8217;s
footnotes.  Default dimensions in the standard document classes (except
<literal>slides</literal>, where it does not appear) is: vertical thickness of
<literal>0.4pt</literal>, and horizontal size of <literal>0.4\columnwidth</literal> long.
Change the rule with something like this.
</para>
<!-- Credit egreg: https://tex.stackexchange.com/a/21917 -->
<screen>\renewcommand{\footnoterule}{% Kerns avoid vertical space
  \kern -3pt                         % This -3 is negative
  \hrule width \textwidth height 1pt % of the sum of this 1
  \kern 2pt}                         % and this 2
</screen>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\footnotesep</primary></indexterm><literal>\footnotesep</literal>
</term><listitem><anchor id="footnote-footnotesep"/><para>The height of the strut placed at the beginning of the footnote
(see <link linkend="_005cstrut">\strut</link>).  By default, this is set to the normal strut for
<literal>\footnotesize</literal> fonts (see <link linkend="Font-sizes">Font sizes</link>), therefore there is no
extra space between footnotes.  This is &#8216;<literal>6.65pt</literal>&#8217; for &#8216;<literal>10pt</literal>&#8217;,
&#8216;<literal>7.7pt</literal>&#8217; for &#8216;<literal>11pt</literal>&#8217;, and &#8216;<literal>8.4pt</literal>&#8217; for &#8216;<literal>12pt</literal>&#8217;.  Change
it as with <literal>\setlength{\footnotesep}{11pt}</literal>.
</para>
</listitem></varlistentry></variablelist>
<para>The <literal>\footnote</literal> command is fragile (see <link linkend="_005cprotect">\protect</link>).
</para>
<para>&latex;&#8217;s default puts many restrictions on where you can use a
<literal>\footnote</literal>; for instance, you cannot use it in an argument to a
sectioning command such as <literal>\chapter</literal> (it can only be used in outer
paragraph mode; see <link linkend="Modes">Modes</link>).  There are some workarounds; see
following sections.
</para>
<indexterm role="cp"><primary>Footnotes, in a minipage</primary></indexterm>
<indexterm role="cp"><primary>mpfootnote counter</primary></indexterm>
<para>In a <literal>minipage</literal> environment the <literal>\footnote</literal> command uses the
<literal>mpfootnote</literal> counter instead of the <literal>footnote</literal> counter, so
they are numbered independently.  They are shown at the bottom of the
environment, not at the bottom of the page.  And by default they are
shown alphabetically. See <link linkend="minipage">minipage</link> and <link linkend="Footnotes-in-a-table">Footnotes in a table</link>.
</para>

</sect1>
<sect1 label="11.2" id="_005cfootnotemark">
<title><literal>\footnotemark</literal></title>

<indexterm role="fn"><primary>\footnotemark</primary></indexterm>

<para>Synopsis, one of:
</para>
<screen>\footnotemark
\footnotemark[<replaceable>number</replaceable>]
</screen>
<para>Put the current footnote mark in the text. To specify associated text
for the footnote see&#160;<link linkend="_005cfootnotetext">\footnotetext</link>.  The optional argument
<replaceable>number</replaceable> causes the command to use that number to determine the
footnote mark. This command can be used in inner paragraph mode
(see <link linkend="Modes">Modes</link>).
</para>
<para>If you use <literal>\footnotemark</literal> without the optional argument then it
increments the footnote counter but if you use the optional <replaceable>number</replaceable>
then it does not. The next example produces several consecutive footnote
markers referring to the same footnote.
</para>
<screen>The first theorem\footnote{Due to Gauss.} 
and the second theorem\footnotemark[\value{footnote}] 
and the third theorem.\footnotemark[\value{footnote}]
</screen>
<para>If there are intervening footnotes then you must remember the value of the
common mark.  This example gives the same institutional affiliation to
both the first and third authors (<literal>\thanks</literal> is a version of
<literal>footnote</literal>), by-hand giving the number of the footnote.
</para>
<screen>\title{A Treatise on the Binomial Theorem}
\author{J Moriarty\thanks{University of Leeds} 
  \and A C Doyle\thanks{Durham University} 
  \and S Holmes\footnotemark[1]}
\begin{document}
\maketitle
</screen>
<para>This uses a counter to remember the footnote number.  The third sentence
is followed by the same footnote marker as the first.
</para>
<screen>\newcounter{footnoteValueSaver}
All babies are illogical.\footnote{%
  Lewis Carroll.}\setcounter{footnoteValueSaver}{\value{footnote}}
Nobody is despised who can manage a crocodile.\footnote{%
  Captain Hook.}
Illogical persons are despised.\footnotemark[\value{footnoteValueSaver}]
Therefore, anyone who can manage a crocodile is not a baby.
</screen>
<indexterm role="cp"><primary>package, <literal>cleveref</literal></primary></indexterm>
<indexterm role="cp"><primary><literal>cleveref</literal> package</primary></indexterm>

<indexterm role="cp"><primary>package, <literal>hyperref</literal></primary></indexterm>
<indexterm role="cp"><primary><literal>hyperref</literal> package</primary></indexterm>

<para>This example accomplishes the same by using the package <filename>cleveref</filename>.
</para>
<!-- from SE user Jake http://tex.stackexchange.com/a/10116/339 -->
<screen>\usepackage{cleveref}[2012/02/15]   % in preamble 
\crefformat{footnote}{#2\footnotemark[#1]#3}
...
The theorem is from Evers.\footnote{\label{fn:TE}Tinker, Evers, 1994.}
The corollary is from Chance.\footnote{Evers, Chance, 1990.}
But the key lemma is from Tinker.\cref{fn:TE}
</screen>
<indexterm role="cp"><primary>package, <literal>hyperref</literal></primary></indexterm>
<indexterm role="cp"><primary><literal>hyperref</literal> package</primary></indexterm>

<para>It will work with the package <filename>hyperref</filename>.
</para>

</sect1>
<sect1 label="11.3" id="_005cfootnotetext">
<title><literal>\footnotetext</literal></title>

<indexterm role="fn"><primary>\footnotetext</primary></indexterm>

<para>Synopsis, one of:
</para>
<screen>\footnotetext{<replaceable>text</replaceable>}
\footnotetext[<replaceable>number</replaceable>]{<replaceable>text</replaceable>}
</screen>
<para>Place <replaceable>text</replaceable> at the bottom of the page as a footnote.  It pairs with
<literal>\footnotemark</literal> (see <link linkend="_005cfootnotemark">\footnotemark</link>) and can come anywhere after
that command, but must appear in outer paragraph mode (see <link linkend="Modes">Modes</link>).
The optional argument <replaceable>number</replaceable> changes the number of the footnote
mark.
</para>
<para>See <link linkend="_005cfootnotemark">\footnotemark</link> and&#160;<link linkend="Footnotes-in-a-table">Footnotes in a table</link> for usage
examples.
</para>

</sect1>
<sect1 label="11.4" id="Footnotes-in-section-headings">
<title>Footnotes in section headings</title>

<indexterm role="cp"><primary>footnote, in section headings</primary></indexterm>
<indexterm role="cp"><primary>table of contents, avoiding footnotes</primary></indexterm>

<para>Putting a footnote in a section heading, as in:
</para>
<screen>\section{Full sets\protect\footnote{This material due to ...}}
</screen>
<para>causes the footnote to appear at the bottom of the page where the
section starts, as usual, but also at the bottom of the table of
contents, where it is not likely to be desired.  The simplest way
to have it not appear on the table of contents is to use the optional
argument to <literal>\section</literal>
</para>
<screen>\section[Please]{Please\footnote{%
  Don't footnote in chapter and section headers!}}
</screen>
<para>No <literal>\protect</literal> is needed in front of <literal>\footnote</literal> here because
what gets moved to the table of contents is the optional argument.
</para>

</sect1>
<sect1 label="11.5" id="Footnotes-in-a-table">
<title>Footnotes in a table</title>

<indexterm role="cp"><primary>footnote, in a table</primary></indexterm>

<para>Inside a <literal>tabular</literal> or <literal>array</literal> environment the <literal>\footnote</literal>
command does not work; there is a footnote mark in the table cell but
the footnote text does not appear.  The solution is to use a
<literal>minipage</literal> environment as here (see <link linkend="minipage">minipage</link>).
</para>
<screen>\begin{center}
  \begin{minipage}{\textwidth} \centering
     \begin{tabular}{l|l}
       \textsc{Ship}           &amp;\textsc{Book} \\ \hline 
       \textit{HMS Sophie}     &amp;Master and Commander  \\ 
       \textit{HMS Polychrest} &amp;Post Captain  \\  
       \textit{HMS Lively}     &amp;Post Captain \\
       \textit{HMS Surprise}   &amp;A number of books\footnote{%
                                  Starting with HMS Surprise.}
     \end{tabular}
  \end{minipage}                              
\end{center}
</screen>
<para>Inside a <literal>minipage</literal>, footnote marks are lowercase letters.  Change
that with something like
<literal>\renewcommand{\thempfootnote}{\arabic{mpfootnote}}</literal>
(see <link linkend="_005calph-_005cAlph-_005carabic-_005croman-_005cRoman-_005cfnsymbol">\alph \Alph \arabic \roman \Roman \fnsymbol</link>).
</para>
<para>The footnotes in the prior example appear at the bottom of the
<literal>minipage</literal>.  To have them appear at the bottom of the main page, as
part of the regular footnote sequence, use the <literal>\footnotemark</literal> and
<literal>\footnotetext</literal> pair and make a new counter.
</para>
<screen>\newcounter{mpFootnoteValueSaver}
\begin{center}
  \begin{minipage}{\textwidth}
    \setcounter{mpFootnoteValueSaver}{\value{footnote}} \centering
     \begin{tabular}{l|l}
       \textsc{Woman}             &amp;\textsc{Relationship} \\ \hline 
       Mona                       &amp;Attached\footnotemark  \\ 
       Diana Villiers             &amp;Eventual wife  \\  
       Christine Hatherleigh Wood &amp;Fiance\footnotemark 
     \end{tabular}
  \end{minipage}%  percent sign keeps footnote text close to minipage
  \stepcounter{mpFootnoteValueSaver}%
    \footnotetext[\value{mpFootnoteValueSaver}]{%
      Little is known other than her death.}%
  \stepcounter{mpFootnoteValueSaver}%
    \footnotetext[\value{mpFootnoteValueSaver}]{%
      Relationship is unresolved in XXI.}
\end{center}
</screen>
<para>For a floating <literal>table</literal> environment (see <link linkend="table">table</link>), use the
<filename>tablefootnote</filename> package.
</para>
<screen>\usepackage{tablefootnote}  % in preamble
   ...
\begin{table}
  \centering
     \begin{tabular}{l|l}
     \textsc{Date}  &amp;\textsc{Campaign} \\ \hline 
     1862           &amp;Fort Donelson \\ 
     1863           &amp;Vicksburg     \\  
     1865           &amp;Army of Northern Virginia\tablefootnote{%
                      Ending the war.}
     \end{tabular}
    \caption{Forces captured by US Grant}
\end{table}
</screen>
<para>The footnote appears at the page bottom and is numbered in sequence with
other footnotes.
</para>

</sect1>
<sect1 label="11.6" id="Footnotes-of-footnotes">
<title>Footnotes of footnotes</title>

<indexterm role="cp"><primary>footnote, of a footnote</primary></indexterm>

<indexterm role="cp"><primary>package, <literal>bigfoot</literal></primary></indexterm>
<indexterm role="cp"><primary><literal>bigfoot</literal> package</primary></indexterm>

<para>Particularly in the humanities, authors can have multiple classes of
footnotes, including having footnotes of footnotes.  The package
<filename>bigfoot</filename> extends &latex;&#8217;s default footnote mechanism in many
ways, including allow these two, as in this example.
</para>
<screen>\usepackage{bigfoot}    % in preamble
\DeclareNewFootnote{Default}
\DeclareNewFootnote{from}[alph]   % create class \footnotefrom{}
 ...
The third theorem is a partial converse of the 
second.\footnotefrom{%
  First noted in Wilson.\footnote{Second edition only.}}
</screen>

</sect1>
</chapter>
<chapter label="12" id="Definitions">
<title>Definitions</title>

<indexterm role="cp"><primary>definitions</primary></indexterm>

<para>&latex; has support for making new commands of many different kinds.
</para>


<sect1 label="12.1" id="_005cnewcommand-_0026-_005crenewcommand">
<title><literal>\newcommand</literal> &amp; <literal>\renewcommand</literal></title>

<indexterm role="fn"><primary>\newcommand</primary></indexterm>
<indexterm role="cp"><primary>commands, defining new ones</primary></indexterm>
<indexterm role="cp"><primary>commands, redefining</primary></indexterm>
<indexterm role="cp"><primary>defining a new command</primary></indexterm>
<indexterm role="cp"><primary>new commands, defining</primary></indexterm>

<para>Synopses, one of:
</para>
<screen>\newcommand{\<replaceable>cmd</replaceable>}{<replaceable>defn</replaceable>}
\newcommand{\<replaceable>cmd</replaceable>}[<replaceable>nargs</replaceable>]{<replaceable>defn</replaceable>}
\newcommand{\<replaceable>cmd</replaceable>}[<replaceable>nargs</replaceable>][<replaceable>optargdefault</replaceable>]{<replaceable>defn</replaceable>}
\newcommand*{\<replaceable>cmd</replaceable>}{<replaceable>defn</replaceable>}
\newcommand*{\<replaceable>cmd</replaceable>}[<replaceable>nargs</replaceable>]{<replaceable>defn</replaceable>}
\newcommand*{\<replaceable>cmd</replaceable>}[<replaceable>nargs</replaceable>][<replaceable>optargdefault</replaceable>]{<replaceable>defn</replaceable>}
</screen>
<para>or one of these.
</para>
<screen>\renewcommand{\<replaceable>cmd</replaceable>}[<replaceable>nargs</replaceable>]{<replaceable>defn</replaceable>}
\renewcommand{\<replaceable>cmd</replaceable>}[<replaceable>nargs</replaceable>]{<replaceable>defn</replaceable>}
\renewcommand{\<replaceable>cmd</replaceable>}[<replaceable>nargs</replaceable>][<replaceable>optargdefault</replaceable>]{<replaceable>defn</replaceable>}
\renewcommand*{\<replaceable>cmd</replaceable>}{<replaceable>defn</replaceable>}
\renewcommand*{\<replaceable>cmd</replaceable>}[<replaceable>nargs</replaceable>]{<replaceable>defn</replaceable>}
\renewcommand*{\<replaceable>cmd</replaceable>}[<replaceable>nargs</replaceable>][<replaceable>optargdefault</replaceable>]{<replaceable>defn</replaceable>}
</screen>
<para>Define or redefine a command.  See also the discussion of
<literal>\DeclareRobustCommand</literal> in <link linkend="Class-and-package-commands">Class and package commands</link>.
<indexterm role="cp"><primary>starred form, defining new commands</primary></indexterm>
<indexterm role="cp"><primary>*-form, defining new commands</primary></indexterm>
The starred form of these two requires that the arguments not contain
multiple paragraphs of text (in plain &tex; terms that it not be
<literal>\long</literal>).
</para>
<para>These are the parameters:
</para>
<variablelist><varlistentry><term><replaceable>cmd</replaceable>
</term><listitem>
<para>Required; the command name.  It must begin with a backslash, <literal>\</literal>,
and must not begin with the four letter string <literal>\end</literal>.  For
<literal>\newcommand</literal>, it must not be already defined.  For
<literal>\renewcommand</literal>, this name must already be defined.
</para>
</listitem></varlistentry><varlistentry><term><replaceable>nargs</replaceable>
</term><listitem><para>Optional; an integer from 0 to 9, specifying the number of arguments
that the command takes, including any optional argument.  Omitting this
argument is the same as specifying 0, meaning that the command has no
arguments.  If you redefine a command, the new version can have a
different number of arguments than the old version.
</para>
</listitem></varlistentry><varlistentry><term><replaceable>optargdefault</replaceable>
</term><listitem><para>Optional; if this argument is present then the first argument of
<literal>\<replaceable>cmd</replaceable></literal> is optional, with default value <replaceable>optargdefault</replaceable>
(which may be the empty string).  If this argument is not present then
<literal>\<replaceable>cmd</replaceable></literal> does not take an optional argument.
</para>
<indexterm role="cp"><primary>positional parameter</primary></indexterm>
<para>That is, if <literal>\<replaceable>cmd</replaceable></literal> is used with square brackets, as in
<literal>\<replaceable>cmd</replaceable>[<replaceable>optval</replaceable>]{...}...</literal>, then within <replaceable>defn</replaceable> the
parameter&#160;<literal>#1</literal> is set to the value of <replaceable>optval</replaceable>.  On the
other hand, if <literal>\<replaceable>cmd</replaceable></literal> is called without the square brackets
then within <replaceable>defn</replaceable> the parameter <literal>#1</literal> is set to the value of
<replaceable>optargdefault</replaceable>.  In either case, the required arguments start with
<literal>#2</literal>.
</para>
<para>Omitting <literal>[<replaceable>optargdefault</replaceable>]</literal> is different from having the
square brackets with no contents, as in <literal>[]</literal>.  The former sets
<literal>#1</literal> to the value of <replaceable>optargdefault</replaceable>; the latter sets <literal>#1</literal>
to the empty string.
</para>
</listitem></varlistentry><varlistentry><term><replaceable>defn</replaceable>
</term><listitem><para>Required; the text to be substituted for every occurrence of
<literal>\<replaceable>cmd</replaceable></literal>. The parameters <literal>#1</literal>, <literal>#2</literal>,
... <literal>#<replaceable>nargs</replaceable></literal> are replaced by the values that you supply when
you call the command (or by the default value if there is an optional
argument and you don&#8217;t exercise the option).
</para>
</listitem></varlistentry></variablelist>
<para>&tex; ignores spaces in the source following an alphabetic control
sequence, as in &#8216;<literal>\cmd </literal>&#8217;.  If you actually want a space there, one
solution is to type <literal>{}</literal> after the command (&#8216;<literal>\cmd{} </literal>&#8217;, and
another solution is to use an explicit control space (&#8216;<literal>\cmd\ </literal>&#8217;).
</para>
<para>A simple example of defining a new command:
<literal>\newcommand{\RS}{Robin Smith}</literal> results in <literal>\RS</literal> being
replaced by the longer text.  Redefining an existing command is similar:
<literal>\renewcommand{\qedsymbol}{{\small QED}}</literal>.
</para>
<para>If you try to define a command and the name has already been used then
you get something like &#8216;<literal>LaTeX Error: Command \fred already
defined. Or name \end... illegal, see p.192 of the manual</literal>&#8217;.  If you try
to redefine a command and the name has not yet been used then you get
something like &#8216;<literal>LaTeX Error: \hank undefined</literal>&#8217;.
</para>
<para>Here the first command definition has no arguments, and the second has
one required argument.
</para>
<screen>\newcommand{\student}{Ms~O'Leary}
\newcommand{\defref}[1]{Definition~\ref{#1}}
</screen>
<para>Use the first as in <literal>I highly recommend \student{} to you</literal>.  The
second has a variable, so that <literal>\defref{def:basis}</literal> expands to
<literal>Definition~\ref{def:basis}</literal>, which ultimately expands to
something like &#8216;<literal>Definition~3.14</literal>&#8217;.
</para>
<para>Similarly, but with two required arguments:
<literal>\newcommand{\nbym}[2]{$#1 \times #2$}</literal> is invoked as
<literal>\nbym{2}{k}</literal>.
</para>
<para>This example has an optional argument.
</para>
<screen>\newcommand{\salutation}[1][Sir or Madam]{Dear #1:}
</screen>
<para>Then <literal>\salutation</literal> gives &#8216;<literal>Dear Sir or Madam:</literal>&#8217; while
<literal>\salutation[John]</literal> gives &#8216;<literal>Dear John:</literal>&#8217;.  And
<literal>\salutation[]</literal> gives &#8216;<literal>Dear :</literal>&#8217;.
</para>
<para>This example has an optional argument and two required arguments.
</para>
<screen>\newcommand{\lawyers}[3][company]{#2, #3, and~#1}
I employ \lawyers[Howe]{Dewey}{Cheatem}.
</screen>
<para>The output is &#8216;<literal>I employ Dewey, Cheatem, and Howe</literal>&#8217;.  The optional
argument, the <literal>Howe</literal>, is associated with <literal>#1</literal>, while
<literal>Dewey</literal> and <literal>Cheatem</literal> are associated with <literal>#2</literal>
and&#160;<literal>#3</literal>.  Because of the optional argument,
<literal>\lawyers{Dewey}{Cheatem}</literal> will give the output &#8216;<literal>I employ
Dewey, Cheatem, and company</literal>&#8217;.
</para>
<para>The braces around <replaceable>defn</replaceable> do not define a group, that is, they do not
delimit the scope of the result of expanding <replaceable>defn</replaceable>.  For example,
with <literal>\newcommand{\shipname}[1]{\it #1}</literal>, in this sentence,
</para>
<screen>The \shipname{Monitor} met the \shipname{Merrimac}.
</screen>
<para>the words &#8216;<literal>met the</literal>&#8217; would incorrectly be in italics.  The solution
is to put another pair of braces inside the definition:
<literal>\newcommand{\shipname}[1]{{\it #1}}</literal>.
</para>

</sect1>
<sect1 label="12.2" id="_005cprovidecommand">
<title><literal>\providecommand</literal></title>

<indexterm role="fn"><primary>\providecommand</primary></indexterm>
<indexterm role="cp"><primary>commands, defining new ones</primary></indexterm>
<indexterm role="cp"><primary>defining a new command</primary></indexterm>
<indexterm role="cp"><primary>new commands, defining</primary></indexterm>

<para>Synopses, one of:
</para>
<screen>\providecommand{<replaceable>cmd</replaceable>}{<replaceable>defn</replaceable>}
\providecommand{<replaceable>cmd</replaceable>}[<replaceable>nargs</replaceable>]{<replaceable>defn</replaceable>}
\providecommand{<replaceable>cmd</replaceable>}[<replaceable>nargs</replaceable>][<replaceable>optargdefault</replaceable>]{<replaceable>defn</replaceable>}
\providecommand*{<replaceable>cmd</replaceable>}{<replaceable>defn</replaceable>}
\providecommand*{<replaceable>cmd</replaceable>}[<replaceable>nargs</replaceable>]{<replaceable>defn</replaceable>}
\providecommand*{<replaceable>cmd</replaceable>}[<replaceable>nargs</replaceable>][<replaceable>optargdefault</replaceable>]{<replaceable>defn</replaceable>}
</screen>
<para>Defines a command, as long as no command of this name already exists.
If no command of this name already exists then this has the same effect
as <literal>\newcommand</literal>.  If a command of this name already exists then
this definition does nothing.  This is particularly useful in a file
that may be loaded more than once, such as a style file.
See <link linkend="_005cnewcommand-_0026-_005crenewcommand">\newcommand &amp; \renewcommand</link> for the description of the arguments.
</para>
<para>This example
</para>
<screen>\providecommand{\myaffiliation}{Saint Michael's College}
\providecommand{\myaffiliation}{Saint Michael's College}
From \myaffiliation.
</screen>
<para>outputs &#8216;<literal>From Saint Michael's College</literal>&#8217;.  Unlike <literal>\newcommand</literal>,
the repeated use of <literal>\providecommand</literal> does not give an error.
</para>

</sect1>
<sect1 label="12.3" id="_005cnewcounter">
<title><literal>\newcounter</literal>: Allocating a counter</title>

<indexterm role="fn"><primary>\newcounter</primary></indexterm>
<indexterm role="cp"><primary>counters, defining new</primary></indexterm>

<para>Synopsis, one of:
</para>
<screen>\newcounter{<replaceable>countername</replaceable>}
\newcounter{<replaceable>countername</replaceable>}[<replaceable>supercounter</replaceable>]
</screen>
<para>Globally defines a new counter named <replaceable>countername</replaceable> and initialize it
to zero (see <link linkend="Counters">Counters</link>).
</para>
<para>The name <replaceable>countername</replaceable> must consist of letters only. It does not
begin with a backslash.  This name must not already be in use by another
counter.
</para>
<para>When you use the optional argument <literal>[<replaceable>supercounter</replaceable>]</literal> then the
counter <replaceable>countername</replaceable> will be reset to zero whenever
<replaceable>supercounter</replaceable> is incremented.  For example, ordinarily
<literal>subsection</literal> is numbered within <literal>section</literal> so that any time you
increment <replaceable>section</replaceable>, either with <literal>\stepcounter</literal>
(see <link linkend="_005cstepcounter">\stepcounter</link>) or <literal>\refstepcounter</literal>
(see <link linkend="_005crefstepcounter">\refstepcounter</link>), then &latex; will reset <replaceable>subsection</replaceable> to
zero.
</para>
<para>This example
</para>
<screen>\newcounter{asuper}  \setcounter{asuper}{1}
\newcounter{asub}[asuper] \setcounter{asub}{3}   % Note `asuper'
The value of asuper is \arabic{asuper} and of asub is \arabic{asub}.
\stepcounter{asuper}
Now asuper is \arabic{asuper} while asub is \arabic{asub}.
</screen>
<para>produces &#8216;<literal>The value of asuper is 1 and that of asub is 3</literal>&#8217; and
&#8216;<literal>Now asuper is 2 while asub is 0</literal>&#8217;.
</para>
<para>If the counter already exists, for instance by entering <literal>asuper</literal>
twice, then you get something like &#8216;<literal>LaTeX Error: Command \c@asuper
already defined.  Or name \end... illegal, see p.192 of the manual.</literal>&#8217;.
</para>
<para>If you use the optional argument then the super counter must already
exist. Entering <literal>\newcounter{jh}[lh]</literal> when <literal>lh</literal> is not a
defined counter will get you &#8216;<literal>LaTeX Error: No counter 'lh'
defined.</literal>&#8217;
</para>

</sect1>
<sect1 label="12.4" id="_005cnewlength">
<title><literal>\newlength</literal></title>

<indexterm role="fn"><primary>\newlength</primary></indexterm>
<indexterm role="cp"><primary>lengths, allocating new</primary></indexterm>
<indexterm role="cp"><primary>rubber lengths, defining new</primary></indexterm>
<indexterm role="cp"><primary>skip register, plain &tex;</primary></indexterm>
<indexterm role="cp"><primary>glue register, plain &tex;</primary></indexterm>

<para>Synopsis:
</para>
<screen>\newlength{<replaceable>arg</replaceable>}
</screen>
<para>Allocate a new length register (see <link linkend="Lengths">Lengths</link>).  The required argument
<replaceable>arg</replaceable> must begin with a backslash, <literal>\</literal>.  The new register holds
rubber lengths such as <literal>72.27pt</literal> or <literal>1in plus.2in minus.1in</literal>
(a &latex; length register is what plain &tex; calls a <literal>skip</literal>
register).  The initial value is zero.  The control sequence
<literal>\<replaceable>arg</replaceable></literal> must not be already defined.
</para>
<para>An example:
</para>
<screen>\newlength{\graphichgt}
</screen>
<para>If you forget the backslash then you get &#8216;<literal>Missing control sequence
inserted</literal>&#8217;.  If the command sequence already exists then you get
something like &#8216;<literal>LaTeX Error: Command \graphichgt already defined.
Or name \end... illegal, see p.192 of the manual</literal>&#8217;.
</para>

</sect1>
<sect1 label="12.5" id="_005cnewsavebox">
<title><literal>\newsavebox</literal></title>

<indexterm role="fn"><primary>\newsavebox</primary></indexterm>
<indexterm role="cp"><primary>box, allocating new</primary></indexterm>

<para>Synopsis:
</para>
<screen>\newsavebox{<replaceable>cmd</replaceable>}
</screen>
<para>Define <literal>\<replaceable>cmd</replaceable></literal> to refer to a new &#8220;bin&#8221; for storing boxes.
Such a box is for holding typeset material, to use multiple times or to
measure or manipulate (see <link linkend="Boxes">Boxes</link>).  The required bin name
<literal><replaceable>cmd</replaceable></literal> must start with a backslash, <literal>\</literal>, and must not
already be defined.  This command is fragile (see <link linkend="_005cprotect">\protect</link>).
</para>
<para>The first line here sets you up to save the material for later use. 
</para>
<screen>\newsavebox{\logobox}
\savebox{\logobox}{LoGo}
Our logo is \usebox{\logobox}. 
</screen>
<para>The output is &#8216;<literal>Our logo is LoGo</literal>&#8217;.
</para>
<para>If there is an already defined bin then you get something like
&#8216;<literal>LaTeX Error: Command \logobox already defined. Or name
\end... illegal, see p.192 of the manual</literal>&#8217;.
</para>
<para>The allocation of a box is global.
</para>

</sect1>
<sect1 label="12.6" id="_005cnewenvironment-_0026-_005crenewenvironment">
<title><literal>\newenvironment</literal> &amp; <literal>\renewenvironment</literal></title>

<indexterm role="fn"><primary>\newenvironment</primary></indexterm>
<indexterm role="fn"><primary>\renewenvironment</primary></indexterm>
<indexterm role="cp"><primary>environments, defining</primary></indexterm>
<indexterm role="cp"><primary>defining new environments</primary></indexterm>
<indexterm role="cp"><primary>redefining environments</primary></indexterm>

<para>Synopses, one of:
</para>
<screen>\newenvironment{<replaceable>env</replaceable>}{<replaceable>begdef</replaceable>}{<replaceable>enddef</replaceable>}
\newenvironment{<replaceable>env</replaceable>}[<replaceable>nargs</replaceable>]{<replaceable>begdef</replaceable>}{<replaceable>enddef</replaceable>}
\newenvironment{<replaceable>env</replaceable>}[<replaceable>nargs</replaceable>][<replaceable>optargdefault</replaceable>]{<replaceable>begdef</replaceable>}{<replaceable>enddef</replaceable>}
\newenvironment*{<replaceable>env</replaceable>}{<replaceable>begdef</replaceable>}{<replaceable>enddef</replaceable>}
\newenvironment*{<replaceable>env</replaceable>}[<replaceable>nargs</replaceable>]{<replaceable>begdef</replaceable>}{<replaceable>enddef</replaceable>}
\newenvironment*{<replaceable>env</replaceable>}[<replaceable>nargs</replaceable>][<replaceable>optargdefault</replaceable>]{<replaceable>begdef</replaceable>}{<replaceable>enddef</replaceable>}
</screen>
<para>or one of these.
</para>
<screen>\renewenvironment{<replaceable>env</replaceable>}{<replaceable>begdef</replaceable>}{<replaceable>enddef</replaceable>}
\renewenvironment{<replaceable>env</replaceable>}[<replaceable>nargs</replaceable>]{<replaceable>begdef</replaceable>}{<replaceable>enddef</replaceable>}
\renewenvironment{<replaceable>env</replaceable>}[<replaceable>nargs</replaceable>][<replaceable>optargdefault</replaceable>]{<replaceable>begdef</replaceable>}{<replaceable>enddef</replaceable>}
\renewenvironment*{<replaceable>env</replaceable>}{<replaceable>begdef</replaceable>}{<replaceable>enddef</replaceable>}
\renewenvironment*{<replaceable>env</replaceable>}[<replaceable>nargs</replaceable>]{<replaceable>begdef</replaceable>}{<replaceable>enddef</replaceable>}
\renewenvironment*{<replaceable>env</replaceable>}[<replaceable>nargs</replaceable>][<replaceable>optargdefault</replaceable>]{<replaceable>begdef</replaceable>}{<replaceable>enddef</replaceable>}
</screen>
<para>Define or redefine the environment <replaceable>env</replaceable>, that is, create the
construct <literal>\begin{<replaceable>env</replaceable>} ... <replaceable>body</replaceable> ... \end{<replaceable>env</replaceable>}</literal>.
</para>
<indexterm role="cp"><primary><literal>*</literal>-form of environment commands</primary></indexterm>
<para>The starred form of these commands requires that the arguments not
contain multiple paragraphs of text.  However, the body of these
environments can contain multiple paragraphs.
</para>
<variablelist><varlistentry><term><replaceable>env</replaceable>
</term><listitem><para>Required; the environment name.  It consists only of letters or the
<literal>*</literal> character, and thus does not begin with backslash, <literal>\</literal>.
It must not begin with the string <literal>end</literal>.  For
<literal>\newenvironment</literal>, the name <replaceable>env</replaceable> must not be the name of an
already existing environment, and also the command <literal>\<replaceable>env</replaceable></literal>
must be undefined.  For <literal>\renewenvironment</literal>, <replaceable>env</replaceable> must be the
name of an existing environment.
</para>
</listitem></varlistentry><varlistentry><term><replaceable>nargs</replaceable>
</term><listitem><para>Optional; an integer from 0 to 9 denoting the number of arguments of
that the environment takes.  When you use the environment these
arguments appear after the <literal>\begin</literal>, as in
<literal>\begin{<replaceable>env</replaceable>}{<replaceable>arg1</replaceable>} ... {<replaceable>argn</replaceable>}</literal>.  Omitting
this is equivalent to setting it to 0; the environment will have no
arguments.  When redefining an environment, the new version can have a
different number of arguments than the old version.
</para>
</listitem></varlistentry><varlistentry><term><replaceable>optargdefault</replaceable>
</term><listitem><para>Optional; if this is present then the first argument of the defined
environment is optional, with default value <replaceable>optargdefault</replaceable> (which
may be the empty string).  If this is not in the definition then the
environment does not take an optional argument.
</para>
<para>That is, when <replaceable>optargdefault</replaceable> is present in the definition of the
environment then you can start the environment with square brackets, as
in <literal>\begin{<replaceable>env</replaceable>}[<replaceable>optval</replaceable>]{...} ... \end{<replaceable>env</replaceable>}</literal>.
In this case, within <replaceable>begdefn</replaceable> the parameter <literal>#1</literal> is set to the
value of <replaceable>optval</replaceable>.  If you call <literal>\begin{<replaceable>env</replaceable>}</literal> without
square brackets, then within <replaceable>begdefn</replaceable> the parameter <literal>#1</literal> is
set to the value of the default <replaceable>optargdefault</replaceable>.  In either case,
any required arguments start with <literal>#2</literal>.
</para>
<para>Omitting <literal>[<replaceable>myval</replaceable>]</literal> in the call is different than having the
square brackets with no contents, as in <literal>[]</literal>.  The former results
in <literal>#1</literal> expanding to <replaceable>optargdefault</replaceable>; the latter results in
<literal>#1</literal> expanding to the empty string.
</para>
</listitem></varlistentry><varlistentry><term><replaceable>begdef</replaceable>
</term><listitem><para>Required; the text expanded at every occurrence of
<literal>\begin{<replaceable>env</replaceable>}</literal>.  Within <replaceable>begdef</replaceable>, the parameters
<literal>#1</literal>, <literal>#2</literal>, ... <literal>#<replaceable>nargs</replaceable></literal>, are replaced by the
values that you supply when you call the environment; see the examples
below.
</para>
</listitem></varlistentry><varlistentry><term><replaceable>enddef</replaceable>
</term><listitem><para>Required; the text expanded at every occurrence of
<literal>\end{<replaceable>env</replaceable>}</literal>.  This may not contain any parameters, that is,
you cannot use <literal>#1</literal>, <literal>#2</literal>, etc., here (but see the final
example below).
</para>
</listitem></varlistentry></variablelist>
<para>All environments, that is to say the <replaceable>begdef</replaceable> code, the environment
body, and the <replaceable>enddef</replaceable> code, are processed within a group.  Thus, in
the first example below, the effect of the <literal>\small</literal> is limited to
the quote and does not extend to material following the environment.
</para>
<para>If you try to define an environment and the name has already been used
then you get something like &#8216;<literal>LaTeX Error: Command \fred already
defined. Or name \end... illegal, see p.192 of the manual</literal>&#8217;.  If you try
to redefine an environment and the name has not yet been used then you
get something like &#8216;<literal>LaTeX Error: Environment hank undefined.</literal>&#8217;.
</para>
<para>This example gives an environment like &latex;&#8217;s <literal>quotation</literal>
except that it will be set in smaller type.
</para>
<screen>\newenvironment{smallquote}{%
  \small\begin{quotation}
}{%
  \end{quotation}
}
</screen>
<para>This has an argument, which is set in boldface at the start of a
paragraph.
</para>
<screen>\newenvironment{point}[1]{%
  \noindent\textbf{#1}
}{%
}
</screen>
<para>This one shows the use of a optional argument; it gives a quotation
environment that cites the author.
</para>
<screen>\newenvironment{citequote}[1][Shakespeare]{%
  \begin{quotation}
  \noindent\textit{#1}: 
}{%
  \end{quotation}
}
</screen>
<para>The author&#8217;s name is optional, and defaults to &#8216;<literal>Shakespeare</literal>&#8217;.  In
the document, use the environment like this.
</para>
<screen>\begin{citequote}[Lincoln]
  ...
\end{citequote}
</screen>
<para>The final example shows how to save the value of an argument to use in 
<replaceable>enddef</replaceable>, in this case in a box (see <link linkend="_005csbox-_0026-_005csavebox">\sbox &amp; \savebox</link>).
</para>
<screen>\newsavebox{\quoteauthor}
\newenvironment{citequote}[1][Shakespeare]{%
  \sbox\quoteauthor{#1}%
  \begin{quotation} 
}{%
  \hspace{1em plus 1fill}---\usebox{\quoteauthor}
  \end{quotation}
}
</screen>

</sect1>
<sect1 label="12.7" id="_005cnewtheorem">
<title><literal>\newtheorem</literal></title>

<indexterm role="fn"><primary>\newtheorem</primary></indexterm>
<indexterm role="cp"><primary>theorems, defining</primary></indexterm>
<indexterm role="cp"><primary>defining new theorems</primary></indexterm>

<indexterm role="cp"><primary>theorem-like environment</primary></indexterm>
<indexterm role="cp"><primary>environment, theorem-like</primary></indexterm>
<para>Synopses:
</para>
<screen>\newtheorem{<replaceable>name</replaceable>}{<replaceable>title</replaceable>}
\newtheorem{<replaceable>name</replaceable>}{<replaceable>title</replaceable>}[<replaceable>numbered_within</replaceable>]
\newtheorem{<replaceable>name</replaceable>}[<replaceable>numbered_like</replaceable>]{<replaceable>title</replaceable>}
</screen>
<para>Define a new theorem-like environment.  You can specify one of
<replaceable>numbered_within</replaceable> and <replaceable>numbered_like</replaceable>, or neither, but not both.
</para>
<para>The first form, <literal>\newtheorem{<replaceable>name</replaceable>}{<replaceable>title</replaceable>}</literal>, creates
an environment that will be labelled with <replaceable>title</replaceable>; see the first
example below.
</para>
<para>The second form,
<literal>\newtheorem{<replaceable>name</replaceable>}{<replaceable>title</replaceable>}[<replaceable>numbered_within</replaceable>]</literal>,
creates an environment whose counter is subordinate to the existing
counter <replaceable>numbered_within</replaceable>, so this counter will be reset when
<replaceable>numbered_within</replaceable> is reset.  See the second example below.
</para>
<para>The third form
<literal>\newtheorem{<replaceable>name</replaceable>}[<replaceable>numbered_like</replaceable>]{<replaceable>title</replaceable>}</literal>,
with optional argument between the two required arguments, creates an
environment whose counter will share the previously defined counter
<replaceable>numbered_like</replaceable>.  See the third example.
</para>
<para>This command creates a counter named <replaceable>name</replaceable>.  In addition, unless
the optional argument <replaceable>numbered_like</replaceable> is used, inside of the
theorem-like environment the current <literal>\ref</literal> value will be that of
<literal>\the<replaceable>numbered_within</replaceable></literal> (see <link linkend="_005cref">\ref</link>).
</para>
<para>This declaration is global.  It is fragile (see <link linkend="_005cprotect">\protect</link>).
</para>
<para>Arguments:
</para>
<variablelist><varlistentry><term><replaceable>name</replaceable>
</term><listitem><para>The name of the environment.  It is a string of letters.  It must not
begin with a backslash, <literal>\</literal>.  It must not be the name of an
existing environment, and the command name <literal>\<replaceable>name</replaceable></literal> must not
already be defined.
</para>
</listitem></varlistentry><varlistentry><term><replaceable>title</replaceable>
</term><listitem><para>The text to be printed at the beginning of the environment, before the
number. For example, &#8216;<literal>Theorem</literal>&#8217;.
</para>
</listitem></varlistentry><varlistentry><term><replaceable>numbered_within</replaceable>
</term><listitem><para>Optional; the name of an already defined counter, usually a sectional
unit such as <literal>chapter</literal> or <literal>section</literal>.  When the
<replaceable>numbered_within</replaceable> counter is reset then the <replaceable>name</replaceable> environment&#8217;s
counter will also be reset.
</para>
<para>If this optional argument is not used then the command
<literal>\the<replaceable>name</replaceable></literal> is set to <literal>\arabic{<replaceable>name</replaceable>}</literal>.
</para>
</listitem></varlistentry><varlistentry><term><replaceable>numbered_like</replaceable>
</term><listitem><para>Optional; the name of an already defined theorem-like environment. The
new environment will be numbered in sequence with <replaceable>numbered_like</replaceable>.
</para>
</listitem></varlistentry></variablelist>
<para>Without any optional arguments the environments are numbered
sequentially.  The example below has a declaration in the preamble that
results in &#8216;<literal>Definition&#160;1</literal>&#8217; and &#8216;<literal>Definition&#160;2</literal>&#8217; in the
output.
</para>
<screen>\newtheorem{defn}{Definition}
\begin{document}
\section{...}
\begin{defn}
  First def 
\end{defn}

\section{...}
\begin{defn}
  Second def
\end{defn}
</screen>
<para>This example has the same document body as the prior one.  But here
<literal>\newtheorem</literal>&#8217;s optional argument <replaceable>numbered_within</replaceable> is given as
<literal>section</literal>, so the output is like &#8216;<literal>Definition&#160;1.1</literal>&#8217; and
&#8216;<literal>Definition&#160;2.1</literal>&#8217;.
</para>
<screen>\newtheorem{defn}{Definition}[section]
\begin{document}
\section{...}
\begin{defn}
  First def 
\end{defn}

\section{...}
\begin{defn}
  Second def
\end{defn}
</screen>
<para>In the next example there are two declarations in the preamble, the
second of which calls for the new <literal>thm</literal> environment to use the same
counter as <literal>defn</literal>.  It gives &#8216;<literal>Definition&#160;1.1</literal>&#8217;, followed
by &#8216;<literal>Theorem&#160;2.1</literal>&#8217; and &#8216;<literal>Definition&#160;2.2</literal>&#8217;.
</para>
<screen>\newtheorem{defn}{Definition}[section]
\newtheorem{thm}[defn]{Theorem}
\begin{document}
\section{...}
\begin{defn}
  First def 
\end{defn}

\section{...}
\begin{thm}
  First thm
\end{thm}

\begin{defn}
  Second def
\end{defn}
</screen>

</sect1>
<sect1 label="12.8" id="_005cnewfont">
<title><literal>\newfont</literal></title>

<indexterm role="fn"><primary>\newfont</primary></indexterm>
<indexterm role="cp"><primary>fonts, new commands for</primary></indexterm>
<indexterm role="cp"><primary>defining new fonts</primary></indexterm>

<!-- @findex .fd @r{file} -->
<para>This command is obsolete. This description is here only to help with old
documents.  New documents should define fonts in families through the
New Font Selection Scheme which allows you to, for example, associate a
boldface with a roman (see <link linkend="Fonts">Fonts</link>).
<!-- This is done either by using -->
<!-- @file{.fd} files or through the use of an engine that can access system -->
<!-- fonts such as Xe@LaTeX{} (@pxref{@TeX{} engines}). -->
</para>
<para>Synopsis:
</para>
<screen>\newfont{\<replaceable>cmd</replaceable>}{<replaceable>font description</replaceable>}
</screen>
<para>Define a command <literal>\<replaceable>cmd</replaceable></literal> that will change the current font.
The control sequence must must not already be defined.  It must begin
with a backslash, <literal>\</literal>.
</para>
<indexterm role="cp"><primary>at clause, in font definitions</primary></indexterm>
<indexterm role="cp"><primary>design size, in font definitions</primary></indexterm>
<para>The <replaceable>font description</replaceable> consists of a <replaceable>fontname</replaceable> and an optional
<firstterm>at clause</firstterm>.  &latex; will look on your system for a file named
<filename><replaceable>fontname</replaceable>.tfm</filename>.  The at clause can have the form either
<literal>at <replaceable>dimen</replaceable></literal> or <literal>scaled <replaceable>factor</replaceable></literal>, where a
<replaceable>factor</replaceable> of &#8216;<literal>1000</literal>&#8217; means no scaling.  For &latex;&#8217;s purposes,
all this does is scale all the character and other font dimensions
relative to the font&#8217;s design size, which is a value defined in the
<filename>.tfm</filename> file.
</para>
<para>This defines two equivalent fonts and typesets a few characters in each.
</para>
<screen>\newfont{\testfontat}{cmb10 at 11pt}
\newfont{\testfontscaled}{cmb10 scaled 1100}
\testfontat abc
\testfontscaled abc
</screen>

</sect1>
<sect1 label="12.9" id="_005cprotect">
<title><literal>\protect</literal></title>

<indexterm role="fn"><primary>\protect</primary></indexterm>
<indexterm role="cp"><primary>fragile commands</primary></indexterm>
<indexterm role="cp"><primary>robust commands</primary></indexterm>

<para>All &latex; commands are either <firstterm>fragile</firstterm> or <firstterm>robust</firstterm>.  A
fragile command can break when it is used in the argument to certain
other commands.  Commands that contain data that &latex; writes to an
auxiliary file and re-reads later are fragile.  This includes material
that goes into a table of contents, list of figures, list of tables,
etc.  Fragile commands also include line breaks, any command that has an
optional argument, and many more. To prevent such commands from
breaking, one solution is to preceded them with the command
<literal>\protect</literal>.
</para>
<para>For example, when &latex; runs the <literal>\section{<replaceable>section
name</replaceable>}</literal> command it writes the <replaceable>section name</replaceable> text to the
<filename>.aux</filename> auxiliary file, moving it there for use elsewhere in the
document such as in the table of contents.  Any argument that is
internally expanded by &latex; without typesetting it directly is
referred to as a 
<indexterm role="cp"><primary>moving arguments</primary></indexterm>
<firstterm>moving argument</firstterm>.  A command is fragile if it can
expand during this process into invalid &tex; code.  Some examples of
moving arguments are those that appear in the <literal>\caption{...}</literal>
command (see <link linkend="figure">figure</link>), in the <literal>\thanks{...}</literal> command
(see <link linkend="_005cmaketitle">\maketitle</link>), and in @-expressions in the <literal>tabular</literal> and
<literal>array</literal> environments (see <link linkend="tabular">tabular</link>).
</para>
<para>If you get strange errors from commands used in moving arguments, try
preceding it with <literal>\protect</literal>.  Every fragile commands must be
protected with their own <literal>\protect</literal>.  
</para>
<para>Although usually a <literal>\protect</literal> command doesn&#8217;t hurt, length
commands are robust and should not be preceded by a <literal>\protect</literal>
command. Nor can a <literal>\protect</literal> command be used in the argument to
<literal>\addtocounter</literal> or <literal>\setcounter</literal> command.
</para>
<para>In this example the <literal>\caption</literal> command gives a mysterious error
about an extra curly brace.  Fix the problem by preceding each
<literal>\raisebox</literal> command with <literal>\protect</literal>.
</para>
<screen>\begin{figure}
  ...
  \caption{Company headquarters of A\raisebox{1pt}{B}\raisebox{-1pt}{C}}
\end{figure}
</screen>
<para>In the next example the <literal>\tableofcontents</literal> command gives an error
because the <literal>\(..\)</literal> in the section title expands to illegal &tex;
in the <filename>.toc</filename> file.  You can solve this by changing <literal>\(..\)</literal>
to <literal>\protect\(..\protect\)</literal>.
</para>
<screen>\begin{document}
\tableofcontents
...
\section{Einstein's \( e=mc^2 \)}
...
</screen>

</sect1>
<sect1 label="12.10" id="_005cignorespaces-_0026-_005cignorespacesafterend">
<title><literal>\ignorespaces &amp; \ignorespacesafterend</literal></title>

<indexterm role="fn"><primary>\ignorespaces</primary></indexterm>
<indexterm role="fn"><primary>\ignorespacesafterend</primary></indexterm>
<indexterm role="cp"><primary>spaces, ignore around commands</primary></indexterm>
<indexterm role="cp"><primary>commands, ignore spaces</primary></indexterm>

<para>Synopsis:
</para>
<screen>\ignorespaces
</screen>
<para>or
</para>
<screen>\ignorespacesafterend
</screen>
<para>Both commands cause &latex; to ignore spaces after the end of the
command up until the first non-space character.  The first is a command
from Plain &tex;, and the second is &latex;-specific.
</para>
<para>The <literal>ignorespaces</literal> is often used when defining commands via
<literal>\newcommand</literal>, or <literal>\newenvironment</literal>, or <literal>\def</literal>.  The
example below illustrates.  It allows a user to show the points values
for quiz questions in the margin but it is inconvenient because, as
shown in the <literal>enumerate</literal> list, users must not put any space between
the command and the question text.
</para>
<screen>\newcommand{\points}[1]{\makebox[0pt]{\makebox[10em][l]{#1~pts}}
\begin{enumerate} 
  \item\points{10}no extra space output here
  \item\points{15} extra space between the number and the `extra'
\end{enumerate} 
</screen>
<para>The solution is to change to this.
</para>
<screen>\newcommand{\points}[1]{%
  \makebox[0pt]{\makebox[10em][l]{#1~pts}}\ignorespaces}
</screen>
<para>A second example shows spaces being removed from the front of text.  The
commands below allow a user to uniformly attach a title to names.  But,
as given, if a title accidentally starts with a space then
<literal>\fullname</literal> will reproduce that.
</para>
<screen>\makeatletter
\newcommand{\honorific}[1]{\def\@honorific{#1}} % remember title
\newcommand{\fullname}[1]{\@honorific~#1}       % put title before name
\makeatother
\begin{tabular}{|l|}
\honorific{Mr/Ms}  \fullname{Jones} \\  % no extra space here
\honorific{ Mr/Ms} \fullname{Jones}     % extra space before title
\end{tabular}
</screen>
<para>To fix this, change to
<literal>\newcommand{\fullname}[1]{\ignorespaces\@honorific~#1}</literal>.
</para>
<para>The <literal>\ignorespaces</literal> is also often used in a <literal>\newenvironment</literal>
at the end of the <replaceable>begin</replaceable> clause, that is, as part of the second
argument, as in <literal>\begin{newenvironment}{<replaceable>env
name</replaceable>}{... \ignorespaces}{...}</literal>.
</para>
<para>To strip spaces off the end of an environment use
<literal>\ignorespacesafterend</literal>.  An example is that this will show a much
larger vertical space between the first and second environments than
between the second and third.
</para>
<screen>\newenvironment{eq}{\begin{equation}}{\end{equation}}
\begin{eq}
e=mc^2
\end{eq}
\begin{equation}
F=ma
\end{equation}
\begin{equation}
E=IR
\end{equation}
</screen>
<para>Putting a comment character&#160;<literal>%</literal> immediately after the
<literal>\end{eq}</literal> will make the vertical space disappear, but that is
inconvenient.  The solution is to change to
<literal>\newenvironment{eq}{\begin{equation}}{\end{equation}\ignorespacesafterend}</literal>.
</para>

</sect1>
</chapter>
<chapter label="13" id="Counters">
<title>Counters</title>

<indexterm role="cp"><primary>counters, a list of</primary></indexterm>
<indexterm role="cp"><primary>variables, a list of</primary></indexterm>

<para>Everything &latex; numbers for you has a counter associated with
it. The name of the counter is often the same as the name of the
environment or command associated with the number, except that the
counter&#8217;s name has no backslash&#160;<literal>\</literal>.  Thus, associated with
the <literal>\chapter</literal> command is the <literal>chapter</literal> counter that keeps
track of the chapter number.  
</para>
<para>Below is a list of the counters used in &latex;&#8217;s standard document
classes to control numbering.
</para>
<screen>part            paragraph       figure          enumi
chapter         subparagraph    table           enumii
section         page            footnote        enumiii
subsection      equation        mpfootnote      enumiv
subsubsection
</screen>
<para>The <literal>mpfootnote</literal> counter is used by the <literal>\footnote</literal> command
inside of a minipage (see <link linkend="minipage">minipage</link>).  The counters <literal>enumi</literal>
through <literal>enumiv</literal> are used in the <literal>enumerate</literal> environment, for
up to four levels of nesting (see <link linkend="enumerate">enumerate</link>).
</para>
<para>Counters can have any integer value but they are typically positive.
</para>
<para>New counters are created with <literal>\newcounter</literal>.  See <link linkend="_005cnewcounter">\newcounter</link>.
</para>



<sect1 label="13.1" id="_005calph-_005cAlph-_005carabic-_005croman-_005cRoman-_005cfnsymbol">
<title><literal>\alph \Alph \arabic \roman \Roman \fnsymbol</literal>: Printing counters</title>

<indexterm role="cp"><primary>counters, printing</primary></indexterm>

<para>Print the value of a counter, in a specified style.  For instance, if
the counter <replaceable>counter</replaceable> has the value 1 then a
<literal>\alph{<replaceable>counter</replaceable>}</literal> in your source will result in a lowercase
letter&#160;a appearing in the output.
</para>
<para>All of these commands take a single counter as an argument, for
instance, <literal>\alph{enumi}</literal>.  Note that the counter name does not
start with a backslash.
</para>
<variablelist><varlistentry><term><indexterm role="fn"><primary>\alph{<replaceable>counter</replaceable>}</primary></indexterm><literal>\alph{<replaceable>counter</replaceable>}</literal>
</term><listitem><para>Print the value of <replaceable>counter</replaceable> in lowercase letters: &#8216;a&#8217;, &#8216;b&#8217;,
... If the counter&#8217;s value is less than 1 or more than 26 then
you get &#8216;<literal>LaTeX Error: Counter too large.</literal>&#8217;
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\Alph{<replaceable>counter</replaceable>}</primary></indexterm><literal>\Alph{<replaceable>counter</replaceable>}</literal>
</term><listitem><para>Print in uppercase letters: &#8216;A&#8217;, &#8216;B&#8217;, ... If the counter&#8217;s value
is less than 1 or more than 26 then you get &#8216;<literal>LaTeX Error: Counter
too large.</literal>&#8217;
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\arabic{<replaceable>counter</replaceable>}</primary></indexterm><literal>\arabic{<replaceable>counter</replaceable>}</literal>
</term><listitem><para>Print in Arabic numbers such as &#8216;<literal>5</literal>&#8217; or &#8216;<literal>-2</literal>&#8217;.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\roman{<replaceable>counter</replaceable>}</primary></indexterm><literal>\roman{<replaceable>counter</replaceable>}</literal>
</term><listitem><para>Print in lowercase roman numerals: &#8216;i&#8217;, &#8216;ii&#8217;, ... If the
counter&#8217;s value is less than 1 then you get no warning or error but
&latex; does not print anything in the output.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\Roman{<replaceable>counter</replaceable>}</primary></indexterm><literal>\Roman{<replaceable>counter</replaceable>}</literal>
</term><listitem><para>Print in uppercase roman numerals: &#8216;I&#8217;, &#8216;II&#8217;, ... If the
counter&#8217;s value is less than 1 then you get no warning or error but
&latex; does not print anything in the output.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\fnsymbol{<replaceable>counter</replaceable>}</primary></indexterm><literal>\fnsymbol{<replaceable>counter</replaceable>}</literal>
</term><listitem><para>Prints the value of <replaceable>counter</replaceable> using a sequence of nine symbols that
are traditionally used for labeling footnotes.  The value of
<replaceable>counter</replaceable> should be between&#160;1 and&#160;9, inclusive.  If the
counter&#8217;s value is less than 0 or more than 9 then you get &#8216;<literal>LaTeX
Error: Counter too large</literal>&#8217;, while if it is 0 then you get no error or
warning but &latex; does not output anything.
</para>
<para>Here are the symbols:
</para>
<informaltable><tgroup cols="4"><colspec colwidth="10*"></colspec><colspec colwidth="30*"></colspec><colspec colwidth="30*"></colspec><colspec colwidth="30*"></colspec><thead><row><entry><para>Number</para></entry><entry><para>Name</para></entry><entry><para>Command</para></entry><entry><para>Symbol
</para></entry></row></thead><tbody><row><entry><para>1</para></entry><entry><para>asterisk</para></entry><entry><para><literal>\ast</literal></para></entry><entry><para>*<!-- -->
</para></entry></row><row><entry><para>2</para></entry><entry><para>dagger</para></entry><entry><para><literal>\dagger</literal></para></entry><entry><para>&#x2020;
</para></entry></row><row><entry><para>3</para></entry><entry><para>ddagger</para></entry><entry><para><literal>\ddagger</literal></para></entry><entry><para>&#x2021;
</para></entry></row><row><entry><para>4</para></entry><entry><para>section-sign</para></entry><entry><para><literal>\S</literal></para></entry><entry><para>&#x00A7;
</para></entry></row><row><entry><para>5</para></entry><entry><para>paragraph-sign</para></entry><entry><para><literal>\P</literal></para></entry><entry><para>&#x00B6;
</para></entry></row><row><entry><para>6</para></entry><entry><para>double-vert</para></entry><entry><para><literal>\parallel</literal></para></entry><entry><para>&#x2016;
</para></entry></row><row><entry><para>7</para></entry><entry><para>double-asterisk</para></entry><entry><para><literal>\ast\ast</literal></para></entry><entry><para>**<!-- -->
</para></entry></row><row><entry><para>8</para></entry><entry><para>double-dagger</para></entry><entry><para><literal>\dagger\dagger</literal></para></entry><entry><para>&#x2020;&#x2020;
</para></entry></row><row><entry><para>9</para></entry><entry><para>double-ddagger</para></entry><entry><para><literal>\ddagger\ddagger</literal></para></entry><entry><para>&#x2021;&#x2021;
</para></entry></row></tbody></tgroup></informaltable>
</listitem></varlistentry></variablelist>

</sect1>
<sect1 label="13.2" id="_005cusecounter">
<title><literal>\usecounter</literal></title>

<indexterm role="fn"><primary>\usecounter</primary></indexterm>
<indexterm role="cp"><primary>list items, specifying counter</primary></indexterm>
<indexterm role="cp"><primary>numbered items, specifying counter</primary></indexterm>

<para>Synopsis:
</para>
<screen>\usecounter{<replaceable>counter</replaceable>}
</screen>
<para>Used in the second argument of the <literal>list</literal> environment
(see <link linkend="list">list</link>), this declares that list items will be numbered by
<replaceable>counter</replaceable>.  It initializes <replaceable>counter</replaceable> to zero, and arranges that
when <literal>\item</literal> is called without its optional argument then
<replaceable>counter</replaceable> is incremented by <literal>\refstepcounter</literal>, making its value
be the current <literal>ref</literal> value (see <link linkend="_005cref">\ref</link>).  This command is fragile
(see <link linkend="_005cprotect">\protect</link>).
</para>
<para>Put in the document preamble, this example makes a new list environment
enumerated with <replaceable>testcounter</replaceable>:
</para>
<screen>\newcounter{testcounter}
\newenvironment{test}{%
  \begin{list}{}{%
    \usecounter{testcounter}
  }
}{%
  \end{list}
}
</screen>

</sect1>
<sect1 label="13.3" id="_005cvalue">
<title><literal>\value</literal></title>

<indexterm role="fn"><primary>\value</primary></indexterm>
<indexterm role="cp"><primary>counters, getting value of</primary></indexterm>

<para>Synopsis:
</para>
<screen>\value{<replaceable>counter</replaceable>}
</screen>
<para>Expands to the value of the counter <replaceable>counter</replaceable>. (Note that the name
of a counter does not begin with a backslash.)
</para>
<para>This example outputs &#8216;<literal>Test counter is&#160;6. Other counter
is&#160;5.</literal>&#8217;.
</para>
<screen>\newcounter{test} \setcounter{test}{5}
\newcounter{other} \setcounter{other}{\value{test}}
\addtocounter{test}{1}

Test counter is \arabic{test}.
Other counter is \arabic{other}.
</screen>
<para>The <literal>\value</literal> command is not used for typesetting the value of the
counter.  For that, see <link linkend="_005calph-_005cAlph-_005carabic-_005croman-_005cRoman-_005cfnsymbol">\alph \Alph \arabic \roman \Roman \fnsymbol</link>.
</para>
<para>It is often used in <literal>\setcounter</literal> or <literal>\addtocounter</literal> but
<literal>\value</literal> can be used anywhere that &latex; expects a number, such
as in <literal>\hspace{\value{foo}\parindent}</literal>.  It must not be
preceded by <literal>\protect</literal> (see <link linkend="_005cprotect">\protect</link>).
</para>
<para>This example inserts <literal>\hspace{4\parindent}</literal>.
</para>
<screen>\setcounter{myctr}{3} \addtocounter{myctr}{1}
\hspace{\value{myctr}\parindent}
</screen>

</sect1>
<sect1 label="13.4" id="_005csetcounter">
<title><literal>\setcounter</literal></title>

<indexterm role="fn"><primary>\setcounter</primary></indexterm>
<indexterm role="cp"><primary>counters, setting</primary></indexterm>
<indexterm role="cp"><primary>setting counters</primary></indexterm>

<para>Synopsis:
</para>
<screen>\setcounter{<replaceable>counter</replaceable>}{<replaceable>value</replaceable>}
</screen>
<para>Globally set the counter <replaceable>counter</replaceable> to have the value of the
<replaceable>value</replaceable> argument, which must be an integer.  Thus, you can set a
counter&#8217;s value as <literal>\setcounter{section}{5}</literal>.  Note that the
counter name does not start with a backslash.
</para>
<para>In this example if the counter <literal>theorem</literal> has value 12 then the
second line will print &#8216;<literal>XII</literal>&#8217;.
</para>
<screen>\setcounter{exercise}{\value{theorem}}
Here it is in Roman: \Roman{exercise}.
</screen>

</sect1>
<sect1 label="13.5" id="_005caddtocounter">
<title><literal>\addtocounter</literal></title>

<indexterm role="fn"><primary>\addtocounter</primary></indexterm>

<para>Synopsis:
</para>
<screen>\addtocounter{<replaceable>counter</replaceable>}{<replaceable>value</replaceable>&#160;</screen>
<para>Globally increment <replaceable>counter</replaceable> by the amount specified by the
<replaceable>value</replaceable> argument, which may be negative.
</para>
<para>In this example the section value appears as &#8216;<literal>VII</literal>&#8217;.
</para>
<screen>\setcounter{section}{5}
\addtocounter{section}{2}
Here it is in Roman: \Roman{section}.
</screen>

</sect1>
<sect1 label="13.6" id="_005crefstepcounter">
<title><literal>\refstepcounter</literal></title>

<indexterm role="fn"><primary>\refstepcounter</primary></indexterm>

<para>Synopsis:
</para>
<screen>\refstepcounter{<replaceable>counter</replaceable>}
</screen>
<para>Globally increments the value of <replaceable>counter</replaceable> by one, as does
<literal>\stepcounter</literal> (see <link linkend="_005cstepcounter">\stepcounter</link>).  The difference is that this
command resets the value of any counter numbered within it.  (For the
definition of &#8220;counters numbered within&#8221;, see <link linkend="_005cnewcounter">\newcounter</link>.)
</para>
<para>In addition, this command also defines the current <literal>\ref</literal> value
to be the result of <literal>\thecounter</literal>.
</para>
<para>While the counter value is set globally, the <literal>\ref</literal> value is set
locally, i.e., inside the current group.
</para>

</sect1>
<sect1 label="13.7" id="_005cstepcounter">
<title><literal>\stepcounter</literal></title>

<indexterm role="fn"><primary>\stepcounter</primary></indexterm>

<para>Synopsis:
</para>
<screen>\stepcounter{<replaceable>counter</replaceable>}
</screen>
<para>Globally adds one to <replaceable>counter</replaceable> and resets all counters numbered
within it.  (For the definition of &#8220;counters numbered within&#8221;,
see <link linkend="_005cnewcounter">\newcounter</link>.)
</para>
<para>This command differs from <literal>\refstepcounter</literal> in that this one does
not influence references&#160;&#8212; it does not define the current
<literal>\ref</literal> value to be the result of <literal>\thecounter</literal>
(see <link linkend="_005crefstepcounter">\refstepcounter</link>).
</para>

</sect1>
<sect1 label="13.8" id="_005cday-_0026-_005cmonth-_0026-_005cyear">
<title><literal>\day</literal> &amp; <literal>\month</literal> &amp; <literal>\year</literal></title>

<indexterm role="fn"><primary>\day</primary></indexterm>
<indexterm role="fn"><primary>\month</primary></indexterm>
<indexterm role="fn"><primary>\year</primary></indexterm>

<para>&latex; defines the counter <literal>\day</literal> for the day of the month
(nominally with value between 1 and 31), <literal>\month</literal> for the month of
the year (nominally with value between 1 and 12), and year <literal>\year</literal>.
When &tex; starts up, they are set from the current values on the
system.  The related command <literal>\today</literal> produces a string
representing the current day (see <link linkend="_005ctoday">\today</link>).
</para>
<para>They counters are not updated as the job progresses so in principle they
could be incorrect by the end.  In addition, &tex; does no sanity
check:
</para>
<screen>\day=-2 \month=13 \year=-4 \today
</screen>
<para>gives no error or warning and results in the output &#8216;<literal>-2, -4</literal>&#8217; (the
bogus month value produces no output).
</para>

</sect1>
</chapter>
<chapter label="14" id="Lengths">
<title>Lengths</title>

<indexterm role="cp"><primary>lengths, defining and using</primary></indexterm>

<para>A <firstterm>length</firstterm> is a measure of distance.  Many &latex; commands take a
length as an argument.
</para>
<para>This shows a box of the given length.
</para>
<screen>\newcommand{\blackbar}[1]{\rule{#1}{10pt}}   % make a bar 
\newcommand{\showhbox}[2]{\fboxsep=0pt\fbox{\hbox to #1{#2}}} % box it
XXX\showhbox{100pt}{\blackbar{100pt}}YYY
</screen>
<para>It produces a black bar 100&#160;points long between &#8216;<literal>XXX</literal>&#8217; and
&#8216;<literal>YYY</literal>&#8217;.
</para>
<para>Lengths come in two types.  A <firstterm>rigid length</firstterm> (what Plain &tex;
calls a <firstterm>dimen</firstterm>) such as <literal>10pt</literal> does not contain a <literal>plus</literal>
or <literal>minus</literal> component.  The above example shows a rigid length. A
<firstterm>rubber length</firstterm> (what Plain &tex; calls a <firstterm>skip</firstterm>) can contain
those components, as with <literal>1cm plus0.05cm minus0.01cm</literal>.  Here the
<literal>1cm</literal> is the <firstterm>natural length</firstterm> while the other two, the
<literal>plus</literal> and <literal>minus</literal> components, allow the length to stretch or
shrink.
</para>
<para>Shrinking is simpler: with <literal>1cm minus 0.05cm</literal>, the natural length
is 1cm but if smaller is needed then &tex; can shrink it down as
far as 0.95cm.  Beyond that, &tex; refuses to shrink any more.
Thus, below the first one works fine, producing a space of
98&#160;points between the two bars.
</para>
<screen>XXX\showhbox{300pt}{%
  \blackbar{101pt}\hspace{100pt minus 2pt}\blackbar{101pt}}YYY

XXX\showhbox{300pt}{%
  \blackbar{105pt}\hspace{100pt minus 1pt}\blackbar{105pt}}YYY
</screen>
<para>But the second one gets a warning like &#8216;<literal>Overfull \hbox (1.0pt too
wide) detected at line 17</literal>&#8217;.  In the output the first &#8216;<literal>Y</literal>&#8217; is
overwritten by the end of the black bar, because the box&#8217;s material is
wider than the 300pt allocated, as &tex; has refused to shrink
the total to less than 309&#160;points.
</para>
<para>Stretching is like shrinking except that if &tex; is asked to stretch
beyond the given amount, it won&#8217;t refuse.  Here the first line is fine,
producing a space of 110&#160;points between the bars.
</para>
<screen>XXX\showhbox{300pt}{%
  \blackbar{95pt}\hspace{100pt plus 10pt}\blackbar{95pt}}YYY

XXX\showhbox{300pt}{%
  \blackbar{95pt}\hspace{100pt plus 1pt}\blackbar{95pt}}YYY
</screen>
<para>In the second line &tex; needs a stretch of 10&#160;points and only
1&#160;point was specified.  In this situation, &tex; stretches the
space to the required length, but it complains with a warning like
&#8216;<literal>Underfull \hbox (badness 10000) detected at line 22</literal>&#8217;.  (We won&#8217;t
discuss badness; the point is that the system was not given as much
stretch as needed.)
</para>
<para>You can put both stretch and shrink in the same length, as in
<literal>1ex plus 0.05ex minus 0.02ex</literal>.
</para>
<para>If &tex; is setting two or more rubber lengths then it allocates the
stretch or shrink in proportion.
</para>
<screen>XXX\showhbox{300pt}{\blackbar{100pt}%  left
  \hspace{0pt plus 50pt}\blackbar{80pt}\hspace{0pt plus 10pt}%  middle
  \blackbar{100pt}}YYY  % right
</screen>
<para>The outside bars take up 100&#160;points, so the middle needs another
100.  In the middle the bar takes up 80&#160;points, so the two
<literal>\hspace</literal>&#8217;s must stretch 20&#160;points.  Because the two say
<literal>plus 50pt</literal> and <literal>plus 10pt</literal>, &tex; gets 5/6 of the
stretch from the first space and 1/6 from the second.
</para>
<para>The <literal>plus</literal> or <literal>minus</literal> component of a rubber length can contain
a <firstterm>fill</firstterm> component, as in <literal>1in plus2fill</literal>.  This gives the
length infinite stretchability or shrinkability so that &tex; could set
it to any distance.  Here the two figures will be equal-spaced across
the page.
</para>
<screen>\begin{minipage}{\linewidth}
  \hspace{0pt plus 1fill}\includegraphics{godel.png}%
  \hspace{0pt plus 1fill}\includegraphics{einstein.png}%
  \hspace{0pt plus 1fill}
\end{minipage}
</screen>
<para>&tex; actually has three infinite glue components <literal>fil</literal>,
<literal>fill</literal>, and <literal>filll</literal>.  The later ones are more infinite than
the earlier ones.  Ordinarily document authors only use the middle one
(see <link linkend="_005chfill">\hfill</link> and see <link linkend="_005cvfill">\vfill</link>).
</para>
<para>Multiplying a rubber length by a number turns it into a rigid length, so
that after <literal>\setlength{\ylength}{1in plus 0.2in}</literal> and
<literal>\setlength{\zlength}{3\ylength}</literal> then the value of
<literal>\zlength</literal> is <literal>3in</literal>.
</para>


<sect1 label="14.1" id="Units-of-length">
<title>Units of length</title>

<indexterm role="cp"><primary>units, of length</primary></indexterm>

<para>&tex; and &latex; know about these units both inside and outside of
math mode.
</para>
<variablelist><varlistentry><term><literal>pt</literal>
</term><listitem><indexterm role="fn"><primary>pt</primary></indexterm>
<indexterm role="cp"><primary>Point</primary></indexterm>
<anchor id="units-of-length-pt"/><para>Point 1/72.27 inch.  The conversion to metric units, to two decimal
places, is 1point = 2.85mm = 28.45cm. 
</para>
</listitem></varlistentry><varlistentry><term><literal>pc</literal>
</term><listitem><indexterm role="cp"><primary>pica</primary></indexterm>
<indexterm role="fn"><primary>pc</primary></indexterm>
<anchor id="units-of-length-pc"/><para>Pica, 12 pt
</para>
</listitem></varlistentry><varlistentry><term><literal>in</literal>
</term><listitem><indexterm role="fn"><primary>in</primary></indexterm>
<indexterm role="fn"><primary>inch</primary></indexterm>
<anchor id="units-of-length-in"/><para>Inch,  72.27 pt
</para>
</listitem></varlistentry><varlistentry><term><literal>bp</literal>
</term><listitem><indexterm role="fn"><primary>bp</primary></indexterm>
<indexterm role="cp"><primary>Big point</primary></indexterm>
<anchor id="units-of-length-bp"/><para>Big point, 1/72 inch.  This length is the definition of a point in
PostScript and many desktop publishing systems.
</para>
</listitem></varlistentry><varlistentry><term><literal>cm</literal>
</term><listitem><indexterm role="cp"><primary>Centimeter</primary></indexterm>
<indexterm role="fn"><primary>cm</primary></indexterm>
<anchor id="units-of-length-cm"/><para>Centimeter
</para>
</listitem></varlistentry><varlistentry><term><literal>mm</literal>
</term><listitem><indexterm role="cp"><primary>Millimeter</primary></indexterm>
<indexterm role="fn"><primary>mm</primary></indexterm>
<anchor id="units-of-length-mm"/><para>Millimeter
</para>
</listitem></varlistentry><varlistentry><term><literal>dd</literal>
</term><listitem><indexterm role="cp"><primary>Didot point</primary></indexterm>
<indexterm role="fn"><primary>dd</primary></indexterm>
<anchor id="units-of-length-dd"/><para>Didot point, 1.07 pt
</para>
</listitem></varlistentry><varlistentry><term><literal>cc</literal>
</term><listitem><indexterm role="cp"><primary>Cicero</primary></indexterm>
<indexterm role="fn"><primary>cc</primary></indexterm>
<anchor id="units-of-length-cc"/><para>Cicero, 12 dd
</para>
</listitem></varlistentry><varlistentry><term><literal>sp</literal>
</term><listitem><indexterm role="cp"><primary>Scaled point</primary></indexterm>
<indexterm role="fn"><primary>sp</primary></indexterm>
<anchor id="units-of-length-sp"/><para>Scaled point, 1/65536 pt
</para>
</listitem></varlistentry></variablelist>
<indexterm role="cp"><primary>ex</primary></indexterm>
<indexterm role="cp"><primary>x-height</primary></indexterm>
<indexterm role="fn"><primary>ex</primary></indexterm>
<indexterm role="cp"><primary>m-width</primary></indexterm>
<indexterm role="cp"><primary>em</primary></indexterm>
<indexterm role="fn"><primary>em</primary></indexterm>
<anchor id="Lengths_002fem"/><anchor id="Lengths_002fen"/><anchor id="Lengths_002fex"/><anchor id="units-of-length-em"/><anchor id="units-of-length-en"/><anchor id="units-of-length-ex"/><para>Two other lengths that are often used are values set by the designer of
the font.  The x-height of the current font <firstterm>ex</firstterm>, traditionally the
height of the lowercase letter x, is often used for vertical
lengths. Similarly <firstterm>em</firstterm>, traditionally the width of the capital
letter M, is often used for horizontal lengths (there is also
<literal>\enspace</literal>, which is <literal>0.5em</literal>).  Use of these can help make a
definition work better across font changes.  For example, a definition
of the vertical space between list items given as
<literal>\setlength{\itemsep}{1ex plus 0.05ex minus 0.01ex}</literal> is more
likely to still be reasonable if the font is changed than a definition
given in points.
</para>
<indexterm role="cp"><primary>mu, math unit</primary></indexterm>
<indexterm role="fn"><primary>mu</primary></indexterm>
<para>In math mode, many definitions are expressed in terms of the math unit
<firstterm>mu</firstterm> given by 1 em = 18 mu, where the em is taken from the current
math symbols family.  See <link linkend="Spacing-in-math-mode">Spacing in math mode</link>.
</para>

</sect1>
<sect1 label="14.2" id="_005csetlength">
<title><literal>\setlength</literal></title>

<indexterm role="fn"><primary>\setlength</primary></indexterm>
<indexterm role="cp"><primary>lengths, setting</primary></indexterm>

<para>Synopsis:
</para>
<screen>\setlength{<replaceable>len</replaceable>}{<replaceable>amount</replaceable>}
</screen>
<para>Set the length <replaceable>len</replaceable> to <replaceable>amount</replaceable>.  The length name <replaceable>len</replaceable>
must begin with a backslash, <literal>\</literal>.  The <literal>amount</literal> can be a
rubber length (see <link linkend="Lengths">Lengths</link>).  It can be positive, negative or zero,
and can be in any units that &latex; understands (see <link linkend="Units-of-length">Units of
length</link>).
</para>
<para>Below, with &latex;&#8217;s defaults the first paragraph will be indented
while the second will not.
</para>
<screen>I told the doctor I broke my leg in two places.

\setlength{\parindent}{0em}
He said stop going to those places.
</screen>
<para>If there is no such length <replaceable>len</replaceable> then you get something like
&#8216;<literal>Undefined control sequence. &lt;argument&gt; \praindent</literal>&#8217;.
</para>

</sect1>
<sect1 label="14.3" id="_005caddtolength">
<title><literal>\addtolength</literal></title>

<indexterm role="fn"><primary>\addtolength</primary></indexterm>
<indexterm role="cp"><primary>lengths, adding to</primary></indexterm>

<para>Synopsis:
</para>
<screen>\addtolength{<replaceable>len</replaceable>}{<replaceable>amount</replaceable>}
</screen>
<para>Increment the length <replaceable>len</replaceable> by <replaceable>amount</replaceable>.  The length name
<replaceable>len</replaceable> begins with a backslash, <literal>\</literal>.  The <literal>amount</literal> is a
rubber length (see <link linkend="Lengths">Lengths</link>).  It can be positive, negative or zero,
and can be in any units that &latex; understands (see <link linkend="Units-of-length">Units of
length</link>).
</para>
<para>Below, if <literal>\parskip</literal> starts with the value <literal>0pt plus 1pt</literal> 
</para>
<screen>\addtolength{\parskip}{1pt}
Doctor: how is the boy who swallowed the silver dollar?

Nurse: no change.
</screen>
<para>then it has the value <literal>1pt plus 1pt</literal> for the second paragraph.
</para>
<para>If there is no such length <replaceable>len</replaceable> then you get something like
&#8216;<literal>Undefined control sequence. &lt;argument&gt; \praindent</literal>&#8217;.  If you leave
off the backslash at the start of <replaceable>len</replaceable>, as in
<literal>\addtolength{parindent}{1pt}</literal>, then you get something like
&#8216;<literal>You can't use `the letter p' after \advance</literal>&#8217;.
</para>

</sect1>
<sect1 label="14.4" id="_005csettodepth">
<title><literal>\settodepth</literal></title>

<indexterm role="fn"><primary>\settodepth</primary></indexterm>

<para>Synopsis:
</para>
<screen>\settodepth{<replaceable>len</replaceable>}{<replaceable>text</replaceable>}
</screen>
<para>Set the length <replaceable>len</replaceable> to the depth of box that &latex; gets on
typesetting the <replaceable>text</replaceable> argument.  The length name <replaceable>len</replaceable> must
begin with a backslash, <literal>\</literal>.
</para>
<para>This will show how low the character descenders go.
</para>
<screen>\newlength{\alphabetdepth}
\settodepth{\alphabetdepth}{abcdefghijklmnopqrstuvwxyz}
\the\alphabetdepth
</screen>
<para>If there is no such length <replaceable>len</replaceable> then you get something like
&#8216;<literal>Undefined control sequence. &lt;argument&gt; \alphabetdepth</literal>&#8217;.  If you
leave the backslash out of <replaceable>len</replaceable>, as in
<literal>\settodepth{alphabetdepth}{...}</literal> then you get something like
&#8216;<literal>Missing number, treated as zero. &lt;to be read again&gt; \setbox</literal>&#8217;.
</para>

</sect1>
<sect1 label="14.5" id="_005csettoheight">
<title><literal>\settoheight</literal></title>

<indexterm role="fn"><primary>\settoheight</primary></indexterm>

<para>Synopsis:
</para>
<screen>\settoheight{<replaceable>len</replaceable>}{text}
</screen>
<para>Sets the length <replaceable>len</replaceable> to the height of box that &latex; gets on
typesetting the <literal>text</literal> argument.  The length name <replaceable>len</replaceable> must
begin with a backslash, <literal>\</literal>.
</para>
<para>This will show how high the characters go.
</para>
<screen>\newlength{\alphabetheight}
\settoheight{\alphabetheight}{abcdefghijklmnopqrstuvwxyz}
\the\alphabetheight
</screen>
<para>If there is no such length <replaceable>len</replaceable> then you get something like
&#8216;<literal>Undefined control sequence. &lt;argument&gt; \alphabetheight</literal>&#8217;.  If you
leave the backslash out of <replaceable>len</replaceable>, as in
<literal>\settoheight{alphabetheight}{...}</literal> then you get something like
&#8216;<literal>Missing number, treated as zero. &lt;to be read again&gt; \setbox</literal>&#8217;.
</para>

</sect1>
<sect1 label="14.6" id="_005csettowidth">
<title><literal>\settowidth</literal></title>

<indexterm role="fn"><primary>\settowidth</primary></indexterm>

<para>Synopsis:
</para>
<screen>\settowidth{<replaceable>len</replaceable>}{<replaceable>text</replaceable>}
</screen>
<para>Set the length <replaceable>len</replaceable> to the width of the box that &latex; gets on
typesetting the <replaceable>text</replaceable> argument.  The length name <replaceable>len</replaceable> must
begin with a backslash, <literal>\</literal>.
</para>
<para>This measures the width of the lowercase ASCII alphabet.
</para>
<screen>\newlength{\alphabetwidth}
\settowidth{\alphabetwidth}{abcdefghijklmnopqrstuvwxyz}
\the\alphabetwidth
</screen>
<para>If there is no such length <replaceable>len</replaceable> then you get something like
&#8216;<literal>Undefined control sequence. &lt;argument&gt; \alphabetwidth</literal>&#8217;.  If you
leave the backslash out of <replaceable>len</replaceable>, as in
<literal>\settoheight{alphabetwidth}{...}</literal> then you get something like
&#8216;<literal>Missing number, treated as zero. &lt;to be read again&gt; \setbox</literal>&#8217;.
</para>

<!-- @node Predefined lengths -->
<!-- @section Predefined lengths -->

<!-- @cindex lengths, predefined -->
<!-- @cindex predefined lengths -->

<!-- @code{\width} -->
<!-- @findex \width -->

<!-- @code{\height} -->
<!-- @findex \height -->

<!-- @code{\depth} -->
<!-- @findex \depth -->

<!-- @code{\totalheight} -->
<!-- @findex \totalheight -->

<!-- These length parameters can be used in the arguments of the box-making -->
<!-- commands (@pxref{Boxes}).  They specify the natural width, etc., of the -->
<!-- text in the box. @code{\totalheight} equals -->
<!-- @math{@code{@backslashchar{}height} + @code{@backslashchar{}depth}}. To -->
<!-- make a box with the text stretched to double the natural size, e.g., say -->

<!-- @example -->
<!-- \makebox[2\width]@{Get a stretcher@} -->
<!-- @end example -->


</sect1>
</chapter>
<chapter label="15" id="Making-paragraphs">
<title>Making paragraphs</title>

<indexterm role="cp"><primary>making paragraphs</primary></indexterm>
<indexterm role="cp"><primary>paragraphs</primary></indexterm>

<para>Once &latex; has all of a paragraph&#8217;s contents it divides it into
lines, in a way that is optimized over the entire paragraph (see <link linkend="Line-breaking">Line
breaking</link>).  To end the current paragraph, put an empty line.
</para>
<screen>It is a truth universally acknowledged, that a single man in possession
of a good fortune, must be in want of a wife.

However little known the feelings or views of such a man may be on his
first entering a neighbourhood, this truth is so well fixed in the minds
of the surrounding families, that he is considered the rightful property
of some one or other of their daughters.

``My dear Mr. Bennet,'' said his lady to him one day,
``have you heard that Netherfield Park is let at last?''
</screen>
<para>The separator lines must be empty, including not containing a comment
character, <literal>%</literal>.
</para>
<para>There are places where a new paragraph is not permitted.  Don&#8217;t put a
blank line in math mode (see <link linkend="Modes">Modes</link>); here the line before the
<literal>\end{equation}</literal>
</para>
<screen>\begin{equation}
  2^{|S|} &gt; |S|

\end{equation}
</screen>
<para>will get you the error &#8216;<literal>Missing $ inserted</literal>&#8217;.  Similarly, the blank
line in this <literal>section</literal> argument
</para>
<screen>\section{aaa

bbb}
</screen>
<para>gets &#8216;<literal>Runaway argument? {aaa ! Paragraph ended before \@sect was
complete</literal>&#8217;.
</para>



<sect1 label="15.1" id="_005cpar">
<title><literal>\par</literal></title>

<indexterm role="fn"><primary>\par</primary></indexterm>
<indexterm role="cp"><primary>paragraph, ending</primary></indexterm>

<para>Synopsis (note that while reading the input &tex;, converts two
consecutive newlines to a <literal>\par</literal>):
</para>
<screen>\par
</screen>
<para>End the current paragraph.  The usual way to separate paragraphs is with
a blank line but the <literal>\par</literal> command is entirely equivalent.  This
command is robust (see <link linkend="_005cprotect">\protect</link>).
</para>
<para>This example uses <literal>\par</literal> rather than a blank line simply for
readability.
</para>
<screen>\newcommand{\syllabusLegalese}{%
  \whatCheatingIs\par\whatHappensWhenICatchYou}
</screen>
<para>The <literal>\par</literal> command does nothing in LR mode or a vertical mode but
it terminates paragraph mode, bringing &latex; to vertical mode
(see <link linkend="Modes">Modes</link>).
</para>
<para>You cannot use the <literal>\par</literal> command in math mode or in the argument
of many commands, such as the <literal>\section</literal> command (see <link linkend="Making-paragraphs">Making
paragraphs</link> and see <link linkend="_005cnewcommand-_0026-_005crenewcommand">\newcommand &amp; \renewcommand</link>).
</para>
<para>The <literal>\par</literal> command differs from the <literal>\paragraph</literal> command in
that the latter is, like <literal>\section</literal> or <literal>\subsection</literal>, a
sectioning unit used by the standard &latex; documents.
</para>
<para>The <literal>\par</literal> command differs from <literal>\newline</literal> and the line break
double backslash, <literal>\\</literal>, in that \par ends the paragraph not just
the line.  It also triggers the addition of the between-paragraph
vertical space <literal>\parskip</literal> (see <link linkend="_005cparindent-_0026-_005cparskip">\parindent &amp; \parskip</link>).
</para>
<para>The output from this example
</para>
<screen>xyz

\setlength{\parindent}{3in}
\setlength{\parskip}{5in}
\noindent test\indent test1\par test2 
</screen>
<para>is: after &#8216;<literal>xyz</literal>&#8217; there is a vertical skip of 5&#160;inches and then
&#8216;<literal>test</literal>&#8217; appears, aligned with the left margin.  On the same line,
there is an empty horizontal space of 3&#160;inches and then
&#8216;<literal>test1</literal>&#8217; appears.  Finally. there is a vertical space of
5&#160;inches, followed by a fresh paragraph with a paragraph indent of
3&#160;inches, and then &latex; puts the text &#8216;<literal>test2</literal>&#8217;.
</para>

</sect1>
<sect1 label="15.2" id="_005cindent-_0026-_005cnoindent">
<title><literal>\indent</literal> &amp; <literal>\noindent</literal></title>

<indexterm role="fn"><primary>\indent</primary></indexterm>
<indexterm role="fn"><primary>\noindent</primary></indexterm>
<indexterm role="fn"><primary>\parindent</primary></indexterm>
<indexterm role="cp"><primary>indent, forcing</primary></indexterm>

<para>Synopsis:
</para>
<screen>\indent
</screen>
<para>or
</para>
<screen>\noindent
</screen>
<para>Go into horizontal mode (see <link linkend="Modes">Modes</link>).  The <literal>\indent</literal> command
first outputs an empty box whose width is <literal>\parindent</literal>.  These
commands are robust (see <link linkend="_005cprotect">\protect</link>).
</para>
<para>Ordinarily you create a new paragraph by putting in a blank line.
See <link linkend="_005cpar">\par</link> for the difference between this command and <literal>\par</literal>.  To
start a paragraph without an indent, or to continue an interrupted
paragraph, use <literal>\noindent</literal>.
</para>
<para>In the middle of a paragraph the <literal>\noindent</literal> command has no effect,
because &latex; is already in horizontal mode there.  The
<literal>\indent</literal> command&#8217;s only effect is to output a space.
</para>
<para>This example starts a fresh paragraph.
</para>
<screen>... end of the prior paragraph.

\noindent This paragraph is not indented.
</screen>
<para>and this continues an interrupted paragraph.
</para>
<screen>The data 

\begin{center}
  \begin{tabular}{rl} ... \end{tabular}
\end{center}

\noindent shows this clearly.
</screen>
<para>To omit indentation in the entire document put
<literal>\setlength{\parindent}{0pt}</literal> in the preamble.  If you do that,
you may want to also set the length of spaces between paragraphs,
<literal>\parskip</literal> (see <link linkend="_005cparindent-_0026-_005cparskip">\parindent &amp; \parskip</link>).
</para>
<indexterm role="cp"><primary>package, <literal>indentfirst</literal></primary></indexterm>
<indexterm role="cp"><primary><literal>indentfirst</literal> package</primary></indexterm>

<para>Default &latex; styles have the first paragraph after a section that is
not indented, as is traditional typesetting in English.  To change that,
look on CTAN for the package <literal>indentfirst</literal>.
</para>

</sect1>
<sect1 label="15.3" id="_005cparindent-_0026-_005cparskip">
<title><literal>\parindent</literal> &amp; <literal>\parskip</literal></title>

<indexterm role="fn"><primary>\parindent</primary></indexterm>
<indexterm role="fn"><primary>\parskip</primary></indexterm>
<indexterm role="cp"><primary>paragraph indentation</primary></indexterm>
<indexterm role="cp"><primary>vertical space before paragraphs</primary></indexterm>

<para>Synopsis:
</para>
<screen>\setlength{\parskip}{<replaceable>horizontal len</replaceable>}
\setlength{\parinden}{<replaceable>vertical len</replaceable>}
</screen>
<para>Both are a rubber lengths (see <link linkend="Lengths">Lengths</link>).  They give the indentation
of ordinary paragraphs, not paragraphs inside minipages
(see <link linkend="minipage">minipage</link>), and the vertical space between paragraphs.
</para>
<para>This, put in the preamble,
</para>
<screen>\setlength{\parindent}{0em}
\setlength{\parskip}{1ex}
</screen>
<para>arranges that the document will have paragraphs that are not indented,
but instead are vertically separated by about the height of a lowercase
&#8216;<literal>x</literal>&#8217;.
</para>
<para>In standard &latex; documents, the default value for <literal>\parindent</literal>
in one-column documents is <literal>15pt</literal> when the default text size is
<literal>10pt</literal> , <literal>17pt</literal> for <literal>11pt</literal>, and <literal>1.5em</literal> for
<literal>12pt</literal>.  In two-column documents it is <literal>1em</literal>.  The default
value for <literal>\parskip</literal> in &latex;&#8217;s standard document styles is
<literal>0pt plus1pt</literal>.
</para>

</sect1>
<sect1 label="15.4" id="Marginal-notes">
<title>Marginal notes</title>

<indexterm role="cp"><primary>marginal notes</primary></indexterm>
<indexterm role="cp"><primary>notes in the margin</primary></indexterm>
<indexterm role="cp"><primary>remarks in the margin</primary></indexterm>
<indexterm role="fn"><primary>\marginpar</primary></indexterm>

<para>Synopsis, one of:
</para>
<screen>\marginpar{<replaceable>right</replaceable>}
\marginpar[<replaceable>left</replaceable>]{<replaceable>right</replaceable>}
</screen>
<para>Create a note in the margin.  The first line of the note will have the
same baseline as the line in the text where the <literal>\marginpar</literal>
occurs.
</para>
<para>The margin that &latex; uses for the note depends on the current layout
(see <link linkend="Document-class-options">Document class options</link>) and also on <literal>\reversemarginpar</literal>
(see below).  If you are using one-sided layout (document option
<literal>oneside</literal>) then it goes in the right margin.  If you are using
two-sided layout (document option <literal>twoside</literal>) then it goes in the
outside margin.  If you are in two-column layout (document option
<literal>twocolumn</literal>) then it goes in the nearest margin.
</para>
<indexterm role="fn"><primary>\reversemarginpar</primary></indexterm>
<indexterm role="fn"><primary>\normalmarginpar</primary></indexterm>
<para>If you declare <literal>\reversemarginpar</literal> then &latex; will place
subsequent marginal notes in the opposite margin to that given in the
prior paragraph.  Revert that to the default position with
<literal>\normalmarginpar</literal>.
</para>
<para>When you specify the optional argument <replaceable>left</replaceable> then it is used for a
note in the left margin, while the mandatory argument <replaceable>right</replaceable> is
used for a note in the the right margin.
</para>
<para>Normally, a note&#8217;s first word will not be hyphenated.  You can enable
hyphenation there by beginning <replaceable>left</replaceable> or <replaceable>right</replaceable> with
<literal>\hspace{0pt}</literal>.
</para>
<para>These parameters affect the formatting of the note:
</para>
<variablelist><varlistentry><term><indexterm role="fn"><primary>\marginparpush</primary></indexterm><literal>\marginparpush</literal>
</term><listitem><anchor id="marginal-notes-marginparpush"/><para>Minimum vertical space between notes; default &#8216;<literal>7pt</literal>&#8217; for
&#8216;<literal>12pt</literal>&#8217; documents, &#8216;<literal>5pt</literal>&#8217; else.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\marginparsep</primary></indexterm><literal>\marginparsep</literal>
</term><listitem><anchor id="marginal-notes-marginparsep"/><para>Horizontal space between the main text and the note; default
&#8216;<literal>11pt</literal>&#8217; for &#8216;<literal>10pt</literal>&#8217; documents, &#8216;<literal>10pt</literal>&#8217; else.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\marginparwidth</primary></indexterm><literal>\marginparwidth</literal>
</term><listitem><anchor id="marginal-notes-marginparwidth"/><para>Width of the note itself; default for a one-sided &#8216;<literal>10pt</literal>&#8217; document
is &#8216;<literal>90pt</literal>&#8217;, &#8216;<literal>83pt</literal>&#8217; for &#8216;<literal>11pt</literal>&#8217;, and &#8216;<literal>68pt</literal>&#8217; for
&#8216;<literal>12pt</literal>&#8217;; &#8216;<literal>17pt</literal>&#8217; more in each case for a two-sided document.
In two column mode, the default is &#8216;<literal>48pt</literal>&#8217;.
</para>
</listitem></varlistentry></variablelist>
<para>The standard &latex; routine for marginal notes does not prevent
notes from falling off the bottom of the page.
<!--  @TeX{} FAQ entry on this topic (xx when there): -->
<!-- @url{http://www.tex.ac.uk/cgi-bin/texfaq2html?label=marginparside}. -->
<!-- (+marginfix) -->
</para>

</sect1>
</chapter>
<chapter label="16" id="Math-formulas">
<title>Math formulas</title>

<indexterm role="cp"><primary>math formulas</primary></indexterm>
<indexterm role="cp"><primary>formulas, math</primary></indexterm>
<indexterm role="cp"><primary>math mode, entering</primary></indexterm>
<indexterm role="fn"><primary>environment, <literal>math</literal></primary></indexterm>
<indexterm role="fn"><primary><literal>math</literal> environment</primary></indexterm>
        
<indexterm role="fn"><primary>environment, <literal>displaymath</literal></primary></indexterm>
<indexterm role="fn"><primary><literal>displaymath</literal> environment</primary></indexterm>
        
<indexterm role="fn"><primary>environment, <literal>equation</literal></primary></indexterm>
<indexterm role="fn"><primary><literal>equation</literal> environment</primary></indexterm>
        

<para>Produce mathematical text by putting &latex; into math mode or display
math mode (see <link linkend="Modes">Modes</link>).  This example shows both.
</para>
<screen>The wave equation for \( u \) is
\begin{displaymath}
  \frac{\partial^2u}{\partial t^2} = c^2\nabla^2u
\end{displaymath}
where \( \nabla^2 \) is the spatial Laplacian and \( c \) is constant.
</screen>
<para>Math mode is for inline mathematics.  In the above example it is invoked
by the starting <literal>\(</literal> and finished by the matching ending <literal>\)</literal>.
Display math mode is for displayed equations and here is invoked by the
<literal>displaymath</literal> environment.  Note that any mathematical text
whatever, including mathematical text consisting of just one character,
is handled in math mode.
</para>
<para>When in math mode or display math mode, &latex; handles many aspects of
your input text differently than in other text modes.  For example,
</para>
<screen>contrast x+y with \( x+y \)
</screen>
<para>in math mode the letters are in italics and the spacing around the plus
sign is different.
</para>
<para>There are three ways to make inline formulas, to put &latex; in math
mode.
</para>
<screen>\( <replaceable>mathematical material</replaceable> \)
$ <replaceable>mathematical material</replaceable> $
\begin{math} <replaceable>mathematical material</replaceable> \end{math}
</screen>
<para>The first form is preferred and the second is quite common, but the
third form is rarely used.  You can sometimes use one and sometimes
another, as in <literal>\(x\) and $y$</literal>.  You can use these in paragraph
mode or in LR mode (see <link linkend="Modes">Modes</link>).
</para>
<para>To make displayed formulas, put &latex; into display math mode with
either:
</para>
<screen>\begin{displaymath}
  <replaceable>mathematical material</replaceable>
\end{displaymath}
</screen>
<para>or
</para>
<screen>\begin{equation}
  <replaceable>mathematical material</replaceable>
\end{equation}
</screen>
<para>(see <link linkend="displaymath">displaymath</link>, see <link linkend="equation">equation</link>).  The only difference is that
with the <literal>equation</literal> environment, &latex; puts a formula number
alongside the formula.  The construct <literal>\[ <replaceable>math</replaceable> \]</literal> is
equivalent to <literal>\begin{displaymath} <replaceable>math</replaceable>
\end{displaymath}</literal>.  These environments can only be used in paragraph
mode (see <link linkend="Modes">Modes</link>).
</para>
<indexterm role="fn"><primary>\displaystyle</primary></indexterm>
<para>The two mathematics modes are similar, but there are some differences.
One involves the placement of subscripts and superscripts; in display
math mode they are further apart and in inline math mode they are closer
together.
</para>
<para>Sometimes you want the display math typographical treatment to happen in
the inline math mode.  For this, the <literal>\displaystyle</literal> declaration
forces the size and style of the formula to be that of
<literal>displaymath</literal>.  Thus <literal>\(\displaystyle \sum_{n=0}^\infty
x_n\)</literal> will have the limits above and below the summation sign, not next
to it.  Another example is that
</para>
<screen>\begin{tabular}{r|cc}
  \textsc{Name}  &amp;\textsc{Series}  &amp;\textsc{Sum}  \\  \hline
  Arithmetic     &amp;\( a+(a+b)+(a+2b)+\cdots+(a+(n-1)b) \)
                   &amp;\( na+(n-1)n\cdot\frac{b}{2}\)  \\
  Geometric      &amp;\( a+ab+ab^2+\cdots+ab^{n-1} \)
                   &amp;\(\displaystyle a\cdot\frac{1-b^n}{1-b}\)  \\
\end{tabular}
</screen>
<para>because it has no <literal>\displaystyle</literal>, the &#8216;<literal>Arithmetic</literal>&#8217; line&#8217;s
fraction will be scrunched.  But, because of its <literal>\displaystyle</literal>,
the &#8216;<literal>Geometric</literal>&#8217; line&#8217;s fraction will be easy to read, with
characters the same size as in the rest of the line.
</para>
<indexterm role="cp"><primary>package, <literal>amsmath</literal></primary></indexterm>
<indexterm role="cp"><primary><literal>amsmath</literal> package</primary></indexterm>

<indexterm role="cp"><primary>package, <literal>amsfonts</literal></primary></indexterm>
<indexterm role="cp"><primary><literal>amsfonts</literal> package</primary></indexterm>

<indexterm role="cp"><primary>package, <literal>mathtools</literal></primary></indexterm>
<indexterm role="cp"><primary><literal>mathtools</literal> package</primary></indexterm>

<para>The American Mathematical Society has made freely available a set of
packages that greatly expand your options for writing mathematics,
<filename>amsmath</filename> and <filename>amssymb</filename> (also be aware of the <filename>mathtools</filename>
package that is an extension to, and loads, <filename>amsmath</filename>).  New
documents that will have mathematical text should use these packages.
Descriptions of these packages is outside the scope of this document;
see their documentation on CTAN.
</para>


<sect1 label="16.1" id="Subscripts-_0026-superscripts">
<title>Subscripts &amp; superscripts</title>

<indexterm role="cp"><primary>superscript</primary></indexterm>
<indexterm role="cp"><primary>subscript</primary></indexterm>
<indexterm role="cp"><primary>exponent</primary></indexterm>
<indexterm role="fn"><primary>_</primary></indexterm>
<indexterm role="fn"><primary>^</primary></indexterm>

<para>Synopsis (in math mode or display math mode), one of:
</para>
<screen><replaceable>base</replaceable>^<replaceable>exp</replaceable>
<replaceable>base</replaceable>^{<replaceable>exp</replaceable>}
</screen>
<para>or, one of:
</para>
<screen><replaceable>base</replaceable>_<replaceable>exp</replaceable>
<replaceable>base</replaceable>_{<replaceable>exp</replaceable>}
</screen>
<para>Make <replaceable>exp</replaceable> appear as a superscript of <replaceable>base</replaceable> (with the caret
character,&#160;<literal>^</literal>) or a subscript (with
underscore,&#160;<literal>_</literal>).
</para>
<para>In this example the <literal>0</literal>&#8217;s and <literal>1</literal>&#8217;s are subscripts while the
<literal>2</literal>&#8217;s are superscripts.
</para>
<screen>\( (x_0+x_1)^2 \leq (x_0)^2+(x_1)^2 \)
</screen>
<para>To have the subscript or superscript contain more than one character,
surround the expression with curly braces, as in <literal>e^{-2x}</literal>.
This example&#8217;s fourth line shows curly braces used to group an expression
for the exponent.
</para>
<screen>\begin{displaymath}
  (3^3)^3=27^3=19\,683
  \qquad
  3^{(3^3)}=3^{27}=7\,625\,597\,484\,987
\end{displaymath}
</screen>
<para>&latex; knows how to handle a superscript on a superscript, or a
subscript on a subscript, or supers on subs, or subs on supers.  So,
expressions such as <literal>e^{x^2}</literal> and <literal>x_{i_0}</literal> give correct
output.  Note the use in those expressions of curly braces to give the
<replaceable>base</replaceable> a determined <replaceable>exp</replaceable>.  If you enter <literal>\(3^3^3\)</literal> then
you get &#8216;<literal>Double superscript</literal>&#8217;.
</para>
<para>&latex; does the right thing when something has both a subscript and a
superscript.  In this example the integral has both.  They come out in
the correct place without any author intervention.
</para>
<screen>\begin{displaymath}
  \int_{x=a}^b f'(x)\,dx = f(b)-f(a)
\end{displaymath}
</screen>
<para>Note the parentheses around <literal>x=a</literal> to make the entire expression a
subscript.
</para>
<para>To put a superscript or subscript before a symbol, use a construct like
<literal>{}_t K^2</literal>.  The empty curly braces <literal>{}</literal> give the
subscript something to attach to and keeps it from accidentally
attaching to a prior symbols.
</para>
<para>Using the subscript or superscript command outside of math mode or
display math mode, as in <literal>the expression x^2</literal>, will get you
the error &#8216;<literal>Missing $ inserted</literal>&#8217;.
</para>
<indexterm role="cp"><primary>package, <literal>mhchem</literal></primary></indexterm>
<indexterm role="cp"><primary><literal>mhchem</literal> package</primary></indexterm>

<para>A common reason to want subscripts outside of a mathematics mode is to
typeset chemical formulas.  There are packages for that such as
<filename>mhchem</filename>; see CTAN.
</para>

</sect1>
<sect1 label="16.2" id="Math-symbols">
<title>Math symbols</title>

<indexterm role="cp"><primary>math symbols</primary></indexterm>
<indexterm role="cp"><primary>symbols, math</primary></indexterm>
<indexterm role="cp"><primary>greek letters</primary></indexterm>

<indexterm role="cp"><primary>package, <literal>symbols</literal></primary></indexterm>
<indexterm role="cp"><primary><literal>symbols</literal> package</primary></indexterm>


<para>&latex; provides almost any mathematical or technical symbol that
anyone uses.  For example, if you include <literal>$\pi$</literal> in your source,
you will get the pi symbol &#x03C0;.  See the <filename>Comprehensive
&latex; Symbol List</filename> at
<ulink url="https://ctan.org/tex-archive/info/symbols/comprehensive/">https://ctan.org/tex-archive/info/symbols/comprehensive/</ulink>.
</para>
<para>Here is a list of commonly-used symbols.  It is by no means exhaustive.
Each symbol is described with a short phrase, and its symbol class,
which determines the spacing around it, is given in parenthesis.  Unless
said otherwise, the commands for these symbols can be used only in math
mode.  To redefine a command so that it can be used whatever the current
mode, see <link linkend="_005censuremath">\ensuremath</link>.
</para>
<!-- xx Add Negation: @code{} for negations of relevant symbols -->
<!-- Useful: http://www.w3.org/TR/WD-math-970515/section6.html -->

<variablelist><varlistentry><term><indexterm role="fn"><primary>\|</primary></indexterm><literal>\|</literal>
</term><listitem><para>&#x2225; Parallel (relation). Synonym:&#160;<literal>\parallel</literal>.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\aleph</primary></indexterm><literal>\aleph</literal>
</term><listitem><para>&#x2135; Aleph, transfinite cardinal (ordinary).
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\alpha</primary></indexterm><literal>\alpha</literal>
</term><listitem><para>&#x03B1; Lowercase Greek letter alpha (ordinary).
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\amalg</primary></indexterm><literal>\amalg</literal>
</term><listitem><para>&#x2A3F; Disjoint union (binary)
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\angle</primary></indexterm><literal>\angle</literal>
</term><listitem><para>&#x2220; Geometric angle (ordinary). Similar: less-than
sign&#160;<literal>&lt;</literal> and angle bracket&#160;<literal>\langle</literal>.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\approx</primary></indexterm><literal>\approx</literal>
</term><listitem><para>&#x2248; Almost equal to (relation).
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\ast</primary></indexterm><literal>\ast</literal>
</term><listitem><para>&#x2217; Asterisk operator, convolution, six-pointed
(binary). Synonym:&#160;<literal>*</literal>, which is often a superscript or
subscript, as in the Kleene star. Similar:&#160;<literal>\star</literal>, which is
five-pointed, and is sometimes used as a general binary operation, and
sometimes reserved for cross-correlation.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\asymp</primary></indexterm><literal>\asymp</literal>
</term><listitem><para>&#x224D; Asymptotically equivalent (relation).
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\backslash</primary></indexterm><literal>\backslash</literal>
</term><listitem><para>\ Backslash (ordinary).  Similar: set minus&#160;<literal>\setminus</literal>, and
<literal>\textbackslash</literal> for backslash outside of math mode.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\beta</primary></indexterm><literal>\beta</literal>
</term><listitem><para>&#x03B2; Lowercase Greek letter beta (ordinary).
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\bigcap</primary></indexterm><literal>\bigcap</literal>
</term><listitem><para>&#x22C2; Variable-sized, or n-ary, intersection (operator). Similar:
binary intersection&#160;<literal>\cap</literal>.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\bigcirc</primary></indexterm><literal>\bigcirc</literal>
</term><listitem><para>&#x26AA; Circle, larger (binary).  Similar: function
composition&#160;<literal>\circ</literal>.
<!-- bb Best unicode symbol for this? -->
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\bigcup</primary></indexterm><literal>\bigcup</literal>
</term><listitem><para>&#x22C3; Variable-sized, or n-ary, union (operator). Similar: binary
union&#160;<literal>\cup</literal>.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\bigodot</primary></indexterm><literal>\bigodot</literal>
</term><listitem><para>&#x2A00; Variable-sized, or n-ary, circled dot operator (operator).
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\bigoplus</primary></indexterm><literal>\bigoplus</literal>
</term><listitem><para>&#x2A01; Variable-sized, or n-ary, circled plus operator (operator).
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\bigotimes</primary></indexterm><literal>\bigotimes</literal>
</term><listitem><para>&#x2A02; Variable-sized, or n-ary, circled times operator (operator).
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\bigtriangledown</primary></indexterm><literal>\bigtriangledown</literal>
</term><listitem><para>&#x25BD; Variable-sized, or n-ary, open triangle pointing down
(operator).
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\bigtriangleup</primary></indexterm><literal>\bigtriangleup</literal>
</term><listitem><para>&#x25B3; Variable-sized, or n-ary, open triangle pointing up (operator).
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\bigsqcup</primary></indexterm><literal>\bigsqcup</literal>
</term><listitem><para>&#x2A06; Variable-sized, or n-ary, square union (operator).
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\biguplus</primary></indexterm><literal>\biguplus</literal>
</term><listitem><para>&#x2A04; Variable-sized, or n-ary, union operator with a plus
(operator).  (Note that the name has only one p.)
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\bigvee</primary></indexterm><literal>\bigvee</literal>
</term><listitem><para>&#x22C1; Variable-sized, or n-ary, logical-and (operator).
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\bigwedge</primary></indexterm><literal>\bigwedge</literal>
</term><listitem><para>&#x22C0; Variable-sized, or n-ary, logical-or (operator).
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\bot</primary></indexterm><literal>\bot</literal>
</term><listitem><para>&#x22A5; Up tack, bottom, least element of a partially ordered
set, or a contradiction (ordinary).  See also&#160;<literal>\top</literal>.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\bowtie</primary></indexterm><literal>\bowtie</literal>
</term><listitem><para>&#x22C8; Natural join of two relations (relation).
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\Box</primary></indexterm><literal>\Box</literal>
</term><listitem><para>&#x25A1; Modal operator for necessity; square open box
(ordinary).  Not available in plain &tex;. In &latex; you need to load the <filename>amssymb</filename> package.
<!-- bb Best Unicode equivalent? -->
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\bullet</primary></indexterm><literal>\bullet</literal>
</term><listitem><indexterm role="cp"><primary>bullet symbol</primary></indexterm>
<para>&#x2022; Bullet (binary).  Similar: multiplication
dot&#160;<literal>\cdot</literal>.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\cap</primary></indexterm><literal>\cap</literal>
</term><listitem><para>&#x2229; Intersection of two sets (binary).  Similar: variable-sized
operator&#160;<literal>\bigcap</literal>.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\cdot</primary></indexterm><literal>\cdot</literal>
</term><listitem><para>&#x22C5; Multiplication (binary).  Similar: Bullet
dot&#160;<literal>\bullet</literal>.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\chi</primary></indexterm><literal>\chi</literal>
</term><listitem><para>&#x03C7; Lowercase Greek chi (ordinary).
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\circ</primary></indexterm><literal>\circ</literal>
</term><listitem><para>&#x2218; Function composition, ring operator (binary).  Similar:
variable-sized operator&#160;<literal>\bigcirc</literal>.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\clubsuit</primary></indexterm><literal>\clubsuit</literal>
</term><listitem><para>&#x2663; Club card suit (ordinary).
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\complement</primary></indexterm><literal>\complement</literal>
</term><listitem><para>&#x2201; Set complement, used as a superscript as in
<literal>$S^\complement$</literal> (ordinary).  Not available in plain &tex;. In &latex; you need to load the <filename>amssymb</filename> package. Also used:
<literal>$S^{\mathsf{c}}$</literal> or&#160;<literal>$\bar{S}$</literal>.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\cong</primary></indexterm><literal>\cong</literal>
</term><listitem><para>&#x2245; Congruent (relation).
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\coprod</primary></indexterm><literal>\coprod</literal>
</term><listitem><para>&#x2210; Coproduct (operator).
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\cup</primary></indexterm><literal>\cup</literal>
</term><listitem><para>&#x222A; Union of two sets (binary).  Similar: variable-sized
operator&#160;<literal>\bigcup</literal>.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\dagger</primary></indexterm><literal>\dagger</literal>
</term><listitem><para>&#x2020; Dagger relation (binary).
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\dashv</primary></indexterm><literal>\dashv</literal>
</term><listitem><para>&#x22A3; Dash with vertical, reversed turnstile (relation).  Similar:
turnstile&#160;<literal>\vdash</literal>.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\ddagger</primary></indexterm><literal>\ddagger</literal>
</term><listitem><para>&#x2021; Double dagger relation (binary).
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\Delta</primary></indexterm><literal>\Delta</literal>
</term><listitem><para>&#x0394; Greek uppercase delta, used for increment (ordinary).
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\delta</primary></indexterm><literal>\delta</literal>
</term><listitem><para>&#x03B4; Greek lowercase delta (ordinary).
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\Diamond</primary></indexterm><literal>\Diamond</literal>
</term><listitem><para>&#x25C7; Large diamond operator (ordinary).  Not available in plain &tex;. In &latex; you need to load the <filename>amssymb</filename> package.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\diamond</primary></indexterm><literal>\diamond</literal>
</term><listitem><para>&#x22C4; Diamond operator (binary).  Similar: large
diamond&#160;<literal>\Diamond</literal>, circle bullet&#160;<literal>\bullet</literal>.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\diamondsuit</primary></indexterm><literal>\diamondsuit</literal>
</term><listitem><para>&#x2662; Diamond card suit (ordinary).
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\div</primary></indexterm><literal>\div</literal>
</term><listitem><para>&#x00F7; Division sign (binary).
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\doteq</primary></indexterm><literal>\doteq</literal>
</term><listitem><para>&#x2250; Approaches the limit (relation).  Similar: geometrically equal
to&#160;<literal>\Doteq</literal>.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\downarrow</primary></indexterm><literal>\downarrow</literal>
</term><listitem><para>&#x2193; Down arrow, converges (relation).  Similar:
<literal>\Downarrow</literal> double line down arrow.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\Downarrow</primary></indexterm><literal>\Downarrow</literal>
</term><listitem><para>&#x21D3; Double line down arrow (relation).  Similar:
<literal>\downarrow</literal> single line down arrow.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\ell</primary></indexterm><literal>\ell</literal>
</term><listitem><para>&#x2113; Lowercase cursive letter l (ordinary).
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\emptyset</primary></indexterm><literal>\emptyset</literal>
</term><listitem><para>&#x2205; Empty set symbol (ordinary).  The variant form is
<literal>\varnothing</literal>.
<!-- bb Why Unicode has \revemptyset but no \emptyset? -->
</para>	
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\epsilon</primary></indexterm><literal>\epsilon</literal>
</term><listitem><para>&#x03F5; Lowercase lunate epsilon (ordinary). Similar to
Greek text letter. More widely used in mathematics is the script small
letter epsilon <literal>\varepsilon</literal>&#160;&#x03B5;. Related:
the set membership relation <literal>\in</literal>&#160;&#x2208;.
<!-- src: David Carlisle http://tex.stackexchange.com/a/98018/339 and -->
<!-- Unicode referenced there asserts varepsilon is much more widely used. -->
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\equiv</primary></indexterm><literal>\equiv</literal>
</term><listitem><para>&#x2261; Equivalence (relation).
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\eta</primary></indexterm><literal>\eta</literal>
</term><listitem><para>&#x03B7; Lowercase Greek letter (ordinary).
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\exists</primary></indexterm><literal>\exists</literal>
</term><listitem><para>&#x2203; Existential quantifier (ordinary).
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\flat</primary></indexterm><literal>\flat</literal>
</term><listitem><para>&#x266D; Musical flat (ordinary).
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\forall</primary></indexterm><literal>\forall</literal>
</term><listitem><para>&#x2200; Universal quantifier (ordinary).
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\frown</primary></indexterm><literal>\frown</literal>
</term><listitem><para>&#x2322; Downward curving arc (ordinary).
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\Gamma</primary></indexterm><literal>\Gamma</literal>
</term><listitem><para>&#x0393; uppercase Greek letter (ordinary).
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\gamma</primary></indexterm><literal>\gamma</literal>
</term><listitem><para>&#x03B3; Lowercase Greek letter (ordinary).
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\ge</primary></indexterm><literal>\ge</literal>
</term><listitem><para>&#x2265; Greater than or equal to (relation).  This is a synonym
for&#160;<literal>\geq</literal>.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\geq</primary></indexterm><literal>\geq</literal>
</term><listitem><para>&#x2265; Greater than or equal to (relation).  This is a synonym
for&#160;<literal>\ge</literal>.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\gets</primary></indexterm><literal>\gets</literal>
</term><listitem><para>&#x2190; Is assigned the value (relation).
Synonym:&#160;<literal>\leftarrow</literal>.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\gg</primary></indexterm><literal>\gg</literal>
</term><listitem><para>&#x226B; Much greater than (relation).  Similar: much less
than&#160;<literal>\ll</literal>.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\hbar</primary></indexterm><literal>\hbar</literal>
</term><listitem><para>&#x210F; Planck constant over two pi (ordinary).
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\heartsuit</primary></indexterm><literal>\heartsuit</literal>
</term><listitem><para>&#x2661; Heart card suit (ordinary).
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\hookleftarrow</primary></indexterm><literal>\hookleftarrow</literal>
</term><listitem><para>&#x21A9; Hooked left arrow (relation).
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\hookrightarrow</primary></indexterm><literal>\hookrightarrow</literal>
</term><listitem><para>&#x21AA; Hooked right arrow (relation).
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\iff</primary></indexterm><literal>\iff</literal>
</term><listitem><para>&#x27F7; If and only if (relation).  It is <literal>\Longleftrightarrow</literal>
with a <literal>\thickmuskip</literal> on either side.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\Im</primary></indexterm><literal>\Im</literal>
</term><listitem><para>&#x2111; Imaginary part (ordinary).  See: real part&#160;<literal>\Re</literal>.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\imath</primary></indexterm><literal>\imath</literal>
</term><listitem><indexterm role="cp"><primary>dotless i, math</primary></indexterm>
<para>Dotless i; used when you are putting an accent on an i (see <link linkend="Math-accents">Math
accents</link>).
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\in</primary></indexterm><literal>\in</literal>
</term><listitem><para>&#x2208; Set element (relation).  See also: lowercase lunate
epsilon&#160;<literal>\epsilon</literal>&#x03F5; and small letter script
epsilon&#160;<literal>\varepsilon</literal>.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\infty</primary></indexterm><literal>\infty</literal>
</term><listitem><para>&#x221E; Infinity (ordinary).
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\int</primary></indexterm><literal>\int</literal>
</term><listitem><para>&#x222B; Integral (operator).
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\iota</primary></indexterm><literal>\iota</literal>
</term><listitem><para>&#x03B9; Lowercase Greek letter (ordinary).
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\Join</primary></indexterm><literal>\Join</literal>
</term><listitem><para>&#x2A1D; Condensed bowtie symbol (relation).  Not available in Plain
&tex;.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\jmath</primary></indexterm><literal>\jmath</literal>
</term><listitem><indexterm role="cp"><primary>dotless j, math</primary></indexterm>
<para>Dotless j; used when you are putting an accent on a j (see <link linkend="Math-accents">Math
accents</link>).
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\kappa</primary></indexterm><literal>\kappa</literal>
</term><listitem><para>&#x03BA; Lowercase Greek letter (ordinary).
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\Lambda</primary></indexterm><literal>\Lambda</literal>
</term><listitem><para>&#x039B; uppercase Greek letter (ordinary).
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\lambda</primary></indexterm><literal>\lambda</literal>
</term><listitem><para>&#x03BB; Lowercase Greek letter (ordinary).
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\land</primary></indexterm><literal>\land</literal>
</term><listitem><para>&#x2227; Logical and (binary).  This is a synonym for <literal>\wedge</literal>.
See also logical or&#160;<literal>\lor</literal>.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\langle</primary></indexterm><literal>\langle</literal>
</term><listitem><para>&#x27E8; Left angle, or sequence, bracket (opening).  Similar:
less-than&#160;<literal>&lt;</literal>. Matches&#160;<literal>\rangle</literal>.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\lbrace</primary></indexterm><literal>\lbrace</literal>
</term><listitem><para>&#x007B; Left curly brace
(opening). Synonym:&#160;<literal>\{</literal>. Matches&#160;<literal>\rbrace</literal>.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\lbrack</primary></indexterm><literal>\lbrack</literal>
</term><listitem><para>&#x005B; Left square bracket (opening).
Synonym:&#160;<literal>[</literal>. Matches&#160;<literal>\rbrack</literal>.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\lceil</primary></indexterm><literal>\lceil</literal>
</term><listitem><para>&#x2308; Left ceiling bracket, like a square bracket but with the bottom
shaved off (opening). Matches&#160;<literal>\rceil</literal>.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\le</primary></indexterm><literal>\le</literal>
</term><listitem><para>&#x2264; Less than or equal to (relation).  This is a synonym
for&#160;<literal>\leq</literal>.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\leadsto</primary></indexterm><literal>\leadsto</literal>
</term><listitem><para>&#x21DD; Squiggly right arrow (relation).  Not available in plain &tex;. In &latex; you need to load the <filename>amssymb</filename> package.
To get this symbol outside of math mode you can put
<literal>\newcommand*{\Leadsto}{\ensuremath{\leadsto}}</literal> in the
preamble and then use <literal>\Leadsto</literal> instead.
<!-- bb Best Unicode equivalent? -->
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\Leftarrow</primary></indexterm><literal>\Leftarrow</literal>
</term><listitem><para>&#x21D0; Is implied by, double-line left arrow (relation).  Similar:
single-line left arrow&#160;<literal>\leftarrow</literal>.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\leftarrow</primary></indexterm><literal>\leftarrow</literal>
</term><listitem><para>&#x2190; Single-line left arrow (relation).
Synonym:&#160;<literal>\gets</literal>. Similar: double-line left
arrow&#160;<literal>\Leftarrow</literal>.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\leftharpoondown</primary></indexterm><literal>\leftharpoondown</literal>
</term><listitem><para>&#x21BD; Single-line left harpoon, barb under bar (relation).
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\leftharpoonup</primary></indexterm><literal>\leftharpoonup</literal>
</term><listitem><para>&#x21BC; Single-line left harpoon, barb over bar (relation).
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\Leftrightarrow</primary></indexterm><literal>\Leftrightarrow</literal>
</term><listitem><para>&#x21D4; Bi-implication; double-line double-headed arrow (relation).
Similar: single-line double headed arrow&#160;<literal>\leftrightarrow</literal>.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\leftrightarrow</primary></indexterm><literal>\leftrightarrow</literal>
</term><listitem><para>&#x2194; Single-line double-headed arrow (relation).  Similar:
double-line double headed arrow&#160;<literal>\Leftrightarrow</literal>.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\leq</primary></indexterm><literal>\leq</literal>
</term><listitem><para>&#x2264; Less than or equal to (relation).  This is a synonym
for&#160;<literal>\le</literal>.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\lfloor</primary></indexterm><literal>\lfloor</literal>
</term><listitem><para>&#x230A; Left floor bracket (opening). Matches:&#160;<literal>\floor</literal>.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\lhd</primary></indexterm><literal>\lhd</literal>
</term><listitem><para>&#x25C1; Arrowhead, that is, triangle, pointing left (binary).
Not available in plain &tex;. In &latex; you need to load the <filename>amssymb</filename> package. For the normal subgroup symbol you should load
<filename>amssymb</filename> and use&#160;<literal>\vartriangleleft</literal> (which is a relation
and so gives better spacing).
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\ll</primary></indexterm><literal>\ll</literal>
</term><listitem><para>&#x226A; Much less than (relation).  Similar: much greater
than&#160;<literal>\gg</literal>.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\lnot</primary></indexterm><literal>\lnot</literal>
</term><listitem><para>&#x00AC; Logical negation (ordinary). Synonym:&#160;<literal>\neg</literal>.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\longleftarrow</primary></indexterm><literal>\longleftarrow</literal>
</term><listitem><para>&#x27F5; Long single-line left arrow (relation).  Similar: long
double-line left arrow&#160;<literal>\Longleftarrow</literal>.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\longleftrightarrow</primary></indexterm><literal>\longleftrightarrow</literal>
</term><listitem><para>&#x27F7; Long single-line double-headed arrow (relation).  Similar: long
double-line double-headed arrow&#160;<literal>\Longleftrightarrow</literal>.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\longmapsto</primary></indexterm><literal>\longmapsto</literal>
</term><listitem><para>&#x27FC; Long single-line left arrow starting with vertical bar
(relation).  Similar: shorter version&#160;<literal>\mapsto</literal>.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\longrightarrow</primary></indexterm><literal>\longrightarrow</literal>
</term><listitem><para>&#x27F6; Long single-line right arrow (relation).  Similar: long
double-line right arrow&#160;<literal>\Longrightarrow</literal>.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\lor</primary></indexterm><literal>\lor</literal>
</term><listitem><para>&#x2228; Logical or (binary).  Synonym: wedge&#160;<literal>\wedge</literal>. 
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\mapsto</primary></indexterm><literal>\mapsto</literal>
</term><listitem><para>&#x21A6; Single-line left arrow starting with vertical bar (relation).
Similar: longer version&#160;<literal>\longmapsto</literal>.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\mho</primary></indexterm><literal>\mho</literal>
</term><listitem><para>&#x2127; Conductance, half-circle rotated capital omega (ordinary).
Not available in plain &tex;. In &latex; you need to load the <filename>amssymb</filename> package.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\mid</primary></indexterm><literal>\mid</literal>
</term><listitem><para>&#x2223; Single-line vertical bar (relation).  A typical use of
<literal>\mid</literal> is for a set <literal>\{\, x \mid x\geq 5 \,\}</literal>.
</para>
<para>Similar: <literal>\vert</literal> and&#160;<literal>|</literal> produce the same single-line
vertical bar symbol but without any spacing (they fall in class
ordinary) and you should not use them as relations but instead only as
ordinals, i.e., footnote symbols.  For absolute value, see the entry
for&#160;<literal>\vert</literal> and for norm see the entry for&#160;<literal>\Vert</literal>.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\models</primary></indexterm><literal>\models</literal>
</term><listitem><para>&#x22A8; Entails, or satisfies; double turnstile, short double dash
(relation).  Similar: long double dash&#160;<literal>\vDash</literal>.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\mp</primary></indexterm><literal>\mp</literal>
</term><listitem><para>&#x2213; Minus or plus (relation).
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\mu</primary></indexterm><literal>\mu</literal>
</term><listitem><para>&#x03BC; Lowercase Greek letter (ordinary).
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\nabla</primary></indexterm><literal>\nabla</literal>
</term><listitem><para>&#x2207; Hamilton&#8217;s del, or differential, operator (ordinary).
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\natural</primary></indexterm><literal>\natural</literal>
</term><listitem><para>&#x266E; Musical natural notation (ordinary).
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\ne</primary></indexterm><literal>\ne</literal>
</term><listitem><para>&#x2260; Not equal (relation). Synonym:&#160;<literal>\neq</literal>.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\nearrow</primary></indexterm><literal>\nearrow</literal>
</term><listitem><para>&#x2197; North-east arrow (relation).
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\neg</primary></indexterm><literal>\neg</literal>
</term><listitem><para>&#x00AC; Logical negation (ordinary).
Synonym:&#160;<literal>\lnot</literal>. Sometimes instead used for
negation:&#160;<literal>\sim</literal>.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\neq</primary></indexterm><literal>\neq</literal>
</term><listitem><para>&#x2260; Not equal (relation). Synonym:&#160;<literal>\ne</literal>.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\ni</primary></indexterm><literal>\ni</literal>
</term><listitem><para>&#x220B; Reflected membership epsilon; has the member
(relation). Synonym:&#160;<literal>\owns</literal>. Similar: is a member
of&#160;<literal>\in</literal>.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\not</primary></indexterm><literal>\not</literal>
</term><listitem><!-- the "@ "s put in spaces so the not slash doesn't hit the next char. -->
<para>&#x0020;&#160;&#160;&#160;&#160;Long solidus, or slash, used to overstrike a
following operator (relation).
</para>
<para>Many negated operators are available that don&#8217;t require <literal>\not</literal>,
particularly with the <filename>amssymb</filename> package. For example, <literal>\notin</literal>
is typographically preferable to <literal>\not\in</literal>.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\notin</primary></indexterm><literal>\notin</literal>
</term><listitem><para>&#x2209; Not an element of (relation).  Similar: not subset
of&#160;<literal>\nsubseteq</literal>.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\nu</primary></indexterm><literal>\nu</literal>
</term><listitem><para>&#x03BD; Lowercase Greek letter (ordinary).
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\nwarrow</primary></indexterm><literal>\nwarrow</literal>
</term><listitem><para>&#x2196; North-west arrow (relation).
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\odot</primary></indexterm><literal>\odot</literal>
</term><listitem><para>&#x2299; Dot inside a circle (binary).  Similar: variable-sized
operator&#160;<literal>\bigodot</literal>.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\oint</primary></indexterm><literal>\oint</literal>
</term><listitem><para>&#x222E; Contour integral, integral with circle in the middle
(operator).
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\Omega</primary></indexterm><literal>\Omega</literal>
</term><listitem><para>&#x03A9; uppercase Greek letter (ordinary).
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\omega</primary></indexterm><literal>\omega</literal>
</term><listitem><para>&#x03C9; Lowercase Greek letter (ordinary).
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\ominus</primary></indexterm><literal>\ominus</literal>
</term><listitem><para>&#x2296; Minus sign, or dash, inside a circle (binary).
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\oplus</primary></indexterm><literal>\oplus</literal>
</term><listitem><para>&#x2295; Plus sign inside a circle (binary).  Similar: variable-sized
operator&#160;<literal>\bigoplus</literal>.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\oslash</primary></indexterm><literal>\oslash</literal>
</term><listitem><para>&#x2298; Solidus, or slash, inside a circle (binary).
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\otimes</primary></indexterm><literal>\otimes</literal>
</term><listitem><para>&#x2297; Times sign, or cross, inside a circle (binary).  Similar:
variable-sized operator&#160;<literal>\bigotimes</literal>.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\owns</primary></indexterm><literal>\owns</literal>
</term><listitem><para>&#x220B; Reflected membership epsilon; has the member
(relation). Synonym:&#160;<literal>\ni</literal>. Similar: is a member
of&#160;<literal>\in</literal>.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\parallel</primary></indexterm><literal>\parallel</literal>
</term><listitem><para>&#x2225; Parallel (relation). Synonym:&#160;<literal>\|</literal>.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\partial</primary></indexterm><literal>\partial</literal>
</term><listitem><para>&#x2202; Partial differential (ordinary).
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\perp</primary></indexterm><literal>\perp</literal>
</term><listitem><para>&#x27C2; Perpendicular (relation).  Similar:&#160;<literal>\bot</literal> uses the
same glyph but the spacing is different because it is in the class
ordinary.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\phi</primary></indexterm><literal>\phi</literal>
</term><listitem><para>&#x03D5; Lowercase Greek letter (ordinary).  The variant form is
<literal>\varphi</literal>&#160;&#x03C6;.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\Pi</primary></indexterm><literal>\Pi</literal>
</term><listitem><para>&#x03A0; uppercase Greek letter (ordinary).
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\pi</primary></indexterm><literal>\pi</literal>
</term><listitem><para>&#x03C0; Lowercase Greek letter (ordinary).  The variant form is
<literal>\varpi</literal>&#160;&#x03D6;.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\pm</primary></indexterm><literal>\pm</literal>
</term><listitem><para>&#x00B1; Plus or minus (binary).
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\prec</primary></indexterm><literal>\prec</literal>
</term><listitem><para>&#x227A; Precedes (relation). Similar: less than&#160;<literal>&lt;</literal>.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\preceq</primary></indexterm><literal>\preceq</literal>
</term><listitem><para>&#x2AAF; Precedes or equals (relation). Similar: less than or
equals&#160;<literal>\leq</literal>.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\prime</primary></indexterm><literal>\prime</literal>
</term><listitem><para>&#x2032; Prime, or minute in a time expression (ordinary).
Typically used as a superscript: <literal>$f^\prime$</literal>; <literal>$f^\prime$</literal>
and <literal>$f'$</literal> produce the same result.  An advantage of the second
is that <literal>$f'''$</literal> produces the desired symbol, that is, the same
result as <literal>$f^{\prime\prime\prime}$</literal>, but uses rather less
typing.  You can only use <literal>\prime</literal> in math mode.  Using the right
single quote&#160;<literal>'</literal> in text mode produces a different character
(apostrophe).
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\prod</primary></indexterm><literal>\prod</literal>
</term><listitem><para>&#x220F; Product (operator).
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\propto</primary></indexterm><literal>\propto</literal>
</term><listitem><para>&#x221D; Is proportional to (relation)
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\Psi</primary></indexterm><literal>\Psi</literal>
</term><listitem><para>&#x03A8; uppercase Greek letter (ordinary).
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\psi</primary></indexterm><literal>\psi</literal>
</term><listitem><para>&#x03C8; Lowercase Greek letter (ordinary).
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\rangle</primary></indexterm><literal>\rangle</literal>
</term><listitem><para>&#x27E9; Right angle, or sequence, bracket (closing).
Similar: greater than&#160;<literal>&gt;</literal>. Matches:<literal>\langle</literal>.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\rbrace</primary></indexterm><literal>\rbrace</literal>
</term><listitem><para>&#x007D; Right curly brace
(closing). Synonym:&#160;<literal>\}</literal>. Matches&#160;<literal>\lbrace</literal>.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\rbrack</primary></indexterm><literal>\rbrack</literal>
</term><listitem><para>&#x005D; Right square bracket
(closing). Synonym:&#160;<literal>]</literal>. Matches&#160;<literal>\lbrack</literal>.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\rceil</primary></indexterm><literal>\rceil</literal>
</term><listitem><para>&#x2309; Right ceiling bracket (closing). Matches&#160;<literal>\lceil</literal>.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\Re</primary></indexterm><literal>\Re</literal>
</term><listitem><para>&#x211C; Real part, real numbers, cursive capital R (ordinary). Related:
double-line, or blackboard bold, R&#160;<literal>\mathbb{R}</literal>; to access
this, load the <filename>amsfonts</filename> package.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\restriction</primary></indexterm><literal>\restriction</literal>
</term><listitem><para>&#x21BE; Restriction of a function (relation). Synonym:
<literal>\upharpoonright</literal>.  Not available in plain &tex;. In &latex; you need to load the <filename>amssymb</filename> package.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\revemptyset</primary></indexterm><literal>\revemptyset</literal>
</term><listitem><para>&#x29B0; Reversed empty set symbol (ordinary).  Related:
<literal>\varnothing</literal>. Not available in plain &tex;. In &latex; you need to load the <filename>stix</filename> package.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\rfloor</primary></indexterm><literal>\rfloor</literal>
</term><listitem><para>&#x230B; Right floor bracket, a right square bracket with the top cut
off (closing). Matches&#160;<literal>\lfloor</literal>.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\rhd</primary></indexterm><literal>\rhd</literal>
</term><listitem><para>&#x25C1; Arrowhead, that is, triangle, pointing right (binary).
Not available in plain &tex;. In &latex; you need to load the <filename>amssymb</filename> package. For the normal subgroup symbol you should instead
load <filename>amssymb</filename> and use&#160;<literal>\vartriangleright</literal> (which is a
relation and so gives better spacing).
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\rho</primary></indexterm><literal>\rho</literal>
</term><listitem><para>&#x03C1; Lowercase Greek letter (ordinary).  The variant form is
<literal>\varrho</literal>&#160;&#x03F1;.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\Rightarrow</primary></indexterm><literal>\Rightarrow</literal>
</term><listitem><para>&#x21D2; Implies, right-pointing double line arrow
(relation). Similar: right single-line arrow&#160;<literal>\rightarrow</literal>.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\rightarrow</primary></indexterm><literal>\rightarrow</literal>
</term><listitem><para>&#x2192; Right-pointing single line arrow (relation).
Synonym:&#160;<literal>\to</literal>. Similar: right double line
arrow&#160;<literal>\Rightarrow</literal>.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\rightharpoondown</primary></indexterm><literal>\rightharpoondown</literal>
</term><listitem><para>&#x21C1; Right-pointing harpoon with barb below
the line (relation). 
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\rightharpoonup</primary></indexterm><literal>\rightharpoonup</literal>
</term><listitem><para>&#x21C0; Right-pointing harpoon with barb above the
line (relation).
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\rightleftharpoons</primary></indexterm><literal>\rightleftharpoons</literal>
</term><listitem><para>&#x21CC; Right harpoon up above left harpoon down
(relation).
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\searrow</primary></indexterm><literal>\searrow</literal>
</term><listitem><para>&#x2198; Arrow pointing southeast (relation).
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\setminus</primary></indexterm><literal>\setminus</literal>
</term><listitem><para>&#x29F5; Set difference, reverse solidus or reverse slash,
like \ (binary). Similar: backslash&#160;<literal>\backslash</literal> and also
<literal>\textbackslash</literal> outside of math mode.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\sharp</primary></indexterm><literal>\sharp</literal>
</term><listitem><para>&#x266F; Musical sharp (ordinary).
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\Sigma</primary></indexterm><literal>\Sigma</literal>
</term><listitem><para>&#x03A3; uppercase Greek letter (ordinary).
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\sigma</primary></indexterm><literal>\sigma</literal>
</term><listitem><para>&#x03C3; Lowercase Greek letter (ordinary). The variant form is
<literal>\varsigma</literal>&#160;&#x03C2;.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\sim</primary></indexterm><literal>\sim</literal>
</term><listitem><para>&#x223C; Similar, in a relation (relation).
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\simeq</primary></indexterm><literal>\simeq</literal>
</term><listitem><para>&#x2243; Similar or equal to, in a relation (relation).
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\smallint</primary></indexterm><literal>\smallint</literal>
</term><listitem><para>&#x222B; Integral sign that does not change to a larger size in a
display (operator).
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\smile</primary></indexterm><literal>\smile</literal>
</term><listitem><para>&#x2323; Upward curving arc, smile (ordinary).
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\spadesuit</primary></indexterm><literal>\spadesuit</literal>
</term><listitem><para>&#x2660; Spade card suit (ordinary).
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\sqcap</primary></indexterm><literal>\sqcap</literal>
</term><listitem><para>&#x2293; Square intersection symbol (binary). Similar:
intersection&#160;<literal>cap</literal>.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\sqcup</primary></indexterm><literal>\sqcup</literal>
</term><listitem><para>&#x2294; Square union symbol (binary). Similar:
union&#160;<literal>cup</literal>. Related: variable-sized
operator&#160;<literal>\bigsqcup</literal>.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\sqsubset</primary></indexterm><literal>\sqsubset</literal>
</term><listitem><para>&#x228F; Square subset symbol (relation). Similar:
subset&#160;<literal>\subset</literal>. Not available in plain &tex;. In &latex; you need to load the <filename>amssymb</filename> package.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\sqsubseteq</primary></indexterm><literal>\sqsubseteq</literal>
</term><listitem><para>&#x2291; Square subset or equal symbol (binary). Similar: subset or
equal to&#160;<literal>\subseteq</literal>.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\sqsupset</primary></indexterm><literal>\sqsupset</literal>
</term><listitem><para>&#x2290; Square superset symbol (relation). Similar:
superset&#160;<literal>\supset</literal>. Not available in plain &tex;. In &latex; you need to load the <filename>amssymb</filename> package.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\sqsupseteq</primary></indexterm><literal>\sqsupseteq</literal>
</term><listitem><para>&#x2292; Square superset or equal symbol (binary).
Similar: superset or equal&#160;<literal>\supseteq</literal>.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\star</primary></indexterm><literal>\star</literal>
</term><listitem><para>&#x22C6; Five-pointed star, sometimes used as a general binary
operation but sometimes reserved for cross-correlation (binary).
Similar: the synonyms asterisk&#160;<literal>*</literal> and <literal>\ast</literal>, which
are six-pointed, and more often appear as a superscript or subscript,
as with the Kleene star.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\subset</primary></indexterm><literal>\subset</literal>
</term><listitem><para>&#x2282; Subset (occasionally, is implied by) (relation).
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\subseteq</primary></indexterm><literal>\subseteq</literal>
</term><listitem><para>&#x2286; Subset or equal to (relation).
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\succ</primary></indexterm><literal>\succ</literal>
</term><listitem><para>&#x227B; Comes after, succeeds (relation). Similar: is less
than&#160;<literal>&gt;</literal>.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\succeq</primary></indexterm><literal>\succeq</literal>
</term><listitem><para>&#x2AB0; Succeeds or is equal to (relation). Similar: less
than or equal to&#160;<literal>\leq</literal>.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\sum</primary></indexterm><literal>\sum</literal>
</term><listitem><para>&#x2211; Summation (operator). Similar: Greek capital
sigma&#160;<literal>\Sigma</literal>.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\supset</primary></indexterm><literal>\supset</literal>
</term><listitem><para>&#x2283; Superset (relation).
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\supseteq</primary></indexterm><literal>\supseteq</literal>
</term><listitem><para>&#x2287; Superset or equal to (relation).
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\surd</primary></indexterm><literal>\surd</literal>
</term><listitem><para>&#x221A; Radical symbol (ordinary).  The &latex; command
<literal>\sqrt{...}</literal> typesets the square root of the argument, with a bar
that extends to cover the argument.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\swarrow</primary></indexterm><literal>\swarrow</literal>
</term><listitem><para>&#x2199; Southwest-pointing  arrow (relation).
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\tau</primary></indexterm><literal>\tau</literal>
</term><listitem><para>&#x03C4; Lowercase Greek letter (ordinary).
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\theta</primary></indexterm><literal>\theta</literal>
</term><listitem><para>&#x03B8; Lowercase Greek letter (ordinary). The variant form is
<literal>\vartheta</literal>&#160;&#x03D1;.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\times</primary></indexterm><literal>\times</literal>
</term><listitem><para>&#x00D7; Primary school multiplication sign (binary). See
also&#160;<literal>\cdot</literal>.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\to</primary></indexterm><literal>\to</literal>
</term><listitem><para>&#x2192; Right-pointing single line arrow (relation).
Synonym:&#160;<literal>\rightarrow</literal>.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\top</primary></indexterm><literal>\top</literal>
</term><listitem><para>&#x22A4; Top, greatest element of a partially ordered set
(ordinary). See also&#160;<literal>\bot</literal>.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\triangle</primary></indexterm><literal>\triangle</literal>
</term><listitem><para>&#x25B3; Triangle (ordinary).
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\triangleleft</primary></indexterm><literal>\triangleleft</literal>
</term><listitem><para>&#x25C1; Not-filled triangle pointing left
(binary). Similar:&#160;<literal>\lhd</literal>. For the normal subgroup symbol you
should load <filename>amssymb</filename> and use&#160;<literal>\vartriangleleft</literal> (which
is a relation and so gives better spacing).
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\triangleright</primary></indexterm><literal>\triangleright</literal>
</term><listitem><para>&#x25B7; Not-filled triangle pointing right
(binary). For the normal subgroup symbol you should instead load
<filename>amssymb</filename> and use&#160;<literal>\vartriangleright</literal> (which is a
relation and so gives better spacing).
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\unlhd</primary></indexterm><literal>\unlhd</literal>
</term><listitem><para>&#x22B4; Left-pointing not-filled underlined arrowhead, that is,
triangle, with a line under (binary). Not available in plain &tex;. In &latex; you need to load the <filename>amssymb</filename> package. For the
normal subgroup symbol load <filename>amssymb</filename> and
use&#160;<literal>\vartrianglelefteq</literal> (which is a relation and so gives
better spacing).
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\unrhd</primary></indexterm><literal>\unrhd</literal>
</term><listitem><para>&#x22B5; Right-pointing not-filled underlined arrowhead, that is,
triangle, with a line under (binary). Not available in plain &tex;. In &latex; you need to load the <filename>amssymb</filename> package. For the
normal subgroup symbol load <filename>amssymb</filename> and
use&#160;<literal>\vartrianglerighteq</literal> (which is a relation and so gives
better spacing).
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\Uparrow</primary></indexterm><literal>\Uparrow</literal>
</term><listitem><para>&#x21D1; Double-line upward-pointing arrow
(relation). Similar: single-line up-pointing
arrow&#160;<literal>\uparrow</literal>.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\uparrow</primary></indexterm><literal>\uparrow</literal>
</term><listitem><para>&#x2191; Single-line upward-pointing arrow, diverges
(relation). Similar: double-line up-pointing
arrow&#160;<literal>\Uparrow</literal>.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\Updownarrow</primary></indexterm><literal>\Updownarrow</literal>
</term><listitem><para>&#x21D5; Double-line upward-and-downward-pointing arrow
(relation). Similar: single-line upward-and-downward-pointing
arrow&#160;<literal>\updownarrow</literal>.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\updownarrow</primary></indexterm><literal>\updownarrow</literal>
</term><listitem><para>&#x2195; Single-line upward-and-downward-pointing arrow
(relation). Similar: double-line upward-and-downward-pointing
arrow&#160;<literal>\Updownarrow</literal>.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\upharpoonright</primary></indexterm><literal>\upharpoonright</literal>
</term><listitem><para>&#x21BE; Up harpoon, with barb on right side
(relation). Synonym:&#160;<literal>&#92;restriction</literal>.
Not available in plain &tex;. In &latex; you need to load the <filename>amssymb</filename> package.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\uplus</primary></indexterm><literal>\uplus</literal>
</term><listitem><para>&#x228E; Multiset union, a union symbol with a plus symbol in
the middle (binary). Similar: union&#160;<literal>\cup</literal>. Related:
variable-sized operator&#160;<literal>\biguplus</literal>.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\Upsilon</primary></indexterm><literal>\Upsilon</literal>
</term><listitem><para>&#x03A5; uppercase Greek letter (ordinary).
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\upsilon</primary></indexterm><literal>\upsilon</literal>
</term><listitem><para>&#x03C5; Lowercase Greek letter (ordinary).
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\varepsilon</primary></indexterm><literal>\varepsilon</literal>
</term><listitem><para>&#x03B5; Small letter script epsilon (ordinary).  This is
more widely used in mathematics than the non-variant lunate epsilon form
<literal>\epsilon</literal>&#160;&#x03F5;. Related: set
membership&#160;<literal>\in</literal>.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\vanothing</primary></indexterm><literal>\vanothing</literal>
</term><listitem><para>&#x2205; Empty set symbol. Similar: <literal>\emptyset</literal>. Related:
<literal>\revemptyset</literal>. Not available in plain &tex;. In &latex; you need to load the <filename>amssymb</filename> package.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\varphi</primary></indexterm><literal>\varphi</literal>
</term><listitem><para>&#x03C6; Variant on the lowercase Greek letter (ordinary).
The non-variant form is <literal>\phi</literal>&#160;&#x03D5;.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\varpi</primary></indexterm><literal>\varpi</literal>
</term><listitem><para>&#x03D6; Variant on the lowercase Greek letter (ordinary).
The non-variant form is <literal>\pi</literal>&#160;&#x03C0;.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\varrho</primary></indexterm><literal>\varrho</literal>
</term><listitem><para>&#x03F1; Variant on the lowercase Greek letter (ordinary).
The non-variant form is <literal>\rho</literal>&#160;&#x03C1;.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\varsigma</primary></indexterm><literal>\varsigma</literal>
</term><listitem><para>&#x03C2; Variant on the lowercase Greek letter
(ordinary).  The non-variant form is
<literal>\sigma</literal>&#160;&#x03C3;.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\vartheta</primary></indexterm><literal>\vartheta</literal>
</term><listitem><para>&#x03D1; Variant on the lowercase Greek letter
(ordinary).  The non-variant form is
<literal>\theta</literal>&#160;&#x03B8;.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\vdash</primary></indexterm><literal>\vdash</literal>
</term><listitem><para>&#x22A2; Provable; turnstile, vertical and a dash
(relation). Similar: turnstile rotated a
half-circle&#160;<literal>\dashv</literal>.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\vee</primary></indexterm><literal>\vee</literal>
</term><listitem><para>&#x2228; Logical or; a downwards v shape (binary). Related:
logical and&#160;<literal>\wedge</literal>. Similar: variable-sized
operator&#160;<literal>\bigvee</literal>.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\Vert</primary></indexterm><literal>\Vert</literal>
</term><listitem><para>&#x2016; Vertical double bar (ordinary). Similar: vertical single
bar&#160;<literal>\vert</literal>.
</para>
<para>For a norm symbol, you can use the <filename>mathtools</filename> package and put in
your preamble
<literal>\DeclarePairedDelimiter\norm{\lVert}{\rVert}</literal>. This gives you
three command variants for double-line vertical bars that are correctly
horizontally spaced: if in the document body you write the starred
version <literal>$\norm*{M^\perp}$</literal> then the height of the vertical bars
will match the height of the argument, whereas with
<literal>\norm{M^\perp}</literal> the bars do not grow with the height of the
argument but instead are the default height, and <literal>\norm[<replaceable>size
command</replaceable>]{M^\perp}</literal> also gives bars that do not grow but are set to
the size given in the <replaceable>size command</replaceable>, e.g., <literal>\Bigg</literal>.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\vert</primary></indexterm><literal>\vert</literal>
</term><listitem><para>&#x007C; Single line vertical bar (ordinary). Similar: double-line
vertical bar&#160;<literal>\Vert</literal>. For such that, as in the definition of a 
set, use&#160;<literal>\mid</literal> because it is a relation.  
</para>
<para>For absolute value you can use the <filename>mathtools</filename> package and in your
preamble put
<literal>\DeclarePairedDelimiter\abs{\lvert}{\rvert}</literal>. This gives you
three command variants for single-line vertical bars that are correctly
horizontally spaced: if in the document body you write the starred
version <literal>$\abs*{\frac{22}{7}}$</literal> then the height of the
vertical bars will match the height of the argument, whereas with
<literal>\abs{\frac{22}{7}}</literal> the bars do not grow with the height of
the argument but instead are the default height, and
<literal>\abs[<replaceable>size command</replaceable>]{\frac{22}{7}}</literal> also gives bars that
do not grow but are set to the size given in the <replaceable>size command</replaceable>,
e.g., <literal>\Bigg</literal>.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\wedge</primary></indexterm><literal>\wedge</literal>
</term><listitem><para>&#x2227; Logical and (binary).  Synonym:&#160;<literal>\land</literal>.  See also
logical or <literal>\vee</literal>. Similar: variable-sized
operator&#160;<literal>\bigwedge</literal>.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\wp</primary></indexterm><literal>\wp</literal>
</term><listitem><para>&#x2118; Weierstrass p (ordinary).
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\wr</primary></indexterm><literal>\wr</literal>
</term><listitem><para>&#x2240; Wreath product (binary).
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\Xi</primary></indexterm><literal>\Xi</literal>
</term><listitem><para>&#x039E; uppercase Greek letter (ordinary).
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\xi</primary></indexterm><literal>\xi</literal>
</term><listitem><para>&#x03BE; Lowercase Greek letter (ordinary).
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\zeta</primary></indexterm><literal>\zeta</literal>
</term><listitem><para>&#x03B6; Lowercase Greek letter (ordinary).
</para>
</listitem></varlistentry></variablelist>
<para>The following symbols are most often used in plain text but &latex;
provides versions to use in mathematical text.
</para>
<variablelist><varlistentry><term><indexterm role="fn"><primary>\mathdollar</primary></indexterm><literal>\mathdollar</literal>
</term><listitem><para>Dollar sign in math mode: $.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\mathparagraph</primary></indexterm><literal>\mathparagraph</literal>
</term><listitem><para>Paragraph sign (pilcrow) in math mode, &#x00B6;.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\mathsection</primary></indexterm><literal>\mathsection</literal>
</term><listitem><para>Section sign in math mode &#x00A7;.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\mathsterling</primary></indexterm><literal>\mathsterling</literal>
</term><listitem><para>Sterling sign in math mode: &#163;.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\mathunderscore</primary></indexterm><literal>\mathunderscore</literal>
</term><listitem><para>Underscore in math mode: _.
</para>
</listitem></varlistentry></variablelist>


<sect2 label="16.2.1" id="Blackboard-bold">
<title>Blackboard bold</title>

<indexterm role="cp"><primary>blackboard bold</primary></indexterm>
<indexterm role="cp"><primary>doublestruck</primary></indexterm>

<para>Synopsis:
</para>
<screen>\usepackage{amssymb}   % in preamble
  ... 
\mathbb{<replaceable>uppercase-letter</replaceable>}
</screen>
<para>Provide blackboard bold symbols, sometimes also known as doublestruck
letters, used to denote number sets such as the natural numbers, the
integers, etc.
</para>
<para>Here 
</para>
<screen>\( \forall n \in \mathbb{N}, n^2 \geq 0 \)
</screen>
<para>the <literal>\mathbb{N}</literal> gives blackboard bold symbol &#x2115;
representing the natural numbers.
</para>
<para>If you use other than an uppercase letter then you do not get an error
but you get strange results, including unexpected characters.
</para>
<para>There are packages that give access to symbols other than just the
capital letters; look on CTAN.
</para>

</sect2>
<sect2 label="16.2.2" id="Calligraphic">
<title>Calligraphic</title>

<indexterm role="cp"><primary>calligraphic fonts</primary></indexterm>
<indexterm role="cp"><primary>script fonts</primary></indexterm>
<indexterm role="cp"><primary>fonts, script</primary></indexterm>

<para>Synopsis:
</para>
<screen>\mathcal{<replaceable>uppercase-letters</replaceable>}
</screen>
<para>Use a script-like font.
</para>
<para>In this example the graph identifier is output in a cursive font.
</para>
<screen>Let the graph be \( \mathcal{G} \).
</screen>
<para>If you use something other than an uppercase letter then you do not get
an error.  Instead you get unexpected output.  For instance,
<literal>\mathcal{g}</literal> outputs a close curly brace symbol, while
<literal>\mathcal{+}</literal> outputs a plus sign.  
</para>

</sect2>
<sect2 label="16.2.3" id="_005cboldmath-_0026-_005cunboldmath">
<title><literal>\boldmath</literal> &amp; <literal>\unboldmath</literal></title>

<indexterm role="cp"><primary>boldface mathematics</primary></indexterm>
<indexterm role="cp"><primary>symbols, boldface</primary></indexterm>
<indexterm role="fn"><primary>\boldmath</primary></indexterm>
<indexterm role="fn"><primary>\unboldmath</primary></indexterm>

<para>Synopsis (used in paragraph mode or LR mode):
</para>
<screen>\boldmath \( <replaceable>math</replaceable> \)
</screen>
<para>or
</para>
<screen>\unboldmath \( <replaceable>math</replaceable> \)
</screen>
<indexterm role="fn"><primary>\boldmath</primary></indexterm>
<indexterm role="fn"><primary>\unboldmath</primary></indexterm>
<para>Declarations to change the letters and symbols in <replaceable>math</replaceable> to be in
a bold font, or to countermand that and bring back the regular
(non-bold) default. They must be used when not in math mode or display
math mode (see <link linkend="Modes">Modes</link>). Both commands are fragile
(see <link linkend="_005cprotect">\protect</link>).
</para>
<para>In this example each <literal>\boldmath</literal> command takes place inside an
<literal>\mbox</literal>,
</para>
<screen>we have $\mbox{\boldmath \( v \)} = 5\cdot\mbox{\boldmath \( u \)$}$
</screen>
<para>which means <literal>\boldmath</literal> is only called in a text mode, here LR
mode, and explains why &latex; must switch to math mode to set <literal>v</literal>
and <literal>u</literal>.
</para>
<para>If you use either command inside math mode, as with <literal>Trouble: \(
\boldmath x \)</literal>, then you get something like &#8216;<literal>LaTeX Font Warning:
Command \boldmath invalid in math mode on input line 11</literal>&#8217; and &#8216;<literal>LaTeX
Font Warning: Command \mathversion invalid in math mode on input line
11</literal>&#8217;.
</para>
<indexterm role="cp"><primary>package, <literal>bm</literal></primary></indexterm>
<indexterm role="cp"><primary><literal>bm</literal> package</primary></indexterm>

<para>There are many issues with <literal>\boldmath</literal>.  New documents should use
the <filename>bm</filename> package provided by the &latex; Project team.  A complete
description is outside the scope of this document (see the full
documentation on CTAN) but even this small example
</para>
<screen>\usepackage{bm}   % in preamble
...
we have $\bm{v} = 5\cdot\bm{u}$
</screen>
<para>shows that it is an improvement over <literal>\boldmath</literal>.
</para>

</sect2>
<sect2 label="16.2.4" id="Dots">
<title>Dots, horizontal or vertical</title>

<indexterm role="cp"><primary>ellipses</primary></indexterm>
<indexterm role="cp"><primary>dots</primary></indexterm>

<para>Ellipses are the three dots (usually three) indicating that a pattern
continues.
</para>
<screen>\begin{array}{cccc}
  a_{0,0}    &amp;a_{0,1}   &amp;a_{0,2} &amp;\ldots \\
  a_{1,0}    &amp;\ddots                     \\
  \vdots
\end{array}
</screen>
<para>&latex; provides these.
</para>
<variablelist><anchor id="ellipses-cdots"/><varlistentry><term><indexterm role="fn"><primary>\cdots</primary></indexterm><literal>\cdots</literal>
</term><listitem><para>Horizontal ellipsis with the dots raised to the center of the line, as
in &#x22EF;.  Used as: <literal>\( a_0\cdot a_1\cdots a_{n-1}
\)</literal>.
</para>
<anchor id="ellipses-ddots"/></listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\ddots</primary></indexterm><literal>\ddots</literal>
</term><listitem><para>Diagonal ellipsis, &#x22F1;.  See the above array example for a
usage.
</para>
<anchor id="ellipses-ldots"/></listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\ldots</primary></indexterm><literal>\ldots</literal>
</term><listitem><para>Ellipsis on the baseline, &#x2026;.  Used as: <literal>\(
x_0,\ldots x_{n-1} \)</literal>.  Another example is the above array example. A
synonym is <literal>\mathellipsis</literal>.  A synonym from the <filename>amsmath</filename>
package is <literal>\hdots</literal>.
</para>
<para>You can also use this command outside of mathematical text, as in
<literal>The gears, brakes, \ldots{} are all broken</literal>.  (In a paragraph
mode or LR mode a synonym for <literal>\ldots</literal> is <literal>\dots</literal>.)
</para>
<anchor id="ellipses-vdots"/></listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\vdots</primary></indexterm><literal>\vdots</literal>
</term><listitem><para>Vertical ellipsis, &#x22EE;.  See the above array example for a
usage.
</para>
</listitem></varlistentry></variablelist>
<indexterm role="cp"><primary>package, <literal>amsmath</literal></primary></indexterm>
<indexterm role="cp"><primary><literal>amsmath</literal> package</primary></indexterm>

<para>The <filename>amsmath</filename> package has the command <literal>\dots</literal> to semantically
mark up ellipses.  This example produces two different-looking outputs
for the first two uses of the <literal>\dots</literal> command.
</para>
<screen>\usepackage{amsmath}  % in preamble
  ...
Suppose that \( p_0, p_1, \dots, p_{n-1} \) lists all of the primes.
Observe that \( p_0\cdot p_1 \dots \cdot p_{n-1} +1 \) is not a
  multiple of any \( p_i \).
Conclusion: there are infinitely many primes \( p_0, p_1, \dotsc \).
</screen>
<para>In the first line &latex; looks to the comma following <literal>\dots</literal> to
determine that it should output an ellipsis on the baseline.  The second
line has a <literal>\cdot</literal> following <literal>\dots</literal> so &latex; outputs an
ellipsis that is on the math axis, vertically centered.  However, the
third usage has no follow-on character so you have to tell &latex; what
to do.  You can use one of the commands: <literal>\dotsc</literal> if you need the
ellipsis appropriate for a comma following, <literal>\dotsb</literal> if you need
the ellipses that fits when the dots are followed by a binary operator
or relation symbol, <literal>\dotsi</literal> for dots with integrals, or
<literal>\dotso</literal> for others.
</para>

</sect2>
</sect1>
<sect1 label="16.3" id="Math-functions">
<title>Math functions</title>

<indexterm role="cp"><primary>math functions</primary></indexterm>
<indexterm role="cp"><primary>functions, math</primary></indexterm>

<para>These commands produce roman function names in math mode with proper
spacing.
</para>
<variablelist><varlistentry><term><indexterm role="fn"><primary>\arccos</primary></indexterm><literal>\arccos</literal>
</term><listitem><para>Inverse cosine
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\arcsin</primary></indexterm><literal>\arcsin</literal>
</term><listitem><para>Inverse sine
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\arctan</primary></indexterm><literal>\arctan</literal>
</term><listitem><para>Inverse tangent
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\arg</primary></indexterm><literal>\arg</literal>
</term><listitem><para>Angle between the real axis and a point in the complex plane
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\bmod</primary></indexterm><literal>\bmod</literal>
</term><listitem><para>Binary modulo operator, used as in <literal>\( 5\bmod 3=2 \)</literal>
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\cos</primary></indexterm><literal>\cos</literal>
</term><listitem><para>Cosine
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\cosh</primary></indexterm><literal>\cosh</literal>
</term><listitem><para>Hyperbolic cosine
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\cot</primary></indexterm><literal>\cot</literal>
</term><listitem><para>Cotangent
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\coth</primary></indexterm><literal>\coth</literal>
</term><listitem><para>Hyperbolic cotangent
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\csc</primary></indexterm><literal>\csc</literal>
</term><listitem><para>Cosecant
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\deg</primary></indexterm><literal>\deg</literal>
</term><listitem><para>Degrees
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\det</primary></indexterm><literal>\det</literal>
</term><listitem><para>Determinant
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\dim</primary></indexterm><literal>\dim</literal>
</term><listitem><para>Dimension
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\exp</primary></indexterm><literal>\exp</literal>
</term><listitem><para>Exponential
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\gcd</primary></indexterm><literal>\gcd</literal>
</term><listitem><para>Greatest common divisor
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\hom</primary></indexterm><literal>\hom</literal>
</term><listitem><para>Homomorphism
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\inf</primary></indexterm><literal>\inf</literal>
</term><listitem><para>Infinum
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\ker</primary></indexterm><literal>\ker</literal>
</term><listitem><para>Kernel
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\lg</primary></indexterm><literal>\lg</literal>
</term><listitem><para>Base 2 logarithm
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\lim</primary></indexterm><literal>\lim</literal>
</term><listitem><para>Limit
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\liminf</primary></indexterm><literal>\liminf</literal>
</term><listitem><para>Limit inferior
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\limsup</primary></indexterm><literal>\limsup</literal>
</term><listitem><para>Limit superior
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\ln</primary></indexterm><literal>\ln</literal>
</term><listitem><para>Natural logarithm
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\log</primary></indexterm><literal>\log</literal>
</term><listitem><para>Logarithm
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\max</primary></indexterm><literal>\max</literal>
</term><listitem><para>Maximum
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\min</primary></indexterm><literal>\min</literal>
</term><listitem><para>Minimum
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\pmod</primary></indexterm><literal>\pmod</literal>
</term><listitem><para>Parenthesized modulus, as used in <literal>\( 5\equiv 2\pmod 3 \)</literal>
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\Pr</primary></indexterm><literal>\Pr</literal>
</term><listitem><para>Probability
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\sec</primary></indexterm><literal>\sec</literal>
</term><listitem><para>Secant
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\sin</primary></indexterm><literal>\sin</literal>
</term><listitem><para>Sine
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\sinh</primary></indexterm><literal>\sinh</literal>
</term><listitem><para>Hyperbolic sine
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\sup</primary></indexterm><literal>\sup</literal>
</term><listitem><para>sup
<!-- don't try to use \sup with dvi/pdf output since that turned into a -->
<!-- Texinfo command and it's not worth hassling with different versions -->
<!-- when it's just three roman letters anyway. -->
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\tan</primary></indexterm><literal>\tan</literal>
</term><listitem><para>Tangent
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\tanh</primary></indexterm><literal>\tanh</literal>
</term><listitem><para>Hyperbolic tangent
</para>
</listitem></varlistentry></variablelist>
<indexterm role="cp"><primary>package, <literal>amsmath</literal></primary></indexterm>
<indexterm role="cp"><primary><literal>amsmath</literal> package</primary></indexterm>

<para>The <filename>amsmath</filename> package adds improvements on some of these, and also
allows you to define your own.  The full documentation is on CTAN, but
briefly, you can define an identity operator with
<literal>\DeclareMathOperator{\identity}{id}</literal> that is like the ones
above but prints as &#8216;<literal>id</literal>&#8217;.  The starred form
<literal>\DeclareMathOperator*{\op}{op}</literal> sets any limits above and
below, as is traditional with <literal>\lim</literal>, <literal>\sup</literal>, or <literal>\max</literal>.
</para>

</sect1>
<sect1 label="16.4" id="Math-accents">
<title>Math accents</title>

<indexterm role="cp"><primary>math accents</primary></indexterm>
<indexterm role="cp"><primary>accents, mathematical</primary></indexterm>

<para>&latex; provides a variety of commands for producing accented letters
in math.  These are different from accents in normal text
(see <link linkend="Accents">Accents</link>).
</para>
<variablelist><varlistentry><term><indexterm role="fn"><primary>\acute</primary></indexterm><literal>\acute</literal>
</term><listitem><indexterm role="cp"><primary>acute accent, math</primary></indexterm>
<para>Math acute accent
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\bar</primary></indexterm><literal>\bar</literal>
</term><listitem><indexterm role="cp"><primary>bar-over accent, math</primary></indexterm>
<indexterm role="cp"><primary>macron accent, math</primary></indexterm>
<para>Math bar-over accent
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\breve</primary></indexterm><literal>\breve</literal>
</term><listitem><indexterm role="cp"><primary>breve accent, math</primary></indexterm>
<para>Math breve accent
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\check</primary></indexterm><literal>\check</literal>
</term><listitem><indexterm role="cp"><primary>check accent, math</primary></indexterm>
<indexterm role="cp"><primary>h&#225;&#269;ek accent, math</primary></indexterm>
<para>Math h&#225;&#269;ek (check) accent
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\ddot</primary></indexterm><literal>\ddot</literal>
</term><listitem><indexterm role="cp"><primary>double dot accent, math</primary></indexterm>
<para>Math dieresis accent
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\dot</primary></indexterm><literal>\dot</literal>
</term><listitem><indexterm role="cp"><primary>overdot accent, math</primary></indexterm>
<indexterm role="cp"><primary>dot over accent, math</primary></indexterm>
<para>Math dot accent
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\grave</primary></indexterm><literal>\grave</literal>
</term><listitem><indexterm role="cp"><primary>grave accent, math</primary></indexterm>
<para>Math grave accent
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\hat</primary></indexterm><literal>\hat</literal>
</term><listitem><indexterm role="cp"><primary>hat accent, math</primary></indexterm>
<indexterm role="cp"><primary>circumflex accent, math</primary></indexterm>
<para>Math hat (circumflex) accent
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\mathring</primary></indexterm><literal>\mathring</literal>
</term><listitem><indexterm role="cp"><primary>ring accent, math</primary></indexterm>
<para>Math ring accent  <!-- don't bother implementing in texinfo -->
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\tilde</primary></indexterm><literal>\tilde</literal>
</term><listitem><indexterm role="cp"><primary>tilde accent, math</primary></indexterm>
<para>Math tilde accent
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\vec</primary></indexterm><literal>\vec</literal>
</term><listitem><indexterm role="cp"><primary>vector symbol, math</primary></indexterm>
<para>Math vector symbol
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\widehat</primary></indexterm><literal>\widehat</literal>
</term><listitem><indexterm role="cp"><primary>wide hat accent, math</primary></indexterm>
<para>Math wide hat accent
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\widetilde</primary></indexterm><literal>\widetilde</literal>
</term><listitem><indexterm role="cp"><primary>wide tilde accent, math</primary></indexterm>
<para>Math wide tilde accent
</para>
</listitem></varlistentry></variablelist>
<para>When you are putting an accent on an i or a j, the tradition is to use
one without a dot, <literal>\imath</literal> or <literal>jmath</literal> (see <link linkend="Math-symbols">Math symbols</link>).
</para>

</sect1>
<sect1 label="16.5" id="Over_002d-and-Underlining">
<title>Over- and Underlining</title>

<indexterm role="cp"><primary>overlining</primary></indexterm>
<indexterm role="cp"><primary>underlining</primary></indexterm>

<para>&latex; provides commands for making overlines or underlines, or
putting braces over or under some material.
</para>
<variablelist>
<varlistentry><term><indexterm role="fn"><primary>\underline{<replaceable>text</replaceable>}</primary></indexterm><literal>\underline{<replaceable>text</replaceable>}</literal>
</term><listitem><para>Underline <replaceable>text</replaceable>.  Works inside math mode, and outside.
The line is always completely below the text, taking account of
descenders, so in <literal>\(\underline{y}\)</literal> the line is lower than in
<literal>\(\underline{x}\)</literal>.  This command is fragile (see <link linkend="_005cprotect">\protect</link>).
</para>
<indexterm role="cp"><primary>package, <literal>ulem</literal></primary></indexterm>
<indexterm role="cp"><primary><literal>ulem</literal> package</primary></indexterm>

<para>Note that the package <filename>ulem</filename> does text mode underlining and allows
line breaking as well as a number of other features.  See the
documentation on CTAN.  See also&#160;<link linkend="_005chrulefill-_0026-_005cdotfill">\hrulefill &amp; \dotfill</link> for
producing a line, for such things as a signature.
</para>        
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\overline{<replaceable>text</replaceable>}</primary></indexterm><literal>\overline{<replaceable>text</replaceable>}</literal>
</term><listitem><para>Put a horizontal line over <replaceable>text</replaceable>.  Works inside math mode, and
outside.  For example, <literal>\overline{x+y}</literal>.
Note that this differs from the command <literal>\bar</literal> (see <link linkend="Math-accents">Math
accents</link>).
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\underbrace{<replaceable>math</replaceable>}</primary></indexterm><literal>\underbrace{<replaceable>math</replaceable>}</literal>
</term><listitem><para>Put a brace under <replaceable>math</replaceable>. For example, this
<literal>(1-\underbrace{1/2)+(1/2}-1/3)</literal> emphasizes the telescoping part.
Attach text to the brace by using subscript, <literal>_</literal>, or superscript,
<literal>^</literal>, as here.
</para>
<screen>\begin{displaymath}
  1+1/2+\underbrace{1/3+1/4}_{&gt;1/2}+
       \underbrace{1/5+1/6+1/7+1/8}_{&gt;1/2}+\cdots
\end{displaymath}
</screen>
<para>The superscript appears on top of the expression, and so can look
unconnected to the underbrace.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\overbrace{<replaceable>math</replaceable>}</primary></indexterm><literal>\overbrace{<replaceable>math</replaceable>}</literal>
</term><listitem><para>Put a brace over <replaceable>math</replaceable>, as with
<literal>\overbrace{x+x+\cdots+x}^{\mbox{\(k\) times}}</literal>.  See also
<literal>\underbrace</literal>.
</para>
</listitem></varlistentry></variablelist>
<indexterm role="cp"><primary>package, <literal>mathtools</literal></primary></indexterm>
<indexterm role="cp"><primary><literal>mathtools</literal> package</primary></indexterm>

<para>The package <filename>mathtools</filename> adds an over- and underbrace, as well as
some improvements on the braces.  See the documentation on CTAN.
</para>

</sect1>
<sect1 label="16.6" id="Spacing-in-math-mode">
<title>Spacing in math mode</title>

<indexterm role="cp"><primary>spacing within math mode</primary></indexterm>
<indexterm role="cp"><primary>math mode, spacing</primary></indexterm>

<para>When typesetting mathematics, &latex; puts in spacing according to the
normal rules for mathematics texts.  If you enter <literal>y=m x</literal> then
&latex; ignores the space and in the output the m is next to the x,
as <inlineequation><mathphrase>y=mx</mathphrase></inlineequation>.
</para>
<para>But &latex;&#8217;s rules sometimes need tweaking.  For example, in an
integral the tradition is to put a small extra space between the
<literal>f(x)</literal> and the <literal>dx</literal>, here done with the <literal>\,</literal> command.
</para>
<screen>\int_0^1 f(x)\,dx
</screen>
<para>&latex; provides the commands that follow for use in math mode.  Many
of these spacing definitions are expressed in terms of the math unit
<firstterm>mu</firstterm>.  It is defined as 1/18em, where the em is taken from the
current math symbols family (see <link linkend="Units-of-length">Units of length</link>).  Thus, a
<literal>\thickspace</literal> is something like 5/18 times the width of
a&#160;&#8216;<literal>M</literal>&#8217;.
</para>
<variablelist><varlistentry><term><literal>\;</literal>
</term><listitem><indexterm role="fn"><primary>\;</primary></indexterm>
<indexterm role="fn"><primary>\thickspace</primary></indexterm>
<anchor id="spacing-in-math-mode-thickspace"/><para>Synonym: <literal>\thickspace</literal>.  Normally <literal>5.0mu plus 5.0mu</literal>.  Math
mode only.
</para>
</listitem></varlistentry><varlistentry><term><literal>\:</literal>
</term><term><literal>\&gt;</literal>
</term><listitem><indexterm role="fn"><primary>\:</primary></indexterm>
<indexterm role="fn"><primary>\&gt;</primary></indexterm>
<indexterm role="fn"><primary>\medspace</primary></indexterm>
<anchor id="spacing-in-math-mode-medspace"/><para>Synonym: <literal>\medspace</literal>.  Normally <literal>4.0mu plus 2.0mu minus 4.0mu</literal>.
Math mode only.
</para>
</listitem></varlistentry><varlistentry><term><literal>\,</literal>
</term><listitem><indexterm role="fn"><primary>\,</primary></indexterm>
<indexterm role="fn"><primary>\thinspace</primary></indexterm>
<indexterm role="cp"><primary>thin space</primary></indexterm>
<anchor id="Spacing-in-math-mode_002f_005cthinspace"/><anchor id="spacing-in-math-mode-thinspace"/><para>Synonym: <literal>\thinspace</literal>.  Normally <literal>3mu</literal>, which is 1/6em.
Can be used in both math mode and text mode (see <link linkend="_005cthinspace-_0026-_005cnegthinspace">\thinspace &amp;
\negthinspace</link>).
</para>
<para>This space is widely used, for instance between the function and the
infinitesimal in an integral <literal>\int f(x)\,dx</literal> and, if an author does
this, before punctuation in a displayed equation.
</para>
<screen>The antiderivative is
\begin{equation}
  3x^{-1/2}+3^{1/2}\,.
\end{equation}
</screen>
</listitem></varlistentry><varlistentry><term><literal>\!</literal>
</term><listitem><indexterm role="fn"><primary>\!</primary></indexterm>
<indexterm role="fn"><primary>\negthinspace</primary></indexterm>
<indexterm role="cp"><primary>thin space, negative</primary></indexterm>
<anchor id="spacing-in-math-mode-negthinspace"/><para>A negative thin space. Normally <literal>-3mu</literal>.  The <literal>\!</literal> command is
math mode only but the <literal>\negthinspace</literal> command is available for
text mode (see <link linkend="_005cthinspace-_0026-_005cnegthinspace">\thinspace &amp; \negthinspace</link>).
</para>
</listitem></varlistentry><varlistentry><term><literal>\quad</literal>
</term><listitem><indexterm role="cp"><primary>quad</primary></indexterm>
<indexterm role="fn"><primary>\quad</primary></indexterm>
<anchor id="spacing-in-math-mode-quad"/><para>This is 18mu, that is, 1em. This is often used for space
surrounding equations or expressions, for instance for the space between
two equations inside a <literal>displaymath</literal> environment.  It is available
in both text and math mode.
</para>
</listitem></varlistentry><varlistentry><term><literal>\qquad</literal>
</term><listitem><indexterm role="fn"><primary>\qquad</primary></indexterm>
<anchor id="spacing-in-math-mode-qquad"/><para>A length of 2 quads, that is, 36mu = 2em.  It is available in
both text and math mode.
</para></listitem></varlistentry></variablelist>

</sect1>
<sect1 label="16.7" id="Math-miscellany">
<title>Math miscellany</title>

<indexterm role="cp"><primary>math miscellany</primary></indexterm>

<para>&latex; contains a wide variety of mathematics facilities.  Here are
some that don&#8217;t fit into other categories.
</para>


<sect2 label="16.7.1" id="Colon-character-_0026-_005ccolon">
<title>Colon character <literal>:</literal> &amp; <literal>\colon</literal></title>

<indexterm role="cp"><primary>:</primary></indexterm>
<indexterm role="cp"><primary>colon character</primary></indexterm>
<indexterm role="fn"><primary>:</primary></indexterm>
<indexterm role="fn"><primary>\colon</primary></indexterm>

<para>Synopsis, one of:
</para>
<screen>:
\colon
</screen>
<para>In mathematics, the colon character, <literal>:</literal>, is a relation.
</para>
<screen>With side ratios \( 3:4 \) and \( 4:5 \), the triangle is right.
</screen>
<indexterm role="cp"><primary>package, <literal>amsmath</literal></primary></indexterm>
<indexterm role="cp"><primary><literal>amsmath</literal> package</primary></indexterm>

<para>Ordinary &latex; defines <literal>\colon</literal> to produce the colon character
with the spacing appropriate for punctuation, as in set-builder notation
<literal>\{x\colon 0\leq x&lt;1\}</literal>.
</para>
<indexterm role="cp"><primary>package, <literal>amsmath</literal></primary></indexterm>
<indexterm role="cp"><primary><literal>amsmath</literal> package</primary></indexterm>

<para>But the widely-used <filename>amsmath</filename> package defines <literal>\colon</literal> for use
in the definition of functions <literal>f\colon D\to C</literal>.  So if you want
the colon character as a punctuation then use <literal>\mathpunct{:}</literal>.
</para>

</sect2>
<sect2 label="16.7.2" id="_005c_002a">
<title><literal>\*</literal></title>

<indexterm role="cp"><primary>multiplication, discretionary</primary></indexterm>
<indexterm role="cp"><primary>breaks, multiplication discretionary</primary></indexterm>
<indexterm role="cp"><primary>line breaks, multiplication discretionary</primary></indexterm>
<indexterm role="cp"><primary>discretionary breaks, multiplication</primary></indexterm>
<indexterm role="fn"><primary>\*</primary></indexterm>

<para>Synopsis:
</para>
<screen>\*
</screen>
<para>A multiplication symbol that allows a line break.  If there is a break
then &latex; puts a <literal>\times</literal> symbol, &#x00D7;, before
that break.
</para>
<para>In <literal>\( A_1\* A_2\* A_3\* A_4 \)</literal>, if there is no line break then
&latex; outputs it as though it were <literal>\( A_1 A_2 A_3 A_4 \)</literal>.  If
a line break does happen, for example between the two middle ones, then
&latex; sets it like <literal>\( A_1 A_2 \times \)</literal>, followed by the
break, followed by <literal>\( A_3 A_4 \)</literal>.
</para>

</sect2>
<sect2 label="16.7.3" id="_005cfrac">
<title><literal>\frac</literal></title>

<indexterm role="cp"><primary>fraction</primary></indexterm>
<indexterm role="fn"><primary>\frac</primary></indexterm>

<para>Synopsis:
</para>
<screen>\frac{<replaceable>numerator</replaceable>}{<replaceable>denominator</replaceable>}
</screen>
<para>Produces the fraction.  Used as: <literal>\begin{displaymath}
\frac{1}{\sqrt{2\pi\sigma}} \end{displaymath}</literal>.  In inline math
mode it comes out small; see the discussion of <literal>\displaystyle</literal>
(see <link linkend="Math-formulas">Math formulas</link>).
</para>

</sect2>
<sect2 label="16.7.4" id="_005cleft-_0026-_005cright">
<title><literal>\left</literal> &amp; <literal>\right</literal></title>

<indexterm role="cp"><primary>delimiters, paired</primary></indexterm>
<indexterm role="cp"><primary>paired delimiters</primary></indexterm>
<indexterm role="cp"><primary>matching parentheses</primary></indexterm>
<indexterm role="cp"><primary>matching brackets</primary></indexterm>
<indexterm role="cp"><primary>null delimiter</primary></indexterm>
<indexterm role="fn"><primary>\left</primary></indexterm>
<indexterm role="fn"><primary>\right</primary></indexterm>

<para>Synopsis:
</para>
<screen>\left <replaceable>delimiter1</replaceable> ... \right <replaceable>delimiter2</replaceable>
</screen>
<para>Make matching parentheses, braces, or other delimiters.  The delimiters
are sized according to the math they enclose.  This makes a unit vector
surrounded by appropriate-height parentheses.
</para>
<screen>\begin{equation}
  \left(\begin{array}{c}
    1   \\
    0   \\
  \end{array}\right)
</screen>
<para>Every <literal>\left</literal> must have a matching <literal>\right</literal>.  Leaving out the
<literal>\left(</literal> in the above gets &#8216;<literal>Extra \right</literal>&#8217;.  Leaving off the
<literal>\right)</literal> gets &#8216;<literal>You can't use `\eqno' in math mode</literal>&#8217;.
</para>
<para>However, the two delimiters <replaceable>delimiter1</replaceable> and <replaceable>delimiter2</replaceable> need
not match.  A common case is that you want an unmatched brace, as
below. Use a period, &#8216;<literal>.</literal>&#8217;, as a null delimiter.
</para>
<screen>\begin{equation}
  f(n)=\left\{\begin{array}{ll}
                1             &amp;\mbox{--if \(n=0\)} \\
                f(n-1)+3n^2   &amp;\mbox{--else}       
       \end{array}\right.
\end{equation}
</screen>
<para>Note that to get a curly brace as a delimiter you must prefix it with a
backslash, <literal>\{</literal>.
</para>

</sect2>
<sect2 label="16.7.5" id="_005csqrt">
<title><literal>\sqrt</literal></title>

<indexterm role="cp"><primary>square root</primary></indexterm>
<indexterm role="cp"><primary>roots</primary></indexterm>
<indexterm role="cp"><primary>radical</primary></indexterm>
<indexterm role="fn"><primary>\sqrt</primary></indexterm>

<para>Synopsis, one of:
</para>
<screen>\sqrt{<replaceable>arg</replaceable>}
\sqrt[<replaceable>root-number</replaceable>]{<replaceable>arg</replaceable>}
</screen>
<para>The square root, or optionally other roots, of <replaceable>arg</replaceable>.  The optional
argument <replaceable>root-number</replaceable> gives the root, i.e., enter the cube root of
<literal>x+y</literal> as <literal>\sqrt[3]{x+y}</literal>.
The radical grows with the size of <replaceable>arg</replaceable> (as the height of the
radical grows, the angle on the leftmost part gets steeper, until for
a large enough <literal>arg</literal>, it is vertical).
</para>
<para>&latex; has a separate <literal>\surd</literal> character (see <link linkend="Math-symbols">Math symbols</link>).
</para>

</sect2>
<sect2 label="16.7.6" id="_005cstackrel">
<title><literal>\stackrel</literal></title>

<indexterm role="cp"><primary>stack math</primary></indexterm>
<indexterm role="cp"><primary>relation, text above</primary></indexterm>
<indexterm role="fn"><primary>\stackrel</primary></indexterm>

<para>Synopsis, one of:
</para>
<screen>\stackrel{<replaceable>text</replaceable>}{<replaceable>relation</replaceable>}
</screen>
<para>Put <replaceable>text</replaceable> above <replaceable>relation</replaceable>.  To put a function name above an
arrow enter <literal>\stackrel{f}{\longrightarrow}</literal>.
</para>

</sect2>
</sect1>
</chapter>
<chapter label="17" id="Modes">
<title>Modes</title>

<indexterm role="cp"><primary>modes</primary></indexterm>

<para>As &latex; processes your document, at any point it is in one of six
modes.  They fall into three categories of two each, the horizontal
modes, the math modes, and the vertical modes. Some commands only work
in one mode or another (in particular, many commands only work in one of
the math modes), and error messages will refer to these.
</para>
<itemizedlist><listitem><anchor id="modes-paragraph-mode"/><indexterm role="cp"><primary>paragraph mode</primary></indexterm>
<para><firstterm>Paragraph mode</firstterm> is what &latex; is in when processing ordinary
text.  It breaks the input text into lines and breaks the lines into
pages.  This is the mode &latex; is in most of the time.
</para>
<indexterm role="cp"><primary>left-to-right mode</primary></indexterm>
<indexterm role="cp"><primary>LR mode</primary></indexterm>
<anchor id="modes-lr-mode"/><para><firstterm>LR mode</firstterm> (for left-to-right mode; in plain &tex; this is called
<firstterm>restricted horizontal mode</firstterm>) is in effect when &latex; starts
making a box with an <literal>\mbox</literal> command.  As in paragraph mode,
&latex;&#8217;s output is a string of words with spaces between them.  Unlike
in paragraph mode, in LR mode &latex; never starts a new line, it just
keeps going from left to right.  (Although &latex; will not complain
that the LR box is too long, when it is finished and next tries to put
that box into a line, it could very well complain that the finished LR
box won&#8217;t fit there.)
</para>
</listitem><listitem><indexterm role="cp"><primary>math mode</primary></indexterm>
<anchor id="modes-math-mode"/><para><firstterm>Math mode</firstterm> is when &latex; is generating
an inline mathematical formula.
</para>
<indexterm role="cp"><primary>display math mode</primary></indexterm>
<para><firstterm>Display math mode</firstterm> is when &latex; is generating a displayed
mathematical formula.  (Displayed formulas differ somewhat from inline
ones.  One example is that the placement of the subscript on <literal>\int</literal>
differs in the two situations.)
</para>
</listitem><listitem><indexterm role="cp"><primary>vertical mode</primary></indexterm>
<anchor id="modes-vertical-mode"/><para><firstterm>Vertical mode</firstterm> is when &latex; is building the list of lines and
other material making the output page.  This is the mode &latex; is in
when it starts a document.
</para>
<indexterm role="cp"><primary>internal vertical mode</primary></indexterm>
<anchor id="modes-internal-vertical-mode"/><para><firstterm>Internal vertical mode</firstterm> is in effect when &latex; starts making a
<literal>\vbox</literal>.  This is the vertical analogue of LR mode.
</para>
</listitem></itemizedlist>
<para>For instance, if you begin a &latex; article with &#8216;<literal>Let \( x \) be
...</literal>&#8217; then these are the modes: first &latex; starts every document in
vertical mode, then it reads the &#8216;<literal>L</literal>&#8217; and switches to paragraph
mode, then the next switch happens at the &#8216;<literal>\(</literal>&#8217; where &latex;
changes to math mode, and then when it leaves the formula it pops
back to paragraph mode.
</para>
<indexterm role="cp"><primary>inner paragraph mode</primary></indexterm>
<indexterm role="cp"><primary>outer paragraph mode</primary></indexterm>
<anchor id="modes-inner-paragraph-mode"/><anchor id="modes-outer-paragraph-mode"/><para>Paragraph mode has two subcases.  If you use a <literal>\parbox</literal> command or
or a <literal>minipage</literal> then &latex; is put into paragraph mode.  But it
will not put a page break here.  Inside one of these boxes, called a
<firstterm>parbox</firstterm>, &latex; is in <firstterm>inner paragraph mode</firstterm>.  Its more usual
situation, where it can put page breaks, is <firstterm>outer paragraph mode</firstterm>
(see <link linkend="Page-breaking">Page breaking</link>).
</para>

<sect1 label="17.1" id="_005censuremath">
<title><literal>\ensuremath</literal></title>

<para>Synopsis:
</para>
<screen>\ensuremath{<replaceable>formula</replaceable>}
</screen>
<para>Ensure that <replaceable>formula</replaceable> is typeset in math mode.
</para>
<para>For instance, you can redefine commands that ordinarily can be used only
in math mode, so that they can be used both in math and in plain text.
</para>
<screen>\newcommand{\dx}{\ensuremath{dx}}
In $\int f(x)\, \dx$, the \dx{} is an infinitesimal.
</screen>
<para>Caution: the <literal>\ensuremath</literal> command is useful but not a panacea.
</para>
<screen>\newcommand{\alf}{\ensuremath{\alpha}}
You get an alpha in text mode: \alf.
But compare the correct spacing in $\alf+\alf$ with that in \alf+\alf.
</screen>
<para>Best is to typeset math things in a math mode.
</para>

</sect1>
</chapter>
<chapter label="18" id="Page-styles">
<title>Page styles</title>

<indexterm role="cp"><primary>styles, page</primary></indexterm>
<indexterm role="cp"><primary>page styles</primary></indexterm>

<para>The style of a page determines where &latex; places the components of
that page, such as headers and footers, and the text body.  This
includes pages in the main part of the document but also includes
special pages such as the title page of a book, a page from an index, or
the first page of an article.
</para>
<indexterm role="cp"><primary>package, <literal>fancyhdr</literal></primary></indexterm>
<indexterm role="cp"><primary><literal>fancyhdr</literal> package</primary></indexterm>

<para>The package <filename>fancyhdr</filename> is very helpful for constructing page
styles.  See its documentation on CTAN.
</para>


<sect1 label="18.1" id="_005cmaketitle">
<title><literal>\maketitle</literal></title>

<indexterm role="cp"><primary>titles, making</primary></indexterm>
<indexterm role="fn"><primary>\maketitle</primary></indexterm>

<para>Synopsis:
</para>
<screen>\maketitle
</screen>
<para>Generate a title.  In the standard classes the title appears on a
separate page, except in the <literal>article</literal> class where it is at the top
of the first page.  (See <link linkend="Document-class-options">Document class options</link> for information about
the <literal>titlepage</literal> document class option.)
</para>
<para>This example shows <literal>\maketitle</literal> appearing in its usual place,
immediately after <literal>\begin{document}</literal>.
</para>
<screen>\documentclass{article}
\title{Constructing a Nuclear Reactor Using Only Coconuts}
\author{Jonas Grumby\thanks{%
    With the support of a Ginger Grant from the Roy Hinkley Society.} \\
  Skipper, \textit{Minnow}
  \and
  Willy Gilligan\thanks{%
    Thanks to the Mary Ann Summers foundation
    and to Thurston and Lovey Howell.}           \\
  Mate, \textit{Minnow}
  }
\date{1964-Sep-26}
\begin{document}
\maketitle
Just sit right back and you'll hear a tale, a tale of a fateful trip.
That started from this tropic port, aboard this tiny ship. The mate was
a mighty sailin' man, the Skipper brave and sure. Five passengers set
sail that day for a three hour tour. A three hour tour.
  ...
</screen>
<para>You tell &latex; the information used to produce the title by making
the following declarations.  These must come before the
<literal>\maketitle</literal>, either in the preamble or in the document body.
</para>
<variablelist><varlistentry><term><indexterm role="fn"><primary>\author{<replaceable>name1</replaceable> \and <replaceable>name2</replaceable> \and ...}</primary></indexterm><literal>\author{<replaceable>name1</replaceable> \and <replaceable>name2</replaceable> \and ...}</literal>
</term><listitem><indexterm role="cp"><primary>author, for titlepage</primary></indexterm>
<indexterm role="fn"><primary>\\ for <literal>\author</literal></primary></indexterm>
<indexterm role="fn"><primary>\and for <literal>\author</literal></primary></indexterm>
<para>Required.  Declare the document author or authors.  The argument is a
list of authors separated by <literal>\and</literal> commands.  To separate lines
within a single author&#8217;s entry, for instance to give the author&#8217;s
institution or address, use a double backslash, <literal>\\</literal>.  If you omit
the <literal>\author</literal> declaration then you get &#8216;<literal>LaTeX Warning: No
\author given</literal>&#8217;.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\date{<replaceable>text</replaceable>}</primary></indexterm><literal>\date{<replaceable>text</replaceable>}</literal>
</term><listitem><indexterm role="cp"><primary>date, for titlepage</primary></indexterm>
<para>Optional.  Declare <replaceable>text</replaceable> to be the document&#8217;s date.  The <replaceable>text</replaceable>
doesn&#8217;t need to be in a date format; it can be any text at all.  If you
omit <literal>\date</literal> then &latex; uses the current date (see <link linkend="_005ctoday">\today</link>).
To have no date, instead use <literal>\date{}</literal>.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\thanks{<replaceable>text</replaceable>}</primary></indexterm><literal>\thanks{<replaceable>text</replaceable>}</literal>
</term><listitem><indexterm role="cp"><primary>thanks, for titlepage</primary></indexterm>
<indexterm role="cp"><primary>credit footnote</primary></indexterm>
<para>Optional.  Produce a footnote.  You can use it in the author information
for acknowledgements, as illustrated below, but you can also use it in
the title, or any place a footnote makes sense.  It can be any text so
you can use it to print an email address, or for any purpose.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\title{<replaceable>text</replaceable>}</primary></indexterm><literal>\title{<replaceable>text</replaceable>}</literal>
</term><listitem><indexterm role="cp"><primary>title, for titlepage</primary></indexterm>
<indexterm role="fn"><primary>\\ for <literal>\title</literal></primary></indexterm>
<para>Required.  Declare <replaceable>text</replaceable> to be the title of the document.  Get line
breaks inside <replaceable>text</replaceable> with a double backslash, <literal>\\</literal>.  If you
omit the <literal>\title</literal> declaration then you get &#8216;<literal>LaTeX Error: No
\title given</literal>&#8217;.
</para>
</listitem></varlistentry></variablelist>
<para>Many publishers will provide a class to use in place of <literal>article</literal>
in that example, that formats the title according to their house
requirements.  To make your own, see&#160;<link linkend="titlepage">titlepage</link>. You can
either create this as a one-off or you can include it as part of a
renewed <literal>\maketitle</literal> command.
</para>

</sect1>
<sect1 label="18.2" id="_005cpagenumbering">
<title><literal>\pagenumbering</literal></title>

<indexterm role="fn"><primary>\pagenumbering</primary></indexterm>
<indexterm role="cp"><primary>page numbering style</primary></indexterm>

<para>Synopsis:
</para>
<screen>\pagenumbering{<replaceable>number-style</replaceable>}
</screen>
<para>Specifies the style of page numbers, and resets the page number.  The
numbering style is reflected on the page, and also in the table of
contents and other page references.  This declaration has global scope
so its effect is not delimited by braces or environments.
</para>
<para>In this example, before the Main section the pages are numbered
&#8216;<literal>a</literal>&#8217;, etc.  Starting on the page containing that section, the pages
are numbered &#8216;<literal>1</literal>&#8217;, etc.
</para>
<screen>\begin{document}\pagenumbering{alph}
  ...
\section{Main}\pagenumbering{arabic}
  ...
</screen>
<para>The argument <replaceable>number-style</replaceable> is one of the following (see
also&#160;<link linkend="_005calph-_005cAlph-_005carabic-_005croman-_005cRoman-_005cfnsymbol">\alph \Alph \arabic \roman \Roman \fnsymbol</link>).
</para>
<variablelist><varlistentry><term><literal>arabic</literal>
</term><listitem><para>arabic numerals: 1, 2, &#8230; 
</para>
</listitem></varlistentry><varlistentry><term><literal>roman</literal>
</term><listitem><para>lowercase Roman numerals: i, ii, &#8230;
</para>
</listitem></varlistentry><varlistentry><term><literal>Roman</literal>
</term><listitem><para>uppercase Roman numerals: I, II, &#8230;
</para>
</listitem></varlistentry><varlistentry><term><literal>alph</literal>
</term><listitem><para>lowercase letters: a, b, &#8230; If you have more than 26 pages then you
get &#8216;<literal>LaTeX Error: Counter too large</literal>&#8217;.
</para>
</listitem></varlistentry><varlistentry><term><literal>Alph</literal>
</term><listitem><para>uppercase letters: A, B, &#8230; If you have more than 26 pages then you
get &#8216;<literal>LaTeX Error: Counter too large</literal>&#8217;.
</para>
</listitem></varlistentry><varlistentry><term><literal>gobble</literal>
</term><listitem><indexterm role="cp"><primary>package, <literal>hyperref</literal></primary></indexterm>
<indexterm role="cp"><primary><literal>hyperref</literal> package</primary></indexterm>
<para>&latex; does not output a page number, although it
does get reset.  References to that page also are blank.  (This does not
work with the popular package <filename>hyperref</filename> so to have the page number
not appear you may want to instead use <literal>\pagestyle{empty}</literal> or
<literal>\thispagestyle{empty}</literal>.)
</para>
</listitem></varlistentry></variablelist>
<para>Traditionally, if a document has front matter&#8212;preface, table of
contents, etc.&#8212;then it is numbered with lowercase Roman numerals. The
main matter of a document uses arabic.  See <link linkend="_005cfrontmatter-_0026-_005cmainmatter-_0026-_005cbackmatter">\frontmatter &amp; \mainmatter
&amp; \backmatter</link>.
</para>
<para>If you want to address where the page number appears on the page,
see&#160;<link linkend="_005cpagestyle">\pagestyle</link>.  If you want to change the value of page
number then you will manipulate the <literal>page</literal> counter
(see <link linkend="Counters">Counters</link>).
</para>

</sect1>
<sect1 label="18.3" id="_005cpagestyle">
<title><literal>\pagestyle</literal></title>

<indexterm role="fn"><primary>\pagestyle</primary></indexterm>
<indexterm role="cp"><primary>header style</primary></indexterm>
<indexterm role="cp"><primary>footer style</primary></indexterm>
<indexterm role="cp"><primary>running header and footer style</primary></indexterm>

<para>Synopsis:
</para>
<screen>\pagestyle{<replaceable>style</replaceable>}
</screen>
<para>Declaration that specifies how the page headers and footers are typeset,
from the current page onwards.
</para>
<indexterm role="cp"><primary>package, <literal>fancyhdr</literal></primary></indexterm>
<indexterm role="cp"><primary><literal>fancyhdr</literal> package</primary></indexterm>

<para>A discussion with an example is below.  Note first that the package
<filename>fancyhdr</filename> is now the standard way to manipulate headers and
footers.  New documents that need to do anything other than one of the
standard options below should use this package.  See its documentation
on CTAN.
</para>
<para>Values for <replaceable>style</replaceable>:
</para>
<variablelist><varlistentry><term><literal>plain</literal>
</term><listitem><para>The header is empty.  The footer contains only a page number, centered.
</para>
</listitem></varlistentry><varlistentry><term><literal>empty</literal>
</term><listitem><para>The header and footer is empty.
</para>
</listitem></varlistentry><varlistentry><term><literal>headings</literal>
</term><listitem><para>Put running headers and footers on each page.  The document style
specifies what goes in there; see the discussion below.
</para>
</listitem></varlistentry><varlistentry><term><literal>myheadings</literal>
</term><listitem><para>Custom headers, specified via the <literal>\markboth</literal> or the
<literal>\markright</literal> commands.
</para>
</listitem></varlistentry></variablelist>
<para>Some discussion of the motivation for &latex;&#8217;s mechanism will help you
work with the options <literal>headings</literal> or <literal>myheadings</literal>.  The
document source below produces an article, two-sided, with the pagestyle
<literal>headings</literal>.  On this document&#8217;s left hand pages, &latex; wants (in
addition to the page number) the title of the current section.  On its
right hand pages &latex; wants the title of the current subsection.
When it makes up a page, &latex; gets this information from the
commands <literal>\leftmark</literal> and <literal>\rightmark</literal>.  So it is up to
<literal>\section</literal> and <literal>\subsection</literal> to store that information there.
</para>
<screen>\documentclass[twoside]{article}
\pagestyle{headings}
\begin{document}
  ... \section{Section 1} ... \subsection{Subsection 1.1} ...
\section{Section 2}
  ...
\subsection{Subsection 2.1}
  ...
\subsection{Subsection 2.2}
  ...
</screen>
<para>Suppose that the second section falls on a left page.  Although when the
page starts it is in the first section, &latex; will put
&#8216;<literal>Section&#160;2</literal>&#8217; in the left page header.  As to the right header,
if no subsection starts before the end of the right page then &latex;
blanks the right hand header.  If a subsection does appear before the
right page finishes then there are two cases.  If at least one
subsection starts on the right hand page then &latex; will put in the
right header the title of the first subsection starting on that right
page.  If at least one of 2.1, 2.2, &#8230;, starts on the left page but
none starts on the right then &latex; puts in the right hand header the
title of the last subsection to start, that is, the one in effect during
the right hand page.
</para>
<para>To accomplish this, in a two-sided article, &latex; has <literal>\section</literal>
issue a command <literal>\markboth</literal>, setting <literal>\leftmark</literal>
to &#8216;<literal>Section&#160;2</literal>&#8217; and setting <literal>\rightmark</literal> to blank.
And, &latex; has <literal>\subsection</literal> issue a command <literal>\markright</literal>,
setting <literal>\rightmark</literal> to &#8216;<literal>Subsection&#160;2.1</literal>&#8217;, etc.
</para>
<para>Here are the descriptions of <literal>\markboth</literal> and <literal>\markright</literal>:
</para>
<variablelist><varlistentry><term><indexterm role="fn"><primary>\markboth{<replaceable>left-head</replaceable>}{<replaceable>right-head</replaceable>}</primary></indexterm><literal>\markboth{<replaceable>left-head</replaceable>}{<replaceable>right-head</replaceable>}</literal>
</term><listitem><para>Sets both the right hand and left hand heading information for either a
page style of <literal>headings</literal> or <literal>myheadings</literal>.  A left hand page
heading <replaceable>left-head</replaceable> is generated by the last <literal>\markboth</literal>
command before the end of the page.  A right hand page heading
<replaceable>right-head</replaceable> is generated by the first <literal>\markboth</literal> or
<literal>\markright</literal> that comes on the page if there is one, otherwise by
the last one that came before that page.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\markright{<replaceable>right</replaceable>}</primary></indexterm><literal>\markright{<replaceable>right</replaceable>}</literal>
</term><listitem><para>Sets the right hand page heading, leaving the left unchanged.
</para>
</listitem></varlistentry></variablelist>

</sect1>
<sect1 label="18.4" id="_005cthispagestyle">
<title><literal>\thispagestyle</literal></title>

<indexterm role="fn"><primary>\thispagestyle</primary></indexterm>
<indexterm role="cp"><primary>page style, this page</primary></indexterm>

<para>Synopsis:
</para>
<screen>\thispagestyle{<replaceable>style</replaceable>}
</screen>
<para>Works in the same way as the <literal>\pagestyle</literal> (see <link linkend="_005cpagestyle">\pagestyle</link>),
except that it changes to <replaceable>style</replaceable> for the current page only.  This
declaration has global scope, so its effect is not delimited by braces
or environments.
</para>
<para>Often the first page of a chapter or section has a different style.  For
example, this &latex; book document has the first page of the first
chapter in in <literal>plain</literal> style, as is the default (see <link linkend="Page-styles">Page
styles</link>).
</para>
<screen>\documentclass{book}
\pagestyle{headings}
\begin{document}
\chapter{First chapter}
  ...
\chapter{Second chapter}\thispagestyle{empty}
  ...
</screen>
<para>The <literal>plain</literal> style has a page number on it, centered in the footer.
To make the page entirely empty, the command
<literal>\thispagestyle{empty}</literal> immediately follows the second
<literal>\chapter</literal>.
</para>

</sect1>
</chapter>
<chapter label="19" id="Spaces">
<title>Spaces</title>

<indexterm role="cp"><primary>spaces</primary></indexterm>
<indexterm role="cp"><primary>white space</primary></indexterm>

<para>&latex; has many ways to produce white (or filled) space.  Some of
these are best suited to mathematical text; see&#160;<link linkend="Spacing-in-math-mode">Spacing in
math mode</link>.  Some spacing commands are suitable for both regular text
and mathematical text; versions of some of these commands are in this
chapter.
</para>


<sect1 label="19.1" id="_005censpace-_0026-_005cquad-_0026-_005cqquad">
<title><literal>\enspace</literal> &amp; <literal>\quad</literal> &amp; <literal>\qquad</literal></title>

<indexterm role="fn"><primary>\enspace</primary></indexterm>
<indexterm role="fn"><primary>\quad</primary></indexterm>
<indexterm role="fn"><primary>\qquad</primary></indexterm>

<para>Synopsis, one of:
</para>
<screen>\enspace
\quad
\qquad
</screen>
<para>Insert a horizontal space of 1/2em, 1em, or 2em. The
em is a length defined by a font designer, often thought of as being the
width of a capital&#160;M.  One advantage of describing space in ems is
that it can be more portable across documents than an absolute
measurement such as points (see <link linkend="Lengths_002fem">Lengths/em</link>).
</para>
<para>This puts a suitable gap between two graphics.
</para>
<screen>\begin{center}
  \includegraphics{womensmile.png}%
  \qquad\includegraphics{mensmile.png}
\end{center}
</screen>
<para>See <link linkend="Spacing-in-math-mode">Spacing in math mode</link> for <literal>\quad</literal> and <literal>\qquad</literal>.  These
are lengths from centuries of typesetting and so may be a better choice
in many circumstances than arbitrary lengths, such as you get with
<literal>\hspace</literal>.
</para>

</sect1>
<sect1 label="19.2" id="_005chspace">
<title><literal>\hspace</literal></title>

<indexterm role="fn"><primary>\hspace</primary></indexterm>

<para>Synopsis, one of:
</para>
<screen>\hspace{<replaceable>length</replaceable>}
\hspace*{<replaceable>length</replaceable>}
</screen>
<para>Insert the horizontal space <replaceable>length</replaceable>.  The <replaceable>length</replaceable> can be
positive, negative, or zero; adding negative space is like backspacing.
It is a rubber length, that is, it may contain a <literal>plus</literal> or
<literal>minus</literal> component, or both (see <link linkend="Lengths">Lengths</link>).  Because the space is
stretchable and shrinkable, it is sometimes called <firstterm>glue</firstterm>.
</para>
<para>This makes a line with &#8216;<literal>Name:</literal>&#8217; an inch from the right margin.
</para>
<screen>\noindent\makebox[\linewidth][r]{Name:\hspace{1in}}
</screen>
<para>The <literal>*</literal>-version inserts horizontal space that non-discardable.
More precisely, when &tex; breaks a paragraph into lines any white
space&#8212;glues and kerns&#8212;that come at a line break are discarded.  The
<literal>*</literal>-version avoids that (technically, it adds a non-discardable
invisible item in front of the space).
</para>
<para>In this example
</para>
<screen>\parbox{0.8\linewidth}{%
  Fill in each blank: Four \hspace*{1in} and seven years ago our
  fathers brought forth on this continent, a new \hspace*{1in},
  conceived in \hspace*{1in}, and dedicated to the proposition
  that all men are created \hspace*{1in}.}
</screen>
<para>the 1&#160;inch blank following &#8216;<literal>conceived in</literal>&#8217; falls at the start
of a line.  If you erase the <literal>*</literal> then &latex; discards the blank.
</para>
<para>Here, the <literal>\hspace</literal> separates the three graphics.
</para>
<screen>\begin{center}
  \includegraphics{lion.png}%   comment keeps out extra space
  \hspace{1cm minus 0.25cm}\includegraphics{tiger.png}%
  \hspace{1cm minus 0.25cm}\includegraphics{bear.png}
\end{center}
</screen>
<para>Because the argument to each <literal>\hspace</literal> has <literal>minus 0.25cm</literal>,
each can shrink a little if the three figures are too wide.  But each
space won&#8217;t shrink more than 0.25cm (see <link linkend="Lengths">Lengths</link>).
</para>

</sect1>
<sect1 label="19.3" id="_005chfill">
<title><literal>\hfill</literal></title>

<indexterm role="fn"><primary>\hfill</primary></indexterm>

<indexterm role="cp"><primary>stretch, infinite horizontal</primary></indexterm>
<indexterm role="cp"><primary>infinite horizontal stretch</primary></indexterm>

<para>Synopsis:
</para>
<screen>\hfill
</screen>
<para>Produce a rubber length which has no natural space but that can stretch
horizontally as far as needed (see <link linkend="Lengths">Lengths</link>).
</para>
<para>This creates a one-line paragraph with &#8216;<literal>Name:</literal>&#8217; on the left side
of the page and &#8216;<literal>Quiz One</literal>&#8217; on the right.
</para>
<screen>\noindent Name:\hfill Quiz One
</screen>
<indexterm role="fn"><primary>\fill</primary></indexterm>
<para>The <literal>\hfill</literal> command is equivalent to <literal>\hspace{\fill}</literal> and
so the space can be discarded at line breaks.  To avoid that instead use
<literal>\hspace*{\fill}</literal> (see <link linkend="_005chspace">\hspace</link>).
</para>
<para>Here the graphs are evenly spaced in the middle of the figure.
</para>
<screen>\newcommand*{\vcenteredhbox}[1]{\begin{tabular}{@{}c@{}}#1\end{tabular}}
  ...
\begin{figure}
  \hspace*{\fill}%
  \vcenteredhbox{\includegraphics{graph0.png}}%
    \hfill\vcenteredhbox{\includegraphics{graph1.png}}%
  \hspace*{\fill}%
  \caption{Comparison of two graphs} \label{fig:twographs}
\end{figure}
</screen>
<para>Note the <literal>\hspace*</literal>&#8217;s where the space could otherwise be dropped.
</para>

</sect1>
<sect1 label="19.4" id="_005chss">
<title><literal>\hss</literal></title>

<indexterm role="fn"><primary>\hss</primary></indexterm>
<indexterm role="cp"><primary>horizontal space</primary></indexterm>
<indexterm role="cp"><primary>horizontal space, stretchable</primary></indexterm>
<indexterm role="cp"><primary>space, inserting horizontal</primary></indexterm>

<para>Synopsis:
</para>
<screen>\hss
</screen>
<para>Produce a horizontal space that is infinitely shrinkable as well as
infinitely stretchable (this command is a &tex; primitive).  &latex;
authors should reach first for the <literal>\makebox</literal> command to get the
effects of <literal>\hss</literal> (see <link linkend="_005cmbox-_0026-_005cmakebox">\mbox &amp; \makebox</link>).
</para>
<para>Here, the first line&#8217;s <literal>\hss</literal> makes the Z stick out to the right,
overwriting the Y.  In the second line the Z sticks out to the left,
overwriting the X.
</para>
<screen>X\hbox to 0pt{Z\hss}Y
X\hbox to 0pt{\hss Z}Y
</screen>
<para>Without the <literal>\hss</literal> you get something like &#8216;<literal>Overfull \hbox
(6.11111pt too wide) detected at line 20</literal>&#8217;.
</para>

</sect1>
<sect1 label="19.5" id="_005cspacefactor">
<title><literal>\spacefactor</literal></title>

<para>Synopsis:
</para>
<screen>\spacefactor=<replaceable>integer</replaceable>
</screen>
<indexterm role="fn"><primary>\spacefactor</primary></indexterm>
<para>Influence &latex;&#8217;s glue stretch and shrink behavior.  Most user-level
documents do not use this command.
</para>
<para>While &latex; is laying out the material, it may stretch or shrink the
gaps between words.  (This space is not a character; it is called the
<firstterm>interword glue</firstterm>; see <link linkend="_005chspace">\hspace</link>).  The <literal>\spacefactor</literal> command
(from Plain&#160;&tex;) allows you to, for instance, have the space
after a period stretch more than the space after a word-ending letter.
</para>
<para>After &latex; places each character, or rule or other box, it sets a
parameter called the <firstterm>space factor</firstterm>.  If the next thing in the input
is a space then this parameter affects how much stretching or shrinking
can happen.  A space factor that is larger than the normal value means
that the glue can stretch more and shrink less.  Normally, the space
factor is 1000. This value is in effect following most characters, and
any non-character box or math formula.  But it is 3000 after a period,
exclamation mark, or question mark, it is 2000 after a colon, 1500 after
a semicolon, 1250 after a comma, and 0 after a right parenthesis or
bracket, or closing double quote or single quote.  Finally, it is 999
after a capital letter.
</para>
<para>If the space factor&#160;<replaceable>f</replaceable> is 1000 then the glue gap will be the
font&#8217;s normal space value (for Computer Modern Roman 10 point this is
3.3333&#160;points).  Otherwise, if the space factor <replaceable>f</replaceable> is greater
than 2000 then &tex; adds the font&#8217;s extra space value (for Computer
Modern Roman 10 point this is 1.11111&#160;points), and then the font&#8217;s
normal stretch value is multiplied by <inlineequation><mathphrase>f /1000</mathphrase></inlineequation> and the normal
shrink value is multiplied by <inlineequation><mathphrase>1000/f</mathphrase></inlineequation> (for Computer Modern Roman
10 point these are 1.66666 and 1.11111&#160;points).
</para>
<para>For example, consider the period ending <literal>A man's best friend is his
dog.</literal>  After it, &tex; puts in a fixed extra space, and also allows the
glue to stretch 3 times as much and shrink 1/3 as much, as the glue
after <literal>friend</literal>, which does not end in a period.
</para>
<para>The rules for space factors are even more complex because they play
additional roles.  In practice, there are two consequences.  First, if a
period or other punctuation is followed by a right parenthesis or
bracket, or right single or double quote then the spacing effect of that
period carries through those characters (that is, the following glue
will have increased stretch and shrink).  Second, if
punctuation comes after a capital letter then its effect is not in place
so you get an ordinary space.  This second case also affects abbreviations
that do not end in a capital letter (see <link linkend="_005c_0040">\@</link>).
</para>
<para>You can only use <literal>\spacefactor</literal> in paragraph mode or LR mode
(see <link linkend="Modes">Modes</link>).  You can see the current value with
<literal>\the\spacefactor</literal> or <literal>\showthe\spacefactor</literal>.
</para>
<para>(Comment, not really related to <literal>\spacefactor</literal>: if you get errors
like &#8216;<literal>You can't use `\spacefactor' in vertical mode</literal>&#8217;, or &#8216;<literal>You
can't use `\spacefactor' in math mode.</literal>&#8217;, or &#8216;<literal>Improper \spacefactor</literal>&#8217;
then you have probably tried to redefine an internal command.
See <link linkend="_005cmakeatletter-_0026-_005cmakeatother">\makeatletter &amp; \makeatother</link>.)
</para>



<sect2 label="19.5.1" id="_005c_0040">
<title><literal>\@</literal></title>

<indexterm role="fn"><primary>\@</primary></indexterm>
<indexterm role="fn"><primary>at-sign</primary></indexterm>
<indexterm role="cp"><primary>period, sentence-ending</primary></indexterm>
<indexterm role="cp"><primary>period, abbreviation-ending</primary></indexterm>
<indexterm role="cp"><primary>period, spacing after</primary></indexterm>
<anchor id="_005cAT"/><!-- old name -->

<para>Synopsis:
</para>
<screen><replaceable>capital-letter</replaceable>\@.    
</screen>
<para>Treat a period as sentence-ending, where &latex; would otherwise think
it is part of an abbreviation.  &latex; thinks that a period ends an
abbreviation if the period comes after a capital letter, and otherwise
thinks the period ends the sentence.  By default, in justifying a line
&latex; adjusts the space after a sentence-ending period (or a question
mark, exclamation point, comma, or colon) more than it adjusts the space
between words (see <link linkend="_005cspacefactor">\spacefactor</link>).
</para>
<para>This example shows the two cases to remember.
</para>
<screen>The songs \textit{Red Guitar}, etc.\ are by Loudon Wainwright~III\@.
</screen>
<para>The second period ends the sentence, despite that it is preceded by a
capital.  We tell &latex; that it ends the sentence by putting
<literal>\@</literal> before it.  The first period ends the abbreviation
&#8216;<literal>etc.</literal>&#8217; but not the sentence.  The backslash-space, <literal>\ </literal>,
produces a mid-sentence space.
</para>
<para>So: if you have a capital letter followed by a period that ends the
sentence, then put <literal>\@</literal> before the period.  This holds even if
there is an intervening right parenthesis or bracket, or right single or
double quote, because the spacing effect of that period carries through
those characters.  For example, this
</para>
<screen>Use the \textit{Instructional Practices Guide},
(a book by the MAA)\@.
</screen>
<para>will have correct inter-sentence spacing after the period.
</para>
<para>The <literal>\@</literal> command is only for a text mode. If you use it outside of
a text mode then you get &#8216;<literal>You can't use `\spacefactor' in vertical
mode</literal>&#8217; (see <link linkend="Modes">Modes</link>).
</para>
<para>Comment: the converse case is a period ending an abbreviation whose last
letter is not a capital letter, and that abbreviation is not the last
word in the sentence.  For that case follow the period with a
backslash-space, (<literal>\ </literal>), or a tie, (<literal>~</literal>), or <literal>\@</literal>.
Examples are <literal>Nat.\ Acad.\ Science</literal>, and <literal>Mr.~Bean</literal>, and
<literal>(manure, etc.\@) for sale</literal> (note in the last one that the
<literal>\@</literal> comes before the closing parenthesis).
</para>

</sect2>
<sect2 label="19.5.2" id="_005cfrenchspacing">
<title><literal>\frenchspacing</literal></title>

<indexterm role="fn"><primary>\frenchspacing</primary></indexterm>
<indexterm role="fn"><primary>\nonfrenchspacing</primary></indexterm>
<indexterm role="cp"><primary>spacing, inter-sentence</primary></indexterm>

<para>Synopsis, one of:
</para>
<screen>\frenchspacing
\nonfrenchspacing
</screen>
<para>The first declaration causes &latex; to treat spacing between sentences
in the same way as spacing between words in the middle of a sentence.
The second causes spacing between sentences to stretch or shrink more
(see <link linkend="_005cspacefactor">\spacefactor</link>); this is the default.
</para>
<para>Some typographic traditions, including English, prefer to adjust the
space between sentences (or spaces following a question mark,
exclamation point, comma, or colon) more than the space between words
that are in the middle of a sentence.  Declaring <literal>\frenchspacing</literal>
(the command is from Plain &tex;) switches to the tradition that all
spaces are treated equally.
</para>

</sect2>
<sect2 label="19.5.3" id="_005cnormalsfcodes">
<title><literal>\normalsfcodes</literal></title>

<indexterm role="fn"><primary>\normalsfcodes</primary></indexterm>
<indexterm role="cp"><primary>spacing, inter-sentence</primary></indexterm>

<para>Synopsis:
</para>
<screen>\normalsfcodes
</screen>
<para>Reset the &latex; space factor values to the default
(see <link linkend="_005cspacefactor">\spacefactor</link>).
</para>

</sect2>
</sect1>
<sect1 label="19.6" id="_005c_0028SPACE_0029">
<title>Backslash-space, <literal>\ </literal></title>

<indexterm role="cp"><primary>\NEWLINE</primary></indexterm>
<indexterm role="cp"><primary>\SPACE</primary></indexterm>
<indexterm role="cp"><primary>\TAB</primary></indexterm>

<para>This section refers to the command consisting of two characters, a
backslash followed by a space. Synopsis:
</para>
<screen>\    
</screen>
<para>Produce a space. By default it produces white space of length
3.33333pt plus 1.66666pt minus 1.11111pt.
</para>
<para>A blank is not a space.  When you type a blank between words, &latex;
produces white space.  That&#8217;s different from an explicit space.  This
illustrates.
</para>
<screen>\begin{tabular}{l}
Three blanks:   in a row \\
Three spaces:\ \ \ in a row \\
\end{tabular}
</screen>
<para>On the first line &latex; collapses the three blanks to output one
whitespace (it would be the same with a single blank or, for instance,
with a blank and an tab and a blank, or a blank and a newline and a
blank). But the second line asks for three spaces so the white area is
wider.  Thus, the backslash-space command will create some horizontal
space.  (But the best way to create horizontal space is with
<literal>\hspace</literal>; See <link linkend="_005chspace">\hspace</link>.)
</para>
<para>The backslash-space command has two main uses.  First, it is often used
after control sequences to keep them from gobbling the space that
follows, as in <literal>\TeX\ is nice</literal>.  (But the approach of using curly
parentheses, as in <literal>\TeX{} is nice</literal>, has the advantage of still
working if the next character is a period.)
</para>
<para>The second common use is that
it mark  a period as ending an abbreviation instead of ending
a sentence, as in <literal>So says Prof.\ Smith</literal> (see <link linkend="_005c_0040">\@</link>).
</para>
<para>Under normal circumstances, <literal>\</literal><keycap>tab</keycap> and <literal>\</literal><keycap>newline</keycap>
are equivalent to backslash-space, <literal>\ </literal>.
</para>
<!-- @PkgIndex{xspace} -->
<!-- Some individual commands, notably those defined with the @code{xspace}, -->
<!-- package do not follow the standard behavior. -->


</sect1>
<sect1 label="19.7" id="_007e">
<title><literal>~</literal></title>

<indexterm role="fn"><primary>~</primary></indexterm>
<indexterm role="cp"><primary>tie</primary></indexterm>
<indexterm role="cp"><primary>space, hard</primary></indexterm>
<indexterm role="cp"><primary>space, unbreakable</primary></indexterm>
<indexterm role="cp"><primary>NBSP</primary></indexterm>

<para>Synopsis:
</para>
<screen><replaceable>before</replaceable>~<replaceable>after</replaceable>
</screen>
<para>The tie character, <literal>~</literal>, produces a space between <replaceable>before</replaceable> and
<replaceable>after</replaceable> at which the line will not be broken. By default the white
space has length 3.33333pt plus 1.66666pt minus
1.11111pt (see <link linkend="Lengths">Lengths</link>).
</para>
<para>Here &latex; will not break the line between the final two words.
</para>
<screen>Thanks to Prof.~Lerman.
</screen>
<para>In addition, despite the period, &latex; does not use the
end-of-sentence spacing (see <link linkend="_005c_0040">\@</link>).
</para>
<para>Ties prevent the end of line separation of things where that could cause
confusion. But they also reduce &latex;&#8217;s options when it breaks lines
into paragraphs, so you can use too many. They are also matters of
taste, sometimes alarmingly dogmatic taste, among readers. Nevertheless,
here are some usage models, many of them from the &tex;book.
</para>
<itemizedlist><listitem><para>Between an enumerator and its item, such as in references:
<literal>Chapter~12</literal>, or <literal>Theorem~\ref{th:Wilsons}</literal>, or
<literal>Figure~\ref{fig:KGraph}</literal>. When cases are enumerated inline:
<literal>(b)~Show that $f(x)$ is (1)~continuous, and (2)~bounded</literal>.
</para>
<indexterm role="cp"><primary>package, <literal>siunitx</literal></primary></indexterm>
<indexterm role="cp"><primary><literal>siunitx</literal> package</primary></indexterm>

</listitem><listitem><para>Between a number and its unit: <literal>$745.7.8$~watts</literal> (the
<filename>siunitx</filename> package has a special facility for this) or
<literal>144~eggs</literal>. This includes between a month and a date:
<literal>October~12</literal> or <literal>12~Oct</literal>. In general, in any expressions where
numbers and abbreviations or symbols are separated by a space:
<literal>AD~565</literal>, or <literal>2:50~pm</literal>, or <literal>Boeing~747</literal>, or
<literal>268~Plains Road</literal>, or <literal>\$$1.4$~billion</literal>.
</para>
</listitem><listitem><para>When mathematical phrases are rendered in words: <literal>equals~$n$</literal>, or
<literal>less than~$\epsilon$</literal>, or <literal>given~$X$</literal>, or <literal>modulo~$p^e$
for all large~$n$</literal> (but compare <literal>is~$15$</literal> with <literal>is $15$~times
the height</literal>). Between mathematical symbols in apposition with nouns:
<literal>dimension~$d$</literal> or <literal>function~$f(x)$</literal> (but compare <literal>with
length $l$~or more</literal>). When a symbol is a tightly bound object of a
preposition: <literal>of~$x$</literal>, or <literal>from $0$ to~$1$</literal>, or <literal>in
common with~$m$</literal>.
</para>
</listitem><listitem><para>Between symbols in series: <literal>$1$,~$2$, or~$3$</literal> or <literal>$1$,~$2$,
\ldots,~$n$</literal>.
</para>
</listitem><listitem><para>Between a person&#8217;s forenames and between multiple surnames:
<literal>Donald~E. Knuth</literal>, or <literal>Luis~I. Trabb~Pardo</literal>, or
<literal>Charles~XII</literal> (but you must give TeX places to break the line so
you may do <literal>Charles Louis Xavier~Joseph de~la Vall\'ee~Poussin</literal>).
</para>
</listitem><listitem><para>Before a dash: <literal>pages 12~--14</literal> or <literal>it is~--- it must be
said~--- plausible</literal>.
</para>
</listitem></itemizedlist>

</sect1>
<sect1 label="19.8" id="_005cthinspace-_0026-_005cnegthinspace">
<title><literal>\thinspace</literal> &amp; <literal>\negthinspace</literal></title>

<indexterm role="fn"><primary>\thinspace</primary></indexterm>
<indexterm role="fn"><primary>\negthinspace</primary></indexterm>
<indexterm role="cp"><primary>thin space</primary></indexterm>
<indexterm role="cp"><primary>space, thin</primary></indexterm>
<indexterm role="cp"><primary>thin space, negative</primary></indexterm>
<indexterm role="cp"><primary>space, negative thin</primary></indexterm>

<para>Synopsis, one of:
</para>
<screen>\thinspace
\negthinspace
</screen>
<para>Produce an unbreakable and unstretchable space of 1/6em and
-1/6em.  These are the text mode equivalents of <literal>\,</literal> and
<literal>\!</literal> (see <link linkend="Spacing-in-math-mode_002f_005cthinspace">Spacing in math mode/\thinspace</link>).  You can use
<literal>\,</literal> as a synonym for <literal>\thinspace</literal> in text mode.
</para>
<para>The <literal>\negthinspace</literal> command is used in text mode mostly for
fiddling with spaces.  One common use of <literal>\thinspace</literal> is as the
space between nested quotes.
</para>
<screen>Killick replied, ``I heard the Captain say, `Ahoy there.'\thinspace''
</screen>
<para>Another use is that some style guides call for a <literal>\thinspace</literal>
between an ellipsis and a sentence ending period (other style guides,
though, think the three dots are quite enough already).  Still another
use is between initials, as in <literal>D.\thinspace E.\ Knuth</literal>.
</para>

</sect1>
<sect1 label="19.9" id="_005c_002f">
<title><literal>\/</literal></title>

<indexterm role="fn"><primary>\/</primary></indexterm>
<indexterm role="cp"><primary>italic correction</primary></indexterm>

<para>Synopsis:
</para>
<screen><replaceable>before-character</replaceable>\/<replaceable>after-character</replaceable>
</screen>
<para>Insert an <firstterm>italic correction</firstterm>, a small space defined by the font
designer for each character, to avoid the character colliding with
whatever follows. When you use <literal>\/</literal>, &latex; takes the correction
from the font metric file, scales it by any scaling that has been
applied to the font, and then inserts that much horizontal space.
</para>
<para>Here, were it not for the <literal>\/</literal>, the <replaceable>before-character</replaceable>
italic&#160;f would hit the <replaceable>after-character</replaceable> roman&#160;H
</para>
<screen>\newcommand{\companylogo}{{\it f}\/H}
</screen>
<para>because the italic letter leans far to the right.
</para>
<para>If <replaceable>after-character</replaceable> is a period or comma then don&#8217;t insert an
italic correction since those punctuation symbols have a very small
height.  However, with semicolons or colons as well as with normal
letters, the italic correction can help.
</para>
<para>When you use commands such as <literal>\textit</literal> or <literal>\itshape</literal> to
change fonts, &latex; will automatically insert any needed italic
correction (see <link linkend="Font-styles">Font styles</link>).
</para>
<para>Roman characters can also have an italic correction.  An example is in
the name <literal>pdf\/\TeX</literal>.
</para>
<para>There is no concept of italic correction in math mode; spacing is done
in a different way.
</para>

</sect1>
<sect1 label="19.10" id="_005chrulefill-_0026-_005cdotfill">
<title><literal>\hrulefill</literal> &amp; <literal>\dotfill</literal></title>

<indexterm role="fn"><primary>\hrulefill</primary></indexterm>
<indexterm role="fn"><primary>\dotfill</primary></indexterm>

<para>Synopsis, one of:
</para>
<screen>\hrulefill
\dotfill
</screen>
<para>Produce an infinite horizontal rubber length (see <link linkend="Lengths">Lengths</link>) that
&latex; fills with a rule (that is, a line) or with dots, instead of
white space.
</para>
<para>This outputs a line 2 inches long.
</para>
<screen>Name:~\makebox[2in]{\hrulefill}
</screen>
<para>This example, when placed between blank lines, creates a paragraph that
is left and right justified and where the middle is filled with evenly
spaced dots.
</para>
<screen>\noindent John Aubrey, RN \dotfill{} Melbury Lodge
</screen>
<para>To make the rule or dots go to the line&#8217;s end use <literal>\null</literal> at the
start or end.  
</para>
<para>To change the rule&#8217;s thickness, copy the definition and adjust it, as
here
</para>
<screen>\renewcommand{\hrulefill}{%
  \leavevmode\leaders\hrule height 1pt\hfill\kern\z@}
</screen>
<para>which changes the default thickness of 0.4pt to 1pt.
Similarly, adjust the dot spacing as with
</para>
<screen>\renewcommand{\dotfill}{%
  \leavevmode\cleaders\hb@xt@1.00em{\hss .\hss }\hfill\kern\z@}
</screen>
<para>which changes the default length of 0.33em to 1.00em.
</para>
<para>This example produces a line for a signature.
</para>
<screen>\begin{minipage}{4cm}
  \centering
  \hrulefill\\
  Signed
\end{minipage}
</screen>
<para>The line is 4cm long.
</para>

</sect1>
<sect1 label="19.11" id="_005cbigskip-_0026-_005cmedskip-_0026-_005csmallskip">
<title><literal>\bigskip</literal> &amp; <literal>\medskip</literal> &amp; <literal>\smallskip</literal></title>

<para>Synopsis, one of:
</para>
<screen>\bigskip
\medskip
\smallskip
</screen>
<para>Produce an amount of vertical space, large or medium-sized or
small. These commands are fragile (see <link linkend="_005cprotect">\protect</link>).
</para>
<para>Here the skip suggests the passage of time (from <emphasis>The Golden Ocean</emphasis> by
O&#8217;Brian).
</para>
<screen>Mr Saumarez would have something rude to say to him, no doubt: he
was at home again, and it was delightful.

\bigskip
``A hundred and fifty-seven miles and one third, in twenty-four hours,''
said Peter.
</screen>
<para>Each command is associated with a length defined in the document class
file.
</para>
<variablelist><anchor id="bigskip"/><varlistentry><term><indexterm role="fn"><primary>\bigskip</primary></indexterm><literal>\bigskip</literal>
</term><listitem><indexterm role="fn"><primary>\bigskipamount</primary></indexterm>
<para>The same as <literal>\vspace{\bigskipamount}</literal>, ordinarily about one line
space, with stretch and shrink.  The default for the <literal>book</literal> and
<literal>article</literal> classes is <literal>12pt plus 4pt minus 4pt</literal>.
</para>
<anchor id="medskip"/></listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\medskip</primary></indexterm><literal>\medskip</literal>
</term><listitem><indexterm role="fn"><primary>\medskipamount</primary></indexterm>
<para>The same as <literal>\vspace{\medskipamount}</literal>, ordinarily about half of a
line space, with stretch and shrink.  The default for the <literal>book</literal>
and <literal>article</literal> classes is <literal>6pt plus 2pt minus 2pt</literal>.
</para>
<anchor id="smallskip"/></listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\smallskip</primary></indexterm><literal>\smallskip</literal>
</term><listitem><indexterm role="fn"><primary>\smallskipamount</primary></indexterm>
<para>The same as <literal>\vspace{\smallskipamount}</literal>, ordinarily about a
quarter of a line space, with stretch and shrink.  The default for the
<literal>book</literal> and <literal>article</literal> classes is <literal>3pt plus 1pt minus 1pt</literal>.
</para>
</listitem></varlistentry></variablelist>
<para>Because each command is a <literal>\vspace</literal>, if you use on in mid-paragraph
then it will insert its vertical space between the line in which you use
it and the next line, not necessarily at the place that you use it.  So
these are best between paragraphs.
</para>
<para>The commands <literal>\bigbreak</literal>, <literal>\medbreak</literal>, and <literal>\smallbreak</literal>
are similar but also suggest to &latex; that this is a good place to
put a page break (see <link linkend="_005cbigbreak-_0026-_005cmedbreak-_0026-_005csmallbreak">\bigbreak &amp; \medbreak &amp; \smallbreak</link>.
</para>

</sect1>
<sect1 label="19.12" id="_005cbigbreak-_0026-_005cmedbreak-_0026-_005csmallbreak">
<title><literal>\bigbreak</literal> &amp; <literal>\medbreak</literal> &amp; <literal>\smallbreak</literal></title>

<para>Synopsis, one of:
</para>
<screen>\bigbreak
\medbreak
\smallbreak
</screen>
<para>Produce a vertical space that is big or medium-sized or small, and
suggest to &latex; that this is a good place to break the page.  (The
associated penalties are -200, -100, and -50.)
</para>
<para>See <link linkend="_005cbigskip-_0026-_005cmedskip-_0026-_005csmallskip">\bigskip &amp; \medskip &amp; \smallskip</link>, for more.  These commands
produce the same vertical space but differ in that they also remove a
preceding vertical space if it is less than what they would insert (as
with <literal>\addvspace</literal>).  In addition, they terminate a paragraph where
they are used: this example
</para>
<screen>abc\bigbreak def ghi

jkl mno pqr
</screen>
<para>will output three paragraphs, the first ending in &#8216;<literal>abc</literal>&#8217; and the
second starting, after an extra vertical space and a paragraph indent,
with &#8216;<literal>def</literal>&#8217;.
</para>

</sect1>
<sect1 label="19.13" id="_005cstrut">
<title><literal>\strut</literal></title>

<indexterm role="fn"><primary>\strut</primary></indexterm>
<indexterm role="cp"><primary>strut</primary></indexterm>

<para>Synopsis:
</para>
<screen>\strut
</screen>
<para>Ensure that the current line has height at least <literal>0.7\baselineskip</literal>
and depth at least <literal>0.3\baselineskip</literal>.  Essentially, &latex;
inserts into the line a rectangle having zero width,
<literal>\rule[-0.3\baselineskip]{0pt}{\baselineskip}</literal> (see <link linkend="_005crule">\rule</link>).
The <literal>\baselineskip</literal> changes with the current font and fontsize.
</para>
<para>In this example the <literal>\strut</literal> keeps the box inside the frame from
having zero height.
</para>
<screen>\setlength{\fboxsep}{0pt}\framebox[2in]{\strut}
</screen>
<para>This example has four lists.  In the first there is a much bigger gap
between items 2 and&#160;3 than there is between items 1 and&#160;2.
The second list fixes that with a <literal>\strut</literal> at the end of its first
item&#8217;s second line.
</para>
<screen>\setlength{\fboxsep}{0pt}
\noindent\begin{minipage}[t]{0.2\linewidth}
\begin{enumerate}
  \item \parbox[t]{15pt}{test \\ test}
  \item test
  \item test
\end{enumerate}
\end{minipage}%
\begin{minipage}[t]{0.2\linewidth}
\begin{enumerate}
  \item \parbox[t]{15pt}{test \\ test\strut}
  \item test
  \item test
\end{enumerate}
\end{minipage}%
\begin{minipage}[t]{0.2\linewidth}
\begin{enumerate}
  \item \fbox{\parbox[t]{15pt}{test \\ test}}
  \item \fbox{test}
  \item \fbox{test}
\end{enumerate}
\end{minipage}%
\begin{minipage}[t]{0.2\linewidth}
\begin{enumerate}
  \item \fbox{\parbox[t]{15pt}{test \\ test\strut}}
  \item \fbox{test}
  \item \fbox{test}
\end{enumerate}
\end{minipage}%
</screen>
<para>The final two lists use <literal>fbox</literal> to show what&#8217;s happening.  The third
list&#8217;s <literal>\parbox</literal> goes only to the bottom of its second &#8216;<literal>test</literal>&#8217;,
which happens not have any characters that descend below the baseline.
The fourth list adds the strut that gives the needed extra
below-baseline space.
</para>
<indexterm role="cp"><primary>package, <literal>TikZ</literal></primary></indexterm>
<indexterm role="cp"><primary><literal>TikZ</literal> package</primary></indexterm>
<indexterm role="cp"><primary>package, <literal>Asymptote</literal></primary></indexterm>
<indexterm role="cp"><primary><literal>Asymptote</literal> package</primary></indexterm>

<para>The <literal>\strut</literal> command is often useful in graphics, such as in
<filename>TikZ</filename> or <filename>Asymptote</filename>.  For instance, you may have a command
such as <literal>\graphnode{<replaceable>node-name</replaceable>}</literal> that fits a circle around
<replaceable>node-name</replaceable>.  However, unless you are careful the <replaceable>node-name</replaceable>&#8217;s
&#8216;<literal>x</literal>&#8217; and &#8216;<literal>y</literal>&#8217; will produce different-diameter circles because
the characters are different sizes.  A careful <literal>\graphnode</literal> might
insert a <literal>\strut</literal>, then <replaceable>node-name</replaceable>, and then draw the circle.
</para>
<para>The general approach of using a zero width <literal>\rule</literal> is useful in
many circumstances.  In this table, the zero-width rule keeps the top of
the first integral from hitting the <literal>\hline</literal>.  Similarly, the
second rule keeps the second integral from hitting the first.
</para>
<screen>\begin{tabular}{rl}
  \textsc{Integral}   &amp;\textsc{Value}           \\
  \hline
  $\int_0^x t\, dt$   &amp;$x^2/2$  \rule{0em}{2.5ex} \\  
  $\int_0^x t^2\, dt$ &amp;$x^3/3$  \rule{0em}{2.5ex}
\end{tabular}
</screen>
<para>(Although the line-ending double backslash command has an available
optional argument to put in more vertical room, that won&#8217;t work here.
Changing the first double backslash to something like <literal>\\[2.5ex]</literal>
will put the room between the header line and the <literal>\hline</literal>, and the
integral would still hit the line.)
</para>

</sect1>
<sect1 label="19.14" id="_005cvspace">
<title><literal>\vspace</literal></title>

<indexterm role="fn"><primary>\vspace</primary></indexterm>
<indexterm role="cp"><primary>vertical space</primary></indexterm>
<indexterm role="cp"><primary>space, vertical</primary></indexterm>

<para>Synopsis, one of:
</para>
<screen>\vspace{<replaceable>length</replaceable>}
\vspace*{<replaceable>length</replaceable>}
</screen>
<para>Add the vertical space <replaceable>length</replaceable>.  The <replaceable>length</replaceable> can be positive,
negative, or zero.  It is a rubber length&#8212;it may contain a <literal>plus</literal>
or <literal>minus</literal> component (see <link linkend="Lengths">Lengths</link>).
</para>
<para>This puts space between the two paragraphs.
</para>
<screen>And I slept.

\vspace{1ex plus 0.5ex}
The new day dawned cold.
</screen>
<para>(See <link linkend="_005cbigskip-_0026-_005cmedskip-_0026-_005csmallskip">\bigskip &amp; \medskip &amp; \smallskip</link> for common inter-paragraph
spaces.)
</para>
<para>The <literal>*</literal>-version inserts vertical space that non-discardable.  More
precisely, &latex; discards vertical space at a page break and the
<literal>*</literal>-version causes the space to stay.  This example leaves space
between the two questions.
</para>
<screen>Question: Find the integral of \( 5x^4+5 \).

\vspace*{2cm plus 0.5cm}
Question: Find the derivative of \( x^5+5x+9 \).
</screen>
<para>That space will be present even if the page break happens to fall
between the questions.
</para>
<para>If you use <literal>\vspace</literal> in the middle of a paragraph (i.e., in
horizontal mode) then the space is inserted after the line containing
the <literal>\vspace</literal> command; it does not start a new paragraph at the
<literal>\vspace</literal> command.
</para>
<para>In this example the two questions will be evenly spaced vertically on
the page, with at least one inch of space below each.
</para>
<screen>\begin{document}
1) Who put the bomp in the bomp bah bomp bah bomp?
\vspace{1in plus 1fill}

2) Who put the ram in the rama lama ding dong?
\vspace{1in plus 1fill}
\end{document}
</screen>

</sect1>
<sect1 label="19.15" id="_005cvfill">
<title><literal>\vfill</literal></title>

<indexterm role="fn"><primary>\vfill</primary></indexterm>

<indexterm role="cp"><primary>stretch, infinite vertical</primary></indexterm>
<indexterm role="cp"><primary>infinite vertical stretch</primary></indexterm>

<para>Synopsis:
</para>
<screen>\vfill
</screen>
<para>End the current paragraph and insert a vertical rubber length that is
infinite, so it can stretch or shrink as far as needed
(see <link linkend="Lengths">Lengths</link>).
</para>
<para>It is often used in the same way as <literal>\vspace{\fill}</literal>, except that
<literal>\vfill</literal> ends the current paragraph whereas <literal>\vspace{\fill}</literal>
adds the infinite vertical space below its line, irrespective of the
paragraph structure.  In both cases that space will disappear at a page
boundary; to circumvent this see the starred option
in&#160;<link linkend="_005cvspace">\vspace</link>.
</para>
<para>In this example the page is filled, so the top and bottom lines contain
the text &#8216;<literal>Lost Dog!</literal>&#8217; and the second &#8216;<literal>Lost Dog!</literal>&#8217; is exactly
halfway between them.
</para> 
<screen>\begin{document}
Lost Dog!
\vfill
Lost Dog!  % perfectly in the middle
\vfill
Lost Dog!
\end{document}
</screen>

</sect1>
<sect1 label="19.16" id="_005caddvspace">
<title><literal>\addvspace</literal></title>

<indexterm role="fn"><primary>\addvspace</primary></indexterm>
<indexterm role="cp"><primary>vertical space</primary></indexterm>
<indexterm role="cp"><primary>space, inserting vertical</primary></indexterm>

<para>Synopsis:
</para>
<screen>\addvspace{<replaceable>vert-length</replaceable>}
</screen>
<para>Add a vertical space of <replaceable>vert-length</replaceable>.  However, if there are two or
more <literal>\addvspace</literal>&#8217;s in a sequence then together they only add the
space needed to make the natural length equal to the maximum of the
<replaceable>vert-length</replaceable>&#8217;s in that sequence. This command is fragile
(see <link linkend="_005cprotect">\protect</link>).  The <replaceable>vert-length</replaceable> is a rubber length
(see <link linkend="Lengths">Lengths</link>).
</para>
<para>This example illustrates.  The <literal>picture</literal> draws a scale.  In a
standard &latex; article the length <literal>\baselineskip</literal> is 12pt.
The two rules here are 22pt apart: the sum of the
<literal>\baselineskip</literal> and the 10pt from the first <literal>addvspace</literal>.
</para>
<screen>\documentclass{article}
\usepackage{color}
\begin{document}
\setlength{\unitlength}{2pt}%
\noindent\begin{picture}(0,0)%
  \multiput(0,0)(0,-1){25}{{\color{blue}\line(1,0){1}}}
  \multiput(0,0)(0,-5){6}{{\color{red}\line(1,0){2}}}
\end{picture}%
\rule{0.25\linewidth}{0.1pt}% 
\par\addvspace{10pt}% \addvspace{20pt}%
\par\noindent\rule{0.25\linewidth}{0.1pt}%
\end{document}
</screen>
<para>Now uncomment the second <literal>\addvspace</literal>.  It does not make the gap
20pt longer; instead the gap is the sum of <literal>\baselineskip</literal>
and 20pt.  So <literal>\addvspace</literal> in a sense does the opposite of
its name &#8212; it makes sure that multiple vertical spaces do not
accumulate, but instead that only the largest one is used.
</para>
<para>&latex; uses this command to adjust the vertical space above or below
an environment that starts a new paragraph.  For instance, a
<literal>theorem</literal> environment begins and ends with <literal>\addvspace</literal> so
that two consecutive <literal>theorem</literal>&#8217;s are separated by one vertical
space, not two.
</para>
<para>A error &#8216;<literal>Something's wrong--perhaps a missing \item</literal>&#8217; pointing to an
<literal>\addvspace</literal> means that you were not in vertical mode when you hit
this command. One way to change that is to precede <literal>\addvspace</literal>
with a <literal>\par</literal> command (see <link linkend="_005cpar">\par</link>), as in the above example.
</para>

</sect1>
</chapter>
<chapter label="20" id="Boxes">
<title>Boxes</title>

<indexterm role="cp"><primary>boxes</primary></indexterm>

<para>At its core, &latex; puts things in boxes and then puts the boxes on a
page.  So these commands are central.
</para>
<para>There are many packages on CTAN that are useful for manipulating boxes.
One useful adjunct to the commands here is <filename>adjustbox</filename>.
</para>


<sect1 label="20.1" id="_005cmbox-_0026-_005cmakebox">
<title><literal>\mbox</literal> &amp; <literal>\makebox</literal></title>

<indexterm role="fn"><primary>\mbox</primary></indexterm>
<indexterm role="fn"><primary>\makebox</primary></indexterm>
<indexterm role="cp"><primary>box</primary></indexterm>
<indexterm role="cp"><primary>make a box</primary></indexterm>
<indexterm role="cp"><primary>hyphenation, preventing</primary></indexterm>

<para>Synopsis, one of:
</para>
<screen>\mbox{<replaceable>text</replaceable>}
\makebox{<replaceable>text</replaceable>}
\makebox[<replaceable>width</replaceable>]{<replaceable>text</replaceable>}
\makebox[<replaceable>width</replaceable>][<replaceable>position</replaceable>]{<replaceable>text</replaceable>}
</screen>
<para>Create a box, a container for material.  The <replaceable>text</replaceable> is is typeset in
LR mode (see <link linkend="Modes">Modes</link>) so it is not broken into lines.  The
<literal>\mbox</literal> command is robust, while <literal>\makebox</literal> is fragile
(see <link linkend="_005cprotect">\protect</link>).
</para>
<para>Because <literal>text</literal> is not broken into lines, you can use <literal>\mbox</literal>
to prevent hyphenation.  In this example, &latex; will not hyphenate
the table name, &#8216;<literal>T-4</literal>&#8217;.
</para>
<screen>See Table~\mbox{T-4}
</screen>
<para>The first two command versions, <literal>\mbox</literal> and <literal>\makebox</literal>, are
roughly equivalent.  They create a box just wide enough to contain the
<replaceable>text</replaceable>.  (They are like plain &tex;&#8217;s <literal>\hbox</literal>.)
</para>
<para>In the third version the optional argument <replaceable>width</replaceable> specifies the
width of the box.  Note that the space occupied by the text need not
equal the width of the box.  For one thing, <replaceable>text</replaceable> can be too small;
this creates a full-line box
</para>
<screen>\makebox[\linewidth]{Chapter Exam}
</screen>
<para>with &#8216;<literal>Chapter Exam</literal>&#8217; centered.  But <replaceable>text</replaceable> can also be too wide
for <replaceable>width</replaceable>.  See the example below of zero-width boxes.
</para>
<anchor id="mbox-makebox-depth"/><anchor id="mbox-makebox-height"/><anchor id="mbox-makebox-width"/><anchor id="mbox-makebox-totalheight"/><para>In the <replaceable>width</replaceable> argument you can use the following lengths that refer
to the dimension of the box that &latex; gets on typesetting
<replaceable>text</replaceable>: <literal>\depth</literal>, <literal>\height</literal>, <literal>\width</literal>,
<literal>\totalheight</literal> (this is the box&#8217;s height plus its depth).  For
example, to make a box with the text stretched to double the natural
size you can say this.
</para>
<screen>\makebox[2\width]{Get a stretcher}
</screen>
<para>For the fourth command version the optional argument <replaceable>position</replaceable>
gives position of the text within the box. It may take the following
values:
</para>
<variablelist><varlistentry><term><literal>c</literal>
</term><listitem><para>The <replaceable>text</replaceable> is centered (default).
</para>
</listitem></varlistentry><varlistentry><term><literal>l</literal>
</term><listitem><para>The <replaceable>text</replaceable> is flush left.
</para>
</listitem></varlistentry><varlistentry><term><literal>r</literal>
</term><listitem><para>Flush right.
</para>
</listitem></varlistentry><varlistentry><term><literal>s</literal>
</term><listitem><para>Stretch the interword space in <replaceable>text</replaceable> across the entire <replaceable>width</replaceable>.
The <replaceable>text</replaceable> must contain stretchable space for this to work.  For
instance, this could head a press release:
<literal>\noindent\makebox[\textwidth][s]{\large\hfil IMMEDIATE\hfil
RELEASE\hfil}</literal>
</para></listitem></varlistentry></variablelist>
<para>A common use of <literal>\makebox</literal> is to make zero-width text boxes.  This
puts the value of the quiz questions to the left of those questions.
</para>
<screen>\newcommand{\pts}[1]{\makebox[0em][r]{#1 points\hspace*{1em}}}
\pts{10}What is the air-speed velocity of an unladen swallow?

\pts{90}An African or European swallow?
</screen>
<para><indexterm role="cp"><primary>package, <literal>TikZ</literal></primary></indexterm>
<indexterm role="cp"><primary><literal>TikZ</literal> package</primary></indexterm>
</para>
<indexterm role="cp"><primary>package, <literal>Asymptote</literal></primary></indexterm>
<indexterm role="cp"><primary><literal>Asymptote</literal> package</primary></indexterm>

<para>The right edge of the output &#8216;<literal>10 points </literal>&#8217; (note the ending space)
will be just before the &#8216;<literal>What</literal>&#8217; (note the space after
&#8216;<literal>points</literal>&#8217;).  You can use <literal>\makebox</literal> similarly when making
graphics, such as in <filename>TikZ</filename> or <filename>Asymptote</filename>, where you put the
edge of the text at a known location, regardless of the length of that
text.
</para>
<para>For boxes with frames see&#160;<link linkend="_005cfbox-_0026-_005cframebox">\fbox &amp; \framebox</link>. For colors
see&#160;<link linkend="Colored-boxes">Colored boxes</link>.
</para>
<para>There is a related version of <literal>\makebox</literal> that is used within the
<literal>picture</literal> environment, where the length is given in terms of
<literal>\unitlength</literal> (see <link linkend="_005cmakebox-_0028picture_0029">\makebox (picture)</link>).
</para>
<para>If you put a double-backslash into <replaceable>text</replaceable> then &latex; will not
give you a new line; for instance <literal>\makebox{abc def \\ ghi}</literal>
outputs &#8216;<literal>abc defghi</literal>&#8217; while <literal>\makebox{abc def \par ghi}</literal>
outputs &#8216;<literal>abc def ghi</literal>&#8217;, but neither go to a second line.  To get
multiple lines see&#160;<link linkend="_005cparbox">\parbox</link> and&#160;<link linkend="minipage">minipage</link>.
</para>

</sect1>
<sect1 label="20.2" id="_005cfbox-_0026-_005cframebox">
<title><literal>\fbox</literal> &amp; <literal>\framebox</literal></title>

<indexterm role="fn"><primary>\fbox</primary></indexterm>
<indexterm role="fn"><primary>\framebox</primary></indexterm>

<para>Synopses, one of:
</para>
<screen>\fbox{<replaceable>text</replaceable>}
\framebox{<replaceable>text</replaceable>}
\framebox[<replaceable>width</replaceable>]{<replaceable>text</replaceable>}
\framebox[<replaceable>width</replaceable>][<replaceable>position</replaceable>]{<replaceable>text</replaceable>}
</screen>
<para>Create a box with an enclosing frame, four lines surrounding the space.
These commands are the same as <literal>\mbox</literal> and <literal>\makebox</literal> except
for the frame (see <link linkend="_005cmbox-_0026-_005cmakebox">\mbox &amp; \makebox</link>). The <literal>\fbox</literal> command is
robust, the <literal>\framebox</literal> command is fragile (see <link linkend="_005cprotect">\protect</link>).
</para>
<screen>\fbox{Warning! No work shown, no credit given.}
</screen>
<para>&latex; puts the text into a box that cannot be split or hyphenated.
Around that box, separated from it by a small gap, are four lines making
a frame.
</para>
<para>The first two command invocations, <literal>\fbox{...}</literal> and
<literal>\framebox{...}</literal>, are roughly the same.  As to the third and
fourth invocations, the optional arguments allow you to specify the box
width as <replaceable>width</replaceable> and the position of the text inside that box as
<replaceable>position</replaceable>.  See <link linkend="_005cmbox-_0026-_005cmakebox">\mbox &amp; \makebox</link> for the full description but
here is an example creating an empty box that is 1/4in wide.
</para>
<screen>\setlength{\fboxsep}{0pt}\framebox[0.25in]{\strut}}
</screen>
<para>The <literal>\strut</literal> inserts a vertical height of <literal>\baselineskip</literal>
(see <link linkend="_005cstrut">\strut</link>).
</para>
<para>These parameters determine the frame layout.
</para>
<variablelist><anchor id="fbox-framebox-fboxrule"/><varlistentry><term><indexterm role="fn"><primary>\fboxrule</primary></indexterm><literal>\fboxrule</literal>
</term><listitem><indexterm role="fn"><primary>frame, line width</primary></indexterm>
<indexterm role="fn"><primary>frame rule width</primary></indexterm>
<indexterm role="cp"><primary>\fboxrule</primary></indexterm>
<para>The thickness of the lines around the enclosed box.  The default is
0.2pt.  Change it with a command such as
<literal>\setlength{\fboxrule}{0.8pt}</literal> (see <link linkend="_005csetlength">\setlength</link>).
</para>
<anchor id="fbox-framebox-fboxsep"/></listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\fboxsep</primary></indexterm><literal>\fboxsep</literal>
</term><listitem><indexterm role="fn"><primary>frame, separation from contents</primary></indexterm>
<indexterm role="cp"><primary>\fboxsep</primary></indexterm>
<para>The distance from the frame to the enclosed box. The default is 3pt.
Change it with a command such as <literal>\setlength{\fboxsep}{0pt}</literal>
(see <link linkend="_005csetlength">\setlength</link>).  Setting it to 0pt is useful sometimes:
this will put a frame around the picture with no white border.
</para>
<screen>{\setlength{\fboxsep}{0pt}
 \framebox{%
   \includegraphics[width=0.5\textwidth]{prudence.jpg}}}
</screen>
<para>The extra curly braces keep the effect of the <literal>\setlength</literal> local.
</para>
</listitem></varlistentry></variablelist>
<para>As with <literal>\mbox</literal> and <literal>\makebox</literal>, &latex; will not break lines
in <replaceable>text</replaceable>.  But this example has &latex; break lines to make a
paragraph, and then frame the result.
</para>
<screen>\framebox{%
  \begin{minipage}{0.6\linewidth}
    My dear, here we must run as fast as we can, just to stay in place.
    And if you wish to go anywhere you must run twice as fast as that.
  \end{minipage}}
</screen>
<para>See <link linkend="Colored-boxes">Colored boxes</link> for colors other than black and white.
</para>
<para>The <literal>picture</literal> environment has a version of this command where the
units depend on <literal>picture</literal>&#8217;s <literal>\unitlength</literal> (see <link linkend="_005cframebox-_0028picture_0029">\framebox
(picture)</link>).
</para>

</sect1>
<sect1 label="20.3" id="_005cparbox">
<title><literal>\parbox</literal></title>

<indexterm role="fn"><primary>\parbox</primary></indexterm>
<indexterm role="cp"><primary>paragraph mode</primary></indexterm>
<indexterm role="cp"><primary>paragraph, in a box</primary></indexterm>

<para>Synopses, one of:
</para>
<screen>\parbox{<replaceable>width</replaceable>}{<replaceable>contents</replaceable>}
\parbox[<replaceable>position</replaceable>]{<replaceable>width</replaceable>}{<replaceable>contents</replaceable>}
\parbox[<replaceable>position</replaceable>][<replaceable>height</replaceable>]{<replaceable>width</replaceable>}{<replaceable>contents</replaceable>}
\parbox[<replaceable>position</replaceable>][<replaceable>height</replaceable>][<replaceable>inner-pos</replaceable>]{<replaceable>width</replaceable>}{<replaceable>contents</replaceable>}
</screen>
<para>Produce a box of text that is <replaceable>width</replaceable> wide. Use this command to make
a box of small pieces of text, of a single paragraph.  This command is
fragile (see <link linkend="_005cprotect">\protect</link>).
</para>
<screen>\begin{picture}(0,0)
  ...
  \put(1,2){\parbox{1.75in}{\raggedright Because the graph is a line on
                         this semilog paper, the relationship is
                         exponential.}}
\end{picture}
</screen>
<para>The <replaceable>contents</replaceable> are processed in a text mode (see <link linkend="Modes">Modes</link>) so
&latex; will break lines to make a paragraph.  But it won&#8217;t make
multiple paragraphs; for that, use a <literal>minipage</literal> environment
(see <link linkend="minipage">minipage</link>).
</para>
<para>The options for <literal>\parbox</literal> (except for <replaceable>contents</replaceable>) are the same
as those for <literal>minipage</literal>.  For convenience a summary of the options
is here but see&#160;<link linkend="minipage">minipage</link> for a complete description.
</para>
<para>There are two required arguments.  The <replaceable>width</replaceable> is a rigid length
(see <link linkend="Lengths">Lengths</link>).  It sets the width of the box into which &latex;
typesets <replaceable>contents</replaceable>.  The <replaceable>contents</replaceable> is the text that is placed
in that box.  It should not have any paragraph-making components.
</para>
<para>There are three optional arguments, <replaceable>position</replaceable>, <replaceable>height</replaceable>, and
<replaceable>inner-pos</replaceable>.  The <replaceable>position</replaceable> gives the vertical alignment of the
<literal>parbox</literal> with respect to the surrounding material.  The possible
values are <literal>c</literal> or <literal>m</literal> to make the vertical center of the
<literal>parbox</literal> lines up with the center of the adjacent line (this is the
default), or <literal>t</literal> to match the top line of the <literal>parbox</literal> with
the baseline of the surrounding material, or <literal>b</literal> to match the
bottom line.
</para>
<para>The optional argument <replaceable>height</replaceable> overrides the natural height of the
box.
</para>
<para>The optional argument <replaceable>inner-pos</replaceable> controls the placement of
<replaceable>content</replaceable> inside the <literal>parbox</literal>. Its default is the value of
<replaceable>position</replaceable>.  Its possible values are: <literal>t</literal> to put the
<replaceable>content</replaceable> at the top of the box, <literal>c</literal> to put it in the vertical
center, <literal>b</literal> to put it at the bottom of the box, and <literal>s</literal> to
stretch it out vertically (for this, the text must contain vertically
stretchable space).
</para>
</sect1>
<sect1 label="20.4" id="_005craisebox">
<title><literal>\raisebox</literal></title>

<indexterm role="fn"><primary>\raisebox</primary></indexterm>

<para>Synopsis, one of:
</para>
<screen>\raisebox{<replaceable>distance</replaceable>}{<replaceable>text</replaceable>}
\raisebox{<replaceable>distance</replaceable>}[<replaceable>height</replaceable>]{<replaceable>text</replaceable>}
\raisebox{<replaceable>distance</replaceable>}[<replaceable>height</replaceable>][<replaceable>depth</replaceable>]{<replaceable>text</replaceable>}
</screen>
<para>Raise or lower <replaceable>text</replaceable>. This command is fragile (see <link linkend="_005cprotect">\protect</link>).
</para>
<para>This example makes a command for the restriction of a function by
lowering the vertical bar symbol.
</para>
<!-- credit: egreg https://tex.stackexchange.com/a/278631/121234 -->
<screen>\newcommand\restricted[1]{\raisebox{-.5ex}{$|$}_{#1}}
$f\restricted{A}$
</screen>
<para>The first mandatory argument <replaceable>distance</replaceable> specifies how far to raise
the second mandatory argument <replaceable>text</replaceable>. This is a rigid length
(see <link linkend="Lengths">Lengths</link>).  If it is negative then it lowers <replaceable>text</replaceable>.  The
<replaceable>text</replaceable> is processed in LR mode so it cannot contain line breaks
(see <link linkend="Modes">Modes</link>).
</para>
<para>The optional arguments <replaceable>height</replaceable> and <replaceable>depth</replaceable> are dimensions.  If
they are specified, they override the natural height and depth of the
box &latex; gets by typesetting <replaceable>text</replaceable>.
</para>
<anchor id="raisebox-depth"/><anchor id="raisebox-height"/><anchor id="raisebox-width"/><anchor id="raisebox-totalheight"/><para>In the arguments <replaceable>distance</replaceable>, <replaceable>height</replaceable>, and <replaceable>depth</replaceable> you can
use the following lengths that refer to the dimension of the box that
&latex; gets on typesetting <replaceable>text</replaceable>: <literal>\depth</literal>, <literal>\height</literal>,
<literal>\width</literal>, <literal>\totalheight</literal> (this is the box&#8217;s height plus its
depth).
</para>
<para>This will align two graphics on their top (see <link linkend="Graphics">Graphics</link>).
</para>
<!-- credit: FAQ https://texfaq.org/FAQ-topgraph -->
<screen>\usepackage{graphicx} \usepackage{calc}  % in preamble
   ...
\begin{center}
  \raisebox{1ex-\height}{%
    \includegraphics[width=0.4\linewidth]{lion.png}}
  \qquad
  \raisebox{1ex-\height}{%
    \includegraphics[width=0.4\linewidth]{meta.png}}
\end{center}
</screen>
<para>The first <literal>\height</literal> is the height of <filename>lion.png</filename> while the
second is the height of <filename>meta.png</filename>.
</para>

</sect1>
<sect1 label="20.5" id="_005csbox-_0026-_005csavebox">
<title><literal>\sbox</literal> &amp; <literal>\savebox</literal></title>

<indexterm role="fn"><primary>\sbox</primary></indexterm>
<indexterm role="fn"><primary>\savebox</primary></indexterm>
<indexterm role="cp"><primary>box, save</primary></indexterm>

<para>Synopsis, one of:
</para>
<screen>\sbox{<replaceable>box-cmd</replaceable>}{<replaceable>text</replaceable>}
\savebox{<replaceable>box-cmd</replaceable>}{<replaceable>text</replaceable>}
\savebox{<replaceable>box-cmd</replaceable>}[<replaceable>width</replaceable>]{<replaceable>text</replaceable>}
\savebox{<replaceable>box-cmd</replaceable>}[<replaceable>width</replaceable>][<replaceable>pos</replaceable>]{<replaceable>text</replaceable>}
</screen>
<para>Typeset <replaceable>text</replaceable> just as with <literal>\makebox</literal> (see <link linkend="_005cmbox-_0026-_005cmakebox">\mbox &amp;
\makebox</link>) except that &latex; does not output it but instead saves it
in a storage bin named <replaceable>box-cmd</replaceable>.  The bin name <replaceable>box-cmd</replaceable> begins
with a backslash, <literal>\</literal>.  You must have previously allocated the bin
<replaceable>box-cmd</replaceable> with <literal>\newsavebox</literal> (see <link linkend="_005cnewsavebox">\newsavebox</link>).The
<literal>\sbox</literal> command is robust while <literal>\savebox</literal> is fragile
(see <link linkend="_005cprotect">\protect</link>).
</para>
<para>This creates and uses a bin.
</para>
<screen>\newsavebox{\fullname}
\sbox{\fullname}{John Jacob Jingleheimer Schmidt}
  ...
\usebox{\fullname}! His name is my name, too!
Whenever we go out, the people always shout!
There goes \\usebox{\fullname}!  Ya da da da da da da.
</screen>
<para>One advantage of using and reusing a bin over a <literal>\newcommand</literal> is
efficiency, that &latex; need not repeatedly retypeset the contents.
See the example below.
</para>
<para>The first two command invocations,
<literal>\sbox{<replaceable>box-cmd</replaceable>}{<replaceable>text</replaceable>}</literal> and
<literal>\savebox{<replaceable>box-cmd</replaceable>}{<replaceable>text</replaceable>}</literal>, are roughly equivalent.
As to the third and fourth, the optional arguments allow you to specify
the box width as <replaceable>width</replaceable>, and the position of the text inside that
box as <replaceable>position</replaceable>.  See <link linkend="_005cmbox-_0026-_005cmakebox">\mbox &amp; \makebox</link> for the full
description.
</para>
<para>In the <literal>\sbox</literal> and <literal>\savebox</literal> commands the <replaceable>text</replaceable> is
typeset in LR mode so it does not have line breaks (see <link linkend="Modes">Modes</link>).  If
you use these then &latex; doesn&#8217;t give you an error but it ignores
what you want: if you enter <literal>\sbox{\newbin}{test \\ test}</literal> and
<literal>\usebox{\newbin}</literal> then you get &#8216;<literal>testtest</literal>&#8217;, while if you
enter <literal>\sbox{\newbin}{test \par test}</literal> and
<literal>\usebox{\newbin}</literal> then you get &#8216;<literal>test test</literal>&#8217;, but no error or
warning.  To fix this use a <literal>\parbox</literal> or <literal>minipage</literal> as here.
</para>
<!-- credit: egreg https://tex.stackexchange.com/a/41668/121234 -->
<screen>\savebox{\abin}{%
  \begin{minipage}{\linewidth}
    \begin{enumerate}
      \item First item
      \item Second item
    \end{enumerate}
  \end{minipage}}
  ...
\usebox{\abin}
</screen>
<para>As an example of the efficiency of reusing a bin&#8217;s contents, this puts
the same picture on each page of the document by putting it in the
header.  &latex; only typesets it once.
</para>
<screen>\usepackage{graphicx}  % all this in the preamble
\newsavebox{\sealbin}
\savebox{\sealbin}{%
  \setlength{\unitlength}{1in}%
  \begin{picture}(0,0)%
     \put(1.5,-2.5){%
       \begin{tabular}{c}
          \includegraphics[height=2in]{companylogo.png} \\
          Office of the President
       \end{tabular}}
  \end{picture}%
}
\markright{\usebox{\sealbin}}
\pagestyle{headings}
</screen>
<para>The <literal>picture</literal> environment is good for fine-tuning the placement.
</para>
<para>If the bin has not already been defined then you get something like
&#8216;<literal>Undefined control sequence. &lt;argument&gt; \nobin</literal>&#8217;.
</para>

</sect1>
<sect1 label="20.6" id="lrbox">
<title><literal>lrbox</literal></title>

<indexterm role="fn"><primary>lrbox</primary></indexterm>

<para>Synopsis:
</para>
<screen>\begin{lrbox}{<replaceable>box-cmd</replaceable>} 
  <replaceable>text</replaceable>
\end{lrbox}
</screen>
<para>The <replaceable>text</replaceable> inside the environment is saved in the bin
<literal><replaceable>box-cmd</replaceable></literal>.  The <replaceable>box-cmd</replaceable> must begin with a
backslash. You must create this bin in advance with <literal>\newsavebox</literal>
(see <link linkend="_005cnewsavebox">\newsavebox</link>).  This is the environment form of the <literal>\sbox</literal>
and <literal>\savebox</literal> commands, and is equivalent to them. See <link linkend="_005csbox-_0026-_005csavebox">\sbox &amp;
\savebox</link> for the full information.
</para>
<para>In this example the environment is convenient for entering the
<literal>tabular</literal>.
</para>
<screen>\newsavebox{\jhbin}
\begin{lrbox}{\jhbin}
  \begin{tabular}{c}
    \includegraphics[height=1in]{jh.png} \\
    Jim Hef{}feron
  \end{tabular}
\end{lrbox}    
  ...
\usebox{\jhbin}
</screen>

</sect1>
<sect1 label="20.7" id="_005cusebox">
<title><literal>\usebox</literal></title>

<indexterm role="fn"><primary>\usebox</primary></indexterm>
<indexterm role="cp"><primary>box, use saved box</primary></indexterm>

<para>Synopsis:
</para>
<screen>\usebox{<replaceable>box-cmd</replaceable>}
</screen>
<para>Produce the box most recently saved in the bin <replaceable>box-cmd</replaceable> by the
commands <literal>\sbox</literal> or <literal>\savebox</literal>, or the <literal>lrbox</literal>
environment. See <link linkend="_005csbox-_0026-_005csavebox">\sbox &amp; \savebox</link> for more information and examples.
(Note that <replaceable>box-cmd</replaceable> starts with a backslash.)  This command is
robust (see <link linkend="_005cprotect">\protect</link>).
</para>

</sect1>
</chapter>
<chapter label="21" id="Color">
<title>Color</title>

<indexterm role="cp"><primary>color</primary></indexterm>

<para>You can add color to text, rules, etc.  You can also have color in a box
or on an entire page and write text on top of it.
</para>
<para>Color support comes as an additional package.  So all the commands below
will only work if your document preamble contains
<literal>\usepackage{color}</literal>, that brings in the standard package.
</para>
<para>Many other packages also supplement &latex;&#8217;s color abilities.
Particularly worth mentioning is <filename>xcolor</filename>, which is widely used and
significantly extends the capabilities described here, including adding
&#8216;<literal>HTML</literal>&#8217; and &#8216;<literal>Hsb</literal>&#8217; color models.
</para>


<sect1 label="21.1" id="Color-package-options">
<title><literal>color</literal> package options</title>

<indexterm role="cp"><primary>color package options</primary></indexterm>
<indexterm role="cp"><primary>options, color package</primary></indexterm>

<para>Synopsis (must be in the document preamble):
</para>
<screen>\usepackage[<replaceable>comma-separated option list</replaceable>]{color}
</screen>
<para>When you load the <filename>color</filename> package there are two kinds of available
options.
</para>
<para>The first specifies the <firstterm>printer driver</firstterm>.  &latex; doesn&#8217;t contain
information about different output systems but instead depends on
information stored in a file.  Normally you should not specify the
driver option in the document, and instead rely on your system&#8217;s
default. One advantage of this is that it makes the document portable
across systems.  For completeness we include a list of the drivers.  The
currently relevant ones are: <filename>dvipdfmx</filename>, <filename>dvips</filename>,
<filename>dvisvgm</filename>, <filename>luatex</filename>, <filename>pdftex</filename>, <filename>xetex</filename>.  The two
<filename>xdvi</filename> and <filename>oztex</filename> are essentially aliases for <filename>dvips</filename>
(and <filename>xdvi</filename> is monochrome).  Ones that should not be used for new
systems are: <filename>dvipdf</filename>, <filename>dvipdfm</filename>, <filename>dviwin</filename>,
<filename>dvipsone</filename>, <filename>emtex</filename>, <filename>pctexps</filename>, <filename>pctexwin</filename>,
<filename>pctexhp</filename>, <filename>pctex32</filename>, <filename>truetex</filename>, <filename>tcidvi</filename>,
<filename>vtex</filename> (and <filename>dviwindo</filename> is an alias for <filename>dvipsone</filename>).
</para>
<para>The second kind of options, beyond the drivers, are below.
</para>
<variablelist>
<varlistentry><term><literal>monochrome</literal>
</term><listitem><para>Disable the color commands, so that they do not generate errors but do
not generate color either.
</para>
</listitem></varlistentry><varlistentry><term><literal>dvipsnames</literal>
</term><listitem><para>Make available a list of 68 color names that are often used,
particularly in legacy documents.  These color names were originally
provided by the <filename>dvips</filename> driver, giving the option name.
</para>
</listitem></varlistentry><varlistentry><term><literal>nodvipsnames</literal>
</term><listitem><para>Do not load that list of color names, saving &latex; a tiny amount of
memory space.
</para>
</listitem></varlistentry></variablelist>

</sect1>
<sect1 label="21.2" id="Color-models">
<title>Color models</title>

<indexterm role="cp"><primary>color models</primary></indexterm>

<para>A <firstterm>color model</firstterm> is a way of representing colors.  &latex;&#8217;s
capabilities depend on the printer driver.  However, the <filename>pdftex</filename>,
<filename>xetex</filename>, and <filename>luatex</filename> printer drivers are today by far the
most commonly used.  The models below work for those drivers.  All but
one of these is also supported by essentially all other printer drivers
used today.
</para>
<para>Note that color combination can be additive or subtractive.  Additive
mixes colors of light, so that for instance combining full intensities
of red, green, and blue produces white.  Subtractive mixes pigments,
such as with inks, so that combining full intensity of cyan, magenta,
and yellow makes black.
</para>
<variablelist><anchor id="color-models-cmyk"/><varlistentry><term><literal>cmyk</literal>
</term><listitem><para>A comma-separated list with four real numbers between 0 and 1,
inclusive.  The first number is the intensity of cyan, the second is
magenta, and the others are yellow and black.  A number value of 0 means
minimal intensity, while a 1 is for full intensity.  This model is often
used in color printing.  It is a subtractive model.
</para>
<anchor id="color-models-gray"/></listitem></varlistentry><varlistentry><term><literal>gray</literal>
</term><listitem><para>A single real number between 0 and 1, inclusive.  The colors are shades
of grey.  The number 0 produces black while 1 gives white.
</para>
<anchor id="color-models-rgb"/></listitem></varlistentry><varlistentry><term><literal>rgb</literal>
</term><listitem><para>A comma-separated list with three real numbers between 0 and 1,
inclusive.  The first number is the intensity of the red component, the
second is green, and the third the blue.  A number value of 0 means that
none of that component is added in, while a 1 means full intensity.
This is an additive model.
</para>
<anchor id="color-models-RGB"/></listitem></varlistentry><varlistentry><term><literal>RGB</literal>
</term><listitem><para>(<filename>pdftex</filename>, <filename>xetex</filename>, <filename>luatex</filename> drivers) A comma-separated
list with three integers between 0 and 255, inclusive.  This model is a
convenience for using <literal>rgb</literal> since outside of &latex; colors are
often described in a red-green-blue model using numbers in this range.
The values entered here are converted to the <literal>rgb</literal> model by
dividing by 255.
</para>
<anchor id="color-models-named"/></listitem></varlistentry><varlistentry><term><literal>named</literal>
</term><listitem><para>Colors are accessed by name, such as &#8216;<literal>PrussianBlue</literal>&#8217;.  The list of
names depends on the driver, but all support the names &#8216;<literal>black</literal>&#8217;,
&#8216;<literal>blue</literal>&#8217;, &#8216;<literal>cyan</literal>&#8217;, &#8216;<literal>green</literal>&#8217;, &#8216;<literal>magenta</literal>&#8217;, &#8216;<literal>red</literal>&#8217;,
&#8216;<literal>white</literal>&#8217;, and &#8216;<literal>yellow</literal>&#8217; (See the <literal>dvipsnames</literal> option in
<link linkend="Color-package-options">Color package options</link>).
</para>
</listitem></varlistentry></variablelist>

</sect1>
<sect1 label="21.3" id="Commands-for-color">
<title>Commands for color</title>

<indexterm role="cp"><primary>color package commands</primary></indexterm>

<para>These are the commands available with the <filename>color</filename> package.
</para>


<sect2 label="21.3.1" id="Define-colors">
<title>Define colors</title>

<indexterm role="cp"><primary>color</primary></indexterm>
<indexterm role="cp"><primary>define color</primary></indexterm>
<indexterm role="cp"><primary>color, define</primary></indexterm>

<para>Synopsis:
</para>
<screen>\definecolor{<replaceable>name</replaceable>}{<replaceable>model</replaceable>}{<replaceable>specification</replaceable>}
</screen>
<para>Give the name <replaceable>name</replaceable> to the color.  For example, after this
</para>
<screen>\definecolor{silver}{rgb}{0.75,0.75,0.74}
</screen>
<para>you can use that color name with <literal>Hi ho,
\textcolor{silver}{Silver}!</literal>.
</para>
<para>This example gives the color a more abstract name, so it could change and
not be misleading.
</para>
<screen>\definecolor{logocolor}{RGB}{145,92,131}    % RGB needs pdflatex
\newcommand{\logo}{\textcolor{logocolor}{Bob's Big Bagels}}
</screen>
<para>Often a document&#8217;s colors are defined in the preamble, or in the class
or style, rather than in the document body.
</para>

</sect2>
<sect2 label="21.3.2" id="Colored-text">
<title>Colored text</title>

<indexterm role="cp"><primary>color</primary></indexterm>
<indexterm role="cp"><primary>colored text</primary></indexterm>

<para>Synopses:
</para>
<screen>\textcolor{<replaceable>name</replaceable>}{...}
\textcolor[<replaceable>color model</replaceable>]{<replaceable>color specification</replaceable>}{...}
</screen>
<para>or
</para>
<screen>\color{<replaceable>name</replaceable>}
\color[<replaceable>color model</replaceable>]{<replaceable>specification</replaceable>}
</screen>
<para>The affected text gets the color.  This line
</para>
<screen>\textcolor{magenta}{My name is Ozymandias, king of kings:}
Look on my works, ye Mighty, and despair!
</screen>
<para>causes the first half to be in magenta while the rest is in black.  You
can use a color declared with <literal>\definecolor</literal> in exactly the same
way that we just used the builtin color &#8216;<literal>magenta</literal>&#8217;.
</para>
<screen>\definecolor{MidlifeCrisisRed}{rgb}{1.0,0.11,0.0}
I'm thinking about getting a \textcolor{MidlifeCrisisRed}{sports car}.
</screen>
<para>The two <literal>\textcolor</literal> and <literal>\color</literal> differ in that the first is
a command form, enclosing the text to be colored as an argument.  Often
this form is more convenient, or at least more explicit.  The second
form is a declaration, as in <literal>The moon is made of {\color{green}
green} cheese</literal>, so it is in effect until the end of the current group
or environment.  This is sometimes useful when writing macros or as
below where it colors everything inside the <literal>center</literal> environment,
including the vertical and horizontal lines.
</para>
<screen>\begin{center} \color{blue}
  \begin{tabular}{l|r}
    UL &amp;UR \\ \hline
    LL &amp;LR 
  \end{tabular}
\end{center}
</screen>
<para>You can use color in equations.  A document might have this definition
in the preamble
</para>
<screen>\definecolor{highlightcolor}{RGB}{225,15,0}
</screen>
<para>and then contain this equation.
</para>
<screen>\begin{equation}
  \int_a^b \textcolor{highlightcolor}{f'(x)}\,dx=f(b)-f(a)
\end{equation}
</screen>
<para>Typically the colors used in a document are declared in a class or style
but sometimes you want a one-off.  Those are the second forms in the
synopses.
</para>
<screen>Colors of \textcolor[rgb]{0.33,0.14,0.47}{Purple} and
{\color[rgb]{0.72,0.60,0.37} Gold} for the team.
</screen>
<para>The format of <replaceable>color specification </replaceable> depends on the color model
(see <link linkend="Color-models">Color models</link>).  For instance, while <literal>rgb</literal> takes three
numbers, <literal>gray</literal> takes only one.
</para>
<screen>The selection was \textcolor[gray]{0.5}{grayed out}.
</screen>
<para>Colors inside colors do not combine.  Thus
</para>
<screen>\textcolor{green}{kind of \textcolor{blue}{blue}}
</screen>
<para>has a final word that is blue, not a combination of blue and green.
</para>
<!-- xx address coloring a line of a table? -->


</sect2>
<sect2 label="21.3.3" id="Colored-boxes">
<title>Colored boxes</title>

<indexterm role="cp"><primary>color</primary></indexterm>
<indexterm role="cp"><primary>colored boxes</primary></indexterm>
<indexterm role="cp"><primary>box, colored</primary></indexterm>

<para>Synopses:
</para>
<screen>\colorbox{<replaceable>name</replaceable>}{...}
\colorbox[<replaceable>model name</replaceable>]{<replaceable>box background color</replaceable>}{...}
</screen>
<para>or
</para>
<screen>\fcolorbox{<replaceable>frame color</replaceable>}{<replaceable>box background color</replaceable>}{...}
\fcolorbox[<replaceable>model name</replaceable>]{<replaceable>frame color</replaceable>}{<replaceable>box background color</replaceable>}{...}
</screen>
<para>Make a box with the stated background color.  The <literal>\fcolorbox</literal>
command puts a frame around the box.  For instance this
</para>
<screen>Name:~\colorbox{cyan}{\makebox[5cm][l]{\strut}}
</screen>
<para>makes a cyan-colored box that is five centimeters long and gets its
depth and height from the <literal>\strut</literal> (so the depth is
<literal>-.3\baselineskip</literal> and the height is <literal>\baselineskip</literal>).  This
puts white text on a blue background.
</para>
<screen>\colorbox{blue}{\textcolor{white}{Welcome to the machine.}}
</screen>
<para>The <literal>\fcolorbox</literal> commands use the same parameters as <literal>\fbox</literal>
(see <link linkend="_005cfbox-_0026-_005cframebox">\fbox &amp; \framebox</link>), <literal>\fboxrule</literal> and <literal>\fboxsep</literal>, to
set the thickness of the rule and the boundary between the box interior
and the surrounding rule.  &latex;&#8217;s defaults are <literal>0.4pt</literal> and
<literal>3pt</literal>, respectively.
</para>
<para>This example changes the thickness of the border to 0.8 points.  Note
that it is surrounded by curly braces so that the change ends at the end
of the second line.
</para>
<screen>{\setlength{\fboxrule}{0.8pt}
\fcolorbox{black}{red}{Under no circumstances turn this knob.}}
</screen>

</sect2>
<sect2 label="21.3.4" id="Colored-pages">
<title>Colored pages</title>

<indexterm role="cp"><primary>color</primary></indexterm>
<indexterm role="cp"><primary>colored page</primary></indexterm>
<indexterm role="cp"><primary>page, colored</primary></indexterm>
<indexterm role="cp"><primary>background, colored</primary></indexterm>

<para>Synopses:
</para>
<screen>\pagecolor{<replaceable>name</replaceable>}
\pagecolor[<replaceable>color model</replaceable>]{<replaceable>color specification</replaceable>}
\nopagecolor
</screen>
<para>The first two set the background of the page, and all subsequent pages,
to the color.  For an explanation of the specification in the second
form see <link linkend="Colored-text">Colored text</link>.  The third returns the background to normal,
which is a transparent background.  (If that is not supported use
<literal>\pagecolor{white}</literal>, although that will make a white background
rather than the default transparent background.)
</para>
<screen> ...
\pagecolor{cyan}
 ...
\nopagecolor
</screen>

</sect2>
</sect1>
</chapter>
<chapter label="22" id="Graphics">
<title>Graphics</title>

<indexterm role="cp"><primary>graphics</primary></indexterm>
<indexterm role="cp"><primary>graphics package</primary></indexterm>

<para>You can use graphics such as PNG or PDF files in your &latex; document.
You need an additional package, which comes standard with &latex;.
This example is the short how-to.
</para>
<screen>\include{graphicx}  % goes in the preamble
  ...
\includegraphics[width=0.5\linewidth]{plot.pdf}
</screen>
<para>To use the commands described here your document preamble must contain
either <literal>\usepackage{graphicx}</literal> or
<literal>\usepackage{graphics}</literal>. Most of the time, <filename>graphicx</filename> is the
better choice.
</para>
<para>Graphics come in two main types, raster and vector.  &latex; can use
both.  In raster graphics the file contains an entry for each location
in an array, describing what color it is.  An example is a photograph,
in JPG format.  In vector graphics, the file contains a list of
instructions such as &#8216;<literal>draw a circle with this radius and that
center</literal>&#8217;.  An example is a line drawing produced by the Asymptote
program, in PDF format.  Generally vector graphics are more useful
because you can rescale their size without pixelation or other problems,
and because they often have a smaller size.
</para>
<para>There are systems particularly well-suited to make graphics for a
&latex; document.  For example, these allow you to use the same fonts
as in your document.  &latex; comes with a <literal>picture</literal> environment
(see <link linkend="picture">picture</link>) that has simple capabilities.  Besides that, there are
other ways to include the graphic-making commands in the document.  Two
such systems are the PSTricks and TikZ packages.  There are also systems
external to &latex;, that generate a graphic that you include using the
commands of this chapter.  Two that use a programming language are
Asymptote and MetaPost.  One that uses a graphical interface is Xfig.
Full description of these systems is outside the scope of this document;
see their documentation on CTAN.
</para>


<sect1 label="22.1" id="Graphics-package-options">
<title><literal>graphics</literal> package options</title>

<indexterm role="cp"><primary>graphics package options</primary></indexterm>
<indexterm role="cp"><primary>options, graphics package</primary></indexterm>

<para>Synopsis (must be in the document preamble):
</para>
<screen>\usepackage[<replaceable>comma-separated option list</replaceable>]{graphics}
</screen>
<para>or
</para>
<screen>\usepackage[<replaceable>comma-separated option list</replaceable>]{graphicx}
</screen>
<para>The <literal>graphicx</literal> package has a format for optional arguments to the
<literal>\includegraphics</literal> command that is convenient (it is the key-value
format), so it is the better choice for new documents.  When you load
the <filename>graphics</filename> or <filename>graphicx</filename> package with <literal>\usepackage</literal>
there are two kinds of available options.
</para>
<para>The first is that &latex; does not contain information about different
output systems but instead depends on information stored in a
<firstterm>printer driver</firstterm> file.  Normally you should not specify the driver
option in the document, and instead rely on your system&#8217;s default. One
advantage of this is that it makes the document portable across systems.
</para>
<para>For completeness here is a list of the drivers.  The currently relevant
ones are: <filename>dvipdfmx</filename>, <filename>dvips</filename>, <filename>dvisvgm</filename>, <filename>luatex</filename>,
<filename>pdftex</filename>, <filename>xetex</filename>.  The two <filename>xdvi</filename> and <filename>oztex</filename> are
essentially aliases for <filename>dvips</filename> (and <filename>xdvi</filename> is monochrome).
Ones that should not be used for new systems are: <filename>dvipdf</filename>,
<filename>dvipdfm</filename>, <filename>dviwin</filename>, <filename>dvipsone</filename>, <filename>emtex</filename>,
<filename>pctexps</filename>, <filename>pctexwin</filename>, <filename>pctexhp</filename>, <filename>pctex32</filename>,
<filename>truetex</filename>, <filename>tcidvi</filename>, <filename>vtex</filename> (and <filename>dviwindo</filename> is an
alias for <filename>dvipsone</filename>).  These are stored in files with a
<filename>.def</filename> extension, such as <filename>pdftex.def</filename>.
</para>
<para>The second kind of options are below.
</para>
<variablelist><varlistentry><term><literal>demo</literal>
</term><listitem><para>Instead of an image file, &latex; puts in a 150&#160;pt by 100&#160;pt
rectangle (unless another size is specified in the
<literal>\includegraphics</literal> command).
</para>
</listitem></varlistentry><varlistentry><term><literal>draft</literal>
</term><listitem><para>For each graphic file, it is not shown but instead the file name is
printed in a box of the correct size.  In order to determine the size,
the file must be present.
</para>
</listitem></varlistentry><varlistentry><term><literal>final</literal>
</term><listitem><para>(Default) Override any previous <literal>draft</literal> option, so that the
document shows the contents of the graphic files.
</para>
</listitem></varlistentry><varlistentry><term><literal>hiderotate</literal>
</term><listitem><para>Do not show rotated text.  (This allows for the possibility that a
previewer does not have the capability to rotate text.)
<!-- what does it show? -->
</para>
</listitem></varlistentry><varlistentry><term><literal>hidescale</literal>
</term><listitem><para>Do not show scaled text.  (This allows for the possibility that a
previewer does not have the capability to scale.)
<!-- what does it show? -->
</para>
</listitem></varlistentry><varlistentry><term><literal>hiresbb</literal>
</term><listitem><para>In a PS or EPS file the graphic size may be specified in two ways.  The
<literal>%%BoundingBox</literal> lines describe the graphic size using integer
multiples of a PostScript point, that is, integer multiples of 1/72
inch.  A later addition to the PostScript language allows decimal
multiples, such as 1.23, in <literal>%%HiResBoundingBox</literal> lines.  This
option has &latex; to read the size from the latter.
</para>
</listitem></varlistentry></variablelist>

</sect1>
<sect1 label="22.2" id="Graphics-package-configuration">
<title><literal>graphics</literal> package configuration</title>

<indexterm role="cp"><primary>graphics</primary></indexterm>
<indexterm role="cp"><primary>graphics package</primary></indexterm>
<indexterm role="cp"><primary>configuration, graphics package</primary></indexterm>
<indexterm role="cp"><primary>EPS files</primary></indexterm>
<indexterm role="cp"><primary>JPEG files</primary></indexterm>
<indexterm role="cp"><primary>JPG files</primary></indexterm>
<indexterm role="cp"><primary>PDF graphic files</primary></indexterm>
<indexterm role="cp"><primary>PNG files</primary></indexterm>

<para>These commands configure the way &latex; searches the file system for
the graphic.
</para>
<para>The behavior of file system search code is necessarily platform
dependent.  In this document we cover GNU/Linux, Macintosh, and Windows, as
those systems are typically configured.  For other situations consult
the documentation in <filename>grfguide.pdf</filename>, or the &latex; source, or your
&tex; distribution&#8217;s documentation.
</para>


<sect2 label="22.2.1" id="_005cgraphicspath">
<title><literal>\graphicspath</literal></title>

<indexterm role="fn"><primary>\graphicspath</primary></indexterm>

<para>Synopsis:
</para>
<screen>\graphicspath{<replaceable>list of dir names inside curly brackets</replaceable>}
</screen>
<para>Declare a list of directories to search for graphics files.  This allows
you to later say something like <literal>\includegraphics{lion.png}</literal>
instead of having to give its path.
</para>
<para>&latex; always looks for graphic files first in the current directory.
The declaration below tells the system to then look in the subdirectory
<filename>pix</filename>, and then <filename>../pix</filename>.
</para>
<screen>\usepackage{graphicx}   % or graphics; put in preamble
  ...
\graphicspath{ {pix/} {../pix/} }
</screen>
<para>The <literal>\graphicspath</literal> declaration is optional.  If you don&#8217;t include
it then &latex;&#8217;s default is to search all of the places that it
usually looks for a file (it uses &latex;&#8217;s <literal>\input@path</literal>).  In
particular, in this case one of the places it looks is the current
directory.
</para>
<para>Enclose each directory name in curly braces; for example, above it says
&#8216;<literal><literal>{pix}</literal></literal>&#8217;.  Do this even if there is only one directory.
Each directory name must end in a forward slash, <filename>/</filename>.  This is true
even on Windows, where good practice is to use forward slashes for all
the directory separators since it makes the document portable to other
platforms.  If you have spaces in your directory name then use double
quotes, as with <literal>{&quot;my docs/&quot;}</literal>. Getting one of these rules wrong
will cause &latex; to report <literal>Error: File `<replaceable>filename</replaceable>' not
found</literal>.
</para>
<para>Basically, the algorithm is that with this example, after looking in the
current directory,
</para>
<screen>\graphicspath{ {pix/} {../pix/} }
...
\usepackage{lion.png}
</screen>
<para>for each of the listed directories, &latex; concatenates it with the
file name and searches for the result, checking for <filename>pix/lion.png</filename>
and then <filename>../pix/lion.png</filename>.  This algorithm means that the
<literal>\graphicspath</literal> command does not recursively search subdirectories:
if you issue <literal>\graphicspath{{a/}}</literal> and the graphic is in
<filename>a/b/lion.png</filename> then &latex; will not find it.  It also means that
you can use absolute paths such as
<literal>\graphicspath{{/home/jim/logos/}}</literal> or
<literal>\graphicspath{{C:/Users/Albert/Pictures/}}</literal>.  However, using
these means that the document is not portable.  (You could preserve
portability by adjusting your &tex; system settings configuration file
parameter <literal>TEXINPUTS</literal>; see the documentation of your system.)
</para>
<para>You can use <literal>\graphicspath</literal> anywhere in the document.  You can use
it more than once.  Show its value with
<literal>\makeatletter\typeout{\Ginput@path}\makeatother</literal>.
</para>
<para>The directories are taken with respect to the base file.  That is,
suppose that you are working on a document based on <filename>book/book.tex</filename>
and it contains <literal>\include{chapters/chap1}</literal>.  If in
<filename>chap1.tex</filename> you put <literal>\graphicspath{{plots/}}</literal> then
&latex; will not search for graphics in <filename>book/chapters/plots</filename>, but
instead in <filename>book/plots</filename>.
</para>

</sect2>
<sect2 label="22.2.2" id="_005cDeclareGraphicsExtensions">
<title><literal>\DeclareGraphicsExtensions</literal></title>

<indexterm role="fn"><primary>\DeclareGraphicsExtensions</primary></indexterm>

<para>Synopses:
</para>
<screen>\DeclareGraphicsExtensions{<replaceable>comma-separated list of file extensions</replaceable>}
</screen>
<para>Declare the filename extensions to try.  This allows you to specify the
order in which to choose graphic formats when you include graphic files
by giving the filename without the extension, as in
<literal>\includegraphics{functionplot}</literal>.
</para>
<para>In this example, &latex; will find files in the PNG format before PDF
files.
</para>
<screen>\DeclareGraphicsExtensions{.png,PNG,.pdf,.PDF}
  ...
\includegraphics{lion}   % will find <filename>lion.png</filename> before <filename>lion.pdf</filename> 
</screen>
<para>Because the file name <filename>lion</filename> does not have a period, &latex; uses
the extension list.  For each directory in the graphics path
(see <link linkend="_005cgraphicspath">\graphicspath</link>), &latex; will try the extensions in the order
given.  If it does not find such a file after trying all the directories
and extensions then it reports &#8216;<literal>! LaTeX Error: File `<filename>lion</filename>'
not found</literal>&#8217;.  Note that you must include the periods at the start of the
extensions.
</para>
<para>Because GNU/Linux and Macintosh filenames are case sensitive, the list of
file extensions is case sensitive on those platforms.  The Windows
platform is not case sensitive.
</para>
<para>You are not required to include <literal>\DeclareGraphicsExtensions</literal> in
your document; the printer driver has a sensible default.  For example,
the most recent <filename>pdftex.def</filename> has this extension list.
</para>
<screen>.png,.pdf,.jpg,.mps,.jpeg,.jbig2,.jb2,.PNG,.PDF,.JPG,.JPEG,.JBIG2,.JB2
</screen>
<para>You can use this command anywhere in the document.  You can use it more
than once.  Show its value with
<literal>\makeatletter\typeout{\Gin@extensions}\makeatother</literal>.
</para>

</sect2>
<sect2 label="22.2.3" id="_005cDeclareGraphicsRule">
<title><literal>\DeclareGraphicsRule</literal></title>

<indexterm role="fn"><primary>\DeclareGraphicsRule</primary></indexterm>

<para>Synopsis:
</para>
<screen>\DeclareGraphicsRule{<replaceable>extension</replaceable>}{<replaceable>type</replaceable>}{<replaceable>size-file extension</replaceable>}{<replaceable>command</replaceable>}
</screen>
<para>Declare how to handle graphic files whose names end in <replaceable>extension</replaceable>.
</para>
<para>This example declares that all files with names have the form
<filename>filename-without-dot.mps</filename> will be treated as output from MetaPost,
meaning that the printer driver will use its MetaPost-handling code to
input the file.
</para>
<screen>\DeclareGraphicsRule{.mps}{mps}{.mps}{}
</screen>
<para>This
</para>
<screen>\DeclareGraphicsRule{*}{mps}{*}{}
</screen>
<para>tells &latex; that it should handle as MetaPost output any file with an
extension not covered by another rule, so it covers <filename>filename.1</filename>,
<filename>filename.2</filename>, etc.
</para>
<para>This describes the four arguments.
</para>
<variablelist><varlistentry><term><replaceable>extension</replaceable>
</term><listitem><para>The file extension to which this rule applies. The extension is anything
after and including the first dot in the filename.  Use the Kleene star,
<literal>*</literal>, to denote the default behaviour for all undeclared extensions.
</para>
</listitem></varlistentry><varlistentry><term><replaceable>type</replaceable>
</term><listitem><para>The type of file involved.  This type is a string that must be defined
in the printer driver. For instance, files with extensions <filename>.ps</filename>,
<filename>.eps</filename>, or <filename>.ps.gz</filename> may all be classed as type <literal>eps</literal>.
All files of the same type will be input with the same internal command
by the printer driver. For example, the file types that <filename>pdftex</filename>
recognizes are: <literal>jpg</literal>, <literal>jbig2</literal>, <literal>mps</literal>, <literal>pdf</literal>,
<literal>png</literal>, <literal>tif</literal>.
</para>
</listitem></varlistentry><varlistentry><term><replaceable>size-file extension</replaceable>
</term><listitem><para>The extension of the file to be read to determine the size of the
graphic, if there is such a file. It may be the same as <replaceable>extension</replaceable>
but it may be different.
</para>
<para>As an example, consider a PostScript graphic.  To make it smaller, it
might be compressed into a <filename>.ps.gz</filename> file.  Compressed files are not
easily read by &latex; so you can put the bounding box information in a
separate file. If <replaceable>size-file extension</replaceable> is empty then you must
specify size information in the arguments of <literal>\includegraphics</literal>.
</para>
<para>If the driver file has a procedure for reading size files for
<literal>type</literal> then that will be used, otherwise it will use the procedure
for reading <filename>.eps</filename> files.  (Thus you may specify the size of bitmap
files in a file with a PostScript style <literal>%%BoundingBox</literal> line if no
other format is available.)
</para>
</listitem></varlistentry><varlistentry><term><replaceable>command</replaceable>
</term><listitem><para>A command that will be applied to the
file. This is very often left blank. This command must start with a
single backward quote.  Thus,
<literal>\DeclareGraphicsRule{.eps.gz}{eps}{.eps.bb}{`gunzip -c
#1}</literal> specifies that any file with the extension <filename>.eps.gz</filename> should
be treated as an <literal>eps</literal> file, with the the BoundingBox information
stored in the file with extension <filename>.eps.bb</filename>, and that the command
<literal>gunzip -c</literal> will run on your platform to decompresses the file.
</para>
<para>Such a command is specific to your platform.  In addition, your &tex;
system must allow you to run external commands; as a security measure
modern systems restrict running commands unless you explicitly allow it.
See the documentation for your &tex; distribution.
</para>
</listitem></varlistentry></variablelist>
</sect2>
</sect1>
<sect1 label="22.3" id="Commands-for-graphics">
<title>Commands for graphics</title>

<indexterm role="cp"><primary>graphics package commands</primary></indexterm>
<indexterm role="cp"><primary>commands, graphics package</primary></indexterm>

<para>These are the commands available with the <filename>graphics</filename> and
<filename>graphicx</filename> packages.
</para>


<sect2 label="22.3.1" id="_005cincludegraphics">
<title><literal>\includegraphics</literal></title>

<indexterm role="cp"><primary>graphics</primary></indexterm>
<indexterm role="cp"><primary>graphics package</primary></indexterm>
<indexterm role="cp"><primary>including graphics</primary></indexterm>
<indexterm role="cp"><primary>importing graphics</primary></indexterm>
<indexterm role="cp"><primary>EPS files</primary></indexterm>
<indexterm role="cp"><primary>JPEG files</primary></indexterm>
<indexterm role="cp"><primary>JPG files</primary></indexterm>
<indexterm role="cp"><primary>PDF graphic files</primary></indexterm>
<indexterm role="cp"><primary>PNG files</primary></indexterm>
<indexterm role="fn"><primary>\includegraphics</primary></indexterm>

<para>Synopses for <filename>graphics</filename> package:
</para>
<screen>\includegraphics{<replaceable>filename</replaceable>}
\includegraphics[<replaceable>urx</replaceable>,<replaceable>ury</replaceable>]{<replaceable>filename</replaceable>}
\includegraphics[<replaceable>llx</replaceable>,<replaceable>lly</replaceable>][<replaceable>urx</replaceable>,<replaceable>ury</replaceable>]{<replaceable>filename</replaceable>}
\includegraphics*{<replaceable>filename</replaceable>}
\includegraphics*[<replaceable>urx</replaceable>,<replaceable>ury</replaceable>]{<replaceable>filename</replaceable>}
\includegraphics*[<replaceable>llx</replaceable>,<replaceable>lly</replaceable>][<replaceable>urx</replaceable>,<replaceable>ury</replaceable>]{<replaceable>filename</replaceable>}
</screen>
<para>Synopses for <filename>graphicx</filename> package:
</para>
<screen>\includegraphics{<replaceable>filename</replaceable>}
\includegraphics[<replaceable>key-value list</replaceable>]{<replaceable>filename</replaceable>}
\includegraphics*{<replaceable>filename</replaceable>}
\includegraphics*[<replaceable>key-value list</replaceable>]{<replaceable>filename</replaceable>}
</screen>
<para>Include a graphics file.  The starred form <literal>\includegraphics*</literal> will
clip the graphic to the size specified, while for the unstarred form any
part of the graphic that is outside the box of the specified size will
over-print the surrounding area.
</para>
<para>This 
</para>
<screen>\usepackage{graphicx}  % in preamble
  ...
\begin{center}
  \includegraphics{plot.pdf}
\end{center}
</screen>
<para>will incorporate into the document the graphic in <filename>plot.pdf</filename>,
centered and at its nominal size.  You can also give a path to the file,
as with <literal>\includegraphics{graphics/plot.pdf}</literal>.  To specify a list
of locations to search for the file, see <link linkend="_005cgraphicspath">\graphicspath</link>.
</para>
<para>If your filename includes spaces then put it in double quotes. An example
is <literal>\includegraphics{&quot;sister picture.jpg&quot;}</literal>.
</para>
<para>The <literal>\includegraphics{<replaceable>filename</replaceable>}</literal> command decides on the
type of graphic by splitting <replaceable>filename</replaceable> on the first dot.  You can
use <replaceable>filename</replaceable> with no dot, as in <literal>\includegraphics{turing}</literal>
and then &latex; tries a sequence of extensions such as <literal>.png</literal> and
<literal>.pdf</literal> until it finds a file with that extension
(see <link linkend="_005cDeclareGraphicsExtensions">\DeclareGraphicsExtensions</link>).
</para>
<para>If your file name contains dots before the extension then you can hide
them with curly braces, as in
<literal>\includegraphics{{plot.2018.03.12.a}.pdf}</literal>.  Or, if you use
the <filename>graphicx</filename> package then you can use the options <literal>type</literal> and
<literal>ext</literal>; see below.  This and other filename issues are also handled
with the package <filename>grffile</filename>.
</para>
<para>This example puts a graphic in a figure environment so &latex; can
move it to the next page if fitting it on the current page is awkward
(see <link linkend="figure">figure</link>).
</para>
<screen>\begin{figure}
  \centering
  \includegraphics[width=3cm]{lungxray.jpg}
  \caption{The evidence is overwhelming: don't smoke.}  \label{fig:xray}
\end{figure}
</screen>
<para>This places a graphic that will not float, so it is sure to appear at
this point in the document even if makes &latex; stretch the text or
resort to blank areas on the page.  It will be centered and will have a
caption.
</para>
<screen>\usepackage{caption}  % in preamble
  ...
\begin{center}
  \includegraphics{pix/nix.png}
  \captionof{figure}{The spirit of the night} \label{pix:nix} % optional
\end{center}
</screen>
<para>This example puts a box with a graphic side by side with one having
text, with the two vertically centered.
</para>
<screen>\newcommand*{\vcenteredhbox}[1]{\begin{tabular}{@{}c@{}}#1\end{tabular}}
  ...
\begin{center}
  \vcenteredhbox{\includegraphics[width=0.4\textwidth]{plot}}
  \hspace{1em}
  \vcenteredhbox{\begin{minipage}{0.4\textwidth}
                   \begin{displaymath}
                     f(x)=x\cdot \sin (1/x)
                   \end{displaymath}
                 \end{minipage}}
\end{center}
</screen>
<para>If you use the <filename>graphics</filename> package then the only options involve the
size of the graphic (but see <link linkend="_005crotatebox">\rotatebox</link> and <link linkend="_005cscalebox">\scalebox</link>).
When one optional argument is present then it is
<literal>[<replaceable>urx</replaceable>,<replaceable>ury</replaceable>]</literal> and it gives the coordinates of the top
right corner of the image, as a pair of &tex; dimensions (see <link linkend="Units-of-length">Units
of length</link>).  If the units are omitted they default to <literal>bp</literal>.  In
this case, the lower left corner of the image is assumed to be at (0,0).
If two optional arguments are present then the leading one is
<literal>[<replaceable>llx</replaceable>,<replaceable>lly</replaceable>]</literal>, specifying the coordinates of the image&#8217;s
lower left.  Thus, <literal>\includegraphics[1in,0.618in]{...}</literal> calls for
the graphic to be placed so it is 1&#160;inch wide and 0.618&#160;inches
tall and so its origin is at (0,0).
</para>
<para>The <filename>graphicx</filename> package gives you many more options.  Specify them
in a key-value form, as here.
</para>
<screen>\begin{center}
  \includegraphics[width=1in,angle=90]{lion}
  \hspace{2em}
  \includegraphics[angle=90,width=1in]{lion}
\end{center}
</screen>
<para>The options are read left-to-right. So the first graphic above is made
one inch wide and then rotated, while the second is rotated and then
made one inch wide.  Thus, unless the graphic is perfectly square, the
two will end with different widths and heights.
</para>
<para>There are many options.  The primary ones are listed first.
</para>
<para>Note that a graphic is placed by &latex; into a box, which is
traditionally referred to as its bounding box (distinct from the
PostScript BoundingBox described below).  The graphic&#8217;s printed area may
go beyond this box, or sit inside this box, but when &latex; makes up a
page it puts together boxes and this is the box allocated for the
graphic.
</para>
<variablelist><anchor id="includegraphics-width"/><varlistentry><term><literal>width</literal>
</term><listitem><para>The graphic will be shown so its bounding box is this width.  An example
is <literal>\includegraphics[width=1in]{plot}</literal>.  You can use the standard
&tex; dimensions (see <link linkend="Units-of-length">Units of length</link>) and also convenient is
<literal>\linewidth</literal>, or in a two-column document, <literal>\columnwidth</literal>
(see <link linkend="Page-layout-parameters">Page layout parameters</link>).  An example is that by using the
<filename>calc</filename> package you can make the graphic be 1&#160;cm narrow than
the width of the text with
<literal>\includegraphics[width=\linewidth-1.0cm]{hefferon.jpg}</literal>.
</para>
</listitem></varlistentry><varlistentry><term><literal>height</literal>
</term><listitem><anchor id="includegraphics-height"/><para>The graphic will be shown so its bounding box is this height.  You can
use the standard &tex; dimensions (see <link linkend="Units-of-length">Units of length</link>), and also
convenient are <literal>\pageheight</literal> and <literal>\textheight</literal> (see <link linkend="Page-layout-parameters">Page
layout parameters</link>).  For instance, the command
<literal>\includegraphics[height=0.25\textheight]{godel}</literal> will make the
graphic a quarter of the height of the text area.
</para>
</listitem></varlistentry><varlistentry><term><literal>totalheight</literal>
</term><listitem><anchor id="includegraphics-totalheght"/><para>The graphic will be shown so its bounding box has this height plus
depth.  This differs from the height if the graphic was rotated. For
instance, if it has been rotated by -90 then it will have zero height
but a large depth.
</para>
</listitem></varlistentry><varlistentry><term><literal>keepaspectratio</literal>
</term><listitem><anchor id="includegraphics-keepaspectratio"/><para>If set to <literal>true</literal>, or just specified as here
</para>
<screen><literal>\includegraphics[...,keepaspectratio,...]{...}</literal>
</screen>
<para>and you give as options both <literal>width</literal> and <literal>height</literal> (or
<literal>totalheight</literal>), then &latex; will make the graphic is as large as
possible without distortion. That is, &latex; will ensure that neither
is the graphic wider than <literal>width</literal> nor taller than <literal>height</literal> (or
<literal>totalheight</literal>).
</para>
</listitem></varlistentry><varlistentry><term><literal>scale</literal>
</term><listitem><para>Factor by which to scale the graphic.  To make a graphic twice its
nominal size, enter <literal>\includegraphics[scale=2.0]{...}</literal>.  This
number may be any value; a number between 1 and&#160;0 will shrink the
graphic and a negative number will reflect it.
</para>
</listitem></varlistentry><varlistentry><term><literal>angle</literal>
</term><listitem><para>Rotate the graphic.  The angle is taken in degrees and counterclockwise.
The graphic is rotated about its <literal>origin</literal>; see that option.  For a
complete description of how rotated material is typeset,
see <link linkend="_005crotatebox">\rotatebox</link>.
</para>
</listitem></varlistentry><varlistentry><term><literal>origin</literal>
</term><listitem><para>The point of the graphic about which the rotation happens.  Possible
values are any string containing one or two of: <literal>l</literal> for left,
<literal>r</literal> for right, <literal>b</literal> for bottom, <literal>c</literal> for center, <literal>t</literal>
for top, and <literal>B</literal> for baseline.  Thus, entering the command
<literal>\includegraphics[angle=180,origin=c]{moon}</literal> will turn the
picture upside down about that picture&#8217;s center, while the command
<literal>\includegraphics[angle=180,origin=lB]{LeBateau}</literal> will turn its
picture upside down about its left baseline.  (The character <literal>c</literal>
gives the horizontal center in <literal>bc</literal> or <literal>tc</literal>, but gives the
vertical center in <literal>lc</literal> or <literal>rc</literal>.)  The default is <literal>lB</literal>.
</para>
<para>To rotate about an arbitrary point, see <link linkend="_005crotatebox">\rotatebox</link>.
</para>
</listitem></varlistentry></variablelist>
<para>These are lesser-used options.
</para>
<variablelist><anchor id="includegraphics-viewport"/><varlistentry><term><literal>viewport</literal>
</term><listitem><para>Pick out a subregion of the graphic to show.  Takes four arguments,
separated by spaces and given in &tex; dimensions, as with
<literal>\includegraphics[.., viewport=0in 0in 1in 0.618in]{...}</literal>.  The
dimensions default to big points,&#160;<literal>bp</literal>.  They are taken
relative to the origin specified by the bounding box.  See also the
<literal>trim</literal> option.
</para>
<anchor id="includegraphics-trim"/></listitem></varlistentry><varlistentry><term><literal>trim</literal>
</term><listitem><para>Gives parts of the graphic to not show.  Takes four arguments, separated
by spaces, that are given in &tex; dimensions, as with
<literal>\includegraphics[.., trim= 0in 0.1in 0.2in 0.3in, ...]{...}</literal>.
These give the amounts of the graphic not to show, that is, &latex;
will crop the picture by 0&#160;inches on the left, 0.1&#160;inches on
the bottom, 0.2&#160;inches on the right, and 0.3&#160;inches on the
top.  See also the <literal>viewport</literal> option.
</para>
<anchor id="includegraphics-clip"/></listitem></varlistentry><varlistentry><term><literal>clip</literal>
</term><listitem><para>If set to <literal>true</literal>, or just specified as here
</para>
<screen>\includegraphics[...,clip,...]{...}
</screen>
<para>then the graphic is cropped to the bounding box.  This is the same as
using the starred form of the command,
<literal>\includegraphics*[...]{...}</literal>.
</para>
<anchor id="includegraphics-page"/></listitem></varlistentry><varlistentry><term><literal>page</literal>
</term><listitem><para>Give the page number of a multi-page PDF file.  The default is
<literal>page=1</literal>.
</para>
<anchor id="includegraphics-pagebox"/></listitem></varlistentry><varlistentry><term><literal>pagebox</literal>
</term><listitem><para>Specifies which bounding box to use for PDF files from among
<literal>mediabox</literal>, <literal>cropbox</literal>, <literal>bleedbox</literal>, <literal>trimbox</literal>, or
<literal>artbox</literal>.  PDF files do not have the BoundingBox that PostScript
files have, but may specify up to four predefined rectangles. The
MediaBox gives the boundaries of the physical medium. The CropBox is the
region to which the contents of the page are to be clipped when
displayed.  The BleedBox is the region to which the contents of the page
should be clipped in production. The TrimBox is the intended dimensions
of the finished page.  The ArtBox is the extent of the page&#8217;s meaningful
content.  The driver will set the image size based on CropBox if
present, otherwise it will not use one of the others, with a
driver-defined order of preference.  MediaBox is always present.
</para>
<anchor id="includegraphics-interpolate"/></listitem></varlistentry><varlistentry><term><literal>interpolate</literal>
</term><listitem><para>Enable or disable interpolation of raster images by the viewer.  Can be
set with <literal>interpolate=true</literal> or just specified as here.
</para>
<screen>\includegraphics[...,interpolate,...]{...}
</screen>
<anchor id="includegraphics-quiet"/></listitem></varlistentry><varlistentry><term><literal>quiet</literal>
</term><listitem><para>Do not write information to the log.  You can set it with
<literal>quiet=true</literal> or just specified it with
<literal>\includegraphics[...,quite,...]{...}</literal>,
</para>
<anchor id="includegraphics-draft"/></listitem></varlistentry><varlistentry><term><literal>draft</literal>
</term><listitem><para>If you set it with <literal>draft=true</literal> or just specify it with
</para>
<screen>\includegraphics[...,draft,...]{...}
</screen>
<para>then the graphic will not appear in the document, possibly saving color
printer ink.  Instead, &latex; will put an empty box of the correct
size with the filename printed in it.
</para>
</listitem></varlistentry></variablelist>
<para>These options address the bounding box for Encapsulated PostScript
graphic files, which have a size specified with a line
<literal>%%BoundingBox</literal> that appears in the file.  It has four values,
giving the lower <inlineequation><mathphrase>x</mathphrase></inlineequation> coordinate, lower <inlineequation><mathphrase>y</mathphrase></inlineequation> coordinate, upper
<inlineequation><mathphrase>x</mathphrase></inlineequation> coordinate, and upper <inlineequation><mathphrase>y</mathphrase></inlineequation> coordinate.  The units are
PostScript points, equivalent to &tex;&#8217;s big points, 1/72&#160;inch.
For example, if an <filename>.eps</filename> file has the line <literal>%%BoundingBox 10
20 40 80</literal> then its natural size is 30/72&#160;inch wide by
60/72&#160;inch tall.
</para>
<variablelist><anchor id="includegraphics-bb"/><varlistentry><term><literal>bb</literal>
</term><listitem><para>Specify the bounding box of the displayed region.  The argument is four
dimensions separated by spaces, as with <literal>\includegraphics[.., bb=
0in 0in 1in 0.618in]{...}</literal>.  Usually <literal>\includegraphics</literal> reads the
BoundingBox numbers from the EPS file automatically, so this option is
only useful if the bounding box is missing from that file or if you want
to change it.
</para>
<anchor id="includegraphics-bbllx"/><anchor id="includegraphics-bblly"/><anchor id="includegraphics-bburx"/><anchor id="includegraphics-bbury"/></listitem></varlistentry><varlistentry><term><literal>bbllx, bblly, bburx, bbury</literal>
</term><listitem><para>Set the bounding box.  These four are obsolete, but are retained for
compatibility with old packages.
</para>
<anchor id="includegraphics-natwidth"/><anchor id="includegraphics-natheight"/></listitem></varlistentry><varlistentry><term><literal>natwidth, natheight</literal>
</term><listitem><para>An alternative for <literal>bb</literal>.  Setting
</para>
<screen>\includegraphics[...,natwidth=1in,natheight=0.618in,...]{...}
</screen>
<para>is the same as setting <literal>bb=0 0 1in 0.618in</literal>.
</para> 
<anchor id="includegraphics-hiresbb"/></listitem></varlistentry><varlistentry><term><literal>hiresbb</literal>
</term><listitem><para>If set to <literal>true</literal>, or just specified as with
</para>
<screen>\includegraphics[...,hiresbb,...]{...}
</screen>
<para>then &latex; will look for <literal>%%HiResBoundingBox</literal> lines instead of
<literal>%%BoundingBox</literal> lines.  (The <literal>BoundingBox</literal> lines use only
natural numbers while the <literal>HiResBoundingBox</literal> lines use decimals;
both use units equivalent to &tex;&#8217;s big points, 1/72&#160;inch.) To
override a prior setting of <literal>true</literal>, you can set it to <literal>false</literal>.
</para>
</listitem></varlistentry></variablelist>
<para>These following options allow a user to override &latex;&#8217;s method of
choosing the graphic type based on the filename extension.  An example
is that <literal>\includegraphics[type=png,ext=.xxx,read=.xxx]{lion}</literal>
will read the file <filename>lion.xxx</filename> as though it were
<filename>lion.png</filename>. For more on these, see <link linkend="_005cDeclareGraphicsRule">\DeclareGraphicsRule</link>.
</para>
<variablelist><anchor id="includegraphics-type"/><varlistentry><term><literal>type</literal>
</term><listitem><para>Specify the graphics type.
</para>
<anchor id="includegraphics-ext"/></listitem></varlistentry><varlistentry><term><literal>ext</literal>
</term><listitem><para>Specify the graphics extension.
Only use this in conjunction with the option <literal>type</literal>.
</para>
<anchor id="includegraphics-read"/></listitem></varlistentry><varlistentry><term><literal>read</literal>
</term><listitem><para>Specify the file extension of the read file.
Only use this in conjunction with the option <literal>type</literal>.
</para>
<anchor id="includegraphics-command"/></listitem></varlistentry><varlistentry><term><literal>command</literal>
</term><listitem><para>Specify a command to be applied to this file.  Only use this in
conjunction with the option <literal>type</literal>.  See <link linkend="Command-line-options">Command line options</link>
for a discussion of enabling the <literal>\write18</literal> functionality to run
external commands.
</para>
</listitem></varlistentry></variablelist>

</sect2>
<sect2 label="22.3.2" id="_005crotatebox">
<title><literal>\rotatebox</literal></title>

<indexterm role="cp"><primary>rotation</primary></indexterm>
<indexterm role="cp"><primary>rotating graphics</primary></indexterm>
<indexterm role="cp"><primary>rotating text</primary></indexterm>
<indexterm role="fn"><primary>\rotatebox</primary></indexterm>

<para>Synopsis for <filename>graphics</filename> package:
</para>
<screen>\rotatebox{<replaceable>angle</replaceable>}{<replaceable>material</replaceable>}
</screen>
<para>Synopses for <filename>graphicx</filename> package:
</para>
<screen>\rotatebox{<replaceable>angle</replaceable>}{<replaceable>material</replaceable>}
\rotatebox[<replaceable>key-value list</replaceable>]{<replaceable>angle</replaceable>}{<replaceable>material</replaceable>}
</screen>
<para>Put <replaceable>material</replaceable> in a box and rotate it <replaceable>angle</replaceable> degrees
counterclockwise.
</para>
<para>This example rotates the table column heads forty five degrees.
</para>
<screen>\begin{tabular}{ll}
  \rotatebox{45}{Character} &amp;\rotatebox{45}{NATO phonetic}   \\
  A                         &amp;AL-FAH  \\
  B                         &amp;BRAH-VOH
\end{tabular}
</screen>
<para>The <replaceable>material</replaceable> can be anything that goes in a box, including a graphic.
</para>
<screen>  \rotatebox[origin=c]{45}{\includegraphics[width=1in]{lion}}
</screen>
<para>To place the rotated material, the first step is that &latex; sets
<replaceable>material</replaceable> in a box, with a reference point on the left baseline.
The second step is the rotation, by default about the reference point.
The third step is that &latex; computes a box to bound the rotated
material.  Fourth, &latex; moves this box horizontally so that the left
edge of this new bounding box coincides with the left edge of the box
from the first step (they need not coincide vertically).  This new
bounding box, in its new position, is what &latex; uses as the box when
typesetting this material.
</para>
<para>If you use the <filename>graphics</filename> package then the rotation is about the
reference point of the box.  If you use the <filename>graphicx</filename> package then
then these are the options that can go in the <replaceable>key-value list</replaceable>, but
note that you can get the same effect without needing this package,
except for the <literal>x</literal> and <literal>y</literal> options (see <link linkend="_005cincludegraphics">\includegraphics</link>).
</para>
<variablelist><varlistentry><term><literal>origin</literal>
</term><listitem><para>The point of the <replaceable>material</replaceable>&#8217;s box about which the rotation happens.
Possible value is any string containing one or two of: <literal>l</literal> for
left, <literal>r</literal> for right, <literal>b</literal> for bottom, <literal>c</literal> for center,
<literal>t</literal> for top, and <literal>B</literal> for baseline.  Thus, the first line here
</para>
<screen>\includegraphics[angle=180,origin=c]{moon}
\includegraphics[angle=180,origin=lB]{LeBateau}
</screen>
<para>will turn the picture upside down from the center while the second will
turn its picture upside down about its left baseline.  (The character
<literal>c</literal> gives the horizontal center in <literal>bc</literal> or <literal>tc</literal> but gives
the vertical center in <literal>lc</literal> or <literal>rc</literal>.)  The default is
<literal>lB</literal>.
</para>
</listitem></varlistentry><varlistentry><term><literal>x, y</literal>
</term><listitem><para>Specify an arbitrary point of rotation with
<literal>\rotatebox[x=<replaceable>&tex; dimension</replaceable>,y=<replaceable>&tex;
dimension</replaceable>]{...}</literal> (see <link linkend="Units-of-length">Units of length</link>).  These give the offset
from the box&#8217;s reference point.
</para>
</listitem></varlistentry><varlistentry><term><literal>units</literal>
</term><listitem><para>This key allows you to change the default of degrees counterclockwise.
Setting <literal>units=-360</literal> changes the direction to degrees clockwise and
setting <literal>units=6.283185</literal> changes to radians counterclockwise.
</para>
</listitem></varlistentry></variablelist>

</sect2>
<sect2 label="22.3.3" id="_005cscalebox">
<title><literal>\scalebox</literal></title>

<indexterm role="cp"><primary>graphics, scaling</primary></indexterm>
<indexterm role="cp"><primary>graphics, resizing</primary></indexterm>
<indexterm role="cp"><primary>scaling</primary></indexterm>
<indexterm role="cp"><primary>resizing</primary></indexterm>
<indexterm role="cp"><primary>text, scaling</primary></indexterm>
<indexterm role="cp"><primary>text, resizing</primary></indexterm>
<indexterm role="fn"><primary>\scalebox</primary></indexterm>
<indexterm role="fn"><primary>\reflectbox</primary></indexterm>

<para>Synopses:
</para>
<screen>\scalebox{<replaceable>horizontal factor</replaceable>}{<replaceable>material</replaceable>}
\scalebox{<replaceable>horizontal factor</replaceable>}[<replaceable>vertical factor</replaceable>]{<replaceable>material</replaceable>}
\reflectbox{<replaceable>material</replaceable>}
</screen>
<para>Scale the <replaceable>material</replaceable>.
</para>
<para>This example halves the size, both horizontally and vertically, of the
first text and doubles the size of the second.
</para>
<screen>\scalebox{0.5}{DRINK ME} and \scalebox{2.0}{Eat Me}
</screen>
<para>If you do not specify the optional <replaceable>vertical factor</replaceable> then it
defaults to the same value as the <replaceable>horizontal factor</replaceable>.
</para>
<para>You can use this command to resize a graphic, as here.
</para>
<screen>\scalebox{0.5}{\includegraphics{lion}}
</screen>
<para>If you use the <filename>graphicx</filename> package then you can accomplish the same
thing with optional arguments to <literal>\includegraphics</literal>
(see <link linkend="_005cincludegraphics">\includegraphics</link>).
</para>
<para>The <literal>\reflectbox</literal> command abbreviates
<literal>\scalebox{-1}[1]{<replaceable>material</replaceable>}</literal>.  Thus, <literal>Able was
I\reflectbox{Able was I}</literal> will show the phrase &#8216;<literal>Able was I</literal>&#8217;
immediately followed by its mirror reflection.
</para>

</sect2>
<sect2 label="22.3.4" id="_005cresizebox">
<title><literal>\resizebox</literal></title>

<indexterm role="cp"><primary>graphics, scaling</primary></indexterm>
<indexterm role="cp"><primary>graphics, resizing</primary></indexterm>
<indexterm role="cp"><primary>scaling</primary></indexterm>
<indexterm role="cp"><primary>resizing</primary></indexterm>
<indexterm role="cp"><primary>text, scaling</primary></indexterm>
<indexterm role="cp"><primary>text, resizing</primary></indexterm>
<indexterm role="fn"><primary>\resizebox</primary></indexterm>

<para>Synopses:
</para>
<screen>\resizebox{<replaceable>horizontal length</replaceable>}{<replaceable>vertical length</replaceable>}{<replaceable>material</replaceable>}
\resizebox*{<replaceable>horizontal length</replaceable>}{<replaceable>vertical length</replaceable>}{<replaceable>material</replaceable>}
</screen>
<para>Given a size, such as <literal>3cm</literal>, transform <replaceable>material</replaceable> to make it
that size.  If either <replaceable>horizontal length</replaceable> or <replaceable>vertical length</replaceable>
is an exclamation point&#160;<literal>!</literal>  then the other argument is used
to determine a scale factor for both directions.
</para>
<para>This example makes the graphic be a half inch wide and scales it
vertically by the same factor to keep it from being distorted.
</para>
<screen>\resizebox{0.5in}{!}{\includegraphics{lion}}
</screen>
<para>The unstarred form <literal>\resizebox</literal> takes <replaceable>vertical length</replaceable> to be
the box&#8217;s height while the starred form <literal>\resizebox*</literal> takes it to
be height+depth.  For instance, make the text have a height+depth of a
quarter inch with <literal>\resizebox*{!}{0.25in}{\parbox{1in}{This
box has both height and depth.}}</literal>.
</para>
<para>You can use <literal>\depth</literal>, <literal>\height</literal>, <literal>\totalheight</literal>, and
<literal>\width</literal> to refer to the original size of the box.  Thus, make the
text two inches wide but keep the original height with
<literal>\resizebox{2in}{\height}{Two inches}</literal>.
</para>

</sect2>
</sect1>
</chapter>
<chapter label="23" id="Special-insertions">
<title>Special insertions</title>

<indexterm role="cp"><primary>special insertions</primary></indexterm>
<indexterm role="cp"><primary>insertions of special characters</primary></indexterm>

<para>&latex; provides commands for inserting characters that have a
special meaning do not correspond to simple characters you can type.
</para>


<sect1 label="23.1" id="Reserved-characters">
<title>Reserved characters</title>

<indexterm role="cp"><primary>reserved characters</primary></indexterm>
<indexterm role="cp"><primary>characters, reserved</primary></indexterm>
<indexterm role="cp"><primary>special characters</primary></indexterm>
<indexterm role="cp"><primary>characters, special</primary></indexterm>
<para>&latex; sets aside the following characters for special purposes. For
example, the percent sign&#160;<literal>%</literal> is for comments.  They are
called <firstterm>reserved characters</firstterm> or <firstterm>special characters</firstterm>.
</para>
<screen># $ % &amp; { } _ ~ ^ \ 
</screen>
<indexterm role="fn"><primary>\#</primary></indexterm>
<indexterm role="fn"><primary>\$</primary></indexterm>
<indexterm role="fn"><primary>\%</primary></indexterm>
<indexterm role="fn"><primary>\&amp;</primary></indexterm>
<indexterm role="fn"><primary>\_</primary></indexterm>
<indexterm role="fn"><primary>\{</primary></indexterm>
<indexterm role="fn"><primary>\}</primary></indexterm>
<para>If you want a reserved character to be printed as itself, in the text
body font, for all but the final three characters in that list simply
put a backslash&#160;<literal>\</literal> in front of the character.  Thus,
typing <literal>\$1.23</literal> will produce&#160;<literal>$1.23</literal> in your output.
</para>
<indexterm role="fn"><primary>\~</primary></indexterm>
<indexterm role="fn"><primary>\^</primary></indexterm>
<indexterm role="fn"><primary>\textbackslash</primary></indexterm>
<para>As to the last three characters, to get a tilde in the text body font
use <literal>\~{}</literal> (omitting the curly braces would result in the next
character receiving a tilde accent).  Similarly, to get a get a text
body font circumflex use <literal>\^{}</literal>.  To get a backslash in the font
of the text body, enter <literal>\textbackslash{}</literal>.
</para>
<para>To produce the reserved characters in a typewriter font use
<literal>\verb!!</literal> as below (the double backslash&#160;<literal>\\</literal> is only
there to split the lines).
</para>
<screen>\begin{center}
  \# \$ \% \&amp; \{ \} \_ \~{} \^{} \textbackslash \\
  \verb!# $ % &amp; { } _ ~ ^ \!
\end{center}
</screen>

</sect1>
<sect1 label="23.2" id="Upper-and-lower-case">
<title>Upper and lower case</title>

<indexterm role="cp"><primary>uppercase</primary></indexterm>
<indexterm role="cp"><primary>lowercase</primary></indexterm>
<indexterm role="cp"><primary>characters, case of</primary></indexterm>

<para>Synopsis:
</para>
<screen>\uppercase{<replaceable>text</replaceable>}
\lowercase{<replaceable>text</replaceable>}
\MakeUppercase{<replaceable>text</replaceable>}
\MakeLowercase{<replaceable>text</replaceable>}
</screen>
<para>Change the case of characters.  The &tex; primitives commands
<literal>\uppercase</literal> and <literal>\lowercase</literal> only work for American
characters. The &latex; commands <literal>\MakeUppercase</literal> and
<literal>\MakeLowercase</literal> commands also change characters accessed by
commands such as <literal>\ae</literal> or <literal>\aa</literal>.  The commands
<literal>\MakeUppercase</literal> and <literal>\MakeLowercase</literal> are robust but they have
moving arguments (see <link linkend="_005cprotect">\protect</link>).
</para>
<para>These commands do not change the case of letters used in the name of a
command within <replaceable>text</replaceable>.  But they do change the case of every other
Latin letter inside the argument <replaceable>text</replaceable>.  Thus,
<literal>\MakeUppercase{Let $y=f(x)$</literal>} produces &#8216;<literal>LET Y=F(X)</literal>&#8217;.  Another
example is that the name of an environment will be changed, so that
<literal>\MakeUppercase{\begin{tabular} ... \end{tabular}}</literal> will
produce an error because the first half is changed to
<literal>\begin{TABULAR}</literal>.
</para>
<para>&latex; uses the same fixed table for changing case throughout a
document, The table used is designed for the font encoding T1; this
works well with the standard &tex; fonts for all Latin alphabets but
will cause problems when using other alphabets.
</para>
<para>To change the case of text that results from a macro inside <replaceable>text</replaceable>
you need to do expansion.  Here the <literal>\Schoolname</literal> produces
&#8216;<literal>COLLEGE OF MATHEMATICS</literal>&#8217;.
</para>
<screen>\newcommand{\schoolname}{College of Mathematics}
\newcommand{\Schoolname}{\expandafter\MakeUppercase
                           <!-- /@w -->\expandafter{\schoolname}}
</screen>
<indexterm role="cp"><primary>package, <literal>textcase</literal></primary></indexterm>
<indexterm role="cp"><primary><literal>textcase</literal> package</primary></indexterm>

<para>The <filename>textcase</filename> package brings some of the missing feature of the
standard &latex; commands <literal>\MakeUppercase</literal> and
<literal>\MakeLowerCase</literal>.
</para>
<indexterm role="cp"><primary>package, <literal>mfirstuc</literal></primary></indexterm>
<indexterm role="cp"><primary><literal>mfirstuc</literal> package</primary></indexterm>
 
<para>To uppercase only the first letter of words, you can use the package
<filename>mfirstuc</filename>.
</para>

</sect1>
<sect1 label="23.3" id="Symbols-by-font-position">
<title>Symbols by font position</title>

<indexterm role="fn"><primary>\symbol</primary></indexterm>
<indexterm role="cp"><primary>accessing any character of a font</primary></indexterm>
<indexterm role="cp"><primary>font symbols, by number</primary></indexterm>

<para>You can access any character of the current font using its number with
the <literal>\symbol</literal> command. For example, the visible space character
used in the <literal>\verb*</literal> command has the code decimal 32, so it can be
typed as <literal>\symbol{32}</literal>.
</para>
<para>You can also specify numbers in octal (base 8) by using a <literal>'</literal>
prefix, or hexadecimal (base 16) with a <literal>&quot;</literal> prefix, so the previous
example could also be written as <literal>\symbol{'40}</literal> or
<literal>\symbol{&quot;20}</literal>.
</para>

</sect1>
<sect1 label="23.4" id="Text-symbols">
<title>Text symbols</title>

<indexterm role="cp"><primary>text symbols</primary></indexterm>
<indexterm role="cp"><primary>symbols, text</primary></indexterm>

<indexterm role="fn"><primary>textcomp package</primary></indexterm>
<para>&latex; provides commands to generate a number of non-letter symbols in
running text.  Some of these, especially the more obscure ones, are not
available in OT1. Unless you are using Xe&latex; or Lua&latex; then
you may need to load the <literal>textcomp</literal> package.
</para>
<variablelist><varlistentry><term><indexterm role="fn"><primary>\copyright</primary></indexterm><literal>\copyright</literal>
</term><term><indexterm role="fn"><primary>\textcopyright</primary></indexterm><literal>\textcopyright</literal>
</term><listitem><indexterm role="cp"><primary>copyright symbol</primary></indexterm>
<para>&#169; The copyright symbol.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\dag</primary></indexterm><literal>\dag</literal>
</term><listitem><indexterm role="cp"><primary>dagger, in text</primary></indexterm>
<para>&#x2020; The dagger symbol (in text).
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\ddag</primary></indexterm><literal>\ddag</literal>
</term><listitem><indexterm role="cp"><primary>double dagger, in text</primary></indexterm>
<para>&#x2021; The double dagger symbol (in text).
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\LaTeX</primary></indexterm><literal>\LaTeX</literal>
</term><listitem><indexterm role="cp"><primary>&latex; logo</primary></indexterm>
<indexterm role="cp"><primary>logo, &latex;</primary></indexterm>
<para>The &latex; logo.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\LaTeXe</primary></indexterm><literal>\LaTeXe</literal>
</term><listitem><indexterm role="cp"><primary>&latex;2e logo</primary></indexterm>
<indexterm role="cp"><primary>logo, &latex;2e</primary></indexterm>
<para>The &latex;2e logo.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\guillemotleft (&#171;)</primary></indexterm><literal>\guillemotleft (&#171;)</literal>
</term><term><indexterm role="fn"><primary>\guillemotright (&#187;)</primary></indexterm><literal>\guillemotright (&#187;)</literal>
</term><term><indexterm role="fn"><primary>\guilsinglleft (&#8249;)</primary></indexterm><literal>\guilsinglleft (&#8249;)</literal>
</term><term><indexterm role="fn"><primary>\guilsinglright (&#8250;)</primary></indexterm><literal>\guilsinglright (&#8250;)</literal>
</term><listitem><indexterm role="cp"><primary>double guillemets</primary></indexterm>
<indexterm role="cp"><primary>single guillemets</primary></indexterm>
<indexterm role="cp"><primary>left angle quotation marks</primary></indexterm>
<indexterm role="cp"><primary>right angle quotation marks</primary></indexterm>
<indexterm role="cp"><primary>double angle quotation marks</primary></indexterm>
<indexterm role="cp"><primary>single angle quotation marks</primary></indexterm>
<indexterm role="cp"><primary>French quotation marks</primary></indexterm>
<indexterm role="cp"><primary>quotation marks, French</primary></indexterm>
<para>&#171;, &#187;, &#8249;, &#8250;
Double and single angle quotation marks, commonly used in French.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\ldots</primary></indexterm><literal>\ldots</literal>
</term><term><indexterm role="fn"><primary>\dots</primary></indexterm><literal>\dots</literal>
</term><term><indexterm role="fn"><primary>\textellipsis</primary></indexterm><literal>\textellipsis</literal>
</term><listitem><indexterm role="cp"><primary>ellipsis</primary></indexterm>
<para>&#8230;  An ellipsis (three dots at the baseline):  <literal>\ldots</literal>
and <literal>\dots</literal> also work in math mode.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\lq</primary></indexterm><literal>\lq</literal>
</term><listitem><indexterm role="cp"><primary>left quote</primary></indexterm>
<indexterm role="cp"><primary>opening quote</primary></indexterm>
<para>&#8216; Left (opening) quote.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\P</primary></indexterm><literal>\P</literal>
</term><term><indexterm role="fn"><primary>\textparagraph</primary></indexterm><literal>\textparagraph</literal>
</term><listitem><indexterm role="cp"><primary>paragraph symbol</primary></indexterm>
<indexterm role="cp"><primary>pilcrow</primary></indexterm>
<para>&#x00B6; Paragraph sign (pilcrow).
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\pounds</primary></indexterm><literal>\pounds</literal>
</term><term><indexterm role="fn"><primary>\textsterling</primary></indexterm><literal>\textsterling</literal>
</term><listitem><indexterm role="cp"><primary>pounds symbol</primary></indexterm>
<indexterm role="cp"><primary>sterling symbol</primary></indexterm>
<para>&#163; English pounds sterling.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\quotedblbase (&#8222;)</primary></indexterm><literal>\quotedblbase (&#8222;)</literal>
</term><term><indexterm role="fn"><primary>\quotesinglbase (&#8218;)</primary></indexterm><literal>\quotesinglbase (&#8218;)</literal>
</term><listitem><indexterm role="cp"><primary>double low-9 quotation mark</primary></indexterm>
<indexterm role="cp"><primary>single low-9 quotation mark</primary></indexterm>
<indexterm role="cp"><primary>low-9 quotation marks, single and double</primary></indexterm>
<para>&#8222; and &#8218;
Double and single quotation marks on the baseline.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\rq</primary></indexterm><literal>\rq</literal>
</term><listitem><indexterm role="cp"><primary>right quote</primary></indexterm>
<indexterm role="cp"><primary>closing quote</primary></indexterm>
<para>&#8217; Right (closing) quote.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\S</primary></indexterm><literal>\S</literal>
</term><term><indexterm role="fn"><primary>\textsection</primary></indexterm><literal>\textsection</literal>
</term><listitem><indexterm role="cp"><primary>section symbol</primary></indexterm>
<para>&#x00A7; Section sign.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\TeX</primary></indexterm><literal>\TeX</literal>
</term><listitem><indexterm role="cp"><primary>&tex; logo</primary></indexterm>
<indexterm role="cp"><primary>logo, &tex;</primary></indexterm>
<para>The &tex; logo.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\textasciicircum</primary></indexterm><literal>\textasciicircum</literal>
</term><listitem><indexterm role="cp"><primary>circumflex, ASCII, in text</primary></indexterm>
<indexterm role="cp"><primary>ASCII circumflex, in text</primary></indexterm>
<para>^ ASCII circumflex.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\textasciitilde</primary></indexterm><literal>\textasciitilde</literal>
</term><listitem><indexterm role="cp"><primary>tilde, ASCII, in text</primary></indexterm>
<indexterm role="cp"><primary>ASCII tilde, in text</primary></indexterm>
<para>~ ASCII tilde.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\textasteriskcentered</primary></indexterm><literal>\textasteriskcentered</literal>
</term><listitem><indexterm role="cp"><primary>asterisk, centered, in text</primary></indexterm>
<indexterm role="cp"><primary>centered asterisk, in text</primary></indexterm>
<para>* Centered asterisk.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\textbackslash</primary></indexterm><literal>\textbackslash</literal>
</term><listitem><indexterm role="cp"><primary>backslash, in text</primary></indexterm>
<para>\ Backslash.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\textbar</primary></indexterm><literal>\textbar</literal>
</term><listitem><indexterm role="cp"><primary>vertical bar, in text</primary></indexterm>
<indexterm role="cp"><primary>bar, vertical, in text</primary></indexterm>
<para>| Vertical bar.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\textbardbl</primary></indexterm><literal>\textbardbl</literal>
</term><listitem><indexterm role="cp"><primary>vertical bar, double, in text</primary></indexterm>
<indexterm role="cp"><primary>bar, double vertical, in text</primary></indexterm>
<indexterm role="cp"><primary>double vertical bar, in text</primary></indexterm>
<para>&#x23F8; Double vertical bar.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\textbigcircle</primary></indexterm><literal>\textbigcircle</literal>
</term><listitem><indexterm role="cp"><primary>big circle symbols, in text</primary></indexterm>
<indexterm role="cp"><primary>circle symbol, big, in text</primary></indexterm>
<para>&#x25EF; Big circle symbol.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\textbraceleft</primary></indexterm><literal>\textbraceleft</literal>
</term><listitem><indexterm role="cp"><primary>left brace, in text</primary></indexterm>
<indexterm role="cp"><primary>brace, left, in text</primary></indexterm>
<para>{ Left brace.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\textbraceright</primary></indexterm><literal>\textbraceright</literal>
</term><listitem><indexterm role="cp"><primary>right brace, in text</primary></indexterm>
<indexterm role="cp"><primary>brace, right, in text</primary></indexterm>
<para>} Right brace.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\textbullet</primary></indexterm><literal>\textbullet</literal>
</term><listitem><indexterm role="cp"><primary>bullet, in text</primary></indexterm>
<para>&#8226; Bullet.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\textcircled{<replaceable>letter</replaceable>}</primary></indexterm><literal>\textcircled{<replaceable>letter</replaceable>}</literal>
</term><listitem><indexterm role="cp"><primary>circled letter, in text</primary></indexterm>
<para>&#x24B6; Circle around <replaceable>letter</replaceable>.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\textcompwordmark</primary></indexterm><literal>\textcompwordmark</literal>
</term><term><indexterm role="fn"><primary>\textcapitalcompwordmark</primary></indexterm><literal>\textcapitalcompwordmark</literal>
</term><term><indexterm role="fn"><primary>\textascendercompwordmark</primary></indexterm><literal>\textascendercompwordmark</literal>
</term><listitem><indexterm role="cp"><primary>composite word mark, in text</primary></indexterm>
<indexterm role="cp"><primary>cap height</primary></indexterm>
<indexterm role="cp"><primary>ascender height</primary></indexterm>
<para>Used to separate letters that would normally ligature. For example,
<literal>f\textcompwordmark i</literal> produces &#8216;<literal>fi</literal>&#8217; without a ligature. This
is most useful in non-English languages.  The
<literal>\textcapitalcompwordmark</literal> form has the cap height of the font
while the <literal>\textascendercompwordmark</literal> form has the ascender height.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\textdagger</primary></indexterm><literal>\textdagger</literal>
</term><listitem><indexterm role="cp"><primary>dagger, in text</primary></indexterm>
<para>&#x2020; Dagger.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\textdaggerdbl</primary></indexterm><literal>\textdaggerdbl</literal>
</term><listitem><indexterm role="cp"><primary>dagger, double, in text</primary></indexterm>
<indexterm role="cp"><primary>double dagger, in text</primary></indexterm>
<para>&#x2021; Double dagger.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\textdollar (or <literal>\$</literal>)</primary></indexterm><literal>\textdollar (or <literal>\$</literal>)</literal>
</term><listitem><indexterm role="cp"><primary>dollar sign</primary></indexterm>
<indexterm role="cp"><primary>currency, dollar</primary></indexterm>
<para>$ Dollar sign.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\textemdash (or <literal>---</literal>)</primary></indexterm><literal>\textemdash (or <literal>---</literal>)</literal>
</term><listitem><indexterm role="cp"><primary>em-dash</primary></indexterm>
<para>&#8212; Em-dash (used for punctuation, as in
<literal>The playoffs --- if you are fortunate enough to make the playoffs ---
is more like a sprint.</literal>).
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\textendash (or <literal>--</literal>)</primary></indexterm><literal>\textendash (or <literal>--</literal>)</literal>
</term><listitem><indexterm role="cp"><primary>e-dash</primary></indexterm>
<para>&#8211; En-dash (used for ranges, as in <literal>See pages 12--14</literal>).
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\texteuro</primary></indexterm><literal>\texteuro</literal>
</term><listitem><indexterm role="cp"><primary>euro symbol</primary></indexterm>
<indexterm role="cp"><primary>currency, euro</primary></indexterm>
<indexterm role="cp"><primary>package, <literal>eurosym</literal></primary></indexterm>
<indexterm role="cp"><primary><literal>eurosym</literal> package</primary></indexterm>

<para>The Euro symbol: &#8364;. For an alternative glyph design, try the
<filename>eurosym</filename> package; also, most fonts nowadays come with their own
Euro symbol (Unicode U+20AC).
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\textexclamdown (or <literal>!`</literal>)</primary></indexterm><literal>\textexclamdown (or <literal>!`</literal>)</literal>
</term><listitem><indexterm role="cp"><primary>exclamation point, upside-down</primary></indexterm>
<para>&#161; Upside down exclamation point.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\textgreater</primary></indexterm><literal>\textgreater</literal>
</term><listitem><indexterm role="cp"><primary>greater than symbol, in text</primary></indexterm>
<para>&gt; Greater than symbol.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\textless</primary></indexterm><literal>\textless</literal>
</term><listitem><indexterm role="cp"><primary>less than symbol, in text</primary></indexterm>
<para>&lt; Less than symbol.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\textleftarrow</primary></indexterm><literal>\textleftarrow</literal>
</term><listitem><indexterm role="cp"><primary>arrow, left, in text</primary></indexterm>
<indexterm role="cp"><primary>left arrow, in text</primary></indexterm>
<para>&#x2190; Left arrow.   
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\textordfeminine</primary></indexterm><literal>\textordfeminine</literal>
</term><term><indexterm role="fn"><primary>\textordmasculine</primary></indexterm><literal>\textordmasculine</literal>
</term><listitem><indexterm role="cp"><primary>feminine ordinal symbol</primary></indexterm>
<indexterm role="cp"><primary>masculine ordinal symbol</primary></indexterm>
<indexterm role="cp"><primary>ordinals, feminine and masculine</primary></indexterm>
<indexterm role="cp"><primary>Spanish ordinals, feminine and masculine</primary></indexterm>
<para>&#170;, &#186; Feminine and masculine ordinal symbols.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\textperiodcentered</primary></indexterm><literal>\textperiodcentered</literal>
</term><listitem><indexterm role="cp"><primary>period, centered, in text</primary></indexterm>
<indexterm role="cp"><primary>centered period, in text</primary></indexterm>
<para>&#x00B7; Centered period.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\textquestiondown (or <literal>?`</literal>)</primary></indexterm><literal>\textquestiondown (or <literal>?`</literal>)</literal>
</term><listitem><indexterm role="cp"><primary>question mark, upside-down</primary></indexterm>
<para>&#191; Upside down question mark.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\textquotedblleft (or <literal>``</literal>)</primary></indexterm><literal>\textquotedblleft (or <literal>``</literal>)</literal>
</term><listitem><indexterm role="cp"><primary>left quote, double</primary></indexterm>
<indexterm role="cp"><primary>double left quote</primary></indexterm>
<para>&#8220; Double left quote.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\textquotedblright (or <literal>''</literal>)</primary></indexterm><literal>\textquotedblright (or <literal>''</literal>)</literal>
</term><listitem><indexterm role="cp"><primary>right quote, double</primary></indexterm>
<indexterm role="cp"><primary>double right quote</primary></indexterm>
<para>&#8221; Double right quote.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\textquoteleft (or <literal>`</literal>)</primary></indexterm><literal>\textquoteleft (or <literal>`</literal>)</literal>
</term><listitem><indexterm role="cp"><primary>left quote, single</primary></indexterm>
<indexterm role="cp"><primary>single left quote</primary></indexterm>
<para>&#8216; Single left quote.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\textquoteright (or <literal>'</literal>)</primary></indexterm><literal>\textquoteright (or <literal>'</literal>)</literal>
</term><listitem><indexterm role="cp"><primary>right quote, single</primary></indexterm>
<indexterm role="cp"><primary>single right quote</primary></indexterm>
<para>&#8217; Single right quote.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\textquotesingle</primary></indexterm><literal>\textquotesingle</literal>
</term><listitem><indexterm role="cp"><primary>quote, single straight</primary></indexterm>
<indexterm role="cp"><primary>straight single quote</primary></indexterm>
<indexterm role="cp"><primary>single quote, straight</primary></indexterm>
<para>&#x0027; Straight single quote.  (From TS1 encoding.)  
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\textquotestraightbase</primary></indexterm><literal>\textquotestraightbase</literal>
</term><term><indexterm role="fn"><primary>\textquotestraightdblbase</primary></indexterm><literal>\textquotestraightdblbase</literal>
</term><listitem><indexterm role="cp"><primary>quote, straight base</primary></indexterm>
<indexterm role="cp"><primary>straight quote, base</primary></indexterm>
<indexterm role="cp"><primary>double quote, straight base</primary></indexterm>
<indexterm role="cp"><primary>straight double quote, base</primary></indexterm>
<!-- Unicode doesn't have these https://en.wikipedia.org/wiki/Quotation_mark -->
<para>Single and double straight quotes on the baseline.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\textregistered</primary></indexterm><literal>\textregistered</literal>
</term><listitem><indexterm role="cp"><primary>registered symbol</primary></indexterm>
<para>&#174; Registered symbol.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\textrightarrow</primary></indexterm><literal>\textrightarrow</literal>
</term><listitem><indexterm role="cp"><primary>arrow, right, in text</primary></indexterm>
<indexterm role="cp"><primary>right arrow, in text</primary></indexterm>
<para>&#x2192; Right arrow.  
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\textthreequartersemdash</primary></indexterm><literal>\textthreequartersemdash</literal>
</term><listitem><indexterm role="cp"><primary>three-quarters em-dash</primary></indexterm>
<indexterm role="cp"><primary>em-dash, three-quarters</primary></indexterm>
<para>&#xFE58; &#8220;Three-quarters&#8221; em-dash, between en-dash and em-dash.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\texttrademark</primary></indexterm><literal>\texttrademark</literal>
</term><listitem><indexterm role="cp"><primary>trademark symbol</primary></indexterm>
<para>&#x2122; Trademark symbol.
</para>
<!-- ?? Diff from \textthreequartersemdash?  In Unicode? -->
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\texttwelveudash</primary></indexterm><literal>\texttwelveudash</literal>
</term><listitem><indexterm role="cp"><primary>two-thirds em-dash</primary></indexterm>
<indexterm role="cp"><primary>em-dash, two-thirds</primary></indexterm>
<para>&#xFE58; &#8220;Two-thirds&#8221; em-dash, between en-dash and em-dash.   
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\textunderscore</primary></indexterm><literal>\textunderscore</literal>
</term><listitem><indexterm role="cp"><primary>underscore, in text</primary></indexterm>
<para>_ Underscore.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>\textvisiblespace</primary></indexterm><literal>\textvisiblespace</literal>
</term><listitem><indexterm role="cp"><primary>visible space symbol, in text</primary></indexterm>
<para>&#x2423; Visible space symbol.  
</para>
</listitem></varlistentry></variablelist>

</sect1>
<sect1 label="23.5" id="Accents">
<title>Accents</title>

<indexterm role="cp"><primary>accents</primary></indexterm>
<indexterm role="cp"><primary>characters, accented</primary></indexterm>
<indexterm role="cp"><primary>letters, accented</primary></indexterm>

<indexterm role="cp"><primary>package, <literal>babel</literal></primary></indexterm>
<indexterm role="cp"><primary><literal>babel</literal> package</primary></indexterm>

<indexterm role="cp"><primary>package, <literal>polyglossia</literal></primary></indexterm>
<indexterm role="cp"><primary><literal>polyglossia</literal> package</primary></indexterm>

<indexterm role="cp"><primary>multilingual support</primary></indexterm>
<para>&latex; has wide support for many of the world&#8217;s scripts and languages,
through the <literal>babel</literal> package and related support if you are using
pdf&latex;, or <filename>polyglossia</filename> if you are using Xe&latex; or
Lua&latex;.  This section does not cover that support. It only lists
the core &latex; commands for creating accented characters.  The
<literal>\capital...</literal> commands shown here produce alternative forms for use
with capital letters.  These are not available with OT1.
</para>
<para>Below, to make them easier to find, the accents are all illustrated with
lowercase &#8216;<literal>o</literal>&#8217;.
</para>
<indexterm role="fn"><primary>\i (dotless i)</primary></indexterm>
<indexterm role="cp"><primary>dotless i</primary></indexterm>
<para>Note that <literal>\i</literal> produces a dotless i,
<!-- @dotless{i}, -->
<indexterm role="fn"><primary>\j (dotless j)</primary></indexterm>
<indexterm role="cp"><primary>dotless j</primary></indexterm>
and <literal>\j</literal> produces a dotless j.
<!-- @dotless{j}. -->
These are often used in place of their dotted counterparts when they are
accented.
</para>
<variablelist><varlistentry><term><literal>\&quot;</literal>
</term><term><literal>\capitaldieresis</literal>
</term><listitem><indexterm role="fn"><primary>\&quot; (umlaut accent)</primary></indexterm>
<indexterm role="fn"><primary>\capitaldieresis</primary></indexterm>
<indexterm role="cp"><primary>umlaut accent</primary></indexterm>
<indexterm role="cp"><primary>dieresis accent</primary></indexterm>
<para>&#246; Umlaut (dieresis).
</para>
</listitem></varlistentry><varlistentry><term><literal>\'</literal>
</term><term><literal>\capitalacute</literal>
</term><listitem><indexterm role="fn"><primary>\' (acute accent)</primary></indexterm>
<indexterm role="fn"><primary>\capitalacute</primary></indexterm>
<indexterm role="cp"><primary>acute accent</primary></indexterm>
<para>&#243; Acute accent.
</para>
</listitem></varlistentry><varlistentry><term><literal>\.</literal>
</term><listitem><indexterm role="fn"><primary>\. (dot-over accent)</primary></indexterm>
<indexterm role="cp"><primary>dot accent</primary></indexterm>
<indexterm role="cp"><primary>dot-over accent</primary></indexterm>
<para>&#559;  Dot accent.
</para>
</listitem></varlistentry><varlistentry><term><literal>\=</literal>
</term><term><literal>\capitalmacron</literal>
</term><listitem><indexterm role="fn"><primary>\= (macron accent)</primary></indexterm>
<indexterm role="fn"><primary>\capitalmacron</primary></indexterm>
<indexterm role="cp"><primary>macron accent</primary></indexterm>
<indexterm role="cp"><primary>overbar accent</primary></indexterm>
<indexterm role="cp"><primary>bar-over accent</primary></indexterm>
<para>&#333; Macron (overbar) accent.
</para>
</listitem></varlistentry><varlistentry><term><literal>\^</literal>
</term><term><literal>\capitalcircumflex</literal>
</term><listitem><indexterm role="fn"><primary>\^ (circumflex accent)</primary></indexterm>
<indexterm role="fn"><primary>\capitalcircumflex</primary></indexterm>
<indexterm role="cp"><primary>circumflex accent</primary></indexterm>
<indexterm role="cp"><primary>hat accent</primary></indexterm>
<para>&#244; Circumflex (hat) accent.
</para>
</listitem></varlistentry><varlistentry><term><literal>\`</literal>
</term><term><literal>\capitalgrave</literal>
</term><listitem><indexterm role="fn"><primary>\` (grave accent)</primary></indexterm>
<indexterm role="fn"><primary>\capitalgrave</primary></indexterm>
<indexterm role="cp"><primary>grave accent</primary></indexterm>
<para>&#242; Grave accent.
</para>
</listitem></varlistentry><varlistentry><term><literal>\~</literal>
</term><term><literal>\capitaltilde</literal>
</term><listitem><indexterm role="fn"><primary>\~ (tilde accent)</primary></indexterm>
<indexterm role="fn"><primary>\capitaltilde</primary></indexterm>
<indexterm role="cp"><primary>tilde accent</primary></indexterm>
<para>&#241; Tilde accent.
</para>
</listitem></varlistentry><varlistentry><term><literal>\b</literal>
</term><listitem><indexterm role="fn"><primary>\b (bar-under accent)</primary></indexterm>
<indexterm role="cp"><primary>bar-under accent</primary></indexterm>
<para>o_ Bar accent underneath.
</para>
<indexterm role="fn"><primary>\underbar</primary></indexterm>
<indexterm role="cp"><primary>underbar</primary></indexterm>
<para>Related to this, <literal>\underbar{<replaceable>text</replaceable>}</literal> produces a bar under
<replaceable>text</replaceable>.  The argument is always processed in LR mode
(see <link linkend="Modes">Modes</link>).  The bar is always a fixed position under the baseline,
thus crossing through descenders.  See also <literal>\underline</literal> in
<link linkend="Math-miscellany">Math miscellany</link>.
</para>
</listitem></varlistentry><varlistentry><term><literal>\c</literal>
</term><term><literal>\capitalcedilla</literal>
</term><listitem><indexterm role="fn"><primary>\c (cedilla accent)</primary></indexterm>
<indexterm role="fn"><primary>\capitalcedilla</primary></indexterm>
<indexterm role="cp"><primary>cedilla accent</primary></indexterm>
<para>&#231; Cedilla accent underneath.
</para>
</listitem></varlistentry><varlistentry><term><literal>\d</literal>
</term><term><literal>\capitaldotaccent</literal>
</term><listitem><indexterm role="fn"><primary>\d (dot-under accent)</primary></indexterm>
<indexterm role="fn"><primary>\capitaldotaccent</primary></indexterm>
<indexterm role="cp"><primary>dot-under accent</primary></indexterm>
<para>&#7885; Dot accent underneath.
</para>
</listitem></varlistentry><varlistentry><term><literal>\H</literal>
</term><term><literal>\capitalhungarumlaut</literal>
</term><listitem><indexterm role="fn"><primary>\H (Hungarian umlaut accent)</primary></indexterm>
<indexterm role="fn"><primary>\capitalhungarumlaut</primary></indexterm>
<indexterm role="cp"><primary>hungarian umlaut accent</primary></indexterm>
<para>&#337; Long Hungarian umlaut accent.
</para>
</listitem></varlistentry><varlistentry><term><literal>\k</literal>
</term><term><literal>\capitalogonek</literal>
</term><listitem><indexterm role="fn"><primary>\k (ogonek)</primary></indexterm>
<indexterm role="fn"><primary>\capitalogonek</primary></indexterm>
<indexterm role="cp"><primary>ogonek</primary></indexterm>
<para>&#491; Ogonek.  Not available in the OT1 encoding.
</para>
</listitem></varlistentry><varlistentry><term><literal>\r</literal>
</term><term><literal>\capitalring</literal>
</term><listitem><indexterm role="fn"><primary>\r (ring accent)</primary></indexterm>
<indexterm role="fn"><primary>\capitalring</primary></indexterm>
<indexterm role="cp"><primary>ring accent</primary></indexterm>
<para>o* Ring accent.
</para>
</listitem></varlistentry><varlistentry><term><literal>\t</literal>
</term><term><literal>\capitaltie</literal>
</term><term><literal>\newtie</literal>
</term><term><literal>\capitalnewtie</literal>
</term><listitem><indexterm role="fn"><primary>\t (tie-after accent)</primary></indexterm>
<indexterm role="fn"><primary>\capitaltie</primary></indexterm>
<indexterm role="fn"><primary>\newtie</primary></indexterm>
<indexterm role="fn"><primary>\capitalnewtie</primary></indexterm>
<indexterm role="cp"><primary>tie-after accent</primary></indexterm>
<para>oo[ Tie-after accent.  The <literal>\newtie</literal> form is centered in
its box.
</para>
</listitem></varlistentry><varlistentry><term><literal>\u</literal>
</term><term><literal>\capitalbreve</literal>
</term><listitem><indexterm role="fn"><primary>\u (breve accent)</primary></indexterm>
<indexterm role="fn"><primary>\capitalbreve</primary></indexterm>
<indexterm role="cp"><primary>breve accent</primary></indexterm>
<para>&#335; Breve accent.
</para>
</listitem></varlistentry><varlistentry><term><literal>\v</literal>
</term><term><literal>\capitalcaron</literal>
</term><listitem><indexterm role="fn"><primary>\v (breve accent)</primary></indexterm>
<indexterm role="fn"><primary>\capitalcaron</primary></indexterm>
<indexterm role="cp"><primary>hacek accent</primary></indexterm>
<indexterm role="cp"><primary>check accent</primary></indexterm>
<indexterm role="cp"><primary>caron accent</primary></indexterm>
<para>&#466;  H&#225;&#269;ek (check, caron) accent.
</para>
</listitem></varlistentry></variablelist>

</sect1>
<sect1 label="23.6" id="Additional-Latin-letters">
<title>Additional Latin letters</title>

<anchor id="Non_002dEnglish-characters"/><indexterm role="cp"><primary>Latin letters, additional</primary></indexterm>
<indexterm role="cp"><primary>letters, additional Latin</primary></indexterm>
<indexterm role="cp"><primary>extended Latin</primary></indexterm>
<indexterm role="cp"><primary>special characters</primary></indexterm>
<indexterm role="cp"><primary>non-English characters</primary></indexterm>
<indexterm role="cp"><primary>characters, non-English</primary></indexterm>

<para>Here are the basic &latex; commands for inserting letters beyond
A&#8211;Z that extend the Latin alphabet, used primarily in languages other
than English.
</para>
<variablelist><varlistentry><term><literal>\aa</literal>
</term><term><literal>\AA</literal>
</term><listitem><indexterm role="fn"><primary>\aa (&#229;)</primary></indexterm>
<indexterm role="fn"><primary>\AA (&#197;)</primary></indexterm>
<indexterm role="cp"><primary>aring</primary></indexterm>
<para>&#229; and &#197;.
</para>
</listitem></varlistentry><varlistentry><term><literal>\ae</literal>
</term><term><literal>\AE</literal>
</term><listitem><indexterm role="fn"><primary>\ae (&#230;)</primary></indexterm>
<indexterm role="fn"><primary>\AE (&#198;)</primary></indexterm>
<indexterm role="cp"><primary>ae ligature</primary></indexterm>
<para>&#230; and &#198;.
</para>
</listitem></varlistentry><varlistentry><term><literal>\dh</literal>
</term><term><literal>\DH</literal>
</term><listitem><indexterm role="fn"><primary>\dh (&#240;)</primary></indexterm>
<indexterm role="fn"><primary>\DH (&#208;)</primary></indexterm>
<indexterm role="cp"><primary>Icelandic eth</primary></indexterm>
<indexterm role="cp"><primary>eth, Icelandic letter</primary></indexterm>
<para>Icelandic letter eth: &#240; and &#208;. Not available with OT1
encoding, you need the <filename>fontenc</filename> package to select an alternate
font encoding, such as T1.
</para>
</listitem></varlistentry><varlistentry><term><literal>\dj</literal>
</term><term><literal>\DJ</literal>
</term><listitem><indexterm role="fn"><primary>\dj</primary></indexterm>
<indexterm role="fn"><primary>\DJ</primary></indexterm>
<para>Crossed d and D, a.k.a. capital and small letter d with stroke.  Not
available with OT1 encoding, you need the <filename>fontenc</filename> package to
select an alternate font encoding, such as T1.
</para>
</listitem></varlistentry><varlistentry><term><literal>\ij</literal>
</term><term><literal>\IJ</literal>
</term><listitem><indexterm role="fn"><primary>\ij (ij)</primary></indexterm>
<indexterm role="fn"><primary>\IJ (IJ)</primary></indexterm>
<indexterm role="cp"><primary>ij letter, Dutch</primary></indexterm>
<para>ij and IJ (except somewhat closer together than appears here).
</para>
</listitem></varlistentry><varlistentry><term><literal>\l</literal>
</term><term><literal>\L</literal>
</term><listitem><indexterm role="fn"><primary>\l (&#322;)</primary></indexterm>
<indexterm role="fn"><primary>\L (&#321;)</primary></indexterm>
<indexterm role="cp"><primary>polish l</primary></indexterm>
<para>&#322; and &#321;.
</para>
</listitem></varlistentry><varlistentry><term><literal>\ng</literal>
</term><term><literal>\NG</literal>
</term><listitem><indexterm role="fn"><primary>\ng</primary></indexterm>
<indexterm role="fn"><primary>\NG</primary></indexterm>
<para>Lappish letter eng, also used in phonetics.
</para>
</listitem></varlistentry><varlistentry><term><literal>\o</literal>
</term><term><literal>\O</literal>
</term><listitem><indexterm role="fn"><primary>\o (&#248;)</primary></indexterm>
<indexterm role="fn"><primary>\O (&#216;)</primary></indexterm>
<indexterm role="cp"><primary>oslash</primary></indexterm>
<para>&#248; and &#216;.
</para>
</listitem></varlistentry><varlistentry><term><literal>\oe</literal>
</term><term><literal>\OE</literal>
</term><listitem><indexterm role="fn"><primary>\oe (&#339;)</primary></indexterm>
<indexterm role="fn"><primary>\OE (&#338;)</primary></indexterm>
<indexterm role="cp"><primary>oe ligature</primary></indexterm>
<para>&#339; and &#338;.
</para>
</listitem></varlistentry><varlistentry><term><literal>\ss</literal>
</term><term><literal>\SS</literal>
</term><listitem><indexterm role="fn"><primary>\ss (&#223;)</primary></indexterm>
<indexterm role="fn"><primary>\SS (SS)</primary></indexterm>
<indexterm role="cp"><primary>es-zet German letter</primary></indexterm>
<indexterm role="cp"><primary>sharp S letters</primary></indexterm>
<para>&#223; and SS.
</para>
</listitem></varlistentry><varlistentry><term><literal>\th</literal>
</term><term><literal>\TH</literal>
</term><listitem><indexterm role="fn"><primary>\th (&#254;)</primary></indexterm>
<indexterm role="fn"><primary>\TH (&#222;)</primary></indexterm>
<indexterm role="cp"><primary>Icelandic thorn</primary></indexterm>
<indexterm role="cp"><primary>thorn, Icelandic letter</primary></indexterm>
<para>Icelandic letter thorn: &#254; and &#222;. Not available with OT1
encoding, you need the <filename>fontenc</filename> package to select an alternate
font encoding, such as T1.
</para>
</listitem></varlistentry></variablelist>

</sect1>
<sect1 label="23.7" id="_005crule">
<title><literal>\rule</literal></title>

<indexterm role="fn"><primary>\rule</primary></indexterm>

<para>Synopsis, one of:
</para>
<screen>\rule{<replaceable>width</replaceable>}{<replaceable>thickness</replaceable>}
\rule[<replaceable>raise</replaceable>]{<replaceable>width</replaceable>}{<replaceable>thickness</replaceable>}
</screen>
<para>Produce a <firstterm>rule</firstterm>, a filled-in rectangle.
</para>
<indexterm role="cp"><primary>Halmos symbol</primary></indexterm>
<indexterm role="cp"><primary>tombstone</primary></indexterm>
<para>This produces a rectangular blob, sometimes called a Halmos symbol,
often used to mark the end of a proof.
</para>
<screen>\newcommand{\qedsymbol}{\rule{0.4em}{2ex}}
</screen>
<indexterm role="cp"><primary>package, <literal>amsthm</literal></primary></indexterm>
<indexterm role="cp"><primary><literal>amsthm</literal> package</primary></indexterm>

<para>The <filename>amsthm</filename> package includes this command, with a somewhat
different-looking symbol.
</para>
<para>The mandatory arguments give the horizontal <replaceable>width</replaceable> and vertical
<replaceable>thickness</replaceable> of the rectangle.  They are rigid lengths
(see <link linkend="Lengths">Lengths</link>).  The optional argument <replaceable>raise</replaceable> is also a rigid
length, and tells &latex; how much to raise the rule above the
baseline, or lower it if the length is negative.
</para>
<para>This produces a line, a rectangle that is wide but not tall.
</para>
<screen>\noindent\rule{\textwidth}{0.4pt}
</screen>
<para>The line is the width of the page and 0.4&#160;points tall.  This line
thickness is common in &latex;.
</para>
<para>A rule that has zero width, or zero thickness, will not show up in the
output, but can cause &latex; to change the output around it.
See <link linkend="_005cstrut">\strut</link> for examples.
</para>

</sect1>
<sect1 label="23.8" id="_005ctoday">
<title><literal>\today</literal></title>

<indexterm role="fn"><primary>\today</primary></indexterm>
<indexterm role="cp"><primary>date, today&#8217;s</primary></indexterm>
<indexterm role="cp"><primary>today&#8217;s date</primary></indexterm>

<para>Synopsis:
</para>
<screen>\today
</screen>
<para>Produce today&#8217;s date in the format &#8216;<literal><replaceable>month</replaceable> <replaceable>dd</replaceable>,
<replaceable>yyyy</replaceable></literal>&#8217;.  An example of a date in that format is &#8216;<literal>July 4,
1976</literal>&#8217;.
</para>
<indexterm role="cp"><primary>package, <literal>babel</literal></primary></indexterm>
<indexterm role="cp"><primary><literal>babel</literal> package</primary></indexterm>

<indexterm role="cp"><primary>package, <literal>polyglossia</literal></primary></indexterm>
<indexterm role="cp"><primary><literal>polyglossia</literal> package</primary></indexterm>

<para>Multilingual packages such as <filename>babel</filename> or <filename>polyglossia</filename>, or
classes such as <filename>lettre</filename>, will localize <literal>\today</literal>. For example,
the following will output &#8216;<literal>4 juillet 1976</literal>&#8217;:
</para>
<screen>\year=1976 \month=7 \day=4
\documentclass{minimal}
\usepackage[french]{babel}
\begin{document}
\today
\end{document}
</screen>
<para><literal>\today</literal> uses the counters <literal>\day</literal>, <literal>\month</literal>, and
<literal>\year</literal> (see <link linkend="_005cday-_0026-_005cmonth-_0026-_005cyear">\day &amp; \month &amp; \year</link>).
</para>
<indexterm role="cp"><primary>package, <literal>datetime</literal></primary></indexterm>
<indexterm role="cp"><primary><literal>datetime</literal> package</primary></indexterm>

<para>A number of package on CTAN work with dates.  One is <filename>datetime</filename> package
which can produce a wide variety of date formats, including ISO standards.
</para>
<para>The date is not updated as the &latex; process runs, so in principle the
date could be incorrect by the time the program finishes.
</para>

</sect1>
</chapter>
<chapter label="24" id="Splitting-the-input">
<title>Splitting the input</title>

<indexterm role="cp"><primary>splitting the input file</primary></indexterm>
<indexterm role="cp"><primary>input file</primary></indexterm>

<para>&latex; lets you split a large document into several smaller ones.
This can simplify editing or allow multiple authors to work on the
document.  It can also speed processing.
</para>
<para>Regardless of how many separate files you use, there is always one
<indexterm role="cp"><primary>root file</primary></indexterm>
<indexterm role="cp"><primary>file, root</primary></indexterm>
<firstterm>root file</firstterm>, on which &latex; compilation starts.  This shows such
a file with five included files.
</para>
<screen>\documentclass{book}
\includeonly{  % comment out lines below to omit compiling
  pref,
  chap1,
  chap2,
  append,
  bib
  }
\begin{document}
\frontmatter
\include{pref} 
\mainmatter
\include{chap1} 
\include{chap2}
\appendix
\include{append}
\backmatter
\include{bib} 
\end{document}
</screen>
<para>This will bring in material from <filename>pref.tex</filename>, <filename>chap1.tex</filename>,
<filename>chap2.tex</filename>, <filename>append.tex</filename>, and <filename>bib.tex</filename>.  If you compile
this file, and then comment out all of the lines inside
<literal>\includeonly{...}</literal> except for <literal>chap1,</literal> and compile again,
then &latex; will only process the material in the first chapter.
Thus, your output will appear more quickly and be shorter to print.
However, the advantage of the <literal>\includeonly</literal> command is that
&latex; will retain the page numbers and all of the cross reference
information from the other parts of the document so these will appear in
your output correctly.
</para>
<para>See <link linkend="Larger-book-template">Larger book template</link> for another example of <literal>\includeonly</literal>.
</para>



<sect1 label="24.1" id="_005cendinput">
<title><literal>\endinput</literal></title>

<indexterm role="fn"><primary>\endinput</primary></indexterm>

<para>Synopsis:
</para>
<screen>\endinput
</screen>
<para>When you <literal>\include{filename}</literal>, inside <filename>filename.tex</filename> the
material after <literal>\endinput</literal> will not be included.  This command is
optional; if <filename>filename.tex</filename> has no <literal>\endinput</literal> then &latex;
will read all of the file.
</para>
<para>For example, suppose that a document&#8217;s root file has
<literal>\input{chap1}</literal> and this is <filename>chap1.tex</filename>.
</para>
<screen>\chapter{One}
This material will appear in the document. 
\endinput
This will not appear.
</screen>
<para>This can be useful for putting documentation or comments at the end of a
file, or for avoiding junk characters that can be added during mailing.
It is also useful for debugging: one strategy to localize errors is to
put <literal>\endinput</literal> halfway through the included file and see if the
error disappears.  Now, knowing which half contains the error, moving
<literal>\endinput</literal> to halfway through that area further narrows down the
location. This process rapidly finds the offending line.
</para>
<para>After reading <literal>\endinput</literal>, &latex; continues to read to the end of
the line, so something can follow this command and be read nonetheless.
This allows you, for instance, to close an <literal>\if...</literal> with a
<literal>\fi</literal>.
</para>

</sect1>
<sect1 label="24.2" id="_005cinclude-_0026-_005cincludeonly">
<title><literal>\include</literal> &amp; <literal>\includeonly</literal></title>

<indexterm role="fn"><primary>\include</primary></indexterm>
<indexterm role="fn"><primary>\includeonly</primary></indexterm>

<para>Synopsis:
</para>
<screen>\includeonly{  % in document preamble
  ...
  <replaceable>filename</replaceable>,
  ...
  }
  ...
\include{<replaceable>filename</replaceable>}  % in document body 
</screen>
<para>Bring material from the external file <filename><replaceable>filename</replaceable>.tex</filename> into a
&latex; document.
</para>
<para>The <literal>\include</literal> command does three things: it executes
<literal>\clearpage</literal> (see <link linkend="_005cclearpage-_0026-_005ccleardoublepage">\clearpage &amp; \cleardoublepage</link>), then it
inputs the material from <filename><replaceable>filename</replaceable>.tex</filename> into the document,
and then it does another <literal>\clearpage</literal>.  This command can only
appear in the document body.  The <literal>\includeonly</literal> command controls
which files will be read by &latex; under subsequent <literal>\include</literal>
commands.  Its list of filenames is comma-separated, and it can only
appear in the preamble.
</para>
<para>This example root document, <filename>constitution.tex</filename>, brings in
three files, <filename>preamble.tex</filename>, <filename>articles.tex</filename>, and
<filename>amendments.tex</filename>.
</para>
<screen>\documentclass{book}
\includeonly{
  preamble,
  articles,
  amendments
  }
\begin{document}
\include{preamble}
\include{articles}
\include{amendments}
\end{document}
</screen>
<para>The file <filename>preamble.tex</filename> contains no special code; you have just
excerpted the chapter from <filename>consitution.tex</filename> and put it in a
separate file just for editing convenience.
</para>
<screen>\chapter{Preamble}
We the People of the United States,
in Order to form a more perfect Union, ...
</screen>
<para>Running &latex; on <filename>constitution.tex</filename> makes the material from the
three files appear in the document but also generates the auxiliary
files <filename>preamble.aux</filename>, <filename>articles.aux</filename>, and
<filename>amendments.tex</filename>. These contain information such as page numbers
and cross-references (see <link linkend="Cross-references">Cross references</link>).  If you now comment out
<literal>\includeonly</literal>&#8217;s lines with <literal>preamble</literal> and <literal>amendments</literal>
and run &latex; again then the resulting document shows only the
material from <filename>articles.tex</filename>, not the material from
<filename>preamble.tex</filename> or <filename>amendments.tex</filename>.  Nonetheless, all of the
auxiliary information from the omitted files is still there, including
the starting page number of the chapter.
</para>
<para>If the document preamble does not have <literal>\includeonly</literal> then
&latex; will include all the files you call for with <literal>\include</literal>
commands.
</para>
<para>The <literal>\include</literal> command makes a new page.  To avoid that, see
<link linkend="_005cinput">\input</link> (which, however, does not retain the auxiliary
information).
</para>
<para>See <link linkend="Larger-book-template">Larger book template</link> for another example using <literal>\include</literal>
and <literal>\includeonly</literal>.  That example also uses <literal>\input</literal> for some
material that will not necessarily start on a new page.
</para>
<para>File names can involve paths.
</para>
<screen>\documentclass{book}
\includeonly{
  chapters/chap1,
  }
\begin{document}
\include{chapters/chap1}
\end{document}
</screen>
<para>To make your document portable across distributions and platforms you
should avoid spaces in the file names. The tradition is to instead use
dashes or underscores.  Nevertheless, for the name &#8216;<literal>amo amas amat</literal>&#8217;,
this works under &tex; Live on GNU/Linux:
</para>
<screen>\documentclass{book}
\includeonly{
  &quot;amo\space amas\space amat&quot;
  }
\begin{document}
\include{&quot;amo\space amas\space amat&quot;}
\end{document}
</screen>
<para>and this works under MiK&tex; on Windows:
</para>
<screen>\documentclass{book}
\includeonly{
  {&quot;amo amas amat&quot;}
  }
\begin{document}
\include{{&quot;amo amas amat&quot;}}
\end{document}
</screen>
<indexterm role="cp"><primary>nested <literal>\include</literal>, not allowed</primary></indexterm>
<para>You cannot use <literal>\include</literal> inside a file that is being included or
you get &#8216;<literal>LaTeX Error: \include cannot be nested.</literal>&#8217;  The
<literal>\include</literal> command cannot appear in the document preamble; you will
get &#8216;<literal>LaTeX Error: Missing \begin{document}</literal>&#8217;.
</para>
<para>If a file that you <literal>\include</literal> does not exist, for instance if you
<literal>\include{athiesm}</literal> but you meant <literal>\include{atheism}</literal>,
then &latex; does not give you an error but will warn you &#8216;<literal>No file
athiesm.tex.</literal>&#8217;  (It will also create <filename>athiesm.aux</filename>.)
</para>
<para>If you <literal>\include</literal> the root file in itself then you first get
&#8216;<literal>LaTeX Error: Can be used only in preamble.</literal>&#8217;  Later runs get
&#8216;<literal>TeX capacity exceeded, sorry [text input levels=15]</literal>&#8217;.  To fix
this, you must remove the inclusion <literal>\include{root}</literal> but also
delete the file <filename><replaceable>root</replaceable>.aux</filename> and rerun &latex;.
</para>

</sect1>
<sect1 label="24.3" id="_005cinput">
<title><literal>\input</literal></title>

<indexterm role="fn"><primary>\input</primary></indexterm>

<para>Synopsis:
</para>
<screen>\input{<replaceable>filename</replaceable>}
</screen>
<para>&latex; processes the file as if its contents were inserted in the
current file.  For a more sophisticated inclusion mechanism see
<link linkend="_005cinclude-_0026-_005cincludeonly">\include &amp; \includeonly</link>.
</para>
<para>If <replaceable>filename</replaceable> does not end in &#8216;<literal>.tex</literal>&#8217; then &latex; first tries
the filename with that extension; this is the usual case.  If
<replaceable>filename</replaceable> ends with &#8216;<literal>.tex</literal>&#8217; then &latex; looks for the
filename as it is.
</para>
<para>For example, this
</para>
<screen>\input{macros}
</screen>
<para>will cause &latex; to first look for <filename>macros.tex</filename>. If it finds
that file then it processes its contents as thought they had been
copy-pasted in.  If there is no file of the name <filename>macros.tex</filename> then
&latex; tries the name <filename>macros</filename>, without an extension. (This may
vary by distribution.)
</para>
<para>To make your document portable across distributions and platforms you
should avoid spaces in the file names. The tradition is to instead use
dashes or underscores.  Nevertheless, for the name &#8216;<literal>amo amas amat</literal>&#8217;,
this works under &tex; Live on GNU/Linux:
</para>
<screen>\input{&quot;amo\space amas\space amat&quot;}
</screen>
<para>and this works under MiK&tex; on Windows:
</para>
<screen>\input{{&quot;amo amas amat&quot;}}
</screen>

</sect1>
</chapter>
<chapter label="25" id="Front_002fback-matter">
<title>Front/back matter</title>



<sect1 label="25.1" id="Table-of-contents-etc_002e">
<title>Table of contents etc.</title>

<indexterm role="cp"><primary>table of contents, creating</primary></indexterm>

<indexterm role="fn"><primary>\tableofcontents</primary></indexterm>
<indexterm role="fn"><primary>.toc file</primary></indexterm>
<indexterm role="fn"><primary>\listoffigures</primary></indexterm>
<indexterm role="fn"><primary>\listoftables</primary></indexterm>
<indexterm role="fn"><primary>.lof file</primary></indexterm>
<indexterm role="fn"><primary>.lot file</primary></indexterm>

<para>Synopsis, one of:
</para>
<screen>\tableofcontents
\listoffigures
\listoftables
</screen>
<para>Produce a table of contents, or list of figures, or list of tables.  Put
the command in the input file where you want the table or list to
go. You do not type the entries; for example, typically the table of
contents entries are automatically generated from the sectioning
commands <literal>\chapter</literal>, etc.
</para>
<para>This example illustrates the first command, <literal>\tableofcontents</literal>.
&latex; will produce a table of contents on the book&#8217;s first page.
</para>
<screen>\documentclass{book}
% \setcounter{tocdepth}{1}
\begin{document}
\tableofcontents\newpage
  ...
\chapter{...}
  ...
\section{...}
  ...
\subsection{...}
  ...
\end{document}
</screen>
<para>Uncommenting the second line would cause that table to contain chapter
and section listings but not subsection listings, because the
<literal>\section</literal> command has level&#160;1.  See <link linkend="Sectioning">Sectioning</link> for level
numbers of the sectioning units.  For more on the <literal>tocdepth</literal>
see <link linkend="Sectioning_002ftocdepth">Sectioning/tocdepth</link>.
</para>
<para>Another example of the use of <literal>\tableofcontents</literal> is in <link linkend="Larger-book-template">Larger
book template</link>.
</para>
<para>If you want a page break after the table of contents, write a
<literal>\newpage</literal> command after the <literal>\tableofcontents</literal> command, as
above.
</para>
<para>To make the table of contents &latex; stores the information in an
auxiliary file named <filename><replaceable>root-file</replaceable>.toc</filename> (see <link linkend="Splitting-the-input">Splitting the
input</link>).  For example, this &latex; file <filename>test.tex</filename>
</para>
<screen>\documentclass{article}
\begin{document}
\tableofcontents\newpage
\section{First section}
\subsection{First subsection}
  ...
</screen>
<para>writes the following line to <filename>test.toc</filename>.
</para>
<screen>\contentsline {section}{\numberline {1}First section}{2}
\contentsline {subsection}{\numberline {1.1}First subsection}{2}
</screen>
<para>The <literal>section</literal> or <literal>subsection</literal> is the sectioning unit.  The
hook <literal>\numberline</literal> lets you to change how the information appears
in the table of contents.  Of its two arguments, <literal>1</literal> or <literal>1.1</literal>
is the sectioning unit number and <literal>First section</literal> or <literal>First
subsection</literal> is the title.  Finally, <literal>2</literal> is the page number on which
the sectioning units start.
</para>
<para>One consequence of this auxiliary file storage strategy is that to get the
contents page correct you must run &latex; twice, once to store the
information and once to get it.  In particular, the first time that you
run &latex; on a new document, the table of contents page will be empty
except for its &#8216;<literal>Contents</literal>&#8217; header.  Just run it again.
</para>
<para>The commands <literal>\listoffigures</literal> and <literal>\listoftables</literal> produce a
list of figures and a list of tables.  They work the same way as the
contents commands; for instance, these work with information stored in
<filename>.lof</filename> and <filename>.lot</filename> files.
</para>
<indexterm role="cp"><primary>package, <literal>babel</literal></primary></indexterm>
<indexterm role="cp"><primary><literal>babel</literal> package</primary></indexterm>

<indexterm role="cp"><primary>package, <literal>polyglossia</literal></primary></indexterm>
<indexterm role="cp"><primary><literal>polyglossia</literal> package</primary></indexterm>

<para>To change the header for the table of contents page do something like
the first line here.
</para>
<screen>\renewcommand{\contentsname}{Table of contents}
\renewcommand{\listfigurename}{Plots}
\renewcommand{\listtablename}{Tables}
</screen>
<para>Similarly, the other two lines will do the other two.
Internationalization packages such as <filename>babel</filename> or <filename>polyglossia</filename>
will change the headers depending on the chosen base language.
</para>
<indexterm role="cp"><primary>package, <literal>tocloft</literal></primary></indexterm>
<indexterm role="cp"><primary><literal>tocloft</literal> package</primary></indexterm>

<indexterm role="cp"><primary>package, <literal>tocbibbind</literal></primary></indexterm>
<indexterm role="cp"><primary><literal>tocbibbind</literal> package</primary></indexterm>

<para>CTAN has many packages for the table of contents and lists of figures
and tables.  One convenient one for adjusting some aspects of the
default, such as spacing, is <filename>tocloft</filename>.  And, <filename>tocbibbind</filename>
will automatically add the bibliography, index, etc. to the table of
contents.
</para>


<sect2 label="25.1.1" id="_005caddcontentsline">
<title><literal>\addcontentsline</literal></title>

<indexterm role="fn"><primary>\addcontentsline</primary></indexterm>
<indexterm role="cp"><primary>table of contents entry, manually adding</primary></indexterm>

<para>Synopsis:
</para>
<screen>\addcontentsline{<replaceable>ext</replaceable>}{<replaceable>unit</replaceable>}{<replaceable>text</replaceable>}
</screen>
<indexterm role="fn"><primary>\contentsline</primary></indexterm>
<para>Add an entry to the file specified by <replaceable>ext</replaceable>.  Usually <replaceable>ext</replaceable> is
one of <literal>toc</literal> for the table of contents, <literal>lof</literal> for the list of
figures, or <literal>lot</literal> for the list of tables.
</para>
<para>The following will result in an &#8216;<literal>Appendices</literal>&#8217; line in the table of
contents.
</para>
<screen>\addcontentsline{toc}{section}{\protect\textbf{Appendices}}
</screen>
<para>It will appear at the same indentation level as the sections, will be in
boldface, and will be assigned the page number associated with the point
where it appears in the input file.
</para>
<para>The <literal>\addcontentsline</literal> command writes information to the file
<filename><replaceable>root-name</replaceable>.<replaceable>ext</replaceable></filename>.  It writes that information as the
text of the command
<literal>\contentsline{<replaceable>unit</replaceable>}{<replaceable>text</replaceable>}{<replaceable>num</replaceable>}</literal>, where
<literal><replaceable>num</replaceable></literal> is the current value of counter <literal><replaceable>unit</replaceable></literal>.  The
most common case is the table of contents and there <replaceable>num</replaceable> is the
page number of the first page of <replaceable>unit</replaceable>.
</para>
<para>This command is invoked by the sectioning commands <literal>\chapter</literal>,
etc., and also by <literal>\caption</literal> inside a float environment.  But it is
also used by authors.  For example, in a book to have the preface
unnumbered, you may use the starred <literal>\chapter*</literal>.  But that does not
put in table of contents information, so you can enter it manually, as
here.
</para>
<screen>\chapter*{Preface}
\addcontentsline{toc}{chapter}{\protect\numberline{}Preface}
</screen>
<para>In the <filename>.toc</filename> file &latex; will put the line <literal>\contentsline
{chapter}{\numberline {}Preface}{3}</literal>; note the page number
&#8216;<literal>3</literal>&#8217;.
</para>
<!-- xx how hardwired are these values?  other unit names? -->

<para>All of the arguments for <literal>\addcontentsline</literal> are required.
</para>
<variablelist><varlistentry><term><replaceable>ext</replaceable>
</term><listitem><para>Typically one of the strings <literal>toc</literal> for the table of contents,
<literal>lof</literal> for the list of figures, or <literal>lot</literal> for the list of
tables. The filename extension of the information file.
</para>
</listitem></varlistentry><varlistentry><term><replaceable>unit</replaceable>
</term><listitem><para>A string that depends on the value of the <replaceable>ext</replaceable> argument:
</para>
<variablelist><varlistentry><term><literal>toc</literal>
</term><listitem><para>For the table of contents, this is the name of a sectional unit:
<literal>part</literal>, <literal>chapter</literal>, <literal>section</literal>, <literal>subsection</literal>, etc.
</para>
</listitem></varlistentry><varlistentry><term><literal>lof</literal>
</term><listitem><para>For the list of figures: <literal>figure</literal>.
</para>
</listitem></varlistentry><varlistentry><term><literal>lot</literal>
</term><listitem><para>For the list of tables: <literal>table</literal>.
</para></listitem></varlistentry></variablelist>
</listitem></varlistentry><varlistentry><term><replaceable>text</replaceable>
</term><listitem><para>The text of the entry.  You must <literal>\protect</literal> any commands that are
fragile (see <link linkend="_005cprotect">\protect</link>).
</para></listitem></varlistentry></variablelist>
<para>The <literal>\addcontentsline</literal> command has an interaction with
<literal>\include</literal> (see <link linkend="_005cinclude-_0026-_005cincludeonly">\include &amp; \includeonly</link>). If you use them at
the same level, as with
<literal>\addcontentsline{...}{...}{...}\include{...}</literal> then lines
in the table of contents can come out in the wrong order.  The solution
is to move <literal>\addcontentsline</literal> into the file being included.
</para>
<para>If you use a <replaceable>unit</replaceable> that &latex; does not recognize, as here
</para>
<screen>\addcontentsline{toc}{setcion}{\protect\textbf{Appendices}}
</screen>
<para>then you don&#8217;t get an error but the formatting in the table of contents
will not make sense.
</para>

</sect2>
<sect2 label="25.1.2" id="_005caddtocontents">
<title><literal>\addtocontents</literal></title>

<indexterm role="fn"><primary>\addtocontents{<replaceable>ext</replaceable>}{<replaceable>text</replaceable>}</primary></indexterm>

<para>Synopsis:
</para>
<screen>\addtocontents{<replaceable>ext</replaceable>}{<replaceable>text</replaceable>}
</screen>
<para>Add <replaceable>text</replaceable>, which may be text or formatting commands, directly to
the auxiliary file with extension <replaceable>ext</replaceable>.  This is most commonly used
for the table of contents so that is the discussion here, but this also
applies to the list of figures and list of tables.
</para>
<para>This will put some vertical space in the table of contents after the
&#8216;<literal>Contents</literal>&#8217; header.
</para>
<screen>\tableofcontents\newpage
\addtocontents{toc}{\protect\vspace*{3ex}}
</screen>
<para>The <literal>\addtocontents</literal> command has two arguments.  Both are
required.
</para>
<variablelist><varlistentry><term><replaceable>ext</replaceable>
</term><listitem><para>Typically one of: <filename>toc</filename> for the table of contents, <filename>lof</filename> for
the list of figures, or <filename>lot</filename> for the list of tables.  The
extension of the file holding the information.
</para>
</listitem></varlistentry><varlistentry><term><replaceable>text</replaceable>
</term><listitem><para>The text, and possibly commands, to be written.
</para></listitem></varlistentry></variablelist>
<para>The sectioning commands such as <literal>\chapter</literal> use the
<literal>\addcontentsline</literal> command to store information.  This command
creates lines in the <filename>.toc</filename> auxiliary file containing the
<literal>\contentsline</literal> command (see <link linkend="_005caddcontentsline">\addcontentsline</link>).  In contrast,
the command <literal>\addtocontents</literal> puts material directly in that file.
</para>
<para>The <literal>\addtocontents</literal> command has an interaction with
<literal>\include</literal> (see <link linkend="_005cinclude-_0026-_005cincludeonly">\include &amp; \includeonly</link>). If you use them at
the same level, as with
<literal>\addtocontents{...}{...}\include{...}</literal> then lines in the
table of contents can come out in the wrong order.  The solution is to
move <literal>\addtocontents</literal> into the file being included.
</para>

</sect2>
<sect2 label="25.1.3" id="_005cnofiles">
<title><literal>\nofiles</literal></title>

<indexterm role="fn"><primary>\nofiles</primary></indexterm>

<para>Synopsis:
</para>
<screen>\nofiles
</screen>
<para>Prevent &latex; from writing any auxiliary files.  The only output will
be the <filename>.log</filename> and <filename>.pdf</filename> (or <filename>.dvi</filename>) files.  This command
must go in the preamble.
</para>
<para>Because of the <literal>\nofiles</literal> command this example will not produce a
<filename>.toc</filename> file.
</para>
<screen>\documentclass{book}
\nofiles
\begin{document}
\tableofcontents\newpage
\chapter{...}
  ...
</screen>
<para>&latex; will not erase any existing auxiliary files, so if you insert
the <literal>\nofiles</literal> command after you have run the file and gotten
a <filename>.toc</filename> then the table of contents page will continue to show
the old information.
</para>

</sect2>
</sect1>
<sect1 label="25.2" id="Indexes">
<title>Indexes</title>

<indexterm role="cp"><primary>indexes</primary></indexterm>

<indexterm role="fn"><primary>\makeindex</primary></indexterm>
<indexterm role="fn"><primary>\index</primary></indexterm>
<indexterm role="cp"><primary><filename>.idx</filename> file</primary></indexterm>

<para>This document has an index.
</para>
<screen>\documentclass{article}
\usepackage{makeidx} \makeindex
  ...
\begin{document}
  ...
Recall Wilson's Theorem: \index{Wilson's Theorem}
a number \( n&gt;1 \) is prime if and only if the factorial of \( n-1 \)
is congruent to \( -1 \) modulo~\( n \).
  ...
\printindex
  ...
</screen>
<para>The <literal>\usepackage{makeidx}</literal> and <literal>\makeindex</literal> in the preamble
bring in the relevant commands.
</para>
<para>Producing an index is a three stage process.  First, in the document
body you declare index entries with the <literal>\index</literal> command
(see <link linkend="_005cindex">\index</link>).  When you run &latex;, the <literal>\index</literal> writes its
information to an auxiliary file <filename><replaceable>root-name</replaceable>.idx</filename>.  Next, to
alphabetize and to do other manipulations you run an external command,
typically <command>makeindex</command> or <command>xindy</command> (see <link linkend="makeindex">makeindex</link>).
These output a file <filename><replaceable>root-name</replaceable>.ind</filename>.  Finally, you bring the
information back into your document and typeset it with the
<literal>\printindex</literal> command (see <link linkend="_005cprintindex">\printindex</link>).
</para>
<indexterm role="cp"><primary>package, <literal>showidx</literal></primary></indexterm>
<indexterm role="cp"><primary><literal>showidx</literal> package</primary></indexterm>

<indexterm role="cp"><primary>package, <literal>multind</literal></primary></indexterm>
<indexterm role="cp"><primary><literal>multind</literal> package</primary></indexterm>

<para>There are many packages that apply to indexing commands.  The
<literal>showidx</literal> package causes each index entries to be shown in the
margin on the page where the entry appears.  This can help in preparing
the index.  The <literal>multind</literal> package supports multiple indexes.  See
also the &tex; FAQ entry on this topic,
<ulink url="http://www.tex.ac.uk/cgi-bin/texfaq2html?label=multind">http://www.tex.ac.uk/cgi-bin/texfaq2html?label=multind</ulink>.
</para>


<sect2 label="25.2.1" id="_005cindex">
<title><literal>\index</literal></title>

<indexterm role="cp"><primary>index entry</primary></indexterm>
<indexterm role="fn"><primary>\index</primary></indexterm>

<para>Synopsis:
</para>
<screen>\index{<replaceable>index-entry-string</replaceable>}
</screen>
<para>Declare an entry in the index.  This command is fragile
(see <link linkend="_005cprotect">\protect</link>).
</para>
<para>For example, as described in <link linkend="Indexes">Indexes</link>, one way to get an index from
what&#8217;s below is to compile the document with <literal>pdflatex test</literal>, then
process the index entries with <literal>makeindex test</literal>, and then compile
again with <literal>pdflatex test</literal>.
</para>
<screen>W~Ackermann (1896--1962).\index{Ackermann}
  ...
Ackermann function\index{Ackermann!function}
  ...
rate of growth\index{Ackermann!function!growth rate}
</screen>
<para>All three index entries will get a page number, such as &#8216;<literal>Ackermann,
22</literal>&#8217;.  &latex; will format the second as a subitem of the first, on the
line below it and indented, and the third as a subitem of the second.
Three levels deep is as far as you can nest subentries.  (If you add
<literal>\index{Ackermann!function!growth rate!comparison}</literal> then
<command>makeindex</command> says &#8216;<literal>Scanning input file test.idx....done (4
entries accepted, 1 rejected)</literal>&#8217; and nothing appears in the index).
</para>
<para>If you enter a second <literal>\index</literal> with the same
<replaceable>index-entry-string</replaceable> then you will get a single index entry with two
page numbers (unless they happen to fall on the same page).  Thus,
adding <literal>as for Ackermann.\index{Ackermann}</literal> later in the same
document as above will give an index entry like &#8216;<literal>Ackermann, 22,
151</literal>&#8217;.  Also, you can enter the index entries in any order, so for
instance <literal>\index{Ackermann!function}</literal> could come before
<literal>\index{Ackermann}</literal>.
</para>
<indexterm role="cp"><primary>index, page range</primary></indexterm>
<para>Get a page range in the output, like &#8216;<literal>Hilbert, 23--27</literal>&#8217;, as here.
</para>
<screen>W~Ackermann (1896--1962).\index{Ackermann}
  ...
D~Hilbert (1862--1943)\index{Ackermann!Hilbert\(}
  ...
disapproved of his marriage.\index{Ackermann!Hilbert\)}
</screen>
<para>If the beginning and ending of the page range are equal then the system
just gives a single page entry, not a range.
</para>
<para>If you index subentries but not a main entry, as with
<literal>\index{Jones!program}</literal> and <literal>\index{Jones!results}</literal>, then
the output is the item &#8216;<literal>Jones</literal>&#8217; with no comma or page number,
followed by two subitems, like &#8216;<literal>program, 50</literal>&#8217; and &#8216;<literal>results,
51</literal>&#8217;.
</para>
<indexterm role="cp"><primary>&#8216;see&#8217; and &#8216;see also&#8217; index entries</primary></indexterm>
<indexterm role="cp"><primary>index entries, &#8216;see&#8217; and &#8216;see also&#8217;</primary></indexterm>
<indexterm role="fn"><primary>\seename</primary></indexterm>
<indexterm role="fn"><primary>\alsoname</primary></indexterm>
<indexterm role="cp"><primary>package, <literal>babel</literal></primary></indexterm>
<indexterm role="cp"><primary><literal>babel</literal> package</primary></indexterm>
<indexterm role="cp"><primary>package, <literal>polyglossia</literal></primary></indexterm>
<indexterm role="cp"><primary><literal>polyglossia</literal> package</primary></indexterm>


<para>Generate a index entry that says &#8216;<literal>See</literal>&#8217; by using a vertical bar
character: <literal>\index{Ackermann!function|see{P\'eter's
function}}</literal>.  You can instead get &#8216;<literal>See also</literal>&#8217; with <literal>seealso</literal>.
(The text &#8216;<literal>See</literal>&#8217; is defined by <literal>\seename</literal>, and &#8216;<literal>See also</literal>&#8217;
by <literal>\alsoname</literal>.  You can redefine these either by using an
internationalization package such as <filename>babel</filename> or <filename>polyglossia</filename>,
or directly as with <literal>\renewcommand{\alsoname}[1]{Also see
#1}</literal>.)
</para>
<para>The &#8216;<literal>See</literal>&#8217; feature is part of a more general functionality.  After
the vertical bar you can put the name of a one-input command, as in
<literal>\index{group|textit}</literal> (note the missing backslash on the
<literal>\textit</literal> command) and the system will apply that command to the
page number, here giving something like <literal>\textit{7}</literal>.  You can
define your own one-input commands, such as
<literal>\newcommand{\definedpage}[1]{{\color{blue}#1}}</literal> and then
<literal>\index{Ackermann!function|definedpage}</literal> will give a blue page
number (see <link linkend="Color">Color</link>).  Another, less practical, example is this,
</para>
<!-- credit Ian Thompson https://tex.stackexchange.com/a/272572/121234 -->
<screen>\newcommand\indexownpage[1]{#1, \thepage}
  ... Epimenides.\index{self-reference|indexownpage}
</screen>
<para>which creates an entry citing the page number of its own index listing.
</para>
<para>The two functions just described combine, as here
</para>
<screen>\index{Ackermann!function|(definedpage}
  ...
\index{Ackermann!function|)}
</screen>
<para>which outputs an index entry like &#8216;<literal>function, 23--27</literal>&#8217; where the page
number range is in blue.
</para>
<para>Consider an index entry such as &#8216;<literal>&#x03B1;-ring</literal>&#8217;.  Entering
it as <literal>$\alpha$-ring</literal> will cause it to be alphabetized according to
the dollar sign. You can instead enter it using an at-sign, as
<literal>\index{alpha-ring@$\alpha$-ring}</literal>.  If you specify an entry
with an at-sign separating two strings, <literal><replaceable>pos</replaceable>@<replaceable>text</replaceable></literal>,
then <replaceable>pos</replaceable> gives the alphabetical position of the entry while
<replaceable>text</replaceable> produces the text of the entry.  Another example is that
<literal>\index{Saint Michael's College@SMC}</literal> produces an index entry
&#8216;<literal>SMC</literal>&#8217; alphabetized into a different location than its spelling
would naturally give it.
</para>
<para>To put a <literal>!</literal>, or <literal>@</literal>, or <literal>|</literal> character in an index
entry, preceding it with a double quote, <literal>&quot;</literal>.  (The double quote
gets deleted before alphabetization.)
</para>
<indexterm role="cp"><primary>package, <literal>index</literal></primary></indexterm>
<indexterm role="cp"><primary><literal>index</literal> package</primary></indexterm>

<para>A number of packages on CTAN have additional functionality beyond that
provided by <filename>makeidx</filename>.  One is <filename>index</filename>, which allows for
multiple indices and contains a command
<literal>\index*{<replaceable>index-entry-string</replaceable>}</literal> that prints the
<replaceable>index-entry-string</replaceable> as well as indexing it.
</para>
<indexterm role="fn"><primary>\indexentry</primary></indexterm>
<indexterm role="cp"><primary>idx file</primary></indexterm>
<para>The <literal>\index</literal> command writes the indexing information to the file
<filename><replaceable>root-name</replaceable>.idx</filename> file.  Specifically, it writes text of the
command
<literal>\indexentry{<replaceable>index-entry-string</replaceable>}{<replaceable>page-num</replaceable>}</literal>, where
where <replaceable>page-num</replaceable> is the value of the <literal>\thepage</literal> counter.  On
occasion, when the <literal>\printindex</literal> command is confused, you have to
delete this file to start with a fresh slate.
</para>
<para>If you omit the closing brace of an <literal>\index</literal> command then you get a
message like this.
</para>
<screen>Runaway argument?  {Ackermann!function
!  Paragraph ended before \@wrindex was complete.
</screen>

</sect2>
<sect2 label="25.2.2" id="makeindex">
<title><command>makeindex</command></title>

<indexterm role="cp"><primary>index, processing</primary></indexterm>
<indexterm role="fn"><primary>makeindex</primary></indexterm>
<indexterm role="cp"><primary><command>makeindex</command> program</primary></indexterm>
<indexterm role="cp"><primary><filename>.ind</filename> file</primary></indexterm>
<indexterm role="cp"><primary><filename>.idx</filename> file</primary></indexterm>

<para>Synopsis, one of:
</para>
<screen>makeindex <replaceable>filename</replaceable>
makeindex -s <replaceable>style-file</replaceable> <replaceable>filename</replaceable>
makeindex <replaceable>options</replaceable> <replaceable>filename0</replaceable> ...
</screen>
<para>Sort, and otherwise process, the index information in the auxiliary file
<replaceable>filename</replaceable>.  This is a command line program.  It takes one or more
raw index files, <filename><replaceable>filename</replaceable>.idx</filename> files, and produces the
actual index file, the <filename><replaceable>filename</replaceable>.ind</filename> file that is input by
<literal>\printindex</literal> (see <link linkend="_005cprintindex">\printindex</link>).
</para>
<indexterm role="cp"><primary><filename>.isty</filename> file</primary></indexterm>
<indexterm role="fn"><primary>index, style file</primary></indexterm>
<indexterm role="fn"><primary>makeindex, style file</primary></indexterm>
<para>The first form of the command suffices for many uses.  The second allows
you to format the index by using an <firstterm>index style file</firstterm>, a
<filename>.isty</filename> file.  The third form is the most general; see the full
documentation on CTAN.
</para>
<para>This is a simple <filename>.isty</filename> file.
</para>
<screen>% book.isty 
%   $ makeindex -s book.isty -p odd book.idx
% creates the index as book.ind, starting on an odd page. 
preamble
&quot;\\pagestyle{empty}
\\small
\\begin{theindex}
\\thispagestyle{empty}&quot;

postamble
&quot;\n
\\end{theindex}&quot;
</screen>
<para>The description here covers only some of the index formatting
possibilities in <replaceable>style-file</replaceable>. For a full list see the documentation
on CTAN.
</para>
<para>A style file consists of a list of pairs: <replaceable>specifier</replaceable> and
<replaceable>attribute</replaceable>.  These can appear in the file in any order.  All of the
<replaceable>attributes</replaceable> are strings, except where noted.  Strings are
surrounded with double quotes, <literal>&quot;</literal>, and the maximum length of a
string is 144 characters.  The <literal>\n</literal> is for a newline and <literal>\t</literal>
is for a tab.  Backslashes are escaped with another backslash,
<literal>\\</literal>.  If a line begins with a percent sign, <literal>%</literal>, then it is a
comment.
</para>
<variablelist><anchor id="makeindex-preamble"/><varlistentry><term><indexterm role="fn"><primary>preamble</primary></indexterm><literal>preamble</literal>
</term><listitem><para>Preamble of the output file. Defines the context in which the index is
formatted.  Default: <literal>&quot;\\begin{theindex}\n&quot;</literal>.
</para>
<anchor id="makeindex-postamble"/></listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>postamble</primary></indexterm><literal>postamble</literal>
</term><listitem><para>Postamble of the output file.  Default: <literal>&quot;\n\n\\end{theindex}\n&quot;</literal>.
</para>
<anchor id="makeindex-group-skip"/></listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>group_skip</primary></indexterm><literal>group_skip</literal>
</term><listitem><indexterm role="fn"><primary>\indexspace</primary></indexterm>
<para>Traditionally index items are broken into groups, typically a group for
entries starting with &#8216;<literal>a</literal>&#8217;, etc.  This specifier gives what is
inserted when a new group begins.  Default: <literal>&quot;\n\n \\indexspace\n&quot;</literal>
(<literal>\indexspace</literal> is a rubber length with default value <literal>10pt
plus5pt minus3pt</literal>).
</para>
<anchor id="makeindex-letheadflag"/></listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>lethead_flag</primary></indexterm><literal>lethead_flag</literal>
</term><listitem><para>An integer.  It governs what is inserted for a new group or letter.  If
it is 0 (which is the default) then other than <literal>group_skip</literal> nothing
will be inserted before the group. If it is is positive then at a new
letter the <literal>lethead_prefix</literal> and <literal>lethead_suffix</literal> will be
inserted, with that letter in uppercase between them.  If it is negative
then what will be inserted is the letter in lowercase.  The default
is&#160;0.
</para>
<anchor id="makeindex-lethead-prefix"/></listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>lethead_prefix</primary></indexterm><literal>lethead_prefix</literal>
</term><listitem><para>If a new group begins with a different letter then this is the prefix
inserted before the new letter header.  Default: <literal>&quot;&quot;</literal>
</para>
<anchor id="makeindex-lethead-suffix"/></listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>lethead_suffix</primary></indexterm><literal>lethead_suffix</literal>
</term><listitem><para>If a group begins with a different letter then this is the suffix
inserted after the new letter header.  Default: <literal>&quot;&quot;</literal>.
</para>
<anchor id="makeindex-item-0"/></listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>item_0</primary></indexterm><literal>item_0</literal>
</term><listitem><para>What is put between two level&#160;0 items.  Default: <literal>&quot;\n \\item
&quot;</literal>.
</para>
<anchor id="makeindex-item-1"/></listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>item_1</primary></indexterm><literal>item_1</literal>
</term><listitem><para>Put between two level&#160;1 items.  Default: <literal>&quot;\n \\subitem &quot;</literal>.
</para>
<anchor id="makeindex-item-2"/></listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>item_2</primary></indexterm><literal>item_2</literal>
</term><listitem><para>put between two level&#160;2 items.  Default: <literal>&quot;\n \\subsubitem &quot;</literal>.
</para>
<anchor id="makeindex-item-01"/></listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>item_01</primary></indexterm><literal>item_01</literal>
</term><listitem><para>What is put between a level&#160;0 item and a level&#160;1 item.
Default: <literal>&quot;\n \\subitem &quot;</literal>.
</para>
<anchor id="makeindex-item-x1"/></listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>item_x1</primary></indexterm><literal>item_x1</literal>
</term><listitem><para>What is put between a level&#160;0 item and a level&#160;1 item in the
case that the level&#160;0 item doesn&#8217;t have any page numbers (as in
<literal>\index{aaa|see{bbb}}</literal>).  Default: <literal>&quot;\n \\subitem &quot;</literal>.
</para>
<anchor id="makeindex-item-12"/></listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>item_12</primary></indexterm><literal>item_12</literal>
</term><listitem><para>What is put between a level&#160;1 item and a level&#160;2 item.
Default: <literal>&quot;\n \\subsubitem &quot;</literal>.
</para>
<anchor id="makeindex-item-x2"/></listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>item_x2</primary></indexterm><literal>item_x2</literal>
</term><listitem><para>What is put between a level&#160;1 item and a level&#160;2 item, if the
level&#160;1 item doesn&#8217;t have page numbers.  Default: <literal>&quot;\n
\\subsubitem &quot;</literal>.
</para>
<anchor id="makeindex-delim-0"/></listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>delim_0</primary></indexterm><literal>delim_0</literal>
</term><listitem><para>Delimiter put between a level&#160;0 key and its first page
number. Default: a comma followed by a blank, <literal>&quot;, &quot;</literal>.
</para>
<anchor id="makeindex-delim-1"/></listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>delim_1</primary></indexterm><literal>delim_1</literal>
</term><listitem><para>Delimiter put between a level&#160;1 key and its first page
number. Default: a comma followed by a blank, <literal>&quot;, &quot;</literal>.
</para>
<anchor id="makeindex-delim-2"/></listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>delim_2</primary></indexterm><literal>delim_2</literal>
</term><listitem><para>Delimiter between a level&#160;2 key and its first page number. Default:
a comma followed by a blank, <literal>&quot;, &quot;</literal>.
</para>
<anchor id="makeindex-delim-n"/></listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>delim_n</primary></indexterm><literal>delim_n</literal>
</term><listitem><para>Delimiter between two page numbers for the same key (at any
level). Default: a comma followed by a blank, <literal>&quot;, &quot;</literal>.
</para>
<anchor id="makeindex-delim-r"/></listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>delim_r</primary></indexterm><literal>delim_r</literal>
</term><listitem><para>What is put between the starting and ending page numbers of a range.
Default: <literal>&quot;--&quot;</literal>.
</para>
<anchor id="makeindex-line-max"/></listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>line_max</primary></indexterm><literal>line_max</literal>
</term><listitem><para>An integer.  Maximum length of an index entry&#8217;s line in the output,
beyond which the line wraps.  Default: <literal>72</literal>.
</para>
<anchor id="makeindex-indent-space"/></listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>indent_space</primary></indexterm><literal>indent_space</literal>
</term><listitem><para>What is inserted at the start of a wrapped line.  Default:
<literal>&quot;\t\t&quot;</literal>.
</para>
<anchor id="makeindex-indent-length"/></listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>indent_length</primary></indexterm><literal>indent_length</literal>
</term><listitem><para>A number.  The length of the wrapped line indentation.  The default
<literal>indent_space</literal> is two tabs and each tab is eight spaces so the
default here is <literal>16</literal>.
</para>
<anchor id="makeindex-page-precedence"/></listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>page_precedence</primary></indexterm><literal>page_precedence</literal>
</term><listitem><para>A document may have pages numbered in different ways.  For example, a
book may have front matter pages numbered in lowercase roman while main
matter pages are in arabic.  This string specifies the order in which
they will appear in the index.  The <command>makeindex</command> command supports
five different types of numerals: lowercase roman <literal>r</literal>, and numeric
or arabic <literal>n</literal>, and lowercase alphabetic <literal>a</literal>, and uppercase
roman <literal>R</literal>, and uppercase alphabetic <literal>A</literal>.  Default:
<literal>&quot;rnaRA&quot;</literal>.
</para>
</listitem></varlistentry></variablelist>
<indexterm role="fn"><primary>xindy</primary></indexterm>
<indexterm role="cp"><primary><command>xindy</command> program</primary></indexterm>
<para>There are a number of other programs that do the job <command>makeindex</command>
does.  One is <command>xindy</command>, which does internationalization and can
process indexes for documents marked up using &latex; and a number of
other languages.  It is is highly configurable, both in markup terms and
in terms of the collating order of the text.  See the documentation on
CTAN.
</para>

</sect2>
<sect2 label="25.2.3" id="_005cprintindex">
<title><command>\printindex</command></title>

<indexterm role="cp"><primary>index, printing</primary></indexterm>
<indexterm role="fn"><primary>\printindex</primary></indexterm>

<para>Synopsis:
</para>
<screen>\printindex
</screen>
<indexterm role="fn"><primary>\printindex</primary></indexterm>
<para>Place the index into the output.
</para>
<para>To get an index you must first include
<literal>\usepackage{makeidx}\makeindex</literal> in the document preamble and
compile the document, then run the system command <command>makeindex</command>,
and then compile the document again.  See <link linkend="Indexes">Indexes</link> for further
discussion and an example of the use of <literal>\printindex</literal>.
</para>

</sect2>
</sect1>
<sect1 label="25.3" id="Glossaries">
<title>Glossaries</title>

<indexterm role="cp"><primary>glossary</primary></indexterm>
<indexterm role="cp"><primary>glossaries</primary></indexterm>
<indexterm role="cp"><primary>acronyms, list of</primary></indexterm>
<indexterm role="fn"><primary>\makeglossary</primary></indexterm>
<indexterm role="fn"><primary>\printglossaries</primary></indexterm>

<para>Synopsis:
</para>
<screen>\usepackage{glossaries} \makeglossaries
  ...
\newglossaryentry{<replaceable>label</replaceable>}{<replaceable>settings</replaceable>}  
  ...
\gls{<replaceable>label</replaceable>}.  
  ...
\printglossaries
</screen>
<para>The <filename>glossaries</filename> package allows you to make glossaries, including
multiple glossaries, as well as lists of acronyms.
</para>
<para>To get the output from this example, compile the document (for instance
with <literal>pdflatex filename</literal>), then run the command line command
<literal>makeglossaries filename</literal>, and then compile the document again.
</para>
<screen>\documentclass{...}  
\usepackage{glossaries} \makeglossaries
\newglossaryentry{tm}{%
  name={Turing machine},
  description={A model of a machine that computes.  The model is simple
               but can compute anything any existing device can compute.
               It is the standard model used in Computer Science.},
  }  
\begin{document}  
Everything begins with the definition of a \gls{tm}.  
  ...
\printglossaries
\end{document}
</screen>
<para>That gives two things.  In the main text it outputs &#8216;<literal>... definition
of a Turing machine</literal>&#8217;.  In addition, in a separate sectional unit headed
&#8216;<literal>Glossary</literal>&#8217; there appears a description list.  In boldface it says
&#8216;<literal>Turing machine</literal>&#8217; and the rest of the item says in normal type
&#8216;<literal>A model of a machine &#8230; Computer Science</literal>&#8217;.
</para>
<indexterm role="fn"><primary>\makeglossary</primary></indexterm>
<indexterm role="fn"><primary>\printglossaries</primary></indexterm>
<indexterm role="cp"><primary><filename>.glo</filename> file</primary></indexterm>
<para>The command <literal>\makeglossary</literal> opens the file that will contain the
entry information, <filename><replaceable>root-file</replaceable>.glo</filename>.  Put the
<literal>\printglossaries</literal> command where you want the glossaries to appear
in your document.
</para>
<para>The <filename>glossaries</filename> package is very powerful.  For instance, besides
the commands <literal>\newglossaryentry</literal> and <literal>\gls</literal>, there are similar
commands for a list of acronyms.  See the package documentations on
CTAN.
</para>


<sect2 label="25.3.1" id="_005cnewglossaryentry">
<title><literal>\newglossaryentry</literal></title>

<indexterm role="cp"><primary>glossary, entries</primary></indexterm>
<indexterm role="fn"><primary>\newglossaryentry</primary></indexterm>

<para>Synopsis, one of: 
</para>
<screen>\newglossaryentry{<replaceable>label</replaceable>} 
{ 
  name={<replaceable>name</replaceable>}, 
  description={<replaceable>description</replaceable>}, 
  <replaceable>other options</replaceable>, ... 
}
</screen>
<para>or
</para>
<screen>\longnewglossaryentry{<replaceable>label</replaceable>} 
{ 
  name={<replaceable>name</replaceable>}, 
  <replaceable>other options</replaceable> ..., 
} 
{<replaceable>description</replaceable>}
</screen>
<para>Declare a new entry for a glossary.  The <replaceable>label</replaceable> must be unique for
the document.  The settings associated with the label are pairs:
<literal><replaceable>key</replaceable>=<replaceable>value</replaceable></literal>.
</para>
<para>This puts the blackboard bold symbol for the real numbers &#x211D; in the
glossary.
</para>
<screen>\newglossaryentry{R}  
{  
  name={\ensuremath{\mathbb{R}}},  
  description={the real numbers},  
}
</screen>
<para>Use the second command form if the <replaceable>description</replaceable> spans more than one
paragraph.
</para>
<para>For a full list of <replaceable>key</replaceable>s see the package documentation on CTAN but
here are a few.
</para>
<variablelist><varlistentry><term><indexterm role="fn"><primary>name</primary></indexterm><literal>name</literal>
</term><listitem><para>(Required.) The word, phrase, or symbol that you are defining.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>description</primary></indexterm><literal>description</literal>
</term><listitem><para>(Required.) The description that will appear in the glossary.
If this has more than one paragraph then you must use the second command
form given in the synopsis.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>plural</primary></indexterm><literal>plural</literal>
</term><listitem><para>The plural form of <replaceable>name</replaceable>.  Refer to the plural form using
<literal>\glspl</literal> or <literal>\Glspl</literal> (see <link linkend="_005cgls">\gls</link>).
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>sort</primary></indexterm><literal>sort</literal>
</term><listitem><para>How to place this entry in the list of entries that the glossary holds.
</para>
</listitem></varlistentry><varlistentry><term><indexterm role="fn"><primary>symbol</primary></indexterm><literal>symbol</literal>
</term><listitem><para>A symbol, such as a mathematical symbol, besides the name.
</para>
</listitem></varlistentry></variablelist>

</sect2>
<sect2 label="25.3.2" id="_005cgls">
<title><literal>\gls</literal></title>

<indexterm role="cp"><primary>glossary, entry reference</primary></indexterm>
<indexterm role="fn"><primary>\gls</primary></indexterm>

<para>Synopsis, one of: 
</para>
<screen>\gls{<replaceable>label</replaceable>}
\glspl{<replaceable>label</replaceable>}
\Gls{<replaceable>label</replaceable>}
\Glspl{<replaceable>label</replaceable>}
</screen>
<para>Refer to a glossary entry.  The entries are declared with
<literal>\newglossaryentry</literal> (see <link linkend="_005cnewglossaryentry">\newglossaryentry</link>).
</para>
<para>This
</para>
<screen>\newglossaryentry{N}{%
  name={the natural numbers},
  description={The numbers $0$, $1$, $2$, $\ldots$\@},
  symbol={\ensuremath{\mathbb{N}}},
  }  
  ...
Consider \gls{N}.
</screen>
<para>gives the output &#8216;<literal>Consider the natural numbers</literal>&#8217;.
</para>
<para>The second command form <literal>\glspl{<replaceable>label</replaceable>}</literal> produces the plural
of <replaceable>name</replaceable> (by default it tries adding an &#8216;<literal>s</literal>&#8217;).  The third form
capitalizes the first letter of <replaceable>name</replaceable>, as does the fourth form,
which also takes the plural.
</para>

</sect2>
</sect1>
</chapter>
<chapter label="26" id="Letters">
<title>Letters</title>

<indexterm role="cp"><primary>letters, writing</primary></indexterm>
<indexterm role="cp"><primary>writing letters</primary></indexterm>

<para>Synopsis: 
</para>
<screen>\documentclass{letter}
\address{<replaceable>senders address</replaceable>}   % return address
\signature{<replaceable>sender name</replaceable>}
\begin{document}
\begin{letter}{<replaceable>recipient address</replaceable>}
\opening{<replaceable>salutation</replaceable>}
  <replaceable>letter body</replaceable>
\closing{<replaceable>closing text</replaceable>}
\end{letter}
   ...
\end{document}
</screen>
<para>Produce one or more letters.  
</para>
<para>Each letter is in a separate <literal>letter</literal> environment, whose argument
<replaceable>recipient address</replaceable> often contains multiple lines separated with a
double backslash,&#160;(<literal>\\</literal>).  For example, you might have:
</para>
<screen> \begin{letter}{Ninon de l'Enclos \\
                l'h\^otel Sagonne}
   ...
 \end{letter}
</screen>
<para>The start of the <literal>letter</literal> environment resets the page number to 1,
and the footnote number to 1 also.
</para>
<para>The <replaceable>sender address</replaceable> and <replaceable>sender name</replaceable> are common to all of the
letters, whether there is one or more, so these are best put in the
preamble.  As with the recipient address, often <replaceable>sender address</replaceable>
contains multiple lines separated by a double
backslash&#160;(<literal>\\</literal>).  &latex; will put the <replaceable>sender name</replaceable>
under the closing, after a vertical space for the traditional
hand-written signature.
</para>
<para>Each <literal>letter</literal> environment body begins with a required
<literal>\opening</literal> command such as <literal>\opening{Dear Madam or Sir:}</literal>.
The <replaceable>letter body</replaceable> text is ordinary &latex; so it can contain
everything from enumerated lists to displayed math, except that commands
such as <literal>\chapter</literal> that make no sense in a letter are turned off.
Each <literal>letter</literal> environment body typically ends with a
<literal>\closing</literal> command such as <literal>\closing{Yours,}</literal>.
</para>
<indexterm role="fn"><primary>\\ for letters</primary></indexterm>
<para>Additional material may come after the <literal>\closing</literal>.  You can say who
is receiving a copy of the letter with a command like <literal>\cc{the
Boss \\ the Boss's Boss}</literal>.  There&#8217;s a similar <literal>\encl</literal> command for
a list of enclosures.  And, you can add a postscript with <literal>\ps</literal>.
</para>
<para>&latex;&#8217;s default is to indent the sender name and the closing above it
by a length of <literal>\longindentation</literal>.  By default this is
<literal>0.5\textwidth</literal>. To make them flush left, put
<literal>\setlength{\longindentation}{0em}</literal> in your preamble.
</para>
<para>To set a fixed date use something like
<literal>\renewcommand{\today}{1958-Oct-12}</literal>.  If put in your preamble
then it will apply to all the letters.
</para>
<para>This example shows only one <literal>letter</literal> environment.  The three lines
marked as optional are typically omitted.
</para>
<screen>\documentclass{letter}
\address{Sender's street \\ Sender's town}
\signature{Sender's name \\ Sender's title}
% optional: \location{Mailbox 13}
% optional: \telephone{(102) 555-0101}
\begin{document}
\begin{letter}{Recipient's name \\ Recipient's address}
\opening{Sir:}
% optional: \thispagestyle{firstpage}
I am not interested in entering a business arrangement with you.
\closing{Your most humble, etc.,}
\end{letter}
\end{document}
</screen>
<para>These commands are used with the <literal>letter</literal> class.
</para>


<sect1 label="26.1" id="_005caddress">
<title><literal>\address</literal></title>

<indexterm role="fn"><primary>\address</primary></indexterm>

<para>Synopsis: 
</para>
<screen>\address{<replaceable>senders address</replaceable>}
</screen>
<para>Specify the return address, as it appears on the letter and on the
envelope.  Separate multiple lines in <replaceable>senders address</replaceable> with a
double backslash,&#160;<literal>\\</literal>.
</para>
<para>Because it can apply to multiple letters this declaration is often put
in the preamble.  However, it can go anywhere, including inside an
individual <literal>letter</literal> environment.
</para>
<para>This command is optional: if you do not use it then the letter is
formatted with some blank space on top, for copying onto pre-printed
letterhead paper.  If you do use the <literal>\address</literal> declaration then it
is formatted as a personal letter.
</para>
<para>Here is an example.
</para>
<screen>\address{Stephen Maturin \\
         The Grapes of the Savoy}
</screen>

</sect1>
<sect1 label="26.2" id="_005ccc">
<title><literal>\cc</literal></title>

<indexterm role="fn"><primary>\cc</primary></indexterm>
<indexterm role="cp"><primary>cc list, in letters</primary></indexterm>

<para>Synopsis:
</para>
<screen>\cc{<replaceable>name0</replaceable> \\ 
     ... }
</screen>
<para>Produce a list of names to which copies of the letter were sent.  This
command is optional.  If it appears then typically it comes after
<literal>\closing</literal>.  Put the names on different lines by separating them
with a double backslash, <literal>\\</literal>, as in:
</para>
<screen>\cc{President \\
    Vice President}
</screen>

</sect1>
<sect1 label="26.3" id="_005cclosing">
<title><literal>\closing</literal></title>

<indexterm role="fn"><primary>\closing</primary></indexterm>
<indexterm role="cp"><primary>letters, ending</primary></indexterm>
<indexterm role="cp"><primary>closing letters</primary></indexterm>

<para>Synopsis:
</para>
<screen>\closing{<replaceable>text</replaceable>}
</screen>
<para>Produce the letter&#8217;s closing.  This is optional, but usual.  It appears
at the end of a letter, above a handwritten signature.  For example:
</para>
<screen>\closing{Regards,}
</screen>

</sect1>
<sect1 label="26.4" id="_005cencl">
<title><literal>\encl</literal></title>

<indexterm role="fn"><primary>\encl</primary></indexterm>
<indexterm role="cp"><primary>enclosure list</primary></indexterm>

<para>Synopsis:
</para>
<screen>\encl{<replaceable>first enclosed object</replaceable> \\ 
       ... }
</screen>
<para>Produce a list of things included with the letter. This command is
optional; when it is used, it typically is put after <literal>\closing</literal>.
Separate multiple lines with a double backslash, <literal>\\</literal>.
</para>
<screen>\encl{License \\
      Passport}
</screen>

</sect1>
<sect1 label="26.5" id="_005clocation">
<title><literal>\location</literal></title>

<indexterm role="fn"><primary>\location</primary></indexterm>

<para>Synopsis:
</para>
<screen>\location{<replaceable>text</replaceable>}
</screen>
<para>The <replaceable>text</replaceable> appears centered at the bottom of the page.  It only
appears if the page style is <literal>firstpage</literal>.
</para>

</sect1>
<sect1 label="26.6" id="_005cmakelabels">
<title><literal>\makelabels</literal></title>

<indexterm role="fn"><primary>\makelabels</primary></indexterm>

<para>Synopsis:
</para>
<screen>\makelabels   % in preamble
</screen>
<para>Optional, for a document that contains <literal>letter</literal> environments.  If
you just put <literal>\makelabels</literal> in the preamble then at the end of the
document you will get a sheet with labels for all the recipients, one
for each letter environment, that you can copy to a sheet of peel-off
address labels.
</para>
<para>Customize the labels by redefining the commands <literal>\startlabels</literal>,
<literal>\mlabel</literal>, and <literal>\returnaddress</literal> (and perhaps <literal>\name</literal>) in
the preamble.  The command <literal>\startlabels</literal> sets the width, height,
number of columns, etc., of the page onto which the labels are printed.
The command <literal>\mlabel{<replaceable>return address</replaceable>}{<replaceable>recipient
address</replaceable>}</literal> produces the two labels (or one, if you choose to ignore the
<replaceable>return address</replaceable>) for each letter environment. The first argument,
<replaceable>return address</replaceable>, is the value returned by the macro
<literal>\returnaddress</literal>.  The second argument, <replaceable>recipient address</replaceable>, is
the value passed in the argument to the <literal>letter</literal> environment.  By
default <literal>\mlabel</literal> ignores the first argument, the <replaceable>return
address</replaceable>, causing the default behavior described in the prior paragraph.
</para>
<para>This illustrates customization.  Its output includes a page with two
columns having two labels each.
</para>
<screen>\documentclass{letter}
\renewcommand*{\returnaddress}{Fred McGuilicuddy \\
                               Oshkosh, Mineola 12305}
\newcommand*\originalMlabel{}
\let\originalMlabel\mlabel
\def\mlabel#1#2{\originalMlabel{}{#1}\originalMlabel{}{#2}}
\makelabels
  ...
\begin{document}
\begin{letter}{A Einstein \\
               112 Mercer Street \\
               Princeton, New Jersey, USA 08540}
  ...
\end{letter}
\begin{letter}{K G\&quot;odel \\
               145 Linden Lane \\
               Princeton, New Jersey, USA 08540}
  ...
\end{letter}
\end{document}
</screen>
<para>The first column contains the return address twice.  The second column
contains the address for each recipient.
</para>
<indexterm role="cp"><primary>package, <literal>envlab</literal></primary></indexterm>
<indexterm role="cp"><primary><literal>envlab</literal> package</primary></indexterm>

<para>The package <literal>envlab</literal> makes formatting the labels easier, with
standard sizes already provided.  The preamble lines
<literal>\usepackage[personalenvelope]{envlab}</literal> and <literal>\makelabels</literal>
are all that you need to print envelopes.
</para>

</sect1>
<sect1 label="26.7" id="_005cname">
<title><literal>\name</literal></title>

<indexterm role="fn"><primary>\name</primary></indexterm>

<para>Synopsis:
</para>
<screen>\name{<replaceable>name</replaceable>}
</screen>
<para>Optional.  Sender&#8217;s name, used for printing on the envelope together
with the return address.
</para>

</sect1>
<sect1 label="26.8" id="_005copening">
<title><literal>\opening</literal></title>

<indexterm role="fn"><primary>\opening</primary></indexterm>
<indexterm role="cp"><primary>letters, starting</primary></indexterm>

<para>Synopsis:
</para>
<screen>\opening{<replaceable>salutation</replaceable>}
</screen>
<para>Required.  Follows the <literal>\begin{letter}{...}</literal>. The argument
<replaceable>salutation</replaceable> is mandatory.  For instance:
</para>
<screen>\opening{Dear John:}
</screen>

</sect1>
<sect1 label="26.9" id="_005cps">
<title><literal>\ps</literal></title>
<indexterm role="fn"><primary>\ps</primary></indexterm>
<indexterm role="cp"><primary>postscript, in letters</primary></indexterm>

<para>Synopsis:
</para>
<screen>\ps{<replaceable>text</replaceable>}
</screen>
<para>Add a postscript.  This command is optional and usually is used after
<literal>\closing</literal>.
</para>
<screen>\ps{P.S. After you have read this letter, burn it. Or eat it.}
</screen>

</sect1>
<sect1 label="26.10" id="_005csignature">
<title><literal>\signature</literal></title>

<para>Synopsis:
</para>
<screen>\signature{<replaceable>first line</replaceable> \\
            ... }
</screen>
<indexterm role="fn"><primary>\signature</primary></indexterm>

<para>The sender&#8217;s name.  This command is optional, although its inclusion is
usual.
</para>
<para>The argument text appears at the end of the letter, after the closing.
&latex; leaves some vertical space for a handwritten
signature. Separate multiple lines with a double
backslash,&#160;<literal>\\</literal>.  For example:
</para>
<screen>\signature{J Fred Muggs \\
           White House}
</screen>
<para>&latex;&#8217;s default for the vertical space from the <literal>\closing</literal> text
down to the <literal>\signature</literal> text is <literal>6\medskipamount</literal>, which is
six times <literal>\medskipamount</literal> (where <literal>\medskipamount</literal> is equal to
a <literal>\parskip</literal>, which in turn is defined by default here to
0.7em).
</para>
<para>This command is usually in the preamble, to apply to all the letters in
the document.  To have it apply to one letter only, put it inside a
<literal>letter</literal> environment and before the <literal>\closing</literal>.
</para>
<para>You can include a graphic in the signature as here.
</para>
<screen>\signature{\vspace{-6\medskipamount}\includegraphics{sig.png}\\
             My name}
</screen>
<para>For this you must put <literal>\usepackage{graphicx}</literal> in the preamble
(see <link linkend="Graphics">Graphics</link>).
</para>

<!-- I think this is not a user-level command; it is used to keep from breaking -->
<!-- the page between the closing and the signature -->
<!-- @node \stopbreaks and \startbreaks -->
<!-- @section @code{\stopbreaks} and @code{\startbreaks} -->

<!-- @findex \startbreak -->
<!-- @findex \stopbreaks -->

<!-- @example  -->
<!-- @code{\stopbreaks} -->
<!-- text -->
<!-- @code{\startbreaks} -->
<!-- @end example -->

<!-- The @code{\stopbreaks} inhibits page breaking.  The @code{\startbreaks} resumes -->
<!-- normal page breaking. -->
<!-- -->
<!-- -->
</sect1>
<sect1 label="26.11" id="_005ctelephone">
<title><literal>\telephone</literal></title>

<indexterm role="fn"><primary>\telephone</primary></indexterm>

<para>Synopsis:
</para>
<screen>\telephone{<replaceable>number</replaceable>}
</screen>
<para>The sender&#8217;s telephone number.  This is typically in the preamble, where
it applies to all letters.  This only appears if the <literal>firstpage</literal>
pagestyle is selected.  If so, it appears on the lower right of the
page.
</para>

</sect1>
</chapter>
<chapter label="27" id="Terminal-input_002foutput">
<title>Terminal input/output</title>

<indexterm role="cp"><primary>input/output, to terminal</primary></indexterm>
<indexterm role="cp"><primary>terminal input/output</primary></indexterm>



<sect1 label="27.1" id="_005ctypein">
<title><literal>\typein</literal></title>

<indexterm role="fn"><primary>\typein</primary></indexterm>

<para>Synopsis, one of:
</para>
<screen>\typein{<replaceable>prompt-msg</replaceable>}
\typein[<replaceable>cmd</replaceable>]{<replaceable>prompt-msg</replaceable>}
</screen>
<para>Print <replaceable>prompt-msg</replaceable> on the terminal and cause &latex; to stop and
wait for you to type a line of input.  This line of input ends when you
hit the return key.
</para>
<para>For example, this
</para>
<screen>As long as I live I shall never forget \typein{Enter student name:}
</screen>
<para>coupled with this command line interaction
</para>
<screen>Enter student name:

\@typein=Aphra Behn
</screen>
<para>gives the output &#8216;<literal>... never forget Aphra Behn</literal>&#8217;.
</para>
<para>The first command version, <literal>\typein{<replaceable>prompt-msg</replaceable>}</literal>, causes
the input you typed to be processed as if it had been included in the
input file in place of the <literal>\typein</literal> command.
</para>
<para>In the second command version the optional argument <literal><replaceable>cmd</replaceable></literal>
argument must be a command name &#8212; it must begin with a backslash, \.
This command name is then defined or redefined to be the input that you
typed.  For example, this
</para>
<screen>\typein[\student]{Enter student name:}
\typeout{Recommendation for \student .}
</screen>
<para>gives this output on the command line,
</para>
<screen>Enter student name:

\student=John Dee
Recommendation for John Dee.
</screen>
<para>where the user has entered &#8216;<literal>John Dee.</literal>&#8217;
</para>

</sect1>
<sect1 label="27.2" id="_005ctypeout">
<title><literal>\typeout</literal></title>

<indexterm role="fn"><primary>\typeout</primary></indexterm>

<para>Synopsis:
</para>
<screen>\typeout{<replaceable>msg</replaceable>}
</screen>
<para>Print <literal>msg</literal> on the terminal and in the <literal>log</literal> file.
</para>
<para>This
</para>
<screen>\newcommand{\student}{John Dee}
\typeout{Recommendation for \student .}
</screen>
<para>outputs &#8216;<literal>Recommendation for John Dee</literal>&#8217;.  Like what happens here with
<literal>\student</literal>, commands that are defined with <literal>\newcommand</literal> or
<literal>\renewcommand</literal> (among others) are replaced by their definitions
before being printed.
</para>
<para>&latex;&#8217;s usual rules for treating multiple spaces as a single space
and ignoring spaces after a command name apply to <literal>msg</literal>.  As above,
use the command <literal>\space</literal> to get a single space, independent of
surrounding spaces.  Use <literal>^^J</literal> to get a newline.  Get a percent
character with <literal>\csname @percentchar\endcsname</literal>.
</para>
<para>This command can be useful for simple debugging, as here:
</para>
<screen>\newlength{\jhlength}
\setlength{\jhlength}{5pt}
\typeout{The length is \the\jhlength.}
</screen>
<para>produces on the command line &#8216;<literal>The length is 5.0pt</literal>&#8217;.
</para>

</sect1>
</chapter>
<chapter label="28" id="Command-line">
<title>Command line</title>

<indexterm role="cp"><primary>command line</primary></indexterm>

<para>Synopsis (from a terminal command line):
</para>
<screen>pdflatex <replaceable>options</replaceable> <replaceable>argument</replaceable>
</screen>
<para>Run &latex; on <replaceable>argument</replaceable>.  In place of <command>pdflatex</command> you can
also use <command>xelatex</command>, or <literal>lualatex</literal>, or <literal>dviluatex</literal>, or
<literal>latex</literal>.
</para>
<para>For example, this will run &latex; on the file <filename>thesis.tex</filename>,
creating the output <filename>thesis.pdf</filename>.
</para>
<screen>pdflatex thesis
</screen>
<para><indexterm role="fn"><primary>.tex, default extension</primary></indexterm>
Note that <filename>.tex</filename> is the default file extension.
</para>
<para>pdf&tex; is a development of the original &tex; program, as are
Xe&tex; and Lua&tex; (see <link linkend="TeX-engines">&tex; engines</link>).  They are completely
backward compatible.  But the original program had a custom output
format, DVI, while the newer ones can output directly to PDF.  This
allows them to take advantage of the extra features in PDF such as
hyperlinks, support for modern image formats such as JPG and PNG, and
ubiquitous viewing programs.  In short, if you run <command>pdflatex</command> or
<command>xelatex</command> or <command>lualatex</command> then you will by default get PDF
and have access to all its modern features.  If you run <command>latex</command>,
or <literal>dvilualatex</literal>, then you will get DVI.  The description here
assumes pdf&latex;.
</para>
<para>See <link linkend="Command-line-options">Command line options</link>, for a selection of the most useful
command line options. As to <replaceable>argument</replaceable>, the usual case is that it
does not begin with a backslash, so the system takes it to be the name
of a file and it compiles that file. If <replaceable>argument</replaceable> begins with a
backslash then the system will interpret it as a line of &latex;
input, which can be used for special effects (see <link linkend="Command-line-input">Command line
input</link>).
</para>
<para>If you gave no arguments or options then <command>pdflatex</command> prompts for
input from the terminal. You can escape from this by entering
<literal>&lt;control&gt;-D</literal>.
</para>
<para>If &latex; finds an error in your document then by default it stops and
asks you about it.  See <link linkend="Recovering-from-errors">Recovering from errors</link> for an outline of what
to do.
</para>


<sect1 label="28.1" id="Command-line-options">
<title>Command line options</title>

<indexterm role="cp"><primary>options, command line</primary></indexterm>

<para>These are the command-line options relevant to ordinary document
authoring. For a full list, try running &#8216;<literal>latex --help</literal>&#8217; from the
command line.
</para>
<para>With many implementations you can specify command line options by
prefixing them with &#8216;<literal>-</literal>&#8217; or &#8216;<literal>--</literal>&#8217;.  This is the case for
both &tex; Live (and Mac&tex;) and MiK&tex;.  We will use both
conventions interchangeably.
</para>
<variablelist><indexterm role="fn"><primary>--version command-line option</primary></indexterm>
<varlistentry><term><literal>-version</literal>
</term><listitem><para>Show the current version, like &#8216;<literal>pdfTeX 3.14159265-2.6-1.40.16 (TeX
Live 2015/Debian)</literal>&#8217; along with a small amount of additional information,
and exit.
</para>
<indexterm role="fn"><primary>--help command-line option</primary></indexterm>
</listitem></varlistentry><varlistentry><term><literal>-help</literal>
</term><listitem><para>Give a brief usage message that is useful as a prompt and exit.
</para>
<indexterm role="fn"><primary>--interaction command-line option</primary></indexterm>
</listitem></varlistentry><varlistentry><term><literal>-interaction=<replaceable>mode</replaceable></literal>
</term><listitem><para>&tex; compiles a document in one of four interaction modes:
<literal>batchmode</literal>, <literal>nonstopmode</literal>, <literal>scrollmode</literal>,
<literal>errorstopmode</literal>. In <firstterm>errorstop mode</firstterm> (the default), &tex;
stops at each error and asks for user intervention. In <firstterm>batch
mode</firstterm> it prints nothing on the terminal, errors are scrolled as if the
user hit <literal>&lt;return&gt;</literal> at every error, and missing files cause the
job to abort. In <firstterm>nonstop mode</firstterm>, diagnostic message appear on the
terminal but as in batch mode there is no user interaction. In
<firstterm>scroll mode</firstterm>, &tex; only stops for missing files or keyboard
input.
</para>
<para>For instance, starting &latex; with this command line
</para>
<screen>pdflatex -interaction=batchmode <replaceable>filename</replaceable>
</screen>
<para>eliminates most terminal output.
</para>
<indexterm role="fn"><primary>--jobname command-line option</primary></indexterm>
</listitem></varlistentry><varlistentry><term><literal>-jobname=<replaceable>string</replaceable></literal>
</term><listitem><para>Set the value of &tex;&#8217;s <literal>jobname</literal> to the string.  The log file
and output file will then be named <filename><replaceable>string</replaceable>.log</filename> and
<filename><replaceable>string</replaceable>.pdf</filename>.
</para>
<para>When you run <literal><command>pdflatex</command> <replaceable>options</replaceable> <replaceable>argument</replaceable></literal>, if
<replaceable>argument</replaceable> does not start with a backslash then &tex; considers it
the name of a file to input. Otherwise it waits for the first
<literal>\input</literal> instruction and the name of the input file will be the job
name. This is used to name the log file the output file.  This option
overrides that process and directly specifies the name.  See <link linkend="Command-line-input">Command
line input</link> for an example of its use.
</para>
<indexterm role="fn"><primary>--output-directory command-line option</primary></indexterm>
</listitem></varlistentry><varlistentry><term><literal>-output-directory=<replaceable>directory</replaceable></literal>
</term><listitem><para>Write files in the directory <replaceable>directory</replaceable>.  It must already exist.
</para>
<indexterm role="fn"><primary>--shell-escape command-line option</primary></indexterm>
<indexterm role="fn"><primary>--no-shell-escape command-line option</primary></indexterm>
<indexterm role="fn"><primary>--enable-write18 command-line option</primary></indexterm>
<indexterm role="fn"><primary>--disable-write18 command-line option</primary></indexterm>
</listitem></varlistentry><varlistentry><term><literal>shell-escape</literal>
</term><term><literal>no-shell-escape</literal>
</term><term><literal>enable-write18</literal>
</term><term><literal>disable-write18</literal>
</term><listitem><para>Enable or disable <literal>\write18{<replaceable>shell command</replaceable>}</literal>.  The first two
options are for with &tex; Live or Mac&tex; while the second two are
for MiK&tex;.
</para>
<indexterm role="cp"><primary>package, <literal>sagetex</literal></primary></indexterm>
<indexterm role="cp"><primary><literal>sagetex</literal> package</primary></indexterm>

<para>Sometimes you want to run external system commands from inside a
&latex; file. For instance the package <filename>sagetex</filename> allows you to
have the mathematics software system <emphasis>Sage</emphasis> do calculations or draw
graphs and then incorporate that output in your document.  For this
&tex; provides the <literal>\write18</literal> command.
</para>
<para>But with this functionality enabled, security issues could happen if you
compiled a &latex; file from the Internet.  By default <literal>\write18</literal>
is disabled.  (More precisely, by default &tex; Live, Mac&tex;, and
MiK&tex; only allow the execution of a limited number of &tex;-related
programs, which they distribute.)
</para>
<para>If you invoke &latex; with the option <literal>no-shell-escape</literal>, and in
your document you call <literal>\write18{ls -l}</literal>, then you do not get an
error but the log file says &#8216;<literal>runsystem(ls -l)...disabled</literal>&#8217;.
</para>
<indexterm role="fn"><primary>--halt-on-error command-line option</primary></indexterm>
</listitem></varlistentry><varlistentry><term><literal>-halt-on-error</literal>
</term><listitem><para>Stop processing at the first error.
</para>
<indexterm role="fn"><primary>--file-line-error command-line option</primary></indexterm>
<indexterm role="fn"><primary>--no-file-line-error command-line option</primary></indexterm>
</listitem></varlistentry><varlistentry><term><literal>-file-line-error</literal>
</term></varlistentry><varlistentry><term><literal>-no-file-line-error</literal>
</term><listitem><para>Enable or disable <literal><replaceable>filename</replaceable>:<replaceable>lineno</replaceable>:<replaceable>error</replaceable></literal>-style
error messages.  These are only available with &tex; Live or Mac&tex;.
</para></listitem></varlistentry></variablelist>

</sect1>
<sect1 label="28.2" id="Command-line-input">
<title>Command line input</title>

<indexterm role="cp"><primary>input, on command line</primary></indexterm>

<para>As part of the command line invocation <literal>pdflatex <replaceable>options</replaceable>
<replaceable>argument</replaceable></literal> you can specify arbitrary &latex; input by starting
<replaceable>argument</replaceable> with a backslash. This allows you to do some special
effects.
</para>
<indexterm role="cp"><primary>package, <literal>hyperref</literal></primary></indexterm>
<indexterm role="cp"><primary><literal>hyperref</literal> package</primary></indexterm>

<para>For example, this file (which uses the <filename>hyperref</filename> package for
hyperlinks) can produce two kinds of output, one for paper and one for a
PDF.
</para>
<screen>\ifdefined\paperversion        % in preamble
\newcommand{\urlcolor}{black}
\else
\newcommand{\urlcolor}{blue}
\fi
\usepackage[colorlinks=true,urlcolor=\urlcolor]{hyperref}
  ...
\href{https://www.ctan.org}{CTAN}  % in body
  ...
</screen>
<para>Compiling this document <filename>book.tex</filename> with the command line
<literal>pdflatex test</literal> will give the &#8216;<literal>CTAN</literal>&#8217; link in blue.  But
compiling it with <literal>pdflatex &quot;\def\paperversion{}\input test.tex&quot;</literal>
has the link in black.  (Note the use of double quotes to prevent
interpretation of the symbols by the command line shell; your system may
do this differently.)
</para>
<para>In a similar way, from the single file <filename>main.tex</filename> you can compile
two different versions.
</para>
<!-- credit Paul Gaborit: https://tex.stackexchange.com/a/220101/121234 -->
<screen>pdflatex -jobname=students &quot;\def\student{}\input{main}&quot;
pdflatex -jobname=teachers &quot;\def\teachers{}\input{main}&quot;
</screen>
<para>The <literal>jobname</literal> option is there because otherwise both files would be
called <filename>main.pdf</filename> and the second would overwrite the first.
</para>
<para>A final example.  This loads the package <filename>graphicx</filename> with the option
<literal>draft</literal>
</para>
<!-- credit Herbert Voss: https://tex.stackexchange.com/a/17236/121234 -->
<screen>pdflatex -jobname=aa &quot;\RequirePackage[draft]{graphicx}\input{aa.tex}&quot;
</screen>
<para>so the graphic files are read for their size information but not
incorporated into the PDF.  (The <literal>jobname</literal> option is needed because
otherwise the output file would be <filename>graphicx.pdf</filename>, as
<literal>\RequirePackage</literal> does an <literal>\input</literal> of its own.)
</para>

</sect1>
<sect1 label="28.3" id="Recovering-from-errors">
<title>Recovering from errors</title>

<para>If &latex; finds an error in your document then it gives you an error
message and prompts you with a question mark, <literal>?</literal>.  For instance,
running &latex; on this file
</para>
<screen>\newcommand{\NP}{\ensuremath{\textbf{NP}}}
The \PN{} problem is a million dollar one.
</screen>
<para>causes it show this, and wait for input.
</para>
<screen>! Undefined control sequence.
l.5 The \PN
           {} problem is a million dollar one.
? 
</screen>
<para>The simplest thing is to enter &#8216;<literal>x</literal>&#8217; and <literal>&lt;return&gt;</literal> and fix the
typo.  You could instead enter &#8216;<literal>?</literal>&#8217; and <literal>&lt;return&gt;</literal> to see other
options.
</para>
<indexterm role="cp"><primary>&#8216;<literal>*</literal>&#8217; prompt</primary></indexterm>
<indexterm role="cp"><primary>prompt, &#8216;<literal>*</literal>&#8217;</primary></indexterm>
<indexterm role="fn"><primary>\stop</primary></indexterm>
<para>There are two other error scenarios.  The first is that you forgot to
include the <literal>\end{document}</literal> or misspelled it.  In this case
&latex; gives you a &#8216;<literal>*</literal>&#8217; prompt.  You can get back to the command
line by typing <literal>\stop</literal> and <literal>&lt;return&gt;</literal>.
</para>
<para>The last scenario is that you mistyped the file name.  For instance,
instead of <literal>pdflatex test</literal> you might type <literal>pdflatex tste</literal>.
</para>
<screen>! I can't find file `tste'.
&lt;*&gt; tste
        
(Press Enter to retry, or Control-D to exit)
Please type another input file name:
</screen>
<para>The simplest thing is to enter <literal>&lt;Contol&gt;</literal> and &#8216;<literal>d</literal>&#8217; (holding
them down at the same time), and just fix the command line.
</para>

</sect1>
</chapter>
<appendix label="A" id="Document-templates">
<title>Document templates</title>

<indexterm role="cp"><primary>document templates</primary></indexterm>
<indexterm role="cp"><primary>templates, document</primary></indexterm>

<para>Although not reference material, perhaps these document templates will
be useful.  Additional template resources are listed at
<ulink url="http://tug.org/interest.html#latextemplates">http://tug.org/interest.html#latextemplates</ulink>.
</para>


<sect1 label="A.1" id="beamer-template">
<title><literal>beamer</literal> template</title>

<indexterm role="cp"><primary><literal>beamer</literal> template and class</primary></indexterm>
<indexterm role="cp"><primary>template, <literal>beamer</literal></primary></indexterm>

<para>The <literal>beamer</literal> class creates presentation slides.  It has a vast
array of features, but here is a basic template:
</para>
<screen>\documentclass{beamer}

\title{Beamer Class template}
\author{Alex Author}
\date{July 31, 2007}

\begin{document}

\maketitle

% without [fragile], any {verbatim} code gets mysterious errors.
\begin{frame}[fragile]
 \frametitle{First Slide}

\begin{verbatim}
  This is \verbatim!
\end{verbatim}

\end{frame}

\end{document}
</screen>
<para>One web resource for this:
<ulink url="http://robjhyndman.com/hyndsight/beamer/">http://robjhyndman.com/hyndsight/beamer/</ulink>.
</para>

</sect1>
<sect1 label="A.2" id="article-template">
<title><literal>article</literal> template</title>

<indexterm role="cp"><primary>template, <literal>article</literal></primary></indexterm>

<screen>\documentclass{article}
\title{Article Class Template}
\author{Alex Author}

\begin{document}
\maketitle

\section{First section}
Some text.

\subsection{First section, first subsection}
Additional text.

\section{Second section}
Some more text.
\end{document}
</screen>

</sect1>
<sect1 label="A.3" id="book-template">
<title><literal>book</literal> template</title>

<indexterm role="cp"><primary>template, <literal>book</literal></primary></indexterm>

<para>This is a straightforward template for a book.  See See <link linkend="Larger-book-template">Larger book
template</link> for a more elaborate one.
</para>
<screen>\documentclass{book}
\title{Book Class Template}
\author{Alex Author}

\begin{document}
\maketitle

\chapter{First}
Some text.

\chapter{Second}
Some other text.

\section{A subtopic}
The end.
\end{document}
</screen>

</sect1>
<sect1 label="A.4" id="Larger-book-template">
<title>Larger <literal>book</literal> template</title>

<indexterm role="cp"><primary>template, <literal>book</literal></primary></indexterm>

<para>This is a more elaborate template for a book.  It has
<literal>\frontmatter</literal>, <literal>\mainmatter</literal>, and <literal>\backmatter</literal> to
control the typography of the three main areas of a book
(see <link linkend="_005cfrontmatter-_0026-_005cmainmatter-_0026-_005cbackmatter">\frontmatter &amp; \mainmatter &amp; \backmatter</link>).  The book has a
bibliography and an index.
</para>
<para>Notable is that it uses <literal>\include</literal> and <literal>\includeonly</literal>
(see <link linkend="Splitting-the-input">Splitting the input</link>).  While you are working on a chapter you
can comment out all the other chapter entries from the argument to
<literal>\includeonly</literal>.  That will speed up compilation without losing any
information such as cross-references.  (Material that does not need to
come on a new page is brought in with <literal>\input</literal> instead of
<literal>\include</literal>.  You don&#8217;t get the cross-reference benefit this way.)
</para>
<screen>\documentclass[titlepage]{book}
\usepackage{makeidx}\makeindex

\title{Book Class Template}
\author{Alex Author}

\includeonly{%
  frontcover,
  preface,
  chap1,
  ...
  }
\begin{document}
\frontmatter
\include{frontcover}
  % maybe comment out while drafting:
\maketitle \input{dedication} \input{copyright}
\tableofcontents
\include{preface}
\mainmatter
\include{chap1}
...
\appendix
\include{appena}
...
\backmatter
\bibliographystyle{apalike}
\addcontentsline{toc}{chapter}{Bibliography}
\bibliography
\addcontentsline{toc}{chapter}{Index}
\printindex
\include{backcover}
\end{document}
</screen>

</sect1>
<sect1 label="A.5" id="tugboat-template">
<title><literal>tugboat</literal> template</title>

<indexterm role="cp"><primary>template, TUGboat</primary></indexterm>
<indexterm role="cp"><primary>TUGboat template</primary></indexterm>
<indexterm role="cp"><primary><literal>ltugboat</literal> class</primary></indexterm>

<para><citetitle>TUGboat</citetitle> is the journal of the &tex; Users Group,
<ulink url="http://tug.org/TUGboat">http://tug.org/TUGboat</ulink>.
</para>
<screen>\documentclass{ltugboat}

\usepackage{graphicx}
\usepackage{ifpdf}
\ifpdf
\usepackage[breaklinks,hidelinks]{hyperref}
\else
\usepackage{url}
\fi

%%% Start of metadata %%%

\title{Example \TUB\ article}

% repeat info for each author.
\author{First Last}
\address{Street Address \\ Town, Postal \\ Country}
\netaddress{user (at) example dot org}
\personalURL{http://example.org/~user/}

%%% End of metadata %%%

\begin{document}

\maketitle

\begin{abstract}
This is an example article for \TUB{}.
Please write an abstract.
\end{abstract}

\section{Introduction}

This is an example article for \TUB, linked from
\url{http://tug.org/TUGboat/location.html}.

We recommend the \texttt{graphicx} package for image inclusions, and the
\texttt{hyperref} package if active urls are desired (in the \acro{PDF}
output).  Nowadays \TUB\ is produced using \acro{PDF} files exclusively.

The \texttt{ltugboat} class provides these abbreviations (and many more):
% verbatim blocks are often better in \small
\begin{verbatim}[\small]
\AllTeX \AMS \AmS \AmSLaTeX \AmSTeX \aw \AW
\BibTeX \CTAN \DTD \HTML
\ISBN \ISSN \LaTeXe
\mf \MFB
\plain \POBox \PS
\SGML \TANGLE \TB \TP
\TUB \TUG \tug
\UNIX \XeT \WEB \WEAVE

\, \bull \Dash \dash \hyph

\acro{FRED} -&gt; {\small[er] fred}  % please use!
\cs{fred}   -&gt; \fred
\meta{fred} -&gt; &lt;fred&gt;
\nth{n}     -&gt; 1st, 2nd, ...
\sfrac{3/4} -&gt; 3/4
\booktitle{Book of Fred}
\end{verbatim}

For references to other \TUB\ issue, please use the format
\textsl{volno:issno}, e.g., ``\TUB\ 32:1'' for our \nth{100} issue.

This file is just a template.  The \TUB\ style documentation is the
\texttt{ltubguid} document at \url{http://ctan.org/pkg/tugboat}.  (For
\CTAN\ references, where sensible we recommend that form of url, using
\texttt{/pkg/}; or, if you need to refer to a specific file location,
\texttt{http://mirror.ctan.org/\textsl{path}}.)

Email \verb|tugboat@tug.org| if problems or questions.

\bibliographystyle{plain}  % we recommend the plain bibliography style
\nocite{book-minimal}      % just making the bibliography non-empty
\bibliography{xampl}       % xampl.bib comes with BibTeX

\makesignature
\end{document}
</screen>

</sect1>
</appendix>
<chapter label="" id="Index">
<title>Index</title>

<!-- Keep `Command Index' working for ltx-help.el. -->
<anchor id="Command-Index"/>
<index role="cp"></index>

</chapter>
</book>