LaTeX Beginner's Guide La Te X Stefan Kottwitz

User Manual: Pdf

Open the PDF directly: View PDF .
Page Count: 661

Download
Open PDF In Browser	View PDF

LaTeX

Beginner's Guide

Table of Contents
LaTeX
Credits
About the Author
About the Reviewers
www.PacktPub.com
Support files, eBooks, discount offers, and more
Why Subscribe?
Free Access for Packt account holders
Preface
What this book covers
What you need for this book
Who this book is for
Conventions
Time for action - heading
What just happened?
Pop quiz
Have a go hero - heading
Reader feedback
Customer support
Downloading the example code
Errata
Piracy
Questions
1. Getting Started with LaTeX
What is LaTeX?
How we can benefit
The virtues of open source
Separation of form and content
Portability
Protection for your work
Comparing it to word processor software
What are the challenges?
Installing LaTeX
Time for action – installing TeX Live using the net installer wizard
What just happened?
Time for action – installing TeX Live offline
What just happened?
Installation on other operating systems

Creating our first document
Time for action – writing our first document with TeXworks
What just happened?
Have a go hero – checking out advanced LaTeX editors
Summary
2. Formatting Words, Lines, and Paragraphs
Understanding logical formatting
Time for action – titling your document
What just happened?
Exploring the document structure
Understanding LaTeX commands
How LaTeX reads your input
Time for action – trying out the effect of spaces, line breaks, and empty
lines
What just happened?
Commenting your source text
Printing out special symbols
Time for action – writing special characters in our text
What just happened?
Formatting text – fonts, shapes, and styles
Time for action – tuning the font shape
What just happened?
Choosing the font family
Time for action – switching to sans-serif and to typewriter fonts
What just happened?
Switching fonts
Time for action – switching the font family
What just happened?
Summarizing font commands and declarations
Delimiting the effect of commands
Time for action – exploring grouping by braces
What just happened?
Time for action – exploring font sizes
What just happened?
Using environments
Time for action – using an environment to adjust the font size
What just happened?
Saving time and effort – creating your own commands
Time for action – creating our first command using it as anabbreviation
What just happened?
Gentle spacing after commands
Time for action – adding intelligent spacing to command output

What just happened?
Creating more universal commands – using arguments
Time for action – creating a macro for formatting keywords
What just happened?
Using optional arguments
Time for action – marking keywords with optional formatting
What just happened?
Pop quiz – commands
Have a go hero – saving effort using optional arguments
Using boxes to limit the width of paragraphs
Time for action – creating a narrow text column
What just happened?
Common paragraph boxes
Have a go hero – exploring further features of \parbox
Boxes containing more text
Time for action – using the minipage environment
What just happened?
Have a go hero – creating a footnote inside a minipage
Understanding environments
Breaking lines and paragraphs
Improving hyphenation
Time for action – stating division points for words
What just happened?
Have a go hero – exploiting the hyphenat package
Improving the justification further
Time for action – using microtype
What just happened?
Breaking lines manually
Time for action – using line breaks
What just happened?
Have a go hero – exploring line breaking options
Preventing line breaks
Managing line breaks wisely
Exploring the fine details
Time for action – exploring ligatures
What just happened?
Understanding ligatures
Choosing the right dash
Setting dots
Time for action – using differently spaced dots
What just happened?
Time for action – comparing dots to ellipsis

What just happened?
Setting accents
Time for action – experimenting with accents
What just happened?
Using special characters directly in the editor
Time for action – using accents directly
What just happened?
Turning off full justification
Time for action – justifying a paragraph to the left
What just happened?
Creating ragged-left text
Time for action – centering a title
What just happened?
Using environments for justification
Time for action – centering verses
What just happened?
Displaying quotes
Time for action – quoting a scientist
What just happened?
Quoting longer text
Time for action – quoting TeX's benefits
What just happened?
Time for action – spacing between paragraphs instead of indentation
What just happened?
Pop quiz – lines and paragraphs
Summary
3. Designing Pages
Defining the overall layout
Time for action – writing a book with chapters
What just happened?
Reviewing LaTeX's default page layout
Defining the margins yourself
Time for action – specifying margins
What just happened?
Using the geometry package
Choosing the paper size
Specifying the text area
Setting the margins
Obtaining package documentation
Time for action – finding the geometry package manual
What just happened?
Have a go hero – constructing the page layout with typearea

Changing the line spacing
Time for action – increasing line spacing
What just happened?
Have a go hero – examining a package source file
Using class options to configure the document style
Time for action – creating a two-column landscape document
What just happened?
Have a go hero – trying KOMA-Script classes
Creating a table of contents
Time for action – adding a table of contents
What just happened?
Sectioning and the contents
Time for action – shortening the table of content entries
What just happened?
Designing headers and footers
Time for action – customizing headers with the fancyhdr package
What just happened?
Understanding page styles
Customizing header and footer
Using decorative lines in header or footer
Changing LaTeX's header marks
Breaking pages
Time for action – inserting page breaks
What just happened?
Have a go hero – exploring page breaking options
Enlarging a page
Time for action – sparing an almost empty page
What just happened?
Using footnotes
Time for action – using footnotes in text and in headings
What just happened?
Modifying the dividing line
Time for action – redefining the footnote line
What just happened?
Using packages to expand footnote styles
Pop quiz
Summary
4. Creating Lists
Building a bulleted list
Time for action – listing LaTeX packages
What just happened?
Nesting lists

Time for action – listing packages by topic
What just happened?
Creating a numbered list
Time for action – writing a step-by-step tutorial
What just happened?
Customizing lists
Saving space with compact lists
Time for action – shrinking our tutorial
What just happened?
Choosing bullets and numbering format
Time for action – modifying lists using enumitem
What just happened?
Suspending and continuing lists
Producing a definition list
Time for action – explaining capabilities of packages
What just happened?
Have a go hero – adjusting the dimensions of lists
Pop quiz
Summary
5. Creating Tables and Inserting Pictures
Writing in columns
Time for action – lining up information using the tabbing environment
What just happened?
Time for action – lining up font commands
What just happened?
Typesetting tables
Time for action – building a table of font family commands
What just happened?
Drawing lines in tables
Understanding formatting arguments
Increasing the row height
Beautifying tables
Time for action – adding nicer horizontal lines with the booktabs package
What just happened?
Have a go hero – adjusting lengths
Spanning entries over multiple columns
Time for action – merging cells
What just happened?
Inserting code column-wise
Time for action – using the array package
What just happened?
Spanning entries over multiple rows

Time for action – merging cells using the multirow package
What just happened?
Adding captions to tables
Time for action – adding a caption to our font table
What just happened?
Placing captions above
Have a go hero – customizing captions
Auto-fitting columns to the table width
Generating multi-page tables
Coloring tables
Using landscape orientation
Aligning columns at the decimal point
Handling narrow columns
Pop quiz – tables
Inserting pictures
Time for action – including a picture
What just happened?
Scaling pictures
Choosing the optimal file type
Including whole pages
Putting images behind the text
Managing floating environments
Time for action – letting a figure float
What just happened?
Understanding float placement options
Forcing the output of floats
Limiting floating
Avoiding floating at all
Spanning figures and tables over text columns
Letting text flow around figures
Time for action – embedding a picture within text
What just happened?
Breaking figures and tables into pieces
Pop quiz – pictures and floats
Summary
6. Cross-Referencing
Setting labels and referencing
Time for action – referencing items of a top list
What just happened?
Assigning a key
Referring to a key
Referring to a page

Producing intelligent page references
Time for action – introducing variable references
What just happened?
Fine-tuning page references
Referring to page ranges
Using automatic reference names
Time for action – referring cleverly
What just happened?
Combing cleveref and varioref
Referring to labels in other documents
Have a go hero – turning references into hyperlinks
Pop quiz
Summary
7. Listing Content and References
Customizing the table of contents
Time for action – refining an extensive table of contents
What just happened?
Adjusting the depth of the TOC
Shortening entries
Adding entries manually
Creating and customizing lists of figures
Time for action – creating a list of diagrams
What just happened?
Creating a list of tables
Using packages for customization
Generating an index
Time for action – marking words and building the index
What just happened?
Defining index entries and subentries
Specifying page ranges
Using symbols and macros in the index
Referring to other index entries
Fine-tuning page numbers
Designing the index layout
Creating a bibliography
Time for action – citing texts and listing the references
What just happened?
Using the standard bibliography environment
Using bibliography databases with BibTeX
Time for action – creating and using a BibTeX database
What just happened?
Looking at the BibTeX entry fields

Understanding BibTeX entry types
Choosing the bibliography style
Listing references without citing
Changing the headings
Have a go hero – using natbib
Pop quiz
Summary
8. Typing Math Formulas
Writing basic formulas
Time for action – discussing quadratic equations and roots
What just happened?
Embedding math expressions within text
Displaying formulas
Numbering equations
Adding subscripts and superscripts
Extracting roots
Writing fractions
Greek letters
Script letters
Producing an ellipsis
Comparing in-line formulas to displayed formulas
Changing the font, style, and size
Customizing displayed formulas
Time for action – typesetting multi-line formulas
What just happened?
Aligning multi-line equations
Numbering rows in multi-line formulas
Inserting text into formulas
Fine-tuning formulas
Using operators
Exploring the wealth of math symbols
Binary operation symbols
Binary relation symbols
Inequality relation symbols
Subset and superset symbols
Variable sized operators
Arrows
Harpoons
Symbols derived from letters
Variable sized delimiters
Miscellaneous symbols
Writing units

Building math structures
Creating arrays
Writing binomial coefficients:
Typesetting matrices
Stacking expressions
Underlining and overlining
Setting accents
Putting a symbol above another
Writing theorems and definitions
Have a go hero – checking out the mathtools package
Pop quiz
Summary
9. Using Fonts
Preparing the encoding
Time for action – directly using special characters
What just happened?
Installing additional fonts
Choosing the main font
Time for action – comparing Computer Modern to Latin Modern
What just happened?
Loading font packages
Latin Modern – a replacement for the standard font
Kp-fonts – a full set of fonts
Serif fonts
Times Roman
Charter
Palatino
Bookman
New Century Schoolbook
Concrete Roman
Sans-serif fonts
Helvetica
Bera Sans
Computer Modern Bright
Kurier
Typewriter fonts
Courier
Inconsolata
Bera Mono
Exploring the world of LaTeX fonts
Pop quiz
Summary

10. Developing Large Documents
Splitting the input
Time for action – swapping out preamble and chapter contents
What just happened?
Including small pieces of code
Including bigger parts of a document
Compiling parts of a document
Creating front and back matter
Time for action – adding a dedication and an appendix
What just happened?
Designing a title page
Time for action – creating a title page
What just happened?
Working with templates
Time for action – starting with a template
What just happened?
Have a go hero – evaluating and enhancing templates
Pop quiz
Summary
11. Enhancing Your Documents Further
Using hyperlinks and bookmarks
Time for action – adding hyperlinks
What just happened?
Time for action – customizing the hyperlink appearance
What just happened?
Time for action – editing PDF metadata
What just happened?
Creating hyperlinks manually
Creating bookmarks manually
Math formulas and special symbols in bookmarks
Benefitting from other packages
Time for action – visiting the TeX Catalogue Online
What just happened?
Time for action – installing a LaTeX package
What just happened?
Designing headings
Time for action – designing chapter and section headings
What just happened?
Coloring your document
Have a go hero – creating colorful tables
Pop quiz
Summary

12. Troubleshooting
Understanding and fixing errors
Time for action – interpreting and fixing an error
What just happened?
Using commands and environments
Writing math formulas
Handling the preamble and document body
Working with files
Creating tables and arrays
Working with lists
Working with floating figures and tables
General syntax errors
Handling warnings
Time for action – emphasizing on a sans-serif font
What just happened?
Justifying text
Referencing
Choosing fonts
Placing figures and tables
Customizing the document class
Avoiding obsolete classes and packages
General troubleshooting
Have a go hero – examining LaTeX's log files
Pop quiz – troubleshooting
Summary
13. Using Online Resources
Web forums, discussion boards, and Q&A sites
Usenet groups
comp.text.tex
Newsgroups in other languages
Web forums
LaTeX-Community.org
TeX and LaTeX on Stack Exchange
Time for action – asking a question online
What just happened?
Frequently Asked Questions
UK TeX FAQ
Visual LaTeX FAQ
MacTeX FAQ
AMS-Math FAQ
LaTeX Picture FAQ
Mailing lists

texhax
tex-live
texworks
List collections
TeX user group sites
TUG – the TeX users group
The LaTeX project
UK TUG – TeX in the United Kingdom
Local user groups
Homepages of LaTeX software and editors
LaTeX distributions
LaTeX editors
Cross-platform
Windows
Linux
Mac OS X
LaTeX archives and catalogs
CTAN – the Comprehensive TeX Archive Network
The TeX Catalogue Online
The LaTeX Font Catalogue
TeX Resources on the Web
Friends of LaTeX
XeTeX
LuaTeX
ConTeXt
LyX
LaTeX blogs
The TeXblog
Some TeX Developments
LaTeX Alive
LaTeX for Humans
The TeX community aggregator
Summary
A. Pop Quiz Answers
Chapter 2: Formatting Words, Lines, and Paragraphs
Commands
Lines and paragraphs
Chapter 3: Designing Pages
Chapter 4: Creating Lists
Chapter 5: Creating Tables and Inserting Pictures
Tables
Pictures and floats

Chapter 6: Cross-Referencing
Chapter 7: Listing Content and References
Chapter 8: Typing Math Formulas
Chapter 9: Using Fonts
Chapter 10: Developing Large Documents
Chapter 11: Enhancing Your Documents Further
Chapter 12: Troubleshooting
Index

LaTeX

Beginner's Guide

LaTeX
Beginner's Guide
Copyright © 2011 Packt Publishing
All rights reserved. No part of this book may be reproduced, stored in a retrieval
system, or transmitted in any form or by any means, without the prior written
permission of the publisher, except in the case of brief quotations embedded in
critical articles or reviews.
Every effort has been made in the preparation of this book to ensure the accuracy
of the information presented. However, the information contained in this book is
sold without warranty, either express or implied. Neither the author, nor Packt
Publishing, and its dealers and distributors will be held liable for any damages
caused or alleged to be caused directly or indirectly by this book.
Packt Publishing has endeavored to provide trademark information about all of
the companies and products mentioned in this book by the appropriate use of
capitals. However, Packt Publishing cannot guarantee the accuracy of this
information.
First published: March 2011
Production Reference: 1150311
Published by Packt Publishing Ltd.
32 Lincoln Road
Olton
Birmingham, B27 6PA, UK.
ISBN 978-1-847199-86-7
www.packtpub.com
Cover Image by Asher Wishkerman ()

Credits
Author
Stefan Kottwitz
Reviewers
Kevin C. Klement
Joseph Wright
Acquisition Editor
Eleanor Duffy
Development Editor
Hyacintha D'Souza
Technical Editor
Sakina Kaydawala
Copy Editor
Leonard D'Silva
Indexer
Hemangini Bari
Editorial Team Leader
Mithun Sehgal
Project Team Leader
Lata Basantani
Project Coordinator
Vishal Bodwani
Proofreader
Aaron Nash
Graphics
Nilesh Mohite
Production Coordinator

Adline Swetha Jesuthas
Cover Work
Adline Swetha Jesuthas

About the Author
Stefan Kottwitz studied mathematics in Jena and Hamburg. Afterwards, he
worked as an IT Administrator and Communication Officer onboard cruise ships
for AIDA Cruises and for Hapag-Lloyd Cruises. Following 10 years of sailing
around the world, he is now employed as a Network & IT Security Engineer for
AIDA Cruises, focusing on network infrastructure and security such as managing
firewall systems for headquarters and fleet.
In between contracts, he worked as a freelance programmer and typography
designer. For many years he has been providing LaTeX support in online forums.
He became a moderator of the web forum http://latex-community.org/ and of the
site http://golatex.de/. Recently, he began supporting the newly established Q&A
site http://tex.stackexchange.com/ as a moderator.
He publishes ideas and news from the TeX world on his blog at http://texblog.net.
I would like to thank Joseph Wright and Kevin C. Klement for reviewing this
book. Special thanks go to Markus Kohm for his great valuable input. I would
also like to thank the people of Packt Publishing, who worked with me on this
book, in particular my development editor Hyacintha D'Souza.

About the Reviewers
Kevin C. Klement is an Associate Professor of Philosophy at the University of
Massachusetts, Amherst. Besides using LaTeX in his academic work in the
history of logic and analytic philosophy, he is a maintainer of the PhilTeX blog,
and an active participant in many online LaTeX communities, including PhilTeX,
LaTeX Community, and TeX.SE.
Joseph Wright is a research assistant at the University of East Anglia. As well as
using LaTeX for his academic work as a chemist, he is a member of the LaTeX3
Project, runs the blog Some TeX Developments and is one of the moderators on
the TeX.SE site.

www.PacktPub.com

Support files, eBooks, discount offers,
and more
You might want to visit www.PacktPub.com for support files and downloads
related to your book.
Did you know that Packt offers eBook versions of every book published, with
PDF and ePub files available? You can upgrade to the eBook version at
www.PacktPub.com and as a print book customer, you are entitled to a discount
on the eBook copy. Get in touch with us at  for more
details.
At www.PacktPub.com, you can also read a collection of free technical articles,
sign up for a range of free newsletters, and receive exclusive discounts and offers
on Packt books and eBooks.

http://PacktLib.PacktPub.com
Do you need instant solutions to your IT questions? PacktLib is Packt's online
digital book library. Here, you can access, read, and search across Packt's entire
library of books.

Why Subscribe?
Fully searchable across every book published by Packt
Copy and paste, print, and bookmark content
On demand and accessible via web browser

Free Access for Packt account holders
If you have an account with Packt at www.PacktPub.com, you can use this to
access PacktLib today and view nine entirely free books. Simply use your login
credentials for immediate access.

Preface
LaTeX is a high-quality open source typesetting software that produces
professional prints and PDF files. However, as LaTeX is a powerful and
complex tool, getting started can be intimidating. There is no official support and
certain aspects such as layout modifications can seem rather complicated. It may
seem more straightforward to use Word or other WYSIWG programs, but once
you've become acquainted, LaTeX's capabilities far outweigh any initial
difficulties. This book guides you through these challenges and makes beginning
with LaTeX easy. If you are writing mathematical, scientific, or business papers,
then this is the perfect book for you.
LaTeX Beginner's Guide offers you a practical introduction to LaTeX. Beginning
with the installation and basic usage, you will learn to typeset documents
containing tables, figures, formulas, and common book elements like
bibliographies, glossaries, and indexes. Lots of step-by-step examples start with
fine-tuning text, formulas and page layout and go on to managing complex
documents and using modern PDF features. It's easy to use LaTeX, when you have
LaTeX Beginner's Guide at hand.
This practical book will guide you through the essential steps of Latex, from
installing LaTeX, formatting, and justification, to page design. Finally, you will
learn how to manage complex documents and how to benefit from modern PDF
features. Right from the beginning, you will learn to use macros and styles to
maintain a consistent document structure while saving typing work. This book
will help you learn to create professional looking tables as well as include
figures and write complex mathematic formulas. You will see how to generate
bibliographies and indexes with ease. Detailed information about online
resources like software archives, web forums, and online compilers complement
this introductory guide.

What this book covers
Chapter 1, Getting Started with LaTeX, introduces LaTeX and explains its
benefits. It guides you through the download and installation of a comprehensive
LaTeX distribution and shows you how to create your first LaTeX document.
Chapter 2, Formatting Words, Lines, and Paragraphs, explains how to vary font,
shape, and style of text. It deals with centering and justification of paragraphs and
how you can improve line breaks and hyphenation. It introduces the concept of
logical formatting and teaches you how to define macros and how to use
environments and packages.
Chapter 3, Designing Pages, shows how you can adjust the margins and change
the line spacing. It demonstrates portrait, landscape, and two-column layout. In
this chapter, we will create dynamic headers and footers and learn how to control
page breaking and how to use footnotes. Along the way, you will also learn about
redefining existing commands and using class options. Furthermore, you will get
familiar with accessing package documentation.
Chapter 4, Creating Lists, deals with arranging text in bulleted, numbered, and
definition lists. We will learn how to choose bullets and numbering style and how
to design the overall layout of lists.
Chapter 5, Creating Tables and Inserting Pictures , shows you how to create
professional-looking tables and how to include external pictures in your
documents. It deals with typesetting captions to tables and figures. We will learn
how to benefit from LaTeX's automated tables and figures placement and how to
fine-tune it.
Chapter 6, Cross-Referencing, introduces means of intelligent referencing to
sections, footnotes, tables, figures, and numbered environments in general.
Chapter 7, Listing Content and References, deals with creating and customizing
of a table of contents and lists of figures and tables. Furthermore, it teaches how
to cite books, how to create bibliographies, and how to generate an index.
Chapter 8, Typing Math Formulas, explains mathematical typesetting in depth. It
starts with basic formulas and continues with centered and numbered equations. It
shows how to align multi-line equations. In detail, it shows how to typeset math
symbols such as roots, arrows, Greek letters, and operators. Moreover, you will
learn how to build complex math structures such as fractions, stacked
expressions, and matrices.
Chapter 9, Using Fonts, takes us into the world of fonts and demonstrates various
fonts for Roman, sans-serif, and typewriter fonts in different shapes. By the way,

you will learn about character encoding and font encoding.
Chapter 10, Developing Large Documents, helps in managing large documents by
splitting them into several files. It shows how to swap out settings, how to reuse
code, and how to compile just parts of a bigger documents. After reading this
chapter, you will be able to create complex projects building upon sub-files.
Furthermore, we deal with front matter and back matter with different page
numbering and separate title pages. We will work it out by creating an example
book. By doing this, you will get familiar with using document templates, finally
being able to write our own thesis, book, or report.
Chapter 11, Enhancing Your Documents Further , brings color into your
documents. It shows you how to modify headings of chapters and all kinds of
sections. We will learn how to create feature-rich PDF documents with
bookmarks, hyperlinks, and meta-data. While doing this, we visit the TeX
Catalogue Online to look out for further useful LaTeX packages and we will go
through a package installation.
Chapter 12, Troubleshooting, provides us with tools for problem-solving. We
will learn about different kinds of LaTeX's errors and warnings and how to deal
with them. After reading this chapter, you will understand LaTeX's messages and
you will know how to use them for fixing errors.
Chapter 13, Using Online Resources, guides you through the vast amount of
LaTeX information on the Internet. We will visit a LaTeX online forum and a
LaTeX Question & Answer site. This chapter points the way to huge LaTeX
software archives, to homepages of TeX user groups, to mailing lists, Usenet
groups, and LaTeX blogs. It tells you where you can download LaTeX capable
editors and where you can find enhanced versions of TeX, such as XeTeX,
LuaTeX, and ConTeXt. Finally, you will know how to access the knowledge of
the world-wide LaTeX community and how to become a part of it.

What you need for this book
You need access to a computer with LaTeX on it. An online connection would be
helpful regarding installation and updates. LaTeX can be installed on most
operating systems, so you can use Windows, Linux, Mac OS X, or Unix.
This book uses the freely available TeX Live distribution, which runs on all
mentioned platforms. You just need an online connection or the TeX Live DVD to
install it. In the book, we work with the cross-platform editor TeXworks, but you
could use any editor you like.

Who this book is for
If you are about to write mathematical or scientific papers, seminar handouts, or
even plan to write a thesis, then this book offers you a fast-paced and practical
introduction. Particularly when studying in school and university you will benefit
a lot, as a mathematician and a physicist as well as an engineer or a humanist.
Everybody with high expectations who plans to write a paper or a book may be
delighted by this stable software.

Conventions
In this book, you will find several headings appearing frequently. To give clear
instructions of how to complete a procedure or task, we use:

Time for action - heading
1. Action 1
2. Action 2
3. Action 3
Instructions often need some extra explanation so that they make sense, so they are
followed with:

What just happened?
This heading explains the working of tasks or instructions that you have just
completed.
You will also find some other learning aids in the book, including:

Pop quiz
These are short multiple choice questions intended to help you test your own
understanding.

Have a go hero - heading
These set practical challenges and give you ideas for experimenting with what
you have learned.
You will also find a number of styles of text that distinguish between different
kinds of information. Here are some examples of these styles, and an explanation
of their meaning.
Code words in text are shown as follows: "The command \chapter produced a
large heading. This command will always begin on a new page."
A block of code is set as follows:
\documentclass[a4paper,12pt]{book}
\usepackage[english]{babel}
\usepackage{blindtext}
\begin{document}
\chapter{Exploring the page layout}
In this chapter we will study the layout of pages.
\section{Some filler text}
\blindtext
\section{A lot more filler text}
More dummy text will follow.
\subsection{Plenty of filler text}
\blindtext[10]
\end{document}

When we wish to draw your attention to a particular part of a code block, the
relevant lines or items are set in bold:
\documentclass[a4paper,11pt]{book}
\usepackage[english]{babel}
\usepackage{blindtext}
\usepackage[a4paper, inner=1.5cm, outer=3cm, top=2cm,
bottom=3cm, bindingoffset=1cm]{geometry}
\begin{document}
\chapter{Exploring the page layout}
In this chapter we will study the layout of pages.
\section{Some filler text}
\blindtext
\section{A lot more filler text}
More dummy text will follow.
\subsection{Plenty of filler text}
\blindtext[3]
\end{document}

Any command-line input or output is written as follows:
texdoc geometry

New terms and important words are shown in bold. Words that you see on the

screen, in menus or dialog boxes for example, appear in the text like this: "Save
the document and Typeset it."

Note
Warnings or important notes appear in a box like this.

Tip
Tips and tricks appear like this.

Reader feedback
Feedback from our readers is always welcome. Let us know what you think about
this book—what you liked or may have disliked. Reader feedback is important
for us to develop titles that you really get the most out of.
To

send

us

general

feedback, simply send an e-mail to
, and mention the book title via the subject of your
message.
If there is a book that you need and would like to see us publish, please send us a
note in the SUGGEST A TITLE form on www.packtpub.com or e-mail
.
If there is a topic that you have expertise in and you are interested in either
writing or contributing to a book on, see our author guide on
www.packtpub.com/authors.

Customer support
Now that you are the proud owner of a Packt book, we have a number of things to
help you to get the most from your purchase.

Downloading the example code
You can download the example code files for all Packt books you have purchased
from your account at http://www.PacktPub.com. If you purchased this book
elsewhere, you can visit http://www.PacktPub.com/support and register to have
the files e-mailed directly to you.

Errata
Although we have taken every care to ensure the accuracy of our content, mistakes
do happen. If you find a mistake in one of our books—maybe a mistake in the text
or the code—we would be grateful if you would report this to us. By doing so,
you can save other readers from frustration and help us improve subsequent
versions of this book. If you find any errata, please report them by visiting
http://www.packtpub.com/support, selecting your book, clicking on the errata
submission form link, and entering the details of your errata. Once your errata are
verified, your submission will be accepted and the errata will be uploaded on our
website, or added to any list of existing errata, under the Errata section of that
title. Any existing errata can be viewed by selecting your title from
http://www.packtpub.com/support.

Piracy
Piracy of copyright material on the Internet is an ongoing problem across all
media. At Packt, we take the protection of our copyright and licenses very
seriously. If you come across any illegal copies of our works, in any form, on the
Internet, please provide us with the location address or website name
immediately so that we can pursue a remedy.
Please contact us at  with a link to the suspected
pirated material.
We appreciate your help in protecting our authors, and our ability to bring you
valuable content.

Questions
You can contact us at  if you are having a problem
with any aspect of the book, and we will do our best to address it.

Chapter 1. Getting Started with
LaTeX
Are you ready to leave those "what you see is what you get" word processors
behind and to enter the world of real, reliable, and high-quality typesetting?
Then let's go together!
It's great that you decided to learn LaTeX. This book will guide you along the
way to help you get the most out of it. Let's speak briefly about LaTeX's benefits
and the challenges, and then we shall prepare our tools.
In this chapter, we will:
Get to know LaTeX and talk about the pros and cons compared to word
processors
Install a complete LaTeX software bundle, including an editor
Write our first LaTeX document
So, let's get started.

What is LaTeX?
LaTeX is a software for typesetting documents. In other words, it's a document
preparation system. LaTeX is not a word processor, but is used as a document
markup language.
LaTeX is a free, open source software. It was originally written by Leslie
Lamport and is based on the TeX typesetting engine by Donald Knuth. People
often refer to it as just TeX, meaning LaTeX. It has a long history; you can read
about it at http://www.tug.org/whatis.html. For now, let's continue by looking at
how we can make the best use of it.

How we can benefit
LaTeX is especially well-suited for scientific and technical documents. Its
superior typesetting of mathematical formulas is legendary. If you are a student or
a scientist, then LaTeX is by far the best choice, and even if you don't need its
scientific capabilities, there are other uses — it produces very high quality output,
it is extremely stable, and handles complex documents easily no matter how large
they are.
Further remarkable strengths of LaTeX are its cross-referencing capabilities, its
automatic numbering and generation of lists of contents, figures and tables,
indexes, glossaries, and bibliographies. It is multilingual with language-specific
features, and it is able to use PostScript and PDF features.
Apart from being perfect for scientists, LaTeX is incredibly flexible—there are
templates for letters, presentations, bills, philosophy books, law texts, music
scores, and even for chess game notations. Hundreds of LaTeX users have written
thousands of templates, styles, and tools useful for every possible purpose. It is
collected and categorized online on archiving servers.
You could benefit from its impressive high quality by starting with its default
styles relying on its intelligent formatting, but you are free to customize and to
modify everything. People of the TeX community have already written a lot of
extensions addressing nearly every formatting need.

The virtues of open source
The sources of LaTeX are completely free and readable for everyone. This
enables you to study and to change everything, from the core of LaTeX to the
latest extension packages. But what does this mean for you as a beginner? There's
a huge LaTeX community with a lot of friendly, helpful people. Even if you
cannot benefit from the open source code directly, they can read the sources and
assist you. Just join a LaTeX web forum and ask your questions there. Helpers
will, if necessary, dig into LaTeX sources and in all probability find a solution
for you, sometimes by recommending a suitable package, often providing a
redefinition of a default command.
Today, we're already profiting from about 30 years of development by the TeX
community. The open source philosophy made it possible, as every user is invited
to study and improve the software and develop it further. Chapter 13, Using
Online Resources, will point the way to the community.

Separation of form and content
A basic principle of LaTeX is that the author should not be distracted too much by
the formatting issues. Usually, the author focuses on the content and formats
logically, for example, instead of writing a chapter title in big bold letters, you
just tell LaTeX that it's a chapter heading—you could let LaTeX design the
heading or you decide in the document's settings what the headings will look like
—just once for the whole document.
LaTeX uses style files extensively called classes and packages , making it easy
to design and to modify the appearance of the whole document and all of its
details.

Portability
LaTeX is available for nearly every operating system, like Windows, Linux, Mac
OS X, and many more. Its file format is plain text—readable and editable, on all
operating systems. LaTeX will produce the same output on all systems. Though
there are different LaTeX software packages, so called TeX distributions , we
will focus on TeX Live , because this distribution is available for Windows,
Linux, and Mac OS X.
LaTeX itself doesn't have a graphical user interface; that's one of the reasons why
it's so portable. You can choose any text editor. There are many editors, even
specialized in LaTeX, for every operating system. Some editors are available for
several systems. For instance, TeXworks runs on Windows, Linux, and Mac OS
X; that's one of the reasons why we will use it in our book. Another very
important reason is that it's probably best-suited for beginners.
LaTeX generates PDF output—printable and readable, on most computers and
looks identical regardless of the operating system. Besides PDF, it supports DVI,
PostScript, and HTML output, preparing the ground for distribution both in print
and online, on screen, electronic book readers, or smart phones.
To sum up, LaTeX is portable in three ways—source, its implementation, and
output.

Protection for your work
LaTeX documents are stored in human readable text format, not in some obscure
word processing format, that may be altered in a different version of the same
software. Try to open a 20 year old document written with a commercial word
processor. What might your modern software show? Even if you can read the file,
its visual appearance would certainly be different than before. LaTeX promises
that the document will always be readable and will result in the same output.
Though it's being further developed, it will remain backwards compatible.
Word processor documents could be infected with viruses, malicious macros
could destroy the data. Did you ever hear of a virus "hiding" in a text file? LaTeX
is not threatened by viruses.

Comparing it to word processor software
We've already described some advantages of the typesetting system LaTeX
compared to word processing software. While LaTeX encourages structured
writing, other word processors may compel you to work inconsistently. They
might hide the real formatting structure and encrypt your document in some
proprietary file format. Compatibility is a big problem, even between versions of
the same software.
There are some interesting articles available online comparing LaTeX to other
software. Of course, they are expressions of opinion. Some are years old and
therefore do not cover the most recent software, but they discuss important points
that are still valid today. You will find them listed in Chapter 13, Using Online
Resources.

What are the challenges?
The learning curve could be steep, but this book will to help you master it.
Though writing LaTeX looks like programming, don't be afraid. Soon you will
know the frequently used commands. Text editors with auto completion and
keyword highlighting will support you. They might even provide menus and
dialogs with commands for you.
Do you now think it will take a long time until you would learn to achieve
creditable results? Don't worry; this book will give you a quick start. You will
learn by practicing with a lot of examples. Many more examples can be read and
downloaded from the Internet. In Chapter 13, we will explore the Internet
resources.
We shall continue with the setup of LaTeX on our computer.

Installing LaTeX
Let's start off with the installation of the LaTeX distribution–TeX Live . This
distribution is available for Windows, Linux, Mac OS X, and other Unix-like
operating systems. TeX Live is well maintained and it is being actively
developed.

Note
Another very good and user-friendly LaTeX distribution for Windows is
MiKTeX. It's easy to install like any other Windows application, but it's not
available for other systems like Linux or Mac OS X. You can download it from
http://miktex.org.
At first, we will visit the TeX Live homepage and take a survey of the installation
possibilities. Feel free to explore the homepage in depth to study the information
offered there.
Open the TeX Live homepage at http://tug.org/texlive.

We will cover two ways of installation. The first will be online and requires an

Internet connection. The other method starts with a huge download, but may be
finished offline.
Let's check out the two installation methods.

Time for action – installing TeX Live
using the net installer wizard
We will download the TeX Live net installer and install the complete TeX Live
distribution on our computer.
1. Click

on downloading

over

the

net
http://tug.org/texlive/acquire-netinstall.

or

navigate

to

2. Download the net installer for Windows by clicking on install-tl.zip.
3. Extract the file install-tl.zip using your favorite archiving program. For
example, WinZip, WinRar, or 7-Zip can do it for you.
4. Open the folder install-tl-* and double-click the Windows batch file
install-tl:

5. The net installer will automatically detect your language. If it's showing the
wrong language, you can force the choice of the language using the lang
option at the command prompt such as install-tl –lang=en:

6. The installation wizard will pop up, as shown in the following screenshot:

7. Click on the Next button, now it offers to change the installation folder, but
it's fine to retain it. In our book, we will refer to this default location:

8. Click on the Next button. As shown in the following screenshot, choose one
of the options, for example, for the creation of shortcuts:

9. Click on the Next button. You can then confirm the settings and actually start
the installation by clicking on the Install button:

10. The next screenshot shows how you can monitor the installation progress:

11. Finally, click on the Finish button and you're done.

What just happened?
You have completed the installation of TeX Live 2010. Now your Start menu
contains a folder called TeX Live 2010 containing six programs:

DVIOUT DVI viewer—a viewer program for the classic LaTeX output
format DVI. Today, most people choose PDF output, so you probably won't
need it.
PS_VIEW—a viewer program for the PostScript format; again you probably
won't need it, except if you would like to use the PostScript language or read
such documents.
TeX Live documentation—well, that's useful regarding setup and use of
your software!
TeX Live Manager—that's your tool for package management, for example,
installation and update of LaTeX packages.
TeXdoc GUI—it's a graphical user interface offering access to a huge
amount of LaTeX-related information. There's a lot of it stored in your
computer by now. Use it to gather information whenever needed; it could be
quicker than searching online.
TeXworks editor—this is an editor developed to create LaTeX documents
comfortably. We will make extensive use of it.

Note
TeXworks is also shipped with MiKTeX 2.8 and higher.
If you would like to stay in control over what should be installed on your
computer, start the install-tl-advanced batch file instead of install-tl:

The TeX documentation available online contains more information for advanced
users.
Now, we will go through the offline installation of TeX Live 2010.

Time for action – installing TeX Live
offline
We will download a compressed ISO image of TeX Live 2010 with a size of
about 1.2 gigabytes. After extraction, we can choose to burn it on DVD or to
extract it to our hard disk drive and run the installation from there:
1. Visit the download area at http://www.tug.org/texlive/acquire-iso.html.
2. Download texlive2010.xz. If possible, use a download manager,
especially if your Internet connection is not stable.
3. Extract texlive2010.xz and you will get the file texlive2010.iso. If
your archiving program doesn't support the .xz file format, obtain, for
instance, the program 7-Zip version 9 or later from http://7zip.org and use it
for extraction.
4. Either burn the ISO file on a DVD using a burning software supporting the
ISO format or extract it to your hard disk drive. 7-Zip is also capable of
doing that for you.
5. Among the extracted files or on your DVD, you will find the installer batch
files install-tl and install-tl-advanced that we've already seen.
Choose one, start it, and go through the installation like in the previous
installation.

What just happened?
It was similar to the first installation, but this time you've got all the data and you
won't need an Internet connection. This complete download is especially
recommended if it's foreseeable that you will do another installation of TeX Live
later or if you would like to give it to friends or colleagues.

Note
After an offline installation, it's recommended to run an update of TeX Live soon,
because packages on a DVD or within an image could already be outdated. Use
the TeX Live Manager to keep your system up-to-date if you are connected to the
Internet.

Installation on other operating systems
If you work on Mac OS X, you may download a customized version of TeX Live
at http://www.tug.org/mactex/. Download the huge .zip file and double-click on
it to install.
On most Linux systems, installation is easy. Use your system's package manager.
With Ubuntu, you may use Synaptic, on SUSE systems use YaST, with Red Hat a
RPM frontend, and on Debian systems use Aptitude. In the respective package
manager, look out for texlive.
If you want to stay on the edge, you could download and install the most current
version of TeX Live from its homepage, instead of the version from the operating
system's repositories. But be aware that installing third party sources may harm
the integrity of your system.
Now that we've prepared the ground, let's start to write LaTeX!

Creating our first document
We have installed TeX and launched the editor; now let's jump in at the deep end
by writing our first LaTeX document.

Time for action – writing our first
document with TeXworks
Our first goal is to create a document that's printing out just one sentence. We
want to use it to understand the basic structure of a LaTeX document.
1. Launch the TeXworks editor by clicking on the desktop icon or open it in the
Start menu.
2. Click on the New button.
3. Enter the following lines:
\documentclass{article}
\begin{document}
This is our first document.
\end{document}

4. Click on the Save button and save the document. Choose a location where
you want to store your LaTeX documents, ideally in its own folder.
5. In the drop-down field in the TeXworks toolbar, choose pdfLaTeX:

6. Click the Typeset button
.
7. The output window will automatically open. Have a look at it:

What just happened?
You have just seen the first few minutes of the life of a LaTeX document. Its
following hours and days will be determined by editing, typesetting, and so on.
Don't forget to save your document frequently.
As announced in contrary to the classic word processor software, you cannot see
the effect of changes immediately—but the result is just one click away.

Have a go hero – checking out advanced LaTeX
editors
Do you have experience in working with complex programs? Do you like using a
feature-rich and powerful editor? Then have a look at these LaTeX editors. Visit
their websites to find screenshots and to read about their features:
TeXnicCenter—
a
http://texniccenter.org/

very

powerful

editor

for

Windows,

Kile— a user-friendly editor for operating systems with KDE, such as Linux,
http://kile.sourceforge.net/
TeXShop—an easy-to-use and very popular editor for Mac OS X,
http://pages.uoregon.edu/koch/texshop/
Texmaker—a cross-platform editor running on Linux, Mac OS X, Unix, and
Windows systems, http://www.xm1math.net/texmaker/
The mentioned editors are free open source software.

Summary
We learned in this chapter about the benefits of LaTeX. It will be our turn to use
the virtues of LaTeX to achieve the best possible results.
Furthermore, we covered:
Installation of TeX Live
Using the editor TeXworks
Creation of a LaTeX document and generation of output
Now that we've got a functional and tested LaTeX system, we're ready to write
our own LaTeX documents. In the next chapter, we will work out the formatting of
text in detail.

Chapter 2. Formatting Words, Lines,
and Paragraphs
In the last chapter, we installed LaTeX and used the TeXworks editor to write
our first document. Now we will speak about the structure of a document and
we will focus on the text details and its formatting.
In this chapter, we shall:
Speak about logical formatting
Learn how to modify font, shape, and style of text
Use boxes to limit the text width
See how to break lines and how to improve hyphenation
Explore justification and formatting of paragraphs
By working with examples and trying out new features, we shall learn some basic
concepts of LaTeX. By the end of this chapter, we will be familiar with
commands and environments. You will even be able to define your own
commands.

Understanding logical formatting
In the previous chapter, we wrote a small example document. Let's extend it a bit
to get an illustrative example for understanding the typical document structure.

Time for
document

action

–

titling

your

We will take the first example and insert some commands that will produce a
nice-looking title.
1. Type the following code in the editor; modify the previous example if you
like:
\documentclass[a4paper,11pt]{article}
\begin{document}
\title{Example 2}
\author{My name}
\date{January 5, 2011}
\maketitle
\section{What's this?}
This is our second document. It contains a title and a
section with text.
\end{document}

2. Click the Typeset button.
3. View the output:

What just happened?
In the first chapter, we talked about logical formatting. First, let's look at this
example from that point of view. We told LaTeX that:
Our document is of the type article. It will be printed on A4 paper using a
size of 11 points for the base font.
The title is "Example 2".
You are the author.
The document was written on January 5, 2011.
Concerning the content of the document:
It begins with a title.
The first section shall have the heading "What's this?"
The following text is "This is our second document."
Note, we did not choose the font size of the title or heading; neither did we make
something bold or centered. Such formatting is done by LaTeX but nevertheless
you're free to tell LaTeX how it actually should look.

Note
We did not need to press the Save button. TeXworks automatically saves the
document if we click the Typeset button.

Exploring the document structure
Let's look at the details. A LaTeX document doesn't stand alone—commonly the
document is based on a versatile template. Such a fundamental template is called
a class . It provides customizable features, usually built for a certain purpose.
There are classes for books, for journal articles, for letters, for presentations, for
posters, and many more; hundreds of reliable classes can be found in Internet
archives but also on your computer now, after you've installed TeX Live. Here
we have chosen the article class, a standard LaTeX class suitable for smaller
documents.
The first line starts with \documentclass. This word begins with a backslash;
such a word is called a command . We used commands to specify the class and to
state document properties: title, author, and date.
This first part of the document is called the preamble of the document. This is
where we choose the class, specify properties, and in general, make documentwide definitions.
\begin{document} marks the end of the preamble and the beginning of the actual
document. \end{document} marks the end of it. Everything that follows would
be ignored by LaTeX. Such a piece of code, framed by a \begin … \end

command pair, is called an environment .
In the actual document, we've used the command \maketitle that prints the title,
author, and date in a nicely formatted manner. By the \section command, we
produced a heading, bigger and bolder than normal text. Then we let some text
follow. What we wrote here, in the document environment, will be printed out.
On the contrary, the preamble will never produce any output.
Let's look at commands in detail.

Understanding LaTeX commands
LaTeX commands begin with a backslash, followed by big or small letters.
LaTeX commands are usually named with small letters and in a descriptive way.
There are exceptions: you will see some commands consisting of a backslash and
just one special character.
Commands may have arguments , given in curly braces or in square brackets.
Calling a command looks like the following:
\command

Or:
\command{argument}

Or:
\command[optional argument]{argument}

There could be several arguments, each of them in braces or brackets. Arguments
in curly braces are mandatory. If a command is defined to require an argument,
one has to be given. For example, calling \documentclass would be futile if we
hadn't stated a class name.

Note
Arguments in square brackets are optional; they may be given but it's not a must. If
no optional argument is provided, the command will use a default one. For
instance, in the first example in Chapter 1, Getting Started with LaTeX, we wrote
\documentclass{article}. This document has been typeset with a base font
size of 10pt, because this is the class default base font size. In the second
document, we wrote \documentclass[a4paper,11pt]{article}; this way, we
replaced default values with the given values, so now the document will be
adjusted for A4 paper using a base font size of 11pt.
There are commands generating output—try \LaTeX—and commands setting
properties, changing fonts or layout. Generally, the names of commands are
chosen according to their purpose. We will have a more detailed look in this
chapter, but first let's see how LaTeX treats what we type.

How LaTeX reads your input
Before we continue writing, let's look at how LaTeX understands what you've
written in the editor.

Time for action – trying out the effect
of spaces, line breaks, and empty lines
We will take the first example and insert spaces and line breaks.
1. Modify the previous example as follows:
\documentclass[a4paper,11pt]{article}
\begin{document}
\title{Example 3}
\author{My name}
\date{January 5, 2011}
\maketitle
\section{What's this?}
This
is
our
second document.
It contains two paragraphs. The first line of a paragraph
will be
indented, but not when it follows a heading.
% Here's a comment.
\end{document}

2. Typeset.
3. View the output:

What just happened?
Though we've inserted some spaces, the distances between the words in the
output remained the same. The reason is that LaTeX treats multiple spaces just
like a single space. Also, a single line break has the same effect like a single
space. It doesn't matter how you arrange your text in the editor using spaces or
breaks, the output will stay the same.
A blank line denotes a paragraph break. Like spaces, multiple empty lines are
treated as one.
Briefly said, spaces separate words, empty lines separate paragraphs.

Commenting your source text
You've seen that the last line seems to be missing in the output. That's because the
percent sign introduces a comment . Everything following a percent sign until the
end of the line will be ignored by LaTeX and won't be printed out. This enables
you to insert notes into your document. It's often used in templates to inform the
user what the template does at that certain place. Note also that the end of the line,
normally behaving like a space, will be ignored after a percent sign.

Tip
Easing experimenting by trial and error
If you want to disable a command temporarily, it may be favorable to insert a
percent sign instead of deleting the command. That way, you're able to undo this
change easily by removing the %.
If the percent sign behaves that way, what should we do if we want to write
100% in our text? Let's figure out how to do it.

Printing out special symbols
Common text mostly contains upper and lowercase letters, digits, and punctuation
characters. Simply type them with your editor. However, some characters are
reserved for LaTeX commands; they cannot be used directly. We already
encountered such characters, and besides the percent sign, there are the curly
braces and so on. There are LaTeX commands to print such symbols.

Time for action – writing special
characters in our text
We will write a very short example printing out an amount of dollars and a
percent number, then we shall try more symbols:
1. Create a new document and enter the following lines:
\documentclass{article}
\begin{document}
Statement \#1:
50\% of \$100 makes \$50.
More special symbols are \&, \_, \{ and \}.
\end{document}

2. Typeset and view the output:

What just happened?
By putting a backslash before such a special symbol, we turned it into a LaTeX
command. This command has the only purpose of printing out that symbol.

Note
The command for printing a backslash is \textbackslash. If you would like to
know what \\ might be used for, it is used as a shortcut for a line break command.
That's a bit odd, but line breaks occur frequently whereas backslashes are rarely
needed in the output, therefore this shortcut has been chosen.
There's a wealth of symbols that we can use for math formulas, chess notation,
zodiac signs, music scores, and more. We don't need to deal with those symbols
for now, but we shall return to that subject in Chapter 8, Typing Math Formulas,
when we will need symbols to typeset math formulas.
Now that we know how to enter pure text, let's find out how we can format it.

Formatting text – fonts, shapes, and
styles
LaTeX already does some formatting. For example, we've seen that section
headings are bigger than normal text and bold faced. Now we will learn how to
modify the appearance of the text ourselves.

Time for action – tuning the font
shape
We will emphasize an important word in a text and we will see how to make
words appear bold, italic, or slanted. We shall figure out how to highlight words
in a part of some text that's already emphasized:
1. Create a new document containing the following code:
\documentclass{article}
\begin{document}
Text can be \emph{emphasized}.
Besides being \textit{italic} words could be \textbf{bold},
\textsl{slanted} or typeset in \textsc{Small Caps}.
Such commands can be \textit{\textbf{nested}}.
\emph{See how \emph{emphasizing} looks when nested.}
\end{document}

2. Typeset and have a look at the output:

What just happened?
At first, we used the command \emph, giving one word as an argument to this
command. This argument will be typeset in italic shape, because this is the default
way how LaTeX emphasizes text.
Text-formatting commands usually look like \text**{argument}, where **
stands for a two letter abbreviation like bf for bold face, it for italic, and sl for
slanted. The argument will then be formatted accordingly like we've seen. After
the command, the subsequent text will be typeset as it was before the command—
precisely after the closing curly brace marking the end of the argument. We
checked it out.
We've nested the commands \textit and \textbf , which allowed us to achieve
a combination of those styles, and the text appears both italic and bold.
Most font commands will show the same effect if they are applied twice like
\textbf{\textbf{words}}: the words won't become bolder. Only \emph
behaves differently. We've seen that emphasized text will be italic, but if we use
\emph onto a piece of this text again, it will change from italic to normal font.
Imagine an important theorem completely typeset in italics—you should still have
the opportunity to highlight words inside this theorem.
is so called semantic markup , because it refers to the meaning, not just to
the appearance of text.
\emph

Tip
Emphasizing twice, such as marking bold and italic at the same time, might be
considered to be a questionable style. Change the font shape wisely—and
consistently.

Choosing the font family
Compare the font of our examples and the standard font you see in this book.
While the LaTeX font has a decorative appearance, the text font of this book looks
simple and clean. Our code examples are different in another way: every letter
has the same width. Let's see how we can implement this in our writings.

Time for action – switching to sansserif and to typewriter fonts
Imagine that we start to write an article about LaTeX's Internet resources. To get
a clearly readable heading, we shall use a font without frills. The body text will
contain a web address; we choose a typewriter font to stress it:
1. Create a LaTeX document with the following code:
\documentclass{article}
\begin{document}
\section{\textsf{\LaTeX\ resources on the internet}}
The best place for downloading LaTeX related software is
CTAN.
Its address is \texttt{http://www.ctan.org}.
\end{document}

2. Typeset and look at the result:

What just happened?
We encountered more font commands. By using \textsf, we've chosen the sansserif font in the section heading. We used the command \texttt to get the
typewriter font for the web address. Those commands can be used just like the
font commands we've learned before.
The letters in the LaTeX standard font have so-called serifs , those small
decorative details at the end of a letter's strokes. Serifs shall improve readability
by leading the reader's eyes along the line. Therefore, they are widely used in
body text. Such fonts are also called Roman fonts . This name led to the command
\textrm for the Roman text—the default font with serifs.
Headings are often done without serifs; the used font is called a sans-serif font .
Such fonts are also a good choice for screen text because of the better readability
on lower resolutions. So you might want to choose a sans-serif font when you
produce an e-book.
If every letter of a font has the same width, the font is called monospaced or a
typewriter font. Besides, on typewriters, such fonts were used with early
computers; today they are still preferred for writing source code of computer
programs, both in print and in text editors. So if you want to typeset a program
listing or LaTeX source code, consider using a typewriter font. Like we did in the
previous example, this book is using a typewriter font to distinguish code and
web addresses from the normal text.

Switching fonts
Putting too much text into a command's argument could be unhandy. Sometimes we
would like to set font properties of longer passages of text. LaTeX provides other
commands, which work like switches.

Time for action – switching the font
family
We will modify the previous example using font family switching commands:
1. Modify the example to get the following code:
\documentclass{article}
\begin{document}
\section{\sffamily\LaTeX\ resources in the internet}
The best place for downloading LaTeX related software is
CTAN.
Its address is \ttfamily http://www.ctan.org\rmfamily.
\end{document}

2. Typeset and compare the output to the previous one; it's the same.

What just happened?
By using the command \sffamily, we switched over to sans serif font. This
change has been made inside an argument, so it's valid only there.
We used the command \ttfamily to switch to a typewriter font. The typewriter
font will be used from this point onwards. By using \rmfamily, we returned to
Roman font.
These commands don't produce any output, but they will affect the following text.
We will call such a command a declaration .
Now have a closer look at the section number: it's a digit with serifs, which
doesn't match the remaining sans-serif heading. Moreover, changing the font
within a \section command feels wrong—and rightly so! The better way is to
declare the section heading font once for the complete document. We will learn
how to globally modify heading fonts in Chapter 11, Enhancing Your Documents
Further, after we prepared some more tools.

Summarizing font commands and declarations
Let's list the font commands and their corresponding declarations together with
their meanings:

Note
The corresponding declaration to \emph is \em.

Delimiting the effect of commands
In the previous example, we've reversed the effect of \ttfamily by writing
\rmfamily. To be safe, we could write \normalfont to switch back to the base
font. However, there's an easier way.

Time for action – exploring grouping
by braces
We shall use curly braces to tell LaTeX where to apply a command and where to
stop that:
1. Modify our first font shape example to get this code:
\documentclass{article}
\begin{document}
{\sffamily
Text can be {\em emphasized}.
Besides being {\itshape italic} words could be {\bfseries
bold},
{\slshape slanted} or typeset in {\scshape Small Caps}.
Such commands can be {\itshape\bfseries nested}.}
{\em See how {\em emphasizing} looks when nested.}
\end{document}

2. Typeset and check out the output:

What just happened?
We started with an opening curly brace. The effect of the following command
\sffamily lasted until we stopped it with the corresponding closing brace. That
closing brace came at the end of the highlighted code. This highlighting shows the
area of the code where \sffamily is valid.
We replaced every font command by the corresponding declaration. Remember,
\em is the declaration version of \emph. Further, we surrounded every
declaration and the affected text by curly braces.
An opening curly brace tells LaTeX to begin a so called group . The following
commands are valid for the subsequent text until a closing curly brace appears
causing LaTeX to stop using the commands or declarations written in this group.
Till a command is valid, that's called its scope .
Groups can be nested as follows:
Normal text, {\sffamily sans serif text {\bfseries and bold}}.

We have to be careful to close each group; opening and closing braces should
match.

Note
Braces which enclose an argument of a command don't form a group. Together
with the argument, these braces are gobbled by the command. If necessary, use
additional braces.

Time for action – exploring font sizes
We will try out every font size available by LaTeX's default font size commands.
For testing, we exceptionally use them in the body text—their main use is in the
macro definitions:
1. Create a document with the following code:
\documentclass{article}
\begin{document}
\noindent\tiny We \scriptsize start \footnotesize \small
small,
\normalsize get \large big \Large and \LARGE bigger,
\huge huge and \Huge gigantic!
\end{document}

2. Typeset and observe the output:

What just happened?
At first, we used \noindent . This command suppresses the paragraph
indentation. Then we used all 10 available size declarations, starting small with
\tiny and ending really big with \Huge. There are no corresponding commands
taking arguments, so we would have to use curly braces to delimit their scope, as
we learned to before.
The actual resulting font size depends on the base font. If your document has a
base font of 12 pt, then \tiny would result in text bigger than with a base font of
10 pt. We have to see it in relation. Use \footnotesize , if you wish to get the
same size like LaTeX uses for footnotes; use \scriptsize , if you create a style
with a size matching LaTeX subscripts and superscripts. It's still a kind of logical
formatting though they are quite low-level commands.
Normally, font size declarations are used only in definitions of macros in the
preamble, just as it does apply to the other font commands. You will rarely
encounter font size or shape commands in good body texts, except freely designed
passages like title pages—and test examples like the ones here.

Using environments
When you use several declarations and you group them by curly braces, will you
always know which closing brace matches which of the previously entered
declarations? An environment forms a group. Using an environment instead of just
curly braces improves the readability of your code.

Time for action – using an
environment to adjust the font size
We will produce a title with a larger and bigger font:
1. Create another small document with this code:
\documentclass{article}
\begin{document}
\begin{huge}
\bfseries
A small example
\end{huge}
\bigskip
This is just another small illustrative example.
\end{document}

2. Typeset to see the result:

What just happened?
By writing \begin{huge}, we told LaTeX to switch to a huge font size just like
with the \huge command. \end{huge} informs LaTeX that this size change
should end now. From this point onwards, the font size is the same as before
\begin{huge}.
Inside this environment, we use the declaration \bfseries . Note that the effect
of this declaration ended together with the end of the environment.
The empty line before \end{huge} denotes a paragraph break. Using the
command \bigskip , we skipped some space vertically.
For every declaration there's a corresponding environment carrying the same
name except the backslash. Using environments instead of braces might make
complex code easily understandable.

Note
Commonly it's advisable to end the paragraph before a font size change, not after
it. That's because TeX calculates the interline spacing depending on the current
font size when it reaches the end of the paragraph. That's the reason we used the
blank line before \end{huge} instead of afterwards.

Saving time and effort – creating your
own commands
If you're frequently using the same term in your document, it would be annoying to
type it again and again. What if you later decided to change that term or its
formatting? To avoid searching and replacing in the whole document, LaTeX
allows you to define your own commands in your preamble. They are also called
macros . Give it a name. Later in the document you just need to use this name
whenever you want to change this term. We need to do it just once in the macro
definition. This will affect the whole document. Let's see how it works.

Time for action – creating our first
command using it as anabbreviation
We will define a short command printing out the name of the TeX Users Group:
1. Type this code into a new document:
\documentclass{article}
\newcommand{\TUG}{TeX Users Group}
\begin{document}
\section{The \TUG}
The \TUG\ is an organization for people who are interested
in \TeX\ or \LaTeX.
\end{document}

2. Typeset and look at the result:

What just happened?
in the highlighted line defines our command. The first argument is
the name we chose for it, and the second argument is the text we want it to put out
in the document.
\newcommand

Now, whenever we type \TUG in our document, the complete name will appear. If
we later decide to change the name or its formatting, we just need to change this
\newcommand line. Then it will be applied for the complete document.
You may use formatting commands inside your command definition. Let's say you
would like to change the formatting of all occurrences of this name to be typeset
in small caps; just change the definition to the following:
\newcommand{\TUG}{\textsc{TeX Users Group}}

You have also seen that we've used the command \TeX . This command just prints
out the name of the basic typesetting system formatted like in its logo. It's an
abbreviation command like we've written now and \LaTeX works similarly. Note
that we had to use a backslash after \TeX! The following space would just
separate the command from the following text; it won't produce a space in the
output. Using the backslash followed by a space forces the output of the space that
would otherwise be ignored. You may try omitting that backslash to be convinced.

Gentle spacing after commands
A backslash following a command could easily be forgotten. Can't we modify the
command in order to automate that? Tasks like this, which aren't supported by
LaTeX directly, could be solved by using packages. These are collections of
styles and commands.

Time for action – adding intelligent
spacing to command output
We will load the package xspace. Its only purpose is to fulfill this need.
1. Modify the previous example to get the following code:
\documentclass{article}
\usepackage{xspace}
\newcommand{\TUG}{\textsc{TeX Users Group}\xspace}
\begin{document}
\section{The \TUG}
The \TUG is an organization for people who are interested in
\TeX\ or \LaTeX.
\end{document}

2. Typeset, see that the spacing between the words is correct, even without the
backslash:

What just happened?
tells LaTeX to load the package called xspace and to
read in all of its definitions. From now on we may use all commands contained in
that package. This package provides the command \xspace that inserts a space
depending on the following character: If a dot, a comma, an exclamation, or a
quotation mark follows, it won't insert a space, but if a normal letter follows, then
it will. Usually, that's exactly what we want.
\usepackage{xspace}

Imagine you've mentioned the TUG several times in your document and now
you've got the idea to use the TeX logo style in its name. There's no need for
changes in the document. Now only the command in the preamble needs
adjustment:
\newcommand{\TUG}{\textsc{\TeX\ Users Group}\xspace}

Tip
The heading in our last example doesn't contain small caps—have you seen it?
Not all font properties can be combined, depending on the chosen font. For
instance, fonts with small caps together with variations like bold and italic are
rare. Therefore, some people even fake small caps if they don't want to change to
a more complex font.
By defining and using commands, you can ensure that the formatting remains
consistent throughout your whole document.

Creating more
arguments

universal

commands

–

using

Imagine that your text contains a lot of keywords that you want to be printed in
bold. If you use the \textbf{} command on all the keywords, what would
happen if you later decide to use an italic shape instead or a typewriter font? You
would have to change that formatting for each keyword. There's a better way:
defining your own macro and using \textbf{} only inside that.

Time for action – creating a macro
for formatting keywords
We will use \newcommand again, but this time we will introduce a parameter that
will contain our keyword. Let's just use it on some terms that we've got to know in
this chapter:
1. Type this code example:
\documentclass{article}
\newcommand{\keyword}[1]{\textbf{#1}}
\begin{document}
\keyword{Grouping} by curly braces limits the
\keyword{scope} of \keyword{declarations}.
\end{document}

2. Typeset and notice the look of the keywords in the output:

What just happened?
Let's look at the highlighted line in the code. The number 1 in the square brackets
marks the number of arguments that we want to use in the command. #1 will be
replaced by the value of the first argument. #2 would refer to a second argument,
and so on.
Now if you want to modify the appearance of all keywords to be italic, just
modify the definition of \keyword and the change will be global.

Using optional arguments
In one preceding example, we've used \newcommand with two arguments. In the
previous example, there were three arguments. The additional argument has been
put in square brackets. That's the way we mark optional arguments : those
arguments may be given or may be omitted. If omitted, they would have a default
value. We've already noticed that with the \documentclass command. But how
can we define a command with optional arguments ourselves?

Time for action – marking keywords
with optional formatting
We will use \newcommand another time, this time with an optional parameter
concerning the formatting and a mandatory argument for the keyword:
1. Modify the previous example to get this code:
\documentclass{article}
\newcommand{\keyword}[2][\bfseries]{{#1#2}}
\begin{document}
\keyword{Grouping} by curly braces limits the
\keyword{scope} of \keyword[\itshape]{declarations}.
\end{document}

2. Typeset and check out the result:

What just happened?
Let's look again at the bold marked line in the code. By using [\bfseries], we
introduced an optional parameter. We refer to it with #1. Its default value is
\bfseries. Since we used a declaration this time, we added a pair of braces to
ensure that only the keyword is affected by the declaration.
Later in the document, we gave [\itshape] to \keyword, changing the default
formatting to italics.
Here's the definition of the \newcommand:
\newcommand{command}[arguments][optional]{definition}
command

The name of the new command, starting with a backslash followed by lowercase and/or
uppercase letters or a backslash followed by a single non-letter symbol. That name must not be
already defined and is not allowed to begin with \end.

arguments

An integer from 1 to 9, the number of arguments of the new command. If omitted, the
command will have no arguments.

optional

If this is present, then the first of the arguments would be optional with a default value given
here. Otherwise all arguments are mandatory.

definition

Every occurrence of the command will be replaced by definition and every occurrence of the
form #n will then be replaced by the nth argument.

is our key to introduce logical formatting. We should avoid using
LaTeX font commands inside the document—you are on the right track if they
appear only in the preamble of the document. Use \newcommand to create styles
for keywords, code snippets, web addresses, names, notes, information boxes, or
differently emphasized text. How did we achieve the consistent structure of this
book? Using styles is the key!
\newcommand

Pop quiz – commands
1. Imagine your document contains some addresses of websites. Let's say we
want them to be typeset in typewriter font. According to this book, which of
the following possibilities would be the best way to print out, for instance,
http://ctan.org?
a. \texttt{http://ctan.org}
b. {\ttfamilyhttp://ctan.org}
c . \newcommand{\CTAN}{\texttt{http://ctan.org}} in the preamble
and \CTAN in the body text
d . \newcommand{\site}[1]{\texttt{#1}} and \newcommand{\CTAN}
{http://ctan.org} in the preamble and \site{\CTAN} in the body text
2. Which kind of punctuation marks are used to enclose optional arguments?
a. Parentheses: ()
b. Square brackets: []
c. Curly braces: {}

Have a go hero – saving effort using optional
arguments
Most website addresses begin with http://. But there are sites that differ—
some may start with ftp://, so called FTP server, some sites begin with
https://, so called secure web server. Visit, for instance, ftp://ctan.org, and you
will enter a file server instead of a web server.
Extend the definition of \site of our last pop quiz. Introduce an optional
argument denoting the protocol of the site, that is, how the address starts. Default
should start with http://. It shall work like the following:
\site{www.tug.org}

prints out http://www.tug.org in typewriter font

\site[ftp]{ctan.org}

prints out ftp://ctan.org in typewriter font

Tip
Web addresses in texts
There's a package called url designed for typesetting web addresses. Write
\usepackage{url} in your preamble; this will provide the command \url. This
command takes an address for the argument and will print it out with typewriter
font. Furthermore, it is able to handle special characters in addresses like
underscores and percent signs. It even enables hyphenation in addresses, which is
useful for websites with a very long name.

Using boxes to limit the width of
paragraphs
We won't always write the text just from left to right over the complete text width.
Sometimes, we'd like a paragraph to have a smaller width, for instance, when we
would like to put text and a picture side-by-side.

Time for action – creating a narrow
text column
We would like to explain the acronym TUG in a text column of only 3 cm width:
1. Create a new document containing these four lines:
\documentclass{article}
\begin{document}
\parbox{3cm}{TUG is an acronym. It means \TeX\ Users Group.}
\end{document}

2. Typeset and take a critical look at the output:

What just happened?
We used the command \parbox to create a column. We stated the width of 3 cm
in the first argument and the contained text in the second argument to \parbox.
takes the argument text and formats the output to fit the specified width.
We see that the text is fully justified. Our example shows an obvious problem:
insisting on full justification could lead to undesirable big gaps in the text.
Possible solutions are:
\parbox

Introducing hyphenation: the word acronym could easily be divided
Improving justification: LaTeX could do better
Giving up full justification: narrow text could look better when it's only left
justified
We will check out all of these options. But first, let's see how \parbox is
working.

Common paragraph boxes
Usually we just need a text box with a certain width; occasionally we would like
to have some additional alignment to the surrounding text. So the common
definition of the \parbox command is:
\parbox[alignment]{width}{text}
alignment

Optional argument for the vertical alignment. State t to align at the top line of the box; write b to
align at its bottom line. The default behavior is to place the box such that its center is in line with
the center of the current text line.

width

The width of the box. It can be given for example in ISO units like 3 cm, 44 mm, or 2 in.

text

The text that you want to put in that box. It should be a short piece of common text. For
complicated contents, we will get to know other methods.

Here's a demonstration of the effect of the alignment parameters:
\documentclass{article}
\begin{document}
Text line
\quad\parbox[b]{1.8cm}{this parbox is aligned at its bottom
line}
\quad\parbox{1.5cm}{center-aligned parbox}
\quad\parbox[t]{2cm}{another parbox aligned at its top line}
\end{document}

The command \quad produces some space; we used it to separate the boxes a bit.
Here's the output:

From now on we will call such a box a parbox.

Have a go hero – exploring further features of \parbox
\parbox

is capable of doing even more. Here's the complete definition:

\parbox[alignment][height][inner alignment]{width}{text}
height

If this optional argument isn't given, the box will have just the natural height of the text inside.
Use this argument if you want to change the height of the box to make it bigger or smaller.
Especially, if the height of the box is different to the natural height of the contained text, you
might want to adjust the text position. The argument means:
c—vertically center

inner
alignment

the text in the box

t—place

text at the top of the box

b—place

text at its bottom

s—stretch the

text vertically if possible

If you omit this argument, the alignment argument will be used here as the default value.

Take our previous demonstration example and try the effect of the optional
arguments. Use the command \fbox that helps to visualize the effect; if you write
\fbox{\parbox[...]{...}{text}}, the complete parbox will be framed.

Boxes containing more text
Parboxes are suitable for boxes with only a little text inside. In case of a box
containing a large amount of text, the closing brace could easily be forgotten or
overlooked. The minipage environment would then be a better choice.

Time for action – using the minipage
environment
We will use the minipage environment instead of \parbox to get a text with a
width of just 3 cm.
1. Modify the parbox example to get the following code:
\documentclass{article}
\begin{document}
\begin{minipage}{3cm}
TUG is an acronym. It means \TeX\ Users Group.
\end{minipage}
\end{document}

2. Typeset and look at the output:

What just happened?
By using \begin{minipage}, we started a "page in a page". We specified the
width of 3cm by the mandatory argument. From this point onwards, the text lines
will have a width of 3cm. They will be automatically wrapped and fully justified.
We ended this restriction with \end{minipage}. Any text typed afterwards
would run over the complete body text width.
Note: There won't be a page break in such a "minipage".
Have you been a hero struggling with the \parbox and all of its optional
arguments? The minipage environment accepts all those arguments as well with
the same meaning.

Have a go hero – creating a footnote inside a minipage
Of course, LaTeX supports footnotes. You guessed it right—the corresponding
command is called \footnote{text}. Try it in one of our examples. You will
get a footnote mark inside the text and the text of the footnote will be put into the
bottom of the page.
A minipage is like a small page inside a normal page. Think of using a footnote
inside such a "page"; do you want to see the footnote text inside this small page?
The minipage environment supports it, just try it—use \footnote inside the
minipage in the previous example.

Understanding environments
LaTeX environments are started with \begin and ended with \end. Both
commands require the name of the environment as their argument.
Simple environments look like the following:
\begin{name}
…
\end{name}

Such environments can be used for each declaration called \name.
Like commands, environments may also have arguments. Exactly like in the case
of commands, mandatory arguments are written in curly braces and optional
arguments in square brackets. So you will encounter:
\begin{name}{argument}
…
\end{name}

And:
\begin{name}[optional argument]{argument}
…
\end{name}

Environments are like declarations with a built-in scope. With \begin, the
environment introduces a change in layout, font, or other properties. There must
be an \end command, where this change will be canceled. The effect of the
environment name is delimited to the piece of code between \begin{name} and
\end{name}.
Furthermore, the effect of all local declarations used inside an environment will
end together with the surrounding environment.

Breaking lines and paragraphs
Generally, when you're writing text, you don't need to care about the line
wrapping. Just type the text with your editor; LaTeX will make it fit to the line
and take care of the justification. If you want to begin a new paragraph, in
consequence of getting a line break in the output, just insert an empty line before
you continue with your text.
Now we will find out how to control the line wrapping. First we will see how to
improve the automatic line breaking. Then we will learn commands to insert
breaks directly.

Improving hyphenation
If you look at longer texts, you will notice that it's outstanding how the text is fully
justified by LaTeX and how the spacing between words is evenly distributed on
the lines. If necessary, LaTeX will divide words and put hyphens at the end of the
line in order to break the lines in a better way. LaTeX already uses very good
algorithms to hyphenate words, but it may happen that it can't find an acceptable
way to divide a word. The previous example pointed out this problem: breaking
the word acronym would improve the output, but LaTeX does not know where to
divide it. We shall find out how to solve that.

Time for action – stating division
points for words
No matter how good the justification skill is, text in very narrow columns is
extremely hard to justify. The previous example showed it pitiless. We will tell
LaTeX how a word could be divided:
1. Insert the following line into the preamble of the previous example:
\hyphenation{acro-nym}

2. Typeset and look at the output:

What just happened?
We've told LaTeX that the word acronym may have a division point between
acro and nym. That means a hyphen might be put after acro at the end of the line
and nym goes to the following line.
The \hyphenation command tells LaTeX where the division points of a word
may be. Its argument may contain several words separated by spaces. For each
word, we can indicate several points. For instance we could extend the argument
by more division points and more word variants like this:
\hyphenation{ac-ro-nym ac-ro-nym-ic a-cro-nym-i-cal-ly}

You could also indicate division points in the body text by inserting a backslash
followed by a hyphen, like ac\-ro\-nym. But if you do it in the preamble, you'll
collect all rules there and they will be used consistently. Use it especially in the
rare cases when LaTeX's automation fails.

Tip
Preventing hyphenation
If you want to prevent the hyphenation of a certain word at all, there are two
ways: either declare it in the preamble by using it in the \hyphenation argument
without any division points, like \hyphenation{indivisible}, or you protect it
inside the text using the \mbox command: The following word is
\mbox{indivisible}.

Have a go hero – exploiting the hyphenat package
The hyphenat package extends the possibilities:
\usepackage[none]{hyphenat}

prevents hyphenation throughout the

document.
enables hyphenation for typewriter text.
Otherwise such monospaced fonts won't be hyphenated by LaTeX.
\usepackage[htt]{hyphenat}

Such optional arguments to \usepackage are called package options. They
configure the behavior of a package. The mentioned options may be combined,
separated by commas. Even if you don't use the option none, you can disable
hyphenation for short pieces of text using the command \nohyphens{text}. Try
out these features if you want to profit from them. The package documentation
explains more features that you might need sometimes, such as hyphenation after
special characters like numerals and punctuation characters—TeX would not
break there.

Improving the justification further
Today's most popular TeX compiler is pdfTeX, which directly produces PDF
output. When Hàn Thế Thành developed pdfTeX, he extended TeX by microtypographic capabilities. When we typeset directly to PDF, we're actually using
pdfLaTeX and we can benefit from the new features by using the microtype
package.

Time for action – using microtype
We will improve our previous example by loading the microtype package :
1. Insert the following line into the preamble of the previous example:
\usepackage{microtype}

2. Typeset and look at the output:

What just happened?
We have loaded the microtype package without any options, relying on its
default behavior. It introduces font expansion to tweak the justification and uses
hanging punctuation to improve the optical appearance of the margins. This may
reduce the need of hyphenation and improves the "grayness" of the output. You've
seen its effect on a narrow column—imagine the improvement on wide text—keep
that in mind and try it out later!
Though microtype provides powerful features and options for the advanced
typesetter, we usually won't need to do more than just load it to profit from it.
There's an extensive package documentation if you want to study it in depth.
does nice tweaking, but it's not a cure-all; we should still take care of
proper hyphenation.
microtype

Breaking lines manually
We might choose to end a line overriding the automatism. There are several
commands with different effects.

Time for action – using line breaks
We will type the beginning of a famous poem by Edgar Allan Poe. As the author
has specified where a verse has to end, we shall insert line breaks there.
1. Create a document containing these lines:
\documentclass{article}
\begin{document}
\emph{Annabel Lee}\\
It was many and many a year ago,\\
In a kingdom by the sea,\\
That a maiden there lived whom you may know\\
By the name of Annabel Lee
\end{document}

2. Typeset and view the output:

What just happened?
The very short command \\ ended a line; the following text was put to the next
line. That's different to a paragraph break: we're still in the same paragraph. The
command called \newline has the same effect.
There's another command called \linebreak, which tells LaTeX to end the line
but to keep the full justification. Therefore, the space between the words would
be stretched to reach the right margin. This could cause unpleasant gaps—that's
why that command is rarely used.

Have a go hero – exploring line breaking options
Both introduced commands understand optional arguments.
\\[value] would insert additional
on the value, like \\[3mm].

vertical space after the break depending

is a variation of the same but prevents a page break before the
next line of text.
\\*[value]

\linebreak[number] can be
strongly: If number is 0, a line

used to influence the line break slightly or
break is allowed, 1 means it's desired, 2 and
3 mark more insistent requests, and 4 will force it. The latter is the default
behavior if no number was given.
Change the heading of our poem example to the following:
\emph{Annabel Lee}\\[3mm]

This will insert some space between our heading and the poem fragment. Try the
options further and check out their effects. If you're daring, try \linebreak
instead of \\ to end a poem line just to see its effect.

Preventing line breaks
The command \linebreak has a direct counterpart: \nolinebreak. This
command prevents a line break at the current position. Like its counterpart, it
takes an optional argument. If you write \nolinebreak[0], you recommend to
not break the line there. Using 1, 2, or even 3 makes the request stronger and
\nolinebreak[4] forbids it completely. The latter will be presumed if you don't
provide an argument.
The already mentioned command, \mbox[text], does not only disable
hyphenation of a word, it will also prevent a line break for the complete text.
LaTeX will break lines at spaces between words if meaningful. The symbol ~
stands for an interword space where no break is allowed: if you would write
Dr.~Watson, the title Dr. would never stand lonely at the end of a line.

Managing line breaks wisely
Bad hyphenation could still disappear as the document grows, so stating some
sensible hyphenation rules would not do any harm but could prove to be useful.
But only use \\, \newline, and \linebreak for line adjustment when you're
working on the final version of your document! While you're still editing your
text, you don't need to worry about line breaks. They still may change during the
writing process. Bad looking justification could still change and become better
without intervention. On the other hand, if you break a line manually but later
insert text before, the result could be an unwanted short line.
So don't waste your energy on formatting while you're writing.

Exploring the fine details
Typographic conventions may require paying attention to small details; there are
different dashes, and the space around a dot may vary depending on the context.
The space after some letters may depend on the following one, so much so that
some letters may even be joined to a single one. Such constructions are called
ligatures. Let's have a closer look at them.

Time for action – exploring ligatures
We will check out Example 3 to discover a ligature. Afterwards, we will have a
close look at the default ligatures:
1. Open Example 3 in TeXworks, click the Typeset button. Move the mouse
pointer into the output window, right over the word first, and then click the
left button.

2. Go to the editor window, and then add a paragraph break by with empty line
and the following letter combinations: ff fi flffiffl -- --3. Add a paragraph break, repeat the preceding line, but this time insert \/
between the letters: f\/f f\/i f\/l f\/f\/i f\/f\/l -\/- -\/-\/4. Typeset and compare the letters in the output:

What just happened?
By left-clicking the mouse in the output window, you activated TeXworks'
magnifying glass. It allows inspecting the output in detail as the area around the
mouse pointer will be enlarged. You can see that two or even three letters have
been joined, but not in the last line: the command \/ prevented this.

Understanding ligatures
If the letter i follows the letter f, the dot of the letter i could collide with the head
of the f. It could be avoided in two ways: either by increasing the space between
those two letters or by combining them. The latter will result in a ligature. This
will automatically be done by LaTeX when the used font provides such ligatures.
Sometimes this behavior might be undesirable, for instance, in compound words
like "halflife". In this case, you can suppress this behavior by inserting \/ getting
half\/life.

Curly braces can be used to achieve the same effect. For instance, in -{}- and {}{-}, the dashes cannot be joined because this would cross group boundaries.

Tip
If you don't like ligatures, for instance, because your PDF reader doesn't support
search or copy and paste of ligatures, completely suppressing ligatures could be
done easily by passing the option noligatures to the microtype package:
\usepackage[noligatures]{microtype}.

Choosing the right dash
We've seen several kinds of dashes up until now. Let's sort them out:
A short dash is used to mark hyphenation or compound words. In LaTeX: A longer dash is used to indicate a range of some values, such as 2-3 hours.
Generally this dash has the same width like a digit. In LaTeX: -Even wider is the dash commonly used to mark a parenthetical thought. In
LaTeX: --Hyphenation is always done using the short dash. Regarding the wider dashes, it's
a matter of style which kind of dash should be used, and further whether there
should be a space around the dash or not. You decide—but now you know how to
typeset those dashes.

Setting dots
How wide is the space following a dot? Some styles require additional space
after the period that ends a sentence, but not after the period marking an
abbreviation. Furthermore, three consecutive dots, forming an ellipsis, aren't
usually written by just typing three dots. Let's figure this out.

Time for action – using differently
spaced dots
We shall typeset a short text containing periods to see the default behavior.
Afterwards, we will create an example where LaTeX's automation fails; we shall
see how to correct it. We will learn how to simplify all this spacing using a
different style.
1. Create a new document in TeXworks:
\documentclass{article}
\begin{document}
\TeX\ was designed by Donald E. Knuth.
It has become a de facto standard in mathematics.
\end{document}

2. Typeset to see the output:

3. Now add a paragraph break followed by this line:
Look at the spacing etc. in LaTeX. See it?

4. Check out the spacing after the two periods in the output:
5. Correct the spacing by inserting a backslash and \@:
Look at the spacing etc.\ in LaTeX\@. See it?

6. Now remove the extra \ and \@ we've added, insert the declaration
\frenchspacing into the preamble, and typeset to see the difference:

What just happened?
The first step shows typical text and its output. As you can see, LaTeX adds a
little space after the period ending a sentence. But not following the letter E: in
case of periods after capital letters, LaTeX assumes that it's an abbreviation,
therefore it doesn't add the extra space.
Generally, this automatism is helpful. However, you might encounter exceptions:
in step 3, we constructed such a situation where LaTeX's assumptions are wrong.
We've got unwanted extra space after etc. and missed a space after the last
period.
In step 5, we corrected it: the backslash followed by a space (a "quoted space")
produces the normal interword space, the \@ before the dot states that the
following dot stands at the end of a sentence. Now the spacing is correct.
If you prefer to avoid this extra space or if the style you're following requires it,
you could switch off this default sentence spacing. We've done it in step 6 with
the declaration \frenchspacing. Now the spacing after sentences will be as it is
between words.
The counterpart to \frenchspacing is \nonfrenchspacing, which is default.

Time for action – comparing dots to
ellipsis
We will write an ellipsis in two ways: firstly by simply writing dots, secondly by
using a dedicated command. Let's compare!
1. Create a new document in TeXworks:
\documentclass{article}
\begin{document}
Here are three dots... compare them to the ellipsis\ldots
\end{document}

2. Check out the difference:

What just happened?
We used the command \ldots to print out an ellipsis—three consecutive dots
with a wider spacing. Such dots may indicate a pause, an unfinished thought, or an
omitted word. When we just accumulated dots, they were typeset tightly together.
However, it's common to print those dots wider.

Setting accents
Some languages have letters with accents that you can't simply type with your
editor. In case you need to write such letters: let's see how to do it.

Time for action – experimenting with
accents
We will write some words having letters with accents in Portuguese and in
French.
1. Create a new document:
\documentclass{article}
\begin{document}
N\~{a}o compreendo. H\'{a} aqui algu\'{e}mque fale
ingl\^{e}s?
Comment \c{c}a va? O\`{u} se trouve l'a\'{e}roport?
\end{document}

2. Typeset to see the generated accents:

What just happened?
LaTeX provides some special commands to produce a variety of accents. They
may be combined with any letter. Such a command consists of a backslash
followed by one character. The accent will be put above or below the letter that
has to follow in curly braces. The following table will list these commands and
their effect:

Using special characters directly in the editor
It might be a bit cumbersome to use those accent commands. There's a package
that extends LaTeX basic capabilities. It allows you to enter at least some
commonly used accented and other special characters directly.

Time for action – using accents
directly
We will modify the previous example daring to enter accented letters directly in
the editor.
1. Create a new document:
\documentclass{article}
\usepackage[utf8]{inputenc}
\begin{document}
Não compreendo. Há aqui alguém que fale inglês?
Comment çava? Où se trouve l'aéroport?
\end{document}

2. Typeset and compare to the previous output:

What just happened?
We loaded the inputenc package. The option utf8 tells the package to use
Unicode input encoding, which provides many more symbols than just the ASCII
code. Now we just need to find the symbol on the keyboard and to type it.
TeXworks supports Unicode/UTF-8. Depending on operating system and editor,
you might need to use another option when loading inputenc. A rule of thumb:
utf8 works on most Linux and Unix systems, like Mac OS X, and latin1 works
with most Windows editors.

Note
Today, many Windows editors move to UTF-8. This is seemingly becoming the
cross-platform standard.

Turning off full justification
Though commonly your text will look fine if full justification is used, there may
be occasions when it's not the optimum. For instance, full justification could be
displeasing if the text lines are short. In such a case, it could be sufficient to
justify only to the left side. We shall see how to put this into practice, further how
to right-justify, and how to get centered lines.

Time for action –
paragraph to the left

justifying

a

Remember the first parbox example which was fully justified but had those big
gaps between the words. We shall give up justification to the right side to avoid
such gaps.
1. Create a new document containing these lines:
\documentclass{article}
\begin{document}
\parbox{3cm}{\raggedright
TUG is an acronym. It means \TeX\ Users Group.}
\end{document}

2. Typeset and look at the output:

What just happened?
We inserted the declaration \raggedright . From this point onwards, the text
will be ragged-right. In other words, the text will be moved to the left margin "flushed-left". There won't be hyphenation.
Because we used this declaration inside a box, it's only valid there, like inside
environments. After the box, the text will be fully justified again.
If we want the whole document to be ragged-right, we just need to use
\raggedright in our preamble.

Creating ragged-left text
There might be occasions when we would like to achieve the opposite effect:
flushing the text to the right margin. We can do it similarly by inserting the
declaration \raggedleft. Remember, you're able to control where lines are
broken by inserting \\.

Time for action – centering a title
We shall create a nice looking title for our document. It should contain the title,
the author, and the date. All those lines have to be centered.
1. Write a document containing this code:
\documentclass{article}
\pagestyle{empty}
\begin{document}
{\centering
\huge\bfseries Centered text \\
\Large\normalfont written by me \\
\normalsize\today
}
\end{document}

2. Typeset to see the output:

What just happened?
Because only the title should be centered, we opened a group to limit the
centering. Through the declaration \centering , we made the remaining text of
this group horizontally aligned to the center. We inserted a paragraph break: it's
always a good idea to do it before ending the group when paragraph-affecting
commands were used. By the brace after \today , we ended the group. If you
complement some following text, it will be typeset normally, not centered.
is commonly used when pictures or tables are inserted, further on
title pages and sometimes for headings.
\centering

Using environments for justification
As there's a corresponding environment for every declaration, as we learned, we
could have written \begin{centering} … \end{centering} in our previous
example. It could be done similarly for ragged-right and ragged-left text. There's a
couple of predefined environments acting similarly but starting a new paragraph
at the same time.

Time for action – centering verses
Let's reuse the fragment of the poem "Annabel Lee". This time we shall center all
verses:
1. Create a document containing these lines:
\documentclass{article}
\usepackage{url}
\begin{document}
\noindent This is the beginning of a poem
by Edgar Allan Poe:
\begin{center}
\emph{Annabel Lee}
\end{center}
\begin{center}
It was many and many a year ago,\\
In a kingdom by the sea,\\
That a maiden there lived whom you may know\\
By the name of Annabel Lee
\end{center}
The complete poem can be read on
\url{http://www.online-literature.com/poe/576/}.
\end{document}

2. Typeset and see the output:

What just happened?
We began with \noindent avoiding the paragraph indentation. \begin{center}
started the center environment. It begins a new paragraph, leaving some space to
the preceding text. \end{center} ended this environment. We used the
environment twice. In the second one, we inserted \\ to end the verses.
After the center environment ended, some space followed and the next
paragraph began at the left margin.
The corresponding environment for ragged-right text is called flushleft , and for
ragged-left text it's flushright .

Displaying quotes
Imagine your text contains a quotation of another author. It might be hard to read if
it's just embedded in your words. A common way to improve the readability is
displaying: setting the text off by indenting on both margins.

Time for action – quoting a scientist
We will quote thoughts of famous physicists.
1. Create a new document containing these lines:
\documentclass{article}
\begin{document}
Niels Bohr said: ``An expert is a person who has made
all the mistakes that can be made in a very narrow field.''
Albert Einstein said:
\begin{quote}
Anyone who has never made a mistake has never tried anything
new.
\end{quote}
Errors are inevitable. So, let's be brave trying something
new.
\end{document}

2. Typeset to see the result:

What just happened?
Firstly we quoted inline. `produced a left quotation mark; the character is also
called a backtick . 'gave a right quotation mark. We just typed two such symbols
to get double quotes.
Then we used the quote environment to display a quotation. We did not begin a
new paragraph for it, because the quotation is already set a bit off. That's the
reason we don't use a blank line before and after the environment.

Quoting longer text
When writing short quotations, the quote environment looks very good. However,
when you would like to quote a text containing several paragraphs, you might
wish to have the same paragraph indentation like in your surrounding text. The
quotation environment will do it for you.

Time for action – quoting TeX's
benefits
We will quote some of the benefits of TeX and LaTeX found on a web page on
CTAN.
1. Start a new document. This time, it will be a bit longer:
\documentclass{article}
\usepackage{url}
\begin{document}
The authors of the CTAN team listed ten good reasons
for using \TeX. Among them are:
\begin{quotation}
\TeX\ has the best output. What you end with,
the symbols on the page, is as useable, and beautiful,
as a non-professional can produce.
\TeX\ knows typesetting. As those plain text samples
show, \TeX's has more sophisticated typographical algorithms
such as those for making paragraphs and for hyphenating.
\TeX\ is fast. On today's machines \TeX\ is very fast.
It is easy on memory and disk space, too.
\TeX\ is stable. It is in wide use, with a long history.
It has been tested by millions of users, on demanding
input.
It will never eat your document. Never.
\end{quotation}
The original text can be found on
\url{ http://www.ctan.org/what_is_tex.html}.
\end{document}

2. Typeset and look at the output:

What just happened?
This time, we used the quotation environment to display some paragraphs. As in
normal text, blank lines separate the paragraphs. They are left-indented at their
beginning just like in all our body text.
But what if we don't like that paragraph indentation? Let's check out an
alternative.

Time for action – spacing between
paragraphs instead of indentation
We like to avoid the paragraph indentation. Instead, we shall separate the
paragraphs by some vertical space.
1. Create copy of the previous example and reuse most of it this way:
\documentclass{article}
\usepackage{parskip}
\usepackage{url}
\begin{document}
The authors of the CTAN team listed ten good reasons
for using \TeX. Among them are:
\TeX\ has the best output. What you end with,
the symbols on the page, is as useable, and beautiful,
as a non-professional can produce\ldots
The original text can be found on
\url{ http://www.ctan.org/what_is_tex.html}.
\end{document}

2. See the effect:

What just happened?
The highlighted line shows that we've loaded the parskip package. Its only
purpose is to remove the paragraph indentation completely. At the same time, this
package introduces a skip between paragraphs. But this package doesn't affect the
definition of the quotation environment—you still could use the quote
environment.
In order to distinguish paragraphs, there are two common ways. One is to indent
the beginning of each paragraph; this is the default LaTeX style. The other way is
to insert vertical space between paragraphs while omitting the indentation, which
is suitable for narrow columns where indenting would cost too much width.

Pop quiz – lines and paragraphs
1. Which of the following designates the end of a paragraph?
a. \newline
b. \\
c. A blank line
2. How do the words in a \hyphenation command be separated?
a. By commas
b. By semicolons
c. By spaces
3. Which command switches to left-aligned text?
a. \raggedleft
b. \raggedright
c. \flushright

Summary
In this chapter, we developed the basics: editing, arranging, and formatting of text.
Specifically, we covered:
Modifying shape and style of text and its font
Breaking lines and improving hyphenation
Controlling justification of text
We got to know the basic LaTeX concepts:
Commands and declarations, mandatory and optional arguments
Definition of new commands
Using environments
Using packages, how they can be loaded and options to packages
Keep in mind that even though we've used formatting commands directly in the
text when exploring them, you should use them inside command definitions in the
preamble to allow easy changes for the future. During your learning and writing
process, you probably will get to know further useful commands and packages
that could improve your previously written commands.
We've learned general practices:
As often as possible, create your own macros to achieve a logical structure.
You will be rewarded with consistent formatting and changes could easily
be applied to the whole document.
Deal with line or page breaking issues at the earliest when you go for your
final version.
Now that we've learned about the detailed formatting of text, we're ready to enter
the next chapter that deals with formatting and layout of whole pages and
documents.

Chapter 3. Designing Pages
After the previous chapter, formatting text should be easy for us. So, let's turn
to whole pages!
In this chapter, we will learn how to:
Adjust the margins
Change the line spacing
Section the document
Create a table of contents
Design headers and footers
Control page breaking
Set footnotes and modify their appearance
In learning this, we shall gain a deeper insight into classes and packages.

Defining the overall layout
We shall write an example document spanning over several pages. This will be
our test object for modifying margins, line spacing, headers, footers, and more.

Time for action – writing a book with
chapters
We will start to write a book. At first, we shall choose a class, further we will
use some filler text to work out the page layout.
1. Create a new document with the following code:
\documentclass[a4paper,12pt]{book}
\usepackage[english]{babel}
\usepackage{blindtext}
\begin{document}
\chapter{Exploring the page layout}
In this chapter we will study the layout of pages.
\section{Some filler text}
\blindtext
\section{A lot more filler text}
More dummy text will follow.
\subsection{Plenty of filler text}
\blindtext[10]
\end{document}

2. Save the document and Typeset it. Look at the first page:

What just happened?
We have chosen the document class book. As the name implies, this class is
suitable for book-like documents. Books are commonly two-sided and consist of
chapters which usually start at right-hand pages. They may have a front matter
with one or more title pages and a back matter with bibliography, index, and so
on. The book class supports all of this.
We loaded the babel package. This is useful especially for typesetting in other
languages than English regarding hyphenation, language-specific characters, and
more.
Even for English language, there are several options: USenglish, american,
english, UKenglish, british, canadian, australian, and newzealand.
Obviously, some mean the same, such as UKenglish and british. However,
there are differences in hyphenation rules between USenglish (american,
english) and UKenglish(british).
For now, we just need babel to load blindtext: this package has been developed
to produce filler text. It requires babel to detect the language of the document.
We stated the language English to babel, which means American English.
The command \chapter produced a large heading. This command will always
begin on a new page.
We've already seen the \section command. It's our second sectioning level and
generates a smaller heading than \chapter. It's automatically numbered per
chapter. The command \blindtext followed, printing some dummy text just to
fill the space with some text.
At last, we refined the sectioning with a \subsection command followed by
more dummy text to fill up the page.

Note
There's another popular package for generating dummy text. It's called lipsum and
it produces the famous LoremIpsum text which has been the typesetter's dummy
text for hundreds of years.

Reviewing LaTeX's default page layout
For the example in the current chapter, we used A4 paper and a font size of 12 pt.
We let LaTeX define the margins. Let's look at them:
The right margin is the outer margin of the book, because the chapter starts
on a right-hand page. It might seem to be a bit wide—but good text isn't just
intended to fill as much space as possible: it should be well-readable for our
eyes. Therefore the lines shouldn't be too long.
The left margin is the inner margin on such a right-hand page. A user asked
why the inner margin is smaller than the outer, even further some space will
get lost because of the binding. The explanation is simple, if we imagine the
book lies opened right before us: the inner margins would look joined. It's a
good idea to aim at equal margins—left, middle, and right. In this sense, we
could choose the inner margin to be half of the size of the outer margin—plus
some offset for the binding.
The bottom margin contains the page number.
The top margin looks very tall. It's caused by the chapter heading, there's
always more space left above. By looking at the second page, you will see
that normal pages have a smaller top margin.
If you ever doubt LaTeX's design, look at some books in your shelves and
compare, whether it's regarding margins, ligatures, numbering, or anything else.

Defining the margins yourself
A publisher or a supervisor may request you to follow his specifications. Besides
font size, interline spacing, and other style issues, this might also apply to the
margins. In this case, you would need to override LaTeX's recommendations
specifying the margins precisely. There's a package fulfilling these demands.

Time for action – specifying margins
We shall load the geometry package and state the exact width and height of all
margins.
1. Extend the preamble of the previous example with this command:
\usepackage[a4paper, inner=1.5cm, outer=3cm, top=2cm,
bottom=3cm, bindingoffset=1cm]{geometry}

2. Typeset and examine the margins:

What just happened?
We loaded another package with the name geometry. This package takes care of
our layout regarding the paper size, margins, and more dimensions. We chose A4
paper size, an outer margin of 3 cm, and an inner margin of just 1.5 cm,
remembering that the two inner margins will be perceived as one space when the
two-sided book is opened. We stated the top and the bottom margin. At last, we
specified a value of 1 cm for the binding correction. We need the inner margins to
be wider because we expect to lose this space later because of the binding like
gluing or stapling.

Using the geometry package
In the early days of LaTeX, it was common to manipulate the layout dimensions
directly. This approach had some disadvantages. We could easily make mistakes
in calculating the lengths, for instance left margin + right margin + text width
might not fit to the paper width. The geometry package comes to the rescue. It
provides a comfortable interface for specifying layout parameters. Further, it
provides auto-completion. It calculates missing values to match the paper size. It
even adds missing lengths using a heuristic approach to achieve a good layout.
Let's look at it in detail.
The geometry package understands arguments of the form "key=value", separated
by commas. If you load geometry without arguments, those arguments could
alternatively be used by calling \geometry{argument list}. We shall look at
some of them.

Choosing the paper size
Geometry provides several options to set the paper size and orientation:
states the paper name, for example, paper=a4paper. It may be
abbreviated like we did in our example. The package supports a lot of paper
sizes, such as letterpaper, executivepaper, legalpaper, a0paper, …
,a6paper, b0paper, … , b6paper, and more.
paper=name

paperwidth, paperheight allow you to choose
freely, like paperwidth=7in, paperheight=10in.
papersize={width,height}
papersize={7in,10in}. This
landscape
portrait

the paper dimensions

sets width and height of the paper like
is an example of a double-valued argument.

changes the paper orientation to landscape mode.

switches to portrait mode. This is the default.

Tip
If you already specified the paper name to the document class, geometry will
inherit it. That's valid in general: all document class options will be automatically
given to the packages that recognize them.

Specifying the text area

The text area may be adjusted by these options:
textwidth

sets the width of the text area, like textwidth=140mm.

textheight

states the height of the text area, like textheight=180mm.

lines gives another
like lines=25.

way to specify the text height by the number of lines,

includehead causes the header
area; set false by default.
includefoot causes
set false by default.

of the page to be included into the body

the footer of the page to be included into the body area;

Setting the margins
The size of the visible margins can be specified by these options:
left, right

set the width of the left and the right margin, like left=2cm.
Use it for one-sided documents.
inner, outer set the width of the inner
inner=2cm. Use it for two-sided documents.
top, bottom

and the outer margin, like

set the height of the top and the bottom margin, like top=25mm.

switches to two-sided mode. This means that left and right margins
would be swapped on left-hand pages, also called verso pages.
twoside

reserves space on the left margin (one-side), respectively
the inner margin (two-sided) for the binding.
bindingoffset

That's just a selection of commonly used options—there are many more. You
could choose and set some options intuitively—for instance,
\usepackage[margin=3cm]{geometry} will result in a 3 cm margin on each
edge of the paper and the paper size comes from the document class option.
The auto-completion works like this:
paperwidth

= left + width + right, where width=textwidth by default

paperheight

default

= top + height + bottom, where height=textheight by

If you decide to include marginal notes within the text body when calculating,
width could get wider than textwidth. If two dimensions of the right side of
each formula are given, the missing dimension would be calculated. That's why it
may be enough to specify left and right, top and bottom, respectively. Even if
just one margin is specified, the other dimensions would be determined using
default margin ratios:
top:bottom = 2:3
left:right = 1:1

for one-sided documents

inner:outer = 2:3

for two-sided documents

Sounds complicated? That's just intended to help you to achieve practical
dimensions even if some values were missing.

Note
In

2010,

version

5

of geometry introduced the commands
\newgeometry{argument list} and \restoregeometry, allowing users to
change margin dimensions in the middle of a document. Their auto-completion
may also differ a bit from previous versions.
The geometry package provides an extensive manual. Don't be worried by the
amount of documentation: it's an offer to guide you through the variety of features.

Obtaining package documentation
There are hundreds of LaTeX packages available. No book could ever explain all
of their features on its own. But most of those packages offer good documentation
that you can read for free. If you work your way through this book and supplement
it with the documentation of the mentioned packages, you're on the right track to
become a LaTeX power user.

Time for action – finding
geometry package manual

the

We would like to examine the documentation of the geometry package. At first,
we shall look for it locally on our computer. Afterwards, we will see how to find
the documentation online:
1. Click the Start button and choose All Programs | Accessories | Command
prompt
2. Type the following:
texdoc geometry

3. A PDF viewer should start and show the package manual. The same
command works at the command prompt on Linux/Unix computers.
4. Hit the Start button again; this time, choose All Programs | TeX Live 2010 |
TeXdoc GUI. Click on Search texdoctk's database, enter the keyword
geometry, and press Enter:

5. Have a look at the search result. Open it by double-clicking or hit the View
button. A viewer should open the documentation.
6. Open an Internet browser and visit the site http://ctan.org/pkg/geometry.
Scroll down to Documentation on CTAN and hit the link User's manual.
Again, your PDF viewer or a browser plug-in will show the documentation.

What just happened?
We've tried three ways of obtaining the documentation. At first, we used texdoc .
This is a tool to find and to view LaTeX documentation. It can be used at the
command prompt; just type texdoc followed by a keyword. For this, it's required
to be in the operating system's search path. The TeX distribution installer takes
care of it. The argument to texdoc doesn't have to be the package name, as
texdoc understands aliases. Further, we used a GUI for texdoc providing a
search feature supplemented by a category classification. Perhaps browse for a
little while through the categories.

Tip
I f texdoc doesn't help you, there's another opportunity: visit CTAN. You just
need to know the package name. Simply visit http://ctan.org/pkg/packagename to
get to the package homepage, there you may find documentation.
In case you don't know the package name or you just have a vague idea what
you're
looking
for,
visit
the
TeX
Catalogue
Online
on
http://texcatalogue.sarovar.org/bytopic.html. It lists hundreds of packages sorted
by topic.
Note: The online documentation might match a different package version than the
one on your computer.

Have a go hero – constructing the page layout with
typearea
Now that you know how to open package documentation, you could figure out if
you want to benefit from the typearea package. While the geometry package
allows you to control all margins yourself, typearea provides a different
approach. It supports you in creating a page layout based on proven typography
standards. typearea is explained within the KOMA-Script manual, so the
keywords for texdoc and CTAN would be koma-script or typearea. Reading
Chapter 2 of the manual on Page Layout could give you an insight into typography,
even if you decide not to use typearea.

Changing the line spacing
Without some vertical space between the lines, the readability of a text could
suffer. Adding such space would help lead the eye along the line.

Time for action – increasing line
spacing
Though LaTeX already takes care of good readability by choosing a meaningful
interline spacing, publishers might require different spacing. We shall modify our
example by adding half of a line height to the line spacing.
1. Extend the preamble of our example with this command:
\usepackage[onehalfspacing]{setspace}

2. Typeset to see the change:

What just happened?
We loaded the setspace package. Its only purpose is to adjust the line spacing.
We provided the option onehalfspacing . This increases the spacing by half of
a line height for the whole document.
setspace

understands three options:

is the default. No additional space will be inserted. The
text will be typeset with LaTeX's default interline spacing, which is about 20
percent of the line height.
singlespacing

onehalfspacing

means one-and-a-half spacing like in our example.

can be used for even more spacing: the distance between
the baselines of successive text lines would be twice as high as a single line.
doublespacing

Commonly, we choose the spacing for the complete document. In that case, such
an option would be sufficient. For each of the options, there's a corresponding
environment. They are called singlespace , onehalfspace , and doublespace
. Those environments may be used to adjust the spacing within the document. They
shouldn't be nested. For the need of a different stretching factor, the spacing
environment is provided. It may be used like the following:
\begin{spacing}{2.4}
This text is stretched by a factor of 2.4.
\end{spacing}

Note
In typesetter's jargon, the distance between the baselines of consecutive text lines
is called leading.

Have a go hero – examining a package source file
In some cases the documentation of a package is not easily accessible. This
applies to setspace: the explanation of the options, commands, and environments
is contained in the package source file itself. How can we obtain it? We could
visit CTAN, as we described before, or locate the file on our computer. A
package filename usually ends with .sty, standing for style file . TeX Live
provides a tool for locating such source files. Try it: open the command prompt,
that is, hit the Start button and click All Programs | Accessories | Command
prompt. Then type the following:
kpsewhich setspace.sty

And press Enter. The program will print the location of the file, as follows:
c:/texlive/2010/texmf-dist/tex/latex/setspace/setspace.sty

Open the file using TeXworks or any text editor. There are a lot of comment lines
explaining the package features.

Note
The kpsewhich tool belongs to the kpathsea library.
purpose is filename lookup.

kpathsea's

fundamental

Using class options to configure the
document style
We already know that a document class is the basis of our document. It provides
commands and environments extending the LaTeX standard features. Though the
class provides a default style, it's still customizable with options.

Time for action – creating a twocolumn landscape document
We shall change the orientation of our example to landscape. We would also like
to typeset our text in two columns:
1. Add the options landscape and twocolumn to the \documentclass
statement of our example, as follows:
\documentclass[a4paper,12pt,landscape,twocolumn]{book}

2. Typeset and see how the layout has changed:

What just happened?
By using the landscape option, we switched the orientation. By stating the
twocolumn option, we caused the body text to be divided into two columns.

Note
There's a command \twocolumn[opening text] starting a two-column page
with optional opening text over the whole width. \onecolumn begins a onecolumn page.
If you'd like to balance the columns on the last page or if you wish to have more
than two columns, use the multicols package.
The LaTeX base classes are article, book, report, slides, and letter . As the
name suggests, the last one can be used to write letters, though there are further
suitable classes like scrlttr2. slides can be used to create presentations, but
today there are more powerful and feature rich classes like beamer and
powerdot .
Let's sum up the options of the base classes:
a4paper,
a5paper,
executivepaper: The

b5paper,

letterpaper,

legalpaper,

or
output will be formatted according to this paper size,
that is, A4: 210 mm x 297 mm. letterpaper (8.5 in x 11 in) is the default.
Loading the geometry package allows more sizes.
10pt, 11pt, or 12pt: The size of normal text in the document; the default is
ten points (10pt). The size of headings, footnotes, indexes, and so on, will

be adjusted accordingly.
landscape:

Switches to landscape format; width and height of the output
size will be interchanged.
or twocolumn: Decides whether the pages will be one-column
(default) or two-column. It's not supported by the letter class.
onecolumn

oneside or twoside: Formatting for printing on one
page. oneside is the default, except for the book
usable with the slides class and the letter class.

side or both sides of a
class. twoside is not

or openany: The first decides that chapters have to begin on a
right-hand page (the default for the book class), openany allows chapters to
start on any page (default for the report class). These options are only
supported by the book and the report class because the other classes don't
openright

provide chapters.
titlepage or notitlepage: The first causes a separate title page when
\maketitle is used and is the default, except for the article class. The
default of article is notitlepage, which means that normal text may

follow the title on the same page.
or draft: If draft is set, then LaTeX will mark overfull lines with a
black box, which is helpful in reviewing and improving the output. Some
packages support these option as well, behaving differently then, such as
omitting embedding of graphics and listings when draft has been chosen.
final is the default.
final

openbib:

When this option is set, a bibliography would be formatted in open
style instead of compressed style.
fleqn:

Causes displayed formulas to be left-aligned.

leqno:

For numbered displayed formulas, the number would be put to the
left side. The right side is the default.

Note
Many other classes support these options as well and even more. For the need of
an uncommon base font size, the classes extarticle, extbook, extreport, and
extletter provide base font sizes from 8 pts to 20 pts. The so called KOMAScript classes allow arbitrary base font sizes. They understand such a large
amount of options that they're additionally supporting a "key=value" interface like
we've seen with the geometry package.

Have a go hero – trying KOMA-Script classes
Have you been so brave to check out the typearea package documentation? Then
you've already seen the KOMA-Script manual accessible by texdoc komascript. Open it again; this time, read it with the idea of choosing KOMA classes
for your documents. They may be used like the base classes: for each base class,
there's a corresponding KOMA class. They differ in some meaningful presets; for
instance, headings use a sans-serif font by default. However, the presets can
easily be changed. Those classes extend the base classes very much, providing a
huge amount of commands. Web forums frequently show user questions being
hard to solve using base classes, but being easily achievable by a single KOMA
class command.

Creating a table of contents
A book commonly begins with a table of contents, so let's create one.

Time for action – adding a table of
contents
We shall create a table of contents based on our numbered headings.
1. Let's remove the options landscape and twocolumn. We also don't load the
setspace package. Add the command \tableofcontents right after
\begin{document}. Our code shall now be:
\documentclass[a4paper,12pt]{book}
\usepackage[english]{babel}
\usepackage{blindtext}
\usepackage[a4paper, inner=1.5cm, outer=3cm, top=2cm,
bottom=3cm, bindingoffset=1cm]{geometry}
\begin{document}
\tableofcontents
\chapter{Exploring the page layout}
In this chapter we will study the layout of pages.
\section{Some filler text}
\blindtext
\section{A lot more filler text}
More dummy text will follow.
\subsection{Plenty of filler text}
\blindtext[10]
\end{document}

2. Typeset twice. Afterwards, the first page of your output will contain this
table:

What just happened?
The command \tableofcontents tells LaTeX to produce a file with the
extension .toc. This file will be used to generate a table of contents. We had to
typeset twice: in the first run, the .toc file was written and in the second run,
LaTeX read it and processed it.
The entries are created by the sectioning commands. We used
\section, and \subsection, and we've got an entry for each.

\chapter,

Sectioning and the contents
A heading might be very long; it could span over two or more lines. In that case,
we might wish to shorten its corresponding table of contents entry. Let's see how.

Time for action – shortening the table
of content entries
We will use the optional argument of the sectioning commands to produce shorter
entries, different to the actual headings:
1. Take our example and modify the sectioning commands as shown in the
highlighted lines:
\documentclass[a4paper,12pt]{book}
\usepackage[english]{babel}
\usepackage{blindtext}
\usepackage[a4paper, inner=1.5cm, outer=3cm, top=2cm,
bottom=3cm, bindingoffset=1cm]{geometry}
\begin{document}
\tableofcontents
\chapter[Page layout]{Exploring the page layout}
In this chapter we will study the layout of pages.
\section[Filler text]{Some filler text}
\blindtext
\section[More]{A lot more filler text}
More blindtext will follow.
\subsection[Plenty]{Plenty of filler text}
\blindtext[10]
\end{document}

2. Typeset twice and look at the modified table of contents:

What just happened?
Besides the mandatory argument producing the heading, each sectioning command
understands an optional argument. If this is given, it will be used instead of the
mandatory heading for the table of contents entry.
Again, we had to typeset twice to notice a change. This is not unusual. You will
encounter more situations when several LaTeX runs would be necessary, like
creating an index or a bibliography.
In Chapter 7, Listing Content and References, we shall take a further look and
learn how to customize the table of contents. Let's look again at the sectioning
commands of book, report, and article. There are seven levels in those base
classes:
\part:

For dividing the document in major units. The numbering of other
sectional units is independent of \part. A part heading will use a whole
page in book and report documents.
\chapter: A large heading
book and report classes.

that will start at a new page, available in the

\section, \subsection, \subsubsection:

Bold headings available in all

classes, the following text comes below.
\paragraph, \subparagraph:

Available in all classes, produces a run-in

heading.
Except \part, all sectioning commands reset the counter of the section that's one
level below in the hierarchy. For instance, \chapter resets the section counter.
This way, the sections will be numbered per chapter.
To sum up, though these commands are easy to use, they do a lot, such as:
Cause a page break (\part and \chapter with book and report classes)
Generate a number and a presentation for it, some depending on the higherlevel counters (like 2.1)
Reset the counter of the next-level sectional unit (except \part)
Produce a table of contents entry storing it in the .toc file
Format the heading, usually bold-faced and the larger the higher in the
hierarchy

If needed, save the heading for use in a page header
All sectioning commands provide a starred form, as follows:
\section*{title}

If you use this form, the numbering will be suppressed and there won't be an entry
in the table of contents or in a header. Look at the heading Contents in our
example; this has actually been typeset by \chapter* inside the
\tableofcontents macro.
As page headers have been mentioned, let's explore them now.

Designing headers and footers
Already during testing of the first version of our example, you might have noticed
that except for the page where the chapter started, all pages already showed the
page number, chapter title, and section title in their header:

Though these standard headers are already quite useful, we shall see how to
customize them to meet our individual requirements.

Time for action – customizing headers
with the fancyhdr package
The default shape of the page headings is slanted. Furthermore, they are written in
capital letters. We shall use bold typeface instead and we will use capital letters
only for the chapter title. We will load the fancyhdr package and use its
commands to achieve that:
1. Load the first version of our example. Insert the highlighted lines:
\documentclass[a4paper,12pt]{book}
\usepackage[english]{babel}
\usepackage{blindtext}
\usepackage{fancyhdr}
\fancyhf{}
\fancyhead[LE]{\leftmark}
\fancyhead[RO]{\nouppercase{\rightmark}}
\fancyfoot[LE,RO]{\thepage}
\pagestyle{fancy}
\begin{document}
\chapter{Exploring the page layout}
In this chapter we will study the layout of pages.
\section{Some filler text}
\blindtext
\section{A lot more filler text}
More dummy text will follow.
\subsection{Plenty of filler text}
\blindtext[10]
\end{document}

2. Typeset—the footers will contain the page number on their outer side and
the headers should look like the following:

What just happened?
We loaded the fancyhdr package. Our first action was calling \fancyhf{} ; this
command clears the headers and footers. Further, we used the following:
\leftmark:

Used by the book class to store the chapter title together with
the chapter number. Capital letters are used as default.
\rightmark:

Used by the book class to store the section title together with
its number. Capital letters are used as well.
We used the command \fancyhead with the optional argument LE to put the
chapter title into the header. LE stands for left-even and means that this chapter
title will be put on the left side of the header on even-numbered pages.
Analogous, we called the command \fancyhead with RO to put the section title
into the header. RO stands for right-odd and means that this section heading shall
be displayed on right side of the header on odd-numbered pages.
Afterwards, we used \fancyfoot to display the page number in the footer. This
time, we used LE and RO that showed the page number on even as well as on odd
pages, always on the outer side. The command \thepage prints the page number.
All those commands are used to modify a page style provided by fancyhdr; this
style is called fancy . We had to tell LaTeX to use this style and we did it through
\pagestyle{fancy}.

Note
Emphasizing by writing all letters capitalized, like fancyhdr does by default, is
called all caps . It is widely regarded as a questionable style. If it's used at all,
then it's recommended to slightly increase the spacing between the capital letters.
For adjusting this so called letterspacing , you can use the microtype package
or the soul package.

Understanding page styles
LaTeX and its base classes provide four page styles:
empty:

Neither a header nor a footer is shown.

plain:

No header. The page number will be printed and centered in the

footer.
headings:

The header contains titles of chapters, sections, and/or
subsections, depending on the class and also the page number. The footer is
empty.
myheadings:

The header contains a user-defined text and the page number;
the footer is empty.
fancyhdr

adds one page style:

fancy:

Both the header and footer may be customized by the user.

Two commands may be used to choose the page style:
\pagestyle{name}:

Switches to the page style name from this point

onwards.
\thispagestyle{name}:

Chooses the page style name only or the current
page; the following pages will have the style that's been used before.
You have seen that where a chapter starts, the page style is different to the style of
other pages. Such pages will have plain style. If you thought all pages should use
the same style, look into some books: it's very common that chapter beginnings
differ in style. They usually have a blank header. \thispagestyle could be used
to override that.

Customizing header and footer
Let's divide the header and footer into six pieces: left, center, and right (l, c, r)
both for the header and footer. The commands to modify those areas are as
follows:
For the header: \lhead, \chead, \rhead
For the footer: \lfoot, \cfoot, \rfoot
Each of these commands requires a mandatory argument, like \chead{User's
guide} or \cfoot{\thepage}. This argument will be put into the corresponding
area of the page.
Alternatively, you could use these versatile commands:
For the header: \fancyhead[code]{text}
For the footer: \fancyfoot[code]{text}
Here, code may consist of one or more letters:
Symbol Meaning
L

left

C

center

R

right

E

even page

O

odd page

H

header

F

footer

It doesn't matter if we choose uppercase or lowercase letters. We already used
such combinations in our example.

Using decorative lines in header or footer
We can introduce or delete lines between header and body text and body text and
footer, respectively, with these two commands:
\renewcommand{\headrulewidth}{width}
\renewcommand{\footrulewidth}{width}

Where width may be a value like 1pt, 0.5mm, and so on. Just set it to 0pt if you
don't like such a line. The default is 0.4pt for the header line and 0pt for the
footer line.
works exactly like \newcommand, except that the command must
already exist. It will get the new meaning. Incidentally, we've got to know a new
concept: a lot of LaTeX commands may be redefined in this way. We shall take
advantage of it from now on.
\renewcommand

Tip
There are starred versions-\newcommand, and \renewcommand, respectively.
They are called short commands. Their arguments must not contain paragraph
breaks. This limitation eases troubleshooting, as TeX would be able to pick up an
error earlier. That's why many people use them whenever possible.

Changing LaTeX's header marks
As we already know, LaTeX classes and packages store sectioning numbers and
headings in the macros \leftmark and \rightmark automatically. It will be
done when we call \chapter, \section, or \subsection. So, we could just use
\leftmark and \rightmark in the arguments of the fancyhdr commands.
We will sometimes want to change those entries manually, even if we rely on this
automatism. For instance, the starred sectioning commands like \chapter* and
\section* won't produce a header entry, like indicated earlier. In such a case,
two commands will help us:
\markright{right head}

sets the right heading

\markboth{left head}{right head}

sets both left and right heading

The default style headings is easy-to-use and gives good results. myheadings
can be used together with \markright and \markboth. However, the most
flexible way is given by fancy, especially in combination with \markright and
\markboth.

Note
There's a very good alternative to fancyhdr, which is a package called
scrpage2. It belongs to KOMA-Script, but works with other classes as well. It
provides a similar functionality and offers even more features.

Breaking pages
As you've seen in our example, LaTeX took care of the page breaking. There
might be occasions where we'd like to insert a page break ourselves. LaTeX
offers several commands to do it; with vertical balance or without.

Time for action – inserting page
breaks
We will go back to the first version of our example and we shall manually insert a
page break right before the subsection 1.2.1.
1. Insert the highlighted line into our example:
\documentclass[a4paper,12pt]{book}
\usepackage[english]{babel}
\usepackage{blindtext}
\begin{document}
\chapter{Exploring the page layout}
In this chapter we will study the layout of pages.
\section{Some filler text}
\blindtext
\section{A lot more filler text}
More dummy text will follow.
\pagebreak
\subsection{Plenty of filler text}
\blindtext[10]
\end{document}

2. Typeset and have a look:

3. Replace \pagebreak by \newpage.
4. Typeset and compare:

What just happened?
At first, we inserted the command \pagebreak. Like its name suggests, it causes
a page break. Furthermore, the text has been stretched to fill the page down to the
bottom.
Afterwards, because of the obviously unpleasant whitespace between the
paragraphs and the headings, we replaced \pagebreak with \newpage. This
command breaks the page as well, but it doesn't stretch the text: the remaining
space of the page will stay empty.
So, \pagebreak behaves like \linebreak and \newpage works like \newline,
for pages instead of lines. There's even a command \nopagebreak that's
analogous to \nolinebreak and forbids page breaking.
won't break a line. Furthermore, \nopagebreak doesn't refer to the
middle of a line: both commands apply at the end of the current line. Of course,
they immediately have effect when used between paragraphs.
\pagebreak

If you use the two-column format, both \pagebreak and \newpage would begin
on a new column instead of a new page.

Note
There are two further variants: \clearpage works like \newpage, except that it
will start on a new page, even in two-column mode. \cleardoublepage does the
same, but causes the following text to start on a right-hand page, inserting a blank
page if necessary. The latter is useful for two-sided documents.
More importantly, both commands cause all figures and tables that LaTeX has in
its memory to be printed out immediately.
Sometimes you may notice large gaps between headings and paragraphs. Those
could be the result of a bad page break like in our example. In that case, you could
assist LaTeX by such a page break command or you switch off vertical
justification by the command \raggedbottom. Its counterpart \flushbottom will
switch it on, which is the default in the book class.
What we learned about line breaking is valid for page breaking as well: don't
worry about it while your work is still in progress. Only when you're done with
your document and you're preparing the final version, will it be time to tweak the
page breaks.

Have a go hero – exploring page breaking options
Remember

the optional arguments for \linebreak and \nolinebreak.
\pagebreak and \nopagebreak are able to take an optional argument as well, an
integer between 0 and 4. While 0 recommends, 4 demands. Again, try the
previous example. This time use an optional argument for \pagebreak. Test with
0, 1, 2, 3, 4. Only \pagebreak[4] will cause the page break because the
necessary stretching is so much that LaTeX doesn't like to follow either a weak or
a strong request.

Enlarging a page
LaTeX's rules are not set in stone; they also apply to page balancing. There may
be occasions where we want to put a little more text onto a page, even if the text
would be squeezed a bit or the text height increases. There's a command that will
help us out: \enlargethispage.

Time for action – sparing an almost
empty page
We shall modify our example a bit. This time, we will try to avoid a nearly empty
page by squeezing their text to the preceding page.
1. Remove the command \newpage from our example and switch to 11pt base
font. This time, use less filler text in the subsection:
\documentclass[a4paper,11pt]{book}
\usepackage[english]{babel}
\usepackage{blindtext}
\usepackage[a4paper, inner=1.5cm, outer=3cm, top=2cm,
bottom=3cm, bindingoffset=1cm]{geometry}
\begin{document}
\chapter{Exploring the page layout}
In this chapter we will study the layout of pages.
\section{Some filler text}
\blindtext
\section{A lot more filler text}
More dummy text will follow.
\subsection{Plenty of filler text}
\blindtext[3]
\end{document}

2. Typeset, and the result will consist of two pages:

3. Insert this command right after the \subsection line:
\enlargethispage{\baselineskip}

4. Typeset. Now our document fits in only one page:

What just happened?
We used the command \enlargethispage to squeeze more text onto a page. This
command takes the additionally requested height as its argument. The command
\baselineskip returns the height of a text line that we used as the argument. So,
LaTeX could put one extra line onto the page, and even the remaining line fitted in
as well because LaTeX compressed some whitespace.
We could use factors: write \enlargethispage{2\baselineskip} to get two
more lines on a page. It doesn't even need to be an integer value. Like always,
when you state a length, you could use other units such as 10pt, 0.5in, 1cm, or
5mm, and even negative values.
Only the current page will be affected by this command.

Note
There's a starred version: \enlargethispage* would additionally shrink all
vertical spaces on the page to their minimum.

Using footnotes
As briefly mentioned in Chapter 2, Formatting Words, Lines, and Paragraphs ,
LaTeX provides a command to typeset footnotes. Let's see it in action.

Time for action – using footnotes in
text and in headings
Let's go back to the very first example of this chapter. We shall insert one footnote
in the body text and one in a section heading:
1. Modify the example inserting a footnote, as shown in the highlighted line:
\documentclass[a4paper,12pt]{book}
\usepackage[english]{babel}
\usepackage{blindtext}
\begin{document}
\chapter{Exploring the page layout}
In this chapter we will study the layout of pages.
\section{Some filler text}
\blindtext
\section{A lot more filler text}
More dummy text\footnote{serving as a placeholder} will
follow.
\subsection{Plenty of filler text}
\blindtext[10]
\end{document}

2. Typeset to see how the footnote looks in print:

3. Insert a footnote in the second section header. This time, put \protect right
before:
\section{A lot more filler text\protect\footnote{to fill the
page}}

4. Typeset to check that it works:

What just happened?
The command \footnote{text} placed a superscripted number at the current
position. Further, it prints its argument text into the bottom of the page, marked
by the same number. As we've seen, such notes are separated from the main text
by a horizontal line.
The other \footnote command has been preceded by the command \protect. If
we omitted that, an error would be raised—try it. Such an error may occur when
we use \footnote inside an argument, where it's being further processed.
\protect simply prevents this processing error. As a rule of thumb, if a
command causes an error when it's used inside an argument, like in headings, try
to fix it by putting \protect right before that command. Cases where \protect
would hurt instead of helping are rare.
Generally, a heading might not be a good place for footnotes. If you simply use
\footnote in headings, be aware that they would appear in the table of contents
and perhaps also in page headers. If you want to avoid that, use the optional
parameter of the sectioning commands without \footnote, that is,
\section[title without footnote]{title with footnote}. The same
applies to \chapter, and so on, and even further to \caption.
The complete definition of \footnote is:
\footnote[number]{text} produces
number, an integer. If we don't give the

a footnote marked by this optional
optional number, an internal counter
would be stepped and used. This would be done automatically; we don't
need to worry.
Nevertheless, if we encounter difficulties in placing footnotes, there are two
commands that help us:
produces a superscripted number in the text as a
footnote mark. If the optional argument wasn't given, it's also stepping and
using the internal footnote counter. No footnote will be generated.
\footnotemark[number]

generates a footnote without putting a
footnote mark into the text without stepping the internal footnote counter.
\footnotetext[number]{text}

Both the commands may be used together in circumstances when \footnote
would fail.
Set a footnote command right after the related text. Don't leave a space in-

between. Otherwise, you would get a gap between the text and the following
footnote mark.

Modifying the dividing line
The line that separates footnotes from the text is produced by the command
\footnoterule. If we wish to omit that line or if we want to modify it, we must
redefine it. We learned about \renewcommand—so, let's use it.

Time for action – redefining the
footnote line
We will use \renewcommand to override the default \footnoterule command:
1. Take the previous example and add the following lines to the preamble:
\renewcommand{\footnoterule}
{\noindent\smash{\rule[3pt]{\textwidth}{0.4pt}}}

2. Typeset and see how the line has changed:

What just happened?
Remember the definition of \newcommand—here it's analogous. The existing
command \footnoterule will be replaced by a new definition that we wrote in
the second line. The command \rule[raising]{width}{height} draws a line,
here 0.4 pt thick, and as wide as the text, raised a bit by 3 pt. Through the
command \smash, we let our line pretend to have a height and a depth of zero, so
it's occupying no vertical space at all. This way, the page balancing will not be
affected. You already know \noindent, which avoids the paragraph indentation.
If you want to omit that line completely, you just need to write:
\renewcommand{\footnoterule}{}

Now the command is defined to do nothing.

Using packages to expand footnote styles
There are different habits for setting footnotes. Some styles require footnotes
numbered per page, they might have to be placed in the document as so called
endnotes, and symbols instead of numbers may be used. More demands exist and
therefore several packages have been developed to comply with them. Here's a
selection:
endnotes: Places footnotes at the end of the document
manyfoot: Allows nested footnotes
bigfoot: Replaces and extends manyfoot, improves page break handling
with footnotes
savefnmark: Useful when you need to use footnotes several times
footmisc: All-round package; introduces numbering per page, is able to save
space when many short footnotes are used, offers symbols instead of
numbers as footnote marks, provides hanging indentation and other styles
Have a look at the respective package documentation to learn more—now you
know how to obtain it.

Pop quiz
1. According to the text, which of the following points should be the goal when
designing the page layout?
a. Fill the page as much as possible
b. Achieve an artistic design
c. Improve readability
d. Distinguish the work from other publications
2. Which options are preset by the book class?
a. openright and twoside
b. openany and twoside
c. openright and oneside
d. openany and oneside
3. Which page style is used as default on the first page of a chapter?
a. empty
b. plain
c. headings
d. myheadings
e. fancy
4. One of these page breaking commands is able to stretch the text to the bottom
to fill the page. Which one?
a. \newpage
b. \clearpage
c. \cleardoublepage
d. \pagebreak
5. We would like to put a footnote into a chapter heading. Choose the best way!
a. \chapter{title\footnote{text}}
b. \chapter*{title\footnote{text}}
c. \chapter{title\protect\footnote{text}}
d. \chapter[title]{title\footnote{text}}

Summary
In this chapter, we have worked out how to design the overall layout of a
document.
Specifically, we learned about:
Adjusting the page dimensions and margins
Using landscape orientation and two-column layout
Modifying the spacing between lines
Breaking pages and paying attention to the vertical justification
Designing headers and footers with titles and page numbers
Using footnotes
Further, we covered some general topics:
Obtaining documentation of packages in several ways
Changing document properties by choosing class options
Redefining existing commands
Now it's time to deal with further text structures. In the next chapter, we shall
learn how to create lists for presenting text in an easy-to-read way.

Chapter 4. Creating Lists
Arranging text in the form of a list can be very reader-friendly. You can present
several ideas by a clear structure which is easy to survey. Commonly, three
types of lists are used:
1. Bulleted lists
2. Numbered lists
3. Definition lists
In this chapter, we shall learn how to create such lists.

Building a bulleted list
We shall start with the simplest kind of list. It contains just the items without
numbers. Each item is marked by a bullet.

Time for action – listing LaTeX
packages
We shall create a list of packages that we got to know in the previous chapter:
1. Begin a new document and enter the following code:
\documentclass{article}
\begin{document}
\section*{Useful packages}
LaTeX provides several packages for designing the layout:
\begin{itemize}
\item geometry
\item typearea
\item fancyhdr
\item scrpage2
\item setspace
\end{itemize}
\end{document}

2. Typeset and have a look at the output:

What just happened?
It was not necessary to load any package. We began with a heading followed by
some text. For the actual list, we used an environment called itemize. As we
know about environments, \begin{itemize} starts it and \end{itemize} ends
it. The command \item tells LaTeX that a new item to the list follows. Each item
may contain text of any length and even paragraph breaks. Well, that's pretty easy,
isn't it?

Nesting lists
When a list gets longer, we could make it clearer by dividing it. We just create
lists in a list. It's advisable to use different bullets to keep the list levels apart
easily. LaTeX does it for us automatically.

Time for action – listing packages by
topic
We shall refine the package list by introducing categories:
1. Refine the aforementioned highlighted itemize environment of our example
in the following way:
\begin{itemize}
\item Page layout
\begin{itemize}
\item geometry
\item typearea
\end{itemize}
\item Headers and footers
\begin{itemize}
\item fancyhdr
\item scrpage2
\end{itemize}
\item Line spacing
\begin{itemize}
\item setspace
\end{itemize}
\end{itemize}

2. Typeset to see the new list:

What just happened?
We simply inserted another list inside the list. Up to four levels are possible,
otherwise LaTeX would stop and print out the error message ! LaTeX Error:
Too deeply nested. As we saw, the first level is marked by a bullet, and the
second by an en dash. A third level item would start with an asterisk symbol *.
The fourth and last level would be marked by a centered dot.
Deeply nested lists can rarely be seen; such complicated structures might be hard
to read. In such cases, it could be a good idea to revise the text structure or at
least split the list.

Creating a numbered list
Bulleted lists are useful if the order of the items doesn't matter. However, if the
order is important, we could organize the items by giving them numbers and
creating a sorted list. That would allow a reader to follow our thoughts easily.

Time for action – writing a step-bystep tutorial
Let's prepare a tiny "how-to" about designing the page layout:
1. Start with a new document and enter the following code:
\documentclass{article}
\begin{document}
\begin{enumerate}
\item State the paper size by an option to the document
class
\item Determine the margin dimensions using one of these
packages:
\begin{itemize}
\item geometry
\item typearea
\end{itemize}
\item Customize header and footer by one of these
packages:
\begin{itemize}
\item fancyhdr
\item scrpage2
\end{itemize}
\item Adjust the line spacing for the whole document
\begin{itemize}
\item by using the setspace package
\item or by the command \linespread{factor}
\end{itemize}
\end{enumerate}
\end{document}

2. Typeset to generate the «how-to»:

What just happened?
We used an enumerate environment. Except for the name, we use it just like the
itemize environment; each list item is introduced by the command \item. Again,
we nested two lists, but this time the lists are of a different kind. As we saw,
that's no problem. Even though the unnumbered list is embedded within a
numbered list, it's marked by bullets, because it's the first unnumbered level.
Mixed nesting could go further than four levels, but four is the maximum for each
kind and six in general.
The default numbering scheme for the enumerate environment is as follows:

understands an optional argument; if you write \item[text], then text
will be printed instead of a number or a bullet. This way, you could use any
numbering and any symbol for the bullet.
\item

Customizing lists
The default appearance of lists is meaningful regarding spacing, indentation, and
symbols. Nevertheless, it may be required to use another scheme for the
enumeration, for the bullets, or to modify the line spacing or their indentation.
There are packages helping us both to save space and to customize the symbols;
let's start with the spacing.

Saving space with compact lists
A frequently arising question is how to reduce the space. LaTeX's lists are often
regarded as being too spacious. We shall see how to implement that.

Time for action – shrinking our
tutorial
We shall remove the white space around the list items and before and after the
whole list as well:
1. In the previous example, load the package paralist and replace enumerate
with compactenum and itemize with compactitem:
\documentclass{article}
\usepackage{paralist}
\begin{document}
\begin{compactenum}
\item State the paper size by an option to the document
class
\item Determine the margin dimensions using one of these
packages:
\begin{compactitem}
\item geometry
\item typearea
\end{compactitem}
\item Customize header and footer by one of these
packages:
\begin{compactitem}
\item fancyhdr
\item scrpage2
\end{compactitem}
\item Adjust the line spacing for the whole document
\begin{compactitem}
\item by using the setspace package
\item or by the command \linespread{factor}
\end{compactitem}
\end{compactenum}
\end{document}

2. Typeset and compare the spacing:

3. Now extend the highlighted list item for setspace as follows:
\item by using the setspace package and one of its options:
\begin{inparaenum}
\item singlespacing
\item onehalfspacing
\item double spacing
\end{inparaenum}

4. Typeset and look at the change in the line spacing subject:

What just happened?
The used package paralist provides several new list environments designed to
be typeset within paragraphs or in a very compact look. We loaded this package
and replaced the standard environments with their compact counterparts. In step
3, we used a new environment, where the items are enumerated but stay within the
same paragraph.
For each standard
environments:

environment, paralist

adds

three

corresponding

Numbered lists:
compactenum:

Compact version of the enumerate environment without any
vertical space before or after the list or its items
inparaenum:

An enumerated list typeset within a paragraph

asparaenum:

Every list item is formatted like a separate common LaTeX
paragraph, but numbered
Bulleted lists:
compactitem:
compactenum
inparaitem:

Compact

version

of

the itemize environment like

An itemized list typeset within a paragraph, rarely seen in

print
asparaitem:

Like asparaenum, but with symbols instead of numbers

Choosing bullets and numbering format
To follow language specific habits or certain requirements, we might wish to
enumerate by Roman numbers or alphabetically; parentheses or dots might be
required. Some may prefer dashes instead of bullets. The package enumitem
provides sophisticated features to implement such requirements.

Time for action – modifying lists
using enumitem
Let's change the numbering scheme. We shall number alphabetically using circled
letters. Furthermore, we will replace bullets by dashes:
1. We shall discard paralist and load the package enumitem instead. We
will turn away from the compact environments returning to the standard list
notation:
\documentclass{article}
\usepackage{enumitem}
\setlist{nolistsep}
\setitemize[1]{label=---}
\setenumerate[1]{label=\textcircled{\scriptsize\Alph*},
font=\sffamily}
\begin{document}
\begin{enumerate}
\item State the paper size by an option to the document
class
\item Determine the margin dimensions using one of these
packages:
\begin{itemize}
\item geometry
\item typearea
\end{itemize}
\item Customize header and footer by one of these
packages:
\begin{itemize}
\item fancyhdr
\item scrpage2
\end{itemize}
\item Adjust the line spacing for the whole document
\begin{itemize}
\item by using the setspace package
\item or by the command \linespread{factor}|
\end{itemize}
\end{enumerate}
\end{document}

2. Typeset and see the output:

3. Right above the highlighted line, insert the following lines:
\end{enumerate}
\subsubsection*{Tweaking the line spacing:}
\begin{enumerate}[resume*]

4. Typeset to see the change:

What just happened?
We used the
look:

enumitem

command to specify list properties. Let's take a closer

\setlist{nolistsep}: \setlistsets properties valid for all types of lists.
Here we specified nolistsep to achieve very compact lists analogous to
the compact paralist environment.
\setitemize[1]{label=–--}:

modifies properties of
bulleted lists. Here we chose an em dash as the label to get a leading wide
dash.
\setitemize

\setenumerate[1]
{label=\textcircled{\scriptsize\Alph*},font=\sffamily}:
\setenumerate sets properties valid for numbered lists. We used it to set a
label and a font for the label. The command \Alph* stands for enumeration

in capital letters.
All of those three commands allow arguments of the form key=value. Some
useful parameters are as follows:
Parameter
font

Meaning
Modifies the label font

Values
Any font command

Example
font=\bfseries

May contain
label

Sets the label for the current level

\arabic*,

label=\emph\alph*)

\alph*, \Alph*, \roman*,
\Roman*
label*

Like label but appended to the current label Like label

label*=\arabic)

align

Alignment of the label

left

align=right

start

Number of the first item

Integer

resume,

Let's the counter continue from the previous
lists value

resume

noitemsep

No extra space between items and
paragraphs

noitemsep

nolistsep

No extra vertical spacing at all

nolistsep

resume*

or right

start=10

These options may be set globally like \setenumerate[level]{key=value

list}.

If the optional argument level is missing, this global command will apply
to all levels.
Those options may be used locally like we did with resume*. Other examples
are:
\begin{itemize}[noitemsep]

for a compact bulleted list

\begin{enumerate}[label=\Roman*.,start=3]

numbered by III., IV., ...

\begin{enumerate}[label=\alph*)],nolistsep]

for a very compact

list numbered a), b), c), ...
The labeling commands would achieve a numbering as follows:
\arabic*

1, 2, 3, 4, …

\alph*

a, b, c, d, …

\Alph*

A, B, C, D, …

\roman*

i, ii, iii, iv, …

\Roman*

I, II, III, IV, …

The * has been added to distinguish those commands from LaTeX commands with
the same name. Parentheses and punctuation may be used as wished. Later in the
book, you will learn how to choose between thousands of symbols for labels and
bullets.
There's even a short form: if you load enumitem with the option shortlabels,
you may use a compact syntax like \begin{enumerate}[(i)],
\begin{enumerate}[(1)] where 1, a, A, i, I stand for \arabic*, \alph*,
\Alph*, \roman*, \Roman* respectively. This allows customization quickly and
easily. However, consider using global commands to keep formatting consistent.

Suspending and continuing lists
In step 3 of our example, we interrupted the list. We continued writing normal text
until we restarted the list by \begin{enumerate}[resume*]. The resume option
tells enumitem to continue the list with the next number. The starred variant
resume* does it with the same formatting like before. If you would like to
complete the document, note that a \subsubsection command should be
preceded by a \subsection command. This should follow a \section
command. Just using a lower-level heading without a parent heading could result
in numbering like 0.0.1.

Producing a definition list
We shall proceed to the third kind of list, namely, definition lists, also called
description lists. Every list item consists of a term followed by its description.

Time for action –
capabilities of packages

explaining

Now that we know some packages for creating lists, we shall write a short
overview to show their capabilities. Let's choose some additional packages listed
on http://texcatalogue.sarovar.org:
1. We will use a description environment. Create a document with the
following code:
\documentclass{article}
\begin{document}
\begin{description}
\item[paralist] provides compact lists and list versions
that can
be used within paragraphs, helps to customize labels and
layout
\item[enumitem] gives control over labels and lengths
in all kind of lists
\item[mdwlist] is useful to customize description lists,
it
even allows multi-line labels. It features compact lists
and
the capability to suspend and resume.
\item[desclist] offers more flexibility in definition list
\item[multenum] produces vertical enumeration in multiple
columns
\end{description}
\end{document}

2. Typeset to get the definition list:

What just happened?
We used the description environment like the other lists, except that we used the
optional argument of \item. In the description environment, \item is defined
such that the optional parameter will be typeset in the bold typeface.
The package paralist supports the description environment as well; there are
the compactdesc, inparadesc, and asparadesc versions working analogous to
the other lists.
The

same

is

valid

for enumitem: there's the global command
\setdescription{format} working like \setitemize and \setenumerate.
Furthermore, it also extends the description environment to accept optional
formatting arguments of the form key=value, supporting the short form as well.

Have a go hero – adjusting the dimensions of lists
LaTeX's lists have a meaningful layout. However, there might be occasions when
you would like to modify this layout, for instance, to change the margins or the
item indentation. All layout dimensions are determined by LaTeX macros , so
called lengths .
There's a package which is really great for visualizing layouts, which presents
these length macros. It's called layouts and it has been written by Peter Wilson.
Let's use it to examine LaTeX's list dimensions. We will use this small document:
\documentclass[12pt]{article}
\usepackage{layouts}
\begin{document}
\listdiagram
\end{document}

By simply typesetting it, we will get the following diagram:

Isn't it fantastic? The layouts package can do even more, which you can read
about in its documentation. For now, we stay on topic regarding lists.
Though there are LaTeX commands for customizing those lengths, for example,
\setlength{\labelwidth}{2cm}, applying them to individual lists and certain

nesting depths is hard. If you need to modify the list layout, the enumitem package
comes handy again. Its commands like \setlist and its key=value interface
may be used also for adjusting the values of the lengths you could see in the
previous diagram.
For example, if we would like to remove the space between list items in the
description environment and to reduce the left margin, we could load
enumitem and write the following:
\setdescription{itemsep=0cm,parsep=0cm,leftmargin=0.5cm}

Note, we don't use the backslash for keys. Similarly, \setitemize,
\setenumerate, and \setlist can be used for fine-tuning. Try assigning values
by yourself, and test the effect on our examples. If you would like to learn more,
have a look at the enumitem documentation.

Pop quiz
1. A list environment may be nested to up to how many levels?
a. Two
b. Three
c. Four
d. Any number, just depending on the memory
2. All but one of these packages understand options in the form key=value.
Which does not?
a. geometry
b. paralist
c. enumitem
3. In which environment should \item be used with an optional argument?
a. itemize
b. enumerate
c. description
d. inparaenum

Summary
In this chapter, we got to know a new way to structure our text—lists.
Specifically, we learned to create:
Bulleted lists using the itemize environment
Numbered lists using the enumerate environment
Definition lists using the description environment
Furthermore, we worked with compact versions of such lists provided by the
packages paralist and enumitem. Those packages allowed us to customize our
lists at will.
You may regard those lists as an additional possibility to structure your text. Use
it to clarify your ideas. Though you may design the layout of lists as you wish, it's
highly recommended to use the same bullets and numbering scheme respectively
for the same nesting level.

Chapter 5. Creating Tables and
Inserting Pictures
Scientific documents and others do not just contain plain text; they also
present information and data in tables and use diagrams and drawings to
visualize them. We shall become familiar with that.
In this chapter, we will learn how to perform the following:
Lining up text and data in columns
Typesetting complex tables
Including pictures in our documents
Adding captions to pictures and tables
Controlling the placement of figures and tables
Let's tackle these tasks!

Writing in columns
Remember the time of the typewriter and early word processing software? When
we needed to line up some text in columns, we could use tab stops. LaTeX
provides a similar way to easily align text in columns, namely, the tabbing
environment.

Time for action – lining up
information using the tabbing
environment
We would like to present a quick overview regarding LaTeX. We shall present
one point on each line, aligned at words and double colons:
1. Begin a new document and open a tabbing environment:
\documentclass{article}
\begin{document}
\begin{tabbing}

2. Write the text, set tab stops by \=, and end the line by \\:
\emph{Info:} \= Software \= : \= \LaTeX \\

3. Add further lines, move to the next tab stop by \>, and again end lines by \\:
\> Author \> : \> Leslie Lamport \\
\> Website \> : \> www.latex-project.org

4. Close the tabbing environment and end the document:
\end{tabbing}
\end{document}

5. Typeset to get the output:

What just happened?
The tabbing environment that we used begins a new line. We used three simple
tags for markup:
sets a tab stop. We could put several tab stops on a line. If we use
later, the next awaited tab stop would be reset to this position.
\=

\\

ends a row.

\>

goes to the next tab stop. This could also mean to go backwards.

\=

This way, we can quickly produce columns containing left-aligned text. If the
rows of the tabbing environment would reach the end of a page, it could continue
on the next page. What if a column is too long, running over the tab stop? Let's see
how to solve that.

Time for action – lining up font
commands
In Chapter 2, Formatting Words, Lines, and Paragraphs, we've got to know a lot
of font commands and declarations. We saw a table containing those commands
and example output. Now we shall create such a table ourselves:
1. Begin a new document, like the one in step 1 of our previous example, but
define a command for setting the font of our header:
\documentclass{article}
\newcommand{\head}[1]{\textbf{#1}}
\begin{document}
\begin{tabbing}

2. Write the first row set tab stops by \= and use \> to move to the tab stops.
Use the command \verb|…| to typeset the LaTeX commands:
\begin{tabbing}
Family \= \verb|\textrm{...}| \= \head{Declaration} \= \kill
\> \head{Command} \> \head{Declaration} \>
\head{Example}\\
Family \> \verb|\textrm{...}| \> \verb|\rmfamily|
\> \rmfamily Example text\\
\> \verb|\textsf{...}| \> \verb|\sffamily|
\> \sffamily Example text\\
\> \verb|\texttt{...}| \> \verb|\ttfamily|
\> \ttfamily Example text
\end{tabbing}

3. Typeset and examine:

4. As we can see, the tab stops are too narrow. We shall correct it. Create a
new head row containing the tab stops; this time, we will mark the line by
\kill to hide that line. Use filler text to specify the width between the tab
stops. Complete it with further font commands:
\begin{tabbing}
Family \= \verb|\textrm{...}| \= \head{Declaration} \= \kill
\> \head{Command} \> \head{Declaration} \>
\head{Example}\\
Family \> \verb|\textrm{...}| \> \verb|\rmfamily|
\> \rmfamily Example text\\
\> \verb|\textsf{...}| \> \verb|\sffamily|
\> \sffamily Example text\\
\> \verb|\texttt{...}| \> \verb|\ttfamily|

\> \ttfamily Example text
\end{tabbing}

5. Typeset to get the result:

What just happened?
After we noticed that our tab stops have been set too narrow, we constructed a
new first row containing the tab stops. It consists of words representing the
widest entries of each column. To hide this auxiliary row, we used the command
\kill right at the end of the line; \kill at the end of a line causes this line to
have no output.

Note
Like in this example, the command \verb|code| typesets code "as it is", without
interpreting commands within. Instead of |, any character may be chosen as the
delimiter. \verb cannot be used in arguments of commands including \section
and \footnote, and not in table heads.
For longer, verbatim text, use the environment with the same name: verbatim.
There are further useful commands:
at the end of a line causes each subsequent line to start at the first tab. Use
it twice \+\+ to start at the second tab and so on.
\+

\-

cancels a preceding \+; also using multiply has cumulative effect.

at the beginning of a line cancels the effect of one previous \+ command
for that line.
\<

The mentioned commands already allow a good use of the tabbing environment.
Even more commands can be found in the reference manual:
http://texblog.net/help/latex/tabbing.html.

Note
Inside tabbing environments, declarations are local to the current item. A
following \=, \>, \\, or \kill command would stop the effect.
Tabbing environments cannot be nested.

Typesetting tables
We might need more complicated structures and formatting, like centering in
columns, dividing lines, or even nested structures. LaTeX provides the tabular
environment for typesetting simple and complex tables which can be nested.

Time for action – building a table of
font family commands
We shall create a table like in the previous example, but this time, we would like
to make all entries in a column horizontally centered to each other. We will also
add some horizontal lines to mark the border and the header of the table:
1. Create a new document. Define a command for setting the font for the head
row:
\documentclass{article}
\newcommand{\head}[1]{\textnormal{\textbf{#1}}}
\begin{document}

2. Begin a tabular environment. As a mandatory argument, provide
standing for three centered columns:

ccc

\begin{tabular}{ccc}

3. Write the table head row, write & to separate column entries, and \\ to end
rows. Use \hline to insert horizontal lines:
\hline
\head{Command} & \head{Declaration} & \head{Output}\\
\hline

4. Continue with the table body and end the environment and the document. For
typesetting LaTeX commands, write \verb|\command|:
\verb|\textrm| & \verb|\rmfamily| & \rmfamily Example
text\\
\verb|\textsf| & \verb|\sffamily| & \sffamily Example
text\\
\verb|\texttt| & \verb|\ttfamily| & \ttfamily Example
text\\
\hline
\end{tabular}
\end{document}

5. Typeset to see the table:

What just happened?
In the mandatory argument, we wrote a list of characters. Each character stands
for a formatting option. As we used three characters, we've got three columns. c
stands for centered alignment. Therefore, the entries of all columns have been
centered.
Column entries are separated by &, while rows are terminated by \\. Don't end
the last line by \\ unless you further wish to write a line below. It's also a good
idea to align the ampersands in our source code to keep it readable.
Inside the entries, you may use ordinary text as well as LaTeX commands. As in
the tabbing environment, declarations are local to the entry, like if each entry was
surrounded by curly braces.
Furthermore, tabular has an optional alignment argument just like minipage. So,
the complete definition is:
\begin{tabular}[position]{column specifiers}
row 1 col 1 entry & row 1 col 2 entry ... & row 1 col n entry\\
...
\end{tabular}

In the optional argument, t means alignment at the top row and b means alignment
at the bottom row. The default is vertically centered alignment. This may come in
handy if you would like to place two tables next to each other or within further
text.

Drawing lines in tables
Within tabular, three types of lines may be used:
\hline

draws a horizontal line over the whole width of the table

draws a horizontal line starting at the beginning of column m
and ending at the end of column n
\cline{m-n}

\vline

row

draws a vertical line over the full height and depth of the current

Understanding formatting arguments
Of course, further formatting is possible. Have a look at this example table:
\begin{tabular}{|l|c|r|p{1.7cm}|}
\hline
left & centered & right & a fully justified paragraph cell\\
\hline
l & c & r & p\\
\hline
\end{tabular}

This code would produce the following table:

The options understood by the tabular environment are as follows:
l

for left alignment.

c

for centered alignment.

r

for right alignment.

for a "paragraph" cell of a certain width. . If you place several p
cells next to each other, they will be aligned at their top line. It's equivalent
to using \parbox[t]{width} within a cell.
p{width}

inserts code instead of empty space before or after a column. This
might also be some text or it could be left empty to avoid this space.
@{code}

|

stands for a vertical line.

*{n}{options} is equivalent to n copies of options, where n is a positive
integer and options may consist of one or more column specifiers including
* as well.

Tip
It is very advisable to avoid vertical lines in tables. Lines should subtly support
your information but not make reading difficult.
After loading the array package by \usepackage{array} you may use some
options, such as:

m{width}

is similar to \parbox{width}: the base line is at the middle

b{width}

is like \parbox[b]{width}: the base line is at the bottom

can be used like | but inserts code instead of a vertical line. In
contrast to @{…}, the space between columns will not be suppressed.
!{code}

can be used before an l, c, r, p, m, or b option and inserts code
right at the beginning of each entry of that column
>{code}

can be used after an l, c, r, p, m, or b option and inserts code at the
end of the entry of that column
<{code}

This example shows the effect of @{} and the alignment arguments p, m, and b:
\begin{tabular}{@{}lp{1.2cm}m{1.2cm}b{1.2cm}@{}}
\hline
baseline & aligned at the top & aligned at the middle
& aligned at the bottom\\
\hline
\end{tabular}

The output table is as follows:

Increasing the row height
You may have noticed that horizontal lines nearly touch the letters in the cells,
especially capital letters. The array package introduces a length called
\extrarowheight . If it has a positive value, this will be added to the height of
every row of the table.
The next example, following the very first example of this chapter, shows how to
extend the row height and shows the effect of the remaining array options:
\documentclass{article}
\usepackage{array}
\setlength{\extrarowheight}{4pt}
\begin{document}
\begin{tabular}{@{}>{\itshape}ll!{:}l<{.}@{}}
\hline
Info: & Software & \LaTeX\\
& Author & Leslie Lamport\\
& Website & www.latex-project.org\\
\hline
\end{tabular}
\end{document}

The output is as follows:

Here, >{\itshape} has been used to change the font of a row to italic. >{} is
often used to insert an alignment declaration, but there's a pitfall: such
declarations might change the internal meaning of \\, which is a shortcut for
\tabularnewline within tables. But the array package offers a command to
repair it; in such cases, just add \arraybackslash, for example:
\begin{tabular}{>{\centering\arraybackslash}p{5cm}}

Otherwise, the content of paragraph cells stated by p,
justified.

m,

or b will be fully

After a specific row, you can add vertical space by the optional argument of \\,
such as \\[10pt].
You may even stretch a whole table: the macro \arraystretch contains a
stretching factor with a default value of 1. Just redefine it. For example,
\renewcommand{\arraystretch}{1.5} will increase the height of the rows by
50 percent. You could use it inside a group or an environment to keep the effect
local.

Beautifying tables
Still our tables don't look as perfect as they look in good books. Especially, the
lines and their distances to the text might need improvement. The booktabs
package comes to the rescue; after loading it, you can enhance the quality of your
tables by new line commands replacing \hline and \cline.

Time for action – adding nicer
horizontal lines with the booktabs
package
We shall use the new commands introduced by booktabs:
1. In the previous example, load the package booktabs:
\usepackage{booktabs}

2. Use \toprule, \midrule, and \bottomrule instead of \hline. Specify a
thickness as an optional argument:
\begin{tabular}{ccc}
\toprule[1.5pt]
\head{Command} & \head{Declaration}
\midrule
\verb|\textrm| & \verb|\rmfamily| &
text\\
\verb|\textsf| & \verb|\sffamily| &
text\\
\verb|\texttt| & \verb|\ttfamily| &
text\\
\bottomrule[1.5pt]
\end{tabular}

3. Typeset to see the difference:

& \head{Output}\\
\rmfamily Example
\sffamily Example
\ttfamily Example

What just happened?
British typesetters call a line a rule. The booktabs developer chose this
terminology for the new commands. We used three of them. These are their
definitions:
\toprule[thickness] may be used to draw a horizontal line at the top
the table. If desired, a thickness may be specified, like 1pt or 0.5mm.
\midrule[thickness]

of

draws a horizontal dividing line between rows of a

table.
\bottomrule[thickness]

draws a horizontal line to finish off a table.

\cmidrule[thickness](trim){m–n} draws a horizontal line from column
m to column n. (trim) is optional like thickness, it could be (l) or (r) to
trim the line at its left or right end. Write (lr) to trim at both ends. Even
adding {width}, like in (l{10pt}), is possible and specifies the trim

width.
The package does not define vertical lines. They are not advisable anyway. The
same applies to double rules. Neither vertical nor double rules are recommended.
They are even widely considered to be bad typographic style.
Consider using \toprule and the other line commands without optional
arguments—let's figure out how.

Have a go hero – adjusting lengths
We briefly introduced the command \setlength . Instead of specifying a line
thickness by an optional argument to \toprule, \midrule, \cmidrule, or
\bottomrule, always omit it. Instead, specify it once for your whole document
by \setlength in the preamble.
These are the lengths that may be adjusted:
\heavyrulewidth

for the thickness of the top and bottom lines

\lightrulewidth

for the thickness of the middle lines by \midrule

\cmidrulewidth
\cmidrulekern
\abovetopsep

for the thickness of \cmidrule

for the trimming in \cmidrule

is the space above the top rule; the default is 0pt

\belowbottomsep

is the space below the bottom rule; the default is 0pt

\aboverulesep
\bottomrule

specifies the space above \midrule,

\cmidrule,

and

\belowrulesep
\toprule

stands for the space below \midrule,

\cmidrule,

and

Try to change the thickness of the lines. The lengths already have reasoned values,
but you may change them. So, the adjustment in your preamble would improve all
tables in your document.

Spanning entries over multiple columns
Columns concerning the same subject might be grouped by a common header. In
such a case, two cells in the header should be merged. The command
\multicolumn does it for us.

Time for action – merging cells
Regarding our example table, commands and declarations are both input,
whereas, the remaining column contains output. We shall emphasize that in our
header:
1. In our example, insert another header row. Use \multicolumn to merge
cells. Alter the column formatting argument and the middle rule. Changes are
highlighted:
\begin{tabular}{@{}*3l@{}}
\toprule[1.5pt]
\multicolumn{2}{c}{\head{Input}} &
\multicolumn{1}{c}{\head{Output}}\\
\head{Command} & \head{Declaration}
\cmidrule(r){1-2}\cmidrule(l){3-3}
\verb|\textrm| & \verb|\rmfamily| &
text\\
\verb|\textsf| & \verb|\sffamily| &
text\\
\verb|\texttt| & \verb|\ttfamily| &
text\\
\bottomrule[1.5pt]
\end{tabular}

2. Typeset and see the output:

& \\
\rmfamily Example
\sffamily Example
\ttfamily Example

What just happened?
We used the command \multicolumn twice; once to merge two cells and
surprisingly another time just for one cell. Let's first look at its definition:
\multicolumn{number of columns}{formatting options}{entry text}

The number of columns to be spanned may be a positive integer or just 1. The
formatting options will be applied instead of the options specified in the tabular
definition for this cell. We took advantage of this when we used
\multicolumn{1}{c}{…}, overriding the l option of the column by a c option to
get just this cell centered.
The other change we made concerns \cmidrule. We used it instead of \midrule
together with the trimming argument to get a gap between the input and the output
column.

Inserting code column-wise
There are many more font commands that we would like to add to the table.
Writing \verb|…| in each cell is tiresome. We shall exploit the >{…} feature of
the array package to define the formatting of the entries once for the column.

Time for action – using the array
package
We shall modify the table definition to set our input columns in the typewriter
font. At the same time, we will insert a column on the left, standing for our
command type:
1. Extend the preamble of our example by defining a command \normal. It
shall use \multicolumn to produce an l cell, no matter what the column
formatting is:
\documentclass{article}
\usepackage{array}
\usepackage{booktabs}
\newcommand{\head}[1]{\textnormal{\textbf{#1}}}
\newcommand{\normal}[1]{\multicolumn{1}{l}{#1}}
\begin{document}

2. A s \verb cannot be used in table headers, we shall use \ttfamily.
Preceding it with \textbackslash is enough for our purposes. Use >{…} to
insert it twice. Then add <{Example text} to the last column to save typing
work:
\begin{tabular}{@{}l*2{>{\textbackslash\ttfamily}l}l%
<{Example text}@{}}
\toprule[1.5pt]
& \multicolumn{2}{c}{\head{Input}} &
\multicolumn{1}{c}{\head{Output}}\\

3. We'll use the
header:

\normal

command to avoid the typewriter formatting in the

& \normal{\head{Command}} & \normal{\head{Declaration}}
& \normal{}\\
\cmidrule(lr){2-3}\cmidrule(l){4-4}

4. Now we may continue listing the font command names:
Family & textrm&rmfamily & \rmfamily\\
& textsf & sffamily & \sffamily\\
& texttt & ttfamily & \ttfamily\\
\bottomrule[1.5pt]
\end{tabular}
\end{document}

5. Typeset and look at the result:

What just happened?
Using >{\textbackslash\ttfamily}l defines a left aligned row, where each
entry is preceded by a backslash and by switching to typewriter font. We wrote
*2{…} to define two columns of this style. Because the example text has been
inserted according to our table definition with <{…}, we just had to put the
declarations into the last column without the text.

Spanning entries over multiple rows
We already know how to span text over several columns. But what if text should
cross over several rows? LaTeX doesn't define a command for it. However, the
package multirow does.

Time for action – merging cells using
the multirow package
Before we complement the font table, we would like to center the command type
"Family" vertically, that is, span this cell over three columns. If we figured this
out, we could use it later on:
1. In our previous example, additionally load multirow:
\usepackage{multirow}

2. Replace the word «Family»:
\multirow{3}{*}{Family} & textrm & rmfamily & \rmfamily &
\\

3.

Typeset

to see the small change:

What just happened?
We used the command \multirow to span three rows. Its definition is:
\multirow{number of rows}{width}{entry text}

The entry will span that number of rows from the row on which \multirow has
been used. If the number is negative, it will span the rows above.
You can specify a width or just write * for the natural width. If a width has been
specified, the text would be wrapped accordingly.
understands further optional arguments for fine tuning. They are
described in its documentation.
multirow

Adding captions to tables
Especially with longer text, we would like to add captions and numbers to our
tables. Numbering the tables allows referring to them, whereas captions are
informative, if the table is not exactly placed where we refer to it. LaTeX has
built-in features to achieve that.

Time for action – adding a caption to
our font table
Now it's time to complete our table. We shall list the remaining font commands.
We'll use the first column to describe the category of the font commands: Family,
Weight, Shape, and so on. Then we will add another column to show the effect of
combining font commands.
To finish, we shall center the table and provide a number and a caption:
1. Put a table environment around our example table, use \centering inside,
and insert a \caption command at the end of the table environment. Add
more font commands and add another column at the right containing more
examples:
\documentclass{article}
\usepackage{array}
\usepackage{booktabs}
\usepackage{multirow}
\newcommand{\head}[1]{\textnormal{\textbf{#1}}}
\newcommand{\normal}[1]{\multicolumn{1}{l}{#1}}
\pagestyle{empty}
\begin{document}
\begin{table}
\centering
\begin{tabular}{@{}l*2{>{\textbackslash\ttfamily}l}%
l<{Example text}l@{}}
\toprule[1.5pt]
& \multicolumn{2}{c}{\head{Input}}
& \multicolumn{2}{c}{\head{Output}}\\
& \normal{\head{Command}} & \normal{\head{Declaration}}
& \normal{\head{Single use}} & \head{Combined}\\
\cmidrule(lr){2-3}\cmidrule(l){4-5}
\multirow{3}{*}{Family} & textrm & rmfamily & \rmfamily &
\\
& textsf & sffamily & \sffamily& \\
& texttt & ttfamily & \ttfamily& \\
\cmidrule(lr){2-3}\cmidrule(lr){4-4}
\multirow{2}{1.1cm}{Weight} & textbf & bfseries &
\bfseries
& \multirow{2}{1.8cm}{\sffamily\bfseries Bold and sansserif} \\
& textmd & mdseries & \mdseries & \\
\cmidrule(lr){2-3}\cmidrule(lr){4-4}
\multirow{4}{*}{Shape} & textit & itshape & \itshape & \\
& textsl & slshape & \slshape &
\multirow{2}{1.8cm}{\sffamily\slshape Slanted and sansserif}\\

& textsc & scshape & \scshape & \\
& textup & upshape & \upshape & \\
\cmidrule(lr){2-3}\cmidrule(lr){4-4}
Default & textnormal & normalfont & \normalfont & \\
\bottomrule[1.5pt]
\end{tabular}
\caption{\LaTeX\ font selection}
\end{table}
\end{document}

2. Typeset and our table is now ready:

What just happened?
We put the tabular environment in a table environment. It's used in this way
together with the \caption command:
\begin{table}[placement options]
table body
\caption{table title}
\end{table}

The table environment is a so-called floating environment . Unlike normal text,
they might appear somewhere else other than what is defined by their position in
the source code. The optional placement argument determines where the table
might appear. However, LaTeX decides it. We shall discuss this at the end of this
chapter together with the placement of graphics.
\ caption understands an optional argument as well: if you write
\caption[short text]{long text}, then short text will appear in a list of
tables and long text in the document body. That's useful if you need very long

descriptive captions.
Tables are automatically numbered.

Placing captions above
In typesetting, it's very common to place captions above the tables instead of
below. This can be achieved by writing \caption before the table body.
However, LaTeX expects caption to always be below, resulting in a cramped
look to the table. There's too little space between the caption and the following
table. You might wish to add some space, for instance, by entering
\vspace{10pt} directly after a top caption.
Remember booktabs? If you begin tables with \toprule, just specify the length
\abovetopskip, for example:
\setlength{\abovetopsep}{10pt}

By putting this line into your preamble, 10pt space would be added below the
caption and above the top line of the table.

Have a go hero – customizing captions
By default, the captions look like normal body text; there's no visual difference.
Would you like to have a slight change in font size, a different formatting of the
label, some margins or indentation, or any other customization? The caption
package is the answer to most needs. By using a few options, you could enhance
the visual appearance of all of your captions. Try:
\usepackage[font=small,labelfont=bf,margin=1cm]{caption}

This way, your captions will be smaller than normal text, the label with number
will be bold, and it will not be as wide as normal text. The package offers a lot of
features, both for document wide settings and fine-tuning. It's very well
documented. So, have a look at its documentation.

Auto-fitting columns to the table width
l, c,

and r columns have the width of their content. For p columns, you specify the
width. This way, it's hard to find out the actual width of the table. Wouldn't it be a
good idea to specify the table width and let LaTeX decide how wide the columns
may be? The tabularx package allows that. Using it looks like:
\usepackage{tabularx}
...
\begin{tabularx}{width}{column specifiers}
...
\end{tabularx}

The new environment tabularx requires an additional argument: the width of the
table. It introduces a new column type X. X columns behave like p columns, but
they use all available space. One X column would take all of the available space.
If you use several X columns, they would share the space equally. So you could
write, for instance:
\begin{tabularx}{0.6\textwidth}{lcX}

This way you would get a table occupying 60 percent of the text width, a left
aligned and a centered column as wide as their content, and a paragraph column
as wide as possible until 60 percent is reached.
Though it's easy to use, the tabularx documentation gives further examples,
informs about the derived types, and gives advice like this: don't let
\multicolumn entries cross any X column.
There are two similar approaches:
LaTeX provides a starred version of the tabular environment:
\begin{tabular*}{width}[position]{column specifiers}

The table is set to width, but by modifying the inter-column space.
tabularx has been developed satisfying the need for a more useful way.
The tabulary package provides another sophisticated tabular environment
taking the total width. It's weighting each column width according to the
natural width of the widest cell in the column.

Generating multi-page tables
All tabular environments we've got to know until now cannot cross page
boundaries. The tabbing environment is an exception due to its different nature.
As tables might contain a lot of data, we need a solution. There are several
packages:
longtable provides an environment with the same name that's like a multipage version of tabular. It provides commands to set table captions,
continued captions, and special headers and footers when a page break
occurs. It's probably the easiest way for multi-page tables and therefore the
most popular. The package documentation describes all you need. In
combination with the booktabs package, you will get very good results.
ltxtable provides a combination of longtable and tabularx.
ltablex is another approach to combine the features of longtable and
tabularx.
supertabular offers another multi-page extension of the internally used
tabular environment, providing optional table tails and heads where page
breaks occur.
xtab extends supertabular and reduces some of its weaknesses.
stabular implements a simple way to use page breaks in tabular without
much ado.
Example tables and links to documentation
http://texblog.net/beginners-guide/tables/.

can

be

found

at

Coloring tables
We didn't even color text yet, as this usually isn't what we do first with LaTeX.
But of course, this can be done with text as well as with tables. For coloring text,
use the color package or, better, the extension xcolor. For coloring tables, use the
package colortbl . All can be combined by using:
\usepackage[table]{xcolor}

The package allows coloring columns, rows, single entries, and lines in many
ways. The package documentation may tell you more.

Using landscape orientation
Very wide tables could be typeset in landscape orientation. The rotating package
offers an environment called sidewaystable that you could use instead of the
table environment. Both table and caption would be rotated +-90 degrees and
placed on a separate page. The package provides further rotation-related
environments and commands.

Aligning columns at the decimal point
Columns containing numbers are more readable when the entries are aligned at
the decimal marker and perhaps at an exponent. Several packages support this:
siunitx is primarily intended for typesetting values with units in a consistent
way according to scientific conventions. However, it provides a tabular
column type for such decimal alignment of numbers.
dcolumn offers a column type for aligning at a comma, a period, or another
single character.
rccol defines a column type where numbers are "right-centered", that is, they
are centered with respect to other entries but flushed right to each other. This
way corresponding digits are aligned along the column.
In contrast to dcolumn and rccol, the siunitx package is very new and
powerful.

Handling narrow columns
Text in very narrow columns might require special attention because justification
is difficult if there's little space. Here's some advice:
Have a look at the correct hyphenation. If necessary, improve it like we did
in Chapter 2.
TeX doesn't hyphenate the first word of a line, a box, or a table entry. So, a
long word may cross the column boundary. To enable hyphenation, insert an
empty word: write \hspace{0pt} directly at the beginning.
Load microtype to improve justification, it shows the best effect in narrow
columns.
Full justification in p columns and the like may look bad because of big
gaps. Consider using >{\raggedright\arraybackslash} for such
columns.
From the ragged2e package, using the command \RaggedRight can do even
better and doesn't need \arraybackslash.

Pop quiz – tables
1. Which of these environments cannot cross page boundaries?
a. tabular
b. longtable
c. tabbing
2. If you would like to format one entry different from the specification of its
column, which command may be used?
a. >{…}
b. @{…}
c. \multirow
d. \multicolumn

Inserting pictures
Documents may not consist of just text and tables. You might wish to include
pictures, diagrams, or drawings made with other programs. The graphicx package
is dedicated to this.

Time for action – including a picture
We shall create a short document. Between two paragraphs, we would like to
insert a picture. If there's no picture file available, we will use the demo mode of
graphicx:
1. Begin a new document and load babel and blindtext to print some filler
text:
\documentclass[a5paper]{article}
\usepackage[english]{babel}
\usepackage{blindtext}
\usepackage[demo]{graphicx}
\pagestyle{empty}
\begin{document}
\section{Including a picture}
\blindtext

2. Open a figure environment and declare centering:
\begin{figure}
\centering

3. Use the command \includegraphics with the filename as the argument:
\includegraphics{test}

4. Declare a caption, close the figure environment, and end the document with
filler text:
\caption{Test figure}
\end{figure}
\blindtext
\end{document}

5. Typeset and have a look at the output:

What just happened?
The most important command is \includegraphics. We specified a filename.
This file would be loaded if it exists. Otherwise an error would occur. LaTeX
supports the following file types:
PNG, JPG,
EPS

and PDF if you directly compile to PDF (pdfLaTeX)

if you compile to DVI and convert to PS and PDF (traditional LaTeX)

To clarify, PS means PostScript, EPS means Encapsulated PostScript, and DVI
means Device Independent Format. The latter was the first output format to be
supported by TeX. You definitely know the very popular picture formats PNG and
JPG, often used for screenshots or photos.
You don't need to specify a filename extension, it will be automatically added.
Put the file into the same directory as your document; otherwise specify a full or
relative path name:
\includegraphics{appendix/figure1}

Don't use blanks in the filename or path! Blanks and special characters may cause
problems with \includegraphics. If such symbols in filenames are required,
load the package grffile to try to fix it. Also, in filenames, use slashes / instead of
backslashes \, as the latter begin a LaTeX command.
Because we specified the demo option, graphicx doesn't require a file test.png
or any other file; instead it's just printing a black filled rectangle. This is useful
for testing or if you would like to discuss a LaTeX problem in an online forum,
but don't wish to publish your pictures.
Go ahead; copy a picture of your choice into your document directory, give
\includegraphics its filename, and compile. The picture will be embedded
with its original size.

Scaling pictures
You may choose a different size. For this, let's look at the definition of
includegraphics:
\includegraphics[key=value list]{file name}

The graphicx documentation lists all keys and possible values. Here are the
most popular ones:
width: The graphic
width=0.9\textwidth.
height: The
height=3cm.

would be resized to this width. Example:

graphic would be resized to this height. Example:

scale:

The graphic would be scaled by this factor. Example: scale=0.5.

angle:

The graphic would be turned by this angle. Example: angle=90.

There are options for clipping, but such post processing can be easily done with
any graphics software.
Instead of turning a figure by 90 degrees, you could use the sidewaysfigure
environment of the rotating package. It's analogous to sidewaystable that has
been mentioned before.

Choosing the optimal file type
If you've got the final picture, well, just use this format and include it. However,
if you are free to choose a file format, you could think about it for a moment.
and PDF are both vector graphics format. They are scalable, also looking
good at high resolution or if you zoom in. So, whenever possible, PDF (or EPS)
should be preferred, for instance, when you export drawings or diagrams out of
other office software. For such graphics, vector formats are common.
EPS

and JPG are bitmap formats, also called raster graphics , commonly used for
photos. If you zoom in, you would notice a loss of quality. PNG uses a lossless
compression, whereas JPG pictures may lose quality when they are saved. So, if
you make screenshots, use PNG or at least ensure that there's no loss compression
if you choose JPG.
PNG

Note
Besides supporting vector graphics, both EPS and PDF may contain bitmap
graphics. They are also called container formats.
There are a lot of tools to convert between graphic formats. These three are
especially useful and included in both TeX Live and MiKTeX:
dvips converts DVI files to the PostScript format
ps2pdf converts PostScript files to PDF
epstopdf converts EPS files to PDF, a package with the same name does it
on-the-fly
These are command line tools. Some editors like TeXnicCenter use them to
automate the way TEX => DVI => PS => PDF.
epstopdf is especially useful if you have to include Postscript pictures, but wish
to benefit from pdfLaTeX features like font expansion and character protrusion
accessible by microtype.
For further working with graphics, ImageMagick and GIMP are very capable
open source programs.

Including whole pages
How can we include pictures wider or higher than the text area?
\includegraphics could do it, but LaTeX would complain about width or size
and might put it off to the next page. Oversized images and even whole pages can
be included using the pdfpages package. It provides a command, \includepdf,
which is able to include a complete page and even a multi-page PDF document at
once. Despite its name and not even mentioned in its otherwise good
documentation, it's capable of including PNG and JPG files, not just PDF.
could also be used to resize several PDF pages and to arrange them on
a single sheet.
pdfpages

Putting images behind the text
Do you need watermarks? Background images? Textboxes positioned at arbitrary
positions on the page, preferably not interfering with the other text? The package
eso-pic does it for you.
Another approach is offered by the textpos package, developed for placing boxes
with text or graphics at absolute positions on a page.

Managing floating environments
When a page break occurs, normal text can be broken to continue on the next page.
However, pictures cannot be divided. Simple tables cannot be split either. That's
why LaTeX provides two floating environments, namely, figure and table.
They are briefly called floats . Their content may float to a place where it's the
optimum for the page layout. Let's figure out how to deal with it.

Time for action – letting a figure float
Both the figure and the table environment take an optional argument affecting
the final placement of the figure or the table. We shall test the effect in our
graphics example:
1. Go back to the previous example. This time, add the options h and t:
\begin{figure}[ht]
\centering
\includegraphics{test}
\caption{Test figure}
\end{figure}

2. Typeset, notice the change in the output.
3. Change the options into !b:
\begin{figure}[!b]

4. Typeset, the figure is now forced to float to the bottom. Compare both
results:

What just happened?
Just by adding some characters standing for placement options, we could force the
figure to appear where we wanted it to.

Understanding float placement options
The optional argument of the figure and table environment tells LaTeX where
it's allowed to place the figure or the table. Four letters stand for four possible
places:
stands for here. The float may appear where it's been written in the source
code.
h

t

stands for top. Placing at the top of a page is permitted.

b

stands for bottom. The float may appear at the bottom of a page.

stands for page. The float is allowed to appear on a separate page, where
only floats may reside but no normal text.
p

A fifth option might come in handy:
tells LaTeX to try harder! Some constraints may be ignored, easing the
placement.
!

If you don't specify any option, the float could be placed far away. New LaTeX
users might be surprised. Specifying more options would help to place it as near
as possible. The most flexible is using the placement [!htbp], allowing a float
everywhere. You still could consider removing a placement specifier if you don't
like it.

Forcing the output of floats
If you would like to stop LaTeX from putting the floats, there's a way; the
\clearpage command ends the current page and causes all already defined
figures and tables to be printed out. You can use \cleardoublepage that does the
same but in a two-sided layout it ensures that the next non-float page is a righthand page. . If necessary, a blank page would be inserted.
Immediately ending the page might not be the best thing to do. It could leave a lot
of empty space on the current page. The afterpage package offers a clever
possibility; this package allows deferring the execution of \clearpage until the
current page has ended:
\usepackage{afterpage}
...
body text
\afterpage{\clearpage}

Limiting floating
It may happen that tables and figures float far away, perhaps even into another
section. The placeins package provides a useful command to restrict the floating.
If you load placeins with \usepackage{placeins} and write \FloatBarrier
somewhere in your document, no table or figure could float past it. This macro
keeps floats in their place.
A very convenient way to prevent floats from crossing section boundaries is
stating the section option:
\usepackage[section]{placeins}

This option causes an implicit \FloatBarrier to be used at the beginning of each
section.
Two further options, namely, above and below, allow you to lower the
restrictions, preventing floats from appearing above the start of the current section
or below the start of the next section.
Figures don't float into the next chapter because \chapter implicitly uses
\clearpage.

Avoiding floating at all
Would you like to place a float exactly where you want it? The obvious answer
is: don't use a floating environment! Even a caption doesn't have the same
importance if the table or figure is placed at the current point. However, if you
really need a caption, you may use the \captionof command without a float.
Both the caption package , KOMA-Script classes, and the tiny capt-of package
provide that command:
\usepackage{capt-of}% or caption
…
\begin{center}
\begin{minipage}{\linewidth}%
\centering%
\includegraphics{test}%
\captionof{figure}{Test figure}%
\end{minipage}
\end{center}

The minipage keeps a picture and caption together. Furthermore, it's centered.
Use \captionof within a group or environment like minipage or center. The
definition is the same as \caption, except there is the additional argument
specifying the float type:
\captionof{figure}[short text]{long text}
\captionof{table}[short text]{long text}

Be aware that the numbering could get wrong if you mix real floats and fixed
figures or tables. As you don't benefit from LaTeX's positioning capabilities, you
have to take care that pages are still properly filled.
The float package provides a convenient and consistent looking approach. It
introduces the placement option H causing the float to appear right there:
\usepackage{float}
…
\begin{figure}[H]
\centering
\includegraphics{test}
\caption{Test figure}
\end{figure}

You may choose between these two options. If you wish to exploit further features
of the float package, load it. Otherwise, consider using the one-liner capt-of.
Perhaps not even that is necessary, assuming that you're already using the
caption package or a KOMA-Script class.

Spanning figures and tables over text columns
There are starred forms of floats, namely, figure* and table*. In a two-column
layout, they put the float into a single column. In one-column mode, there's no
difference to the non-starred form.

Letting text flow around figures
Though it's a bit playful, you might wish to let text flow around a table or a figure.
This can be achieved using the wrapfig package and its environments
wrapfigure and wraptable.

Time for action – embedding a
picture within text
We shall modify our picture embedding an example. We would like the picture to
appear on the left side, accompanied by the body text on the right side:
1. In our example, additionally load the wrapfig package:
\documentclass[a5paper]{article}
\usepackage[english]{babel}
\usepackage{blindtext}
\usepackage[demo]{graphicx}
\usepackage{wrapfig}
\pagestyle{empty}
\begin{document}

2. Begin an unnumbered section, place a wrapfig environment within some
filler text:
\section*{Text flowing around a picture}
\blindtext
\begin{wrapfigure}{l}{4.4cm}
\includegraphics[width=4.4cm]{test}
\caption{Test figure}
\end{wrapfigure}
\blindtext
\end{document}

3. Typeset and look:

What just happened?
T he wrapfigure environment understands parameters other than the figure
environment. We used just two of them. If you need more, here's the complete
definition:
\begin{wrapfigure}[number of lines]{placement}[overhang] {width}

The first optional argument states the number of wrapped text lines. If omitted, it
would be automatically calculated from the height. placement can be one of the
characters r, l, i, o for right, left, inner, or outer side or the corresponding
uppercase letters R, L, I, O with the same meaning, but allowing the figure to float.
Only one character for specifying the option is allowed. The other optional
argument, overhang, may specify a width that the figure might stick into the
margin. The final and mandatory argument gives the width of the figure.
The wraptable environment works analogous.

Breaking figures and tables into pieces
For grouping several sub figures or sub tables with captions within a single figure
or table, there are several supporting packages you can choose from:
subfig is a sophisticated package supporting inclusion of small figures and
tables. It takes care of positioning, labeling, and captioning within single
floats.
subcaption is another package for this purpose and belongs to the caption
package.
subfigure is still available, but considered as obsolete since subfig has
appeared.

Pop quiz – pictures and floats
1. This chapter listed all graphic formats supported by LaTeX. Now, identify
from the following file types which is not directly supported by LaTeX!
a. jpg
b. gif
c. png
d. eps
e. pdf
2. If you would like a figure or a table to float but to appear as near as possible
to its source, which of the following float placement options would be
appropriate?
a. h
b. H
c. htbp
d. !htbp

Summary
In this chapter, we have learned to create tables and figures, and how to place
them within our document. Specifically, we dealt with:
Putting text into columns
Typesetting tables with captions
Spanning columns and rows in tables
Using packages to auto-fit columns and to create colored, landscape, and
even multi-page tables
Including pictures with captions
Controlling and tuning placement of figures and tables
LaTeX is able to generate lists of tables and figures like a table of contents. We
shall deal with such lists in Chapter 7, Listing Contents and References.
As our figures and tables are numbered, we can use these numbers to refer to them
within the text. In the next chapter, we shall figure out how to do this, benefitting
from LaTeX's cross-referencing capabilities.

Chapter 6. Cross-Referencing
Our documents contain a lot of numbered things such as pages, sections, list
items, figures, and tables. There's even more we have not covered yet. For
instance, if you would like to write a mathematical text, you may number
equations, theorems, definitions, and many more.
We number things not just to count them, but to refer to them in other places of
our document. If you want to point the reader to the ninth figure in the third
chapter, you might write "See figure 3.9". LaTeX automatically enumerates the
figures for you. If you insert another figure, LaTeX will automatically adjust
the numbering of all figures after it. But what's with the references?
LaTeX is able to take care of all of our cross-references. This is the subject of
this chapter. We shall learn how to:
Set labels that we shall use to refer to
Refer to sections, footnotes, list items, tables, and more
Refer to page numbers and ranges
Make LaTeX refer verbosely to adjacent pages
Automate naming of references
Create references to external documents
Let's figure out how to do all this.

Setting labels and referencing
To be able to refer to a certain point, we have to mark it by a label. The name of
that label will serve us afterwards.

Time for action – referencing items of
a top list
We shall typeset a list of the most used packages for papers on the e-print archive
site http://arXiv.org. Through the command \label, we will mark items that we
can later refer to with the command \ref.
1. Create a new document with this code:
\documentclass{book}
\begin{document}
\chapter{Statistics}
\section{Most used packages on
arXiv.org}\label{sec:packages}
The Top Five packages, used on arXiv.org\footnote{according
to the arXMLiv project\label{fn:project}}:
\begin{enumerate}
\item graphicx
\item amssymb \label{item:amssymb}
\item amsmath \label{item:amsmath}
\item epsfig
\item amsfonts
\end{enumerate}
\chapter{Mathematics}
\emph{amsmath}, on position \ref{item:amsmath} of the top
list
in section~\ref{sec:packages} on
page~\pageref{sec:packages},
is indispensable to high-quality mathematical typesetting in
\LaTeX.\emph{amssymb}, on position \ref{item:amssymb},
provides a huge amount of math symbols.
See also the footnote on page~\pageref{fn:project}.
\end{document}

2. Typeset and have a look at the text on page 3:

3. Do you see the question marks? The references are still missing! Typeset
again and compare:

What just happened?
We created cross-references with just three commands: \label marks the
position, \ref prints the number of the element we refer to, and \pageref prints
the page number of that element. Each command takes the name of the element as
argument. Any name may be chosen by us.
We had to Typeset twice because LaTeX needs one run to produce the references
that can be read in during the next compiler run. If LaTeX cannot resolve a
reference, it prints two question marks instead.
Let's have a closer look.

Assigning a key
The command \label{name} assigns the current position to the key name.
Specifically:
If the \label command appeared in ordinary text, then the current sectional
unit, like the chapter or the section, would be assigned.
If the \label would be placed within a numbered environment, that
environment would be assigned to the key.
So, we cannot mark a section within a table environment. To avoid any problem
because of a possible unsuitable positioning, a good rule of thumb is to place the
\label command right after the position we would like to mark. For instance,
place it directly after the corresponding \chapter or after \section—not
before, of course.
In the figure or table environments, \caption is responsible for the numbering.
That's why \label has to be placed after \caption, not before. Therefore,
typical floating environments look like the following:
\begin{figure}[htp]
\centering
\includegraphics{filename}
\caption{Test figure}\label{fig:name}
\end{figure}

Or as follows:
\begin{table}[hbp]
\centering
\caption{table descripion}\label{tab:name}
\begin{tabular}{cc}
…
\end{tabular}
\end{table}

A key may consist of letters, digits, or punctuation characters. Keys are casesensitive.
If you write larger documents, the number of keys could become very high.
Imagine, you have a section dealing with fonts and a font table—how to
distinguish their keys? We could prefix them with the type of environment. It has
become common practice to label figures with fig:name, tables with tab:name,
sections with sec:name, and similar in other cases.

Referring to a key
Once a label has been set and given a name, we may refer to that name. For this,
we use \ref{name}: this command prints the number that belongs to name. It
could already be used before the corresponding \label command appears in your
code.
Even though it's that simple, it's powerful. Each time we Typeset a document,
LaTeX checks the keys and reassigns the numbers, responding to all changes. If
LaTeX noticed that labels have been changed, it would inform you that a second
Typeset run would be required to update the corresponding labels. If in doubt
Typeset twice.

Referring to a page
The command \pageref{name} works analogous to \ref, except that it prints the
corresponding page number.
Would all the references stay correct if we changed the section and page
numbers? Let's put it to the test! Insert a section and a page break at the beginning:
\chapter{Statistics}
\section{Introduction}
\newpage
\section{Most used packages on arXiv.org}\label{sec:packages}

Typeset once. LaTeX will compile it, but it will show a message: LaTeX
Warning: Label(s) may have changed. Rerun to get cross-references right .
That's what we shall do! Typeset a second time, now all the numbers have been
correctly adjusted:

Using a reference together with the page number reference, you may write:
See figure~\ref{fig:name} on page~\pageref{fig:name}.

As you know how to define a command, you could make such referencing easier:
\newcommand{\fullref}[1]{\ref{#1} on page~\pageref{#1}}
…
See figure~\fullref{fig:name}.

This way, you would get a full reference like "See figure 4.2 on page 32".
However, if the reference appears on the same page, like the figure, writing out
the page number looks a bit odd. How can we avoid that? The varioref package
provides a way.

Producing intelligent page references
The varioref package offers a command being able to add "on the preceding
page", "on the following page", or on the page number to a reference, depending
on the context.

Time for action
variable references

–

introducing

We will use the varioref commands, \vref a n d \vpageref , to achieve
enhanced reference texts:
1. Open our current example. Add the package varioref to your preamble:
\usepackage{varioref}

2. Edit the content of the second chapter:
\emph{amsmath}, on position \ref{item:amsmath} of the top
list
in section~\vref{sec:packages}, is indispensable to highquality
mathematical typesetting in \LaTeX. \emph{amssymb}, on
position
\ref{item:amssymb}, provides a huge amount of math symbols.
See also the footnote \vpageref{fn:project}.

3. Typeset twice and look at the result:

What just happened?
The command \vref checked the distance to the referenced section. As it's on the
facing page, that is, on the preceding page in a two-sided layout, it wrote "1.2 on
the facing page". Similar for \vpageref, it refers to "the preceding page".
\vref{name}

acts in the following way:

If the reference and \label{name} are on the same page, it behaves exactly
like \ref. The page number will not be printed.
If the reference and the corresponding \label are on two successive pages,
\vref prints the referred number and additionally "on the preceding page"
or "on the following page", or "on the facing page". The latter will be chosen
if the document is two-sided, that is, if \label and the reference fall onto a
double-page spread.
Otherwise it will print both \ref and \pageref.
is equivalent to \pageref, but behaves like \vref concerning the
page reference.
\vpageref

Even though varioref defines new commands, you may still use the common
\ref and \pageref.

Fine-tuning page references
If label and reference are very close to each other, they would probably fall on
the same page, but not necessarily. In such cases, we usually know if the label
comes before or after the reference. It allows specifying an optional argument to
\vpageref:
see the figure \vpageref[above]{fig:name}

This will print:
"see the figure above" if the figure is on the same page
"see the figure on the page before" if the figure is on the preceding page
Whereas, with the following:
see the footnote \vpageref[below]{fn:name}

This will print:
"see the footnote below" if the footnote is on the same page
"see the footnote on the following page" if the footnote is on the next page
In fact, \vpageref understands two optional arguments. The second optional
argument would be used if the label and reference would fall on different pages.
So, we could even write:
see the figure \vpageref[above figure][figure]{fig:name}

This would print:
"see the above figure" if the figure is on the same page
"see the figure on the page before" if the figure is on the preceding page
Actually, reversing the word order to "above figure" if necessary.
Sounds complicated? Well, your demands might increase over time, requiring
more sophisticated features, so these features might come in handy some day.

Referring to page ranges
varioref

offers two more commands:

\vpagerefrange[opt]{key1}{key2},

w her e key1 and key2 denote a
range (like a sequence of figures from fig:a to fig:c). If both labels fall
onto the same page, the result is the same as with \vpageref. Otherwise, the
output will be a range like "on pages 32-36". opt would be used if both
labels fall onto the current page.
\vpageref [opt]{key1}{key2} is analogous, but
figures \vpageref{fig:a}{fig:c} may result in

similar to \vref: see
"see figures 4.2 to 4.4

on pages 36-37".
\vref, \vpageref,

and \vpagerefrange have been implemented to remove any
space to their left before they insert space of their own. For example, \vref
inserts a nonbreakable space like we did before with ~. If this is not desired, use
starred forms \vref*, \vpageref*, or \vpagerefrange*.
More information regarding customization may be found in the package manual.

Using automatic reference names
Tired of writing figure~\ref{fig:name} and table~\ref{fig:name} again
and again? Wouldn't it be great if LaTeX knew what type is meant by
\ref{name} writing type and number? What if we desire to abbreviate, say,
fig.~\ref{fig:name} in the whole document? The cleverev package eases the
work: it automatically determines the type of cross-reference and the context in
which it's used.
Basically, you could just use \cref instead of \ref or \Cref if you wish to
capitalize. The corresponding range commands are \crefrange and
\Crefrange.

Time for action – referring cleverly
We shall rewrite our first example using cleveref. To verify that the package
acts cleverly, we consciously omit prefixes in key names to \label and \cref.
1. Modify our first example in this way:
\documentclass{book}
\usepackage{cleveref}
\crefname{enumi}{position}{positions}
\begin{document}
\chapter{Statistics}\label{stats}
\section{Most used packages on arXiv.org}\label{packages}
The Top Five packages, used on arXiv.org\footnote{according
to the arXMLiv project\label{project}}:
\begin{enumerate}
\item graphicx
\item amssymb \label{amssymb}
\item amsmath \label{amsmath}
\item epsfig
\item amsfonts
\end{enumerate}
\begin{table}[tp]
\centering
(Identify obsolete packages on CTAN)
\caption{Obsolete packages}\label{obsolete}
\end{table}
\chapter{Mathematics}
\emph{amsmath}, on \cref{amsmath} of the top list in
\cref{packages} of \cref{stats}, is indispensable to highquality
Mathematical typesetting in \LaTeX. \emph{amssymb}, on
\cref{amssymb}, provides a huge amount of math symbols.
\Cref{obsolete} shows obsolete and outdated packages that
should be avoided. See also the \cref{project}.
\end{document}

2. Typeset twice and check the references for having the correct names:

What just happened?
As we can see, we never needed to specify which object we refer to. \cref
always chooses the right name and the correct number for us. That's really
comfortable!
We used the command \crefname to tell cleveref which name it should use for
enumerated items. The definition of \crefname is:
\crefname{type}{singular}{plural}
type may be one of chapter, section, figure, table, enumi, equation,
theorem, or many other types we did not encounter yet. The singular version will

be used for single references and the plural version for multiple. If you need
capitalized versions, use \Crefname. So, a typical use may be:
\crefname{figure}{fig.}{figs.}
\Crefname{figure}{Fig.}{Figs.}

To sum up your benefits:
You save much typing.
You could use arbitrary labels. The package fancyref does a similar job
but relies on prefixes like chap, fig, tab, and so on.
If you decide to change wordings, it could be done easily by doing this once
in the preamble, having the desired effect in the whole document.
However, it's recommended to use a prefix like fig: or sec: to distinguish the
kind of referenced object; your code would become more understandable—it's
common.

Combing cleveref and varioref
As cleveref fully supports varioref, you may use both to get the most out of
them. cleveref redefines the commands of varioref to use \cref internally.
So, you could use the good page referencing features of varioref together with
the clever naming automatism!
Just load varioref before cleveref:
\usepackage{varioref}
\usepackage{cleveref}

Now, you may use
seems appropriate.

\vref, \cref, \ref,

or the other commands—whichever

Referring to labels in other documents
If you write several related documents that refer to each other, you might want to
use references to labels of another document. The package with the short name xr
(standing for eXternal References) implements it. First load the package:
\usepackage{xr}

If you need to refer to sections or environments in an external document called,
say, doc.tex, insert this command into your preamble:
\externaldocument{doc}

This enables you to additionally refer to anything that has been given a label in
doc.tex. You may do this for several documents. If you need to avoid conflicts
when an external document uses the same \label like the main document, declare
a prefix using the optional argument of \externaldocument:
\externaldocument[D-]{doc}

This way, all references from doc.tex would be prefixed by D- and you could
write \ref{D-name} to refer to name in doc.tex. Instead of D- you may choose
any prefix that transforms your labels such that they become unique.

Have a go hero – turning references into hyperlinks
PDF documents offer bookmarks and hyperlink capabilities. How about
exploiting that ourselves? There's an outstanding package offering hyperlink
support; it's called hyperref . Try it: load hyperref right before cleveref. This
order is important for the references to work. Even without any options or
commands, your document will be hyperlinked as much as possible:
All references become hyperlinks. Click any of those numbers to jump to the
referred table, list item, section, or page.
Each footnote marker is a hyperlink to the footnote text. Click it to jump
down.
If you insert \tableofcontents, you will get a bookmark list for the
document, chapters, and sections listed in a navigation bar of you PDF
reader.
can do even more for you: linking index entries to text passages, backreferencing of bibliography entries, and more. You can finely customize the
behavior using options, for instance, choosing color or frames for hyperlinks. So,
you could keep that valuable package in mind. In Chapter 11, Enhancing Your
Documents Further, we shall return to this topic.
hyperref

Pop quiz
1. For a reference to a table, at which position should the \label be placed?
a. Before \caption
b. After \caption
c. After \end{table}
2. What should we do if we see two unexpected consecutive question marks in
the output?
a. Load the hyperref package.
b. Look out for an undefined command. Check the syntax at that point.
c. Typeset again. If the question marks stay, check the key to the reference at
that point.

Summary
In this chapter, we learned how to reference chapters, sections, footnotes, and
environments by their number or by the number of the corresponding page.
We even got to know some ways of clever and context dependent referencing.
In the next chapter, we shall deal with lists, which consist mainly of references:
table of contents, lists of figures and tables, and bibliographies.

Chapter 7.
References

Listing

Content and

LaTeX makes it very easy to create lists for many purposes. For example, we've
seen that just the simple command \tableofcontents creates a nice looking
table of contents. Let's just call it the TOC from now on. It simply takes the
entries from the headings and from the numbers of the pages they fall on.
In this chapter, we will perform the following:
Decide which headings at which numbering level go to the TOC
Create and customize a list of figures (LOF) and a list of tables (LOT)
Insert arbitrary text and commands into the TOC and into other lists
Cite books and create a bibliography
Create an index
Modify the headings for all those lists
We shall start with the content.

Customizing the table of contents
Besides just calling \tableofcontents to get a pre-designed list of content,
LaTeX provides basic ways to modify it. Let's use some.

Time for action – refining
extensive table of contents

an

We shall build the frame of a document containing some headings. We will
modify the automatically created table of contents to be finer and to contain
additional entries.
In Chapter 3, Designing Pages, we saw the effect of \tableofcontents. The
entries have been collected from the headings. We shall use this down to the
subsubsection level.
Then we shall extend the TOC further. We will manually add entries for some
headings:
1. Create a new document with the following code:
\documentclass{book}
\setcounter{tocdepth}{3}
\begin{document}
\tableofcontents
\part{First Part}
\chapter*{Preface}
\addcontentsline{toc}{chapter}{Preface}
\chapter{First main chapter}
\section{A section}
\section{Another section}
\subsection{A smaller section}
\subsubsection[Deeper level]{This section has an even deeper
level}
\chapter{Second main chapter}
\part{Second part}
\chapter{Third main chapter}
\appendix
\cleardoublepage
\addtocontents{toc}{\bigskip}
\addcontentsline{toc}{part}{Appendix}
\chapter{Glossary}
\chapter{Symbols}
\end{document}

2. Typeset. The first page will just show Contents but no entries.
3. Typeset a second time. Now the table of contents will be displayed:

What just happened?
We structured a document using several sectioning commands. In the first run,
LaTeX read all of our sectioning commands and created a file with the extension
.toc. This file contains the commands and the titles for all entries in the table of
contents. During the first run, that file didn't exist yet, thus the TOC remained
empty.
During the second run, the command \tableofcontents read the .toc file and
printed the TOC.
In this example, we raised the depth of the TOC by one level. We added a
chapter-like entry for the preface and inserted a part-like heading showing the
beginning of the appendix, using \addcontentsline . Through \addtocontents
, we inserted some space before the latter heading. Let's look at these commands
in detail.

Adjusting the depth of the TOC
These are the standard sectioning commands and their so called TOC level:
Command

Level

\part

-1 (book and report class)

\chapter

0 (not available in article class)

\section

1

\subsection

2

\subsubsection

3

\paragraph

4

\subparagraph

5

In the book and in the report class, LaTeX creates TOC entries until level 2, in
t h e article class until level 3. In a book, this means, for example,
\subsubsection doesn't generate a TOC entry. There's a variable representing
the level, namely, \tocdepth. It's an integer variable which we call a counter .
To tell LaTeX to include subsubsections in the TOC, we would have to raise this
counter. There are two basic ways to adjust a counter value:
\setcounter{name}{n}

specifies an integer value of n for the counter name

\addtocounter{name}{n} adds
counter name. n may be negative

the integer value of n to value of the

Thus, the following command would ensure that even \subparagraph gives a
TOC entry:
\setcounter{tocdepth}{5}

Using \addcounter instead you may raise or lower the level without knowing its
number.

Note
In contrast to commands, counter names don't begin with a backslash.

Shortening entries
As you have already learned in Chapter 3, you may choose a text for the TOC
that's different to the heading in the body text. Each sectioning command
understands an optional argument for the TOC entry, which is especially useful if
you wish to use very long headings, but a shorter TOC entry would be sufficient.
In our example, we did this by:
\subsubsection[Deeper level]{This section has an even deeper
level}

The body text shows the long heading while the TOC shows the short one.
Running titles in the headings would use the short entry as well, as the space in
headings is very limited.

Adding entries manually
Starred commands like \chapter* and \section* don't produce a TOC entry. In
our example, we did that manually using this command:
\addcontentsline{file extension}{sectional unit}{text}

This command can be used in several contexts: the file
toc

for the table of contents file

lof

for the list of figures file

lot

for the list of tables file

extension

may be:

Or, any another extension of such a file type known to LaTeX.
The sectional unit determines the formatting of the entry. Specify chapter to
create an entry that's formatted like normal chapter entries and similar for other
sectional units like part, section, or subsection.
The third argument contains the text for the entry. This text may contain
commands. Do you remember the \protect command? For a macro inside the
entries, it's usually a good idea to write \protect right before it.
You may insert text or commands more directly with:
\addtocontents{file extension}{entry}

In contrary to \addcontentsline, the argument entry is written directly to the
file without any additional formatting. You may choose any formatting you like.
This command may even be used for some customization, for example:
\addtocontents{toc}{\protect\enlargethispage{\baselineskip}}

extends the text height such that one additional line fits to the contents page.
causes a page break in the
TOC. For instance, if the automatic page break happens after a chapter entry
and before the following section entries, you might wish to force a page
break already before the chapter entry.
\addtocontents{toc}{\protect\newpage}

changes the
page style of the current TOC page to fancy. As the first page of a chapter is
of plain style by default, the first page of the TOC would be plain as well,
even if you specified \pagestyle{fancy}. This command overrides it.
\addtocontents{toc}{\protect\thispagestyle{fancy}}

Place such commands where they should be effective. To affect the first TOC
page, place it at the beginning of your document. To cause a page break before a
certain chapter, place it right before the corresponding \chapter call.

Creating and customizing lists of
figures
As briefly mentioned in the previous chapter, the two commands for creating lists
of figures and tables are \listoffigures and \listoftables . Depending on
the class, they produce a fine list of all captions together with the figure
respectively the table number and the corresponding page numbers. As with the
TOC, all can be done automatically. However, we may use the same techniques,
like with the TOC, to customize the other lists. Let's try that.

Time for action – creating a list of
diagrams
Suppose all of our figures are diagrams. We shall avoid the term figure and we
will typeset a list of diagrams:
1. Open our current example. Add these lines to your preamble:
\renewcommand{\figurename}{Diagram}
\renewcommand{\listfigurename}{List of Diagrams}

2. Right after \tableofcontents, add:
\listoffigures

3. Add a diagram somewhere in Chapter 1:
\begin{figure}
\centering
\fbox{Diagram placeholder}
\caption{Enterprize Organizational Chart}
\end{figure}

4. In the second part, in the third chapter, we'd like to add network design
diagrams. Let's mark that in the LOF and let the diagrams follow:
\addtocontents{lof}{Network Diagrams:}
\begin{figure}
\centering
\fbox{Diagram placeholder}
\caption{Network overview}
\end{figure}
\begin{figure}
\centering
\fbox{Diagram placeholder}
\caption{WLAN Design}
\end{figure}

5. Typeset twice to get the document and the list:

What just happened?
We renamed the figures and the list heading by redefining LaTeX macros. At the
end of the chapter, you will get a list of names used by LaTeX classes that you
may redefine.
Like with the TOC, we used the command \addtocontents ; this time to insert a
bold heading into the .lof file where LaTeX collects the captions. It works
similar to the TOC.

Creating a list of tables
You already know all you need to create and customize a list of tables! The file,
where LaTeX collects the captions of the tables, has the extension .lot, and that's
why the first argument of \addtocontents would be lot. Everything works
analogous, like \listoftables, \tablename, and \listtablename.

Using packages for customization
Besides the described simple methods, there are packages providing
sophisticated features for customizing the table of contents and the lists of figures
and tables:
tocloft gives extensive control over the typography of TOC, LOF, and LOT.
You may even define new kinds of such lists.
titletoc offers convenient handling of entries and is the companion to
titlesec, a very good package for customizing sectioning headings.
multitoc offers a layout in two or more columns using the multicol package.
minitoc can create small TOCs for each part, chapter, or section.
tocbibind can automatically add bibliography, index, TOC, LOF, and LOT to
the table of contents. It's even capable of using numbered headings instead of
the default unnumbered ones.
Let's continue with the announced index and the bibliography.

Generating an index
Extensive documents often contain an index. It is a list of words or phrases and
page numbers pointing to where related material can be found in the document. In
contrary to a full-text search feature, the index provides selective pointers to
relevant information.
While it's our turn to identify and to mark the words for the index, LaTeX will
collect this information and is able to typeset the index for us.

Time for action – marking words and
building the index
Suppose our example would contain information about an enterprise and its
structure and further about its network structure and design. We shall mark places
in the text where these concepts occur. Finally, we will order LaTeX to typeset
the index:
1. Go back to our example. In the preamble, load the index package and add the
command to create the index:
\usepackage{index}
\makeindex

2. In the caption of our enterprise diagram, index this point with the keyword
enterprise:
\caption{\index{enterprise}Enterprise Organizational Chart}

3. In the third chapter, which contains our diagrams, index by the keyword
network:
\index{network}

4. Directly before \end{document}, create an entry for the index for the table
of contents. To ensure that it shows the correct page number, end the page
before:
\clearpage
\addcontentsline{toc}{chapter}{Index}

5. In the next line, order LaTeX to typeset the index:
\printindex

6. If you're using TeXworks, choose MakeIndex instead of pdfLaTeX in the
drop-down box next to the typeset button. Then click typeset. If you use
another editor, use its MakeIndex feature or type the following at the
command prompt in the document directory:
makeindex documentname

7. Switch back to pdfLaTeX, Typeset, and look at the last page:

What just happened?
We loaded the index package, which improves LaTeX's built-in indexing
capabilities. Alternatively, you could use the makeidx package which is part of
standard LaTeX. The command \makeindex prepared the index. Both commands
must be put into the preamble.
The command \index takes just one argument, namely, the word or the phrase to
be indexed. This will be written to a file with the extension .idx. If you look into
this file, you will find lines like the following:
\indexentry {enterprise}{9}
\indexentry {network}{15}

These stand for the index entries and the corresponding page numbers.
The external program makeindex takes that .idx file and produces an .ind file.
The latter consists of LaTeX code for the index creation. Specifically, it contains
the index list environment together with the items and looks like the following:
\begin{theindex}
\item enterprise, 9
\indexspace
\item network, 15
\end{theindex}

More complex indexes may contain subitems, page ranges, and references to other
items. Let's see how to produce such an index.

Defining index entries and subentries
We already created simple index entries with:
\index{phrase}

Subentries are produced by specifying the main entry followed by the subentry,
separated by an exclamation mark, for example:
\index{network!overview}

Also subentries may have subentries; just use another ! symbol, for example:
\index{enterprise!organization}
\index{enterprise!organization!sales}
\index{enterprise!organization!controlling}
\index{enterprise!organization!operation}

Up to three levels are possible.

Specifying page ranges
If several pages deal with the same concept, you may specify a page range for the
index entry. Suffix the entry with |( where the range starts and add |) where it
ends, as follows:
\index{network|(}

at the beginning of the network chapter and:
\index{network|)}

at the end of this chapter. This results in an entry of the form Network, 15-17.

Using symbols and macros in the index
sorts the entries alphabetically. If you would like to include symbols
in the index, for example, Greek letters, chemical formulas, or math symbols, you
could face the problem of integrating them into the sorting. For this purpose,
\index understands a sort key. Use this key as prefix for the entry, separated by
an @ symbol, for instance:
makeindex

\index{Gamma@$\Gamma$}

Using macros for index entries is generally not really recommended. Macros
would be sorted according to the macro name including the backslash, though it
would be expanded when the index is typeset. Imagine, you've got a macro
\group standing for TeX Users Group. If you write the following, then the entry
TeX Users Group would be treated like \group in the sorting and won't appear
among the entries beginning with T:
\index{\group}

However, you could repair such issues by adding a sort key:
\index{TeX@\group}
\index{schon@sch\"{o}n}

What if you wish to use the symbols |, @, and ! within index entries? Quote them
by using a preceding ":
\index{exclamation ("!)!loud}

And don't worry, " itself may be used within verbatim text as follows:
\index{quote (\verb|""|)}

Referring to other index entries
Different words may stand for the same concept. For such cases, it's possible to
add a cross-reference to the main phrase without a page number. Adding the code
|see{entry list} achieves that, for example:
\index{network|see{WLAN}}
\index{WLAN}

As such references don't print a page number, their position in the text doesn't
matter. You could collect them in one place of your document.

Fine-tuning page numbers
If an index entry refers to several pages, you might want to emphasize one page
number to indicate it as the primary reference. You could define a command for
emphasizing as follows:
\newcommand{\main}[1]{\emph{#1}}

And for the index entry, add a pipe symbol and the command name:
\index{WLAN|main}

Thus, the corresponding page number would be emphasized. Simply writing
\index{WLAN|emph} or \index{WLAN|texbf} is possible as well. However,
defining your own macro is more consistent—remember the concept of separating
form and content.

Designing the index layout
If we extend our example document with the aforementioned example commands,
\printindex gives us this layout, containing subentries, ranges, references, and
emphasized entries:

LaTeX provides some index styles called latex (the default), gind, din, and iso.
To use another style, specify it using the –s option of the makeindex program, for
example:
makeindex –s iso documentname

If you typeset after this call, the index layout would be changed to:

You could even define your own styles. To learn more about indexing and
makeindex, use texdoc at the command prompt:
texdoc index
texdoc makeindex

Tip
Though it seems natural to generate the index while writing the document, this
might lead to inconsistencies in the index. It's recommendable first to finish

writing and afterwards to work out what should appear in the index.

Creating a bibliography
Especially in scientific documents, a list of references or bibliography is very
common. We shall work out how to typeset a bibliography and how to refer to its
entries.

Time for action – citing texts and
listing the references
Using LaTeX's standard features, we shall create a small list of references
containing a book and an article by Donald E. Knuth, the creator of TeX. In our
body text, we will refer to both:
1. Create a new document as follows:
\documentclass{article}
\begin{document}
\section*{Recommended texts}
To study \TeX\ in depth, see \cite{DK86}. For writing math
texts, see \cite{DK89}.
\begin{thebibliography}{8}
\bibitem{DK86} D.E. Knuth, \emph{The {\TeX}book}, 1986
\bibitem{DK89} D.E. Knuth, \emph{Typesetting Concrete
Mathematics}, 1989
\end{thebibliography}
\end{document}

2. Typeset and view the output:

What just happened?
We used an environment called thebibliography to typeset the list of references,
which is similar to a description list as we've seen in Chapter 4, Creating Lists.
Each item of this list has got a key. For citing in the body text, we referred to that
key using the \cite command.
Let's look at these commands in detail.

Using the standard bibliography environment
LaTeX's standard environment for bibliographies has the following form:
\begin{thebibliography}{widest label}
\bibitem[label]{key} author, title, year etc.
\bibitem…
…
\end{thebibliography}

Each item is specified using the command \bibitem. This command requires a
mandatory argument determining the key. We may simply refer to this key by
\cite{key} or \cite{key1,key2}. \cite accepts an optional argument stating
a page range, for example, \cite[p.\,18--20]{key}. You may choose a label
by the optional argument of \bibitem. If no label has been given, LaTeX will
number the items consecutively in square brackets, as we've seen.
Using labels, the environment could look as follows:
\begin{thebibliography}{Knuth89}
\bibitem[Knuth86]{DK86} D.E. Knuth, \emph{The {\TeX}book}, 1986
\bibitem[Knuth89]{DK89} D.E. Knuth, \emph{Typesetting Concrete
Mathematics}, 1989
\end{thebibliography}

And the output would then be as follows:

As you can see, the output of \cite has been automatically adjusted to the new
labels.

Tip
The cite package offers compressed and sorted lists of numerical citations, like
[2,4-6], and further formatting options for in-text citations.
The mandatory item of the environment should contain the widest label for the
alignment of the items. So, for instance, if you have more than 9 but fewer than a
100 items, you may write two digits into the argument.

Using bibliography databases with BibTeX
Manually creating the bibliography is laborious. Especially, if you use references
in several documents, it would be beneficial to use a database and let a program
create the bibliography for you. This sounds more complicated than it really is.
We will try this.

Time for action – creating and using a
BibTeX database
We shall create a separate database file containing the references of our previous
example. We will modify our example to use that database. To make this database
usable, we have to call the external program BibTeX :
1. Create a new document. Begin with writing the entry for the TeXbook:
@book{DK86,
author = "D.E. Knuth",
title = "The {\TeX}book",
publisher = "Addison Wesley",
year = 1986
}

2. For the next entry, that is, the article, we will specify even more fields:
@article{DK89,
author = "D.E. Knuth",
title = "Typesetting Concrete Mathematics",
journal = "TUGboat",
volume = 10,
number = 1,
pages = "31--36",
month = apr,
year = 1989
}

3. Save the file and give it the name tex.bib. Open our example tex document
and modify it as follows:
\documentclass{article}
\begin{document}
\section*{Recommended texts}
To study \TeX\ in depth, see \cite{DK86}. For writing math
texts, see \cite{DK89}.
\bibliographystyle{alpha}
\bibliography{tex}
\end{document}

4. Typeset one time with pdfLaTeX.
5. If you're using TeXworks, choose BibTeX instead of pdfLaTeX, present in
the drop-down box next to the Typeset button, and then click on Typeset. If
you write with another editor, use its BibTeX option or type at the command
prompt in the document directory as follows:
bibtex documentname

6. Typeset twice with pdfLaTeX. Here's the result:

What just happened?
We created a text file containing all bibliography entries. In the next section, we
shall look at its format in depth. In our document, we chose a style called alpha ,
which sorts entries according to the author's name and uses a shortcut consisting
of author and year as label. Then we told LaTeX to load the bibliography file
called tex. The extension .bib has been automatically added.
Afterwards, we called the external program BibTeX. This program knows from
the example .tex file that tex.bib has to be translated. Thus out of this .bib
file, it creates a .bbl file, which contains a LaTeX thebibliography
environment and the final entries.
Finally, we had to Typeset twice, to ensure that all cross-references are correct.
Though we need some more steps to generate the bibliography, there are benefits:
we don't need to fine-tune each entry. We can easily switch between styles. We
can then reuse the .bib file.
So, let's look at the .bib file format. It supports various entry types like book and
article. Furthermore, these entries contain fields like author, title, and year.
Let's first look at the supported fields, and afterwards we shall talk about the
different kinds of entries.

Looking at the BibTeX entry fields
Here's a list of the standard fields. Some fields are common, some are rarely used
—we just list them in alphabetical order, following the BibTeX documentation:
address

Usually the address of the publisher. At least for small publishers, this information might be
useful.

annote

An annotation, not used by the standard bibliography styles. Other styles or macros might
use this.

author

The name(s) of the author(s).

booktitle

The title of a book, if you cite a part of that. For a book, use the title field instead.

chapter

A chapter number.

crossref

The key of the database entry being cross referenced.

edition

The edition (First, Second, and alike) of a book. Commonly it's capitalized.

editor

The name(s) of the editor(s).

howpublished The way of publishing, especially if it's unusual. Capitalize the first word.
institution

Could be a sponsoring institution.

journal

A journal name; you may use common abbreviations.

key

Used for alphabetizing, cross-referencing, and labeling if the author information is missing.
Don't confuse it with the key used in the \cite command which corresponds to the
beginning of the entry.

month

The month in which the work was published or written if it's not yet published. Usually a
three letter abbreviation is used.

note

Any additional useful information. Capitalize the first word.

number

The number of a journal or another kind of work in a series.

organization

Might be a sponsoring organization.

pages

A page number or range of page numbers, like 12-18 or 22+.

publisher

The name of the publisher.

school

Could be the name of the school where the document was written.

series

The name of a series of books or its number of a multi-volume set.

title

The title of the work.

type

The type of the publication.

volume

The volume of a journal or multi-volume book.

year

The year of the publication or the year when it was written if it hasn't been published yet.
Commonly four numerals are used, such as 2010.

You may use any fields possibly supported by other styles and ignored by
standard styles.

Note
Referring to Internet resources
Today we often refer to online sources. To put Internet addresses into BibTeX
fields, use the \url command of the url or hyperref package, for example,
howpublished = {\url{http://texblog.net}}. Some styles offer a url field.

Understanding BibTeX entry types
Firstly, you decide which type of entry you want to add and then you fill in the
fields. Different types may support different fields. Some fields are required,
some are optional and may be omitted, and some are simply ignored.
Usually the name of the entry tells you its meaning. These are the standard entry
types and their required and optional fields, according to the BibTeX reference:
Type

Required fields

Optional fields

author, title, journal,
year

volume, number, pages, month, note

or editor, title,
publisher, year

volume

booklet

title

author, howpublished, address, month, year, note

conference

author, title, booktitle,
year

editor, volume

manual

title

author, organization, address, edition, month, year, note

mastersthesis

author, title, school, year type, address, month, note

misc

none

phdthesis

author, title, school, year type, address, month, note

proceedings

title, year

editor, volume or number, series, address, month,
organization, publisher, note

techreport

author, title,
institution, year

type, number, address, month, note

unpublished

author, title, note

month, year

article

book

author

or number, series, address, edition, month, note

or number, series, pages, address, month,
organization, publisher, note

author, title, howpublished, month, year, note

Have a look at the BibTeX reference for more details:
texdoc bibtex

If no other entry fits, choose misc. It doesn't matter if you use capitals or small
letters; @ARTICLE is understood the same as @article. As the example showed,
entries have the following form:
@entrytype{keyword,
fieldname = {field text},
fieldname = {field text},
…
}

Use braces around the field

text.

Straight quotes instead like in "field

text"

are supported as well. For numbers, you may omit the braces.

Tip
Some styles change the capitalization, which might lead to undesired lowercase
letters. To protect letters or words from becoming lowercase, put additional
braces around them. Preferably around a word instead of just the letter to keep
ligatures and kerning improvement, that is, {WAL} looks better than {W}AL as TeX
moves an A closer to a preceding W. Separating braces hampers TeX's microtypographic improvements.

Choosing the bibliography style
Standard styles are as follows:
plain

Arabic numbers for the labels, sorted according to the names of the authors. The number is written in
square brackets, which also appear with \cite.

unsrt No sorting. All entries appear like they were cited in the text, otherwise it looks like plain.
alpha

Sorting is according to the names of the authors, the labels are shortcuts made of the authors name
and the year of publication. Also here, square brackets are used.

abbrv Like plain, but first names and other field entries are abbreviated.

The style should be chosen after \begin{document} and before \bibliography.
You may write \bibliographystyle right before \bibliography to keep it
together.
There are more styles available in TeX distributions and on the Internet. For
instance, the natbib package provides styles and the capability to cite in a nice
author-year scheme. This package further adds some fields, like ISBN, ISSN, and
URL.

Tip
The package biblatex provides a complete reimplementation of the bibliographic
features offered by BibTeX and LaTeX. Without learning the BibTeX language,
you could create new styles just by using your LaTeX knowledge.

Listing references without citing
BibTeX takes only those references from the database that are cited in the text,
and prints them out. However, you may specify keys for references, which should
appear nevertheless the following for a single reference. Just write the following
for a single reference:
\nocite{key}

Or write the following to list the complete database
\nocite{*}

Changing the headings
Like in our diagram example, if you don't like the heading Contents, you could
easily change it. The text of the heading is stored in the text macro
\contentsname. So, just redefine it as follows:
\renewcommand{\contentsname}{Table of Contents}

Here's a list of all those headings:
List

Heading command

Default heading

Table of contents

\contentsname

Contents

List of figures

\listfigurename

List of figures

List of tables

\listtablename

List of tables

\bibname

in book and report Bibliography in book and report

\refname

in article

Bibliography
Index

\indexname

References in article
Index

Furthermore, as promised, here's a list of further macros for names used by
LaTeX:
Name

Command

Default value

figure

\figurename

Figure

table

\tablename

Table

part

\partname

Part

chapter

\chaptername

Chapter

abstract

\abstractname

Abstract

appendix

\appendixname

Appendix

Not really surprising! Using name macros is especially useful when another
language will be used. For instance, the babel package takes a language option
and redefines all those name macros according to those languages.
However, they are also useful to choose abbreviations like Fig. or Appendices
instead of Appendix.

Have a go hero – using natbib
Load the already mentioned natbib package and try its new styles plainnat,
abbrvnat, and unsrtnat, for instance:
\usepackage{natbib}
\bibliographystyle{plainnat}

Our example would change as follows:

reimplemented the \cite command and offers variations to it, with the
main purpose of supporting author-year citations. It works with most other
available styles. natbib introduces the citation command \citet for textual
citations and the command \citep for parenthetical citations. There are starred
variants printing the full author list, and optional arguments allow adding text
before and after.
natbib

Check out the documentation if you would like to benefit from this fine package.

Pop quiz
1. Which of these commands should be used for adding unnumbered headings to
the table of contents?
a. \addtocontents
b. \addcontentsline
c. \contentsname
2. Which of the following indexing commands has to come before
\begin{document}?
a. \index
b. \printindex
c. \makeindex
3. Which command generates the bibliography?
a. \bibliography
b. \bibliographystyle
c. \bibitem

Summary
In this chapter, we dealt with many kinds of lists. Specifically, we learned about:
Generating and customizing the table of contents and lists of figures and
tables
Producing an index pointing to relevant information for keywords and
phrases
Creating bibliographies, both manually and using a bibliography database
These lists are intended to guide the reader to the information he's/she's looking
for; they aren't just for listing and summarizing. That's why the headings of the list
of figures and the list of tables usually don't appear in the TOC, as they commonly
directly follow the TOC. Sometimes there's even the strange requirement to list
the table of contents within itself. If you are not sure with a design or a
requirement, have a look at a good book in your special field, to see how
exemplary tables of contents, lists, and indexes might look.
In the next chapter, we shall look at scientific writing in depth.

Chapter 8. Typing Math Formulas
Chapter 1, Getting started with LaTeX, claimed that LaTeX offers excellent
quality for mathematical typesetting. Now it's time to prove this.
To make the most of LaTeX's capabilities, we shall learn how to perform the
following:
Writing basic formulas
Embedding formulas within text and text within formulas
Centering and numbering equations
Aligning multi-line equations
Typesetting math symbols such as roots, operators, Greek letters, and arrows
Building fractions
Stacking expressions
Building matrices
That's a great undertaking—let's tackle it!

Writing basic formulas
LaTeX knows three general modes:
The paragraph mode: The text is typeset as a sequence of words in lines,
paragraphs, and pages. That's what we used until now.
The left-to-right mode: The text is also considered to be a sequence of
words, but LaTeX typesets it from left to right without breaking the line. For
instance, the argument of \mbox will be typeset in this mode; that's why
\mbox prevents hyphenation.
The math mode: Letters are treated as math symbols. That's why they're
typeset in italic shape, which is common for variables. A lot of symbols can
be used, most of them exclusively in this mode. Such symbols are roots, sum
signs, relation signs, math accents, arrows, and various delimiters like
brackets and braces. Space characters between letters and symbols are
ignored. Instead, the spacing depends on the type of symbols—distances to
relation signs are different from distances to opening or closing delimiters.
This mode is required for all math expressions.
Now we shall enter the math mode for the first time.

Time for action – discussing
quadratic equations and roots
Our first small math text shall deal with the solutions of quadratic equations. We
need to typeset formulas with constants, variables, superscripts for the square,
and subscripts for the solutions. The solution itself needs a root symbol. Finally,
we will use cross-references to formulas.
1. Start a new document. For now, we don't need any package:
\documentclass{article}
\begin{document}
\section*{Quadratic equations}

2. State the quadratic equation with its conditions. Use an equation
environment for it. Surround small pieces of math within text by \( … \):
The quadratic equation
\begin{equation}
\label{quad}
ax^2 + bx + c = 0,
\end{equation}
where \( a, b \) and \( c \) are constants and \( a \neq 0
\),
has two solutions for the variable \( x \):

3. Use another equation for the solutions. The command for the square root is
\sqrt. The command for a fraction is \frac:
\begin{equation}
\label{root}
x_{1,2} = \frac{-b \pm \sqrt{b^2-4ac}}{2a}.
\end{equation}

4. Let's introduce the discriminant and discuss the case zero. To get an
unnumbered displayed equation, we surround the formula with \[ … \]:
If the \emph{discrimimant} \( \Delta \) with
\[
\Delta = b^2 - 4ac
\]
is zero, then the equation (\ref{quad}) has a double
solution: (\ref{root}) becomes
\[
x = - \frac{b}{2a}.
\]
\end{document}

5. Typeset twice and look at the result:

What just happened?
Just as we said in Chapter 1, writing formulas also looks a lot like programming.
However, this chapter will help you master it, and the results are worth the effort.
The formulas have been built with commands: there are commands with
arguments, like for roots and fractions, and simple commands for symbols, like
for the Greek letter. Most symbols have to be within a math environment and don't
simply work within normal text.
The equation environment created a displayed formula; that formula has been
horizontally centered, some vertical space has been added before and after,
further these formulas are consecutively numbered.
However,
out.

\[ … \]

and \(

… \)

are also, in truth, environments. Let's sort this

Embedding math expressions within text
LaTeX provides the math environment in-text formulas:
\begin{math}
expression
\end{math}

Since it's very laborious to write this environment for each small expression or
symbol, LaTeX offers an alias that's doing the same:
\(
expression
\)

You may write it without line breaks, such as \(expression\).
A third way is by using a shortcut, coming from TeX:
$expression$

A disadvantage of the latter is that the commands for beginning and ending the
math environment are the same, which may easily lead to errors. However, it's
much easier to type, which may be the reason why it's still popular among LaTeX
users.

Tip
might cause problems in moving arguments like in headings. To prevent
any such problems, just load the package fixltx2e , which fixes this as well as
other issues.
\( … \)

Displaying formulas
For displayed formulas, which have to be centered, LaTeX offers the
displaymath environment:
\begin{displaymath}
expression
\end{displaymath}

The effect of this environment is that the paragraph will be ended, some vertical
space follows, then the centered formula plus the following vertical space. As
this math environment takes care of the spacing, don't leave empty lines before
and after it! This would cause additional vertical space because of the
superfluous paragraph breaks.
Also for this environment there's a shortcut. We already used it:
\[
expression
\]

In this case, putting the shortcuts \[ and \] on separate lines commonly improves
the readability as the formula is also kind of displayed in the source code.
There's also a TeX low level command:
$$
expression
$$

However, it's strongly recommended to use
environment handles vertical spacing better.

\[

…

\],

because this LaTeX

For the rest of this chapter, all pieces of code use math mode. Either we explicitly
use a math environment or we imagine that we are already in math mode, for short
pieces of code.

Numbering equations
Equations and formulas in general may be numbered. However, this applies only
to displayed formulas. The equation environment is responsible for this:
\begin{equation}
\label{key}
expression
\end{equation}

It looks similar to displaymath but numbered this time. The number will be
displayed in parentheses on the right side of the equation.

Adding subscripts and superscripts
As exponents and indexes are frequently used, there are very short commands for
typesetting them, for example, _ gives an index or subscript, ^ produces an
exponent or superscript. Use braces to mark the concerned expression. So, the
common forms are as follows:
{expression}_{subscript}

And:
{expression}^{superscript}

This may be nested. If you use subscripts and superscripts to the same expression,
the order of ^ and _ is not important. In the case of single letters, numerals, or
symbols, you can omit the braces. Let's look at an example:
\[ x_1^2 + x_2^2 = 1, \quad 2^{2^x} = 64 \]

Extracting roots
Our example contained a square root: \sqrt{value}. As there are roots of higher
order, this command accepts an optional argument for the order. The complete
definition is:
\sqrt[order]{value}

The size of the root symbol will be automatically adjusted to the height and the
width of the value expression. Roots may be nested. Both can be seen in this
example:
\sqrt[64]{x} = \sqrt{\sqrt{\sqrt{\sqrt{\sqrt{\sqrt{x}}}}}}

Writing fractions
Within text formulas, you may just write / to denote fractions, such as \(
(a+b)/2 \). For larger fractions, there's the \frac command:
\frac{numerator}{denumerator}

This command may also be used for in-text formulas. However, the line spacing
could increase.
\[ \frac{n(n+1)}{2}, \quad \frac{\frac{\sqrt{x}+1}{2}-x}{y^2} \]

Greek letters
Mathematicians like to use Greek letters, for instance, to denote constants. To get
a lowercase Greek letter, just write the name with a backslash for the command.
Here are the lowercase Greek letters with their corresponding LaTeX commands:

For some letters, variants are available:

As the omicron just looks like an o, there's no command for it. It's similar for most
uppercase Greek letters, which are equal to Roman letters. The remaining
uppercase Greek letters are produced as follows:

For upright Greek letters, you may use the upgreek package.

Script letters
For the twenty-six uppercase letters A, B, C, … , Z, there's a calligraphic shape,
produced by \mathcal:
\[ \mathcal{A}, \mathcal{B}, \mathcal{C}, \ldots, \mathcal{Z} \]

Note
There are packages offering different calligraphic fonts, such as zapfino and xits.

Producing an ellipsis
You already know \ldots for a low ellipsis. It also works in math mode. We use
the low ellipsis mainly between letters and commas. Between operation and
relation symbols, a centered ellipsis is commonly used. Furthermore, a matrix
may require a vertical ellipsis. Here's how all of them may be produced:

Comparing in-line formulas to displayed formulas
Writing formulas in-line saves space and allows fluent explanations. This is
recommendable for short math expressions within text.
Formulas in the displayed style are outstanding; they are centered and require
more space. Furthermore, they can be numbered and you may refer to them using
the \label and \ref techniques that you learned in Chapter 5, Creating Tables
and Inserting Pictures.
Choose the style that is optimal for the readability of your text.

Changing the font, style, and size
I n Chapter 2, Formatting Words, Lines, and Paragraphs , we learned how to
modify the font of common text. There are further commands changing the font
style in math mode:

Though letters in math mode are italic, they are considered to be separate
symbols, which results in a different spacing than that of an italic word. For
instance, in math mode, fi may be the product of the variables f and i, but not the
ligature fi. Compare:
\textit{Definition}, \(Definition\)

Also, \mathit treats the argument as text in italic math font. So, for text within
formulas, use a text or math font command, or even better: use \text{…} of
amsmath—we will return to this very soon.
If you wish to switch to bold typeface for a complete math expression, you can
use the declaration \boldmath before the expression, that is, already outside math
mode. The declaration \unboldmath switches back to the normal typeface. The
latter has to be used outside math mode as well.
To make parts of a formula bold, you can switch to the left-to-right mode by
\mbox and by using \boldmath in its argument.
Four math styles are available, determining the way of typesetting and the font
size:
Style

Command

Meaning

display

\displaystyle

Default for letters and symbols in displayed formulas

text

\textstyle

Default for letters and symbols within in-text formulas

script

\scriptstyle

Smaller font size, used for subscripts and for superscripts

scriptscript

\scriptscriptstyle

Even smaller font size, for nested script style

T h e textstyle differs from the displaystyle in mainly two ways; in
textstyle, variable sized symbols are smaller and subscripts and superscripts
are usually placed beside the expression instead of below and above,
respectively. Otherwise the font size is the same.
LaTeX switches the style automatically; if you write a simple exponent, it will be
typeset in script style, that is, with a smaller font size.
You may force a desired style using one of the commands in the aforementioned
table. This allows you, for instance, to:
Type formulas within the text exactly like they would appear in a displayed
formula: bigger fraction, bigger sum signs, further subscripts are set below,
and superscripts are set above. Note, all of this increases the line spacing.
Write exponents or indexes with bigger symbols.

Customizing displayed formulas
There are two options that modify the way the formulas are displayed:
fleqn

"left equation
numbers"

This causes all displayed formulas to be aligned at the left margin.

leqno

"left equation
numbers"

All numbered formulas would get the numbers on the left side instead of the
right.

Often, formulas are not displayed just standalone. We may encounter situations
where:
A formula is too long to fit on one line
Several formulas are listed row-by-row
An equation shall be transformed step-by-step
A chain of inequalities spans over more than one line
Several formulas shall be aligned at relation symbols
We may also encounter similar situations, where we have to write multi-line
formulas, often with some kind of alignment. The amsmath package offers
specialized environments for nearly every such need.

Time for action – typesetting multiline formulas
We shall use the amsmath package to experiment with a very long formula and
with a system of equations:
1. Start a new document on an A6 paper and load the amsmath package.
\documentclass{article}
\usepackage[a6paper]{geometry}
\usepackage{amsmath}
\begin{document}

2. Use the multline environment to span a long equation over three lines. End
the lines with \\:
\begin{multline}
\sum = a + b + c + d + e \\
+ f + g + h + i + j + \\
+ k + l + m + n
\end{multline}
\end{document}

3. Typeset and look at the formula:

4. Now we handle a system of equations. Use the gather environment to add
these equations. Again, end lines with \\:
\begin{gather}
x + y + z = 0 \\
y - z = 1
\end{gather}

5. Typeset and look at the equations:

6. Commonly, equation systems are aligned at the equal sign. Let's do this. Use
the & symbol to mark the point that we wish to align:
\begin{align}
x + y + z &= 0 \\
y - z&= 1
\end{align}

7. Typeset; now the equations are aligned as desired:

What just happened?
Because we loaded the amsmath package, we have access to several multi-line
math environments.
Each line in such an environment is ended by \\, except the last one. The alignment
depends on the environment, as we've seen:
multline: The first line is left-aligned, the last line right-aligned, and all
other lines in between are centered
gather: All lines are centered
align: The lines are aligned at marked relation signs
Let's have a closer look.

Aligning multi-line equations
Here's a list of the amsmath multi-line environments:
Name

Meaning

multline

First line is left-aligned, last line is right-aligned, all others are centered.

gather

Each line is centered.

align

Use & to mark a symbol where the formulas shall be aligned. Use another & to end a
column, if you need several aligned columns.

flalign

Similar to align with more than one column, but the columns are flushed to the left
and the right margin, respectively.

alignat

Alignment at several places, each has to be marked by &.

split

Similar to align, but within another math environment, thus unnumbered.

aligned, gathered,
alignedat

Used for an aligned block within a math environment. This can be displayed math or
in-line math.

Numbering rows in multi-line formulas
In multi-line math environments, each line would be numbered like a normal
equation. If you wish to suppress the numbering of a line, write \notag before the
end of the line.
Use the starred variant like align*, or gather*, if you would like to avoid
numbering completely.

Inserting text into formulas
To insert some text into a formula, standard LaTeX provides the
amsmath offers further commands:

\mbox

command.

inserts text within a math formula. The size is adjusted
according to the current math style, that is, \text produces smaller text
within subscripts or superscripts.
\text{words}

suspends the formula, the text follows in a separate
paragraph, then the multi-line formula is resumed, keeping the alignment. Use
it for longer text.
\intertext{text}

These commands are the best choice, when you would like to use text within math
environments.

Fine-tuning formulas
If we go beyond writing variables and basic math operators, we may need many
symbols for special purposes: certain relation signs, unary and binary operators,
function-like operators, sum and integral symbols and variants of the latter,
arrows, and many more. LaTeX and additional packages offer thousands of
symbols for many purposes. We shall have a look at some of them. But firstly,
let's figure out how to write functions.

Using operators
Trigonometric functions, logarithm functions, and other analytic and algebraic
functions are commonly written with upright Roman letters. Simply typing log
would otherwise look like a product of the three variables, namely, l, o, and g. To
ease the input, there are commands for many common functions or so called
operators. Here's an alphabetical list of the predefined ones:
\arccos, \arcsin, \arctan, \arg, \cos, \cosh, \cot, \coth, \scs, \deg,
\det, \dim, \exp, \gcd, \hom, \inf, \ker, \lg, \lim, \liminf, \limsup, \ln,
\log, \max, \min, \Pr, \sec, \sin, \sinh, \sup, \tan, \tanh

The modulo function may be written in two ways, either by using \bmod for a
binary relation or by using \pmod{argument} for a modulo expression in
parentheses.
Some operators support subscripts which are set as follows:
\[ \lim_{n=1, 2, \ldots} a_n, \qquad \max_{x

---

Source Exif Data:

File Type                       : PDF
File Type Extension             : pdf
MIME Type                       : application/pdf
PDF Version                     : 1.4
Linearized                      : No
Author                          : Stefan Kottwitz
Create Date                     : 2016:07:28 16:10:08+00:00
Producer                        : calibre 1.48.0 [http://calibre-ebook.com]
Description                     : 
Title                           : LaTeX Beginner's Guide
Publisher                       : Packt Publishing
Subject                         : 
Creator                         : Stefan Kottwitz
Date                            : 2011:03:21 00:00:00+08:00
Language                        : en
Identifier Scheme               : mobi-asin
Identifier                      : B0073FTN5I
Metadata Date                   : 2016:07:29 00:10:08.320000+08:00
Timestamp                       : 2016:07:28 23:48:14.821000+08:00
Title sort                      : LaTeX Beginner's Guide
Author sort                     : Kottwitz, Stefan
Page Count                      : 661

EXIF Metadata provided by EXIF.tools