This is an unofficial reference manual for LaTeX. See below for the Table of Contents. If you want a tutorial then please instead visit learnlatex.org or see this long list.

This manual has two versions. One has separate web pages for each section or subsection. It's also available as a single web page and as a pdf. Translations to French and Spanish are available at https://ctan.org/pkg/latex2e-help-texinfo; they're maintained separately.

This document is not official. It has not been reviewed by the LaTeX maintainers. Our ultimate goal is to cover all (non-private) LaTeX commands. Your comments and contributions, including bug reports, are very welcome. See our project page for more, including license information and information on how you can contribute to this manual as well as mirror it.

LaTeX2e: An unofficial reference manual

This document is an unofficial reference manual (version of May 2024) for LaTeX2e, a document preparation system.

Table of Contents


1 About this document

This is an unofficial reference manual for the LaTeX2e document preparation system, which is a macro package for the TeX typesetting program (see Overview of LaTeX).

This document’s home page is https://latexref.xyz; it has separate web pages for each topic. Alternatively. https://latexref.xyz/dev/latex2e.html has the entire document on a single page. For other output formats, the sources, and plenty more information, see https://latexref.xyz/dev/.

In this document, we will mostly just use ‘LaTeX’ rather than ‘LaTeX2e’, since the previous version of LaTeX (2.09) was frozen decades ago.

LaTeX is maintained by a group of volunteers (https://latex-project.org). The official documentation written by the LaTeX project is available from their web site. The present document is completely unofficial and has not been written or reviewed by the LaTeX maintainers. Do not send bug reports or anything else about this document to them. Instead, please send all comments to . This is a public list; you can (un)subscribe, view the archives, etc., at https://lists.tug.org/latexrefman.

This document is a reference, not a tutorial. There is a vast array of other information available about LaTeX, at all levels. Here are a few introductions.

https://ctan.org/pkg/latex-doc-ptr

Two pages of recommended references to LaTeX documentation.

https://ctan.org/pkg/first-latex-doc

Writing your first document, with a bit of both text and math.

https://ctan.org/pkg/lshort

A longer introduction to LaTeX, translated to many languages.

https://tug.org/begin.html

Overview of getting started with TeX and LaTeX.


2 Overview of LaTeX

LaTeX is a system for typesetting documents. It was originally created by Leslie Lamport in 1984, but has been maintained by a group of volunteers for many years now (https://latex-project.org). It is widely used, particularly but not exclusively for mathematical and technical documents.

A LaTeX user writes an input file containing text to be typeset along with interspersed commands. The default encoding for the text is UTF-8 (as of 2018). The commands specify, for example, how the text should be formatted.

LaTeX is implemented as a set of so-called “macros” (a TeX format) which use Donald E. Knuth’s TeX typesetting program or one of its derivatives, collectively known as “engines”. Thus, the user produces output, typically PDF, by giving the input file to a TeX engine. The following sections describe all this in more detail.

The term LaTeX is also sometimes used to mean the language in which the input document is marked up, that is, to mean the set of commands available to a LaTeX user.

The name LaTeX is short for “Lamport TeX”. It is pronounced LAH-teck or LAY-teck, or sometimes LAY-tecks. Inside a document, produce the logo with \LaTeX. Where use of the logo is not sensible, such as in plain text, write it as ‘LaTeX’.


2.1 Starting and ending

LaTeX files have a simple global structure, with a standard beginning and ending. Here is a small example:

\documentclass{article}
\begin{document}
Hello, \LaTeX\ world.
\end{document}

Every LaTeX document has a \begin{document} line and an \end{document} line.

Here, the ‘article’ is the document class. It is implemented in a file article.cls. You can use any document class available on your system. A few document classes are defined by LaTeX itself, and a vast array of others are available. See Document classes.

You can include other LaTeX commands between the \documentclass and the \begin{document} commands. This area is called the preamble.

The \begin{document}\end{document} pair defines an environment; the ‘document’ environment (and no others) is required in all LaTeX documents (see document). LaTeX provides many environments that are documented here (see Environments). Many more are available to you from external packages, most importantly those available at CTAN (see CTAN: The Comprehensive TeX Archive Network).

The following sections discuss how to produce PDF or other output from a LaTeX input file.


2.2 Output files

LaTeX produces a main output file and at least two auxiliary files. The main output file’s name ends in either .dvi or .pdf.

.dvi

If LaTeX is invoked with the system command latex then it produces a DeVice Independent file, with extension .dvi. You can view this file with a command such as xdvi, or convert it to a PostScript .ps file with dvips or to a Portable Document Format .pdf file with dvipdfmx. The contents of the file can be dumped in human-readable form with dvitype. A vast array of other DVI utility programs are available (https://mirror.ctan.org/dviware).

.pdf

If LaTeX is invoked via the system command pdflatex, among other commands (see TeX engines), then the main output is a Portable Document Format (PDF) file. Typically this is a self-contained file, with all fonts and images included.

LaTeX always produces at least two additional files.

.log

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.

.aux

Auxiliary information is used by LaTeX for things such as cross references. For example, the first time that LaTeX finds a forward reference—a cross reference to something that has not yet appeared in the source—it will appear in the output as a doubled question mark ??. When the referred-to spot does eventually appear in the source then LaTeX writes its location information to this .aux 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.

LaTeX may produce yet more files, characterized by the filename ending. These include a .lof file that is used to make a list of figures, a .lot file used to make a list of tables, and a .toc file used to make a table of contents (see Table of contents, list of figures, list of tables). A particular class may create others; the list is open-ended.


2.3 TeX engines

LaTeX is a large set of commands (macros) that is executed by a TeX program (see Overview of LaTeX). Such a set of commands is called a format, and is embodied in a binary .fmt file, which can be read much more quickly than the corresponding TeX source.

This section gives a terse overview of the TeX programs that are commonly available (see also Command line interface).

latex
pdflatex

In TeX Live (https://tug.org/texlive), if LaTeX is invoked via either the system command latex or pdflatex, then the pdfTeX engine is run (https://ctan.org/pkg/pdftex). When invoked as latex, the main output is a .dvi file; as pdflatex, the main output is a .pdf file.

pdfTeX incorporates the e-TeX extensions to Knuth’s original program (https://ctan.org/pkg/etex), 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 etex, but this is plain TeX (and produces .dvi).

In other TeX distributions, latex may invoke e-TeX rather than pdfTeX. In any case, the e-TeX extensions can be assumed to be available in LaTeX, and a few extensions beyond e-TeX, particularly for file manipulation.

lualatex

If LaTeX is invoked via the system command lualatex, the LuaTeX engine is run (https://ctan.org/pkg/luatex). This program allows code written in the scripting language Lua (http://luatex.org) to interact with TeX’s typesetting. LuaTeX handles UTF-8 Unicode input natively, can handle OpenType and TrueType fonts, and produces a .pdf file by default. There is also dvilualatex to produce a .dvi file.

xelatex

If LaTeX is invoked with the system command xelatex, the XeTeX engine is run (https://tug.org/xetex). Like LuaTeX, XeTeX natively supports UTF-8 Unicode and TrueType and OpenType fonts, though the implementation is completely different, mainly using external libraries instead of internal code. XeTeX produces a .pdf file as output; it does not support DVI output.

Internally, XeTeX creates an .xdv file, a variant of DVI, and translates that to PDF using the (x)dvipdfmx program, but this process is automatic. The .xdv file is only useful for debugging.

hilatex

If LaTeX is invoked via the system command hilatex, the HiTeX engine is run (https://ctan.org/pkg/hitex). This program produces its own format, named HINT, designed especially for high-quality typesetting on mobile devices.

platex
uplatex

These commands provide significant additional support for Japanese and other languages; the u variant supports Unicode. See https://ctan.org/pkg/ptex and https://ctan.org/pkg/uptex.

As of 2019, there is a companion -dev command and format for all of the above, except hitex:

dvilualatex-dev
latex-dev
lualatex-dev
pdflatex-dev
platex-dev
uplatex-dev
xelatex-dev

These are candidates for an upcoming LaTeX release. The main purpose is to find and address compatibility problems before an official release.

These -dev formats make it easy for anyone to help test documents and code: you can run, say, pdflatex-dev instead of pdflatex, without changing anything else in your environment. Indeed, it is easiest and most helpful to always run the -dev versions instead of bothering to switch back and forth. During quiet times after a release, the commands will be equivalent.

These are not daily snapshots or untested development code. They undergo the same extensive regression testing by the LaTeX team before being released.

For more information, see “The LaTeX release workflow and the LaTeX dev formats” by Frank Mittelbach, TUGboat 40:2, https://tug.org/TUGboat/tb40-2/tb125mitt-dev.pdf.


2.4 Input text

To a first approximation, most input characters in LaTeX print as themselves. But there are exceptions, as discussed in the following sections.


2.4.1 Input encodings

The input to TeX (or any computer program) ultimately consists of a sequence of bytes. (Nowadays, a byte is almost universally an eight-bit number, i.e., an integer between 0 and 255, inclusive.) The input encoding defines how to interpret that sequence of bytes, and thus how LaTeX behaves.

Today, by far the most common way to encode text is with UTF-8, a so-called “Unicode Transformation Format” which specifies how to transform a sequence of 8-bit bytes to Unicode code points, which are defined independent of any particular representation. The Unicode encoding defines code points for virtually all characters used today in written text.

When TeX was created, Unicode and UTF-8 did not exist and the 7-bit ASCII encoding was by far the most widely used. So TeX does not require Unicode for text input. UTF-8 is a superset of ASCII, so a pure 7-bit ASCII document is also UTF-8.

Since 2018, the default input encoding for LaTeX is UTF-8. Some methods for handling documents written in some other encoding, such as ISO-8859-1 (Latin 1), are explained in inputenc package.

You can easily find more about all these topics in any introductory computer text or online. For example, you might start at: https://en.wikipedia.org/wiki/Unicode.


2.4.2 Ligatures

A ligature combines two or more letters (more generally, characters) into a single glyph. For example, in Latin-based typography, the two letters ‘f’ and ‘i’ are often combined into the glyph ‘fi’.

TeX supports ligatures automatically. To continue the example, if the input has the word ‘fine’, written as four separate ASCII characters, TeX will output the word ‘fine’ (with the default fonts), with three typeset glyphs.

In traditional TeX, the available ligatures, if any, are defined by the current font. TeX also uses the ligature mechanism to produce a few typographical characters which were not available in any computer encoding when TeX was invented. In all, in the original Computer Modern fonts, the following input character sequences are defined to lead to ligatures:

ff

ff (ff ligature, U+FB00)

fi

fi (fi ligature, U+FB01)

fl

fl (fl ligature, U+FB02)

ffi

ffi (ffi ligature, U+FB03)

ffl

ffl (ffl ligature, U+FB04)

``

“ (left double quotation mark, U+201C)

''

” (right double quotation mark, U+201D)

--

– (en-dash, U+2013)

---

— (em-dash, U+2014)

!`

!‘ (inverted exclamation mark, U+00A1)

?`

?‘ (inverted question mark, U+00BF)

(For the f-ligatures above, the text in parentheses shows the individual characters, so in the typeset output you can easily see the difference between the ligature and the original character sequence.)

Nowadays it’s usually possible to directly input the punctuation characters as Unicode characters, and LaTeX supports that (see previous section). But even today, it can still often be useful to use the ASCII ligature input form; for example, the difference between an en-dash and em-dash, as a single glyph, can be all but impossible to discern, but the difference between two and three ASCII hyphen characters is clear. Similarly with quotation marks, in some fonts.

Thus, even the engines with native support for UTF-8, namely LuaTeX and XeTeX, also support the ASCII ligature input sequences by default, independent of the font used. They also need to do so for compatibility.

By the way, the f-ligatures are also available in Unicode (the “Alphabetic Presentation Forms” block starting at U+FB00), but it’s almost never desirable to use them as input characters, since in principle it should be up to the typesetter and the current font whether to use ligatures. Also, in practice, using them will typically cause searches to fail, that is, a search for the two characters ‘fi’ will not be matched by the ligature ‘fi’ at U+FB01.


2.4.3 Special characters: \ { } % $ & _ ^ # ~

Besides ligatures (see previous section), a few individual characters have special meaning to LaTeX. They are called reserved characters or special characters. Here they are:

\

Introduces a command name, as seen throughout this manual.

{
}

Delimits a required argument to a command or a level of grouping, as seen throughout this manual.

%

Starts a comment: the ‘%’ and all remaining characters on the current line are ignored.

$

Starts and ends math mode (see Math formulas).

&

Separates cells in a table (see tabular).

_
^

Introduce a subscript or superscript, respectively, in math (see Subscripts & superscripts); they produce an error outside math mode. As a little-used special feature, two superscript characters in a row can introduce special notation for an arbitrary character.

#

Stands for arguments in a macro definition (see \newcommand & \renewcommand).

~

Produces a nonbreakable interword space (see ~, \nobreakspace).

See Printing special characters, for how to typeset these characters when you need them literally.


2.5 LaTeX command syntax

In the LaTeX input file, a command name starts with a backslash character, \. The name itself then consists of either (a) a string of letters or (b) a single non-letter.

LaTeX commands names are case sensitive; for example, \pagebreak differs from \Pagebreak (the latter is not a standard command). Most command names are lowercase, but in any event you must enter all commands in the same case as they are defined.

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, {...}. Optional arguments are contained in square brackets, [...]. Generally, but not universally, if the command accepts an optional argument, it comes first, before any required arguments; optional arguments could come after required arguments, or both before and after.

Inside of an optional argument, to use the character close square bracket (]) hide it inside curly braces, as in \item[closing bracket {]}]. 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.

LaTeX has the convention that some commands have a * form that is closely related to the form without a *, such as \chapter and \chapter*. The difference in behavior varies from command to command.

This manual describes all accepted options and *-forms for the commands it covers (barring unintentional omissions, a.k.a. bugs).

As of the 2020-10-01 release of LaTeX, the expl3 and xparse packages are part of the LaTeX2e format. They provide a completely different underlying programming language syntax. We won’t try to cover that in this document; see the related package documentation and other LaTeX manuals.


2.6 Environment syntax

Synopsis:

\begin{environment-name}
  ...
\end{environment-name}

An environment is an area of LaTeX source, inside of which there is a distinct behavior. For instance, for poetry in LaTeX put the lines between \begin{verse} and \end{verse}.

\begin{verse}
  There once was a man from Nantucket \\
  ...
\end{verse}

See Environments, for a list of environments. Particularly notable is that every LaTeX document must have a document environment, a \begin{document} ... \end{document} pair.

The environment-name at the beginning must exactly match that at the end. This includes the case where environment-name ends in a star (*); both the \begin and \end texts must include the star.

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).

\begin{tabular}[t]{r|l}
  ... rows-of-table ...
\end{tabular}

2.7 \DocumentMetadata: Producing tagged PDF output

The \DocumentMetadata command was added to LaTeX in 2022. It enables so-called “tagging” of the PDF output, aiding accessibility of the PDF. It is supported best with LuaLaTeX; pdfLaTeX and XeLaTeX are supported as well as possible (see TeX engines).

It is unlike nearly any other command in LaTeX in that it must occur before the \documentclass command that starts a LaTeX document proper (see \documentclass). Therefore it must be called with \RequirePackage rather than \usepackage (see \RequirePackage).

This support is still in development, so we will not try to list all the possible settings. Please see the documentmetadata-support-doc document, part of the latex-lab package (https://ctan.org/pkg/latex-lab). Here is a simple example which enables most tagging currently implemented:

\DocumentMetadata{testphase={phase-III,firstaid}}
\documentclass{article}
...

As you can see from the key name testphase, this is all still in an experimental phase. The LaTeX developers strongly encourage users to give it a try and report problems, so it can be improved.


2.8 CTAN: The Comprehensive TeX Archive Network

The Comprehensive TeX Archive Network, CTAN, is the TeX and LaTeX community’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 https://ctan.org.

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’s specifications.

In addition to its massive holdings, the ctan.org web site offers features such as search by name or by functionality.

CTAN is not a single host, but instead is a set of hosts, one of which is the so-called “master”. The master host actively manages the material, for instance, by accepting uploads of new or updated packages. For many years, it has been 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 master 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 master site. The list of mirrors is at https://ctan.org/mirrors.


3 Document classes

The document’s overall class is defined with the \documentclass command, which is normally the first command in a LaTeX source file.

\documentclass[options]{class}

The following document class names are built into LaTeX. Many other document classes are available as separate packages (see Overview of LaTeX).

article

For a journal article, a presentation, and miscellaneous general use.

book

Full-length books, including chapters and possibly including front matter, such as a preface, and back matter, such as an appendix (see Front/back matter).

letter

Mail, optionally including mailing labels (see Letters).

report

For documents of length between an article and a book, such as technical reports or theses, which may contain several chapters.

slides

For slide presentations—rarely used nowadays. The beamer package is perhaps the most prevalent replacement (https://ctan.org/pkg/beamer). See beamer template, for a small template for a beamer document.

Standard options are described in the next section.


3.1 Document class options

You can specify global options or class options to the \documentclass command by enclosing them in square brackets. To specify more than one option, separate them with a comma.

\documentclass[option1,option2,...]{class}

LaTeX automatically passes options specified for \documentclass on to any other loaded classes that can handle them.

Here is the list of the standard class options.

All of the standard classes except slides accept the following options for selecting the typeface size; the default is 10pt:

10pt  11pt  12pt

All of the standard classes accept these options for selecting the paper size (dimensions are listed height by width):

a4paper

210 by 297mm (about 8.25 by 11.75 inches)

a5paper

148 by 210mm (about 5.8 by 8.3 inches)

b5paper

176 by 250mm (about 6.9 by 9.8 inches)

executivepaper

7.25 by 10.5 inches

legalpaper

8.5 by 14 inches

letterpaper

8.5 by 11 inches (the default)

When using one of the engines pdfLaTeX, LuaLaTeX, or XeLaTeX (see TeX engines), options other than letterpaper set the print area but you must also set the physical paper size. Usually, the geometry package is the best way to do that; it provides flexible ways of setting the print area and physical page size. Otherwise, setting the paper size is engine-dependent. For example, with pdfLaTeX, you could include \pdfpagewidth=\paperwidth and \pdfpageheight=\paperheight in the preamble.

Miscellaneous other options:

draft
final

Mark (draft) or do not mark (final) overfull boxes with a black box in the margin; default is final.

fleqn

Put displayed formulas flush left; default is centered.

landscape

Selects landscape format; default is portrait.

leqno

Put equation numbers on the left side of equations; default is the right side.

openbib

Use “open” bibliography format.

titlepage
notitlepage

Specifies whether there is a separate page for the title information and for the abstract also, if there is one. The default for the report class is titlepage, for the other classes it is notitlepage.

The following options are not available with the slides class.

onecolumn
twocolumn

Typeset in one or two columns; default is onecolumn.

oneside
twoside

Selects one- or two-sided layout; default is oneside, except that in the book class the default is twoside.

For one-sided printing, the text is centered on the page. For two-sided printing, the \evensidemargin (\oddsidemargin) parameter determines the distance on even (odd) numbered pages between the left side of the page and the text’s left margin, with \oddsidemargin being 40% of the difference between \paperwidth and \textwidth, and \evensidemargin is the remainder.

openright
openany

Determines if a chapter should start on a right-hand page; default is openright for book, and openany for report.

The slides class offers the option clock for printing the time at the bottom of each note.


3.2 \usepackage: Additional packages

To load a package pkg, with the package options given in the comma-separated list options:

\usepackage[options]{pkg}[mindate]

To specify more than one package you can separate them with a comma, as in \usepackage{pkg1,pkg2,...}, or use multiple \usepackage commands.

If the mindate optional argument is given, LaTeX gives a warning if the loaded package has an earlier date, i.e., is too old. The mindate argument must be in the form YYYY/MM/DD. More info on this: https://tex.stackexchange.com/questions/47743.

\usepackage must be used in the document preamble, between the \documentclass declaration and the \begin{document}. Occasionally it is necessary to load packages before the \documentclass; see \RequirePackage for that (see \RequirePackage).

Any options given in the global \documentclass command that are unknown to the selected document class are passed on to the packages loaded with \usepackage.


3.3 Class and package creation

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 smcmemo.cls and begin your documents with \documentclass{smcmemo}.

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 be Memo from the SMC Math Department is for a class.

Inside of a class or package definition you can use the at-sign @ as a character in command names without having to surround the code containing that command with \makeatletter and \makeatother (see \makeatletter & \makeatother). This allows you to create commands that users will not accidentally redefine.

It is also highly desirable to prefix class- or package-specific commands with your package name or similar string, to prevent your definitions from clashing with those from other packages. For instance, the class smcmemo might have commands \smc@tolist, \smc@fromlist, etc.


3.3.1 Class and package structure

A class file or package file typically has four parts.

  1. In the identification part, the file says that it is a LaTeX package or class and describes itself, using the \NeedsTeXFormat and \ProvidesClass or \ProvidesPackage commands.
  2. The preliminary declarations part 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 smcmemo class might be called with an option to read in a file with a list of people for the to-head, as \documentclass[mathto]{smcmemo}, and therefore needs to define a command \newcommand{\setto}[1]{\def\@tolist{#1}} used in that file.
  3. In the handle options part the class or package declares and processes its options. Class options allow a user to start their document as \documentclass[option list]{class name}, to modify the behavior of the class. An example is when you declare \documentclass[11pt]{article} to set the default document font size.
  4. Finally, in the more declarations part the class or package usually does most of its work: declaring new variables, commands and fonts, and loading other files.

Here is a starting class file, which should be saved as stub.cls where LaTeX can find it, for example in the same directory as the .tex file.

\NeedsTeXFormat{LaTeX2e}
\ProvidesClass{stub}[2017/07/06 stub to start building classes from]
\DeclareOption*{\PassOptionsToClass{\CurrentOption}{article}}
\ProcessOptions\relax
\LoadClass{article}

It identifies itself, handles the class options via the default of passing them all to the article class, and then loads the article class to provide the basis for this class’s code.

For more, see the official guide for class and package writers, the Class Guide, at https://ctan.org/pkg/clsguide (much of the description here derives from this document), or the tutorial at https://tug.org/TUGboat/tb26-3/tb84heff.pdf.

See Class and package commands, for some of the commands specifically intended for class and package writers.


4 Fonts

LaTeX comes with powerful font capacities. For one thing, its New Font Selection Scheme allows you to work easily with the font families in your document (for instance, see Font styles). And, LaTeX documents can use most fonts that are available today, including versions of Times Roman, Helvetica, Courier, etc. (Note, though, that many fonts do not have support for mathematics.)

The first typeface in the TeX world was the Computer Modern family, developed by Donald Knuth. It is the default for LaTeX documents and is still the most widely used. But changing to another font often only involves a few commands. For instance, putting the following in your preamble gives you a Palatino-like font, which is handsome and more readable online than many other fonts, while still allowing you to typeset mathematics. (This example is from Michael Sharpe, https://math.ucsd.edu/~msharpe/RcntFnts.pdf.)

\usepackage[osf]{newpxtext} % osf for text, not math
\usepackage{cabin} % sans serif
\usepackage[varqu,varl]{inconsolata} % sans serif typewriter
\usepackage[bigdelims,vvarbb]{newpxmath} % bb from STIX
\usepackage[cal=boondoxo]{mathalfa} % mathcal

In addition, the xelatex or lualatex engines allow you to use any fonts on your system that are in OpenType or TrueType format (see TeX engines).

The LaTeX Font Catalogue (https://tug.org/FontCatalogue) shows font sample graphics and copy-and-pasteable source to use many fonts, including many with support for mathematics. It aims to cover all Latin alphabet free fonts available for easy use with LaTeX.

More information is also available from the TeX Users Group, at https://www.tug.org/fonts/.


4.1 fontenc package

Synopsis:

\usepackage[font_encoding]{fontenc}

or

\usepackage[font_encoding1, font_encoding2, ...]{fontenc}

Specify the font encodings. A font encoding is a mapping of the character codes to the font glyphs that are used to typeset your output.

This package only applies if you use the pdflatex engine (see TeX engines). If you use the xelatex or lualatex engine then instead use the fontspec package.

TeX’s original font family, Computer Modern, has a limited character set. For instance, to make common accented characters you must use \accent (see \accent) but this disables hyphenation. TeX users have agreed on a number of standards to access the larger sets of characters provided by modern fonts. If you are using pdflatex then put this in the preamble

\usepackage[T1]{fontenc}

gives you support for the most widespread European languages, including French, German, Italian, Polish, and others. In particular, if you have words with accented letters then LaTeX will hyphenate them and your output can be copied and pasted. (The optional second line allows you to directly enter accented characters into your source file.)

If you are using an encoding such as T1 and the characters appear blurry or do not magnify well then your fonts may be bitmapped, sometimes called raster or Type 3. You want vector fonts. Use a package such as lmodern or cm-super to get a font that extends LaTeX’s default using vector fonts.

For each font_encoding given as an option but not already declared, this package loads the encoding definition files, named font_encodingenc.def. It also sets \encodingdefault to be the last encoding in the option list.

These are the common values for font_encoding:

OT1

The original 7-bit encoding for TeX. Limited to mostly English characters.

OMS, OML

Math symbols and math letters encoding.

T1

TeX text extended. Sometimes called the Cork encoding for the users group meeting where it was developed (1990). Gives access to most European accented characters. The most common option for this package.

TS1

Text Companion encoding.

LaTeX’s default is to load OML, T1, OT1, and then OMS, and set the default to OT1.

Even if you do not use accented letters, you may need to specify a font encoding if your font requires it.

If you use T1 encoded fonts other than the default Computer Modern family then you may need to load the package that selects your fonts before loading fontenc, to prevent the system from loading any T1 encoded fonts from the default.

The LaTeX team reserves encoding names starting with: ‘T’ for the standard text encodings with 256 characters, ‘TS’ for symbols that extend the corresponding T encodings, ‘X’ for test encodings, ‘M’ for standard math encodings with 256 characters, ‘A’ for special applications, ‘OT’ for standard text encodings with 128 characters, and ‘OM’ for standard math encodings with 128 characters (‘O’ stands for ‘obsolete’).

This package provides a number of commands, detailed below. Many of them are encoding-specific, so if you have defined a command that works for one encoding but the current encoding is different then the command is not in effect.


4.1.1 \DeclareFontEncoding

Synopsis:

\DeclareFontEncoding{encoding}{text-settings}{math-settings}

Declare the font encoding encoding. It also saves the value of encoding in \LastDeclaredEncoding (see \LastDeclaredEncoding).

The file t1enc.def contains this line (followed by many others).

\DeclareFontEncoding{T1}{}{}

The text-settings are the commands that LaTeX will run every time it switches from one encoding to another with the \selectfont and \fontencoding commands. The math-settings are the commands that LaTeX will use whenever the font is accessed as a math alphabet.

LaTeX ignores any space characters inside text-settings and math-settings, to prevent unintended spaces in the output.

If you invent an encoding you should pick a two or three letter name starting with ‘L’ for ‘local’, or ‘E’ for ‘experimental’.

Note that output encoding files may be read several times by LaTeX so using, e.g., \newcommand may cause an error. In addition, such files should contain \ProvidesFile line (see Class and package commands).

Note also that you should use the \...Default commands only in a package, not in the encoding definition files, since those files should only contain declarations specific to that encoding.


4.1.2 \DeclareTextAccent

Synopsis:

\DeclareTextAccent{cmd}{encoding}{slot}

Define an accent, to be put on top of other glyphs, in the encoding encoding at the location slot.

A slot is the number identifying a glyph within a font.

This line from t1enc.def declares that to make a circumflex accent as in \^A, the system will put the accent in slot 2 over the ‘A’ character, which is represented in ASCII as 65. (This holds unless there is a relevant DeclareTextComposite or \DeclareTextCompositeCommand declaration; see \DeclareTextComposite.)

\DeclareTextAccent{\^}{T1}{2}

If cmd has already been defined then \DeclareTextAccent does not give an error but it does log the redefinition in the transcript file.


4.1.3 \DeclareTextAccentDefault

Synopsis:

\DeclareTextAccentDefault{\cmd}{encoding}

If there is an encoding-specific accent command \cmd but there is no associated \DeclareTextAccent for that encoding then this command will pick up the slack, by saying to use it as described for encoding.

For example, to make the encoding OT1 be the default encoding for the accent \", declare this.

\DeclareTextAccentDefault{\"}{OT1}

If you issue a \" when the current encoding does not have a definition for that accent then LaTeX will use the definition from OT1

That is, this command is equivalent to this call (see \UseTextSymbol & \UseTextAccent).

\DeclareTextCommandDefault[1]{\cmd}
   {\UseTextAccent{encoding}{\cmd}{#1}}

Note that \DeclareTextAccentDefault works for any one-argument fontenc command, not just the accent command.


4.1.4 \DeclareTextCommand & \ProvideTextCommand

Synopsis, one of:

\DeclareTextCommand{\cmd}{encoding}{defn}
\DeclareTextCommand{\cmd}{encoding}[nargs]{defn}
\DeclareTextCommand{\cmd}{encoding}[nargs][optargdefault]{defn}

or one of:

\ProvideTextCommand{\cmd}{encoding}{defn}
\ProvideTextCommand{\cmd}{encoding}[nargs]{defn}
\ProvideTextCommand{\cmd}{encoding}[nargs][optargdefault]{defn}

Define the command \cmd, which will be specific to one encoding. The command name cmd must be preceded by a backslash, \. These commands can only appear in the preamble. Redefining \cmd does not cause an error. The defined command will be robust even if the code in defn is fragile (see \protect).

For example, the file t1enc.def contains this line.

\DeclareTextCommand{\textperthousand}{T1}{\%\char 24 }

With that, you can express parts per thousand.

\usepackage[T1]{fontenc}  % in preamble
  ...
Legal limit is \( 0.8 \)\textperthousand.

If you change the font encoding to OT1 then you get an error like ‘LaTeX Error: Command \textperthousand unavailable in encoding OT1’.

The \ProvideTextCommand variant does the same, except that it does nothing if \cmd is already defined. The \DeclareTextSymbol command is faster than this one for simple slot-to-glyph association (see \DeclareTextSymbol)

The optional nargs and optargdefault arguments play the same role here as in \newcommand (see \newcommand & \renewcommand). Briefly, nargs is an integer from 0 to 9 specifying the number of arguments that the defined command \cmd takes. This number includes any optional argument. Omitting this argument is the same as specifying 0, meaning that \cmd will have no arguments. And, if optargdefault is present then the first argument of \cmd is optional, with default value optargdefault (which may be the empty string). If optargdefault is not present then \cmd does not take an optional argument.


4.1.5 \DeclareTextCommandDefault & \ProvideTextCommandDefault

Synopsis:

\DeclareTextCommandDefault{\cmd}{defn}

or:

\ProvideTextCommandDefault{\cmd}{defn}

Give a default definition for \cmd, for when that command is not defined in the encoding currently in force. This default should only use encodings known to be available.

This makes \copyright available.

\DeclareTextCommandDefault{\copyright}{\textcircled{c}}

It uses only an encoding (OMS) that is always available.

The \DeclareTextCommandDefault should not occur in the encoding definition files since those files should declare only commands for use when you select that encoding. It should instead be in a package.

As with the related non-default commands, the \ProvideTextCommandDefault has exactly the same behavior as \DeclareTextCommandDefault except that it does nothing if \cmd is already defined (see \DeclareTextCommand & \ProvideTextCommand). So, packages can use it to provide fallbacks that other packages can improve upon.


4.1.6 \DeclareTextComposite

Synopsis:

\DeclareTextComposite{\cmd}{encoding}{simple_object}{slot}

Access an accented glyph directly, that is, without having to put an accent over a separate character.

This line from t1enc.def means that \^o will cause LaTeX to typeset lowercase ‘o’ by taking the character directly from slot 224 in the font.

\DeclareTextComposite{\^}{T1}{o}{244}

See fontenc package, for a list of common encodings. The simple_object should be a single character or a single command. The slot argument is usually a positive integer represented in decimal (although octal or hexadecimal are possible). Normally \cmd has already been declared for this encoding, either with \DeclareTextAccent or with a one-argument \DeclareTextCommand. In t1enc.def, the above line follows the \DeclareTextAccent{\^}{T1}{2} command.


4.1.7 \DeclareTextCompositeCommand

Synopsis:

\DeclareTextCompositeCommand{\cmd}{encoding}{arg}{code}

A more general version of \DeclareTextComposite that runs arbitrary code with \cmd.

This allows accents on ‘i’ to act like accents on dotless i, \i.

\DeclareTextCompositeCommand{\'}{OT1}{i}{\'\i}

See fontenc package, for a list of common encodings. Normally \cmd will have already been declared with \DeclareTextAccent or as a one argument \DeclareTextCommand.


4.1.8 \DeclareTextSymbol

Synopsis:

\DeclareTextSymbol{\cmd}{encoding}{slot}

Define a symbol in the encoding encoding at the location slot. Symbols defined in this way are for use in text, not mathematics.

For example, this line from t1enc.def declares the number of the glyph to use for «, the left guillemet.

\DeclareTextSymbol{\guillemetleft}{T1}{19}

The command \DeclareTextCommand{\guillemetleft}{T1}{\char 19} has the same effect but is slower (see \DeclareTextCommand & \ProvideTextCommand).

See fontenc package, for a list of common encodings. The slot can be specified in decimal, or octal (as in '023), or hexadecimal (as in "13), although decimal has the advantage that single quote or double quote could be redefined by another package.

If \cmd has already been defined then \DeclareTextSymbol does not give an error but it does log the redefinition in the transcript file.


4.1.9 \DeclareTextSymbolDefault

Synopsis:

\DeclareTextSymbolDefault{\cmd}{encoding}

If there is an encoding-specific symbol command \cmd but there is no associated \DeclareTextSymbol for that encoding, then this command will pick up the slack, by saying to get the symbol as described for encoding.

For example, to declare that if the current encoding has no meaning for \textdollar then use the one from OT1, declare this.

\DeclareTextSymbolDefault{\textdollar}{OT1}

That is, this command is equivalent to this call (see \UseTextSymbol & \UseTextAccent).

\DeclareTextCommandDefault{\cmd}
   {\UseTextSymbol{encoding}{\cmd}}

Note that \DeclareTextSymbolDefault can be used to define a default for any zero-argument fontenc command.


4.1.10 \LastDeclaredEncoding

Synopsis:

\LastDeclaredEncoding

Get the name of the most recently declared encoding. The \DeclareFontEncoding command stores the name so that it can be retrieved with this command (see \DeclareFontEncoding).

This relies on \LastDeclaredEncoding rather than give the name of the encoding explicitly.

\DeclareFontEncoding{JH1}{}{}
\DeclareTextAccent{\'}{\LastDeclaredEncoding}{0}

4.1.11 \UseTextSymbol & \UseTextAccent

Synopsis:

\UseTextSymbol{encoding}{\cmd}

or:

\UseTextAccent{encoding}{\cmd}{text}

Use a symbol or accent not from the current encoding.

In general, to use a fontenc command in an encoding where it is not defined, and if the command has no arguments, then you can use it like this:

\UseTextSymbol{OT1}{\ss}

which is equivalent to this (note the outer braces form a group, so LaTeX reverts back to the prior encoding after the \ss):

{\fontencoding{OT1}\selectfont\ss}

Similarly, to use a fontenc command in an encoding where it is not defined, and if the command has one argument, you can use it like this:

\UseTextAccent{OT1}{\'}{a}

which is equivalent to this (again note the outer braces forming a group):

{fontencoding{OT1}\selectfont\'{\fontencoding{enc_in_use}\selectfont a}}

Here, enc_in_use is the encoding in force before this sequence of commands, so that ‘a’ is typeset using the current encoding and only the accent is taken from OT1.


4.2 Font styles

The following type style commands are supported by LaTeX.

In the table below the listed commands, the \text... commands, are used with an argument as in \textit{text}. This is the preferred form. But shown after it in parenthesis is the corresponding declaration form, which is often useful. This form takes no arguments, as in {\itshape text}. 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 \begin{itshape}...\end{itshape}, which we’ll describe further at the end of the section.

These commands, in any of the three forms, are cumulative; for instance you can get bold sans serif by saying either of \sffamily\bfseries or \bfseries\sffamily.

One advantage of these commands is that they automatically insert italic corrections if needed (see \/). Specifically, they insert the italic correction unless the following character is in the list \nocorrlist, which by default consists of period and comma. To suppress the automatic insertion of italic correction, use \nocorr at the start or end of the command argument, such as \textit{\nocorr text} or \textsc{text \nocorr}.

\textrm (\rmfamily)

Roman.

\textit (\itshape)

Italics.

\textmd (\mdseries)

Medium weight (default).

\textbf (\bfseries)

Boldface.

\textup (\upshape)

Upright (default).

\textsl (\slshape)

Slanted.

\textsf (\sffamily)

Sans serif.

\textsc (\scshape)

Small caps.

\texttt (\ttfamily)

Typewriter.

\textnormal (\normalfont)

Main document font.

Although it also changes fonts, the \emph{text} command is semantic, for text to be emphasized, and should not be used as a substitute for \textit. For example, \emph{start text \emph{middle text} end text} will result in the start text and end text in italics, but middle text will be in roman.

LaTeX also provides the following commands, which unconditionally switch to the given style, that is, are not cumulative. They are used as declarations: {\cmd...} instead of \cmd{...}.

(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 what is needed.)

\bf

Switch to bold face.

\cal

Switch to calligraphic letters for math.

\it

Italics.

\rm

Roman.

\sc

Small caps.

\sf

Sans serif.

\sl

Slanted (oblique).

\tt

Typewriter (monospace, fixed-width).

The \em command is the unconditional version of \emph.

The following commands are for use in math mode. They are not cumulative, so \mathbf{\mathit{symbol}} does not create a boldface and italic symbol; instead, it will just be in italics. This is because typically math symbols need consistent typographic treatment, regardless of the surrounding environment.

\mathrm

Roman, for use in math mode.

\mathbf

Boldface, for use in math mode.

\mathsf

Sans serif, for use in math mode.

\mathtt

Typewriter, for use in math mode.

\mathit
(\mit)

Italics, for use in math mode.

\mathnormal

For use in math mode, e.g., inside another type style declaration.

\mathcal

Calligraphic letters, for use in math mode.

In addition, the command \mathversion{bold} can be used for switching to bold letters and symbols in formulas. \mathversion{normal} restores the default.

Finally, the command \oldstylenums{numerals} will typeset so-called “old-style” numerals, which have differing heights and depths (and sometimes widths) from the standard “lining” numerals, which all have the same height as uppercase letters. LaTeX’s default fonts support this, and will respect \textbf (but not other styles; there are no italic old-style numerals in Computer Modern). Many other fonts have old-style numerals also; sometimes package options are provided to make them the default. FAQ entry: https://www.texfaq.org/FAQ-osf.


4.3 Font sizes

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 ‘10pt’, ‘11pt’, and ‘12pt’ document size options, respectively (see Document class options).

Command10pt11pt12pt
\tiny566
\scriptsize788
\footnotesize8910
\small91010.95
\normalsize (default)1010.9512
\large121214.4
\Large14.414.417.28
\LARGE17.2817.2820.74
\huge20.7420.7424.88
\Huge24.8824.8824.88

The commands are listed here in declaration (not environment) form, since that is how they are typically used. For example.

\begin{quotation} \small
  The Tao that can be named is not the eternal Tao.
\end{quotation}

Here, the scope of the \small lasts until the end of the quotation environment. It would also end at the next type style command or the end of the current group, so you could enclose it in curly braces {\small This text is typeset in the small font.}.

Trying to use these commands in math, as with $\small mv^2/2$, results in ‘LaTeX Font Warning: Command \small invalid in math mode’, and the font size doesn’t change. To work with a too-large formula, often the best option is to use the displaymath environment (see Math formulas), or one of the environments from the amsmath package. For inline mathematics, such as in a table of formulas, an alternative is something like {\small $mv^2/2$}. (Sometimes \scriptsize and \scriptstyle are confused. Both change the font size, but the latter also changes a number of other aspects of how mathematics is typeset. See Math styles.)

An environment form of each of these commands is also defined; for instance, \begin{tiny}...\end{tiny}. However, in practice this form can easily lead to unwanted spaces at the beginning and/or end of the environment without careful consideration, so it’s generally less error-prone to stick to the declaration form.

(Aside: Technically, due to the way LaTeX defines \begin and \end, nearly every command that does not take an argument technically has an environment form. But in almost all cases, it would only cause confusion to use it. The reason for mentioning the environment form of the font size declarations specifically is that this particular use is not rare.)


4.4 Low-level font commands

These commands are primarily intended for writers of macros and packages. The commands listed here are only a subset of the available ones.

\fontencoding{encoding}

Select the font encoding, the encoding of the output font. There are a large number of valid encodings. The most common are OT1, Knuth’s original encoding for Computer Modern (the default), and T1, 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 https://ctan.org/pkg/encguide.

\fontfamily{family}

Select the font family. The web page https://tug.org/FontCatalogue/ provides one way to browse through many of the fonts easily used with LaTeX. Here are examples of some common families.

pagAvant Garde
fvsBitstream Vera Sans
pbkBookman
bchCharter
ccrComputer Concrete
cmrComputer Modern
cmssComputer Modern Sans Serif
cmttComputer Modern Typewriter
pcrCourier
phvHelvetica
fi4Inconsolata
lmrLatin Modern
lmssLatin Modern Sans
lmttLatin Modern Typewriter
pncNew Century Schoolbook
pplPalatino
ptmTimes
unclUncial
putUtopia
pzcZapf Chancery
\fontseries{series}

Select the font series. A series combines a weight and a width. Typically, a font supports only a few of the possible combinations. Some common combined series values include:

mMedium (normal)
bBold
cCondensed
bcBold condensed
bxBold extended

The possible values for weight, individually, are:

ulUltra light
elExtra light
lLight
slSemi light
mMedium (normal)
sbSemi bold
bBold
ebExtra bold
ubUltra bold

The possible values for width, individually, are (the meaning and relationship of these terms varies with individual typefaces):

ucUltra condensed
ecExtra condensed
cCondensed
scSemi condensed
mMedium
sxSemi expanded
xExpanded
exExtra expanded
uxUltra expanded

When forming the series string from the weight and width, drop the m that stands for medium weight or medium width, unless both weight and width are m, in which case use just one (‘m’).

\fontshape{shape}

Select font shape. Valid shapes are:

nUpright (normal)
itItalic
slSlanted (oblique)
scSmall caps
uiUpright italics
olOutline

The two last shapes are not available for most font families, and small caps are often missing as well.

\fontsize{size}{skip}

Set the font size and the line spacing. The unit of both parameters defaults to points (pt). The line spacing is the nominal vertical space between lines, baseline to baseline. It is stored in the parameter \baselineskip. The default \baselineskip for the Computer Modern typeface is 1.2 times the \fontsize. Changing \baselineskip directly is inadvisable since its value is reset every time a size change happens; instead use \baselinestretch. (see \baselineskip & \baselinestretch).

\linespread{factor}

Equivalent to \renewcommand{\baselinestretch}{factor}, and therefore must be followed by \selectfont to have any effect. Best specified in the preamble. See \baselineskip & \baselinestretch, for using setspace package instead.

\selectfont

The effects of the font commands described above do not happen until \selectfont is called, as in \fontfamily{familyname}\selectfont. It is often useful to put this in a macro:
\newcommand*{\myfont}{\fontfamily{familyname}\selectfont}
(see \newcommand & \renewcommand).

\usefont{enc}{family}{series}{shape}

The same as invoking \fontencoding, \fontfamily, \fontseries and \fontshape with the given parameters, followed by \selectfont. For example:

\usefont{ot1}{cmr}{m}{n}

5 Layout

Commands for controlling the general page layout.


5.1 \onecolumn

Synopsis:

\onecolumn

Start a new page and produce single-column output. If the document is given the class option onecolumn then this is the default behavior (see Document class options). This command is fragile (see \protect).


5.2 \twocolumn

Synopses:

\twocolumn
\twocolumn[prelim one column text]

Start a new page and produce two-column output. If the document is given the class option twocolumn then this is the default (see Document class options). This command is fragile (see \protect).

If the optional prelim one column text argument is present, it is typeset in one-column mode before the two-column typesetting starts.

These parameters control typesetting in two-column output:

\columnsep

The distance between columns. The default is 35pt. Change it with a command such as \setlength{\columnsep}{40pt}. You must change it before the two column mode starts; in the preamble is a good place.

\columnseprule

The width of the rule between columns. The default is 0pt, meaning that there is no rule. Otherwise, the rule appears halfway between the two columns. Change it with a command such as \setlength{\columnseprule}{0.4pt}, before the two-column mode starts.

\columnwidth

The width of a single column. In one-column mode this is equal to \textwidth. In two-column mode by default LaTeX sets the width of each of the two columns, \columnwidth, to be half of \textwidth minus \columnsep.

In a two-column document, the starred environments table* and figure* are two columns wide, whereas the unstarred environments table and figure take up only one column (see figure and see table). LaTeX places starred floats at the top of a page. The following parameters control float behavior of two-column output.

\dbltopfraction

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 table* or figure* environment must not exceed 0.7\textheight. 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:

  • Use the [tp] 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.
  • Use the [t!] location specifier to override the effect of \dbltopfraction for this particular float.
  • Increase the value of \dbltopfraction to a suitably large number, to avoid going to float pages so soon.

You can redefine it, as with \renewcommand{\dbltopfraction}{0.9}.

\dblfloatpagefraction

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’s default is 0.5. Change it with \renewcommand.

\dblfloatsep

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 12pt plus2pt minus2pt for a document set at 10pt or 11pt, and 14pt plus2pt minus4pt for a document set at 12pt.

\dbltextfloatsep

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 20pt plus2pt minus4pt.

\dbltopnumber

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 2.

This example uses \twocolumn’s optional argument of to create a title that spans the two-column article:

\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  &\authormark{2}Department two \\ 
     School one                   &School two 
  \end{tabular}\\[3em] % space below title part
  }]

Two column text here.

5.3 \flushbottom

Make all pages in the document 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.

If TeX cannot satisfactorily stretch the vertical space in a page then you get a message like ‘Underfull \vbox (badness 10000) has occurred while \output is active’. If you get that, one option is to change to \raggedbottom (see \raggedbottom). Alternatively, you can adjust the textheight to make compatible pages, or you can add some vertical stretch glue between lines or between paragraphs, as in \setlength{\parskip}{0ex plus0.1ex}. Your last option is to, in a final editing stage, adjust the height of individual pages (see \enlargethispage).

The \flushbottom state is the default only if you select the twocolumn document class option (see Document class options), and for indexes made using makeidx.


5.4 \raggedbottom

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 \flushbottom.

This is the default unless you select the twocolumn document class option (see Document class options).


5.5 Page layout parameters

\columnsep
\columnseprule
\columnwidth

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 twocolumn is in effect (see Document class options). See \twocolumn.

\headheight

Height of the box that contains the running head. The default in the article, report, and book classes is ‘12pt’, at all type sizes.

\headsep

Vertical distance between the bottom of the header line and the top of the main text. The default in the article and report classes is ‘25pt’. In the book class the default is: if the document is set at 10pt then it is ‘0.25in’, and at 11pt or 12pt it is ‘0.275in’.

\footskip

Distance from the baseline of the last line of text to the baseline of the page footer. The default in the article and report classes is ‘30pt’. In the book class the default is: when the type size is 10pt the default is ‘0.35in’, while at 11pt it is ‘0.38in’, and at 12pt it is ‘30pt’.

\linewidth

Width of the current line, decreased for each nested list (see list). That is, the nominal value for \linewidth is to equal \textwidth but for each nested list the \linewidth is decreased by the sum of that list’s \leftmargin and \rightmargin (see itemize).

\marginparpush
\marginsep
\marginparwidth

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.

Normally marginal notes appear on the outside of the page, but the declaration \reversemarginpar changes that (and \normalmarginpar changes it back).

The defaults for \marginparpush in both book and article classes are: ‘7pt’ if the document is set at 12pt, and ‘5pt’ if the document is set at 11pt or 10pt.

For \marginsep, in article class the default is ‘10pt’ except if the document is set at 10pt and in two-column mode where the default is ‘11pt’.

For \marginsep in book class the default is ‘10pt’ in two-column mode and ‘7pt’ in one-column mode.

For \marginparwidth in both book and article classes, in two-column mode the default is 60% of \paperwidth − \textwidth, while in one-column mode it is 50% of that distance.

\oddsidemargin
\evensidemargin

The \oddsidemargin length is the extra distance between the left side of the page and the text’s left margin, on odd-numbered pages when the document class option twoside is chosen and on all pages when oneside is in effect. When twoside is in effect, on even-numbered pages the extra distance on the left is \evensidemargin.

LaTeX’s default is that \oddsidemargin is 40% of the difference between \paperwidth and \textwidth, and \evensidemargin is the remainder.

\paperheight

The height of the paper, as distinct from the height of the print area. Normally set with a document class option, as in \documentclass[a4paper]{article} (see Document class options).

\paperwidth

The width of the paper, as distinct from the width of the print area. Normally set with a document class option, as in \documentclass[a4paper]{article} (see Document class options).

\textheight

The normal vertical height of the page body. If the document is set at a nominal type size of 10pt then for an article or report the default is ‘43\baselineskip’, while for a book it is ‘41\baselineskip’. At a type size of 11pt the default is ‘38\baselineskip’ for all document classes. At 12pt it is ‘36\baselineskip’ for all classes.

\textwidth

The full horizontal width of the entire page body. For an article or report document, the default is ‘345pt’ when the chosen type size is 10pt, the default is ‘360pt’ at 11pt, and it is ‘390pt’ at 12pt. For a book document, the default is ‘4.5in’ at a type size of 10pt, and ‘5in’ at 11pt or 12pt.

In multi-column output, \textwidth remains the width of the entire page body, while \columnwidth is the width of one column (see \twocolumn).

In lists (see list), \textwidth remains the width of the entire page body (and \columnwidth the width of the entire column), while \linewidth may decrease for nested lists.

Inside a minipage (see minipage) or \parbox (see \parbox), all the width-related parameters are set to the specified width, and revert to their normal values at the end of the minipage or \parbox.

\hsize

This entry is included for completeness: \hsize is the TeX primitive parameter used when text is broken into lines. It should not be used in normal LaTeX documents.

\topmargin

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: \paperheight − 2in − \headheight − \headsep − \textheight − \footskip, and then divided by two.

\topskip

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., ‘10pt’ at a type size of 10pt.


5.6 \baselineskip & \baselinestretch

The \baselineskip is a rubber length (see Lengths). It gives the leading, the normal distance between lines in a paragraph, from baseline to baseline.

Ordinarily document authors do not directly change \baselineskip while writing. Instead, it is set by the low level font selection command \fontsize (see low level font commands fontsize). The \baselineskip’s value is reset every time a font change happens and so any direct change to \baselineskip would vanish the next time there was a font switch. For how to influence line spacing, see the discussion of \baselinestretch below.

Usually, a font’s size and baseline skip is assigned by the font designer. These numbers are nominal in the sense that if, for instance, a font’s style file has the command \fontsize{10pt}{12pt} then that does not mean that the characters in the font are 10pt tall; for instance, parentheses and accented capitals may be taller. Nor does it mean that if the lines are spaced less than 12pt apart then they risk touching. Rather these numbers are typographic judgements. (Often, the \baselineskip is about twenty percent larger than the font size.)

The \baselineskip is not a property of each line but of the entire paragraph. As a result, large text in the middle of a paragraph, such as a single {\Huge Q}, will be squashed into its line. TeX will make sure it doesn’t scrape up against the line above but won’t change the \baselineskip for that one line to make extra room above. For the fix, use a \strut (see \strut).

The value of \baselineskip that TeX uses for the paragraph is the value in effect at the blank line or command that ends the paragraph unit. So if a document contains this paragraph then its lines will be scrunched together, compared to lines in surrounding paragraphs.

Many people see a page break between text and a displayed equation as 
bad style, so in effect the display is part of the paragraph.
Because this display is in footnotesize, the entire paragraph has the
baseline spacing matching that size.
{\footnotesize $$a+b = c$$}

The process for making paragraphs is that when a new line is added, if the depth of the previous line plus the height of the new line is less than \baselineskip then TeX inserts vertical glue to make up the difference. There are two fine points. The first is that if the lines would be too close together, closer than \lineskiplimit, then TeX instead uses \lineskip as the interline glue. The second is that TeX doesn’t actually use the depth of the previous line. Instead it uses \prevdepth, which usually contains that depth. But at the beginning of the paragraph (or any vertical list) or just after a rule, \prevdepth has the value -1000pt and this special value tells TeX not to insert any interline glue at the paragraph start.

In the standard classes \lineskiplimit is 0pt and \lineskip is 1pt. By the prior paragraph then, the distance between lines can approach zero but if it becomes zero (or less than zero) then the lines jump to 1pt apart.

Sometimes authors must, for editing purposes, put the document in double space or one-and-a-half space. The right way to influence the interline distance is via \baselinestretch. It scales \baselineskip, and has a default value of 1.0. It is a command, not a length, and does not take effect until a font change happens, so set the scale factor like this: \renewcommand{\baselinestretch}{1.5}\selectfont.

The most straightforward way to change the line spacing for an entire document is to put \linespread{factor} in the preamble. For double spacing, take factor to be 1.6 and for one-and-a-half spacing use 1.3. These numbers are rough: for instance, since the \baselineskip is about 1.2 times the font size, multiplying by 1.6 gives a baseline skip to font size ratio of about 2. (The \linespread command is defined as \renewcommand{\baselinestretch}{factor} so it also won’t take effect until a font setting happens. But that always takes place at the start of a document, so there you don’t need to follow it with \selectfont.)

A simpler approach is the setspace package. The basic example:

\usepackage{setspace}
\doublespacing  % or \onehalfspacing for 1.5

In the preamble these will start the document off with that sizing. But you can also use these declarations in the document body to change the spacing from that point forward, and consequently there is \singlespacing to return the spacing to normal. In the document body, a better practice than using the declarations is to use environments, such as \begin{doublespace} ... \end{doublespace}. The package also has commands to do arbitrary spacing: \setstretch{factor} and \begin{spacing}{factor} ... \end{spacing}. This package also keeps the line spacing single-spaced in places where that is typically desirable, such as footnotes and figure captions. See the package documentation.


5.7 Floats

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.

LaTeX can have a number of different classes of floating material. The default is the two classes, figure (see figure) and table (see table), but you can create a new class with the package float.

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.

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.

In addition to changing the parameters, for each float you can tweak where the float placement algorithm tries to place it by using its placement argument. The possible values are a sequence of the letters below. The default for both figure and table, in both article and book classes, is tbp.

t

(Top)—at the top of a text page.

b

(Bottom)—at the bottom of a text page. (However, b is not allowed for full-width floats (figure*) with double-column output. To ameliorate this, use the stfloats or dblfloatfix package, but see the discussion at caveats in the FAQ: https://www.texfaq.org/FAQ-2colfloat.

h

(Here)—at the position in the text where the figure environment appears. However, h is not allowed by itself; t is automatically added.

To absolutely force a float to appear “here”, you can \usepackage{float} and use the H specifier which it defines. For further discussion, see the FAQ entry at https://www.texfaq.org/FAQ-figurehere.

p

(Page of floats)—on a separate float page, which is a page containing no text, only floats.

!

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 ! specifier does not mean “put the float here”; see above.

Note: the order in which letters appear in the placement argument does not change the order in which LaTeX tries to place the float; for instance, btp has the same effect as tbp. All that placement does is that if a letter is not present then the algorithm does not try that location. Thus, LaTeX’s default of tbp is to try every location except placing the float where it occurs in the source.

To prevent LaTeX from moving floats to the end of the document or a chapter you can use a \clearpage command to start a new page and insert all pending floats. If a pagebreak is undesirable then you can use the afterpage package and issue \afterpage{\clearpage}. This will wait until the current page is finished and then flush all outstanding floats.

LaTeX can typeset a float before where it appears in the source (although on the same output page) if there is a t specifier in the placement parameter. If this is not desired, and deleting the t 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 flafter package or using the command \suppressfloats[t], which causes floats for the top position on this page to moved to the next page.

Parameters relating to fractions of pages occupied by float and non-float text (change them with \renewcommand{parameter}{decimal between 0 and 1}):

\bottomfraction

The maximum fraction of the page allowed to be occupied by floats at the bottom; default ‘.3’.

\floatpagefraction

The minimum fraction of a float page that must be occupied by floats; default ‘.5’.

\textfraction

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 ‘.2’.

\topfraction

Maximum fraction at the top of a page that may be occupied before floats; default ‘.7’.

Parameters relating to vertical space around floats (change them with a command of the form \setlength{parameter}{length expression}):

\floatsep

Space between floats at the top or bottom of a page; default ‘12pt plus2pt minus2pt’.

\intextsep

Space above and below a float in the middle of the main text; default ‘12pt plus2pt minus2pt’ for 10 point and 11 point documents, and ‘14pt plus4pt minus4pt’ for 12 point documents.

\textfloatsep

Space between the last (first) float at the top (bottom) of a page; default ‘20pt plus2pt minus4pt’.

Counters relating to the number of floats on a page (change them with a command of the form \setcounter{ctrname}{natural number}):

bottomnumber

Maximum number of floats that can appear at the bottom of a text page; default 1.

dbltopnumber

Maximum number of full-sized floats that can appear at the top of a two-column page; default 2.

topnumber

Maximum number of floats that can appear at the top of a text page; default 2.

totalnumber

Maximum number of floats that can appear on a text page; default 3.

The principal TeX FAQ entry relating to floats https://www.texfaq.org/FAQ-floats contains suggestions for relaxing LaTeX’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’s article “How to influence the position of float environments like figure and table in LaTeX?” (https://www.latex-project.org/publications/2014-FMi-TUB-tb111mitt-float-placement.pdf).


5.7.1 \caption

Synopsis:

\caption{caption-text}

or

\caption[short-caption-text]{caption-text}

Make a caption for a floating environment, such as a figure or table environment (see figure or table).

In this example, LaTeX places a caption below the vertical blank space that is left by the author for the later inclusion of a picture.

\begin{figure}
  \vspace*{1cm}
  \caption{Alonzo Cushing, Battery A, 4th US Artillery.}
  \label{fig:CushingPic}
\end{figure}

The \caption command will label the caption-text with something like ‘Figure 1:’ for an article or ‘Figure 1.1:’ for a book. The text is centered if it is shorter than the text width, or set as an unindented paragraph if it takes more than one line.

In addition to placing the caption-text in the output, the \caption command also saves that information for use in a list of figures or list of tables (see Table of contents, list of figures, list of tables).

Here the \caption command uses the optional short-caption-text, so that the shorter text appears in the list of tables, rather than the longer caption-text.

\begin{table}
  \centering
  \begin{tabular}{|*{3}{c}|}
    \hline
    4  &9  &2 \\
    3  &5  &7 \\ 
    8  &1  &6 \\
    \hline
  \end{tabular}
  \caption[\textit{Lo Shu} magic square]{%
    The \textit{Lo Shu} magic square, which is unique among
    squares of order three up to rotation and reflection.}
  \label{tab:LoShu}
\end{table}

LaTeX will label the caption-text with something like ‘Table 1:’ for an article or ‘Table 1.1:’ for a book.

The caption can appear at the top of the figure or table. For instance, that would happen in the prior example by putting the \caption between the \centering and the \begin{tabular}.

Different floating environments are numbered separately, by default. It is \caption that updates the counter, and so any \label must come after the \caption. The counter for the figure environment is named figure, and similarly the counter for the table environment is table.

The text that will be put in the list of figures or list of tables is moving argument. If you get the LaTeX error ‘! Argument of \@caption has an extra }’ then you must put \protect in front of any fragile commands. See \protect.

The caption package has many options to adjust how the caption appears, for example changing the font size, making the caption be hanging text rather than set as a paragraph, or making the caption always set as a paragraph rather than centered when it is short.


6 Sectioning

Structure your text into divisions: parts, chapters, sections, etc. All sectioning commands have the same form, one of:

sectioning-command{title}
sectioning-command*{title}
sectioning-command[toc-title]{title}

For instance, declare the start of a subsection as with \subsection{Motivation}.

The table has each sectioning-command in LaTeX. All are available in all of LaTeX’s standard document classes book, report, and article, except that \chapter is not available in article.

Sectioning unitCommandLevel
Part\part-1 (book, report), 0 (article)
Chapter\chapter0
Section\section1
Subsection\subsection2
Subsubsection\subsubsection3
Paragraph\paragraph4
Subparagraph\subparagraph5

All these commands have a *-form that prints title as usual but does not number it and does not make an entry in the table of contents. An example of using this is for an appendix in an article. The input \appendix\section{Appendix} gives the output ‘A Appendix’ (see \appendix). You can lose the numbering ‘A’ by instead entering \section*{Appendix} (articles often omit a table of contents and have simple page headers so the other differences from the \section command may not matter).

The section title title 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 Page styles). You may not want the same text in these places as in the main text. All of these commands have an optional argument toc-title for these other places.

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’s level is less than or equal to the value of the counter secnumdepth then the titles for this sectioning command will be numbered (see Sectioning/secnumdepth). And, if level is less than or equal to the value of the counter tocdepth then the table of contents will have an entry for this sectioning unit (see Sectioning/tocdepth).

LaTeX expects that before you have a \subsection you will have a \section and, in a book class document, that before a \section you will have a \chapter. Otherwise you can get something like a subsection numbered ‘3.0.1’.

LaTeX lets you change the appearance of the sectional units. As a simple example, you can change the section numbering to uppercase letters with this (in the preamble):
\renewcommand\thesection{\Alph{section}} . (See \alph \Alph \arabic \roman \Roman \fnsymbol: Printing counters.) CTAN has many packages that make this adjustment easier, notably titlesec.

Two counters relate to the appearance of headings made by sectioning commands.

secnumdepth

Controls which sectioning unit are numbered. Setting the counter with \setcounter{secnumdepth}{level} will suppress numbering of sectioning at any depth greater than level (see \setcounter). See the above table for the level numbers. For instance, if the secnumdepth is 1 in an article then a \section{Introduction} command will produce output like ‘1 Introduction’ while \subsection{Discussion} will produce output like ‘Discussion’, without the number. LaTeX’s default secnumdepth is 3 in article class and  2 in the book and report classes.

tocdepth

Controls which sectioning units are listed in the table of contents. The setting \setcounter{tocdepth}{level} makes the sectioning units at level be the smallest ones listed (see \setcounter). See the above table for the level numbers. For instance, if tocdepth is 1 then the table of contents will list sections but not subsections. LaTeX’s default tocdepth is 3 in article class and  2 in the book and report classes.


6.1 \part

Synopsis, one of:

\part{title}
\part*{title}
\part[toc-title]{title}

Start a document part. The standard LaTeX classes book, report, and article, all have this command.

This produces a document part, in a book.

\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.

In each standard class the \part command outputs a part number such as ‘Part I’, alone on its line, in boldface, and in large type. Then LaTeX outputs title, also alone on its line, in bold and in even larger type. In class book, 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 report it is again alone on a page, but LaTeX won’t force it onto an odd-numbered page. In an article LaTeX does not put it on a fresh page, but instead outputs the part number and part title onto the main document page.

The * form shows title but it does not show the part number, does not increment the part counter, and produces no table of contents entry.

The optional argument toc-title will appear as the part title in the table of contents (see Table of contents, list of figures, list of tables) and in running headers (see Page styles). If it is not present then title will be there. This example puts a line break in title but omits the break in the table of contents.

\part[Up from the bottom; my life]{Up from the bottom\\ my life}

For determining which sectional units are numbered and which appear in the table of contents, the level number of a part is -1 (see Sectioning/secnumdepth, and Sectioning/tocdepth).

In the class article, if a paragraph immediately follows the part title then it is not indented. To get an indent you can use the package indentfirst.

One package to change the behavior of \part is titlesec. See its documentation on CTAN.


6.2 \chapter

Synopsis, one of:

\chapter{title}
\chapter*{title}
\chapter[toc-title]{title}

Start a chapter. The standard LaTeX classes book and report have this command but article does not.

This produces a chapter.

\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.

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 ‘Chapter 1’ in large boldface type (the size is \huge). It then puts title on a fresh line, in boldface type that is still larger (size \Huge). It also increments the chapter counter, adds an entry to the table of contents (see Table of contents, list of figures, list of tables), and sets the running header information (see Page styles).

The * form shows title on a fresh line, in boldface. But it does not show the chapter number, does not increment the chapter counter, produces no table of contents entry, and does not affect the running header. (If you use the page style headings in a two-sided document then the header will be from the prior chapter.) This example illustrates.

\chapter*{Preamble}

The optional argument toc-title will appear as the chapter title in the table of contents (see Table of contents, list of figures, list of tables) and in running headers (see Page styles). If it is not present then title will be there. This shows the full name in the chapter title,

\chapter[Weyl]{Hermann Klaus Hugo (Peter) Weyl (1885--1955)}

but only ‘Weyl’ on the contents page. This puts a line break in the title but that doesn’t work well with running headers so it omits the break in the contents

\chapter[Given it all; my story]{Given it all\\ my story}

For determining which sectional units are numbered and which appear in the table of contents, the level number of a chapter is 0 (see Sectioning/secnumdepth and see Sectioning/tocdepth).

The paragraph that follows the chapter title is not indented, as is a standard typographical practice. To get an indent use the package indentfirst.

You can change what is shown for the chapter number. To change it to something like ‘Lecture 1’, put in the preamble either \renewcommand{\chaptername}{Lecture} or this (see \makeatletter & \makeatother).

\makeatletter
\renewcommand{\@chapapp}{Lecture}
\makeatother

To make this change because of the primary language for the document, see the package babel.

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 \clearpage & \cleardoublepage.

To change the behavior of the \chapter 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 titlesec. See its documentation, but the example below gives a sense of what it can do.

\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

This omits the chapter number ‘Chapter 1’ from the page but unlike \chapter* it keeps the chapter in the table of contents and the running headers.


6.3 \section

Synopsis, one of:

\section{title}
\section*{title}
\section[toc-title]{title}

Start a section. The standard LaTeX classes article, book, and report all have this command.

This produces a section.

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.

For the standard LaTeX classes book and report the default output is like ‘1.2 title’ (for chapter 1, section 2), alone on its line and flush left, in boldface and a larger type (the type size is \Large). The same holds in article except that there are no chapters in that class so it looks like ‘2 title’.

The * form shows title. But it does not show the section number, does not increment the section counter, produces no table of contents entry, and does not affect the running header. (If you use the page style headings in a two-sided document then the header will be from the prior section.)

The optional argument toc-title will appear as the section title in the table of contents (see Table of contents, list of figures, list of tables) and in running headers (see Page styles). If it is not present then title will be there. This shows the full name in the title of the section:

\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.}

but only ‘Elizabeth II’ on the contents page and in the headers. This has a line break in title but that does not work with headers so it is omitted from the contents and headers.

\section[Truth is, I cheated; my life story]{Truth is,
  I cheated\\my life story}

For determining which sectional units are numbered and which appear in the table of contents, the level number of a section is 1 (see Sectioning/secnumdepth and see Sectioning/tocdepth).

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 indentfirst.

In general, to change the behavior of the \section command, there are a number of options. One is the \@startsection command (see \@startsection: Typesetting sectional unit headings). There are also many packages on CTAN that address this, including titlesec. See the documentation but the example below gives a sense of what they can do.

\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]

That puts the section number in the margin.


6.4 \subsection

Synopsis, one of:

\subsection{title}
\subsection*{title}
\subsection[toc-title]{title}

Start a subsection. The standard LaTeX classes article, book, and report all have this command.

This produces a subsection.

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.

For the standard LaTeX classes book and report the default output is like ‘1.2.3 title’ (for chapter 1, section 2, subsection 3), alone on its line and flush left, in boldface and a larger type (the type size is \large). The same holds in article except that there are no chapters in that class so it looks like ‘2.3 title’.

The * form shows title. But it does not show the subsection number, does not increment the subsection counter, and produces no table of contents entry.

The optional argument toc-title will appear as the subsection title in the table of contents (see Table of contents, list of figures, list of tables). If it is not present then title will be there. This shows the full text in the title of the subsection:

\subsection[$\alpha,\beta,\gamma$ paper]{\textit{The Origin of
  Chemical Elements} by R.A.~Alpher, H.~Bethe, and G.~Gamow}

but only ‘α,β,γ paper’ on the contents page.

For determining which sectional units are numbered and which appear in the table of contents, the level number of a subsection is 2 (see Sectioning/secnumdepth and see Sectioning/tocdepth).

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 indentfirst.

There are a number of ways to change the behavior of the \subsection command. One is the \@startsection command (see \@startsection: Typesetting sectional unit headings). There are also many packages on CTAN that address this, including titlesec. See the documentation but the example below gives a sense of what they can do.

\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

That puts the subsection number and title in the first line of text.


6.5 \subsubsection, \paragraph, \subparagraph

Synopsis, one of:

\subsubsection{title}
\subsubsection*{title}
\subsubsection[toc-title]{title}

or one of:

\paragraph{title}
\paragraph*{title}
\paragraph[toc-title]{title}

or one of:

\subparagraph{title}
\subparagraph*{title}
\subparagraph[toc-title]{title}

Start a subsubsection, paragraph, or subparagraph. The standard LaTeX classes article, book, and report all have these commands, although they are not commonly used.

This produces a subsubsection.

\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.

The default output of each of the three does not change over the standard LaTeX classes article, book, and report. For \subsubsection the title is alone on its line, in boldface and normal size type. For \paragraph the title is inline with the text, not indented, in boldface and normal size type. For \subparagraph the title is inline with the text, with a paragraph indent, in boldface and normal size type (Because an article has no chapters its subsubsections are numbered and so it looks like ‘1.2.3 title’, for section 1, subsection 2, and subsubsection 3. The other two divisions are not numbered.)

The * form shows title. But it does not increment the associated counter and produces no table of contents entry (and does not show the number for \subsubsection).

The optional argument toc-title will appear as the division title in the table of contents (see Table of contents, list of figures, list of tables). If it is not present then title will be there.

For determining which sectional units are numbered and which appear in the table of contents, the level number of a subsubsection is 3, of a paragraph is 4, and of a subparagraph is 5 (see Sectioning/secnumdepth and see Sectioning/tocdepth).

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 indentfirst.

There are a number of ways to change the behavior of the these commands. One is the \@startsection command (see \@startsection: Typesetting sectional unit headings). There are also many packages on CTAN that address this, including titlesec. See the documentation on CTAN.


6.6 \appendix

Synopsis:

\appendix

This does not directly produce any output. But in a book or report document it declares that subsequent \chapter commands start an appendix. In an article it does the same, for \section commands. It also resets the chapter and section counters to 0 in a book or report, and in an article resets the section and subsection counters.

In this book

\chapter{One}  ...
\chapter{Two}  ...
 ...
\appendix
\chapter{Three}  ...
\chapter{Four}  ...

the first two will generate output numbered ‘Chapter 1’ and ‘Chapter 2’. After the \appendix the numbering will be ‘Appendix A’ and ‘Appendix B’. See Larger book template, for another example.

The appendix package adds the command \appendixpage to put a separate ‘Appendices’ in the document body before the first appendix, and the command \addappheadtotoc to do the same in the table of contents. You can reset the name ‘Appendices’ with a command like \renewcommand{\appendixname}{Specification}, as well as a number of other features. See the documentation on CTAN.


6.7 \frontmatter, \mainmatter, \backmatter

Synopsis, one or more of:

\frontmatter
...
\mainmatter
...
\backmatter
...

Format a book class document differently according to which part of the document is being produced. All three commands are optional.

Traditionally, a book’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.

The \frontmatter command makes the pages numbered in lowercase roman, and makes chapters not numbered, although each chapter’s title appears in the table of contents; if you use other sectioning commands here, use the *-version (see Sectioning).

The \mainmatter command changes the behavior back to the expected version, and resets the page number.

The \backmatter command leaves the page numbering alone but switches the chapters back to being not numbered.

See Larger book template, for an example using these three commands.


6.8 \@startsection: Typesetting sectional unit headings

Synopsis:

\@startsection{name}{level}{indent}{beforeskip}{afterskip}{style}

Used to help redefine the behavior of commands that start sectioning divisions such as \section or \subsection.

The titlesec package makes manipulation of sectioning easier. Further, while most requirements for sectioning commands can be satisfied with \@startsection, some cannot. For instance, in the standard LaTeX book and report classes the commands \chapter and \report are not constructed using this. To make such a command you may want to use the \secdef command.

The \@startsection macro is used like this:

\@startsection{name}
  {level}
  {indent}
  {beforeskip}
  {afterskip}
  {style}*[toctitle]{title}

so that issuing

\renewcommand{\section}{\@startsection{name}
  {level}
  {indent}
  {beforeskip}
  {afterskip}
  {style}}

redefines \section while keeping its standard calling form \section*[toctitle]{title} (in which, as a reminder, the star * is optional). See Sectioning. This implies that when you write a command like \renewcommand{\section}{...}, the \@startsection{...} must come last in the definition. See the examples below.

name

Name of the counter used to number the sectioning header. This counter must be defined separately. Most commonly this is either section, subsection, or paragraph. Although in those cases the counter name is the same as the sectioning command itself, you don’t have to use the same name.

Then \thename displays the title number and \namemark is for the page headers. See the third example below.

level

An integer giving the depth of the sectioning command. See Sectioning, for the list of standard level numbers.

If level is less than or equal to the value of the counter secnumdepth then titles for this sectioning command will be numbered (see Sectioning/secnumdepth). For instance, if secnumdepth is 1 in an article then the command \section{Introduction} will produce output like “1 Introduction” while \subsection{Discussion} will produce output like “Discussion”, without the number prefix.

If level is less than or equal to the value of the counter tocdepth then the table of contents will have an entry for this sectioning unit (see Sectioning/tocdepth). For instance, in an article, if tocdepth is 1 then the table of contents will list sections but not subsections.

indent

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 0pt. A negative indentation such as -\parindent will move the title into the left margin.

beforeskip

The absolute value of this length is the amount of vertical space that is inserted before this sectioning unit’s title. This space will be discarded if the sectioning unit happens to start at the beginning of a 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. (Example: the negative of 1pt plus 2pt minus 3pt is -1pt plus -2pt minus -3pt.)

For example, if beforeskip is -3.5ex plus -1ex minus -0.2ex 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 plus and minus, is good practice here since it gives LaTeX more flexibility in making up the page (see Lengths).

The full accounting of the vertical space between the baseline of the line prior to this sectioning unit’s header and the baseline of the header is that it is the sum of the \parskip of the text font, the \baselineskip of the title font, and the absolute value of the beforeskip. 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.)

afterskip

This is a length. If afterskip is non-negative then this is the vertical space inserted after the sectioning unit’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 1pt plus 2pt minus 3pt is -1pt plus -2pt minus -3pt.)

As with beforeskip, using a rubber length, with plus and minus components, is good practice here since it gives LaTeX more flexibility in putting together the page.

If afterskip is non-negative then the full accounting of the vertical space between the baseline of the sectioning unit’s header and the baseline of the first line of the following paragraph is that it is the sum of the \parskip of the title font, the \baselineskip of the text font, and the value of after. That space is typically rubber so it may stretch or shrink. (Note that because the sign of afterskip changes the sectioning unit header’s from standalone to run-in, you cannot use a negative afterskip to cancel part of the \parskip.)

style

Controls the styling of the title. See the examples below. Typical commands to use here are \centering, \raggedright, \normalfont, \hrule, or \newpage. The last command in style may be one that takes one argument, such as \MakeUppercase or \fbox that takes one argument. The section title will be supplied as the argument to this command. For instance, setting style to \bfseries\MakeUppercase would produce titles that are bold and uppercase.

These are LaTeX’s defaults for the first three sectioning units that are defined with \@startsection, for the article, book, and report classes.

  • For section: level is 1, indent is 0pt, beforeskip is -3.5ex plus -1ex minus -0.2ex, afterskip is 2.3ex plus 0.2ex, and style is \normalfont\Large\bfseries.
  • For subsection: level is 2, indent is 0pt, beforeskip is -3.25ex plus -1ex minus -0.2ex, afterskip is 1.5ex plus 0.2ex, and style is \normalfont\large\bfseries.
  • For subsubsection: level is 3, indent is 0pt, beforeskip is -3.25ex plus -1ex minus -0.2ex, afterskip is 1.5ex plus 0.2ex, and style is \normalfont\normalsize\bfseries.

Some examples follow. These 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 \makeatletter command and a \makeatother. (Probably the error message You can't use `\spacefactor' in vertical mode. means that you forgot this.) See \makeatletter & \makeatother.

This will put section titles in large boldface type, centered. It says \renewcommand because LaTeX’s standard classes have already defined a \section. For the same reason it does not define a section counter, or the commands \thesection and \l@section.

\renewcommand\section{%
  \@startsection{section}% name
    {1}% level
    {0pt}% indent
    {-3.5ex plus -1ex minus -.2ex}% beforeskip
    {2.3ex plus.2ex}% afterskip
    {\centering\normalfont\Large\bfseries}% style
  }

This will put subsection titles in small caps type, inline with the paragraph.

\renewcommand\subsection{%
  \@startsection{subsection}%  name
    {2}% level
    {0em}% indent
    {-1ex plus 0.1ex minus -0.05ex}% beforeskip
    {-1em plus 0.2em}% afterskip
    {\scshape}% style
  }

The prior examples redefined existing sectional unit title commands. This defines a new one, illustrating the needed counter and macros to display that counter.

\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

7 Cross references

We often want something like ‘See Theorem~31’. But by-hand typing the 31 is poor practice. Instead you should write a label such as \label{eq:GreensThm} and then reference it, as with See equation~\ref{eq:GreensThm}. LaTeX will automatically work out the number, put it into the output, and will change that number later if needed.

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}.

LaTeX tracks cross reference information in a file having the extension .aux and with the same base name as the file containing the \label. So if \label is in calculus.tex then the information is in calculus.aux. LaTeX puts the information in that file every time it runs across a \label.

The behavior described in the prior paragraph results in a quirk that happens when your document has a forward reference, a \ref that appears before the associated \label. If this is the first time that you are compiling the document then you will get ‘LaTeX Warning: Label(s) may have changed. Rerun to get cross references right’ and in the output the forward reference will appear as two question marks ‘??’, 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.

The cleveref package enhances LaTeX’s cross referencing features. You can arrange that if you enter \begin{thm}\label{th:Nerode}...\end{thm} then \cref{th:Nerode} will output ‘Theorem 3.21’, without you having to enter the “Theorem.”


7.1 \label

Synopsis:

\label{key}

Assign a reference number to key. In ordinary text \label{key} assigns to key the number of the current sectional unit. Inside an environment with numbering, such as a table or theorem environment, \label{key} assigns to key the number of that environment. Retrieve the assigned number with the \ref{key} command (see \ref).

A key name can consist of any sequence of letters, digits, or common punctuation characters. Upper and lowercase letters are distinguished, as usual.

A common convention is to use labels consisting of a prefix and a suffix separated by a colon or period. Thus, \label{fig:Post} 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:

ch

for chapters

sec
subsec

for lower-level sectioning commands

fig

for figures

tab

for tables

eq

for equations

In the auxiliary file the reference information is kept as the text of a command of the form \newlabel{label}{{currentlabel}{pagenumber}}. Here currentlabel is the current value of the macro \@currentlabel that is usually updated whenever you call \refstepcounter{counter}.

Below, the key sec:test will get the number of the current section and the key fig:test will get the number of the figure. (Incidentally, put labels after captions in figures and tables.)

\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}.

7.2 \pageref

Synopsis:

\pageref{key}

Produce the page number of the place in the text where the corresponding \label{key} command appears.

If there is no \label{key} then you get something like ‘LaTeX Warning: Reference `th:GrensThm' on page 1 undefined on input line 11.

Below, the \label{eq:main} 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.)

The main result is formula~\ref{eq:main} on page~\pageref{eq:main}.
  ...
\begin{equation} \label{eq:main}
   \mathbf{P}=\mathbf{NP}
\end{equation}

7.3 \ref

Synopsis:

\ref{key}

Produces the number of the sectional unit, equation, footnote, figure, …, of the corresponding \label command (see \label). It does not produce any text, such as the word ‘Section’ or ‘Figure’, just the bare number itself.

If there is no \label{key} then you get something like ‘LaTeX Warning: Reference `th:GrensThm' on page 1 undefined on input line 11.

In this example the \ref{popular} produces ‘2’. Note that it is a forward reference since it comes before \label{popular} so this document would have to be compiled twice.

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}

The cleveref package includes text such as ‘Theorem’ in the reference. See the documentation on CTAN.


7.4 xr package

Synopsis:

\usepackage{xr}
  \externaldocument{document-basename}

or

\usepackage{xr}
  \externaldocument[reference-prefix]{document-basename}

Make cross references to the external document document-basename.tex.

Here is an example. If lectures.tex has this in the preamble

\usepackage{xr}
  \externaldocument{exercises}
  \externaldocument[H-]{hints} 
  \externaldocument{answers}

then it can use cross reference labels from the other three documents. Suppose that exercises.tex has an enumerated list containing this,

\item \label{exer:EulersThm} What if every vertex has odd degree?

and hints.tex has an enumerated list with this,

\item \label{exer:EulersThm} Distinguish the case of two vertices.

and answers.tex has an enumerated list with this,

\item \label{ans:EulersThm} There is no Euler path, except if there
  are exactly two vertices.

After compiling the exercises, hints, and answers documents, entering this in the body of lectures.tex will result in the lectures getting the reference numbers used in the other documents.

See Exercise~\ref{exer:EulersThm}, with Hint~\ref{H-exer:EulersThm}.
The solution is Answer~\ref{ans:EulersThm}.

The prefix H- for the reference from the hints file is needed because the label in the hints file is the same as the label in the exercises file. Without that prefix, both references would get the number from the later file.

Note: if the document uses the hyperref package then in place of xr, put \usepackage{xr-hyper} before the \usepackage{hyperref}. Also, if any of the multiple documents uses hyperref then they all must use it.


8 Environments

LaTeX provides many environments for delimiting certain behavior. An environment begins with \begin and ends with \end, like this:

\begin{environment-name}
  ...
\end{environment-name}

The environment-name at the beginning must exactly match that at the end. For instance, the input \begin{table*}...\end{table} will cause an error like: ‘! LaTeX Error: \begin{table*} on input line 5 ended by \end{table}.

Environments are executed within a group.


8.1 abstract

Synopsis:

\begin{abstract}
...
\end{abstract}

Produce an abstract, possibly of multiple paragraphs. This environment is only defined in the article and report document classes (see Document classes).

Using the example below in the article class produces a displayed paragraph. Document class option titlepage causes the abstract to be on a separate page (see Document class options); this is the default only in the report class.

\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}

The next example produces a one column abstract in a two column document (for a more flexible solution, use the package abstract).

\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.}
}

8.2 array

Synopsis:

\begin{array}{cols}
  column 1 entry &column 2 entry ... &column n entry \\
  ...
\end{array}

or:

\begin{array}[pos]{cols}
  column 1 entry &column 2 entry ... &column n entry \\
  ...
\end{array}

Produce a mathematical array. This environment can only be used in math mode (see Modes), and normally appears within a displayed mathematics environment such as equation (see equation). Inside of each row the column entries are separated by an ampersand, (&). Rows are terminated with double-backslashes (see \\).

This example shows a three by three array.

\begin{displaymath}
  \chi(x) =
  \left|              % vertical bar fence
    \begin{array}{ccc}
      x-a  &-b  &-c  \\
      -d   &x-e &-f  \\
      -g   &-h  &x-i
    \end{array}
 \right|
\end{displaymath}

The required argument cols describes the number of columns, their alignment, and the formatting of the intercolumn regions. For instance, \begin{array}{rcl}...\end{array} gives three columns: the first flush right, the second centered, and the third flush left. See tabular for the complete description of cols and of the other common features of the two environments, including the optional pos argument.

There are two ways that array diverges from tabular. The first is that array entries are typeset in math mode, in textstyle (see Math styles) except if the cols definition specifies the column with p{...}, which causes the entry to be typeset in text mode. The second is that, instead of tabular’s parameter \tabcolsep, LaTeX’s intercolumn space in an array is governed by \arraycolsep, which gives half the width between columns. The default for this is ‘5pt’ so that between two columns comes 10pt of space.

To obtain arrays with braces the standard is to use the amsmath package. It comes with environments pmatrix for an array surrounded by parentheses (...), bmatrix for an array surrounded by square brackets [...], Bmatrix for an array surrounded by curly braces {...}, vmatrix for an array surrounded by vertical bars |...|, and Vmatrix for an array surrounded by double vertical bars ||...||, along with a number of other array constructs.

The next example uses the amsmath package.

\usepackage{amsmath}  % in preamble

\begin{equation}    
  \begin{vmatrix}{cc}  % array with vert lines
    a  &b \\
    c  &d       
  \end{vmatrix}=ad-bc
\end{equation}

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


8.3 center

Synopsis:

\begin{center}
  line1 \\
  line2 \\
  ...
\end{center}

Create a new paragraph consisting of a sequence of lines that are centered within the left and right margins. Use double-backslash, \\, to get a line break (see \\). 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.

This environment inserts space above and below the text body. See \centering to avoid such space, for example inside a figure environment.

This example produces three centered lines. There is extra vertical space between the last two lines.

\begin{center}
  A Thesis Submitted in Partial Fufillment \\
  of the Requirements of \\[0.5ex]
  the School of Environmental Engineering
\end{center}

In this example, depending on the page’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.

\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}

A double backslash after the final line is optional. If present it doesn’t add any vertical space.

In a two-column document the text is centered in a column, not in the entire page.


8.3.1 \centering

Synopsis:

{\centering ... }

or

\begin{group}
  \centering ...
\end{group}

Center the material in its scope. It is most often used inside an environment such as figure, or in a parbox.

This example’s \centering declaration causes the graphic to be horizontally centered.

\begin{figure}
  \centering
  \includegraphics[width=0.6\textwidth]{ctan_lion.png}
  \caption{CTAN Lion}  \label{fig:CTANLion}
\end{figure}

The scope of this \centering ends with the \end{figure}.

Unlike the center environment, the \centering command does not add vertical space above and below the text. That’s its advantage in the above example; there is not an excess of space.

It also does not start a new paragraph; it simply changes how LaTeX formats paragraph units. If ww {\centering xx \\ yy} zz is surrounded by blank lines then LaTeX will create a paragraph whose first line ‘ww xx’ is centered and whose second line, not centered, contains ‘yy zz’. Usually what is desired is for the scope of the declaration to contain a blank line or the \end command of an environment such as figure or table that ends the paragraph unit. Thus, if {\centering xx \\ yy\par} zz is surrounded by blank lines then it makes a new paragraph with two centered lines ‘xx’ and ‘yy’, followed by a new paragraph with ‘zz’ that is formatted as usual.


8.4 description

Synopsis:

\begin{description}
  \item[label of first item] text of first item
  \item[label of second item] text of second item
   ...
\end{description}

Environment to make a list of labeled items. Each item’s label 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 ‘Something's wrong--perhaps a missing \item’.

This example shows the environment used for a sequence of definitions.

\begin{description}
  \item[lama] A priest.
  \item[llama] A beast.
\end{description}

The labels ‘lama’ and ‘llama’ are output in boldface, with the left edge on the left margin.

Start list items with the \item command (see \item: An entry in a list). Use the optional labels, as in \item[Main point], because there is no sensible default. Following the \item is optional text, which may contain multiple paragraphs.

Since the labels are in bold style, if the label text calls for a font change given in argument style (see Font styles) then it will come out bold. For instance, if the label text calls for typewriter with \item[\texttt{label text}] 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: \item[{\tt label text}]. Similarly, get the standard roman font with \item[{\rm label text}].

For other major LaTeX labelled list environments, see itemize and enumerate. Unlike those environments, nesting description environments does not change the default label; it is boldface and flush left at all levels.

For information about list layout parameters, including the default values, and for information about customizing list layout, see list. The package enumitem is useful for customizing lists.

This example changes the description labels to small caps.

\renewcommand{\descriptionlabel}[1]{%
  {\hspace{\labelsep}\textsc{#1}}}

8.5 displaymath

Synopsis:

\begin{displaymath}
  mathematical text
\end{displaymath}

Environment to typeset the mathematical text on its own line, in display style and centered. To make the text be flush-left use the global option fleqn (see Document class options).

In the displaymath environment no equation number is added to the math text. One way to get an equation number is to use the equation environment (see equation).

LaTeX will not break the math text across lines.

The amsmath package defines an equation* environment which is functionally identical to displaymath but allows use of other amsmath facilities. In general, amsmathhas significantly more extensive displayed equation facilities. For example, there are a number of ways in that package for having math text broken across lines; see also the breqn package for that (https://ctan.org/pkg/breqn).

The construct \[ math \] is a synonym for the environment \begin{displaymath} math \end{displaymath} but the latter is easier to work with in the source; for instance, searching for a square bracket may get false positives but the word displaymath will likely be unique.

The construct $$math$$ from plain TeX is sometimes used instead of LaTeX’s displaymath. Although the output is similar, but is not officially supported in LaTeX at all; $$ doesn’t support the fleqn option, has different vertical spacing, and doesn’t perform consistency checks.

The output from this example is centered and alone on its line.

\begin{displaymath}
  \int_1^2 x^2\,dx=7/3
\end{displaymath}

Also, the integral sign is larger than the inline version \( \int_1^2 x^2\,dx=7/3 \) produces.


8.6 document

The document environment encloses the entire body of a document. It is required in every LaTeX document. See Starting and ending.


8.6.1 \AtBeginDocument

Synopsis:

\AtBeginDocument{code}

Save code and execute it when \begin{document} 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.

You can issue this command more than once; the successive code lines will be executed in the order that you gave them.


8.6.2 \AtEndDocument

Synopsis:

\AtEndDocument{code}

Save code and execute it near the end of the document. Specifically, it is executed when \end{document} 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 \clearpage at the appropriate point in code.

You can issue this command more than once; the successive code lines will be executed in the order that you gave them.


8.7 enumerate

Synopsis:

\begin{enumerate}
  \item[optional label of first item] text of first item
  \item[optional label of second item] text of second item
  ...
\end{enumerate}

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 ‘1.’, ‘2.’, etc. Each enumerate list environment must have at least one item; having none causes the LaTeX error ‘Something's wrong--perhaps a missing \item’.

This example gives the first two finishers in the 1908 Olympic marathon. As a top-level list the labels would come out as ‘1.’ and ‘2.’.

\begin{enumerate}
 \item Johnny Hayes (USA)
 \item Charles Hefferon (RSA) 
\end{enumerate}

Start list items with the \item command (see \item: An entry in a list). If you give \item an optional argument by following it with square brackets, as in \item[Interstitial label], then the next item will continue the interrupted sequence (see \item: An entry in a list). That is, you will get labels like ‘1.’, then ‘Interstitial label’, then ‘2.’. Following the \item is optional text, which may contain multiple paragraphs.

Enumerations may be nested within other enumerate environments, or within any paragraph-making environment such as itemize (see itemize), up to four levels deep. This gives LaTeX’s default for the format at each nesting level, where 1 is the top level, the outermost level.

  1. arabic number followed by a period: ‘1.’, ‘2.’, …
  2. lowercase letter inside parentheses: ‘(a)’, ‘(b)’ …
  3. lowercase roman numeral followed by a period: ‘i.’, ‘ii.’, …
  4. uppercase letter followed by a period: ‘A.’, ‘B.’, …

The enumerate environment uses the counters \enumi through \enumiv (see Counters).

For other major LaTeX labeled list environments, see description and itemize. For information about list layout parameters, including the default values, and for information about customizing list layout, see list. The package enumitem is useful for customizing lists.

To change the format of the label use \renewcommand (see \newcommand & \renewcommand) on the commands \labelenumi through \labelenumiv. For instance, this first level list will be labelled with uppercase letters, in boldface, and without a trailing period.

\renewcommand{\labelenumi}{\textbf{\Alph{enumi}}}
\begin{enumerate}
  \item Shows as boldface A
  \item Shows as boldface B
\end{enumerate}

For a list of counter-labeling commands see \alph \Alph \arabic \roman \Roman \fnsymbol: Printing counters.


8.8 eqnarray

The eqnarray environment is obsolete. It has infelicities, including spacing that is inconsistent with other mathematics elements. (See “Avoid eqnarray!” by Lars Madsen https://tug.org/TUGboat/tb33-1/tb103madsen.pdf). New documents should include the amsmath package and use the displayed mathematics environments provided there, such as the align environment. We include a description only for completeness and for working with old documents.

Synopsis:

\begin{eqnarray} 
  first formula left  &first formula middle  &first formula right \\
  ...
\end{eqnarray}

or

\begin{eqnarray*} 
  first formula left  &first formula middle  &first formula right \\
  ...
\end{eqnarray*}

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.

It is similar to a three-column array environment, with items within a row separated by an ampersand (&), and with rows separated by double backslash  \\). The starred form of line break (\\*) can also be used to separate equations, and will disallow a page break there (see \\).

The unstarred form eqnarray places an equation number on every line (using the equation counter), unless that line contains a \nonumber command. The starred form eqnarray* omits equation numbering, while otherwise being the same.

The command \lefteqn is used for splitting long formulas across lines. It typesets its argument in display style flush left in a box of zero width.

This example shows three lines. The first two lines make an inequality, while the third line has not entry on the left side.

\begin{eqnarray*}
  \lefteqn{x_1+x_2+\cdots+x_n}     \\
    &\leq &y_1+y_2+\cdots+y_n      \\
    &=    &z+y_3+\cdots+y_n
\end{eqnarray*}

8.9 equation

Synopsis:

\begin{equation}
  mathematical text
\end{equation}

The same as a displaymath environment (see displaymath) except that LaTeX puts an equation number flush to the right margin. The equation number is generated using the equation counter.

You should have no blank lines between \begin{equation} and \end{equation}, or LaTeX will tell you that there is a missing dollar sign.

The package amsmath package has extensive displayed equation facilities. New documents should include this package.


8.10 figure

Synopsis:

\begin{figure}[placement]
  figure body
  \caption[loftitle]{title}  % optional
  \label{label}              % optional
\end{figure}

or:

\begin{figure*}[placement]
  figure body
  \caption[loftitle]{title}  % optional
  \label{label}              % optional  
\end{figure*}

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 “floats” them to a convenient place, such as the top of a following page (see Floats).

The figure body can consist of imported graphics (see Graphics), or text, LaTeX commands, etc. It is typeset in a parbox of width \textwidth.

The possible values of placement are h for ‘here’, t for ‘top’, b for ‘bottom’, and p for ‘on a separate page of floats’. For the effect of these options on the float placement algorithm, see Floats.

The starred form figure* is used when a document is in double-column mode (see \twocolumn). 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 placement b in Floats.

The label is optional; it is used for cross references (see Cross references). The optional \caption command specifies caption text for the figure (see \caption). By default it is numbered. If loftitle is present, it is used in the list of figures instead of title (see Table of contents, list of figures, list of tables).

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.

\usepackage{graphicx}  % in preamble
  ...
\begin{figure}[t]
  \centering
  \includegraphics[width=0.5\textwidth]{CTANlion.png}
  \caption{The CTAN lion, by Duane Bibby}
\end{figure}

8.11 filecontents

Synopsis:

\begin{filecontents}[option]{filename}
  text
\end{filecontents}

or

\begin{filecontents*}[option]{filename}
  text
\end{filecontents*}

Create a file named filename in the current directory (or the output directory, if specified; see output directory) and write text to it. By default, an existing file is not overwritten.

The unstarred version of the environment filecontents prefixes the content of the created file with a header of TeX comments; see the example below. The starred version filecontents* does not include the header.

The possible options are:

force
overwrite

Overwrite an existing file.

noheader

Omit the header. Equivalent to using filecontents*.

nosearch

Only check the current directory (and the output directory, if specified) for an existing file, not the entire search path.

These options were added in a 2019 release of LaTeX.

This environment can be used anywhere in the preamble, although it often appears before the \documentclass command. It is commonly used to create a .bib or other such data file from the main document source, to make the source file self-contained. Similarly, it can be used to create a custom style or class file, again making the source self-contained.

For example, this document:

\documentclass{article}
\begin{filecontents}{JH.sty}
\newcommand{\myname}{Jim Hef{}feron}
\end{filecontents}
\usepackage{JH}
\begin{document}
Article by \myname.
\end{document}

produces this file JH.sty:

%% LaTeX2e file `JH.sty'
%% generated by the `filecontents' environment
%% from source `test' on 2015/10/12.
%%
\newcommand{\myname}{Jim Hef{}feron}

8.12 flushleft

Synopsis:

\begin{flushleft}
  line1 \\
  line2 \\
  ...
\end{flushleft}

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 interword spaces. To force a new line use a double backslash, \\. For the declaration form see \raggedright.

This creates a box of text that is at most 3 inches wide, with the text flush left and ragged right.

\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}

8.12.1 \raggedright

Synopses:

{\raggedright  ... }

or

\begin{environment} \raggedright
  ...
\end{environment}

A declaration which causes lines to be flush to the left margin and ragged right. It can be used inside an environment such as quote or in a parbox. For the environment form see flushleft.

Unlike the flushleft environment, the \raggedright command does not start a new paragraph; it only changes how LaTeX formats paragraph units. To affect a paragraph unit’s format, the scope of the declaration must contain the blank line or \end command that ends the paragraph unit.

Here \raggedright in each second column keeps LaTeX from doing awkward typesetting to fit the text into the narrow column. Note that \raggedright is inside the curly braces {...} to delimit its effect.

\begin{tabular}{rp{2in}}
  Team alpha  &{\raggedright This team does all the real work.} \\
  Team beta   &{\raggedright This team ensures that the water
                cooler is never empty.}                         \\
\end{tabular}

8.13 flushright

\begin{flushright}
  line1 \\
  line2 \\
  ...
\end{flushright}

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 inter-word spaces. To force a new line use a double backslash, \\. For the declaration form see \raggedleft.

For an example related to this environment, see flushleft, where one just have mutatis mutandis to replace flushleft by flushright.


8.13.1 \raggedleft

Synopses:

{\raggedleft  ... }

or

\begin{environment} \raggedleft
  ...
\end{environment}

A declaration which causes lines to be flush to the right margin and ragged left. It can be used inside an environment such as quote or in a parbox. For the environment form see flushright.

Unlike the flushright environment, the \raggedleft command does not start a new paragraph; it only changes how LaTeX formats paragraph units. To affect a paragraph unit’s formatting, the scope of the declaration must contain the blank line or \end command that ends the paragraph unit.

See \raggedright, for an example related to this environment; just replace \raggedright there by \raggedleft.


8.14 itemize

Synopsis:

\begin{itemize}
  \item[optional label of first item] text of first item
  \item[optional label of second item] text of second item
  ...
\end{itemize}

Produce an unordered list, sometimes called a bullet list. There must be at least one \item within the environment; having none causes the LaTeX error ‘Something's wrong--perhaps a missing \item’.

This gives a two-item list:

\begin{itemize}
 \item Pencil and watercolor sketch by Cassandra
 \item Rice portrait 
\end{itemize}

By default, in a top-level list each label would come out as a bullet, •. The format of the labeling depends on the nesting level; see below.

Many language adaptations change list formatting, in which case this section may apply only partially or not at all. For instance, after this:

\usepackage[french]{babel} % changes list formatting!

the margins are smaller and the item markers are different.

Start list items with the \item command (see \item: An entry in a list). If you give \item an optional argument by following it with square brackets, as in \item[Optional label], then by default Optional label will appear in bold and be flush right, so it could extend into the left margin. For labels that are flush left see the description environment. Following the \item is the text of the item, which may be empty or contain multiple paragraphs.

Unordered lists can be nested within one another, up to four levels deep. They can also be nested within other paragraph-making environments, such as enumerate (see enumerate).

The itemize environment uses the commands \labelitemi through \labelitemiv to produce the default label (note 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.

  1. • (bullet, from \textbullet)
  2. (bold en-dash, from \normalfont\bfseries\textendash)
  3. * (asterisk, from \textasteriskcentered)
  4. \cdot (vertically centered dot, rendered here as a period, from \textperiodcentered)

Change the labels with \renewcommand. For instance, this makes the first level use diamonds.

\renewcommand{\labelitemi}{$\diamond$}

The distance between the left margin of the enclosing environment and the left margin of the itemize list is determined by the parameters \leftmargini through \leftmarginvi. (This also uses the convention of using lowercase roman numerals at the end of the command name to denote the nesting level.) The defaults are: 2.5em in level 1 (2em in two-column mode), 2.2em in level 2, 1.87em in level 3, and 1.7em in level 4, with smaller values for more deeply nested levels. The margin parameters must be overridden before the list starts.

For other major LaTeX labeled list environments, see description and enumerate. The itemize, enumerate and description environment use the same list layout parameters. For a description, including the default values, and for information about customizing list layout, see list. The package enumitem is useful for customizing lists.

This example greatly reduces the margin space for an outermost itemized list:

\setlength{\leftmargini}{1.25em} % default 2.5em
\begin{itemize}
\item ...
\end{itemize}

Especially for lists with short items, it may be desirable to elide space between items. Here is an example defining an itemize* environment with no extra spacing between items, or between paragraphs within a single item (\parskip is not list-specific; see \parindent & \parskip):

\newenvironment{itemize*}%
  {\begin{itemize}%
    \setlength{\itemsep}{0pt}%
    \setlength{\parsep}{0pt}%
    \setlength{\parskip}{0pt}%
  }%
  {\end{itemize}}

8.15 letter environment: writing letters

This environment is used for creating letters. See Letters.


8.16 list

Synopsis:

\begin{list}{labeling}{spacing}
  \item[optional label of first item] text of first item
  \item[optional label of second item] text of second item
  ...
\end{list}

An environment for constructing lists.

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 description, enumerate, and itemize environments (see description, enumerate, and itemize).

Instead, the list 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 list, including quotation, quote, and center (see quotation & quote, see center).

This uses the list environment to define a new custom environment.

\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}

The mandatory first argument labeling specifies the default labeling of list items. It can contain text and LaTeX commands, as above where it contains both ‘Item’ and ‘\Roman{…}’. LaTeX forms the label by putting the labeling argument in a box of width \labelwidth. 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 \item an optional argument by including square braces and the text, as in the above \item[Special label.]; see \item: An entry in a list.

The mandatory second argument spacing has a list of commands. This list can be empty. A command that can go in here is \usecounter{countername} (see \usecounter). 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 \item that does not have an optional argument.

Another command that can go in spacing is \makelabel, 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 Modes). One example of changing its definition is that to the above named example, before the definition of the environment add \newcommand{\namedmakelabel}[1]{\textsc{#1}}, and between the \setlength command and the parenthesis that closes the spacing argument also add \let\makelabel\namedmakelabel. Then the labels will be typeset in small caps. Similarly, changing the second code line to \let\makelabel\fbox puts the labels inside a framed box. Another example of the \makelabel command is below, in the definition of the redlabel environment.

Also often in spacing 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 itemize can be different than the values shown here.) See also the figure that follows the list. Each is a length (see Lengths). The vertical spaces are normally rubber lengths, with plus and minus components, to give TeX flexibility in setting the page. Change each with a command such as \setlength{\itemsep}{2pt plus1pt minus1pt}. For some effects these lengths should be zero or negative.

\itemindent

Extra horizontal space indentation, beyond leftmargin, of the first line each item. Its default value is 0pt.

\itemsep

Vertical space between items, beyond the \parsep. The defaults for the first three levels in LaTeX’s ‘article’, ‘book’, and ‘report’ classes at 10 point size are: 4pt plus2pt minus1pt, \parsep (that is, 2pt plus1pt minus1pt), and \topsep (that is, 2pt plus1pt minus1pt). The defaults at 11 point are: 4.5pt plus2pt minus1pt, \parsep (that is, 2pt plus1pt minus1pt), and \topsep (that is, 2pt plus1pt minus1pt). The defaults at 12 point are: 5pt plus2.5pt minus1pt, \parsep (that is, 2.5pt plus1pt minus1pt), and \topsep (that is, 2.5pt plus1pt minus1pt).

\labelsep

Horizontal space between the label and text of an item. The default for LaTeX’s ‘article’, ‘book’, and ‘report’ classes is 0.5em.

\labelwidth

Horizontal width. The box containing the label is nominally this wide. If \makelabel 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 \makelabel returns text of width less than or equal to \labelwidth then LaTeX’s default is that the label is typeset flush right in a box of this width.

The left edge of the label box is \leftmargin+\itemindent-\labelsep-\labelwidth from the left margin of the enclosing environment.

The default for LaTeX’s ‘article’, ‘book’, and ‘report’ classes at the top level is \leftmargini-\labelsep, (which is 2em in one column mode and 1.5em in two column mode). At the second level it is \leftmarginii-\labelsep, and at the third level it is \leftmarginiii-\labelsep. These definitions make the label’s left edge coincide with the left margin of the enclosing environment.

\leftmargin

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.

In the standard LaTeX document classes, a top-level list has this set to the value of \leftmargini, while a list that is nested inside a top-level list has this margin set to \leftmarginii. More deeply nested lists get the values of \leftmarginiii through \leftmarginvi. (Nesting greater than level five generates the error message ‘Too deeply nested’.)

The defaults for the first three levels in LaTeX’s ‘article’, ‘book’, and ‘report’ classes are: \leftmargini is 2.5em (in two column mode, 2em), \leftmarginii is 2.2em, and \leftmarginiii is 1.87em.

\listparindent

Horizontal space of additional line indentation, beyond \leftmargin, for second and subsequent paragraphs within a list item. A negative value makes this an “outdent”. Its default value is 0pt.

\parsep

Vertical space between paragraphs within an item. The defaults for the first three levels in LaTeX’s ‘article’, ‘book’, and ‘report’ classes at 10 point size are: 4pt plus2pt minus1pt, 2pt plus1pt minus1pt, and 0pt. The defaults at 11 point size are: 4.5pt plus2pt minus1pt, 2pt plus1pt minus1pt, and 0pt. The defaults at 12 point size are: 5pt plus2.5pt minus1pt, 2.5pt plus1pt minus1pt, and 0pt.

\partopsep

Vertical space added, beyond \topsep+\parskip, 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.)

The defaults for the first three levels in LaTeX’s ‘article’, ‘book’, and ‘report’ classes at 10 point size are: 2pt plus1 minus1pt, 2pt plus1pt minus1pt, and 1pt plus0pt minus1pt. The defaults at 11 point are: 3pt plus1pt minus1pt, 3pt plus1pt minus1pt, and 1pt plus0pt minus1pt). The defaults at 12 point are: 3pt plus2pt minus3pt, 3pt plus2pt minus2pt, and 1pt plus0pt minus1pt.

\rightmargin

Horizontal space between the right margin of the list and the right margin of the enclosing environment. Its default value is 0pt. It must be non-negative.

\topsep

Vertical space added to both the top and bottom of the list, in addition to \parskip (see \parindent & \parskip). The defaults for the first three levels in LaTeX’s ‘article’, ‘book’, and ‘report’ classes at 10 point size are: 8pt plus2pt minus4pt, 4pt plus2pt minus1pt, and 2pt plus1pt minus1pt. The defaults at 11 point are: 9pt plus3pt minus5pt, 4.5pt plus2pt minus1pt, and 2pt plus1pt minus1pt. The defaults at 12 point are: 10pt plus4pt minus6pt, 5pt plus2.5pt minus1pt, and 2.5pt plus1pt minus1pt.

This shows the horizontal and vertical distances.

latex2e-figures/list

The lengths shown are listed below. The key relationship is that the right edge of the bracket for h1 equals the right edge of the bracket for h4, so that the left edge of the label box is at h3+h4-(h0+h1).

v0

\topsep + \parskip if the list environment does not start a new paragraph, and \topsep+\parskip+\partopsep if it does

v1

\parsep

v2

\itemsep+\parsep

v3

Same as v0. (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.)

h0

\labelwidth

h1

\labelsep

h2

\listparindent

h3

\leftmargin

h4

\itemindent

h5

\rightmargin

The list’s left and right margins, shown above as h3 and h5, 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 \linewidth (see Page layout parameters). For instance, set the list’s left margin to be one quarter of the distance between the left and right margins of the enclosing environment with \setlength{\leftmargin}{0.25\linewidth}.

Page breaking in a list structure is controlled by the three parameters below. For each, the LaTeX default is -\@lowpenalty, that is, -51. Because it is negative, it somewhat encourages a page break at each spot. Change it with, e.g., \@beginparpenalty=9999; a value of 10000 prohibits a page break.

\@beginparpenalty

The page breaking penalty for breaking before the list (default -51).

\@itempenalty

The page breaking penalty for breaking before a list item (default -51).

\@endparpenalty

The page breaking penalty for breaking after a list (default -51).

The package enumitem is useful for customizing lists.

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 \usecounter.

\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}}

8.16.1 \item: An entry in a list

Synopsis:

\item text of item

or

\item[optional-label] text of item

An entry in a list. The entries are prefixed by a label, whose default depends on the list type.

Because the optional label is surrounded by square brackets ‘[...]’, if you have an item whose text starts with [, you have to hide the bracket inside curly braces, as in: \item {[} is an open square bracket; otherwise, LaTeX will think it marks the start of an optional label.

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: \item[Close square bracket, {]}]. See LaTeX command syntax.

In this example the enumerate list has two items that use the default label and one that uses the optional label.

\begin{enumerate}
  \item Moe
  \item[sometimes] Shemp
  \item Larry
\end{enumerate}

The first item is labelled ‘1.’, the second item is labelled ‘sometimes’, and the third item is labelled ‘2.’. Because of the optional label in the second item, the third item is not labelled ‘3.’.


8.16.2 trivlist: A restricted form of list

Synopsis:

\begin{trivlist}
  ...
\end{trivlist}

A restricted version of the list environment, in which margins are not indented and an \item without an optional argument produces no text. It is most often used in macros, to define an environment where the \item command is part of the environment’s definition. For instance, the center environment is defined essentially like this:

\newenvironment{center}
  {\begin{trivlist}\centering\item\relax}
  {\end{trivlist}}

Using trivlist 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.

Specifically, trivlist uses the current values of the list parameters (see list), except that \parsep is set to the value of \parskip, and \leftmargin, \labelwidth, and \itemindent are set to zero.

This example outputs the items as two paragraphs, except that (by default) they have no paragraph indent and are vertically separated.

\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}

8.17 math

Synopsis:

\begin{math}
math
\end{math}

The math environment inserts given math material within the running text. \(...\) and $...$ are synonyms. See Math formulas.


8.18 minipage

Synopses:

\begin{minipage}{width}
  contents
\end{minipage}

or

\begin{minipage}[position][height][inner-pos]{width}
  contents
\end{minipage}

Put contents into a box that is width 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 minipage is similar to \parbox (see \parbox) but here you can have paragraphs.

This example will be 3 inches wide, and has two paragraphs.

\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}

See below for a discussion of the paragraph indent inside a minipage.

The required argument width is a rigid length (see Lengths). It gives the width of the box into which contents are typeset.

There are three optional arguments, position, height, and inner-pos. You need not include all three. For example, get the default position and set the height with \begin{minipage}[c][2.54cm]{\columnwidth} contents \end{minipage}. (Get the natural height with an empty argument, [].)

The optional argument position governs how the minipage vertically aligns with the surrounding material.

c

(synonym m) Default. Positions the minipage so its vertical center lines up with the center of the adjacent text line.

t

Align the baseline of the top line in the minipage with the baseline of the surrounding text (plain TeX’s \vtop).

b

Align the baseline of the bottom line in the minipage with the baseline of the surrounding text (plain TeX’s \vbox).

To see the effects of these, contrast running this

---\begin{minipage}[c]{0.25in}
  first\\ second\\ third
\end{minipage}

with the results of changing c to b or t.

The optional argument height is a rigid length (see Lengths). It sets the height of the minipage. You can enter any value larger than, or equal to, or smaller than the minipage’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.

The final optional argument inner-pos controls the placement of contents inside the box. These are the possible values are (the default is the value of position).

t

Place contents at the top of the box.

c

Place it in the vertical center.

b

Place it at the box bottom.

s

Stretch contents out vertically; it must contain vertically stretchable space.

The inner-pos argument makes sense when the height option is set to a value larger than the minipage’s natural height. To see the effect of the options, run this example with the various choices in place of b.

Text before
\begin{center}
  ---\begin{minipage}[c][3in][b]{0.25\textwidth}
       first\\ second\\ third
  \end{minipage}              
\end{center}
Text after  

By default paragraphs are not indented in a minipage. Change that with a command such as \setlength{\parindent}{1pc} at the start of contents.

Footnotes in a minipage environment are handled in a way that is particularly useful for putting footnotes in figures or tables. A \footnote or \footnotetext command puts the footnote at the bottom of the minipage instead of at the bottom of the page, and it uses the \mpfootnote counter instead of the ordinary footnote counter (see Counters).

This puts the footnote at the bottom of the table, not the bottom of the page.

\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}  &\textsc{Reign}             \\ \hline
      Elizabeth II      &63 years\footnote{to date} \\
      Victoria          &63 years                   \\
      George III        &59 years
    \end{tabular}
  \end{center}  
\end{minipage}
\end{center}

If you nest minipages then there is an oddness when using footnotes. Footnotes appear at the bottom of the text ended by the next \end{minipage} which may not be their logical place.

This puts a table containing data side by side with a map graphic. They are vertically centered.

% siunitx to have the S column specifier,
% which aligns numbers on their decimal point.
\usepackage{siunitx}
\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|S}
      % \multicolumn to remove vertical bar between column headers
      \multicolumn{1}{r}{Borough} &
      % braces to prevent siunitx from misinterpreting the
      % period as a decimal separator
      {Pop. (million)}  \\ \hline
      The Bronx      &1.5  \\
      Brooklyn       &2.6  \\
      Manhattan      &1.6  \\
      Queens         &2.3  \\
      Staten Island  &0.5  
    \end{tabular}
  \end{minipage}              
\end{center}

8.19 picture

Synopses:

\begin{picture}(width,height)
   picture command
\end{picture}

or

\begin{picture}(width,height)(xoffset,yoffset)
  picture command
\end{picture}

Where there may be any number of picture command’s.

An environment to create simple pictures containing lines, arrows, boxes, circles, and text. This environment is not obsolete, but new documents typically use much more powerful graphics creation systems, such as TikZ, PSTricks, MetaPost, or Asymptote. None of these are covered in this document; see CTAN.

To start, here’s an example showing the parallelogram law for adding vectors.

\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){\vector(1,1){6}}
    \put(3,3){\makebox(0,0)[r]{sum\ }}  
\end{picture}

The picture environment has one required argument, a pair of positive real numbers (width,height). Multiply these by the value \unitlength to get the nominal size of the output, i.e. 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’s box.

This environment also has an optional argument (xoffset,yoffset). 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 a pair of two real numbers, but these may also be negative or null. Multiply these by \unitlength to get the coordinates of the point at the lower-left corner of the picture.

For example, if \unitlength has been set to 1mm, the command

\begin{picture}(100,200)(10,20)

produces a box of width 100 millimeters and height 200 millimeters. The picture’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.

Each picture command tells LaTeX where to put something by providing its position. A position is a pair such as (2.4,-5) giving the x- and y-coordinates. A coordinate 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 \unitlength, so if \unitlength has been set to 1cm, then the coordinate 2.54 specifies a length of 2.54 centimeters.

LaTeX’s default for \unitlength is 1pt. It is a rigid length (see Lengths). Change it with the \setlength command (see \setlength). Make this change only outside of a picture environment.

The picture environment supports using standard arithmetic expressions as well as numbers.

Coordinates are given with respect to an origin, which is by default at the lower-left corner of the picture. Note that when a position appears as an argument, as with \put(1,2){...}, 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, for example, y = 1 is above y = 0.

There are four ways to put things in a picture: \put, \multiput, \qbezier, and \graphpaper. The most often used is \put. This

\put(11.3,-0.3){...}

places the object with its reference point at coordinates (11.3,-0.3). The reference points for various objects will be described below. The \put command creates an LR box (see Modes). Anything that can go in an \mbox (see \mbox & \makebox) can go in the text argument of the \put command. The reference point will be the lower left corner of the box. In this picture

\setlength{\unitlength}{1cm}
...\begin{picture}(1,1)
  \put(0,0){\line(1,0){1}}
  \put(0,0){\line(1,1){1}}
\end{picture}

the three dots are just slightly left of the point of the angle formed by the two lines. (Also, \line(1,1){1} does not call for a line of length one; rather the line has a change in the x coordinate of 1.)

The \multiput, qbezier, and graphpaper commands are described below.

You can also use this environment to place arbitrary material at an exact location. For example:

\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}

The red X will be precisely a quarter of the \textwidth from the left margin, and 0.35\textwidth up from the bottom of the picture. Another example of this usage is to put similar code in the page header to get repeat material on each of a document’s pages.


8.19.1 \put

Synopsis:

\put(xcoord,ycoord){content}

Place content at the coordinate (xcoord,ycoord). See the discussion of coordinates and \unitlength in picture. The content is processed in LR mode (see Modes) so it cannot contain line breaks.

This includes the text into the picture.

\put(4.5,2.5){Apply the \textit{unpoke} move}

The reference point, the location (4.5,2.5), is the lower left of the text, at the bottom left of the ‘A’.


8.19.2 \multiput

Synopsis:

\multiput(x,y)(delta_x,delta_y){num-copies}{obj}

Copy obj a total of num-copies times, with an increment of delta_x,delta_y. The obj first appears at position (x,y), then at (x+\delta_x,y+\delta_y), and so on.

This draws a simple grid with every fifth line in bold (see also \graphpaper).

\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}

8.19.3 \qbezier

Synopsis:

\qbezier(x1,y1)(x2,y2)(x3,y3)
\qbezier[num](x1,y1)(x2,y2)(x3,y3)

Draw a quadratic Bezier curve whose control points are given by the three required arguments (x1,y1), (x2,y2), and (x3,y3). That is, the curve runs from (x1,y1) to (x3,y3), is quadratic, and is such that the tangent line at (x1,y1) passes through (x2,y2), as does the tangent line at (x3,y3).

This draws a curve from the coordinate (1,1) to (1,0).

\qbezier(1,1)(1.25,0.75)(1,0)  

The curve’s tangent line at (1,1) contains (1.25,0.75), as does the curve’s tangent line at (1,0).

The optional argument num gives the number of calculated intermediate points. The default is to draw a smooth curve whose maximum number of points is \qbeziermax (change this value with \renewcommand).

This draws a rectangle with a wavy top, using \qbezier for that curve.

\begin{picture}(8,4)
  \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
  \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}

8.19.4 \graphpaper

Synopsis:

\graphpaper(x_init,y_init)(x_dimen,y_dimen)
\graphpaper[spacing](x_init,y_init)(x_dimen,y_dimen)

Draw a coordinate grid. Requires the graphpap package. The grid’s origin is (x_init,y_init). Grid lines come every spacing units (the default is 10). The grid extends x_dimen units to the right and y_dimen units up. All arguments must be positive integers.

This make a grid with seven vertical lines and eleven horizontal lines.

\usepackage{graphpap}    % in preamble
  ...
\begin{picture}(6,20)    % in document body
  \graphpaper[2](0,0)(12,20)
\end{picture}

The lines are numbered every ten units.


8.19.5 \line

Synopsis:

\line(x_run,y_rise){travel}

Draw a line. It slopes such that it vertically rises y_rise for every horizontal x_run. The travel is the total horizontal change—it is not the length of the vector, it is the change in x. In the special case of vertical lines, where (x_run,y_rise)=(0,1), the travel gives the change in y.

This draws a line starting at coordinates (1,3).

\put(1,3){\line(2,5){4}}

For every over 2, this line will go up 5. Because travel specifies that this goes over 4, it must go up 10. Thus its endpoint is (1,3)+(4,10)=(5,13). In particular, note that travel=4 is not the length of the line, it is the change in x.

The arguments x_run and y_rise are integers that can be positive, negative, or zero. (If both are 0 then LaTeX treats the second as 1.) With \put(x_init,y_init){\line(x_run,y_rise){travel}}, if x_run is negative then the line’s ending point has a first coordinate that is less than x_init. If y_rise is negative then the line’s ending point has a second coordinate that is less than y_init.

If travel is negative then you get LaTeX Error: Bad \line or \vector argument.

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 x_run and y_rise must have integer values from −6 through 6. Also, they must be relatively prime, so that (x_run,y_rise) 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 picture, see the package pict2e (https://ctan.org/pkg/pict2e). Another solution is to use a full-featured graphics system such as TikZ, PSTricks, MetaPost, or Asymptote.


8.19.6 \linethickness

Synopsis:

\linethickness{dim}

Declares the thickness of subsequent horizontal and vertical lines in a picture to be dim, which must be a positive length (see Lengths). It differs from \thinlines and \thicklines in that it does not affect the thickness of slanted lines, circles, or ovals (see \oval).


8.19.7 \thinlines

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 \linethickness or \thicklines.


8.19.8 \thicklines

Declaration to set the thickness of subsequent lines, circles, and ovals in a picture environment to be 0.8pt. See also \linethickness and \thinlines. This command is illustrated in the Trapezoidal Rule example of \qbezier.


8.19.9 \circle

Synopsis:

\circle{diameter}
\circle*{diameter}

Produces a circle with a diameter as close as possible to the specified one. The * form produces a filled-in circle.

This draws a circle of radius 6, centered at (5,7).

\put(5,7){\circle{6}}

The available radii for \circle are, in points, the even numbers from 2 to 20, inclusive. For \circle* they are all the integers from 1 to 15.


8.19.10 \oval

Synopsis:

\oval(width,height)
\oval(width,height)[portion]

Produce a rectangle with rounded corners, hereinafter referred to as an oval. The optional argument portion allows you to produce only half or a quarter of the oval. For half an oval take portion to be one of these.

t

top half

b

bottom half

r

right half

l

left half

Produce only one quarter of the oval by setting portion to tr, br, bl, or tl.

This draws the top half of an oval that is 3 wide and 7 tall.

\put(5,7){\oval(3,7)[t]}

The (5,7) is the center of the entire oval, not just the center of the top half.

These shapes are not ellipses. They are rectangles whose corners are made with quarter circles. These circles have a maximum radius of 20pt (see \circle for the sizes). Thus large ovals are just frames with a small amount of corner rounding.


8.19.11 \shortstack

Synopsis:

\shortstack[position]{line 1 \\ ... }

Produce a vertical stack of objects.

This labels the y axis by writing the word ‘y’ above the word ‘axis’.

\setlength{\unitlength}{1cm}
\begin{picture}(5,2.5)(-0.75,0)
   \put(0,0){\vector(1,0){4}}   % x axis
   \put(0,0){\vector(0,1){2}}   % y
   \put(-0.2,2){\makebox(0,0)[r]{\shortstack[r]{$y$\\ axis}}}
\end{picture}

For a short stack, the reference point is the lower left of the stack. In the above example the \makebox (see \mbox & \makebox) puts the stack flush right in a zero width box so in total the short stack sits slightly to the left of the y axis.

The valid positions are:

r

Make objects flush right

l

Make objects flush left

c

Center objects (default)

Separate objects into lines with \\. These stacks are short in that, unlike in a tabular or array environment, here the rows are not spaced out to be of even baseline skips. Thus, in \shortstack{X\\o\\o\\X} the first and last rows are taller than the middle two, and therefore the baseline skip between the two middle rows is smaller than that between the third and last row. You can adjust row heights and depths either by putting in the usual interline spacing with \shortstack{X\\ \strut o\\o\\X} (see \strut), or explicitly, via an zero-width box \shortstack{X \\ \rule{0pt}{12pt} o\\o\\X} or by using \\’s optional argument \shortstack{X\\[2pt] o\\o\\X}.

The \shortstack command is also available outside the picture environment.


8.19.12 \vector

Synopsis:

\vector(x_run,y_rise){travel}

Draw a line ending in an arrow. The slope of that line is: it vertically rises y_rise for every horizontal x_run. The travel is the total horizontal change—it is not the length of the vector, it is the change in x. In the special case of vertical vectors, if (x_run,y_rise)=(0,1), then travel gives the change in y.

For an example see picture.

For elaboration on x_run and y_rise see \line. As there, the values of x_run and y_rise are limited. For \vector you must chooses integers between −4 and 4, inclusive. Also, the two you choose must be relatively prime. Thus, \vector(2,1){4} is acceptable but \vector(4,2){4} is not (if you use the latter then you get a sequence of arrowheads).


8.19.13 \makebox (picture)

Synopsis:

\makebox(rec-width,rec-height){text}
\makebox(rec-width,rec-height)[position]{text}

Make a box to hold text. This command fits with the picture environment, although you can use it outside of there, because rec-width and rec-height are numbers specifying distances in terms of the \unitlength (see picture). This command is similar to the normal \makebox command (see \mbox & \makebox) except here that you must specify the width and height. This command is fragile (see \protect).

This makes a box of length 3.5 times \unitlength and height 4 times \unitlength.

\put(1,2){\makebox(3.5,4){...}}

The optional argument position specifies where in the box the text 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.

t

Puts text the top of the box.

b

Put text at the bottom.

l

Put text on the left.

r

Put text on the right.


8.19.14 \framebox (picture)

Synopsis:

\framebox(rec-width,rec-height){text}
\framebox(rec-width,rec-height)[position]{text}

This is the same as \makebox (picture) 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 picture environment, although you can use it outside of there, because lengths are numbers specifying the distance in terms of the \unitlength (see picture). This command is fragile (see \protect).

This example creates a frame 2.5 inches by 3 inches and puts the text in the center.

\setlength{\unitlength}{1in}
\framebox(2.5,3){test text}

The required arguments are that the rectangle has overall width rect-width units and height rect-height units.

The optional argument position specifies the position of text; see \makebox (picture) for the values that it can take.

The rule has thickness \fboxrule and there is a blank space \fboxsep between the frame and the contents of the box.

For this command, you must specify the width and height. If you want to just put a frame around some contents whose dimension is determined in some other way then either use \fbox (see \fbox & \framebox) or \frame (see \frame).


8.19.15 \frame

Synopsis:

\frame{contents}

Puts a rectangular frame around contents. The reference point is the bottom left corner of the frame. In contrast to \framebox (see \framebox (picture)), this command puts no extra space between the frame and the object. It is fragile (see \protect).


8.19.16 \dashbox

Synopsis:

\dashbox{dash-len}(rect-width,rect-height){text}
\dashbox{dash-len}(rect-width,rect-height)[position]{text}

Create a dashed rectangle around text. This command fits with the picture environment, although you can use it outside of there, because lengths are numbers specifying the distance in terms of the \unitlength (see picture).

The required arguments are: dashes are dash-len units long, with the same length gap, and the rectangle has overall width rect-width units and height rect-height units.

The optional argument position specifies the position of text; see \makebox (picture) for the values that it can take.

This shows that you can use non-integer value for dash-len.

\put(0,0){\dashbox{0.1}(5,0.5){My hovercraft is full of eels.}}

Each dash will be 0.1\unitlength long, the box’s width is 5\unitlength and its height is 0.5\unitlength.

As in that example, a dashed box looks best when rect-width and rect-height are multiples of the dash-len.


8.20 quotation & quote

Synopsis:

\begin{quotation}
  text
\end{quotation}

or

\begin{quote}
  text
\end{quote}

Include a quotation. Both environments indent margins on both sides by \leftmargin and the text is right-justified.

They differ in how they treat paragraphs. In the quotation environment, paragraphs are indented by 1.5em and the space between paragraphs is small, 0pt plus 1pt. In the quote environment, paragraphs are not indented and there is vertical space between paragraphs (it is the rubber length \parsep).

\begin{quotation} \small\it
  Four score and seven years ago
  ... shall not perish from the earth.
  \hspace{1em plus 1fill}---Abraham Lincoln
\end{quotation}

8.21 tabbing

Synopsis:

\begin{tabbing}
row1col1 \= row1col2 ...  \\
row2col1 \> row2col2 ...  \\
...
\end{tabbing}

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 tabular (see tabular) or array (see array) because in those the width of each column need not be constant and need not be known in advance.

This example has a first line where the tab stops are set to explicit widths, ended by a \kill command (which is described below):

\begin{tabbing}
\hspace{1.2in}\=\hspace{1in}\=\kill
Ship                \>Guns             \>Year    \\ 
\textit{Sophie}     \>14               \>1800    \\
\textit{Polychrest} \>24               \>1803    \\
\textit{Lively}     \>38               \>1804    \\
\textit{Surprise}   \>28               \>1805    \\
\end{tabbing}

Both the tabbing environment and the more widely-used tabular environment put text in columns. The most important distinction is that in tabular the width of columns is determined automatically by LaTeX, while in tabbing the user sets the tab stops. Another distinction is that tabular generates a box, but tabbing can be broken across pages. Finally, while tabular can be used in any mode, tabbing can be used only in paragraph mode and it 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 \hspace command at the beginning of a tabbed row. The right margin of the tabbing environment is the end of line, so that the width of the environment is \linewidth.

The tabbing environment contains a sequence of tabbed rows. The first tabbed row begins immediately after \begin{tabbing} and each row ends with \\ or \kill. The last row may omit the \\ and end with just \end{tabbing}.

At any point the tabbing environment has a current tab stop pattern, a sequence of n > 0 tab stops, numbered 0, 1, etc. These create n corresponding columns. Tab stop 0 is always the left margin, defined by the enclosing environment. Tab stop number i is set if it is assigned a horizontal position on the page. Tab stop number i can only be set if all the stops 0, …, i-1 have already been set; normally later stops are to the right of earlier ones.

By default any text typeset in a tabbing environment is typeset ragged right and left-aligned on the current tab stop. Typesetting is done in LR mode (see Modes).

The following commands can be used inside a tabbing environment. They are all fragile (see \protect).

\\ (tabbing)

End a tabbed line and typeset it.

\= (tabbing)

Sets a tab stop at the current position.

\> (tabbing)

Advances to the next tab stop.

\<

Put following text to the left of the local margin (without changing the margin). Can only be used at the start of the line.

\+

Moves the left margin of the next and all the following commands one tab stop to the right, beginning tabbed line if necessary.

\-

Moves the left margin of the next and all the following commands one tab stop to the left, beginning tabbed line if necessary.

\' (tabbing)

Moves everything that you have typed so far in the current column, i.e., everything from the most recent \>, \<, \', \\, or \kill command, to the previous column and aligned to the right, flush against the current column’s tab stop.

\` (tabbing)

Allows you to put text flush right against any tab stop, including tab stop 0. However, it can’t move text to the right of the last column because there’s no tab stop there. The \` command moves all the text that follows it, up to the \\ or \end{tabbing} command that ends the line, to the right margin of the tabbing environment. There must be no \> or \' command between the \` and the \\ or \end{tabbing} command that ends the line.

\a (tabbing)

In a tabbing environment, the commands \=, \' and \` do not produce accents as usual (see Accents). Instead, use the commands \a=, \a' and \a`.

\kill

Sets tab stops without producing text. Works just like \\ except that it throws away the current line instead of producing output for it. Any \=, \+ or \- commands in that line remain in effect.

\poptabs

Restores the tab stop positions saved by the last \pushtabs.

\pushtabs

Saves all current tab stop positions. Useful for temporarily changing tab stop positions in the middle of a tabbing environment.

\tabbingsep

Distance of the text moved by \' to left of current tab stop.

This example typesets a Pascal function:

\begin{tabbing}
function \= fact(n : integer) : integer;\\
         \> begin \= \+ \\
               \> if \= n > 1 then \+ \\
                        fact := n * fact(n-1) \- \\
                  else \+ \\
                        fact := 1; \-\- \\
            end;\\
\end{tabbing}

The output looks like this.

function fact(n : integer) : integer;
         begin
               if n > 1 then
                  fact := n * fact(n-1);
               else
                  fact := 1;
         end;

This example is just for illustration of the environment. To actually typeset computer code in typewriter like this, a verbatim environment (see verbatim) would normally be best. For pretty-printed code, there are quite a few packages, including algorithm2e, fancyvrb, listings, and minted.


8.22 table

Synopsis:

\begin{table}[placement]
  table body
  \caption[loftitle]{title}  % optional
  \label{label}              % also optional
\end{table}

A class of floats (see Floats). 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.

This example table environment contains a tabular

\begin{table}
  \centering\small
  \begin{tabular}{ll}
    \multicolumn{1}{c}{\textit{Author}}
      &\multicolumn{1}{c}{\textit{Piece}}  \\ \hline
    Bach            &Cello Suite Number 1  \\
    Beethoven       &Cello Sonata Number 3 \\
    Brahms          &Cello Sonata Number 1
  \end{tabular}
  \caption{Top cello pieces}
  \label{tab:cello}
\end{table}

but you can put many different kinds of content in a table: the table body may contain text, LaTeX commands, graphics, etc. It is typeset in a parbox of width \textwidth.

For the possible values of placement and their effect on the float placement algorithm, see Floats.

The label is optional; it is used for cross references (see Cross references). The \caption command is also optional. It specifies caption text title for the table (see \caption). By default it is numbered. If its optional lottitle is present then that text is used in the list of tables instead of title (see Table of contents, list of figures, list of tables).

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.

\begin{table}[b]
  \centering
  \begin{tabular}{r|p{2in}} \hline
    One &The loneliest number \\
    Two &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}

8.23 tabular

Synopsis:

\begin{tabular}[pos]{cols}
  column 1 entry  &column 2 entry  ...  &column n entry \\
  ...
\end{tabular}

or

\begin{tabular*}{width}[pos]{cols}
  column 1 entry  &column 2 entry  ...  &column n entry \\
  ...
\end{tabular*}

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.

\begin{tabular}{l|l}
  \textit{Player name}  &\textit{Career home runs}  \\ 
  \hline
  Hank Aaron  &755 \\
  Babe Ruth   &714
\end{tabular}

The output will have two left-aligned columns with a vertical bar between them. This is specified in tabular’s argument {l|l}. Put the entries into different columns by separating them with an ampersand, &. The end of each row is marked with a double backslash, \\. Put a horizontal rule below a row, after a double backslash, with \hline. After the last row the \\ is optional, unless an \hline command follows to put a rule below the table.

The required and optional arguments to tabular consist of:

pos

Optional. Specifies the table’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: t aligns the table so its top row matches the baseline of the surrounding text, and b aligns on the bottom row.

This only has an effect if there is other text. In the common case of a tabular alone in a center environment this option makes no difference.

cols

Required. Specifies the formatting of columns. It consists of a sequence of the following specifiers, corresponding to the types of column and intercolumn material.

l

A column of left-aligned items.

r

A column of right-aligned items.

c

A column of centered items.

|

A vertical line the full height and depth of the environment.

@{text or space}

Insert text or space at this location in every row. The text or space material is typeset in LR mode. This text is fragile (see \protect).

If between two column specifiers there is no @-expression then LaTeX’s book, article, and report classes will put on either side of each column a space of width \tabcolsep, which by default is 6pt. That is, by default adjacent columns are separated by 12pt (so \tabcolsep is misleadingly named since it is only half of the separation between tabular columns). In addition, a space of \tabcolsep also comes before the first column and after the final column, unless you put a @{...} there.

If you override the default and use an @-expression then LaTeX does not insert \tabcolsep so you must insert any desired space yourself, as in @{\hspace{1em}}.

An empty expression @{} will eliminate the space. In particular, sometimes you want to eliminate 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.

\begin{flushleft}
  \begin{tabular}{@{}l}
    ...
  \end{tabular}
\end{flushleft}

The next example shows text, a decimal point between the columns, arranged so the numbers in the table are aligned on it.

\begin{tabular}{r@{$.$}l}
  $3$ &$14$  \\
  $9$ &$80665$
\end{tabular}

An \extracolsep{wd} command in an @-expression causes an extra space of width wd to appear to the left of all subsequent columns, until countermanded by another \extracolsep. Unlike ordinary intercolumn space, this extra space is not suppressed by an @-expression. An \extracolsep command can be used only in an @-expression in the cols argument. Below, LaTeX inserts the right amount of intercolumn space to make the entire table 4 inches wide.

\begin{tabular*}{4in}{l@{\extracolsep{\fill}}l}
  Seven times down, eight times up \ldots 
  &such is life!
\end{tabular*}

To insert commands that are automatically executed before a given column, load the array package and use the >{...} specifier.

p{wd}

Each item in the column is typeset in a parbox of width wd, as if it were the argument of a \parbox[t]{wd}{...} command.

A line break double backslash \\ may not appear in the item, except inside an environment like minipage, array, or tabular, or inside an explicit \parbox, or in the scope of a \centering, \raggedright, or \raggedleft declaration (when used in a p-column element these declarations must appear inside braces, as with {\centering .. \\ ..}). Otherwise LaTeX will misinterpret the double backslash as ending the tabular row. Instead, to get a line break in there use \newline (see \newline).

*{num}{cols}

Equivalent to num copies of cols, where num is a positive integer and cols is a list of specifiers. Thus the specifier \begin{tabular}{|*{3}{l|r}|} is equivalent to the specifier \begin{tabular}{|l|rl|rl|r|}. Note that cols may contain another *-expression.

width

Required for tabular*, not allowed for tabular. Specifies the width of the tabular* environment. The space between columns should be rubber, as with @{\extracolsep{\fill}}, to allow the table to stretch or shrink to make the specified width, or else you are likely to get the Underfull \hbox (badness 10000) in alignment ... warning.

Parameters that control formatting:

\arrayrulewidth

A length that is the thickness of the rule created by |, \hline, and \vline in the tabular and array environments. The default is ‘.4pt’. Change it as in \setlength{\arrayrulewidth}{0.8pt}.

\arraystretch

A factor by which the spacing between rows in the tabular and array environments is multiplied. The default is ‘1’, for no scaling. Change it as \renewcommand{\arraystretch}{1.2}.

\doublerulesep

A length that is the distance between the vertical rules produced by the || specifier. The default is ‘2pt’.

\tabcolsep

A length that is half of the space between columns. The default is ‘6pt’. Change it with \setlength.

The following commands can be used inside the body of a tabular environment, the first two inside an entry and the second two between lines:


8.23.1 \multicolumn

Synopsis:

\multicolumn{numcols}{cols}{text}

Make an array or tabular entry that spans several columns. The first argument numcols gives the number of columns to span. The second argument cols specifies the formatting of the entry, with c for centered, l for flush left, or r for flush right. The third argument text gives the contents of that entry.

In this example, in the first row, the second and third columns are spanned by the single heading ‘Name’.

\begin{tabular}{lccl} 
  \textit{ID}       &\multicolumn{2}{c}{\textit{Name}} &\textit{Age} \\
  \hline  
  978-0-393-03701-2 &O'Brian &Patrick                  &55           \\
    ...
\end{tabular}

What counts as a column is: the column format specifier for the array or tabular environment is broken into parts, where each part (except the first) begins with l, c, r, or p. So from \begin{tabular}{|r|ccp{1.5in}|} the parts are |r|, c, c, and p{1.5in}|.

The cols argument overrides the array or tabular environment’s intercolumn area default adjoining this multicolumn entry. To affect that area, this argument can contain vertical bars | indicating the placement of vertical rules, and @{...} expressions. Thus if cols is ‘|c|’ 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.

\begin{tabular}{|cc|c|c|}
  \multicolumn{1}{r}{w}       % entry one
    &\multicolumn{1}{|r|}{x}  % entry two 
    &\multicolumn{1}{|r}{y}   % entry three
    &z                        % entry four
\end{tabular}

Before the first entry the output will not have a vertical rule because the \multicolumn has the cols specifier ‘r’ with no initial vertical bar. Between entry one and entry two there will be a vertical rule; although the first cols does not have an ending vertical bar, the second cols does have a starting one. Between entry two and entry three there is a single vertical rule; despite that the cols in both of the surrounding multicolumn’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 cols in the entry three \multicolumn leaves it out, and that takes precedence. Finally, following entry four there is a vertical rule because of the default.

The number of spanned columns numcols can be 1. Besides giving the ability to change the horizontal alignment, this also is useful to override for one row the tabular definition’s default intercolumn area specification, including the placement of vertical rules.

In the example below, in the tabular definition the first column is specified to default to left justified but in the first row the entry is centered with \multicolumn{1}{c}{\textsc{Period}}. Also in the first row, the second and third columns are spanned by a single entry with \multicolumn{2}{c}{\textsc{Span}}, overriding the specification to center those two columns on the page range en-dash.

\begin{tabular}{l|r@{--}l} 
  \multicolumn{1}{c}{\textsc{Period}}  
    &\multicolumn{2}{c}{\textsc{Span}} \\ \hline
  Baroque          &1600           &1760         \\
  Classical        &1730           &1820         \\
  Romantic         &1780           &1910         \\
  Impressionistic  &1875           &1925
\end{tabular}

Although the tabular specification by default puts a vertical rule between the first and second columns, no such vertical rule appears in the first row here. That’s because there is no vertical bar in the cols part of the first row’s first \multicolumn command.


8.23.2 \vline

Draw a vertical line in a tabular or array environment extending the full height and depth of an entry’s row. Can also be used in an @-expression, although its synonym vertical bar | is more common. This command is rarely used in the body of a table; typically a table’s vertical lines are specified in tabular’s cols argument and overridden as needed with \multicolumn (see tabular).

The example below illustrates some pitfalls. In the first row’s second entry the \hfill moves the \vline 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 {c|cc} specifier coming before the one produced by the \vline\hfill. In contrast, the first row’s third entry shows the usual way to put a vertical bar between two columns. In the second row, the ghi is the widest entry in its column so in the \vline\hfill the \hfill has no effect and the vertical line in that entry appears immediately next to the g, with no whitespace.

\begin{tabular}{c|cc}
  x   &\vline\hfill y   &\multicolumn{1}{|r}{z} \\ % row 1  
  abc &def &\vline\hfill ghi                       % row 2  
\end{tabular}

8.23.3 \cline

Synopsis:

\cline{i-j}

In an array or tabular environment, draw a horizontal rule beginning in column i and ending in column j. The dash, -, must appear in the mandatory argument. To span a single column use the number twice, as with \cline{2-2}.

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.

\begin{tabular}{llrr} 
  a &b &c &d \\ \cline{1-1} \cline{3-4} 
  e &f &g &h 
\end{tabular}

8.23.4 \hline

Draw a horizontal line the width of the enclosing tabular or array environment. It’s most commonly used to draw a line at the top, bottom, and between the rows of a table.

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 \hline, the tabular second row’s line ending double backslash \\ is required.

\begin{tabular}{ll} \hline\hline
  Baseball   &Red Sox  \\
  Basketball &Celtics  \\ \hline
\end{tabular}

8.24 thebibliography

Synopsis:

\begin{thebibliography}{widest-label}
  \bibitem[label]{cite_key}
  ...
\end{thebibliography}

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 Using BibTeX, for a more sophisticated approach.

This shows the environment with two entries.

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}

This styles the first reference as ‘[1] Leslie ...’, and so that ... based on \cite{latexdps} produces the matching ‘... based on [1]’. The second \cite produces ‘[1, 2]’. You must compile the document twice to resolve these references.

The mandatory argument widest-label is text that, when typeset, is as wide as the widest item label produced by the \bibitem commands. The tradition is to use 9 for bibliographies with less than 10 references, 99 for ones with less than 100, etc.

The bibliographic list is headed by a title such as ‘Bibliography’. To change it there are two cases. In the book and report classes, where the top level sectioning is \chapter and the default title is ‘Bibliography’, that title is in the macro \bibname. For article, where the class’s top level sectioning is \section and the default is ‘References’, the title is in macro \refname. Change it by redefining the command, as with \renewcommand{\refname}{Cited references}, after \begin{document}.

Language support packages such as babel will automatically redefine \refname or \bibname to fit the selected language.

See list, for the list layout control parameters.


8.24.1 \bibitem

Synopsis:

\bibitem{cite_key}

or

\bibitem[label]{cite_key}

Generate an entry labeled by default by a number generated using the enumi counter. The citation key cite_key can be any string of letters, numbers, and punctuation symbols (but not comma).

See thebibliography, for an example.

When provided, the optional label becomes the entry label and the enumi counter is not incremented. With this

\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}

the first entry will be styled as ‘[Lamport 1993] Leslie ...’ (The amount of horizontal space that LaTeX leaves for the label depends on the widest-label argument of the thebibliography environment; see thebibliography.) Similarly, ... based on \cite{latexdps} will produce ‘... based on [Lamport 1994]’.

If you mix \bibitem entries having a label with those that do not then LaTeX will number the unlabelled ones sequentially. In the example above the texbook entry will appear as ‘[1] Donald ...’, despite that it is the second entry.

If you use the same cite_key twice then you get ‘LaTeX Warning: There were multiply-defined labels’.

Under the hood, LaTeX remembers the cite_key and label information because \bibitem writes it to the auxiliary file jobname.aux (see Jobname). For instance, the above example causes the two \bibcite{latexdps}{Lamport, 1993} and \bibcite{texbook}{1} to appear in that file. The .aux file is read by the \begin{document} command and then the information is available for \cite commands. This explains why you need to run LaTeX twice to resolve references: once to write it out and once to read it in.

Because of this two-pass algorithm, when you add a \bibitem or change its cite_key you may get ‘LaTeX Warning: Label(s) may have changed. Rerun to get cross-references right’. Fix it by recompiling.


8.24.2 \cite

Synopsis:

\cite{keys}

or

\cite[subcite]{keys}

Generate as output a citation to the references associated with keys. The mandatory keys is a citation key, or a comma-separated list of citation keys (see \bibitem).

This

The ultimate source is \cite{texbook}.
  ...
\begin{thebibliography}
\bibitem{texbook} 
  Donald Ervin Knuth. 
  \textit{The \TeX book}. 
  Addison-Wesley, Reading, Massachusetts, 1983.
\end{thebibliography}

produces output like ‘... source is [1]’. You can change the appearance of the citation and of the reference by using bibliography styles if you generate automatically the thebibliography environment. More information in Using BibTeX.

The optional argument subcite is appended to the citation. For example, See 14.3 in \cite[p.~314]{texbook} might produce ‘See 14.3 in [1, p. 314]’.

In addition to what appears in the output, \cite writes information to the auxiliary file jobname.aux (see Jobname). For instance, \cite{latexdps} writes ‘\citation{latexdps}’ to that file. This information is used by BibTeX to include in your reference list only those works that you have actually cited; see \nocite also.

If keys is not in your bibliography information then you get ‘LaTeX Warning: There were undefined references’, 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 \cite{texbok} 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 .aux file (see \bibitem) then the fix may be to run LaTeX again.


8.24.3 \nocite

Synopsis:

\nocite{keys}

Produces no output but writes keys to the auxiliary file jobname.aux (see Jobname).

The mandatory argument keys is a comma-separated list of one or more citation keys (see \bibitem). This information is used by BibTeX to include these works in your reference list even though you have not explicitly cited them (see \cite).


8.24.4 Using BibTeX

As described in thebibliography (see thebibliography), a sophisticated approach to managing bibliographies is provided by the BibTeX program. This is only an introduction; see the full documentation on CTAN (see CTAN: The Comprehensive TeX Archive Network).

With BibTeX, you don’t use the thebibliography environment directly (see thebibliography). Instead, include these lines:

\bibliographystyle{bibstyle}
\bibliography{bibfile1, bibfile2, ...}

The bibstyle refers to a file bibstyle.bst, which defines how your citations will look. The standard bibstyle’s distributed with BibTeX are:

alpha

Labels are formed from name of author and year of publication. The bibliographic items are sorted alphabetically.

plain

Labels are integers. Sort the bibliographic items alphabetically.

unsrt

Like plain, but entries are in order of citation.

abbrv

Like plain, but more compact labels.

Many, many other BibTeX style files exist, tailored to the demands of various publications. See the CTAN topic https://ctan.org/topic/bibtex-sty.

The \bibliography command is what actually produces the bibliography. Its argument is a comma-separated list, referring to files named bibfile1.bib, bibfile2.bib, … These contain your database in BibTeX format. This shows a typical couple of entries in that format.

@book{texbook,
  title     = {The {{\TeX}}book},
  author    = {D.E. Knuth},
  isbn      = {0201134489},
  series    = {Computers \& 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}
}

Only the bibliographic entries referred to via \cite and \nocite will be listed in the document’s bibliography. Thus you can keep all your sources together in one file, or a small number of files, and rely on BibTeX to include in this document only those that you used.

With BibTeX, the keys argument to \nocite can also be the single character ‘*’. This means to implicitly cite all entries from all given bibliographies.


8.24.4.1 BibTeX error messages

If you forget to use \bibliography or \bibliographystyle in your document (or, less likely, any \cite or \nocite command), BibTeX will issue an error message. Because BibTeX can be used with any program, not just LaTeX, the error messages refer to the internal commands read by BibTeX (from the .aux file), rather than the user-level commands described above.

Here is a table showing internal commands mentioned in the BibTeX errors, and the corresponding user-level commands.

\bibdata

\bibliography

\bibstyle

\bibliographystyle

\citation

\cite, \nocite

For example, if your document has no \bibliographystyle command, BibTeX complains as follows:

I found no \bibstyle command---while reading file document.aux

8.25 theorem

Synopsis:

\begin{theorem}
  theorem body
\end{theorem}

Produces ‘Theorem n’ in boldface followed by theorem body in italics. The numbering possibilities for n are described under \newtheorem (see \newtheorem).

\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}

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


8.26 titlepage

Synopsis:

\begin{titlepage}
  ... text and spacing ...
\end{titlepage}

Create a title page, a page with no printed page number or heading and with succeeding pages numbered starting with page one.

In this example all formatting, including vertical spacing, is left to the author.

\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}

To instead produce a standard title page without a titlepage environment, use \maketitle (see \maketitle).


8.27 verbatim

Synopsis:

\begin{verbatim}
literal-text
\end{verbatim}

A paragraph-making environment in which LaTeX produces as output exactly what you type as input. For instance inside literal-text the backslash \ character does not start commands, it produces a printed ‘\’, and carriage returns and blanks are taken literally. The output appears in a monospaced typewriter-like font (\tt).

\begin{verbatim}
Symbol swearing: %&$#?!.
\end{verbatim}

The only restriction on literal-text is that it cannot include the string \end{verbatim}.

You cannot use the verbatim environment in the argument to macros, for instance in the argument to a \section. This is not the same as commands being fragile (see \protect), instead it just cannot work, as the verbatim environment changes the catcode regime before processing its contents, and restore it immediately afterward, nevertheless with a macro argument the content of the argument has already be converted to a token list along the catcode regime in effect when the macro was called. However, the cprotect package can help with this.

One common use of verbatim input is to typeset computer code. Some packages offer many features not provided by the verbatim environment; two of the most popular are listings and minted. For example, they are capable of pretty-printing, line numbering, and selecting parts of files for a continuing listing.

A package that provides many more options for verbatim environments is fancyvrb. Another is verbatimbox.

For a list of all the relevant packages, see CTAN (see CTAN: The Comprehensive TeX Archive Network), particularly the topics listing (https://ctan.org/topic/listing) and verbatim (https://ctan.org/topic/verbatim).


8.27.1 \verb

Synopsis:

\verb char literal-text char
\verb* char literal-text char

Typeset literal-text as it is input, including special characters and spaces, using the typewriter (\tt) font.

This example shows two different invocations of \verb.

This is \verb!literally! the biggest pumpkin ever.
And this is the best squash, \verb+literally!+

The first \verb has its literal-text surrounded with exclamation point, !. The second instead uses plus, +, because the exclamation point is part of literal-text.

The single-character delimiter char surrounds literal-text—it must be the same character before and after. No spaces come between \verb or \verb* and char, or between char and literal-text, or between literal-text and the second occurrence of char (the synopsis shows a space only to distinguish one component from the other). The delimiter must not appear in literal-text. The literal-text cannot include a line break.

The *-form differs only in that spaces are printed with a visible space character.

The output from this will include a visible space on both side of word ‘with’:

The command's first argument is \verb*!filename with extension! and ...

For typesetting Internet addresses, urls, the package url is a better option than the \verb command, since it allows line breaks.

You cannot use \verb in the argument to a macro, for instance in the argument to a \section. It is not a question of \verb being fragile (see \protect), instead it just cannot work, as the \verb command changes the catcode regime before reading its argument, and restore it immediately afterward, nevertheless with a macro argument the content of the argument has already be converted to a token list along the catcode regime in effect when the macro was called. However, the cprotect package can help with this.


8.28 verse

Synopsis:

\begin{verse}
  line1 \\
  line2 \\
  ...
\end{verse}

An environment for poetry.

Here are two lines from Shakespeare’s Romeo and Juliet.

Then plainly know my heart's dear love is set \\
On the fair daughter of rich Capulet.

Separate the lines of each stanza with \\, and use one or more blank lines to separate the stanzas.

\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}

The output has margins indented on the left and the right, paragraphs are not indented, and the text is not right-justified.


9 Line breaking

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).

LaTeX usually does the line (and page) breaking in the text body for you but in some environments you manually force line breaks.

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.


9.1 \\

Synopsis, one of:

\\
\\[morespace]

or one of:

\\*
\\*[morespace]

End the current line. The optional argument morespace specifies extra vertical space to be inserted before the next line. This is a rubber length (see Lengths) 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 \protect).

\title{My story: \\[0.25in]
       a tale of woe}

The starred form, \\*, tells LaTeX not to start a new page between the two lines, by issuing a \nobreak.

Explicit line breaks in the main text body are unusual in LaTeX. In particular, don’t start new paragraphs with \\. Instead leave a blank line between the two paragraphs. And don’t put in a sequence of \\’s to make vertical space. Instead use \vspace{length}, or \leavevmode\vspace{length}, or \vspace*{length} if you want the space to not be thrown out at the top of a new page (see \vspace).

The \\ command is mostly used outside of the main flow of text such as in a tabular or array environment or in an equation environment.

The \\ command is a synonym for \newline (see \newline) under ordinary circumstances (an example of an exception is the p{...} column in a tabular environment; see tabular).

The \\ command is a macro, and its definition changes by context so that its definition in normal text, a center environment, a flushleft environment, and a tabular are all different. In normal text when it forces a linebreak it is essentially a shorthand for \newline. 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.

You get ‘LaTeX Error: There's no line here to end’ if you use \\ to ask for a new line, rather than to end the current line. An example is if you have \begin{document}\\ or, more likely, something like this.

\begin{center}
  \begin{minipage}{0.5\textwidth}
  \\
  In that vertical space put your mark.
  \end{minipage}
\end{center}

Fix it by replacing the double backslash with something like \vspace{\baselineskip}.


9.2 \obeycr & \restorecr

The \obeycr command makes a return in the input file (‘^^M’, internally) the same as \\, followed by \relax. So each new line in the input will also be a new line in the output. The \restorecr command restores normal line-breaking behavior.

This is not the way to show verbatim text or computer code. Use verbatim (see verbatim) instead.

With LaTeX’s usual defaults, this

aaa
bbb

\obeycr
ccc
ddd
   eee

\restorecr
fff
ggg

hhh
iii

produces output like this.

  aaa bbb
  ccc
ddd
eee

fff ggg
  hhh iii

The indents are paragraph indents.


9.3 \newline

In ordinary text, this ends a line in a way that does not right-justify it, so the text before the end of line is not stretched. That is, in paragraph mode (see Modes), the \newline command is equivalent to double-backslash (see \\). This command is fragile (see \protect).

However, the two commands are different inside a tabular or array environment. In a column with a specifier producing a paragraph box such as typically p{...}, \newline 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 \\ or its equivalent \tabularnewline.

This will print ‘Name:’ and ‘Address:’ as two lines in a single cell of the table.

\begin{tabular}{p{1in}@{\hspace{2in}}p{1in}}
  Name: \newline Address: &Date: \\ \hline
\end{tabular}

The ‘Date:’ will be baseline-aligned with ‘Name:’.


9.4 \- (discretionary hyphen)

Tell LaTeX that it may hyphenate the word at that point. When you insert \- commands in a word, the word will only be hyphenated at those points and not at any of the other hyphenation points that LaTeX might otherwise have chosen. This command is robust (see \protect).

LaTeX is good at hyphenating and usually finds most of the correct hyphenation points, while almost never using an incorrect one. The \- command is for exceptional cases.

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.

\begin{tabular}{rp{1.75in}}
  Isaac Asimov &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}

Commenting out the third line and uncommenting the fourth makes a much better fit.

The \- command only allows LaTeX to break there, it does not require that it break there. You can force a split with something like Hef-\linebreak feron. Of course, if you later change the text then this forced break may look out of place, so this approach requires care.


9.5 \slash: breakable ‘/

The \slash command produces a ‘/’ character and then a penalty of the same value as an explicit ‘-’ character (\exhyphenpenalty). This allows TeX to break a line at the ‘/’, similar to a hyphen. Hyphenation is allowed in the word part preceding the ‘/’, but not after. For example:

The input\slash output of the program is complicated.

9.6 \discretionary (generalized hyphenation point)

Synopsis:

\discretionary{pre-break}{post-break}{no-break}

Handle word changes around hyphens. This command is not often used in LaTeX documents.

If a line break occurs at the point where \discretionary appears then TeX puts pre-break at the end of the current line and puts post-break at the start of the next line. If there is no line break here then TeX puts no-break.

In ‘difficult’ the three letters ffi form a ligature. But TeX can nonetheless break between the two ‘f’’s with this.

di\discretionary{f-}{fi}{ffi}cult

Note that users do not have to do this. It is typically handled automatically by TeX’s hyphenation algorithm.


9.7 \fussy & \sloppy

Declarations to make TeX more picky or less picky about line breaking. Declaring \fussy usually avoids too much space between words, at the cost of an occasional overfull box. Conversely, \sloppy avoids overfull boxes while suffering loose interword spacing.

The default is \fussy. Line breaking in a paragraph is controlled by whichever declaration is current at the end of the paragraph, i.e., at the blank line or \par or displayed equation ending that paragraph. So to affect the line breaks, include that paragraph-ending material in the scope of the command.


9.7.1 sloppypar

Synopsis:

\begin{sloppypar}
  ... paragraphs ...
\end{sloppypar}

Typeset the paragraphs with \sloppy in effect (see \fussy & \sloppy). Use this to locally adjust line breaking, to avoid ‘Overfull box’ or ‘Underfull box’ errors.

The example is simple.

\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}

9.8 \hyphenation

Synopsis:

\hyphenation{word1 ...}

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 an ASCII dash character, -.

Here is an example:

\hyphenation{hat-er il-lit-e-ra-ti tru-th-i-ness}

Use lowercase letters. TeX will only hyphenate if the word matches exactly; no inflections are tried. Multiple \hyphenation commands accumulate.


9.9 \linebreak & \nolinebreak

Synopses, one of:

\linebreak
\linebreak[zero-to-four]

or one of these.

\nolinebreak
\nolinebreak[zero-to-four]

Encourage or discourage a line break. The optional zero-to-four is an integer lying between 0 and 4 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, \nolinebreak[1] is a suggestion that another place may be better. The higher the number, the more insistent the request. Both commands are fragile (see \protect).

Here we tell LaTeX that a good place to put a linebreak is after the standard legal text.

\boilerplatelegal{} \linebreak[2]
We especially encourage applications from members of traditionally
underrepresented groups.

When you issue \linebreak, the spaces in the line are stretched out so that the break point reaches the right margin. See \\ and \newline, to have the spaces not stretched out.


10 Page breaking

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.

LaTeX’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.

But LaTeX does not optimize over the entire document’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’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.

See Layout, for more material that is relevant to page breaking.


10.1 \clearpage & \cleardoublepage

Synopsis:

\clearpage

or

\cleardoublepage

End the current page and output all of the pending floating figures and tables (see Floats). 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, \cleardoublepage also makes the next page of content a right-hand page, an odd-numbered page, if necessary inserting a blank page. The \clearpage command is robust while \cleardoublepage is fragile (see \protect).

LaTeX’s page breaks are optimized so ordinarily you only use this command in a document body to polish the final version, or inside commands.

The \cleardoublepage 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.

\let\origdoublepage\cleardoublepage
\newcommand{\clearemptydoublepage}{%
  \clearpage
  {\pagestyle{empty}\origdoublepage}%
}

If you want LaTeX’s standard \chapter command to do this then add the line \let\cleardoublepage\clearemptydoublepage. (Of course this affects all uses of \cleardoublepage, not just the one in \chapter.)

The command \newpage (see \newpage) also ends the current page, but without clearing pending floats. And, if LaTeX is in two-column mode then \newpage ends the current column while \clearpage and \cleardoublepage end the current page.


10.2 \newpage

Synopsis:

\newpage

End the current page. This command is robust (see \protect).

LaTeX’s page breaks are optimized so ordinarily you only use this command in a document body to polish the final version, or inside commands.

While the commands \clearpage and \cleardoublepage also end the current page, in addition they clear pending floats (see \clearpage & \cleardoublepage). And, if LaTeX is in two-column mode then \clearpage and \cleardoublepage end the current page, possibly leaving an empty column, while \newpage only ends the current column.

In contrast with \pagebreak (see \pagebreak & \nopagebreak), the \newpage command will cause the new page to start right where requested. This

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.

makes a new page start after ‘continent’, and the cut-off line is not right justified. In addition, \newpage does not vertically stretch out the page, as \pagebreak does.


10.3 \enlargethispage

Synopsis, one of:

\enlargethispage{size}
\enlargethispage*{size}

Enlarge the \textheight for the current page. The required argument size must be a rigid length (see Lengths). It may be positive or negative. This command is fragile (see \protect).

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.

This will allow one extra line on the current page.

\enlargethispage{\baselineskip}

The starred form \enlargesthispage* 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 \pagebreak.


10.4 \pagebreak & \nopagebreak

Synopses:

\pagebreak
\pagebreak[zero-to-four]

or

\nopagebreak
\nopagebreak[zero-to-four]

Encourage or discourage a page break. The optional zero-to-four 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 \nopagebreak[1] suggests to LaTeX that another spot might be preferable. The higher the number, the more insistent the request. Both commands are fragile (see \protect).

LaTeX’s page endings are optimized so ordinarily you only use these commands in a document body to polish the final version, or inside commands.

If you use these inside a paragraph, they apply to the point following the line in which they appear. So this

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.

does not give a page break at ‘continent’, but instead at ‘nation’, since that is where LaTeX breaks that line. In addition, with \pagebreak 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 \flushbottom is in effect this can cause you to get ‘Underfull \vbox (badness 10000) has occurred while \output is active’. See \newpage, for a command that does not have these effects.

A declaration \samepage and corresponding samepage environment try to only allow breaks between paragraphs. They are not perfectly reliable. For more on keeping material on the same page, see the FAQ entry https://texfaq.org/FAQ-nopagebrk.)


11 Footnotes

Place a footnote at the bottom of the current page, as here.

Noël 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.}

You can put multiple footnotes on a page. If the footnote text becomes too long then it will flow to the next page.

You can also produce footnotes by combining the \footnotemark and the \footnotetext commands, which is useful in special circumstances.

To make bibliographic references come out as footnotes you need to include a bibliographic style with that behavior (see Using BibTeX).


11.1 \footnote

Synopsis, one of:

\footnote{text}
\footnote[number]{text}

Place a footnote text at the bottom of the current page, with a footnote marker at the current position in the text.

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.''}

The optional argument number allows you to specify the number of the footnote. If you use this then LaTeX does not increment the footnote counter.

By default, LaTeX uses arabic numbers as footnote markers. Change this with something like \renewcommand{\thefootnote}{\fnsymbol{footnote}}, which uses a sequence of symbols (see \alph \Alph \arabic \roman \Roman \fnsymbol: Printing counters). 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 \setcounter{footnote}{0}.

LaTeX determines the spacing of footnotes with two parameters.

\footnoterule

Produces the rule separating the main text on a page from the page’s footnotes. Default dimensions in the standard document classes (except slides, where it does not appear) are: vertical thickness of 0.4pt, and horizontal size of 0.4\columnwidth long. Change the rule with something like this.

% \footnoterule is expanded in vertical mode, thus \kern
% commands ensure that no vertical space is created,
% and the rule is separated vertically with 2pt
% above the note text.
\renewcommand*{\footnoterule}{%
  \kern -3pt                         % This -3 is negative
  \hrule width \textwidth height 1pt % of the sum of this 1
  \kern 2pt}                         % and this 2
\footnotesep

The height of the strut placed at the beginning of the footnote (see \strut). By default, this is set to the normal strut for \footnotesize fonts (see Font sizes), therefore there is no extra space between footnotes. This is ‘6.65pt’ for ‘10pt’, ‘7.7pt’ for ‘11pt’, and ‘8.4pt’ for ‘12pt’. Change it as with \setlength{\footnotesep}{11pt}.

The \footnote command is fragile (see \protect).

LaTeX’s default puts many restrictions on where you can use a \footnote; for instance, you cannot use it in an argument to a sectioning command such as \chapter (it can only be used in outer paragraph mode; see Modes). There are some workarounds; see following sections.

In a minipage environment the \footnote command uses the mpfootnote counter instead of the footnote 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 minipage and Footnotes in a table.


11.2 \footnotemark

Synopsis, one of:

\footnotemark
\footnotemark[number]

Put the current footnote mark in the text. To specify associated text for the footnote see \footnotetext. The optional argument number causes the command to use that number to determine the footnote mark. This command can be used in inner paragraph mode (see Modes).

If you use \footnotemark without the optional argument then it increments the footnote counter, but if you use the optional number then it does not. The next example produces several consecutive footnote markers referring to the same footnote.

The first theorem\footnote{Due to Gauss.} 
and the second theorem\footnotemark[\value{footnote}] 
and the third theorem.\footnotemark[\value{footnote}]

If there are intervening footnotes then you must remember the value of the number of the common mark. This example gives the same institutional affiliation to both the first and third authors (\thanks is a version of \footnote), by explicitly specifying the number of the footnote (‘1’).

\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

This example accomplishes the same by using the package cleveref.

\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}

It will work with the package hyperref.

This uses a counter to remember the footnote number. The third sentence is followed by the same footnote marker as the first.

\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.

11.3 \footnotetext

Synopsis, one of:

\footnotetext{text}
\footnotetext[number]{text}

Place text at the bottom of the page as a footnote. It pairs with \footnotemark (see \footnotemark) and can come anywhere after that command, but must appear in outer paragraph mode (see Modes). The optional argument number changes the number of the footnote mark.

See \footnotemark and Footnotes in a table, for usage examples.


11.4 Footnotes in section headings

Putting a footnote in a section heading, as in:

\section{Full sets\protect\footnote{This material due to ...}}

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 \section.

\section[Please]{Please\footnote{%
  Don't footnote in chapter and section headers!}}

No \protect is needed in front of \footnote here because what gets moved to the table of contents is the optional argument.


11.5 Footnotes in a table

Inside a tabular or array environment the \footnote 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 minipage environment as here (see minipage).

\begin{center}
  \begin{minipage}{\textwidth} \centering
     \begin{tabular}{l|l}
       \textsc{Ship}           &\textsc{Book} \\ \hline 
       \textit{HMS Sophie}     &Master and Commander  \\ 
       \textit{HMS Polychrest} &Post Captain  \\  
       \textit{HMS Lively}     &Post Captain \\
       \textit{HMS Surprise}   &A number of books\footnote{%
                                  Starting with \textit{HMS Surprise}.}
     \end{tabular}
  \end{minipage}                              
\end{center}

Inside a minipage, footnote marks are lowercase letters. Change that with something like \renewcommand{\thempfootnote}{\arabic{mpfootnote}} (see \alph \Alph \arabic \roman \Roman \fnsymbol: Printing counters).

The footnotes in the prior example appear at the bottom of the minipage. To have them appear at the bottom of the main page, as part of the regular footnote sequence, use the \footnotemark and \footnotetext pair and make a new counter.

\newcounter{mpFootnoteValueSaver}
\begin{center}
  \begin{minipage}{\textwidth}
    \setcounter{mpFootnoteValueSaver}{\value{footnote}} \centering
     \begin{tabular}{l|l}
       \textsc{Woman}             &\textsc{Relationship} \\ \hline 
       Mona                       &Attached\footnotemark  \\ 
       Diana Villiers             &Eventual wife  \\  
       Christine Hatherleigh Wood &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.}
\end{center}

For a floating table environment (see table), use the tablefootnote package.

\usepackage{tablefootnote}  % in preamble
   ...
\begin{table}
  \centering
     \begin{tabular}{l|l}
     \textsc{Date}  &\textsc{Campaign} \\ \hline 
     1862           &Fort Donelson \\ 
     1863           &Vicksburg     \\  
     1865           &Army of Northern Virginia\tablefootnote{%
                      Ending the war.}
     \end{tabular}
    \caption{Forces captured by US Grant}
\end{table}

The footnote appears at the page bottom and is numbered in sequence with other footnotes.


11.6 Footnotes of footnotes

Particularly in the humanities, authors can have multiple classes of footnotes, including having footnotes of footnotes. The package bigfoot extends LaTeX’s default footnote mechanism in many ways, including allow these two, as in this example.

\usepackage{bigfoot}    % in preamble
\DeclareNewFootnote{Default}
\DeclareNewFootnote{from}[alph]   % create class \footnotefrom{}
 ...
The third theorem is a partial converse of the 
second.\footnotefrom{%
  Noted in Wilson.\footnote{Second edition only.}}

12 Definitions

LaTeX has support for making new commands of many different kinds.


12.1 \newcommand & \renewcommand

Synopses, one of (three regular forms, three starred forms):

\newcommand{\cmd}{defn}
\newcommand{\cmd}[nargs]{defn}
\newcommand{\cmd}[nargs][optargdefault]{defn}
\newcommand*{\cmd}{defn}
\newcommand*{\cmd}[nargs]{defn}
\newcommand*{\cmd}[nargs][optargdefault]{defn}

or the same six possibilities with \renewcommand instead of \newcommand:

\renewcommand{\cmd}{defn}
\renewcommand{\cmd}[nargs]{defn}
\renewcommand{\cmd}[nargs][optargdefault]{defn}
\renewcommand*{\cmd}{defn}
\renewcommand*{\cmd}[nargs]{defn}
\renewcommand*{\cmd}[nargs][optargdefault]{defn}

Define or redefine a command (see also \DeclareRobustCommand in Class and package commands).

The starred form of these two forbids the arguments from containing multiple paragraphs of text (i.e., a \par token; in plain TeX terms: the commands are not \long). With the default form, arguments can be multiple paragraphs.

These are the parameters (examples follow):

cmd

Required; \cmd is the command name. It must begin with a backslash, \, and must not begin with the four character string \end. For \newcommand, it must not be already defined. For \renewcommand, this name must already be defined.

nargs

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.

optargdefault

Optional; if this argument is present then the first argument of \cmd is optional, with default value optargdefault (which may be the empty string). If optargdefault is not present then \cmd does not take an optional argument.

That is, if \cmd is called with a following argument in square brackets, as in \cmd[optval]{...}..., then within defn the parameter #1 is set to optval. On the other hand, if \cmd is called without following square brackets then within defn the parameter #1 is set to optargdefault. In either case, the required arguments start with #2.

Omitting [optargdefault] from the definition is entirely different from giving the square brackets with empty contents, as in []. The former says the command being defined takes no optional argument, so #1 is the first required argument (if nargs ≥ 1); the latter sets the optional argument #1 to the empty string as the default, if no optional argument was given in the call.

Similarly, omitting [optval] from a call is also entirely different from giving the square brackets with empty contents. The former sets #1 to the value of optval (assuming the command was defined to take an optional argument); the latter sets #1 to the empty string, just as with any other value.

If a command is not defined to take an optional argument, but is called with an optional argument, the results are unpredictable: there may be a LaTeX error, there may be incorrect typeset output, or both.

defn

Required; the text to be substituted for every occurrence of \cmd. The parameters #1, #2, …, #nargs are replaced by the values supplied when the command is called (or by optargdefault in the case of an optional argument not specified in the call, as just explained).

TeX ignores blanks in the source following a control word (see Control sequence, control word and control symbol), as in ‘\cmd ’. If you want a space there, one solution is to type {} after the command (‘\cmd{} ’), and another solution is to use an explicit control space (‘\cmd\ ’).

A simple example of defining a new command: \newcommand{\RS}{Robin Smith} results in \RS being replaced by the longer text. Redefining an existing command is similar: \renewcommand{\qedsymbol}{{\small QED}}.

If you use \newcommand and the command name has already been used then you get something like ‘LaTeX Error: Command \fred already defined. Or name \end... illegal, see p.192 of the manual’. Similarly, If you use \renewcommand and the command name has not been defined then you get something like ‘LaTeX Error: \hank undefined’.

Here the first definition creates a command with no arguments, and the second, a command with one required argument:

\newcommand{\student}{Ms~O'Leary}
\newcommand{\defref}[1]{Definition~\ref{#1}}

Use the first as in I highly recommend \student{} to you. The second has a variable argument, so that \defref{def:basis} expands to Definition~\ref{def:basis}, which ultimately expands to something like ‘Definition~3.14’.

Similarly, but with two required arguments: \newcommand{\nbym}[2]{$#1 \times #2$} is invoked as \nbym{2}{k}.

This example has an optional argument.

\newcommand{\salutation}[1][Sir or Madam]{Dear #1:}

Then \salutation gives ‘Dear Sir or Madam:’ while \salutation[John] gives ‘Dear John:’. And \salutation[] gives ‘Dear :’.

This example has an optional argument and two required arguments.

\newcommand{\lawyers}[3][company]{#2, #3, and~#1}
I employ \lawyers[Howe]{Dewey}{Cheatem}.

The output is ‘I employ Dewey, Cheatem, and Howe.’. The optional argument, Howe, is associated with #1, while Dewey and Cheatem are associated with #2 and #3. Because of the optional argument, \lawyers{Dewey}{Cheatem} will give the output ‘I employ Dewey, Cheatem, and company.’.

The braces around defn do not define a group, that is, they do not delimit the scope of the result of expanding defn. For example, with \newcommand{\shipname}[1]{\it #1}, in this sentence,

The \shipname{Monitor} met the \shipname{Merrimac}.

the words ‘met the’, and the period, would incorrectly be in italics. The solution is to put another pair of braces inside the definition: \newcommand{\shipname}[1]{{\it #1}}.


12.1.1 Control sequence, control word and control symbol

When reading input TeX converts the stream of read characters into a sequence of tokens. When TeX sees a backslash \, it will handle the following characters in a special way in order to make a control sequence token.

The control sequences fall into two categories:

  • control word, when the control sequence is gathered from a \ followed by at least one ASCII letter (A-Z and a-z), followed by at least one non-letter.
  • control symbol, when the control sequence is gathered from a \ followed by one non-letter character.

The sequence of characters so found after the \ is also called the control sequence name.

Blanks after a control word are ignored and do not produce any whitespace in the output (see \newcommand & \renewcommand and Backslash-space, \ ).

Just as the \relax command does nothing, the following input will simply print ‘Hello!’ :

Hel\relax   
   lo!

This is because blanks after \relax, including the newline, are ignored, and blanks at the beginning of a line are also ignored (see Leading blanks).


12.2 \providecommand

Synopses, one of:

\providecommand{\cmd}{defn}
\providecommand{\cmd}[nargs]{defn}
\providecommand{\cmd}[nargs][optargdefault]{defn}
\providecommand*{\cmd}{defn}
\providecommand*{\cmd}[nargs]{defn}
\providecommand*{\cmd}[nargs][optargdefault]{defn}

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 \newcommand. 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 \newcommand & \renewcommand, for the description of the arguments.

This example

\providecommand{\myaffiliation}{Saint Michael's College}
\providecommand{\myaffiliation}{Lyc\'ee Henri IV}
From \myaffiliation.

outputs ‘From Saint Michael's College.’. Unlike \newcommand, the repeated use of \providecommand to (try to) define \myaffiliation does not give an error.


12.3 \makeatletter & \makeatother

Synopsis:

\makeatletter
  ... definition of commands with @ in their name ..
\makeatother

Use this pair when you redefine LaTeX commands that are named with an at-sign character ‘@’. The \makeatletter declaration makes the at-sign character have the category code of a letter, code 11. The \makeatother declaration sets the category code of the at-sign to code 12, its default value.

As TeX reads characters, it assigns each one a category code, or catcode. For instance, it assigns the backslash character ‘\’ the catcode 0. Command names consist of a category 0 character, ordinarily backslash, followed by letters, category 11 characters (except that a command name can also consist of a category 0 character followed by a single non-letter symbol).

LaTeX’s source code has the convention that some commands use @ in their name. These commands are mainly intended for package or class writers. The convention prevents authors who are just using a package or class from accidentally replacing such a command with one of their own, because by default the at-sign has catcode 12.

Use the pair \makeatletter and \makeatother inside a .tex file, typically in the preamble, when you are defining or redefining commands named with @, by having them surround your definition. Don’t use these inside .sty or .cls files since the \usepackage and \documentclass commands already arrange that the at-sign has the character code of a letter, catcode 11.

For a comprehensive list of macros with an at-sign in their names see https://ctan.org/pkg/macros2e.

In this example the class file has a command \thesis@universityname that the user wants to change. These three lines should go in the preamble, before the \begin{document}.

\makeatletter
\renewcommand{\thesis@universityname}{Saint Michael's College}
\makeatother

12.4 \@ifstar

Synopsis:

\newcommand{\mycmd}{\@ifstar{\mycmd@star}{\mycmd@nostar}}
\newcommand{\mycmd@nostar}[nostar-num-args]{nostar-body} 
\newcommand{\mycmd@star}[star-num-args]{star-body}

Many standard LaTeX environments or commands have a variant with the same name but ending with a star character *, an asterisk. Examples are the table and table* environments and the \section and \section* commands.

When defining environments, following this pattern is straightforward because \newenvironment and \renewenvironment allow the environment name to contain a star. So you just have to write \newenvironment{myenv} or \newenvironment{myenv*} and continue the definition as usual. For commands the situation is more complex as the star not being a letter cannot be part of the command name. As in the synopsis above, there will be a user-called command, given above as \mycmd, which peeks ahead to see if it is followed by a star. For instance, LaTeX does not really have a \section* command; instead, the \section 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 \mycmd@nostar and \mycmd@star. 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 an at-sign @ in its name must be enclosed inside a \makeatletter ... \makeatother block (see \makeatletter & \makeatother).

This example of \@ifstar defines the command \ciel and a variant \ciel*. Both have one required argument. A call to \ciel{blue} will return "not starry blue sky" while \ciel*{night} will return "starry night sky".

\makeatletter
\newcommand*{\ciel@unstarred}[1]{not starry #1 sky}
\newcommand*{\ciel@starred}[1]{starry #1 sky}
\newcommand*{\ciel}{\@ifstar{\ciel@starred}{\ciel@unstarred}}
\makeatother

In the next example, the starred variant takes a different number of arguments than the unstarred one. With this definition, Agent 007’s ``My name is \agentsecret*{Bond}, \agentsecret{James}{Bond}.'' is equivalent to entering the commands ``My name is \textsc{Bond}, \textit{James} textsc{Bond}.''

\newcommand*{\agentsecret@unstarred}[2]{\textit{#1} \textsc{#2}}
\newcommand*{\agentsecret@starred}[1]{\textsc{#1}}
\newcommand*{\agentsecret}{%
  \@ifstar{\agentsecret@starred}{\agentsecret@unstarred}}

After a command name, a star is handled similarly to an optional argument. (This differs from environment names in which the star is part of the name itself and as such could be in any position.) Thus, it is technically possible to put any number of spaces between the command and the star. Thus \agentsecret*{Bond} and \agentsecret *{Bond} are equivalent. However, the standard practice is not to insert any such spaces.

There are two alternative ways to accomplish the work of \@ifstar. (1) The suffix package allows the construct \newcommand\mycommand{unstarred-variant} followed by \WithSuffix\newcommand\mycommand*{starred-variant}. (2) LaTeX provides the xparse package, which allows this code:

\NewDocumentCommand\foo{s}{\IfBooleanTF#1
  {starred-variant}%
  {unstarred-variant}% 
  }

12.5 \newcounter: Allocating a counter

Synopsis, one of:

\newcounter{countername}
\newcounter{countername}[supercounter]

Globally defines a new counter named countername and initialize it to zero (see Counters).

The name countername must consist of letters only. It does not begin with a backslash. This name must not already be in use by another counter.

When you use the optional argument [supercounter] then the counter countername will be reset to zero whenever supercounter is incremented. For example, ordinarily subsection is numbered within section so that any time you increment section, either with \stepcounter (see \stepcounter) or \refstepcounter (see \refstepcounter), then LaTeX will reset subsection to zero.

This example

\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}.

produces ‘The value of asuper is 1 and that of asub is 3’ and ‘Now asuper is 2 while asub is 0’.

If the counter already exists, for instance by entering asuper twice, then you get something like ‘LaTeX Error: Command \c@asuper already defined. Or name \end... illegal, see p.192 of the manual.’.

If you use the optional argument then the super counter must already exist. Entering \newcounter{jh}[lh] when lh is not a defined counter will get you ‘LaTeX Error: No counter 'lh' defined.


12.6 \newlength

Synopsis:

\newlength{\len}

Allocate a new length register (see Lengths). The required argument \len has to be a control sequence (see Control sequence, control word and control symbol), and as such must begin with a backslash, \ under normal circumstances. The new register holds rubber lengths such as 72.27pt or 1in plus.2in minus.1in (a LaTeX length register is what plain TeX calls a skip register). The initial value is zero. The control sequence \len must not be already defined.

An example:

\newlength{\graphichgt}

If you forget the backslash then you get ‘Missing control sequence inserted’. If the control sequence already exists then you get something like ‘LaTeX Error: Command \graphichgt already defined. Or name \end... illegal, see p.192 of the manual’.


12.7 \newsavebox

Synopsis:

\newsavebox{\cmd}

Define \cmd, the string consisting of a backslash followed by cmd, to refer to a new bin for storing material. These bins hold material that has been typeset, to use multiple times or to measure or manipulate (see Boxes). The bin name \cmd is required, must start with a backslash, \, and must not already be a defined command. This command is fragile (see \protect).

This allocates a bin and then puts typeset material into it.

\newsavebox{\logobox}
\savebox{\logobox}{LoGo}
Our logo is \usebox{\logobox}. 

The output is ‘Our logo is LoGo’.

If there is an already defined bin then you get something like ‘LaTeX Error: Command \logobox already defined. Or name \end... illegal, see p.192 of the manual’.

The allocation of a box is global.


12.8 \newenvironment & \renewenvironment

Synopses, one of:

\newenvironment{env}{begdef}{enddef}
\newenvironment{env}[nargs]{begdef}{enddef}
\newenvironment{env}[nargs][optargdefault]{begdef}{enddef}
\newenvironment*{env}{begdef}{enddef}
\newenvironment*{env}[nargs]{begdef}{enddef}
\newenvironment*{env}[nargs][optargdefault]{begdef}{enddef}

or one of these.

\renewenvironment{env}{begdef}{enddef}
\renewenvironment{env}[nargs]{begdef}{enddef}
\renewenvironment{env}[nargs][optargdefault]{begdef}{enddef}
\renewenvironment*{env}{begdef}{enddef}
\renewenvironment*{env}[nargs]{begdef}{enddef}
\renewenvironment*{env}[nargs][optargdefault]{begdef}{enddef}

Define or redefine the environment env, that is, create the construct \begin{env} ... body ... \end{env}.

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.

env

Required; the environment name. It consists only of letters or the * character, and thus does not begin with backslash, \. It must not begin with the string end. For \newenvironment, the name env must not be the name of an already existing environment, and also the command \env must be undefined. For \renewenvironment, env must be the name of an existing environment.

nargs

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 \begin, as in \begin{env}{arg1} ... {argn}. 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.

optargdefault

Optional; if this is present then the first argument of the defined environment is optional, with default value optargdefault (which may be the empty string). If this is not in the definition then the environment does not take an optional argument.

That is, when optargdefault is present in the definition of the environment then you can start the environment with square brackets, as in \begin{env}[optval]{...} ... \end{env}. In this case, within begdefn the parameter #1 is set to the value of optval. If you call \begin{env} without square brackets, then within begdefn the parameter #1 is set to the value of the default optargdefault. In either case, any required arguments start with #2.

Omitting [myval] in the call is different than having the square brackets with no contents, as in []. The former results in #1 expanding to optargdefault; the latter results in #1 expanding to the empty string.

begdef

Required; the text expanded at every occurrence of \begin{env}. Within begdef, the parameters #1, #2, ... #nargs, are replaced by the values that you supply when you call the environment; see the examples below.

enddef

Required; the text expanded at every occurrence of \end{env}. This may not contain any parameters, that is, you cannot use #1, #2, etc., here (but see the final example below).

All environments, that is to say the begdef code, the environment body, and the enddef code, are processed within a group. Thus, in the first example below, the effect of the \small is limited to the quote and does not extend to material following the environment.

If you try to define an environment and the name has already been used then you get something like ‘LaTeX Error: Command \fred already defined. Or name \end... illegal, see p.192 of the manual’. If you try to redefine an environment and the name has not yet been used then you get something like ‘LaTeX Error: Environment hank undefined.’.

This example gives an environment like LaTeX’s quotation except that it will be set in smaller type.

\newenvironment{smallquote}{%
  \small\begin{quotation}
}{%
  \end{quotation}
}

This has an argument, which is set in boldface at the start of a paragraph.

\newenvironment{point}[1]{%
  \noindent\textbf{#1}
}{%
}

This one shows the use of a optional argument; it gives a quotation environment that cites the author.

\newenvironment{citequote}[1][Shakespeare]{%
  \begin{quotation}
  \noindent\textit{#1}: 
}{%
  \end{quotation}
}

The author’s name is optional, and defaults to ‘Shakespeare’. In the document, use the environment like this.

\begin{citequote}[Lincoln]
  ...
\end{citequote}

The final example shows how to save the value of an argument to use in enddef, in this case in a box (see \sbox & \savebox).

\newsavebox{\quoteauthor}
\newenvironment{citequote}[1][Shakespeare]{%
  \sbox\quoteauthor{#1}%
  \begin{quotation} 
}{%
  \hspace{1em plus 1fill}---\usebox{\quoteauthor}
  \end{quotation}
}

12.9 \newtheorem

Synopses:

\newtheorem{name}{title}
\newtheorem{name}{title}[numbered_within]
\newtheorem{name}[numbered_like]{title}

Define a new theorem-like environment. You can specify one of numbered_within and numbered_like, or neither, but not both.

The first form, \newtheorem{name}{title}, creates an environment that will be labelled with title; see the first example below.

The second form, \newtheorem{name}{title}[numbered_within], creates an environment whose counter is subordinate to the existing counter numbered_within, so this counter will be reset when numbered_within is reset. See the second example below.

The third form \newtheorem{name}[numbered_like]{title}, with optional argument between the two required arguments, creates an environment whose counter will share the previously defined counter numbered_like. See the third example.

This command creates a counter named name. In addition, unless the optional argument numbered_like is used, inside of the theorem-like environment the current \ref value will be that of \thenumbered_within (see \ref).

This declaration is global. It is fragile (see \protect).

Arguments:

name

The name of the environment. It is a string of letters. It must not begin with a backslash, \. It must not be the name of an existing environment, and the command name \name must not already be defined.

title

The text to be printed at the beginning of the environment, before the number. For example, ‘Theorem’.

numbered_within

Optional; the name of an already defined counter, usually a sectional unit such as chapter or section. When the numbered_within counter is reset then the name environment’s counter will also be reset.

If this optional argument is not used then the command \thename is set to \arabic{name}.

numbered_like

Optional; the name of an already defined theorem-like environment. The new environment will be numbered in sequence with numbered_like.

Without any optional arguments the environments are numbered sequentially. The example below has a declaration in the preamble that results in ‘Definition 1’ and ‘Definition 2’ in the output.

\newtheorem{defn}{Definition}
\begin{document}
\section{...}
\begin{defn}
  First def 
\end{defn}

\section{...}
\begin{defn}
  Second def
\end{defn}

This example has the same document body as the prior one. But here \newtheorem’s optional argument numbered_within is given as section, so the output is like ‘Definition 1.1’ and ‘Definition 2.1’.

\newtheorem{defn}{Definition}[section]
\begin{document}
\section{...}
\begin{defn}
  First def 
\end{defn}

\section{...}
\begin{defn}
  Second def
\end{defn}

In the next example there are two declarations in the preamble, the second of which calls for the new thm environment to use the same counter as defn. It gives ‘Definition 1.1’, followed by ‘Theorem 2.1’ and ‘Definition 2.2’.

\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}

12.10 \newfont

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 Fonts).

Synopsis:

\newfont{\cmd}{font description}

Define a command \cmd that will change the current font. The control sequence must not already be defined. It must begin with a backslash, \.

The font description consists of a fontname and an optional at clause. LaTeX will look on your system for a file named fontname.tfm. The at clause can have the form either at dimen or scaled factor, where a factor of ‘1000’ means no scaling. For LaTeX’s purposes, all this does is scale all the character and other font dimensions relative to the font’s design size, which is a value defined in the .tfm file.

This defines two equivalent fonts and typesets a few characters in each.

\newfont{\testfontat}{cmb10 at 11pt}
\newfont{\testfontscaled}{cmb10 scaled 1100}
\testfontat abc
\testfontscaled abc

12.11 \protect

All LaTeX commands are either fragile or robust. A fragile command can break when it is used in the argument to certain other commands, typically those that write material to the table of contents, the cross-reference file, etc. To prevent fragile commands from causing errors, one solution is to precede them with the command \protect.

For example, when LaTeX runs the \section{section name} command it writes the section name text to the .aux auxiliary file, moving it there for use elsewhere in the document such as in the table of contents. Such an argument that is used in multiple places is referred to as a moving argument. 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 \caption{...} command (see figure), in the \thanks{...} command (see \maketitle), and in @-expressions in the tabular and array environments (see tabular).

If you get strange errors from commands used in moving arguments, try preceding it with \protect. Each fragile command must be protected with their own \protect.

Although usually a \protect command doesn’t hurt, length commands such as \parindent should not be preceded by a \protect command (see Lengths. Nor can a \protect command be used in the argument to \addtocounter or \setcounter command (see \setcounter and \addtocounter. These commands are already robust.

As of the October 2019 release of LaTeX (https://www.latex-project.org/news/latex2e-news/ltnews30.pdf), most commands that had been previously fragile were fixed to be robust. For example, any command taking an optional argument, such as \root or \raisebox, was fragile, but is now robust. Similarly, \(...\) math was fragile and is now robust ($...$ has always been robust).

Perhaps the most commonly used remaining fragile command is \verb; for example,

\begin{figure}
  ...
  \caption{This \verb|\command| causes an error.}
\end{figure}

Adding \protect does not help here. It’s usually feasible to rewrite the caption (or section heading or whatever) to use \texttt, often the simplest solution.

Alternatively, to use \verb, you can apply the \cprotect command from cprotect package (https://ctan.org/pkg/cprotect) to the \caption:

\cprotect\caption{This \verb|\command| is ok with \verb|\cprotect|.}

\cprotect also allows use of \begin...\end environments in moving arguments, where they are normally not allowed, via a similar prefix command \cprotEnv.


12.12 \ignorespaces & \ignorespacesafterend

Synopsis:

\ignorespaces

or

\ignorespacesafterend

Both commands cause LaTeX to ignore blanks (that is, characters of catcode 10 such as space or tabulation) after the end of the command up to the first box or non-blank character. The first is a primitive command of TeX, and the second is LaTeX-specific.

The \ignorespaces is often used when defining commands via \newcommand, or \newenvironment, or \def. 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 enumerate list, users must not put any space between the command and the question text.

\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} 

The solution is to change to this.

\newcommand{\points}[1]{%
  \makebox[0pt]{\makebox[10em][l]{#1~pts}}\ignorespaces}

A second example shows blanks 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 \fullname will reproduce that.

\newcommand{\honorific}[1]{\def\honorific{#1}} % remember title
\newcommand{\fullname}[1]{\honorific~#1}       % put title before name

\begin{tabular}{|l|}
\honorific{Mr/Ms}  \fullname{Jones} \\  % no extra space here
\honorific{ Mr/Ms} \fullname{Jones}     % extra space before title
\end{tabular}

To fix this, change to \newcommand{\fullname}[1]{\ignorespaces\honorific~#1}.

The \ignorespaces is also often used in a \newenvironment at the end of the begin clause, as in \begin{newenvironment}{env name}{... \ignorespaces}{...}.

To strip blanks off the end of an environment use \ignorespacesafterend. An example is that this will show a much larger vertical space between the first and second environments than between the second and third.

\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}

Putting a comment character % immediately after the \end{eq} will make the vertical space disappear, but that is inconvenient. The solution is to change to \newenvironment{eq}{\begin{equation}}{\end{equation}\ignorespacesafterend}.


12.13 xspace package

This is an add-on package, not part of core LaTeX. Synopsis:

\usepackage{xspace}
  ...
\newcommand{...}{...\xspace}

The \xspace macro, when used at the end of a command definition, adds a space unless the command is followed by certain punctuation characters.

After a control sequence that is a control word (see Control sequence, control word and control symbol, as opposed to control symbols such as \$), TeX gobbles blank characters. Thus, in the first sentence below, the output has ‘Vermont’ placed snugly against the period, without any intervening space, despite the space in the input.

\newcommand{\VT}{Vermont}
Our college is in \VT .
\VT{} summers are nice.

But because of the gobbling, the second sentence needs the empty curly braces or else there would be no space separating ‘Vermont’ from ‘summers’. (Many authors instead use a backslash-space \ for this. See Backslash-space, \ .)

The xspace package provides \xspace. It is for writing commands which are designed to be used mainly in text. It must be placed at the very end of the definition of these commands. It inserts a space after that command unless what immediately follows is in a list of exceptions. In this example, the empty braces are not needed.

\newcommand{\VT}{Vermont\xspace}
Our college is in \VT .
\VT summers are nice.

The default exception list contains the characters ,.'/?;:!~-), the open curly brace and the backslash-space command discussed above, and the commands \footnote or \footnotemark. You can add to that list as with \xspaceaddexceptions{\myfni \myfnii} which adds \myfni and \myfnii to the list; and you can remove from that list as with \xspaceremoveexception{!}.

A comment: many experts prefer not to use \xspace. Putting it in a definition means that the command will usually get the spacing right. But it isn’t easy to predict when to enter empty braces because \xspace will get it wrong, such as when it is followed by another command, and so \xspace can make editing material harder and more error-prone than instead of always inserting the empty braces.


12.14 Class and package commands

These are commands designed to help writers of classes or packages.


12.14.1 \AtBeginDvi & \AtEndDvi

Synopsis:

\AtBeginDvi{code}
\AtEndDvi{code}

\AtBeginDvi saves, in a box register, code to be executed at the beginning of the shipout of the first page of the document. Despite the name, it applies to DVI, PDF, and XDV output. It fills the shipout/firstpage hook; new code should use that hook directly.

Similarly, \AtEndDvi (previously available only with the atenddvi package) is code executed when finalizing the main output document.


12.14.2 \AtEndOfClass & \AtEndOfPackage

Synopses:

\AtEndOfClass{code}
\AtEndOfPackage{code}

Hooks to insert code to be executed when LaTeX finishes processing the current class resp. package.

These hooks can be used multiple times; each code segment will be executed in the order called. Many packages and classes use these commands.

See also \AtBeginDocument.


12.14.3 \CheckCommand

Synopsis:

\CheckCommand{cmd}[num][default]{definition}
\CheckCommand* (same parameters)

Like \newcommand (see \newcommand & \renewcommand) but does not define cmd; instead it checks that the current definition of cmd is exactly as given by definition and is or is not \long as expected. A long command is a command that accepts \par within an argument.

With the unstarred version of \CheckCommand, cmd is expected to be \long; with the starred version, cmd must not be \long

\CheckCommand raises an error when the check fails. This allows you to check before you start redefining cmd yourself that no other package has already redefined this command.


12.14.4 \ClassError and \PackageError and others

Produce error, warning, and informational messages for classes:

\ClassError{class name}{error-text}{help-text}
\ClassWarning{class name}{warning-text}
\ClassWarningNoLine{class name}{warning-text}
\ClassInfo{class name}{info-text}
\ClassInfoNoLine{class name}{info-text}

and the same for packages:

\PackageError{package name}{error-text}{help-text}
\PackageWarning{package name}{warning-text}
\PackageWarningNoLine{package name}{warning-text}
\PackageInfo{package name}{info-text}
\PackageInfoNoLine{package name}{info-text}

For \ClassError and \PackageError the message is error-text, followed by TeX’s ‘?’ error prompt. If the user then asks for help by typing h, they see the help text.

The four Warning commands are similar except that they write warning-text on the screen with no error prompt. The four Info commands write info-text only in the transcript file. The NoLine versions omit the number of the line generating the message, while the other versions do show that number.

To format the messages, including the help-text: use \protect to stop a command from expanding, get a line break with \MessageBreak, and get a space with \space when a space character is ignore, most commonly after a command.

LaTeX appends a period to the messages.


12.14.5 \CurrentOption

Expands to the name of the option currently being processed. This can only be used within the code argument of either \DeclareOption or \DeclareOption*.


12.14.6 \DeclareOption

Synopsis:

\DeclareOption{option}{code}
\DeclareOption*{option}{code}

Define an option a user can include in their \documentclass command. For example, a class smcmemo could have an option logo allowing users to put the institutional logo on the first page. The document would start with \documentclass[logo]{smcmemo}. To enable this, the class file must contain \DeclareOption{logo}{code} (and later, \ProcessOptions).

If you request an option that has not been declared, by default this will produce a warning like Unused global option(s): [badoption]. This can be changed by using \DeclareOption*{code}, which executes code for any unknown option.

For example, many classes extend an existing class, using code such as \LoadClass{article} (see \LoadClass). In this case, it makes sense to pass any otherwise-unknown options to the underlying class, like this:

\DeclareOption*{%
  \PassOptionsToClass{\CurrentOption}{article}%
}

As another example, our class smcmemo might allow users to keep lists of memo recipients in external files, so the user could invoke \documentclass[math]{smcmemo} and it will read the file math.memo. This code inputs the file if it exists, while if it doesn’t, the option is passed to the article class:

\DeclareOption*{\InputIfFileExists{\CurrentOption.memo}
  {}{%
  \PassOptionsToClass{\CurrentOption}{article}}}

12.14.7 \DeclareRobustCommand

Synopsis:

\DeclareRobustCommand{cmd}[num][default]{definition}
\DeclareRobustCommand* (same parameters

\DeclareRobustCommand and its starred form are generally like \newcommand and \newcommand* (see \newcommand & \renewcommand), with the addition that they define a so-called robust command, even if some code within the definition is fragile. (For a discussion of robust and fragile commands, see \protect.)

Also unlike \newcommand, these do not give an error if macro cmd already exists; instead, a log message is put into the transcript file if a command is redefined. Thus, \DeclareRobustCommand can be used to define new robust commands or to redefine existing commands, making them robust.

The starred form, \DeclareRobustCommand*, disallows the arguments from containing multiple paragraphs, just like the starred form of \newcommand and \renewcommand. The meaning of the arguments is the same.

Commands defined this way are a bit less efficient than those defined using \newcommand so unless the command’s data is fragile and the command is used within a moving argument, use \newcommand.

Related to this, the etoolbox package offers three commands and their starred forms: \newrobustcmd(*) \renewrobustcmd(*), and \providerobustcmd(*). They are similar to \newcommand, \renewcommand, and \providecommand and their own starred forms, but define a robust cmd. They have two possible advantages compared to \DeclareRobustCommand:

  1. They use the low-level e-TeX protection mechanism rather than the higher-level LaTeX \protect mechanism, so they do not incur the slight loss of performance mentioned above, and
  2. They make the same distinction between \new…, \renew…, and \provide…, as the standard commands. That is, they do not just write a log message when you redefine cmd that already exists; you need to use either \renew… or \provide…, or you get an error. This may or may not be a benefit.

12.14.8 \ExecuteOptions

Synopsis:

\ExecuteOptions{option-list}

For each option option in option-list, in order, this command executes the command \ds@option. If this command is not defined then that option is silently ignored.

This can be used to provide a default option list before \ProcessOptions. For example, if in a class file you want the default to be 11pt fonts then you could specify \ExecuteOptions{11pt}\ProcessOptions\relax.


12.14.9 \IfFileExists & \InputIfFileExists

Synopses:

\IfFileExists{filename}{true-code}{false-code}
\InputIfFileExists{filename}{true-code}{false-code}

\IfFileExists executes true-code if LaTeX finds the file filename or false-code otherwise. In the first case it executing true-code and then inputs the file. Thus the command

\IfFileExists{img.pdf}{%
  \includegraphics{img.pdf}}
  {\typeout{!! img.pdf not found}

will include the graphic img.pdf if it is found and otherwise give a warning.

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 \IfFileExists{./filename}{true-code}{false-code}. If you ask for a filename without a .tex extension then LaTeX will first look for the file by appending the .tex; for more on how LaTeX handles file extensions see \input.

\InputIfFileExists is similar, but, as the name states, automatically \inputs filename if it exists. The true-code is executed just before the \input; if the file doesn’t exist, the false-code is executed. An example:

\InputIfFileExists{mypkg.cfg}
  {\PackageInfo{Loading mypkg.cfg for configuration information}}
  {\PackageInfo{No mypkg.cfg found}}

12.14.10 \LoadClass & \LoadClassWithOptions

Synopses:

\LoadClass[options-list]{class-name}[release-date]
\LoadClassWithOptions{class-name}[release-date]

Load a class, as with \documentclass[options-list]{class-name}[release-date]. An example: \LoadClass[twoside]{article}.

The options-list, if present, is a comma-separated list. The release-date is also optional. If present it must have the form YYYY/MM/DD.

If you request release-date 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:

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.

The command version \LoadClassWithOptions uses the list of options for the current class. This means it ignores any options passed to it via \PassOptionsToClass. This is a convenience command that lets you build classes on existing ones, such as the standard article class, without having to track which options were passed.


12.14.11 \NeedsTeXFormat

Synopsis:

\NeedsTeXFormat{format}[format-date]

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: \NeedsTeXFormat{LaTeX2e}. When a document using that class is processed, the format being run must exactly match the format name given, including case. If it does not match then execution stops with an error like ‘This file needs format `LaTeX2e' but this is `plain'.’.

To require a version of the format that you know to have certain features, include the optional format-date on which those features were implemented. If present, it must be in the form YYYY/MM/DD. If the format version installed on your system is earlier than format date then you get a warning like this.

You have requested release `2038/01/20' of LaTeX, but only
release `2016/02/01' is available.

12.14.12 \OptionNotUsed

Adds the current option to the list of unused options. Can only be used within the code argument of either \DeclareOption or \DeclareOption*.


12.14.13 \PassOptionsToClass & \PassOptionsToPackage

Synopses:

\PassOptionsToClass{options}{clsname}
\PassOptionsToPackage{option}{pkgname}

Adds the options in the comma-separated list options to the options used by any future \RequirePackage or \usepackage command for the class clsname or the package pkgname, respectively.

The reason for these commands is that although you may load a package any number of times with no options, if you can specify options only the first time you load the package. Loading a package with options more than once will get you an error like Option clash for package foo.. LaTeX throws an error even if there is no conflict between the options.

If your own code is bringing in a package twice then you can combine the calls; for example, replacing the two

\RequirePackage[landscape]{geometry}
\RequirePackage[margins=1in]{geometry}

with the single command

\RequirePackage[landscape,margins=1in]{geometry}

However, suppose you are loading firstpkg and inside that package it loads secondpkg, and you need secondpkg to be loaded with option draft. Then before load the first package you must tell LaTeX about the desired options for the second package, like this:

\PassOptionsToPackage{draft}{secondpkg}
\RequirePackage{firstpkg}

If firstpkg.sty loads an option in conflict with what you want then you may have to alter its source, or yours.

These commands are useful for general users as well as class and package writers. For instance, suppose a user wants to load the graphicx package with the option draft and also wants to use a class foo that loads the graphicx package, but without that option. The user could start their LaTeX file with \PassOptionsToPackage{draft}{graphicx} \documentclass{foo}.


12.14.14 \ProcessOptions

Synopsis:

\ProcessOptions\@options
\ProcessOptions*\@options

Execute the code for each option that the user has invoked. Invoke it in the class file as \ProcessOptions\relax (because of the existence of the starred version, described below).

Options come in two types. Local options have been specified for this particular package in \usepackage[options], \RequirePackage[options], or the options argument of \PassOptionsToPackage{options}. Global options are those given by the class user in \documentclass[options]. If an option is specified both locally and globally then it is local.

When \ProcessOptions is called for a package pkg.sty, the following happens:

  1. For each option option so far declared with \DeclareOption, \ProcessOptions looks to see if that option is either global or local for pkg. If so, then it executes the declared code. This is done in the order in which these options were given in pkg.sty.
  2. For each remaining local option, it executes the command \ds@option if it has been defined somewhere (other than by a \DeclareOption); otherwise, it executes the default option code given in \DeclareOption*. 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.

When \ProcessOptions is called for a class it works in the same way except that all options are local, and the default code for \DeclareOption* is \OptionNotUsed rather than an error.

The starred version \ProcessOptions* 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.


12.14.15 \ProvidesClass & \ProvidesPackage

Synopses:

\ProvidesClass{clsname}[release-date [info-text]]
\ProvidesPackage{pkgname}[release-date [info-text]]

Identifies the class or package being defined, printing a message to the screen and the log file.

When you load a class or package, for example with \documentclass{smcmemo} or \usepackage{test}, LaTeX inputs a file (smcmemo.cls and test.sty, respectively). 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 \documentclass{smcmemo}, and the file smcmemo.cls has the statement \ProvidesClass{foo} then you get a warning like You have requested document class `smcmemo', but the document class provides 'foo'. This warning does not prevent LaTeX from processing the rest of the class file normally.

If you include the optional argument then you must include a date, before any spaces, of the form YYYY/MM/DD. The rest of the optional argument is free-form, although it traditionally identifies the class. It is written to the screen during compilation and to the log file. Thus, if your file smcmemo.cls contains the line \ProvidesClass{smcmemo}[2008/06/01 v1.0 SMC memo class] and your document’s first line is \documentclass{smcmemo} then you will see Document Class: smcmemo 2008/06/01 v1.0 SMC memo class.

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 release date. For instance, a user could enter \documentclass{smcmemo}[2018/10/12] or \usepackage{foo}[[2017/07/07]] to require a class or package with certain features by specifying that it must be released no earlier than the given date. Perhaps more importantly, the date serves as documentation of the last release. (In practice, package users rarely include a date, and class users almost never do.)


12.14.16 \ProvidesFile

Synopsis:

\ProvidesFile{filename}[info-text]

Declare a file other than the main class and package files, such as a configuration or font definition file. It writes the given information to the log file, essentially like \ProvidesClass and \ProvidesPackage (see the previous section).

For example:

\ProvidesFile{smcmemo.cfg}[2017/10/12 config file for smcmemo.cls]

writes this into the log:

File: smcmemo.cfg 2017/10/12 config file for smcmemo.cls

12.14.17 \RequirePackage & \RequirePackageWithOptions

Synopsis:

\RequirePackage[option-list]{pkgname}[release-date]
\RequirePackageWithOptions{pkgname}[release-date]

Load a package, like the command \usepackage (see Additional packages). An example:
\RequirePackage[landscape,margin=1in]{geometry}

The initial optional argument option-list, if present, must be a comma-separated list. The trailing optional argument release-date, if present, must have the form YYYY/MM/DD. If the release date of the package as installed on your system is earlier than release-date then you get a warning like ‘You have requested, on input line 9, version `2017/07/03' of package jhtest, but only version `2000/01/01' is available’.

The \RequirePackageWithOptions variant uses the list of options for the current class. This means it ignores any options passed to it via \PassOptionsToClass. This is a convenience command to allow easily building classes on existing ones without having to track which options were passed.

The difference between \usepackage and \RequirePackage is small. The \usepackage command is intended to be used in documents, while \RequirePackage is intended for package and class files. The most significant difference in practice is that \RequirePackage can be used in a document before the \documentclass command, while \usepackage gives an error there. The most common need for this nowadays is for the \DocumentMetadata command (see \DocumentMetadata: Producing tagged PDF output).

The LaTeX development team strongly recommends use of these and related commands over plain TeX’s \input; see the Class Guide (https://ctan.org/pkg/clsguide).


13 Counters

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’s name has no backslash \. Thus, associated with the \chapter command is the chapter counter that keeps track of the chapter number.

Below is a list of the counters used in LaTeX’s standard document classes to control numbering.

part            paragraph       figure          enumi
chapter         subparagraph    table           enumii
section         page            footnote        enumiii
subsection      equation        mpfootnote      enumiv
subsubsection

The mpfootnote counter is used by the \footnote command inside of a minipage (see minipage). The counters enumi through enumiv are used in the enumerate environment, for up to four levels of nesting (see enumerate).

Counters can have any integer value but they are typically positive.

New counters are created with \newcounter. See \newcounter: Allocating a counter.


13.1 \alph \Alph \arabic \roman \Roman \fnsymbol: Printing counters

Print the value of a counter, in a specified style. For instance, if the counter counter has the value 1 then a \alph{counter} in your source will result in a lowercase letter a appearing in the output.

All of these commands take a single counter as an argument, for instance, \alph{enumi}. Note that the counter name does not start with a backslash.

\alph{counter}

Print the value of counter in lowercase letters: ‘a’, ‘b’, ... If the counter’s value is less than 1 or more than 26 then you get ‘LaTeX Error: Counter too large.

\Alph{counter}

Print in uppercase letters: ‘A’, ‘B’, ... If the counter’s value is less than 1 or more than 26 then you get ‘LaTeX Error: Counter too large.

\arabic{counter}

Print in Arabic numbers such as ‘5’ or ‘-2’.

\roman{counter}

Print in lowercase roman numerals: ‘i’, ‘ii’, ... If the counter’s value is less than 1 then you get no warning or error but LaTeX does not print anything in the output.

\Roman{counter}

Print in uppercase roman numerals: ‘I’, ‘II’, ... If the counter’s value is less than 1 then you get no warning or error but LaTeX does not print anything in the output.

\fnsymbol{counter}

Prints the value of counter using a sequence of nine symbols that are traditionally used for labeling footnotes. The value of counter should be between 1 and 9, inclusive. If the counter’s value is less than 0 or more than 9 then you get ‘LaTeX Error: Counter too large’, while if it is 0 then you get no error or warning but LaTeX does not output anything.

Here are the symbols:

NumberNameCommandSymbol
1asterisk\ast*
2dagger\dagger
3ddagger\ddagger
4section-sign\S§
5paragraph-sign\P
6double-vert\parallel
7double-asterisk\ast\ast**
8double-dagger\dagger\dagger††
9double-ddagger\ddagger\ddagger‡‡

13.2 \usecounter

Synopsis:

\usecounter{counter}

Used in the second argument of the list environment (see list), this declares that list items will be numbered by counter. It initializes counter to zero, and arranges that when \item is called without its optional argument then counter is incremented by \refstepcounter, making its value be the current ref value (see \ref). This command is fragile (see \protect).

Put in the document preamble, this example makes a new list environment enumerated with testcounter:

\newcounter{testcounter}
\newenvironment{test}{%
  \begin{list}{}{%
    \usecounter{testcounter}
  }
}{%
  \end{list}
}

13.3 \value

Synopsis:

\value{counter}

Expands to the value of the counter counter. (Note that the name of a counter does not begin with a backslash.)

This example outputs ‘Test counter is 6. Other counter is 5.’.

\newcounter{test} \setcounter{test}{5}
\newcounter{other} \setcounter{other}{\value{test}}
\addtocounter{test}{1}

Test counter is \arabic{test}.
Other counter is \arabic{other}.

The \value command is not used for typesetting the value of the counter. For that, see \alph \Alph \arabic \roman \Roman \fnsymbol: Printing counters.

It is often used in \setcounter or \addtocounter but \value can be used anywhere that LaTeX expects a number, such as in \hspace{\value{foo}\parindent}. It must not be preceded by \protect (see \protect).

This example inserts \hspace{4\parindent}.

\setcounter{myctr}{3} \addtocounter{myctr}{1}
\hspace{\value{myctr}\parindent}

13.4 \setcounter

Synopsis:

\setcounter{counter}{value}

Globally set the counter counter to have the value of the value argument, which must be an integer. Thus, you can set a counter’s value as \setcounter{section}{5}. Note that the counter name does not start with a backslash.

In this example if the counter theorem has value 12 then the second line will print ‘XII’.

\setcounter{exercise}{\value{theorem}}
Here it is in Roman: \Roman{exercise}.

13.5 \addtocounter

Synopsis:

\addtocounter{counter}{value}

Globally increment counter by the amount specified by the value argument, which may be negative.

In this example the section value appears as ‘VII’.

\setcounter{section}{5}
\addtocounter{section}{2}
Here it is in Roman: \Roman{section}.

13.6 \refstepcounter

Synopsis:

\refstepcounter{counter}

Globally increments the value of counter by one, as does \stepcounter (see \stepcounter). The difference is that this command resets the value of any counter numbered within it. (For the definition of “counters numbered within”, see \newcounter: Allocating a counter.)

In addition, this command also defines the current \ref value to be the result of \thecounter.

While the counter value is set globally, the \ref value is set locally, i.e., inside the current group.


13.7 \stepcounter

Synopsis:

\stepcounter{counter}

Globally adds one to counter and resets all counters numbered within it. (For the definition of “counters numbered within”, see \newcounter: Allocating a counter.)

This command differs from \refstepcounter in that this one does not influence references; that is, it does not define the current \ref value to be the result of \thecounter (see \refstepcounter).


13.8 \day & \month & \year

LaTeX defines the counter \day for the day of the month (nominally with value between 1 and 31), \month for the month of the year (nominally with value between 1 and 12), and \year for the year. When TeX starts up, they are set from the current values on the system. The related command \today produces a string representing the current day (see \today).

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:

\day=-2 \month=13 \year=-4 \today

gives no error or warning and results in the output ‘-2, -4’ (the bogus month value produces no output).

See Command line input, to force the date to a given value from the command line.


14 Lengths

A length is a measure of distance. Many LaTeX commands take a length as an argument.

Lengths come in two types. A rigid length such as 10pt does not contain a plus or minus component. (Plain TeX calls this a dimen.) A rubber length (what plain TeX calls a skip or glue) such as with 1cm plus0.05cm minus0.01cm can contain either or both of those components. In that rubber length, the 1cm is the natural length while the other two, the plus and minus components, allow TeX to stretch or shrink the length to optimize placement.

The illustrations below use these two commands.

% make a black bar 10pt tall and #1 wide
\newcommand{\blackbar}[1]{\rule{#1}{10pt}}

% Make a box around #2 that is #1 wide (excluding the border)
\newcommand{\showhbox}[2]{%
  \fboxsep=0pt\fbox{\hbox to #1{#2}}} 

This next example uses those commands to show a black bar 100 points long between ‘ABC’ and ‘XYZ’. This length is rigid.

ABC\showhbox{100pt}{\blackbar{100pt}}XYZ

As for rubber lengths, shrinking is simpler one: with 1cm minus 0.05cm, the natural length is 1cm but 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 points between the two bars.

ABC\showhbox{300pt}{%
  \blackbar{101pt}\hspace{100pt minus 2pt}\blackbar{101pt}}YYY

ABC\showhbox{300pt}{%
  \blackbar{105pt}\hspace{100pt minus 1pt}\blackbar{105pt}}YYY

But the second one gets a warning like ‘Overfull \hbox (1.0pt too wide) detected at line 17’. In the output the first ‘Y’ is overwritten by the end of the black bar, because the box’s material is wider than the 300pt allocated, as TeX has refused to shrink the total to less than 309 points.

Stretching is like shrinking except that if TeX is asked to stretch beyond the given amount, it will do it. Here the first line is fine, producing a space of 110 points between the bars.

ABC\showhbox{300pt}{%
  \blackbar{95pt}\hspace{100pt plus 10pt}\blackbar{95pt}}YYY

ABC\showhbox{300pt}{%
  \blackbar{95pt}\hspace{100pt plus 1pt}\blackbar{95pt}}YYY

In the second line TeX needs a stretch of 10 points and only 1 point was specified. TeX stretches the space to the required length but it gives you a warning like ‘Underfull \hbox (badness 10000) detected at line 22’. (We won’t discuss badness.)

You can put both stretch and shrink in the same length, as in 1ex plus 0.05ex minus 0.02ex.

If TeX is setting two or more rubber lengths then it allocates the stretch or shrink in proportion.

ABC\showhbox{300pt}{%
  \blackbar{100pt}%  left
  \hspace{0pt plus 50pt}\blackbar{80pt}\hspace{0pt plus 10pt}%  middle
  \blackbar{100pt}}YYY  % right

The left and right bars take up 100 points, so the middle needs another 100. The middle bar is 80 points so the two \hspace’s must stretch 20 points. Because the two are plus 50pt and plus 10pt, TeX gets 5/6 of the stretch from the first space and 1/6 from the second.

The plus or minus component of a rubber length can contain a fill component, as in 1in plus2fill. This gives the length infinite stretchability or shrinkability so that TeX could set it to any distance. Here the two figures will be equally spaced across the page.

\begin{minipage}{\linewidth}
  \hspace{0pt plus 1fill}\includegraphics{godel.png}%
  \hspace{0pt plus 1fill}\includegraphics{einstein.png}%
  \hspace{0pt plus 1fill}
\end{minipage}

TeX has three levels of infinity for glue components: fil, fill, and filll. The later ones are more infinite than the earlier ones. Ordinarily document authors only use the middle one (see \hfill and see \vfill).

Multiplying a rubber length by a number turns it into a rigid length, so that after \setlength{\ylength}{1in plus 0.2in} and \setlength{\zlength}{3\ylength} then the value of \zlength is 3in.


14.1 Units of length

TeX and LaTeX know about these units both inside and outside of math mode.

pt

Point, 1/72.27 inch. The (approximate) conversion to metric units is 1point = .35146mm = .035146cm.

pc

Pica, 12 pt

in

Inch, 72.27 pt

bp

Big point, 1/72 inch. This length is the definition of a point in PostScript and many desktop publishing systems.

mm

Millimeter, 2.845pt

cm

Centimeter, 10mm

dd

Didot point, 1.07 pt

cc

Cicero, 12 dd

sp

Scaled point, 1/65536 pt

Three other units are defined according to the current font, rather than being an absolute dimension.

ex

The x-height of the current font ex, traditionally the height of the lowercase letter x, is often used for vertical lengths.

em

Similarly em, traditionally the width of the capital letter M, is often used for horizontal lengths. This is also often the size of the current font, e.g., a nominal 10pt font will have 1em = 10pt. LaTeX has several commands to produce horizontal spaces based on the em (see \enspace & \quad & \qquad).

mu

Finally, in math mode, many definitions are expressed in terms of the math unit mu, defined by 1em = 18mu, where the em is taken from the current math symbols family. See Spacing in math mode.

Using these units can help make a definition work better across font changes. For example, a definition of the vertical space between list items given as \setlength{\itemsep}{1ex plus 0.05ex minus 0.01ex} is more likely to still be reasonable if the font is changed than a definition given in points.


14.2 \setlength

Synopsis:

\setlength{\len}{amount}

Set the length \len to amount. The length name \len has to be a control sequence (see Control sequence, control word and control symbol), and as such must begin with a backslash, \ under normal circumstances. The amount can be a rubber length (see Lengths). It can be positive, negative or zero, and can be in any units that LaTeX understands (see Units of length).

Below, with LaTeX’s defaults the first paragraph will be indented while the second will not.

I told the doctor I broke my leg in two places.

\setlength{\parindent}{0em}
He said stop going to those places.

If you did not declare \len with \newlength, for example if you mistype it as in \newlength{\specparindent}\setlength{\sepcparindent}{...}, then you get an error like ‘Undefined control sequence. <argument> \sepcindent’. If you omit the backslash at the start of the length name then you get an error like ‘Missing number, treated as zero.’.


14.3 \addtolength

Synopsis:

\addtolength{\len}{amount}

Increment the length \len by amount. The length name \len has to be a control sequence (see Control sequence, control word and control symbol), and as such must begin with a backslash, \ under normal circumstances. The amount is a rubber length (see Lengths). It can be positive, negative or zero, and can be in any units that LaTeX understands (see Units of length).

Below, if \parskip starts with the value 0pt plus 1pt

Doctor: how is the boy who swallowed the silver dollar?
\addtolength{\parskip}{1pt}

Nurse: no change.

then it has the value 1pt plus 1pt for the second paragraph.

If you did not declare \len with \newlength, for example if you mistype it as in \newlength{\specparindent}\addtolength{\sepcparindent}{...}, then you get an error like ‘Undefined control sequence. <argument> \sepcindent’. If the amount uses some length that has not been declared, for instance if for example you mistype the above as \addtolength{\specparindent}{0.6\praindent}, then you get something like ‘Undefined control sequence. <argument> \praindent’. If you leave off the backslash at the start of \len, as in \addtolength{parindent}{1pt}, then you get something like ‘You can't use `the letter p' after \advance’.


14.4 \settodepth

Synopsis:

\settodepth{\len}{text}

Set the length \len to the depth of box that LaTeX gets on typesetting the text argument. The length name \len has to be a control sequence (see Control sequence, control word and control symbol), and as such must begin with a backslash, \ under normal circumstances.

This will print how low the character descenders go.

\newlength{\alphabetdepth}
\settodepth{\alphabetdepth}{abcdefghijklmnopqrstuvwxyz}
\the\alphabetdepth

If you did not declare \len with \newlength, if for example you mistype the above as \settodepth{\aplhabetdepth}{abc...}, then you get something like ‘Undefined control sequence. <argument> \aplhabetdepth’. If you leave the backslash out of \len, as in \settodepth{alphabetdepth}{...} then you get something like ‘Missing number, treated as zero. <to be read again> \setbox’.


14.5 \settoheight

Synopsis:

\settoheight{\len}{text}

Sets the length \len to the height of box that LaTeX gets on typesetting the text argument. The length name \len has to be a control sequence (see Control sequence, control word and control symbol), and as such must begin with a backslash, \ under normal circumstances.

This will print how high the characters go.

\newlength{\alphabetheight}
\settoheight{\alphabetheight}{abcdefghijklmnopqrstuvwxyz}
\the\alphabetheight

If no such length \len has been declared with \newlength, if for example you mistype as \settoheight{\aplhabetheight}{abc...}, then you get something like ‘Undefined control sequence. <argument> \alphabetheight’. If you leave the backslash out of \len, as in \settoheight{alphabetheight}{...} then you get something like ‘Missing number, treated as zero. <to be read again> \setbox’.


14.6 \settowidth

Synopsis:

\settowidth{\len}{text}

Set the length \len to the width of the box that LaTeX gets on typesetting the text argument. The length name \len has to be a control sequence (see Control sequence, control word and control symbol), and as such must begin with a backslash, \ under normal circumstances.

This prints the width of the lowercase ASCII alphabet.

\newlength{\alphabetwidth}
\settowidth{\alphabetwidth}{abcdefghijklmnopqrstuvwxyz}
\the\alphabetwidth

If no such length \len has been declared with \newlength, if for example you mistype the above as \settowidth{\aplhabetwidth}{abc...}, then you get something like ‘Undefined control sequence. <argument> \aplhabetwidth’. If you leave the backslash out of \len, as in \settoheight{alphabetwidth}{...} then you get something like ‘Missing number, treated as zero. <to be read again> \setbox’.


14.7 \stretch

Synopsis:

\stretch{number}

Produces a rubber length with zero natural length and number times \fill units of stretchability (see Lengths). The number can be positive or negative. This command is robust (see \protect).

It works for both vertical and horizontal spacing. In this horizontal example, LaTeX produces three tick marks, and the distance between the first and second is half again as long as the distance between the second and third.

\rule{0.4pt}{1ex}\hspace{\stretch{1.5}}%
  \rule{0.4pt}{1ex}\hspace{\stretch{1}}%
  \rule{0.4pt}{1ex} 

In this vertical example, the ‘We dedicate …’ will have three times as much space under it as above it.

\newenvironment{dedication}{% in document preamble
  \clearpage\thispagestyle{empty}% 
  \vspace*{\stretch{1}} % stretchable space at top 
  \it
}{%
  \vspace{\stretch{3}}  % space at bot is 3x as at top
  \clearpage
}
  ...
\begin{dedication}  % in document body
We dedicate this book to our wives.
\end{dedication}

14.8 Expressions

Synopsis, one of:

\numexpr expression 
\dimexpr expression
\glueexpr expression
\muglue expression

Any place where you may write an integer, or a TeX dimen, or TeX glue, or muglue, you can instead write an expression to compute that type of quantity.

An example is that \the\dimexpr\linewidth-4pt\relax will produce as output the length that is four points less than width of a line (the only purpose of \the is to show the result in the document). Analogously, \romannumeral\numexpr6+3\relax will produce ‘ix’, and \the\glueexpr 5pt plus 1pt * 2 \relax will produce ‘10.0pt plus 2.0pt’.

A convenience here over doing calculations by allocating registers and then using \advance, etc., is that the evaluation of expressions does not involve assignments and can therefore be performed in places where assignments are not allowed. The next example computes the width of the \parbox.

\newlength{\offset}\setlength{\offset}{2em}
\begin{center}
\parbox{\dimexpr\linewidth-\offset*3}{With malice toward none
with charity for all with firmness in the right as God gives us to see
the right let us strive on to finish the work we are in to bind up the
nation's wounds, to care for him who shall have borne the battle and
for his widow and his orphan \textasciitilde\ to do all which may
achieve and cherish a just and lasting peace among ourselves and with
all nations.  ---Abraham Lincoln, Second Inaugural Address, from the
memorial}
\end{center}

The expression consists of one or more terms of the same type (integer, dimension, etc.) that are added or subtracted. A term that is a type of number, dimension, etc., consists of a factor of that type, optionally multiplied or divided by factors. A factor of a type is either a quantity of that type or a parenthesized subexpression. The expression produces a result of the given type, so that \numexpr produces an integer, \dimexpr produces a dimension, etc.

In the quotation example above, changing to \dimexpr\linewidth-3*\offset gives the error Illegal unit of measure (pt inserted). This is because for \dimexpr and \glueexpr, the input consists of a dimension or glue value followed by an optional multiplication factor, and not the other way around. Thus \the\dimexpr 1pt*10\relax is valid and produces ‘10.0pt’, but \the\dimexpr 10*1pt\relax gives the Illegal unit error.

The expressions absorb tokens and carry out appropriate mathematics up to a \relax (which will be absorbed), or up to the first non-valid token. Thus, \the\numexpr2+3px will print ‘5px’, because LaTeX reads the \numexpr2+3, which is made up of numbers, and then finds the letter p, which cannot be part of a number. It therefore terminates the expression and produces the ‘5’, followed by the regular text ‘px’.

This termination behavior is useful in comparisons. In \ifnum\numexpr\parindent*2 < 10pt Yes\else No\fi, the less than sign terminates the expression and the result is ‘No’ (in a standard LaTeX article).

Expressions may use the operators +, -, * and / along with parentheses for subexpressions, (...). In glue expressions the plus and minus parts do not need parentheses to be affected by a factor. So \the\glueexpr 5pt plus 1pt * 2 \relax results in ‘10pt plus 2pt’.

TeX will coerce other numerical types in the same way as it does when doing register assignment. Thus \the\numexpr\dimexpr 1pt\relax\relax will result in ‘65536’, which is 1pt converted to scaled points (see sp, TeX’s internal unit) and then coerced into an integer. With a \glueexpr here, the stretch and shrink would be dropped. Going the other way, a \numexpr inside a \dimexpr or \glueexpr will need appropriate units, as in \the\dimexpr\numexpr 1 + 2\relax pt\relax, which produces ‘3.0pt’.

The details of the arithmetic: each factor is checked to be in the allowed range, numbers must be less than 2^{31} in absolute value, and dimensions or glue components must be less than 2^{14} points, or mu, or fil, etc. The arithmetic operations are performed individually, except for a scaling operation (a multiplication immediately followed by a division) which is done as one combined operation with a 64-bit product as intermediate value. The result of each operation is again checked to be in the allowed range.

Finally, division and scaling take place with rounding (unlike TeX’s \divide, which truncates). Thus \the\dimexpr 5pt*(3/2)\relax puts ‘10.0pt’ in the document, because it rounds 3/2 to 2, while \the\dimexpr 5pt*(4/3)\relax produces ‘5.0pt’.


15 Making paragraphs

To start a paragraph, just type some text. To end the current paragraph, put an empty line. This is three paragraphs, the separation of which is made by two empty lines.

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?''

A paragraph separator can be made of a sequence of at least one blank line, at least one of which is not terminated by a comment. A blank line is a line that is empty or made only of blank characters such as space or tab. Comments in source code are started with a % and span up to the end of line. In the following example the two columns are identical:

\documentclass[twocolumn]{article}
\begin{document}
First paragraph.

Second paragraph.
\newpage
First paragraph.
      
  % separator lines may contain blank characters.
  		
Second paragraph.
\end{document}

Once LaTeX has gathered all of a paragraph’s contents it divides that content into lines in a way that is optimized over the entire paragraph (see Line breaking).

There are places where a new paragraph is not permitted. Don’t put a blank line in math mode (see Modes); here the blank line before the \end{equation}

\begin{equation}
  2^{|S|} > |S|

\end{equation}

will get you the error ‘Missing $ inserted’. Similarly, the blank line in this \section argument

\section{aaa

bbb}

gets ‘Runaway argument? {aaa ! Paragraph ended before \@sect was complete’.


15.1 \par

Synopsis (note that while reading the input TeX converts any sequence of one or more blank lines to a \par, Making paragraphs):

\par

End the current paragraph. The usual way to separate paragraphs is with a blank line but the \par command is entirely equivalent. This command is robust (see \protect).

This example uses \par rather than a blank line simply for readability.

\newcommand{\syllabusLegalese}{%
  \whatCheatingIs\par\whatHappensWhenICatchYou}

In LR mode the \par command does nothing and is ignored. In paragraph mode, the \par command terminates paragraph mode, switching LaTeX to vertical mode (see Modes).

You cannot use the \par command in a math mode. You also cannot use it in the argument of many commands, such as the sectioning commands, e.g. \section (see Making paragraphs and \newcommand & \renewcommand).

The \par command is not the same as the \paragraph command. The latter is, like \section or \subsection, a sectioning command used by the LaTeX document standard classes (see \subsubsection, \paragraph, \subparagraph).

The \par command is not the same as \newline or the line break double backslash, \\. The difference is that \par ends the paragraph, not just the line, and also triggers the addition of the between-paragraph vertical space \parskip (see \parindent & \parskip).

The output from this example

xyz

\setlength{\parindent}{3in}
\setlength{\parskip}{5in}
\noindent test\indent test1\par test2 

is: after ‘xyz’ there is a vertical skip of 5 inches and then ‘test’ appears, aligned with the left margin. On the same line, there is an empty horizontal space of 3 inches and then ‘test1’ appears. Finally. there is a vertical space of 5 inches, followed by a fresh paragraph with a paragraph indent of 3 inches, and then LaTeX puts the text ‘test2’.


15.2 \indent & \noindent

Synopsis:

\indent

or

\noindent

Go into horizontal mode (see Modes). The \indent command first outputs an empty box whose width is \parindent. These commands are robust (see \protect).

Ordinarily you create a new paragraph by putting in a blank line. See \par, for the difference between this command and \par. To start a paragraph without an indent, or to continue an interrupted paragraph, use \noindent.

In the middle of a paragraph the \noindent command has no effect, because LaTeX is already in horizontal mode there. The \indent command’s only effect is to output a space.

This example starts a fresh paragraph.

... end of the prior paragraph.

\noindent This paragraph is not indented.

and this continues an interrupted paragraph.

The data 

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

\noindent shows this clearly.

To omit indentation in the entire document put \setlength{\parindent}{0pt} in the preamble. If you do that, you may want to also set the length of spaces between paragraphs, \parskip (see \parindent & \parskip).

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 indentfirst.


15.3 \parindent & \parskip

Synopsis:

\setlength{\parindent}{horizontal len}
\setlength{\parskip}{vertical len}

Both are rubber lengths (see Lengths). They affect the indentation of ordinary paragraphs, not paragraphs inside minipages (see minipage), and the vertical space between paragraphs, respectively.

For example, if this is put in the preamble:

\setlength{\parindent}{0em}
\setlength{\parskip}{1ex}

The document will have paragraphs that are not indented, but instead are vertically separated by about the height of a lowercase ‘x’.

In LaTeX standard class documents, the default value for \parindent in one-column documents is 15pt when the default text size is 10pt, 17pt for 11pt, and 1.5em for 12pt. In two-column documents it is 1em. (These values are set before LaTeX calls \normalfont so em is derived from the default font, Computer Modern. If you use a different font then to set \parindent to 1em matching that font, put \AtBeginDocument{\setlength{\parindent}{1em}} in the preamble.)

The default value for \parskip in LaTeX’s standard document classes is 0pt plus1pt.


15.4 Marginal notes

Synopsis, one of:

\marginpar{right}
\marginpar[left]{right}

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 \marginpar occurs.

The margin that LaTeX uses for the note depends on the current layout (see Document class options) and also on \reversemarginpar (see below). If you are using one-sided layout (document option oneside) then it goes in the right margin. If you are using two-sided layout (document option twoside) then it goes in the outside margin. If you are in two-column layout (document option twocolumn) then it goes in the nearest margin.

If you declare \reversemarginpar 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 \normalmarginpar.

When you specify the optional argument left then it is used for a note in the left margin, while the mandatory argument right is used for a note in the right margin.

Normally, a note’s first word will not be hyphenated. You can enable hyphenation there by beginning left or right with \hspace{0pt}.

These parameters affect the formatting of the note:

\marginparpush

Minimum vertical space between notes; default ‘7pt’ for ‘12pt’ documents, ‘5pt’ else. See also page layout parameters marginparpush.

\marginparsep

Horizontal space between the main text and the note; default ‘11pt’ for ‘10pt’ documents, ‘10pt’ else.

\marginparwidth

Width of the note itself; default for a one-sided ‘10pt’ document is ‘90pt’, ‘83pt’ for ‘11pt’, and ‘68pt’ for ‘12pt’; ‘17pt’ more in each case for a two-sided document. In two column mode, the default is ‘48pt’.

The standard LaTeX routine for marginal notes does not prevent notes from falling off the bottom of the page.


16 Math formulas

Produce mathematical text by putting LaTeX into math mode or display math mode (see Modes). This example shows both.

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.

Math mode is for inline mathematics. In the above example it is invoked by the starting \( and finished by the matching ending \). Display math mode is for displayed equations and here is invoked by the displaymath environment. Note that any mathematical text whatever, including mathematical text consisting of just one character, is handled in math mode.

When in math mode or display math mode, LaTeX handles many aspects of your input text differently than in other text modes. For example,

contrast x+y with \( x+y \)

in math mode the letters are in italics and the spacing around the plus sign is different.

There are three ways to make inline formulas, to put LaTeX in math mode.

\( mathematical material \)
$ mathematical material $
\begin{math} mathematical material \end{math}

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 \(x\) and $y$. You can use these in paragraph mode or in LR mode (see Modes).

To make displayed formulas, put LaTeX into display math mode with either:

\begin{displaymath}
  mathematical material
\end{displaymath}

or

\begin{equation}
  mathematical material
\end{equation}

(see displaymath, see equation). The only difference is that with the equation environment, LaTeX puts a formula number alongside the formula. The construct \[ math \] is equivalent to \begin{displaymath} math \end{displaymath}. These environments can only be used in paragraph mode (see Modes).

The American Mathematical Society has made freely available a set of packages that greatly expand your options for writing mathematics, amsmath and amssymb (also be aware of the mathtools package that is an extension to, and loads, amsmath). 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.


16.1 Subscripts & superscripts

Synopsis (in math mode or display math mode), one of:

base^exp
base^{exp}

or, one of:

base_exp
base_{exp}

Make exp appear as a superscript of base (with the caret character, ^) or a subscript (with underscore, _).

In this example the 0’s and 1’s are subscripts while the 2’s are superscripts.

\( (x_0+x_1)^2 \leq (x_0)^2+(x_1)^2 \)

To have the subscript or superscript contain more than one character, surround the expression with curly braces, as in e^{-2x}. This example’s fourth line shows curly braces used to group an expression for the exponent.

\begin{displaymath}
  (3^3)^3=27^3=19\,683
  \qquad
  3^{(3^3)}=3^{27}=7\,625\,597\,484\,987
\end{displaymath}

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 e^{x^2} and x_{i_0} give correct output. Note the use in those expressions of curly braces to give the base a determined exp. If you enter \(3^3^3\), this interpreted as \(3^{3}^{3}\) and then you get TeX error ‘Double superscript’.

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.

\begin{displaymath}
  \int_{x=a}^b f'(x)\,dx = f(b)-f(a)
\end{displaymath}

Note the curly braces around x=a to make the entire expression a subscript.

To put a superscript or subscript before a symbol, use a construct like {}_t K^2. The empty curly braces {} give the subscript something to attach to and keeps it from accidentally attaching to a prior symbols.

Using the subscript or superscript character outside of math mode or display math mode, as in the expression x^2, will get you the TeX error ‘Missing $ inserted’.

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


16.2 Math symbols

LaTeX provides almost any mathematical or technical symbol that anyone uses. For example, if you include $\pi$ in your source, you will get the pi symbol π. See the “Comprehensive LaTeX Symbol List” package at https://ctan.org/pkg/comprehensive.

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 \ensuremath.

\|

∥ Parallel (relation). Synonym: \parallel.

\aleph

ℵ Aleph, transfinite cardinal (ordinary).

\alpha

α Lowercase Greek letter alpha (ordinary).

\amalg

⨿ Disjoint union (binary)

\angle

∠ Geometric angle (ordinary). Similar: less-than sign < and angle bracket \langle.

\approx

≈ Almost equal to (relation).

\ast

∗ Asterisk operator, convolution, six-pointed (binary). Synonym: *, which is often a superscript or subscript, as in the Kleene star. Similar: \star, which is five-pointed, and is sometimes used as a general binary operation, and sometimes reserved for cross-correlation.

\asymp

≍ Asymptotically equivalent (relation).

\backslash

\ Backslash (ordinary). Similar: set minus \setminus, and \textbackslash for backslash outside of math mode.

\beta

β Lowercase Greek letter beta (ordinary).

\bigcap

⋂ Variable-sized, or n-ary, intersection (operator). Similar: binary intersection \cap.

\bigcirc

⚪ Circle, larger (binary). Similar: function composition \circ.

\bigcup

⋃ Variable-sized, or n-ary, union (operator). Similar: binary union \cup.

\bigodot

⨀ Variable-sized, or n-ary, circled dot operator (operator).

\bigoplus

⨁ Variable-sized, or n-ary, circled plus operator (operator).

\bigotimes

⨂ Variable-sized, or n-ary, circled times operator (operator).

\bigtriangledown

▽ Variable-sized, or n-ary, open triangle pointing down (binary). Synonym: \varbigtriangledown.

\bigtriangleup

△ Variable-sized, or n-ary, open triangle pointing up (binary). Synonym: \varbigtriangleup.

\bigsqcup

⨆ Variable-sized, or n-ary, square union (operator).

\biguplus

⨄ Variable-sized, or n-ary, union operator with a plus (operator). (Note that the name has only one p.)

\bigvee

⋁ Variable-sized, or n-ary, logical-or (operator).

\bigwedge

⋀ Variable-sized, or n-ary, logical-and (operator).

\bot

⊥, Up tack, bottom, least element of a partially ordered set, or a contradiction (ordinary). See also \top.

\bowtie

⋈ Natural join of two relations (relation).

\Box

□ Modal operator for necessity; square open box (ordinary). Not available in plain TeX. In LaTeX you need to load the amssymb package.

\bullet

• Bullet (binary). Similar: multiplication dot \cdot.

\cap

∩ Intersection of two sets (binary). Similar: variable-sized operator \bigcap.

\cdot

⋅ Multiplication (binary). Similar: Bullet dot \bullet.

\chi

χ Lowercase Greek chi (ordinary).

\circ

∘ Function composition, ring operator (binary). Similar: variable-sized operator \bigcirc.

\clubsuit

♣ Club card suit (ordinary).

\complement

∁, Set complement, used as a superscript as in $S^\complement$ (ordinary). Not available in plain TeX. In LaTeX you need to load the amssymb package. Also used: $S^{\mathsf{c}}$ or $\bar{S}$.

\cong

≅ Congruent (relation).

\coprod

∐ Coproduct (operator).

\cup

∪ Union of two sets (binary). Similar: variable-sized operator \bigcup.

\dagger

† Dagger relation (binary).

\dashv

⊣ Dash with vertical, reversed turnstile (relation). Similar: turnstile \vdash.

\ddagger

‡ Double dagger relation (binary).

\Delta

Δ Greek uppercase delta, used for increment (ordinary).

\delta

δ Greek lowercase delta (ordinary).

\Diamond

◇ Large diamond operator (ordinary). Not available in plain TeX. In LaTeX you need to load the amssymb package.

\diamond

⋄ Diamond operator (binary). Similar: large diamond \Diamond, circle bullet \bullet.

\diamondsuit

♢ Diamond card suit (ordinary).

\div

÷ Division sign (binary).

\doteq

≐ Approaches the limit (relation). Similar: geometrically equal to \Doteq.

\downarrow

↓ Down arrow, converges (relation). Similar: \Downarrow double line down arrow.

\Downarrow

⇓ Double line down arrow (relation). Similar: \downarrow single line down arrow.

\ell

ℓ Lowercase cursive letter l (ordinary).

\emptyset

∅ Empty set symbol (ordinary). The variant form is \varnothing.

\epsilon

ϵ Lowercase lunate epsilon (ordinary). Similar to Greek text letter. More widely used in mathematics is the script small letter epsilon \varepsilon ε. Related: the set membership relation \in ∈.

\equiv

≡ Equivalence (relation).

\eta

η Lowercase Greek letter (ordinary).

\exists

∃ Existential quantifier (ordinary).

\flat

♭ Musical flat (ordinary).

\forall

∀ Universal quantifier (ordinary).

\frown

⌢ Downward curving arc (ordinary).

\Gamma

Γ uppercase Greek letter (ordinary).

\gamma

γ Lowercase Greek letter (ordinary).

\ge

≥ Greater than or equal to (relation). This is a synonym for \geq.

\geq

≥ Greater than or equal to (relation). This is a synonym for \ge.

\gets

← Is assigned the value (relation). Synonym: \leftarrow.

\gg

≫ Much greater than (relation). Similar: much less than \ll.

\hbar

ℏ Planck constant over two pi (ordinary).

\heartsuit

♡ Heart card suit (ordinary).

\hookleftarrow

↩ Hooked left arrow (relation).

\hookrightarrow

↪ Hooked right arrow (relation).

\iff

⟷ If and only if (relation). It is \Longleftrightarrow with a \thickmuskip on either side.

\Im

ℑ Imaginary part (ordinary). See: real part \Re.

\imath

Dotless i; used when you are putting an accent on an i (see Math accents).

\in

∈ Set element (relation). See also: lowercase lunate epsilon \epsilonϵ and small letter script epsilon \varepsilon.

\infty

∞ Infinity (ordinary).

\int

∫ Integral (operator).

\iota

ι Lowercase Greek letter (ordinary).

\Join

⨝ Condensed bowtie symbol (relation). Not available in Plain TeX.

\jmath

Dotless j; used when you are putting an accent on a j (see Math accents).

\kappa

κ Lowercase Greek letter (ordinary).

\Lambda

Λ uppercase Greek letter (ordinary).

\lambda

λ Lowercase Greek letter (ordinary).

\land

∧ Logical and (binary). Synonym: \wedge. See also logical or \lor.

\langle

⟨ Left angle, or sequence, bracket (opening). Similar: less-than <. Matches \rangle.

\lbrace

{ Left curly brace (opening). Synonym: \{. Matches \rbrace.

\lbrack

[ Left square bracket (opening). Synonym: [. Matches \rbrack.

\lceil

⌈ Left ceiling bracket, like a square bracket but with the bottom shaved off (opening). Matches \rceil.

\le

≤ Less than or equal to (relation). This is a synonym for \leq.

\leadsto

⇝ Squiggly right arrow (relation). To get this symbol outside of math mode you can put \newcommand*{\Leadsto}{\ensuremath{\leadsto}} in the preamble and then use \Leadsto instead.

\Leftarrow

⇐ Is implied by, double-line left arrow (relation). Similar: single-line left arrow \leftarrow.

\leftarrow

← Single-line left arrow (relation). Synonym: \gets. Similar: double-line left arrow \Leftarrow.

\leftharpoondown

↽ Single-line left harpoon, barb under bar (relation).

\leftharpoonup

↼ Single-line left harpoon, barb over bar (relation).

\Leftrightarrow

⇔ Bi-implication; double-line double-headed arrow (relation). Similar: single-line double headed arrow \leftrightarrow.

\leftrightarrow

↔ Single-line double-headed arrow (relation). Similar: double-line double headed arrow \Leftrightarrow.

\leq

≤ Less than or equal to (relation). This is a synonym for \le.

\lfloor

⌊ Left floor bracket (opening). Matches: \floor.

\lhd

◁ Arrowhead, that is, triangle, pointing left (binary). For the normal subgroup symbol you should load amssymb and use \vartriangleleft (which is a relation and so gives better spacing).

\ll

≪ Much less than (relation). Similar: much greater than \gg.

\lnot

¬ Logical negation (ordinary). Synonym: \neg.

\longleftarrow

⟵ Long single-line left arrow (relation). Similar: long double-line left arrow \Longleftarrow.

\longleftrightarrow

⟷ Long single-line double-headed arrow (relation). Similar: long double-line double-headed arrow \Longleftrightarrow.

\longmapsto

⟼ Long single-line left arrow starting with vertical bar (relation). Similar: shorter version \mapsto.

\longrightarrow

⟶ Long single-line right arrow (relation). Similar: long double-line right arrow \Longrightarrow.

\lor

∨ Logical or (binary). Synonym: \vee. See also logical and \land.

\mapsto

↦ Single-line left arrow starting with vertical bar (relation). Similar: longer version \longmapsto.

\mho

℧ Conductance, half-circle rotated capital omega (ordinary).

\mid

∣ Single-line vertical bar (relation). A typical use of \mid is for a set \{\, x \mid x\geq 5 \,\}.

Similar: \vert and | 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 \vert and for norm see the entry for \Vert.

\models

⊨ Entails, or satisfies; double turnstile, short double dash (relation). Similar: long double dash \vDash.

\mp

∓ Minus or plus (relation).

\mu

μ Lowercase Greek letter (ordinary).

\nabla

∇ Hamilton’s del, or differential, operator (ordinary).

\natural

♮ Musical natural notation (ordinary).

\ne

≠ Not equal (relation). Synonym: \neq.

\nearrow

↗ North-east arrow (relation).

\neg

¬ Logical negation (ordinary). Synonym: \lnot. Sometimes instead used for negation: \sim.

\neq

≠ Not equal (relation). Synonym: \ne.

\ni

∋ Reflected membership epsilon; has the member (relation). Synonym: \owns. Similar: is a member of \in.

\not

    Long solidus, or slash, used to overstrike a following operator (relation).

Many negated operators are available that don’t require \not, particularly with the amssymb package. For example, \notin is typographically preferable to \not\in.

\notin

∉ Not an element of (relation). Similar: not subset of \nsubseteq.

\nu

ν Lowercase Greek letter (ordinary).

\nwarrow

↖ North-west arrow (relation).

\odot

⊙ Dot inside a circle (binary). Similar: variable-sized operator \bigodot.

\oint

∮ Contour integral, integral with circle in the middle (operator).

\Omega

Ω uppercase Greek letter (ordinary).

\omega

ω Lowercase Greek letter (ordinary).

\ominus

⊖ Minus sign, or dash, inside a circle (binary).

\oplus

⊕ Plus sign inside a circle (binary). Similar: variable-sized operator \bigoplus.

\oslash

⊘ Solidus, or slash, inside a circle (binary).

\otimes

⊗ Times sign, or cross, inside a circle (binary). Similar: variable-sized operator \bigotimes.

\owns

∋ Reflected membership epsilon; has the member (relation). Synonym: \ni. Similar: is a member of \in.

\parallel

∥ Parallel (relation). Synonym: \|.

\partial

∂ Partial differential (ordinary).

\perp

⟂ Perpendicular (relation). Similar: \bot uses the same glyph but the spacing is different because it is in the class ordinary.

\Phi

Φ Uppercase Greek letter (ordinary).

\phi

ϕ Lowercase Greek letter (ordinary). The variant form is \varphi φ.

\Pi

Π uppercase Greek letter (ordinary).

\pi

π Lowercase Greek letter (ordinary). The variant form is \varpi ϖ.

\pm

± Plus or minus (binary).

\prec

≺ Precedes (relation). Similar: less than <.

\preceq

⪯ Precedes or equals (relation). Similar: less than or equals \leq.

\prime

′ Prime, or minute in a time expression (ordinary). Typically used as a superscript: $f^\prime$; $f^\prime$ and $f'$ produce the same result. An advantage of the second is that $f'''$ produces the desired symbol, that is, the same result as $f^{\prime\prime\prime}$, but uses rather less typing. You can only use \prime in math mode. Using the right single quote ' in text mode produces a different character (apostrophe).

\prod

∏ Product (operator).

\propto

∝ Is proportional to (relation)

\Psi

Ψ uppercase Greek letter (ordinary).

\psi

ψ Lowercase Greek letter (ordinary).

\rangle

⟩ Right angle, or sequence, bracket (closing). Similar: greater than >. Matches:\langle.

\rbrace

} Right curly brace (closing). Synonym: \}. Matches \lbrace.

\rbrack

] Right square bracket (closing). Synonym: ]. Matches \lbrack.

\rceil

⌉ Right ceiling bracket (closing). Matches \lceil.

\Re

ℜ Real part, real numbers, cursive capital R (ordinary). Related: double-line, or blackboard bold, R \mathbb{R}; to access this, load the amsfonts package.

\restriction

↾, Restriction of a function (relation). Synonym: \upharpoonright. Not available in plain TeX. In LaTeX you need to load the amssymb package.

\revemptyset

⦰, Reversed empty set symbol (ordinary). Related: \varnothing. Not available in plain TeX. In LaTeX you need to load the stix package.

\rfloor

⌋ Right floor bracket, a right square bracket with the top cut off (closing). Matches \lfloor.

\rhd

◁ Arrowhead, that is, triangle, pointing right (binary). For the normal subgroup symbol you should instead load amssymb and use \vartriangleright (which is a relation and so gives better spacing).

\rho

ρ Lowercase Greek letter (ordinary). The variant form is \varrho ϱ.

\Rightarrow

⇒ Implies, right-pointing double line arrow (relation). Similar: right single-line arrow \rightarrow.

\rightarrow

→ Right-pointing single line arrow (relation). Synonym: \to. Similar: right double line arrow \Rightarrow.

\rightharpoondown

⇁ Right-pointing harpoon with barb below the line (relation).

\rightharpoonup

⇀ Right-pointing harpoon with barb above the line (relation).

\rightleftharpoons

⇌ Right harpoon up above left harpoon down (relation).

\searrow

↘ Arrow pointing southeast (relation).

\setminus

⧵ Set difference, reverse solidus or reverse slash, like \ (binary). Similar: backslash \backslash and also \textbackslash outside of math mode.

\sharp

♯ Musical sharp (ordinary).

\Sigma

Σ uppercase Greek letter (ordinary).

\sigma

σ Lowercase Greek letter (ordinary). The variant form is \varsigma ς.

\sim

∼ Similar, in a relation (relation).

\simeq

≃ Similar or equal to, in a relation (relation).

\smallint

∫ Integral sign that does not change to a larger size in a display (operator).

\smile

⌣ Upward curving arc, smile (ordinary).

\spadesuit

♠ Spade card suit (ordinary).

\sqcap

⊓ Square intersection symbol (binary). Similar: intersection cap.

\sqcup

⊔ Square union symbol (binary). Similar: union cup. Related: variable-sized operator \bigsqcup.

\sqsubset

⊏, Square subset symbol (relation). Similar: subset \subset. Not available in plain TeX. In LaTeX you need to load the amssymb package.

\sqsubseteq

⊑ Square subset or equal symbol (binary). Similar: subset or equal to \subseteq.

\sqsupset

⊐, Square superset symbol (relation). Similar: superset \supset. Not available in plain TeX. In LaTeX you need to load the amssymb package.

\sqsupseteq

⊒ Square superset or equal symbol (binary). Similar: superset or equal \supseteq.

\star

⋆ Five-pointed star, sometimes used as a general binary operation but sometimes reserved for cross-correlation (binary). Similar: the synonyms asterisk * and \ast, which are six-pointed, and more often appear as a superscript or subscript, as with the Kleene star.

\subset

⊂ Subset (occasionally, is implied by) (relation).

\subseteq

⊆ Subset or equal to (relation).

\succ

≻ Comes after, succeeds (relation). Similar: is less than >.

\succeq

⪰ Succeeds or is equal to (relation). Similar: less than or equal to \leq.

\sum

∑ Summation (operator). Similar: Greek capital sigma \Sigma.

\supset

⊃ Superset (relation).

\supseteq

⊇ Superset or equal to (relation).

\surd

√ Radical symbol (ordinary). The LaTeX command \sqrt{...} typesets the square root of the argument, with a bar that extends to cover the argument.

\swarrow

↙ Southwest-pointing arrow (relation).

\tau

τ Lowercase Greek letter (ordinary).

\theta

θ Lowercase Greek letter (ordinary). The variant form is \vartheta ϑ.

\times

× Primary school multiplication sign (binary). See also \cdot.

\to

→ Right-pointing single line arrow (relation). Synonym: \rightarrow.

\top

⊤ Top, greatest element of a partially ordered set (ordinary). See also \bot.

\triangle

△ Triangle (ordinary).

\triangleleft

◁ Not-filled triangle pointing left (binary). Similar: \lhd. For the normal subgroup symbol you should load amssymb and use \vartriangleleft (which is a relation and so gives better spacing).

\triangleright

▷ Not-filled triangle pointing right (binary). For the normal subgroup symbol you should instead load amssymb and use \vartriangleright (which is a relation and so gives better spacing).

\unlhd

⊴ Left-pointing not-filled underlined arrowhead, that is, triangle, with a line under (binary). For the normal subgroup symbol load amssymb and use \vartrianglelefteq (which is a relation and so gives better spacing).

\unrhd

⊵ Right-pointing not-filled underlined arrowhead, that is, triangle, with a line under (binary). For the normal subgroup symbol load amssymb and use \vartrianglerighteq (which is a relation and so gives better spacing).

\Uparrow

⇑ Double-line upward-pointing arrow (relation). Similar: single-line up-pointing arrow \uparrow.

\uparrow

↑ Single-line upward-pointing arrow, diverges (relation). Similar: double-line up-pointing arrow \Uparrow.

\Updownarrow

⇕ Double-line upward-and-downward-pointing arrow (relation). Similar: single-line upward-and-downward-pointing arrow \updownarrow.

\updownarrow

↕ Single-line upward-and-downward-pointing arrow (relation). Similar: double-line upward-and-downward-pointing arrow \Updownarrow.

\upharpoonright

↾, Up harpoon, with barb on right side (relation). Synonym: \restriction. Not available in plain TeX. In LaTeX you need to load the amssymb package.

\uplus

⊎ Multiset union, a union symbol with a plus symbol in the middle (binary). Similar: union \cup. Related: variable-sized operator \biguplus.

\Upsilon

Υ uppercase Greek letter (ordinary).

\upsilon

υ Lowercase Greek letter (ordinary).

\varepsilon

ε Small letter script epsilon (ordinary). This is more widely used in mathematics than the non-variant lunate epsilon form \epsilon ϵ. Related: set membership \in.

\vanothing

∅, Empty set symbol. Similar: \emptyset. Related: \revemptyset. Not available in plain TeX. In LaTeX you need to load the amssymb package.

\varphi

φ Variant on the lowercase Greek letter (ordinary). The non-variant form is \phi ϕ.

\varpi

ϖ Variant on the lowercase Greek letter (ordinary). The non-variant form is \pi π.

\varrho

ϱ Variant on the lowercase Greek letter (ordinary). The non-variant form is \rho ρ.

\varsigma

ς Variant on the lowercase Greek letter (ordinary). The non-variant form is \sigma σ.

\vartheta

ϑ Variant on the lowercase Greek letter (ordinary). The non-variant form is \theta θ.

\vdash

⊢ Provable; turnstile, vertical and a dash (relation). Similar: turnstile rotated a half-circle \dashv.

\vee

∨ Logical or; a downwards v shape (binary). Related: logical and \wedge. Similar: variable-sized operator \bigvee.

\Vert

‖ Vertical double bar (ordinary). See Delimiters, for how to use the mathtools package to create flexibly-sized norm symbols.

\vert

| Single line vertical bar (ordinary). For “such that”, as in the definition of a set, use \mid because it is a relation. See Delimiters, for how to use the mathtools package to create flexibly-sized absolute-value symbols.

\wedge

∧ Logical and (binary). Synonym: \land. See also logical or \vee. Similar: variable-sized operator \bigwedge.

\wp

℘ Weierstrass p (ordinary).

\wr

≀ Wreath product (binary).

\Xi

Ξ uppercase Greek letter (ordinary).

\xi

ξ Lowercase Greek letter (ordinary).

\zeta

ζ Lowercase Greek letter (ordinary).

The following symbols are most often used in plain text but LaTeX provides versions to use in mathematical text.

\mathdollar

Dollar sign in math mode: $.

\mathparagraph

Paragraph sign (pilcrow) in math mode, ¶.

\mathsection

Section sign in math mode: §.

\mathsterling

Sterling sign in math mode: £.

\mathunderscore

Underscore in math mode: _.


16.2.1 Arrows

These are the arrows that come with standard LaTeX. The latexsym and amsfonts packages contain many more.

SymbolCommand
\Downarrow
\downarrow
\hookleftarrow
\hookrightarrow
\leftarrow
\Leftarrow
\Leftrightarrow
\leftrightarrow
\longleftarrow
\Longleftarrow
\longleftrightarrow
\Longleftrightarrow
\longmapsto
\Longrightarrow
\longrightarrow
\mapsto
\nearrow
\nwarrow
\Rightarrow
\rightarrow, or \to
\searrow
\swarrow
\uparrow
\Uparrow
\updownarrow
\Updownarrow

An example of the difference between \to and \mapsto is: \( f\colon D\to C \) given by \( n\mapsto n^2 \).

For commutative diagrams there are a number of packages, including tikz-cd and amscd.


16.2.2 \boldmath & \unboldmath

Synopsis (used in paragraph mode or LR mode):

\boldmath \( math \)

or

\unboldmath \( math \)

Declarations to change the letters and symbols in math to be in a bold font, or to countermand that and bring back the regular (non-bold) default, respectively. They must be used when not in math mode or display math mode (see Modes). Both commands are fragile (see \protect).

In this example each \boldmath command takes place inside an \mbox,

we have $\mbox{\boldmath \( v \)} = 5\cdot\mbox{\boldmath \( u \)$}$

which means \boldmath is only called in a text mode, here LR mode, and explains why we must switch LaTeX into math mode to set v and u.

If you use either command inside math mode, as with Trouble: \( \boldmath x \), then you get something like ‘LaTeX Font Warning: Command \boldmath invalid in math mode’ and ‘LaTeX Font Warning: Command \mathversion invalid in math mode’.


16.2.2.1 bm: Individual bold math symbols

Specifying \boldmath is the best method for typesetting a whole math expression in bold. But to typeset individual symbols within an expression in bold, the bm package provided by the LaTeX Project team is better. Its usage is outside the scope of this document (see its documentation at https://ctan.org/pkg/bm or in your installation) but the spacing in the output of this small example will show that it is an improvement over \boldmath within an expression:

\usepackage{bm}   % in preamble
...
we have $\bm{v} = 5\cdot\bm{u}$

16.2.2.2 OpenType bold math

Unfortunately, when using the Unicode engines (XeLaTeX, LuaLaTeX), neither \boldmath nor bm usually work well, because the OpenType math fonts normally used with those engines rarely come with a bold companion, and both \boldmath and bm require this. (The implementation of bm relies on \boldmath, so the requirements are the same.) If you do have a bold math font, though, then \boldmath and bm work fine.

If no such font is available, one alternative is to construct fake bold fonts with the fontspec package’s FakeBold=1 parameter (see its documentation, https://ctan.org/pkg/fontspec). This may be acceptable for drafting or informal distribution, but the results are far from a true bold font.

Another alternative to handling bold for OpenType math fonts is to use the \symbf (bold), \symbfit (bold italic), and related commands from the unicode-math package. These do not change the current font, but rather change the (Unicode) “alphabet” used, which in practice is more widely supported than a separate bold font. Many variations are possible, and so there are subtleties to getting the desired output. As usual, see the package documentation (https://ctan.org/pkg/unicode-math).


16.2.3 Blackboard bold

Synopsis:

\usepackage{amssymb}   % in preamble
  ... 
\mathbb{uppercase-letter}

Provide blackboard bold symbols, sometimes also known as doublestruck letters, used to denote number sets such as the natural numbers, the integers, etc.

Here

\( \forall n \in \mathbb{N}, n^2 \geq 0 \)

the \mathbb{N} gives blackboard bold symbol ℕ, representing the natural numbers.

If the argument contains something other than an uppercase letter, you do not get an error but you do get strange results, including unexpected characters.

There are packages that give access to symbols other than just the capital letters; look on CTAN.


16.2.4 Calligraphic

Synopsis:

\mathcal{uppercase-letters}

Use a script-like font.

In this example the graph identifier is output in a cursive font.

Let the graph be \( \mathcal{G} \).

If you use something other than an uppercase letter then you do not get an error but you also do not get math calligraphic output. For instance, \mathcal{g} outputs a close curly brace symbol.


16.2.5 Delimiters

Delimiters are parentheses, braces, or other characters used to mark the start and end of subformulas. This formula has three sets of parentheses delimiting the three subformulas.

(z-z_0)^2 = (x-x_0)^2 + (y-y_0)^2

The delimiters do not need to match, so you can enter \( [0,1) \).

Here are the common delimiters:

DelimiterCommandName
((Left parenthesis
))Right parenthesis
\}{ or \lbraceLeft brace
\{} or \rbraceRight brace
[[ or \lbrackLeft bracket
]] or \rbrackRight bracket
\lfloorLeft floor bracket
\rfloorRight floor bracket
\lceilLeft ceiling bracket
\rceilRight ceiling bracket
\langleLeft angle bracket
\rangleRight angle bracket
//Slash, or forward slash
\\backslashReverse slash, or backslash
|| or \vertVertical bar
\| or \VertDouble vertical bar

The mathtools package allows you to create commands for paired delimiters. For instance, if you put \DeclarePairedDelimiter\abs{\lvert}{\rvert} in your preamble then you get two commands for single-line vertical bars (they only work in math mode). The starred form, such as \abs*{\frac{22}{7}}, has the height of the vertical bars match the height of the argument. The unstarred form, such as \abs{\frac{22}{7}}, has the bars fixed at a default height. This form accepts an optional argument, as in \abs[size command]{\frac{22}{7}}, where the height of the bars is given in size command, such as \Bigg. Using instead \lVert and \rVert as the symbols will give you a norm symbol with the same behavior.


16.2.5.1 \left & \right

Synopsis:

\left delimiter1 ... \right delimiter2

Make matching parentheses, braces, or other delimiters. LaTeX makes the delimiters tall enough to just cover the size of the formula that they enclose.

This makes a unit vector surrounded by parentheses tall enough to cover the entries.

\begin{equation}
  \left(\begin{array}{c}
    1   \\
    0   \\
  \end{array}\right)
\end{equation}

See Delimiters, for a list of the common delimiters.

Every \left must have a matching \right. In the above example, leaving out the \left( gets the error message ‘Extra \right’. Leaving out the \right) gets ‘You can't use `\eqno' in math mode’.

However, delimiter1 and delimiter2 need not match. A common case is that you want an unmatched brace, as below. Use a period, ‘.’, as a null delimiter.

\begin{equation}
  f(n)=\left\{\begin{array}{ll}
                1             &\mbox{--if \(n=0\)} \\
                f(n-1)+3n^2   &\mbox{--else}       
       \end{array}\right.
\end{equation}

Note that to get a curly brace as a delimiter you must prefix it with a backslash, \{ (see Reserved characters). (The packages amsmath and mathtools allow you to get the above construct through in a cases environment.)

The \left ... \right pair make a group. One consequence is that the formula enclosed in the \left ... \right pair cannot have line breaks in the output. This includes both manual line breaks and LaTeX-generated automatic ones. In this example, LaTeX breaks the equation to make the formula fit the margins.

Lorem ipsum dolor sit amet
\( (a+b+c+d+e+f+g+h+i+j+k+l+m+n+o+p+q+r+s+t+u+v+w+x+y+z) \)

But with \left and \right

Lorem ipsum dolor sit amet
\( \left(a+b+c+d+e+f+g+h+i+j+k+l+m+n+o+p+q+r+s+t+u+v+w+x+y+z\right) \)

LaTeX won’t break the line, causing the formula to extend into the margin.

Because \left ... \right make a group, all the usual grouping rules hold. Here, the value of \testlength set inside the equation will be forgotten, and the output is ‘1.2pt’.

\newlength{\testlength} \setlength{\testlength}{1.2pt}
\begin{equation}
  \left( a+b=c \setlength{\testlength}{3.4pt} \right)
  \the\testlength
\end{equation}

The \left ... \right pair affect the horizontal spacing of the enclosed formula, in two ways. The first is that in \( \sin(x) = \sin\left(x\right) \) the one after the equals sign has more space around the x. That’s because \left( ... \right) inserts an inner node while ( ... ) inserts an opening node. The second way that the pair affect the horizontal spacing is that because they form a group, the enclosed subformula will be typeset at its natural width, with no stretching or shrinking to make the line fit better.

TeX scales the delimiters according to the height and depth of the enclosed formula. Here LaTeX grows the brackets to extend the full height of the integral.

\begin{equation}
  \left[ \int_{x=r_0}^{\infty} -G\frac{Mm}{r^2}\, dr \right] 
\end{equation}

Manual sizing is often better. For instance, although below the rule has no depth, TeX will create delimiters that extend far below the rule.

\begin{equation}
  \left( \rule{1pt}{1cm} \right)
\end{equation}

TeX can choose delimiters that are too small, as in \( \left| |x|+|y| \right| \). It can also choose delimiters that are too large, as here.

\begin{equation}
  \left( \sum_{0\leq i<n} i^k \right)
\end{equation}

A third awkward case is when a long displayed formula is on more than one line and you must match the sizes of the opening and closing delimiter; you can’t use \left on the first line and \right on the last because they must be paired.

To size the delimiters manually, see \bigl, \bigr, etc..


16.2.5.2 \bigl, \bigr, etc.

Synopsis, one of:

\bigldelimiter1 ... \bigrdelimiter2
\Bigldelimiter1 ... \bigrdelimiter2
\biggldelimiter1 ... \biggrdelimiter2
\Biggldelimiter1 ... \Biggrdelimiter2  

(as with \bigl[...\bigr]; strictly speaking they need not be paired, see below), or one of:

\bigmdelimiter
\Bigmdelimiter
\biggmdelimiter
\Biggmdelimiter

(as with \bigm|), or one of:

\bigdelimiter
\Bigdelimiter
\biggdelimiter
\Biggdelimiter

(as with \big[).

Produce manually-sized delimiters. For delimiters that are automatically sized see \left & \right).

This produces slightly larger outer vertical bars.

  \bigl| |x|+|y| \bigr|

The commands above are listed in order of increasing size. You can use the smallest size such as \bigl...\bigr in a paragraph without causing LaTeX to spread the lines apart. The larger sizes are meant for displayed equations.

See Delimiters, for a list of the common delimiters. In the family of commands with ‘l’ or ‘r’, delimiter1 and delimiter2 need not match together.

The ‘l’ and ‘r’ commands produce open and close delimiters that insert no horizontal space between a preceding atom and the delimiter, while the commands without ‘l’ and ‘r’ insert some space (because each delimiter is set as an ordinary variable). Compare these two.

\begin{tabular}{l}
  \(\displaystyle \sin\biggl(\frac{1}{2}\biggr) \)  \\  % good
  \(\displaystyle \sin\bigg(\frac{1}{2}\bigg)  \)   \\  % bad
\end{tabular}

The traditional typographic treatment is on the first line. On the second line the output will have some extra space between the \sin and the open parenthesis.

Commands without ‘l’ or ‘r’ do give correct spacing in some circumstances, as with this large vertical line

\begin{equation}
  \int_{x=a}^b x^2\,dx = \frac{1}{3} x^3 \Big|_{x=a}^b
\end{equation}

(many authors would replace \frac with the \tfrac command from the amsmath package), and as with this larger slash.

\begin{equation}
  \lim_{n\to\infty}\pi(n) \big/ (n/\log n) = 1
\end{equation}

Unlike the \left...\right pair (see \left & \right), the commands here with ‘l’ or ‘r’ do not make a group. Strictly speaking they need not be matched so you can write something like this.

\begin{equation}
  \Biggl[ \pi/6 ] 
\end{equation}

The commands with ‘m’ are for relations, which are in the middle of formulas, as here.

\begin{equation}
  \biggl\{ a\in B \biggm| a=\sum_{0\leq i<n}3i^2+4 \biggr\}
\end{equation}  

16.2.6 Dots, horizontal or vertical

Ellipses are the three dots (usually three) indicating that a pattern continues.

\begin{array}{cccc}
  a_{0,0}    &a_{0,1}   &a_{0,2} &\ldots \\
  a_{1,0}    &\ddots                     \\
  \vdots
\end{array}

LaTeX provides these.

\cdots

Horizontal ellipsis with the dots raised to the center of the line, as in ⋯. Used as: \( a_0\cdot a_1\cdots a_{n-1} \).

\ddots

Diagonal ellipsis, ⋱. See the above array example for a usage.

\ldots
\mathellipsis
\dots

Ellipsis on the baseline, …. Used as: \( x_0,\ldots x_{n-1} \). Another example is the above array example. Synonyms are \mathellipsis and \dots. A synonym from the amsmath package is \hdots.

You can also use this command outside of mathematical text, as in The gears, brakes, \ldots{} are all broken.

\vdots

Vertical ellipsis, ⋮. See the above array example for a usage.

The amsmath package has the command \dots to semantically mark up ellipses. This example produces two different-looking outputs for the first two uses of the \dots command.

\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 \( p_0, p_1, \dotsc \).

In the first line LaTeX looks to the comma following \dots to determine that it should output an ellipsis on the baseline. The second line has a \cdot following \dots 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: \dotsc if you need the ellipsis appropriate for a comma following, \dotsb if you need the ellipses that fits when the dots are followed by a binary operator or relation symbol, \dotsi for dots with integrals, or \dotso for others.

The \dots command from amsmath differs from the LaTeX kernel’s \dots command in another way: it outputs a thin space after the ellipsis. Furthermore, the unicode-math package automatically loads amsmath, so amsmath’s \dots may be active even when you did not explicitly load it, thus changing the output from \dots in both text and math mode.

Yet more about the ellipsis commands: when running under Unicode engines (lualatex, xelatex), LaTeX will use the Unicode ellipsis character (U+2026) in the font if it’s available; under traditional TeX engines (pdflatex, latex), it will typeset three spaced periods. Generally, the Unicode single-character ellipsis has almost no space between the three periods, while the spacing of the non-Unicode ellipsis is looser, more in accordance with traditional typography.


16.2.7 Greek letters

The upper case versions of these Greek letters are only shown when they differ from Roman upper case letters.

SymbolCommandName
α\alphaAlpha
β\betaBeta
γ, Γ\gamma, \GammaGamma
δ, Δ\delta, \DeltaDelta
ε, ϵ\varepsilon, \epsilonEpsilon
ζ\zetaZeta
η\etaEta
θ, ϑ\theta, \varthetaTheta
ι\iotaIota
κ\kappaKappa
λ, Λ\lambda, \LambdaLambda
μ\muMu
ν\nuNu
ξ, Ξ\xi, \XiXi
π, Π\pi, \PiPi
ρ, ϱ\rho, \varrhoRho
σ, Σ\sigma, \SigmaSigma
τ\tauTau
ϕ, φ, Φ\phi, \varphi, \PhiPhi
χ\chichi
ψ, Ψ\psi, \PsiPsi
ω, Ω\omega, \OmegaOmega

For omicron, if you are using LaTeX’s default Computer Modern font then enter omicron just as ‘o’ or ‘O’. If you like having the name or if your font shows a difference then you can use something like \newcommand\omicron{o}. The package unicode-math has \upomicron for upright omicron and \mitomicron for math italic.

While the set membership relation symbol ∈ generated by \in is related to epsilon, it is never used for a variable.


16.3 Math functions

These commands produce roman function names in math mode with proper spacing.

\arccos

Inverse cosine

\arcsin

Inverse sine

\arctan

Inverse tangent

\arg

Angle between the real axis and a point in the complex plane

\bmod

Binary modulo operator, used as in \( 5\bmod 3=2 \)

\cos

Cosine

\cosh

Hyperbolic cosine

\cot

Cotangent

\coth

Hyperbolic cotangent

\csc

Cosecant

\deg

Degrees

\det

Determinant

\dim

Dimension

\exp

Exponential

\gcd

Greatest common divisor

\hom

Homomorphism

\inf

Infimum

\ker

Kernel

\lg

Base 2 logarithm

\lim

Limit

\liminf

Limit inferior

\limsup

Limit superior

\ln

Natural logarithm

\log

Logarithm

\max

Maximum

\min

Minimum

\pmod

Parenthesized modulus, as used in \( 5\equiv 2\pmod 3 \)

\Pr

Probability

\sec

Secant

\sin

Sine

\sinh

Hyperbolic sine

\sup

Supremum sup

\tan

Tangent

\tanh

Hyperbolic tangent

The amsmath 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 \DeclareMathOperator{\identity}{id} that is like the ones above but prints as ‘id’. The starred form \DeclareMathOperator*{\op}{op} sets any superscript or subscript to be above and below, as is traditional with \lim, \sup, or \max.


16.4 Math accents

LaTeX provides a variety of commands for producing accented letters in math. These are different from accents in normal text (see Accents).

\acute

Math acute accent

\bar

Math bar-over accent

\breve

Math breve accent

\check

Math háček (check) accent

\ddot

Math dieresis accent

\dot

Math dot accent

\grave

Math grave accent

\hat

Math hat (circumflex) accent

\mathring

Math ring accent

\tilde

Math tilde accent

\vec

Math vector symbol

\widehat

Math wide hat accent

\widetilde

Math wide tilde accent

When you are putting an accent on an i or a j, the tradition is to use one without a dot, \imath or jmath (see Math symbols).


16.5 Over- or under math

LaTeX provides commands for putting lines, braces, and arrows over or under math material.

\underline{math}

Underline math. For example: \underline{x+y}. The line is always completely below the text, taking account of descenders, so in \(\underline{y}\) the line is lower than in \(\underline{x}\). As of approximately 2019, this command and others in this section are robust; before that, they were fragile (see \protect).

The package ulem (https://ctan.org/pkg/uelem) does text mode underlining and allows line breaking as well as a number of other features. See also \hrulefill & \dotfill for producing a line for such things as a signature or placeholder.

\overline{math}

Put a horizontal line over math. For example: \overline{x+y}. This differs from the accent command \bar (see Math accents).

\underbrace{math}

Put a brace under math. For example: (1-\underbrace{1/2)+(1/2}-1/3).

You can attach text to the brace as a subscript (_) or superscript (^) as here:

\begin{displaymath}
  1+1/2+\underbrace{1/3+1/4}_{>1/2}+
       \underbrace{1/5+1/6+1/7+1/8}_{>1/2}+\cdots
\end{displaymath}

The superscript appears on top of the expression, and so can look unconnected to the underbrace.

\overbrace{math}

Put a brace over math. For example:
\overbrace{x+x+\cdots+x}^{\mbox{\(k\) times}}.

\overrightarrow{math}

Put a right arrow over math. For example: \overrightarrow{x+y}.

\overleftarrow{math}

Put a left arrow over math. For example: \overleftarrow{a+b}.

The package mathtools (https://ctan.org/pkg/mathtools) adds an over- and underbracket, as well as some improvements on the braces.


16.6 Spacing in math mode

When typesetting mathematics, LaTeX puts in spacing according to the normal rules for mathematics texts. If you enter y=m x then LaTeX ignores the space and in the output the m is next to the x, as y=mx.

But LaTeX’s rules occasionally need tweaking. For example, in an integral the tradition is to put a small extra space between the f(x) and the dx, here done with the \, command:

\int_0^1 f(x)\,dx

LaTeX provides the commands that follow for use in math mode. Many of these spacing definitions are expressed in terms of the math unit mu. It is defined as 1/18em, where the em is taken from the current math symbols family (see Units of length). Thus, a \thickspace is something like 5/18 times the width of a ‘M’.

\;

Synonym: \thickspace. Normally 5.0mu plus 5.0mu. With the amsmath package, or as of the 2020-10-01 LaTeX release, can be used in text mode as well as math mode; otherwise, in math mode only.

\negthickspace

Normally -5.0mu plus 2.0mu minus 4.0mu. With the amsmath package, or as of the 2020-10-01 LaTeX release, can be used in text mode as well as math mode; otherwise, in math mode only.

\:
\>

Synonym: \medspace. Normally 4.0mu plus 2.0mu minus 4.0mu. With the amsmath package, or as of the 2020-10-01 LaTeX release, can be used in text mode as well as math mode; before that, in math mode only.

\negmedspace

Normally -4.0mu plus 2.0mu minus 4.0mu. With the amsmath package, or as of the 2020-10-01 LaTeX release, can be used in text mode as well as math mode; before that, in math mode only.

\,

Synonym: \thinspace. Normally 3mu, which is 1/6em. Can be used in both math mode and text mode (see \thinspace & \negthinspace).

This space is widely used, for instance between the function and the infinitesimal in an integral \int f(x)\,dx and, if an author does this, before punctuation in a displayed equation.

The antiderivative is
\begin{equation}
  3x^{-1/2}+3^{1/2}\,.
\end{equation}
\!

Synonym: \negthinspace. A negative thin space. Normally -3mu. With the amsmath package, or as of the 2020-10-01 LaTeX release, can be used in text mode as well as math mode; otherwise, the \! command is math mode only but the \negthinspace command has always also worked in text mode (see \thinspace & \negthinspace).

\quad

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 displaymath environment. It is available in both text and math mode.

\qquad

A length of 2 quads, that is, 36mu = 2em. It is available in both text and math mode.


16.6.1 \smash

Synopsis:

\smash{subformula}

Typeset subformula as if its height and depth were zero.

In this example the exponential is so tall that without the \smash command LaTeX would separate its line from the line above it, and the uneven line spacing might be unsightly.

To compute the tetration $\smash{2^{2^{2^2}}}$,
evaluate from the top down, as $2^{2^4}=2^{16}=65536$.

(Because of the \smash the printed expression could run into the line above so you may want to wait until the final version of the document to make such adjustments.)

This pictures the effect of \smash by using \fbox to surround the box that LaTeX will put on the line. The \blackbar command makes a bar extending from 10 points below the baseline to 20 points above.

\newcommand{\blackbar}{\rule[-10pt]{5pt}{30pt}}
\fbox{\blackbar}
\fbox{\smash{\blackbar}}

The first box that LaTeX places is 20 points high and 10 points deep. But the second box is treated by LaTeX as having zero height and zero depth, despite that the ink printed on the page still extends well above and below the line.

The \smash command appears often in mathematics to adjust the size of an element that surrounds a subformula. Here the first radical extends below the baseline while the second lies just on the baseline.

\begin{equation}
\sqrt{\sum_{0\leq k< n} f(k)}
\sqrt{\vphantom{\sum}\smash{\sum_{0\leq k< n}} f(k)}
\end{equation}

Note the use of \vphantom to give the \sqrt command an argument with the height of the \sum (see \phantom & \vphantom & \hphantom).

While most often used in mathematics, the \smash command can appear in other contexts. However, it doesn’t change into horizontal mode. So if it starts a paragraph then you should first put a \leavevmode, as in the bottom line below.

Text above.

\smash{smashed, no indent}  % no paragraph indent

\leavevmode\smash{smashed, with indent}  % usual paragraph indent

The package mathtools has operators that provide even finer control over smashing a subformula box.


16.6.2 \phantom & \vphantom & \hphantom

Synopsis:

\phantom{subformula}

or

\vphantom{subformula}

or

\hphantom{subformula}

The \phantom command creates a box with the same height, depth, and width as subformula, but empty. That is, this command causes LaTeX to typeset the space but not fill it with the material. Here LaTeX will put a blank line that is the correct width for the answer, but will not show that answer.

\begin{displaymath}
  \int x^2\,dx=\mbox{\underline{$\phantom{(1/3)x^3+C}$}} 
\end{displaymath}

The \vphantom variant produces an invisible box with the same vertical size as subformula, the same height and depth, but having zero width. And \hphantom makes a box with the same width as subformula but with zero height and depth.

In this example, the tower of exponents in the second summand expression is so tall that TeX places this expression further down than its default. Without adjustment, the two summand expressions would be at different levels. The \vphantom in the first expression tells TeX to leave as much vertical room as it does for the tower, so the two expressions come out at the same level.

\begin{displaymath}
    \sum_{j\in\{0,\ldots\, 10\}\vphantom{3^{3^{3^j}}}}
      \sum_{i\in\{0,\ldots\, 3^{3^{3^j}}\}} i\cdot j
\end{displaymath}

These commands are often used in conjunction with \smash. See \smash, which includes another example of \vphantom.

The three phantom commands appear often but note that LaTeX provides a suite of other commands to work with box sizes that may be more convenient, including \makebox (see \mbox & \makebox) as well as \settodepth (see \settodepth), \settoheight (see \settoheight), and \settowidth (see \settowidth). In addition, the mathtools package has many commands that offer fine-grained control over spacing.

All three commands produce an ordinary box, without any special mathematics status. So to do something like attaching a superscript you should give it such a status, for example with the \operatorname command from the package amsmath.

While most often used in mathematics, these three can appear in other contexts. However, they don’t cause LaTeX to change into horizontal mode. So if one of these starts a paragraph then you should prefix it with \leavevmode.


16.6.3 \mathstrut

Synopsis:

\mathstrut

The analogue of \strut for mathematics. See \strut.

The input $\sqrt{x} + \sqrt{x^i}$ gives output where the second radical is taller than the first. To add extra vertical space without any horizontal space, so that the two have the same height, use $\sqrt{x\mathstrut} + \sqrt{x^i\mathstrut}$.

The \mathstrut command adds the vertical height of an open parenthesis, (, but no horizontal space. It is defined as \vphantom{(}, so see \phantom & \vphantom & \hphantom for more. An advantage over \strut is that \mathstrut adds no depth, which is often the right thing for formulas. Using the height of an open parenthesis is just a convention; for complete control over the amount of space, use \rule with a width of zero. See \rule.


16.7 Math styles

TeX’s rules for typesetting a formula depend on the context. For example, inside a displayed equation, the input \sum_{0\leq i<n}k^m=\frac{n^{m+1}}{m+1}+\mbox{lower order terms} will give output with the summation index centered below the summation symbol. But if that input is inline then the summation index is off to the right rather than below, so it won’t push the lines apart. Similarly, in a displayed context, the symbols in the numerator and denominator will be larger than for an inline context, and in display math subscripts and superscripts are further apart then they are in inline math.

TeX uses four math styles.

  • Display style is for a formula displayed on a line by itself, such as with \begin{equation} ... \end{equation}.
  • Text style is for an inline formula, as with ‘so we have $ ... $’.
  • Script style is for parts of a formula in a subscript or superscript.
  • Scriptscript style is for parts of a formula at a second level (or more) of subscript or superscript.

TeX determines a default math style but you can override it with a declaration of \displaystyle, or \textstyle, or \scriptstyle, or \scriptscriptstyle.

In this example, the ‘Arithmetic’ line’s fraction will look scrunched.

\begin{tabular}{r|cc}
  \textsc{Name}  &\textsc{Series}  &\textsc{Sum}  \\  \hline
  Arithmetic     &$a+(a+b)+(a+2b)+\cdots+(a+(n-1)b)$
                   &$na+(n-1)n\cdot\frac{b}{2}$  \\
  Geometric      &$a+ab+ab^2+\cdots+ab^{n-1}$
                   &$\displaystyle a\cdot\frac{1-b^n}{1-b}$  \\
\end{tabular}

But because of the \displaystyle declaration, the ‘Geometric’ line’s fraction will be easy to read, with characters the same size as in the rest of the line.

Another example is that, compared to the same input without the declaration, the result of

we get
$\pi=2\cdot{\displaystyle\int_{x=0}^1 \sqrt{1-x^2}\,dx}$

will have an integral sign that is much taller. Note that here the \displaystyle applies to only part of the formula, and it is delimited by being inside curly braces, as ‘{\displaystyle ...}’.

The last example is a continued fraction.

\begin{equation}
a_0+\frac{1}{
       \displaystyle a_1+\frac{\mathstrut 1}{
       \displaystyle a_2+\frac{\mathstrut 1}{
       \displaystyle a_3}}}
\end{equation}

Without the \displaystyle declarations, the denominators would be set in script style and scriptscript style. (The \mathstrut improves the height of the denominators; see \mathstrut.)


16.8 Math miscellany

LaTeX contains a wide variety of mathematics facilities. Here are some that don’t fit into other categories.


16.8.1 Colon character : & \colon

Synopsis, one of:

:
\colon

In mathematics, the colon character, :, is a relation.

With side ratios \( 3:4 \) and \( 4:5 \), the triangle is right.

Ordinary LaTeX defines \colon to produce the colon character with the spacing appropriate for punctuation, as in set-builder notation \{x\colon 0\leq x<1\}.

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


16.8.2 \*

Synopsis:

\*

A multiplication symbol that allows a line break. If there is a break then LaTeX puts a \times symbol, ×, before that break.

In \( A_1\* A_2\* A_3\* A_4 \), if there is no line break then LaTeX outputs it as though it were \( A_1 A_2 A_3 A_4 \). If a line break does happen, for example between the two middle ones, then LaTeX sets it like \( A_1 A_2 \times \), followed by the break, followed by \( A_3 A_4 \).


16.8.3 \frac

Synopsis:

\frac{numerator}{denominator}

Produces the fraction. Used as: \begin{displaymath} \frac{1}{\sqrt{2\pi\sigma}} \end{displaymath}. In inline math mode it comes out small; see the discussion of \displaystyle (see Math formulas).


16.8.4 \sqrt

Synopsis, one of:

\sqrt{arg}
\sqrt[root-number]{arg}

The square root, or optionally other roots, of arg. The optional argument root-number gives the root, i.e., enter the cube root of x+y as \sqrt[3]{x+y}. The size of the radical grows with that of arg (as the height of the radical grows, the angle on the leftmost part gets steeper, until for a tall enough arg, it is vertical).

LaTeX has a separate \surd symbol for making a square root without arg (see Math symbols).


16.8.5 \stackrel

Synopsis:

\stackrel{text}{relation}

Put text above relation. To put a function name above an arrow enter \stackrel{f}{\longrightarrow}.


17 Modes

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.

  • Paragraph mode (in plain TeX this is called horizontal mode) is what LaTeX is in when processing ordinary text. It breaks the input text into lines and finds the positions of line breaks, so that in vertical mode page breaks can be done. This is the mode LaTeX is in most of the time.

    LR mode (for left-to-right mode; in plain TeX this is called restricted horizontal mode) is in effect when LaTeX starts making a box with an \mbox command. As in paragraph mode, LaTeX’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 might well complain that the finished LR box won’t fit there.)

  • Math mode is when LaTeX is generating an inline mathematical formula.

    Display math mode 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 \int differs in the two situations.)

  • Vertical mode is when LaTeX is building the list of lines and other material making the output page, which comprises insertion of page breaks. This is the mode LaTeX is in when it starts a document.

    Internal vertical mode is in effect when LaTeX starts making a \vbox. It has not such thing as page breaks, and as such is the vertical analogue of LR mode.

For instance, if you begin a LaTeX article with ‘Let \( x \) be ...’ then these are the modes: first LaTeX starts every document in vertical mode, then it reads the ‘L’ and switches to paragraph mode, then the next switch happens at the ‘\(’ where LaTeX changes to math mode, and then when it leaves the formula it pops back to paragraph mode.

Paragraph mode has two subcases. If you use a \parbox command or a minipage then LaTeX is put into paragraph mode. But it will not put a page break here. Inside one of these boxes, called a parbox, LaTeX is in inner paragraph mode. Its more usual situation, where it can put page breaks, is outer paragraph mode (see Page breaking).


17.1 \ensuremath

Synopsis:

\ensuremath{formula}

Ensure that formula is typeset in math mode.

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.

\newcommand{\dx}{\ensuremath{dx}}
In $\int f(x)\, \dx$, the \dx{} is an infinitesimal.

Caution: the \ensuremath command is useful but not a panacea.

\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.

Best is to typeset math things in a math mode.


18 Page styles

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.

The package fancyhdr is commonly used for constructing page styles. See its documentation.


18.1 \maketitle

Synopsis:

\maketitle

Generate a title. In the standard classes the title appears on a separate page, except in the article class where it is at the top of the first page. (See Document class options, for information about the titlepage document class option.)

This example shows \maketitle appearing in its usual place, immediately after \begin{document}.

\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.
  ...

You tell LaTeX the information used to produce the title by making the following declarations. These must come before the \maketitle, either in the preamble or in the document body.

\author{name1 \and name2 \and ...}

Required. Declare the document author or authors. The argument is a list of authors separated by \and commands. To separate lines within a single author’s entry, for instance to give the author’s institution or address, use a double backslash, \\. If you omit the \author declaration then you get ‘LaTeX Warning: No \author given’.

\date{text}

Optional. Declare text to be the document’s date. The text doesn’t need to be in a date format; it can be any text at all. If you omit \date then LaTeX uses the current date (see \today). To have no date, instead use \date{}.

\thanks{text}

Optional. Produce a footnote. You can use it in the author information for acknowledgements as illustrated above, but you can also use it in the title, or anywhere that a footnote mark makes sense. It can be any text at all so you can use it for any purpose, such as to print an email address.

\title{text}

Required. Declare text to be the title of the document. Get line breaks inside text with a double backslash, \\. If you omit the \title declaration then the \maketitle command yields error ‘LaTeX Error: No \title given’.

To make your own title page, see titlepage. You can either create this as a one-off or you can include it as part of a renewed \maketitle command. Many publishers will provide a class to use in place of article that formats the title according to their house requirements.


18.2 \pagenumbering

Synopsis:

\pagenumbering{number-style}

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 stopped by an end of group such as a closing brace or an end of environment.

By default, LaTeX numbers pages starting at 1, using Arabic numerals.

The argument number-style is one of the following (see also \alph \Alph \arabic \roman \Roman \fnsymbol: Printing counters).

arabic

Arabic numerals: 1, 2, …

roman

lowercase Roman numerals: i, ii, …

Roman

uppercase Roman numerals: I, II, …

alph

lowercase letters: a, b, … If you have more than 26 pages then you get ‘LaTeX Error: Counter too large’.

Alph

uppercase letters: A, B, … If you have more than 26 pages then you get ‘LaTeX Error: Counter too large’.

gobble

no page number is output, though the number is still reset. References to that page also are blank.

This setting does not work with the popular package hyperref, so to omit page numbers you may want to instead use \pagestyle{empty} or \thispagestyle{empty}.

If you want to typeset the page number in some other way, or change where the page number appears on the page, see \pagestyle (in short: use the fancyhdr package). The list above of LaTeX’s built-in numbering styles cannot be extended.

Traditionally, if a document has front matter—preface, table of contents, etc.—then it is numbered with lowercase Roman numerals. The main matter of a document uses arabic. LaTeX implements this, by providing explicit commands for the different parts (see \frontmatter, \mainmatter, \backmatter).

As an explicit example, before the ‘Main’ section the pages are numbered ‘a’, etc. Starting on the page containing the \pagenumbering call in that section, the pages are numbered ‘1’, etc.

\begin{document}\pagenumbering{alph}
  ...
\section{Main}\pagenumbering{arabic}
  ...

If you want to change the value of the page number, then you manipulate the page counter (see Counters).


18.3 \pagestyle

Synopsis:

\pagestyle{style}

Declaration that specifies how the page headers and footers are typeset, from the current page onwards.

A discussion with an example is below. First, however: the package fancyhdr 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 (https://ctan.org/pkg/fancyhdr).

Values for style:

plain

The header is empty. The footer contains only a page number, centered.

empty

The header and footer are both empty.

headings

Put running headers and footers on each page. The document style specifies what goes in there; see the discussion below.

myheadings

Custom headers, specified via the \markboth or the \markright commands.

Some discussion of the motivation for LaTeX’s mechanism will help you work with the options headings or myheadings. The document source below produces an article, two-sided, with the pagestyle headings. On this document’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 \leftmark and \rightmark. So it is up to \section and \subsection to store that information there.

\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}
  ...

Suppose that the second section falls on a left page. Although when the page starts it is in the first section, LaTeX will put ‘Section 2’ 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, …, 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.

To accomplish this, in a two-sided article, LaTeX has \section issue a command \markboth, setting \leftmark to ‘Section 2’ and setting \rightmark to an empty content. And, LaTeX has \subsection issue a command \markright, setting \rightmark to ‘Subsection 2.1’, etc.

Here are the descriptions of \markboth and \markright:

\markboth{left-head}{right-head}

Sets both the right hand and left hand heading information for either a page style of headings or myheadings. A left hand page heading left-head is generated by the last \markboth command before the end of the page. A right hand page heading right-head is generated by the first \markboth or \markright that comes on the page if there is one, otherwise by the last one that came before that page.

\markright{right-head}

Sets the right hand page heading, leaving the left unchanged.


18.4 \thispagestyle

Synopsis:

\thispagestyle{style}

Works in the same way as the \pagestyle (see \pagestyle), except that it changes to style for the current page only. This declaration has global scope, so its effect is not delimited by braces or environments.

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 plain style, as is the default (see Page styles).

\documentclass{book}
\pagestyle{headings}
\begin{document}
\chapter{First chapter}
  ...
\chapter{Second chapter}\thispagestyle{empty}
  ...

The plain style has a page number on it, centered in the footer. To make the page entirely empty, the command \thispagestyle{empty} immediately follows the second \chapter.


18.5 \thepage

If you want to change the appearance of page numbers only in the page headers, for example by adding an ornament, typesetting in small caps, etc., then the fancyhdr package, as mentioned in a previous section, is the best approach.

On the other hand, you may want to change how page numbers are denoted everywhere, including the table of contents and cross-references, as well as the page headers. In this case, you should redefine \thepage, which is the command LaTeX uses for the representation of page numbers.

However, \thepage should do any typesetting or other complicated maneuvers, but merely expand to the intended page number representation. The results of a complicated redefinition of \thepage are not predictable, but LaTeX’s report of page numbers in diagnostic messages, at least, will become unusable.

There is some discussion of this issue at https://tex.stackexchange.com/questions/687258.


19 Spaces

LaTeX has many ways to produce white space, or filled space. Some of these are best suited to mathematical text; for these see Spacing in math mode.


19.1 \enspace & \quad & \qquad

Synopsis, one of:

\enspace
\quad
\qquad

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 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 Lengths/em).

This puts a suitable gap between two graphics.

\begin{center}
  \includegraphics{womensmile.png}%
  \qquad\includegraphics{mensmile.png}
\end{center}

See Spacing in math mode, for \quad and \qquad. 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 \hspace.


19.2 \hspace

Synopsis, one of:

\hspace{length}
\hspace*{length}

Insert the amount length of horizontal space. The length can be positive, negative, or zero; adding a negative amount of space is like backspacing. It is a rubber length, that is, it may contain a plus or minus component, or both (see Lengths). Because the space is stretchable and shrinkable, it is sometimes called glue.

This makes a line with ‘Name:’ an inch from the right margin.

\noindent\makebox[\linewidth][r]{Name:\hspace{1in}}

The *-form inserts horizontal space that is non-discardable. More precisely, when TeX breaks a paragraph into lines any white space—glues and kerns—that come at a line break are discarded. The *-form avoids that (technically, it adds a non-discardable invisible item in front of the space).

In this example

\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}.}

the 1 inch blank following ‘conceived in’ falls at the start of a line. If you erase the * then LaTeX discards the blank.

Here, the \hspace separates the three graphics.

\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}

Because the argument to each \hspace has minus 0.25cm, each can shrink a little if the three figures are too wide. But each space won’t shrink more than 0.25cm (see Lengths).


19.3 \hfill

Synopsis:

\hfill

Produce a rubber length which has no natural space but that can stretch horizontally as far as needed (see Lengths).

This creates a one-line paragraph with ‘Name:’ on the left side of the page and ‘Quiz One’ on the right.

\noindent Name:\hfill Quiz One

The \hfill command is equivalent to \hspace{\fill} and so the space can be discarded at line breaks. To avoid that instead use \hspace*{\fill} (see \hspace).

Here the graphs are evenly spaced in the middle of the figure.

\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}

Note the \hspace*’s where the space could otherwise be dropped.


19.4 \hss

Synopsis:

\hss

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 \makebox command to get the effects of \hss (see \mbox & \makebox).

Here, the first line’s \hss 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.

X\hbox to 0pt{Z\hss}Y
X\hbox to 0pt{\hss Z}Y

Without the \hss you get something like ‘Overfull \hbox (6.11111pt too wide) detected at line 20’.


19.5 \spacefactor

Synopsis:

\spacefactor=integer

Influence LaTeX’s stretching and shrinking of glue. Few user-level documents need to use this.

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 interword glue; see \hspace). The \spacefactor parameter (a TeX primitive) allows you to, for instance, have the space after a period stretch more than the space after a word-ending letter.

After LaTeX places each character, or rule or other box, it sets a parameter called the space factor. 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, 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.

If the space factor f is 1000 then the glue gap will be the font’s normal space value (for Computer Modern Roman 10 point this is 3.3333pt). Otherwise, if the space factor f is greater than 2000 then TeX adds the font’s extra space value (for Computer Modern Roman 10 point this is 1.11111pt), and then the font’s normal stretch value is multiplied by f /1000 and the normal shrink value is multiplied by 1000/f (for Computer Modern Roman 10 point these are 1.66666 and 1.11111pt).

For example, consider the period ending ‘A man's best friend is his dog.’. 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 friend or any of the other words, since they are not followed by punctuation.

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 the normal effect of the period is does not happen, so you get an ordinary space. This second case also affects abbreviations that do not end in a capital letter (see \@).

You can only use \spacefactor in paragraph mode or LR mode (see Modes). You can see the current value with \the\spacefactor or \showthe\spacefactor.

Finally, not especially related to \spacefactor itself: if you get errors like ‘You can't use `\spacefactor' in vertical mode’, or ‘You can't use `\spacefactor' in math mode.’, or ‘Improper \spacefactor’ then you have probably tried to redefine an internal command. See \makeatletter & \makeatother.


19.5.1 \@

Synopsis:

capital-letter\@.    

Treat a period (or other punctuation) 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.

This example shows the two cases to remember.

The songs \textit{Red Guitar}, etc.\ are by Loudon Wainwright~III\@.

The first period ends the abbreviation ‘etc.’ but not the sentence. The backslash-space, \ , produces a mid-sentence space. The second period ends the sentence, despite it being preceded by a capital letter. We tell LaTeX that it ends the sentence by putting \@ before it.

So: if you have a capital letter followed by a period that ends the sentence, then put \@ 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

Use the \textit{Instructional Practices Guide},
(a book by the MAA)\@.

will have correct inter-sentence spacing after the period.

The \@ command is only for text modes. If you use it outside of a text mode then you get the error ‘You can't use `\spacefactor' in vertical mode’ (see Modes).

All the above applies equally to question marks and exclamation points as periods, since all are sentence-ending punctuation, and LaTeX increases the space after each in the same way, when they end a sentence. LaTeX also increases spacing after colon, semicolon, and comma characters (see \spacefactor).

In addition: the converse case is a period (or other punctuation) that does not end a sentence. For that case, follow the period with a backslash-space, (\ ), or a tie, (~), or \@. Examples are Nat.\ Acad.\ Science, and Mr.~Bean, and (manure, etc.\@) for sale (note in the last one that the \@ comes after the period but before the closing parenthesis).


19.5.2 \frenchspacing & \nonfrenchspacing

Synopsis, one of:

\frenchspacing
\nonfrenchspacing

\frenchspacing causes LaTeX to make spacing after all punctuation, including periods, be the same as the space between words in the middle of a sentence. \nonfrenchspacing switches back to the default handling in which spacing after most punctuation stretches or shrinks differently than a word space (see \spacefactor).

In American English, the typesetting tradition is to adjust, typically increasing, the space after punctuation more than the space between words that are in the middle of a sentence. Declaring \frenchspacing (the command is inherited from plain TeX) switches to the tradition that all spaces are treated equally.

If your LaTeX document specifies the language being used, for example with the babel package, the necessary settings should be taken care of for you.


19.5.3 \normalsfcodes

Synopsis:

\normalsfcodes

Reset the LaTeX space factors to the default values (see \spacefactor).


19.6 Backslash-space, \

This section refers to the command consisting of two characters, a backslash followed by a space. Synopsis:

\ 

Produce a space. By default it produces white space of length 3.33333pt plus 1.66666pt minus 1.11111pt.

When you type one or more blanks between words, LaTeX produces whitespace that is different than an explicit space. This illustrates:

\begin{tabular}{rl}
One blank:& makes some space \\
Three blanks:&   in a row \\
Three spaces:&\ \ \ in a row \\
\end{tabular}

On the first line LaTeX puts some space after the colon. On the second line LaTeX collapses the three blanks to output one whitespace, so you end with the same space after the colon as in the first line. LaTeX would similarly collapse them to a single whitespace if one, two or all of the three blanks were replaced by a tab, or by a newline. However, the bottom line asks for three spaces so the white area is wider. That is, the backslash-space command creates a fixed amount of horizontal space. (Note that you can define a horizontal space of any width at all with \hspace; see \hspace.)

The backslash-space command has two main uses. It is often used after control sequences to keep them from gobbling the blank that follows, as after \TeX in \TeX\ (or \LaTeX). (But using curly braces has the advantage of still working whether the next character is a blank or any other non-letter, as in \TeX{} (or \LaTeX{}) in which {} can be added after \LaTeX as well as after \TeX.) The other common use is that it marks a period as ending an abbreviation instead of ending a sentence, as in Prof.\ Smith or Jones et al.\ (1993) (see \@).

Under normal circumstances, \TAB and \NEWLINE are equivalent to backslash-space, \ .

In order to allow source code indentation, under normal circumstances, TeX ignores leading blanks in a line. So the following prints ‘one word’:

one
 word

where the white space between ‘one’ and ‘word’ is produced by the newline after ‘one’, not by the space before ‘word’.


19.7 ~, \nobreakspace

Synopsis:

before~after

The tie character, ~, produces a space between before and after at which the line will not be broken. By default the white space has length 3.33333pt plus 1.66666pt minus 1.11111pt (see Lengths). The command \nobreakspace and the Unicode input character U+00A0 (also in many 8-bit encodings) are synonyms.

Note that the word ‘tie’ has this meaning in the TeX/Texinfo community; this differs from the typographic term “tie”, which is a diacritic in the shape of an arc, called a “tie-after” accent in The TeXbook.

Here LaTeX will not break the line between the final two words:

Thanks to Prof.~Lerman.

In addition, despite the period, LaTeX does not use the end-of-sentence spacing (see \@).

Ties prevent a line break where that could cause confusion. They also still allow hyphenation (of either of the tied words), so they are generally preferable to putting consecutive words in an \mbox (see \mbox & \makebox).

Exactly where ties should be used is something of a matter of taste, sometimes alarmingly dogmatic taste, among readers. Nevertheless, here are some usage models, many of them from The TeXbook.

  • Between an enumerator label and number, such as in references: Chapter~12, or Theorem~\ref{th:Wilsons}, or Figure~\ref{fig:KGraph}.
  • When cases are enumerated inline: (b)~Show that $f(x)$ is (1)~continuous, and (2)~bounded.
  • Between a number and its unit: $745.7.8$~watts (the siunitx package has a special facility for this) or 144~eggs. This includes between a month and day number in a date: October~12 or 12~Oct. In general, in any expressions where numbers and abbreviations or symbols are separated by a space: AD~565, or 2:50~pm, or Boeing~747, or 268~Plains Road, or \$$1.4$~billion. Other common choices here are a thin space (see \thinspace & \negthinspace) and no space at all.
  • When mathematical phrases are rendered in words: equals~$n$, or less than~$\epsilon$, or given~$X$, or modulo~$p^e$ for all large~$n$ (but compare is~$15$ with is $15$~times the height). Between mathematical symbols in apposition with nouns: dimension~$d$ or function~$f(x)$ (but compare with length $l$~or more). When a symbol is a tightly bound object of a preposition: of~$x$, or from $0$ to~$1$, or in common with~$m$.
  • Between symbols in series: $1$,~$2$, or~$3$ or $1$,~$2$, \ldots,~$n$.
  • Between a person’s given names and between multiple surnames: Donald~E. Knuth, or Luis~I. Trabb~Pardo, or Charles~XII—but you must give TeX places to break the line so you might do Charles Louis Xavier~Joseph de~la Vall\'ee~Poussin.

19.8 \thinspace & \negthinspace

Synopsis, one of:

\thinspace
\negthinspace

These produce unbreakable and unstretchable spaces of 1/6em and -1/6