KOMPENDIUM

Writing science with LATEX

Peter Jansson

Institutionen för naturgeografi October 15, 2019

Contents

1 Scientific writing 5
2 A quick start to writing in LATEX 5
3 Expanding your manuscript 8
3.1 Types of commands . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
3.2 The document structure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
3.3 Document classes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
3.4 Reserved characters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
3.5 Type face styles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
3.6 Typographic features . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
3.7 Breaks and space . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
3.8 Headings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
3.9 Table of contents . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
3.10 Environments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
3.11 cross-referencing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
3.12 Special commands . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
4 Tabular information 17
5 Floats; tables and figures on the run 18
6 Mathematical typesetting 19
7 Extending the functionality of LATEX, packages 21
7.1 inputenc . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
7.2 graphicx . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
7.3 geometry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
7.4 natbib . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
7.5 hyperref . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
7.6 booktabs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
7.7 siunitx . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
7.8 lineno . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
7.9 dcolumn . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
7.10 tikz/pgf . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
7.11 xcolor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28
7.12 caption . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29
7.13 float . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
7.14 amslatex . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
7.15 babel . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
7.16 lipsum . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
8 Managing references 31
8.1 The natbib package . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31
8.2 The fully automated reference system . . . . . . . . . . . . . . . . . . . . . . . . 32
8.3 Integration of Overleaf and Zotero . . . . . . . . . . . . . . . . . . . . . . . . . . 33
8.4 The semi-automated reference system . . . . . . . . . . . . . . . . . . . . . . . . 33
9 Automating LATEX or creating your own commands 34
10 Copying documents from Word to LATEX 35
11 Making presentation slides in LATEX 36
12 Reinventing the wheel? …or not 37
13 How does it really work? 38
14 Saving the worst to last 39
A Running LaTeX on your own computer 41
A.1 LATEX . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41
A.2 LATEX editors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41
A.3 Other software that may be required . . . . . . . . . . . . . . . . . . . . . . . . . 41
A.4 Reference management . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42
A.5 Yet more software that might be of use . . . . . . . . . . . . . . . . . . . . . . . . 42

2
B Install LATEX on your computer 42
B.1 Windows . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42
B.2 Mac . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42
B.3 Linux . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43
C Learning more on LATEX 43
D Revision history 44

Scientific writing with LATEX is copyright © 2019 by Peter Jansson and

available through its doi: http://dx.doi.org/10.17045/sthlmuni.3205753
This work is licensed under the Creative Commons Attribution 4.0
International License. To view a copy of this license, visit
http://creativecommons.org/licenses/by/4.0/

3
Preface
This document is intended to provide you with a good working knowledge of LATEX so that you
can freely use it for your scientific output. LATEX was, and is, created by scientists for scientists
and is widely used by publishers for both journals and books. To have working knowledge of
LATEX should be part of every scientist’s toolbox.
There are two choices for running LATEX, on-line or off-line tools. These should not be
considered mutually exclusive, you may need to work both ways. In this text I will assume you
are working with the on-line platform Overleaf which is what I would also recommend to anyone,
beginner or experienced, since all you need is a web-browser and an internet connection, no
installation on your computer. Overleaf is similar to Google docs in that documents are stored in
the cloud and that several persons can interactively collaborate on a document. The drawback
is of course that you cannot work on your documents without internet access. You can, however,
synchronise your files in the cloud with Dropbox or Git so that you always have updated versions
on your own computer. To work off-line you need to install a LATEX environment locally on
your own computer; this is covered in appendix A.
So I strongly advice the beginner to sign up for and use Overleaf. Note that you may be
covered by an institutional license if you study or work at a university. Please visit the Overleaf
site and look at the introductory video to familiarise yourself with the Overleaf environment.
This document was entirely written on and using Overleaf but does not detail how you use the
interface. Of particular use is the keyboard shortcuts sheet.
Although not covered in any detail in this document, I also strongly urge you to start using
a reference manager for handling your scientific references. Unless you already use a software
I would recommend using the online manager Zotero which integrates with Overleaf. This
integration is briefly covered in section 8.3.
In this text, I will use typewriter type face to highlight LATEX code. In the cases where a
complete section of code is given
it will be presented in a blue field.

Apart from what is covered in this text, you probably should download the LATEX 2ε Cheat
Sheet and the LATEX quick reference and keep them handy. The comprehensive list of LATEX
symbols is also useful to have handy. You should also be aware of the TEX StackExchange site,
which is a question and answer site where probably all your questions are already answered. At
the time of writing there are 181 306 answered questions on the site. You can also find many
solutions to type-setting issues by typing ‘latex’ followed by whatever you are looking for. It is
almost certain some of the highest hits are from the TEX StackExchange.
More details on how different macro packages work can be obtained from their documenta-
tion which is available on the Comprehensive TEX Archive Network (CTAN) site. At the time
of writing there are 5740 macro packages contributed by 2632 different users.

4
1 Scientific writing
Apart from the linguistic aspects of scientific writing, such as clarity, brevity and density,
what distinguishes scientific writing is the use of many technical terms, technical notation and
cross-referencing of citations, figures, tables and equations as well as the need to follow precise
document formatting instructions. Clearly you can accomplish the first aspect with any type of
writing tool from the Goose pen to a modern word processor, it is the second part that requires
some special tools. This is where LATEX fits.
Now LATEX is not the universal response to all writing issues; as with everything there
are definite pro’s and con’s. Since LATEX involves knowing a fair number of commands and
understanding exactly how to use them, it is associated with a learning curve similar to that
of learning a programming language. In fact, it is a programming language (Turing complete).
It is therefore evident that it makes little sense to use LATEX for menial tasks such as writing a
letter or a memo to colleagues or a list for making purchases, that is the realm of Word. LATEX
comes into its own when the document is more complex.
There are probably many different imaginable tasks where LATEX is preferable, I will just
mention a few. When you write a scientific article you greatly benefit from using LATEX because
you can clearly express, for example, complicated notation and mathematics, and it is also
very likely that the journal accepts LATEX manuscripts in their own journal format. LATEX is
also very useful when you try to write a long document, book or similar, where you need cross
referencing, you may need to create a table of contents and perhaps an index; tools built into
LATEX. A third type of document is that which has to be formatted exactly the same time after
time. This could, for example, be a report series from a project or organisation or from ongoing
research. In all such instances, there is little competition.
So the conclusion is that the more complex your writing task the more you will benefit from
using LATEX. Another way to describe this is that what is difficult in Word is simple in LATEX
and vice versa.

2 A quick start to writing in LATEX

In this section I will provide the basics to produce a document in LATEX. In the following
sections, I will go more in-depth with the understanding of how LATEX works. When you read
the remaining part of this section I expect you will follow along and try the examples.
Let us start with a basic example of a document:
\documentclass{article}
\usepackage[utf 8]{ inputenc}

\title{Example}
\author{Peter Jansson}
\date{September 2019}

\begin{document}

\maketitle

\section{Introduction}

\end{document}

In Overleaf you start a new blank project and provide a project name. In my example, I
called it ‘Example’. Overleaf will then open the document in a three pane display as shown
in the figure below. To the left is the project pane which displays what files are included in
the project. By default the file you receive will be named main.tex but this can be changed
to something more meaningful. Next is the LATEX code pane which is where you enter your
text and code to create your document. Finally the is the preview pane which shows how the
document looks.

5
 The simple document we receive in Overleaf shows us several things. First you notice is
that there are commands identifiable by the backslash and often with an argument within
curly braces, such as \documentclass{article}. LATEX produces documents in response to such
commands, similar to how you build a web-page in html. You will with time learn many
commands but it is still a surprising few that are needed to get you far.
If we look at the code we can identify three mandatory commands:
\documentclass{article}

\begin{document}

\end{document}

These must be present in all LATEX documents because they tell LATEX what kind of document
to produce (in this case an article) and where the document text that should be type set begins
and ends. These three lines should always be in any document you start.
In our example we can identify three commands
\title{Example}
\author{Peter Jansson}
\date{September 2019}

entered before the \begin{document}. The first command \title{} allows you to provide a title
for your document and the second \author{} allows you to provide your name to the document.
The thirds allows you to enter a date. In my case Overleaf entered ‘September 2019’ because
this is when I started the project.
The three commands creates the document header we see in the preview pane. However,
they will only display if we enter a fourth command \maketitle which takes the first three and
produces a title for your document. It is possible to remove the \date{} command, \maketitle
will then extract the current date from the computer and display the date instead of the text
provided in the example.
What we now have is the header of the document. We now turn to the content.
In the remaining document text you see the command \section{} which produces your first
heading from whatever you write within the curly braces. Note that headings are numbered
by default. LATEX has predefined the look of the headings so that all you have to do to start a
new section in your text is to use the \section{} command. The heading will always be on a
separate line with spacing above and below the heading.
After the heading you can simply type in the text you want, the text body that belongs to
that heading. When you do so you will probably notice a few unexpected things. If you try
to start a new paragraph by pressing the enter (or return) key ‘Word style’, no new paragraph
will start. Instead the text runs on continuously. This is because LATEX uses an empty line

6
to identify a new paragraph. So, you need one return to end the line and a second to cause a
paragraph break.
Another aspect that can cause confusion is that LATEX determines all relevant spacing in
a document. This means that, for example, if you type in four consecutive spaces LATEX still
interprets this as one space. This also applies if you try to use the tab key to offset text.
It will still result in a single space. So you will not be able to push text around by adding
several spaces. It is actually the same with paragraph breaks. Adding several returns between
paragraphs will still result in a normal paragraph break. We will return later to how we can
create space in a document if we really need to.
A final command that I have not commented on is \usepackage[utf8]{inputenc}. This line
asks LATEX to load a macro package called ‘inputenc’ (input encoding; section 7.1) into your
document. The reason for this is that LATEX by default does not recognise accented letters.
The package allows LATEX to understand accented letters such as the Swedish å, ä and ö. If you
type in accented letters in a LATEX document without having loaded the ‘inputenc’ package,
they will simply not be visible.

Do it
• Add some text of your own and observe what happens if you put many consecutive spaces
or tabs in the text and use the return key to make new paragraphs.

A mostly overused but nevertheless useful feature of a document is the list. A bullet list in
LATEX is referred to as itemize and is invoked by the following code
\begin{itemize} • first item
\item first item
\item second item • second item
\item third item
\end{itemize} • third item
As you can see the list created by the \begin–\end{itemize} environment. Each item, bullet
point, is created by writing the command \item followed by the content of that bullet. There
are other forms of lists and you can create your own but we will cover that in section 3.10.

Do it
• Make your own list. Try nesting lists.

Once we have made a list we may consider adding a table

\begin{tabular }{l c r}
\hline
a &BB &CCC\\ a BB CCC
aa &B &CC\\ aa B CC
aa &BBB &C\\ aa BBB C
\hline
\end{tabular}

A table is thus created by a \begin–\end{tabular} environment. After the \begin{} is a set of

curly braces showing the letters l c r. This tells LATEX there will be three columns and shows
the alignment for them: left, centre and right in this case.
The table itself is made up of several rows of data. Each row has the form & & \\. The
ampersand indicates a tab separating columns and each row of information is ended by a double
backslash, which is the LATEX instruction to end a line. So the first row in our table should be
understood as ‘a - new column - BB - new column - CCC - new line’.
In the table, there is also the command \hline which produces a horizontal line. This basic
table is not exactly beautiful but we will return to tables in section 4 and how to improve on
this later. For now you know the principles to make a table.

Do it
• Make your own larger table. Use your own data or try to mimic something from a journal.

7
 Now let us look at an original strength in LATEX, typesetting mathematical notation and
equations. Type in the following into your document
The definition of
$X_Z$ is given by The definition of XZ is given by
\begin{equation} ∑
X_Z = \sum x_z + \beta^n XZ = xz + β n (1)
\end{equation}

This equation is of course nonsensical but serves the purpose of showing some mathematics.
First, you notice that there is code in the sentence: $X_Z$ which yields XZ . The dollar signs
signals to LATEX that it should switch in and out of mathematical mode. LATEX works in two
modes, mathematical and text mode. Certain commands and features only work in either
mathematical or text mode, rarely in both. If you use them in the wrong mode they give rise
to an error. In this example the underscore _ is LATEX mathematical code for a subscript.
It is very important to remember to keep the dollar signs paired. If you left the code above
without a dollar sign, $X_Z, LATEX would continue to think everything afterwards until the end
of the paragraph is mathematics and eventually yield an error message.
After the sentence there is a construction using the environment \begin–\end{equation}.
This is the way you create so called display mathematics, equations that are on a separate line
and as in this case also centred and numbered. In the equation given by the code X_Z = \sum x_z
+ \betaˆn you see a couple of mathematical commands \sum which yields the summation sign
and \beta which yields the Greek letter β. You also notice the LATEX command for superscript,
ˆ.
More on mathematical typesetting will be discussed in section 6 but for information, use,
for example, the LATEX cheat sheet to see examples of mathematical notation.

Do it
• Make your equations using the cheat sheet.
You now have almost all you need to write a complete scientific manuscript, but you obvi-
ously wonder about figures (graphics) and references. This is a bit more involved and we will
return to these aspects in sections 7.2 and 8.

3 Expanding your manuscript

In the quick start, we briefly touched on several types of formatting in LATEX. You can go a long
way with what you have already learned to write a scientific document, but what you lack is the
knowledge of how to change the layout. It is therefore time to provide a more comprehensive
view.

3.1 Types of commands

LATEX has three different forms of commands. The simplest form is a command that does not
take any argument. An example is \LaTeX which yields the formatted word LATEX. These com-
mands may produce unwanted effects if you just type them in as they are. If you, for example,
type \LaTeX is you will get LATEXis, where the space between ‘LATEX’ and ‘is’ disappears. The
reason for this is that \LaTeX and any simple command needs to signal to LATEX that the com-
mand is completed. A space is one way to accomplish this but then the space is ‘consumed’
and cannot be used to create a space between words. This means that the space we typed
will not be used as an inter-word spacing but just to complete the command and ‘is’ will start
immediately after the command. So to get the space we need we must remember to provide a
completion of the simple commands. This can be accomplished in several other ways. We can
add a backslash after the command as in \LaTeX\ is or add a pair of curly braces as in \LaTeX{}
is. If the simple command is followed by a punctuation such as in \LaTeX. Is and \LaTeX, is
the punctuation acts as completion. Forgetting this detail is the cause for many inadvertent
typographical errors in LATEX.
The second type of command is the command that takes one or more arguments. An simple
example of this is \textit{italics} which yields italics. This type of command uses curly

8
braces, sometimes additional braces for input to the command. These commands do not need
any completion since the braces provide the information LATEX needs to understand that the
command is complete. A classical mistake with these types of commands is to forget to add
the closing brace which then will yield an error.
The third type of command is the ‘environment’ \begin–\end{} where the command has an
opening and a closing statement and where the argument is everything you place between the
two. We have already looked a three such environments document, itemize and equation. In
this case it is important to match any \begin with a matching \end statement.
It is important to know how the different commands should be written and what may be
optional arguments. But you will build your knowledge as you continue working with LATEX.

3.2 The document structure

In section 2 we saw that the basic LATEX document consisted of the following three lines of
commands
\documentclass{article}
% preamble
\begin{document}
% document text
\end {document}

and we also began writing our text in the \begin–\end{document} environment. The general
structure requires a few more comments.
First, the \documentclass command must be the first command encountered in a LATEX
document. It may be preceded by comments but not by any other command or plain text.
Second, all text that we want displayed in a document must occur between the \begin–
\end{document}. In fact anything you write after the \end{document} will be ignored. This can
be quite useful since you can move the \end{document} up into the text in order to compile only
a portion of the text file. It also means you can cut and paste bits of text into your document
file after the \end{document} for later use.
Third, the space between \documentclass and \begin{document} is usually called the preamble
and is reserved for LATEX code that is used to influence the behaviour of what you write in the
document environment. You will see how this part of the document usually gets filled with
several useful tools. It is however, important to realise that no ordinary text can be written in
this part of the document, only commands. Errors will otherwise occur.

3.3 Document classes

When you begin writing a LATEX document, you need to select a class or type of document.
Most commonly we find ourselves working with ‘article’ as we have already seen, but, there are
other classes that are standard in LATEX

Class Description
article Common scientific article
book Book with chapters and parts
letter Letters with letterheads
report Report with chapters
slides Overhead slides
Apart from the standard classes, there are numerous other classes available. Many are
written by enthusiasts for particular purposes such as CV, sheet music, sudoku, etc. Other
classes have been developed by publishers and journals to facilitate authoring in journal specific
formats. For scientific publishing, the latter is probably what we should look for.
To provide some examples, the American Geophysical Union and the Open Access publisher
Copernicus has classes for their journals. Most journals on, for example, Wiley, Springer and
Taylor & Francis also have class files for their own style. Note that classes may contain new or
redefined commands which can lead to unexpected behaviour if you do not carefully check how
the class works. In most cases, you will not notice any differences but not all classes may be
up-to-date or well written.

9
Do it
• Look at some key journals in your field and locate their class files. Or, search Overleaf
for journal templates. You may want to try them out when you feel ready.

3.4 Reserved characters

When you write text in LATEX there are a number of characters that are reserved because they
are used as part of commands and instructions for LATEX. These cannot be used directly in the
text without causing either an error or some unexpected result. You have already come across
most: the comment, the tab stop in tables and the mathematical mode switch as well as super
and subscripts. In the table below you also see the commands you need to use if you wish to
use these reserved characters in the text.
Character LATEX use Command
% comment \%
& tab stop \&
$ mathematical mode \$
ˆ mathematical mode superscript \ˆ{}
_ mathematical mode subscript \_
{} start stop grouping \{ \}
˜ non-breaking space \˜{}
# Parameter in macros \#

3.5 Type face styles

As with all word processors, basic text is typed out with the normal type face. If you need to
change this there are several different choices summarised in the table

Type Command
italics \textit{}
bold face \textbf{}
Small Caps \textsc{}
slanted \textsl{}
sans serif \textsf{}
typewriter \texttt{}
underline \underline{}

The size of the text can also be varied but not in the way you may think. LATEX scales fonts
relative to the ‘normal size’, that is the size of the main text. Thus in order to get larger or
smaller font sizes you can use the following set of commands, relative to the ‘normal size’.

Font size Type Example

5 6 \tiny the quick brown fox

7 8 \scriptsize the quick brown fox

8 10 \footnotesize the quick brown fox
9 11 \small the quick brown fox
10 12 \normalsize the quick brown fox
12 14 \large the quick brown fox
14 17 \Large the quick brown fox
17 20 \LARGE the quick brown fox
20 25 \huge the quick brown fox
25 25 \Huge the quick brown fox
The benefit of this is that as you change the global font size all relative scaling will also change.

10
 The default type size for LATEX is 10pt. The basic sizes you can use are 10, 11, and 12pt
and this can be set using the \documentclass command as an optional argument. If you want
to make an article in 12pt you need to provide
\documentclass [12pt]{ article}

at the beginning of your document. Here the square bracket indicates that the command takes
an optional argument stated within the brackets. If you do not specify any optional argument
LATEX will use a default value.
Changing type faces is not necessarily straight forward; I am almost tempted to state you
should not even try, yet. The reason for the hesitation is that while LATEX typesetting has
remained more or less constant since its infancy, fonts have not. This is because as comput-
ers have gone from 8-bit to 64-bit and capacity has sky-rocketed, Fonts have developed from
containing 128 to now over a million characters. With the backwards compatibility of LATEX
it has not been possible to develop with the new standards but rather find ways to implement
the new while keeping the old.
In Overleaf, you are most likely using an implementation of LATEX called pdfLATEX. This
directly generates pdf files from your LATEX source and is probably the most widely used form
of LATEX. In the core LATEX distributions (incl. Overleaf) many fonts are installed that work
with pdfLATEX.
Another implementation called XeLATEX was developed to make use of the newer Unicode
type faces containing over a million characters. If you run XeLATEX you can quite easily access
your system type faces (the same you access in Word). This is quite tempting but remember
that in order for someone else to run your document, they need to have the same type face. So
there are benefits and pitfalls with going either way.
In section 2 we had issues with accented letters. This problem is easily circumvented by
switching to the XeLATEX compiler in Overleaf since we can then access such accented letters
the way they are entered from our computer keyboard. So if you are writing in languages that
use accented letters it may be useful to use XeLATEX instead of pdfLATEX.
That said, I can mention the LATEXFont Catalogue, which contains an overview of generally
available free type faces and also how to make them usable in your document. An important
point to make is that only certain type faces have support for mathematics. These are identified
in the font catalogue. If you use a type face that does not include mathematics anything written
with mathematics will be written using the default font. In some cases you will probably barely
notice the difference but in other cases it will just be ugly. The freedom to chose can thus be
quite limited.
You can try to change fonts by following the instructions in the font catalogue and try other
solutions found on TEX StackExchange.

Do it
• Play around with the different ways to change font and size, including the switch between
10, 11, and 12 pt in the documentclass command.

3.6 Typographic features

LATEX contains many special characters and features. A list of some of these is

Type Command Function

‘’ single grave accent and apostrophe British open–close quote
“” double grave accent and apostrophe American open–close quote
– -- en-dash
… \ldots –
© \copyright –
® \textregistered –
TEX \TeX –
LATEX \LaTeX –

11
 Most of these need no further comment but I will point out a couple of details. First, when
it comes to using quotes, it is possible to use the common double quote on the keyboard but
it does not look very good. In addition, double quotes in English is American typography so
unless you are American or come from a country that uses the American typesetting standard,
you should stick to the British English single quote.
Second, a common problem with texts is that authors do not know the difference between
a hyphen (dash) and a longer dash (en-dash). The hyphen is part of your key board and used
when you hyphenate words or put two words together. The longer dash is used when you want
the meaning to be ‘to’ as in 4–7 (four to seven). The longer dash looks like a normal minus sign
but LATEX actually have a special minus sign produced in math mode, compare – (en-dash) and
− (math minus). Since it is very easy to distinguish between the different types of dashes in
LATEX you should make an effort to use them appropriately.

3.7 Breaks and space

When you write text you need to occasionally deliberately break the text or words to fit the
space available. One of the strong points of LATEX is to automatically manage these duties but
it is still not possible to automate typesetting to 100%.
Let us start by looking at a few commands that are useful for breaking text over pages and
lines.
Command Function
\newpage page break
\pagebreak page break keeping justification
\\ line break without new paragraph
\newline line break without new paragraph
\linebreak line break keeping justification
\nolinebreak –
\noindent prevent indentation of a paragraph
\indent indent a non-indented paragraph
\- conditional hyphenation
\ keep normal space
\@ end of sentence space after capital letter
Most of these are quite self-explanatory and you can test their function in your own text.
The conditional hyphen and hyphenation as a whole may deserve a comment. If you end up
with a word that is too long and it does not seem to hyphenate at all or perhaps wrong (LATEX
uses a set of rules for hyphenation), then you can insert a conditional hyphen. If we take the
word ‘multispectral’, we can prepare the word with conditional hyphens so that it will know
where breaks can be, the code will be mul\-ti\-spec\-tral. These conditional breaks will not
show unless used.
In addition, if you use a word, for example, a complex scientific term that is unlikely to
be hyphenated correctly you can provide LATEX with information how to treat the word in the
text. By placing
\ hyphenation {mul\-ti\-spec\-tral}

in the preamble, you pass this information to the hyphenation engine to correctly deal with the
word any time it is encountered in the document.
By default LATEX uses English hyphenation rules. If you write in another language you may
therefore get word breaks that do not fit that language. You can correct this problem by using
a language package called ‘Babel’ which we cover in section 7.15.
LATEX uses an intricate system to keep track of good word spacing. By default you will get
a slightly wider space after a period, indicating the end of sentence and beginning of a new.
This rule is, however, not always correct. If we have an abbreviation with periods each period
will be interpreted as an end-of-sentence. If we look at the following example

12
 Example Code
J. Geophys. Res. J. Geophys. Res.
J. Geophys. Res. J. Geophys.\ Res.

You can see that the space between ‘Geophys.’ and ‘Res.’ becomes shorter in the second
case, in fact it is a common inter-word space. This means that you should add a backslash after
abbreviations written in common letters unless the period really is the end of the sentence.
Another issue that can arise is that LATEX by default interprets a period after a capital letter
as an abbreviation and hence provides inter-word spacing. If you end a sentence with such an
abbreviation the space is too short. To solve this you can add the command \@ between the
abbreviation and the period as in …in UNESCO\@. Another ….
Lastly, in the old days of days of monospaced typewriter fonts, it was customary to insert
two spaces between sentences. You sometimes see people still do this in word processors. LATEX
inserts a space that is slightly longer than a single space, but still shorter than two, by default. It
is possible to get single space between sentences by adding the command \frenchspacing at the
beginning of the document. You can turn the French spacing off by adding \nonfrenchspacing.
Since adding two or more spaces between sentences in LATEX does not affect the spacing, it does
not matter if you accidentally hit the space bar a couple of times extra, the only way to really
easily affect the spacing is through the French-spacing commands.

Do it
• Test the different commands affecting spacing within your own text so that you understand
the difference they make. It may be beneficial to work on several identical paragraphs in
parallel so that you can visually compare the results.

3.8 Headings
The headings in LATEX come in two flavours, numbered (default) and unnumbered. The un-
numbered versions can be obtained by using so called ‘starred’ versions of the command. The
section command \section{} is thus converted to unnumbered by writing \section*{}.

Numbered Unnumbered
\part{} –
\chapter{} \chapter*{}
\section{} \section*{}
\subsection{} \subsection*{}
\subsubsection{} \subsubsection*{}
\paragraph{} \paragraph*{}
\subparagraph{} \subparagraph*{}

The levels part and chapter are only available in the book class.
In addition the book class contains several other features. By adding the command \appendix
LATEX will start alphabetic sectioning numbering instead of numeric, which is the common
form for an appendix. There are also two commands called \frontmatter and \mainmatter that
numbers pages with roman numerals from the frontmatter command until the main matter
command. After that numbering will be with Arabic numerals. this is common in type setting
books where pages with roman numerals are used for pages containing material that is not part
of the main book content such as table of contents, preface etc.

Do it
• Try out the different forms of headings. Note, however, that for some to work you need
to switch your document to the book class.

13
3.9 Table of contents
To create a table of contents is very simple in LATEX. You need to do one thing and that is
to type the command \tableofcontents where you wish the table of contents to go. There is,
however, one caveat. The table of contents can only be generated from the numbered sections.
If you, for example, have numbered headings in your document and edit in a star in one of
the heading commands, that heading will no longer be visible in the table of contents. There
is a work-around for this by changing the way the table of contents work. If you add
\setcounter{secnumdepth }{0}

to the preamble this tells LATEX not to number any sections. By adding or removing this line
from your file you can essentially turn numbering on or off.
Another way to accomplish this is to add
\addcontentsline {toc}{ type }{ heading title}

after each heading. type refers to the heading level. If you for example have an unnumbered
section and subsection that should be in the table of contents you will provide the following
after each heading
\section{The section heading}
\addcontentsline {toc}{ section }{The section heading}

\subsection{The subsection heading}

\addcontentsline {toc}{ subsection }{The subsection heading}

The \addcontentsline command tells LATEX to add information about a section to a file that it
keeps to create the table of contents. The first argument is the name of the file. This will be a
file named the same as your document but with the extension .toc. The second argument tells
LATEX what level of heading you add to the file. The third is the heading text.
In LATEX you can also provide similar table of contents for your figures and tables using
the commands \listoffigures and \listoftables. We will get back to details how this can be
accomplished in section 5.
If you have many sections of all levels it may make sense to remove the lowest level (subsub-
section) from the table of contents in order to shorten it. The command \setcounter{tocdepth}{1}
is then useful because it allows you to limit the ‘depth’ of the table of contents. With the ar-
gument ‘1’ only sections and subsections will be visible.

3.10 Environments
In the quick start we came a cross several environments. When you enter text there are several
more that you should be aware of. Common to all is that the use the environment structure
\begin{}–\end{}.

Environment Action
center centers content
flushleft flushes content left
flushright flushes content right
quotation decreases text area width for a quote with indentation
quote decreases text area width for a quote without indentation
verse for poetry
minipage produces a separate environment with its own dimensions
itemize produces bullet lists
enumerate produces numbered lists
description produces lists with key words as ‘bullets’
verbatim reproduces all text including protected symbols as written
Most of these environments are self-explanatory and something you can explore on your own.
The minipage environment is worth looking into a bit deeper. In this text I have used the
minipage to create ‘side-by-side’ looks at code and result. The following display

14
\begin{itemize} • first item
\item first item
\item second item • second item
\item third item
\end{itemize} • third item
is created by
\ noindent \begin{minipage }[c]{0.4\ textwidth}
\begin{verbatim}
\begin{itemize}
\item first item
\item second item
\item third item
\end{itemize}
\end{verbatim}
\end{minipage }\ hfil
\begin{minipage }[c]{0.4\ textwidth}
\begin{itemize}
\item first item
\item second item
\item third item
\end{itemize}
\end{minipage}

This may seem daunting but if you look closely, it consists of two minipage environments typeset
side by side (no paragraph break) and where the mini-pages are set to be 40% of the text width
(0.4\textwidth) each. What is not obvious from the typesetting is that I have pushed the
two mini-pages apart by inserting the command \hfil which is a basic command for inserting
space if there is extra space to be had. There are a whole series of such commands but they
are beyond the scope of this text. I can only urge you to start a little self study on the more
advanced topics that allows you to dive into the interior of LATEX.
The description environment is very useful and works just like the itemize and enumerate
environment but with the exception that you need to provide an argument.

\begin{description} 1 first item

\item [1] first item
\item[two] second item two second item
\item[III] third item
\end{description} III third item

It is also possible to change the look of all list environments. A simple way to change the
bullets in itemize is to add an argument after \item[]. If you want an en-dash instead of the
bullet you simply write \item[--] for each of the items you want to list.

Do it
• Try the different environments. In the case of lists, try to nest them to see the effects of
nesting on their behaviour.

3.11 cross-referencing
LATEX comes with a powerful cross referencing system. The system works on the principle
that you put a label on everything you want cross-referenced and then use a command to take
information about that label and insert it into the text. The following are the commands you
may encounter.

Command Action
\label{} establishes a label
\ref{} takes a label and replaces it with a number
\eqref{} specific references to equation numbers
\pageref{} specific references to page numbers
\cite{} generic literature references (see section 7.4)

15
 The \label command is used to assign a unique name (label) to a section, figure, table or
equation. You need to provide a name for the label and this can be any name you chose. If we,
for example, have a map in our manuscript, I would label this figure \label{fig:map}. First,
I would recommend that you have a descriptive word in the label so that you know what that
figure is. Second, I prefer to add a ‘tag’ ‘fig:’ for figure ‘tab:’ for table and ‘eq:’ for equation
so that I do not confuse a tag for a figure with one for an equation. You can come up with
your own scheme, however, you will be limited to letters a–z, numbers and basic punctuation
for your labels. So no protected symbols or accented letters can be used.
Once you have assigned a label to an object you can reference, for example the map we
labelled above as \ref{fig:map}. In the resulting layout LATEX will take the number of the
figure labelled fig:map and insert that instead of the \ref{fig:map} command. The benefit of
this is that all figures (and tables and equations) are numbered implicitly, which means that if
you rearrange figures the labels will be assigned new appropriate numbers and so will all you
references in the text. You simply do not have to worry.
There are a set of specific reference commands. \eqref{} takes only labels from equations
but you can just as easily use the \ref{} command. The \pageref{} will display the page
number where your label is defined. This means that if you have labelled your map somewhere
and wish to refer to it, not by its figure number but the page on which it occurs, then you
should use \pageref{fig:map}.
Finally, LATEX comes with a cross-referencing system for citations and reference lists. I will
not go into any details here because there is more to the system than just cross-referencing.
Please refer to section 7.4 for a detailed description. At this point it is sufficient to know that
the \cite command exists for literature references coupled to a reference list.

3.12 Special commands

LATEX has several built in commands that perform tasks you want to use. Here is a list of ones
we will look more carefully at

Numbered Unnumbered
\today provides current date
\the allows typesetting of internal variable values from LATEX
\year variable containing current year
\day variable containing current day
\month variable containing current month
\footnote{} typesets footnote
\noindent cancels indentation of a new paragraph
\marginpar{} adds text in the margin where the command occurs
LATEX can take information from the computer and display in documents. The most common
command for this is \today which displays the current date. When you created the title of the
document with \maketitle LATEX implicitly uses the \today command to add a date to the title.
You can use the command anywhere if you want to have the current date in your document but
do not confuse this with a fixed date, it will be updated every time you run your document.
A usually little known but powerful command is \the. This command takes a value from
some parameter within LATEX and displays it as text in your document. Since LATEX has
variables for year, month and day called (of course), \year, \month and \day, it is possible to
extract that information into the text by using the combination of commands \the\year etc.
But, this will only yield the numbers, not, for example, the month in text. It is however,
possible to program LATEX to convert the number to text.
Footnotes are rarely used in scientific articles but is of course an integral part of any text.
LATEX has a footnote command that will take the text you enter as an argument and place it
at the bottom of the page like in the example here1 . To create a footnote you simply add the
command \footnote{} right where you want the reference to occur. LATEX will insert an index
number and add the footnote at the bottom of the page automatically.
1 This is footnote text entered in the paragraph on footnotes above

16
 The marginpar command is useful if you want to, for example, add visual reminders to your
self in the text. I find it useful to define a new command (see section 9) based on the marginpar
where the type size is smaller and perhaps also in some colour to make it more visible.

Do it
• Try the different special commands.

4 Tabular information
Typesetting good tables is not easy anywhere. Quite often we see tables made to look like a
heap of boxes. This is not good type-setting practise. We will thus focus on simple publication
quality tables.
In LATEX, the basic environment for creating tables is called tabular and a simple table
would be made as follows (with output to the right):
\begin{tabular }{l c c}
\hline
1 &2 &3 \\ 1 2 3
\hline
A &B &C \\
A B C
a &b &c \\ a b c
\hline
\end {tabular}

This is neither sophisticated or beautiful but shows the principles. Obviously the tabular
uses a \begin–\end structure. After the \begin{tabular} is a set of curly braces containing
the letters l c c. These letters tells LATEX how to align the content in the three columns,
first column l for left-adjusted and the other two c for centred. There is of course an r for
right-adjusted columns.
The core of the table contains rows such as for example 1 &2 &3 \\. We can see that the
ampersand is used to mark columns 2 and 3 but is not used for column 1. This is how LATEX
uses the ampersand in the tabular environment, it indicates a jump, tab, to the next column.
The line ends with a double backslash which tells LATEX that the line is complete and to switch
to a new line. All lines of data in a table must end with the double backslash. I have also
inserted some horizontal lines using the \hline command. This code does not take a double
backslash to mark that it is on a separate line.
This example shows us a few things worth noting. First, that a table should be as simple
as possible when it comes to lines, this is good typography. Second, LATEX will determine the
width of columns and the width of the table based on the content. This is of course fine as long
as our data and column headers are simple and short.
By loading an extension called booktabs (see section 7.6) you have access to a series of
additional lines, called rules that will help you organise more complex tables. In the following
example we use the non-native commands \toprule, \midrule, \bottomrule and \cmidrule. They
are all self-explanatory except the last.
\cmidrule[](){} takes three arguments (note the different types of brackets used). The first
specifies the thickness of the line and the second if it should be trimmed. Both the thickness
and trim arguments are optional. If not trimmed, the lines will go the full width of the columns
and possibly meet with neighbouring lines. Trimming can be achieved by entering l for left
trim and r for right trim or both into the argument. The third argument takes the column
numbers across which the rule should span.
As an example we can look at this small but relatively complex table.

17
 Year bw Hss bw (k) Swi bc (k) ba (k)
29 28S3 29 28S3 29 28S3 29 28S3
(m w.e.) (m) (m w.e.) (%) (m w.e.)
6 0.06 0.06 0.09 0.12
1997/98 1.35 2.7 2.4 0.03 0.06 7 0.07 0.07 0.10 0.13
8 0.08 0.09 0.11 0.14
6 0.06 0.08 0.09 0.13
1998/99 1.33 2.9 2.6 0.03 0.06 7 0.07 0.09 0.10 0.15
8 0.08 0.10 0.11 0.16

was created by
{\ begin{tabular }{l c c c c c c c c c c}
\toprule
Year &$b_w$ &\ multicolumn {2}{c}{$H_{ss}$} &\ multicolumn {2}{c}{$b_w(k)$}
&$S_{wi}$ &\ multicolumn {2}{c}{$b_c(k)$} &\ multicolumn {2}{c}{$b_a(k)$}\\
\cmidrule(rl){3 -4} \cmidrule(rl){5 -6} \cmidrule(rl){8 -9} \cmidrule(rl ){10 -11}
& &29 &28S3 &29 &28S3 & &29 &28S3 &29 &28S3\\
&(m w.e.) &\ multicolumn {2}{c}{(m)} &\ multicolumn {2}{c}{(m w.e.)}
&(\%) &\ multicolumn {4}{c}{(m w.e .)}\\
\midrule
& & & & & &6 &0.06 &0.06 &0.09 &0.12\\
1997/98 &1.35 &2.7 &2.4 &0.03 &0.06 &7 &0.07 &0.07 &0.10 &0.13\\
& & & & & &8 &0.08 &0.09 &0.11 &0.14\\
& & & & & &6 &0.06 &0.08 &0.09 &0.13\\
1998/99 &1.33 &2.9 &2.6 &0.03 &0.06 &7 &0.07 &0.09 &0.10 &0.15\\
& & & & & &8 &0.08 &0.10 &0.11 &0.16\\
\bottomrule
\end{tabular}

An important command we encounter here is the \multicolumn{}{}{} command. The command

places a single entry across any number of columns. It takes three arguments. The first is the
number of columns it should span. The second is the alignment we want. The third is the text
that should be placed across the multiple columns.

Do it
• Typeset some of your own tables. Start with a simpler table.

5 Floats; tables and figures on the run

LATEX provides special two environments for figures and tables that have the property that they
can be moved by LATEX from the location where you have entered them to a typographically
preferable place. This feature is called a float because they can ‘float around’. Many beginners
can find this feature irritating, not knowing exactly before hand where the figure may appear.
But, in this case, LATEX has much better knowledge of good layout and typesetting than the
average user so the irritation is just a lack of knowledge.
To make a float you use the two environments figure and table. For a figure you would use
the following code
\begin{figure}
\includegraphics {filename}
\caption{This is the figure caption text}
\label{fig:x}
\end{figure}

The figure environment should contain three commands. First is a command \includegraphics{}
which brings in a graphics file into the document. We will discuss this command in more detail
in section 7.2. Second is a new command called \caption which produces a figure caption using
the text you enter as the argument to the command. This caption should always be typeset
below the figure graphics, which is why we place the commands in that order, and the figure–
caption pair will never separate across pages etc. Third is the \label command which allows
you to refer to the figure using the label name (see section 3.11).

18
 A table follows the same basic recipe but the table itself is made up of a tabular environment
(section 4)
\begin{figure}
\caption{This is the figure caption text}
\label{tab:x}
\begin{tabular }{c c}
. . .
\end{tabular}
\end{figure}

In this case we place the caption before the tabular environment because the table caption must
go above the table. We also have a label working the same way as for the figure environment.
Creating these floats is thus quite simple. It is, however, possible to influence the placement
of the floats in the document. To do this, we can add an optional argument to the \begin{}[]
command. This optional argument can be one or a combination of

Option Action
h here, approximately where it occurs in the source text
t top of the page
b bottom of the page
p on a special floats page
! Override internal parameters for ‘good’ float position
H H ere, precisely at the location in the code, similar to !ht1
1
This requires a package called ‘float’, section 7.13
The way float placement works has been described by Frank Mittelbach on TEX StackEx-
change. and I will summarise the flow here.
When a float is encountered, LATEX attempts to place it immediately according to its rules.
If this does not succeed, then LATEX places the float into a holding queue to be considered when
the next page is started. Once a page is completed, LATEX tries to place remaining floats in the
holding queue as best as possible. To do this it will first try to generate as many float pages
as possible. If this is not possible, it will try to place the floats into top and bottom of pages.
It looks at the remaining floats and either places them or defers them to a later page. After
that, it starts processing the text for the page. In the process, it may encounter further floats
and the process goes on. If the end of the document is reached before the queue is emptied,
LATEXwill add pages and dump remaining floats there.
As you can tell the process is quite involved and never random. If there is need to try to
create a page of floats inside a document, you can use the command \clearpage which will
empty the queue onto pages before continuing with the document.
The option H is not generic LATEX and requires additional work which we will discuss in
section ??.

Do it
• Test the different float environments with a simple figure and table and change the po-
sitioning arguments to see their effects. Note that you need to have a few pages of text
with plenty of paragraph breaks to be able to really test the floats. I recommend loading
the lipsum package (section 7.16) and adding text by typing \lipsum[n] where n can be
a number or a range (e.g. 2-5). Check the package documentation for details; lipsum is
useful if you need text to see how a layout works.

6 Mathematical typesetting
Since mathematics is one of LATEX original strong points there are too many introductions to
count on writing mathematics out on the Internet. I will therefore not try to add another one
to the mix but draw up some basics that will allow you to go along way and form a basis for
your own research. As we saw in the quick start, mathematics come in two flavours, in line
and display mathematics. The same commands are used in both cases in order to produce the

19
mathematical notation so we will first focus on the different forms for displaying mathematics
and then onto the details of how to create the equations.
The basic display mathematics environment is equation. As we have seen earlier it works
just like any other environment

\begin{equation }\ label{eq:x}
a^2 = b^2 + c^2 a2 = b2 + c2 (2)
\end{equation}

In this example I have also added the label which means you can now cross-reference the
equation number as \ref{eq:x} in the text. There is also an environment for sets of equations

\begin{eqnarray}
c_1 &=& a_1x + b_1x \label{eq:a} \\
c_1 &=& a_2x + b_2x \label{eq:b}
c1 = a1 x + b1 x (3)
\end{eqnarray} c1 = a2 x + b2 x (4)

The alignment is set up by the ampersands in the two equations, in this case to make sure
the equal signs line up. You can also use the eqnarray to take care of longer equations

\begin{eqnarray}
y = a_1x &+& b_1x + c_1x + d_1x
\nonumber \\ y = a1 x + b1 x + c1 x + d1 x
&+& e_1x + f_1x \label{eq:y}
\end{eqnarray} + e1 x + f1 x (5)

With AMSLATEX (see below), you can access another environment to do what we did with
eqnarray above. This is the align environment

\begin{align}
y = a_1x &+ b_1x + c_1x + d_1x
\nonumber \\ y = a1 x + b1 x + c1 x + d1 x
&+ e_1x + f_1x \label{eq:y}
\end{align}
+ e1 x + f1 x (6)

In this case the alignment is much better. This is but one example of how using AMSLATEX
improves your capabilities.
When you build equations, you need to know the commands that produce all the different
mathematical symbols and notation. A good source is the American Mathematical Society
(AMS). It is worth mentioning at this stage that while LATEX does excellent mathematical
type setting, AMS has improved on the capabilities through their AMSLATEX extension. If
you just need to type set a few formulas and ordinary equations, basic LATEX should be more
than sufficient. But if you need some more special features, they are almost certainly available
through the AMSLATEX extension.
Now turning to formulating an equation. When you type in an equation, you do generally
not need to worry about spacing. As we noted earlier all letters will be typeset in mathematical
italics, this is the norm for mathematical variables. The italics only applies to Latin letters,
Greek letters are not italicised. This means that if you want to type in a function such as sine
you need to be careful as this example shows

\begin{eqnarray}
\tau &=& \rho gh sin\alpha \\
\tau &=& \rho gh \sin\alpha
τ = ρghsinα (7)
\end{eqnarray} τ = ρgh sin α (8)

In the upper row we typed in the letters ‘s i n’ and they came out in italics but with a
similar spacing as between the variables g and h In the second row we used the mathematical
command \sin. In the first case, ‘sin’ is treated as three variable names. LATEX, therefore has
all the mathematical functions established as commands so that they are type set correctly.
Adding letters after each other is just interpreted as a series of variables multiplied by each
other. This has a bearing on those who persist to form variables more or less at abbreviations

20
such as ET for evapotranspiration or worse NDVI for Normalized Difference Vegetation Index
N IR − V IS
N DV I = (9)
N IR + V IS
where NIR is the near infrared band and VIS is the visible. This clearly does not look good
partly because of the way LATEX handles text i mathematical mode but also because it is a poor
way to defined variable names. So let this be a warning when you define your own variables
and equations.
Despite the poor look of equation 9 we see that we can produce division. This is easily
accomplished with the \frac{}{} command. frac takes two arguments. In the first is everything
that goes in the numerator and in the second, everything that goes in the denominator.

\begin{equation }\ label{eq:NDVIb}
I_\ mathrm{NDV} = \frac{E_\ mathrm{IR}
- E_\ mathrm{VIS}} ENIR − EVIS
{E_\ mathrm{IR} INDV = (10)
ENIR + EVIS
+ E_\ mathrm{VIS}}
\end{equation}

In this example we can see how the \frac command works but we can also see that it is
possible to type in the equation in almost any form, in this case to try to make it structured,
sine it has no consequence for the output. You can also see a new command \mathrm{} which
produces regular text inside mathematical mode.
When you use the frac command and you need to put a parenthesis around the equation,
you can obtain scalable parenthesis using the \left and \right commands

\begin{equation }\ label{eq:NDVIb}
I_\ mathrm{NDV} = \left(
\frac{E_\ mathrm{IR}
- E_\ mathrm{VIS}} ( )
ENIR − EVIS
{E_\ mathrm{IR} INDV = (11)
+ E_\ mathrm{VIS}} ENIR + EVIS
\right)
\end{equation}

The \left and \right commands must pair up. You can use ( ), [ ] and { } and also | with
these commands. It is also possible to set only one side of the pairs you should use a period for
the side you want to omit: \left. \right. . Note that when you use \left and \right, both
must be included.
Making mathematical type setting is otherwise mostly finding the right commands to obtain
what you want. There are many lists floating around that summarizes these so I will not take
up space here. I would argue that mathematics is one of the easiest things you can do in LATEX.

Do it
• Typeset some of your own equations or try to reproduce an equation out of an article or
book, or both.

7 Extending the functionality of LATEX, packages

Because LATEX is OpenSource, many enthusiasts contribute to LATEX functionality by providing
extensions, so called packages. The the Comprehensive TEX Archive Network (CTAN) con-
tains packages and their documentation and is thus a good source for information. I strongly
recommend you to look at any package documentation before using any package.
Out of all the available packages, only a small fraction will likely be of use to you. A well
known effect of being new to LATEX is to go ‘package crazy’ and load almost anything. If this
happens to you, then it will pass; but more severely, you may end up with unforeseen problems
due to conflicts between certain packages. Some packages are also tailored to solve a particular

21
problem for a given situation that may not apply to you. Much of this is stated in the package
documentation.
In the following sections I will provide information on some packages that vary from a must
to being of general interest. A package is loaded by adding the command \usepackage to the
preamble:
\usepackage []{}

As can be seen the command has a normal argument field (denoted by the curly braces) and an
optional input field (denoted by the square brackets). Many packages can be loaded without any
options which means you can ignore the square brackets. Other packages require or has options
for providing specific features. Such options are described in the package documentation.

7.1 inputenc
While scientific writing is primarily done in English and LATEX is made with English in mind it
is often necessary to include words, for example, place names with accented letters in the text.
Generic LATEX does, for example, not support the å, ä, and ö on the Swedish keyboard. You
can of course create these letters through LATEX accenting commands, in this case by typing
\aa, \"a and \"o, respectively. If this happens a few times it is not a major issue. There is
however a way to ensure that what is on your keyboard is also possible to handle in LATEX.
The package inputenc (short for ‘input encoding’) allows LATEX to use an extended encoding,
that is beyond the letters and numbers of the English keyboard. this is not limited to Swedish
letters but to a large variety of other accented letters. The package should be called with
the option utf8. The UTF8 encoding allows a computer to manage over a million different
characters which includes all characters in modern Unicode type faces. By adding the following
to your preamble, you have full use of your font character set. The package is loaded by
\usepackage[utf 8]{ inputenc}

Note that this package is not necessary when you use XeLATEX to compile your document.

7.2 graphicx
The graphicx package is a must. Without this package you will not be able to include graphics
into your document. The package is loaded by
\usepackage{graphicx}

in the preamble of your document.

When you have graphicx loaded into your document you have access to the command for
including graphics, appropriately named \includegraphics. The command come with several
optional arguments and looks as follows
\includegrphics[width =0.5\ textwidth ,angle =45]{ fname.ext}

The file name is provided as the main argument in curly braces. It is not necessary to type
the extension. It is also possible to provide a path to the graphics file if it is not in the folder
where your LATEX document resides. valid graphics formats are JPEG, PNG, and PDF. Note
that SVG and TIFF are not supported.
The optional arguments in the example provide LATEX with the information that the figure
should be reproduced with a width that is 0.5× the width of the text. This type of scaling is
convenient but you can also specify a fixed width in, say millimetres, by writing, for example
[width=57mm]. In addition we have also asked LATEX to rotate the figure by 45◦ , which, of course,
is unusual. In addition you can scale the figure by its height in a similar way to the width.
There is a command \textheight that allows you to scale relative to the height of the text area.
If you provide only one of width= or height= the figure will be scaled proportionally in the
other directions. You can scale an image disproportionally by providing different scaling factors
for width and height.

Do it
• Add a figure and test the scaling and rotation capabilities.

22
7.3 geometry
The geometry package provides an easier interaction with LATEX page layout settings. The
package is loaded by
\usepackage[a4paper ]{ geometry}

In this example, the option a4paper has been added in order for margins etc. to be scaled
correctly for that paper size. This is the minimum of what you need to add to your preamble.
It is worth noting that the documentclass command also takes a4paper as an argument but I
prefer to use it in the call to geometry.
If you need to change your margins, you do this by simply using the following
\usepackage[a4paper ,left =35mm ,right =30mm ,top =30mm ,bottom =25mm]{ geometry}

which just happens to be the call made for this document. This sets the left and right margins
to 25 mm wide and the top and bottom margins to 30 mm. You should study the geometry
package information for more details

Do it
• Use the arguments to the geometry package to change margin widths, paper size etc.

7.4 natbib
The natbib package is a must if you need Harvard style author-year referencing system in your
work. The package is loaded by
\uepackage{natbib}

I will not dwell on natbib functionality here since we will go into much more detail when
discussing referencing in section 8. What may be of more interest here is a section of code
that I recommend you to use in conjunction with the natbib package. This code referred to as
natbibspacing.sty
\ newdimen \bibspacing
\setlength\bibspacing\z@
\renewenvironment {thebibliography }[1]{%
\bibfont\bibsection\ parindent \z@\list
{\ @biblabel {\ arabic{NAT@ctr }}}{\ @bibsetup {#1}%
\setcounter{NAT@ctr }{0}}%
\ifNAT@openbib
\renewcommand\newblock {\ par}
\else
\renewcommand\newblock {\ hskip .11em \@plus .33em \@minus .07em}%
\fi
\sloppy\ clubpenalty 4000\ widowpenalty 4000
\sfcode `\.=1000\ relax
\let\citeN\cite \let\shortcite\cite
\let\citeasnoun\cite
\itemsep\bibspacing %
\parsep\z@skip %
}{\ def\@noitemerr{%
\PackageWarning{natbib}
{Empty `thebibliography ' environment }}%
\endlist\vskip -\ lastskip }
% ---------------------
\setlength {\ bibspacing }{0pt}

You may be able to load this code by adding the name to the call for natbib
\uepackage{natbib ,natbibspacing}

but only if it exists in your LATEXdistribution. Otherwise, you can enter the code into your
preamble after you have loaded natbib.
The natbibspacing is an example of pure code and may get an insight into how much details
can be changed but also that it involves learning LATEX as a programming language in detail.
Fortunately, others do these things for you in the community.

23
7.5 hyperref
The hyperref package enables you to make workable links in your pdf output. The package is
loaded by
\usepackage{natbib}

When you use hyperref all the cross-links in your document becomes clickable and allows you
to move around in the document. The table of contents is also clickable by default as are any
references. In addition, you have access to the command \href{}{} which allow you to produce
links to web URLs outside of your document. To show an example how this works we can set
up a link to TEX StackExchange
\href{http :// tex.stackexchange.com }{\ TeX\ StackExchange}

The command takes two arguments. The first is the complete URL and the second is the text
that will be highlighted as a link in the text.
You can modify the way links are shown by the command hypersetup
\hypersetup{
pdftoolbar=true , % show Acrobat 's toolbar ?
pdfmenubar=true , % show Acrobat 's menu?
pdffitwindow=true , % window fit to page when opened
pdfstartview ={ FitH}, % fits the height of the page to the window
pdftitle ={ title}, % title
pdfauthor ={ name}, % author
pdfsubject ={ subject}, % subject of the document
pdfkeywords ={ keyword 1} {key2} {key3}, % list of keywords
pdfnewwindow=true , % links in new window
colorlinks=true , % false : boxed links ; true: colored links
linkcolor=black , % color of internal links
citecolor=SUBlue , % color of links to bibliography
filecolor=blue , % color of file links
urlcolor=blue % color of external links
}

Note that you do not need to list all settings, simply the ones you wish to change from default
values. Just remember to separate them by a comma. With this setting, the links to external
ages are blue, internal links are black except citations which are in a blue defined as SUblue (see
section 7.11). As you can tell, the hypersetup also dictates how the pdf should be opened by
Acrobat reader. You should refer to the hyperref package documentation for more information.

Do it
• Try the hyperref package by adding some urls and also changing the parameters in the
hypersetup command.

7.6 booktabs
The booktabs package is a small package that improves table formatting. The package is loaded
by
\usepackage{booktabs}

It is the booktabs package that allows you to use the commands \toprule,\midrule and \bottomrule
in tables. I have consistently used these for the tables in this text unless stated otherwise. The
package provides a few other commands as well but the booktabs package documentation is
well worth reading because it outlines ideas around good practises in table design.

Do it
• Try the different rules in your tables to replace the \hline.

24
7.7 siunitx
The siunitx package is a very complex package that deals with how to write numbers and units
(with focus on SI units). The package is loaded by
\usepackage{siunitx}

This package may to many seem a bit over the top. Let us look at some of the core ideas. Let
us say we want to write a complex unit such as square volt cubic lumen per farad (V2 lm3 F−1 ).
This seems like a disaster waiting to happen. With the siunitx package and its command \si{}
we will get the correct units through the following
\si{\ square\volt\cubic\lumen\per\farad}

As you can see all the words we used to describe the unit in text is available as a command and
the result is a perfect SI unit form. If you need to add a number to the unit there is a second
command \SI{}
\SI {203}{\ kilo\gram\metre\per\second}

which yields 203 kg m s−1 . Now why is this good? There are several benefits. First it becomes
quite clear what your unit is. Second, the spacing between numbers and units and between the
units themselves are constant and of accurate length. It is true that you end up writing a lot
but the benefits definitely outweighs the disadvantages.
The package contains much more on representing data in text so you should study the
extensive siunitx package documentation.

Do it
• Try siunitx package by typing in some units that you are familiar with.

7.8 lineno
The lineno package does one thing, it allows you do add line numbers to, for example a
manuscript. Some journals, for example, ask for such manuscripts. The package is loaded by
\usepackage{lineno}

You turn the line numbers on by inserting the command \linenumbers where you want the
numbers to start. You can stop line numbers by placing the command \nolinenumbers where
you want them to stop. You should look at the lineno package documentation for more details
on different options.

Do it
• Try lineno package in your document

7.9 dcolumn
The dcolumn package provides means to define new alignments for tabular environments. The
package is loaded by
\usepackage{dcolumn}

To provide an example of what can be accomplished with dcolumn we can define a new column
type called ‘.’ (period) using the command \newcolumntype
\newcolumntype {.}{D{.}{.}{ -1}}

This new column type will align numbers on the period. As you can see the command takes
many arguments. The first argument is the name of the column type (‘.’). The second argu-
ment is divided into three parts, the first part indicates what character should be used for the
alignment (the period), the second part shows what symbol should be used for the separator
(usually the same as the first), the third part indicates how many decimal places should be
shown. With -1 as the third argument, the column will be centred on the decimal point. An

25
example
\begin{tabular }{. . .}
\toprule
45.73 &43.894 &5.463\\ 45.73 43.894 5.463
33 &0.0001 &0.02\\ 33 0.0001 0.02
7.76 &.2 &9.75\\ 7.76 .2 9.75
\bottomrule
\end{tabular}

And then the same table but with a different column type called ‘:’ (colon) which has the
\cdot) as decimal separator and 4 decimal places (the maximum number of decimals in the
table; \newcolumntype{:}{D{.}{\cdot}{4}})
\begin{tabular }{: : :}
\toprule
45.73 &43.894 &5.463\\ 45·73 43·894 5·463
33 &0.0001 &0.02\\ 33 0·0001 0·02
7.76 &.2 &9.75\\ 7·76 ·2 9·75
\bottomrule
\end{tabular}

As you realize you can do a lot with this package and define your own numeric column types.
The dcolumn package documentation is quite brief so experimentation is best way to learn this
package.
The siunitx package also has table alignment support which may be useful to study.

Do it
• Try the dcolumn package by making your own definition or changing parameters in exam-
ples given above and using the new column definitions in your own table.

7.10 tikz/pgf
The tikz/pgf package is actually two packages. Tikz is a front end for the drawing capabilities
set up by pgf. TikZ is a recursive name, Tikz ist kein zeichenprogram. The package is loaded
by
\usepackage{tikz}

The tikz package documentation is 1302 pages (at the time of writing, version 3.1.4b) and
hence extremely detailed. The manual doubles as a tutorial and dictionary. We will briefly
look at two examples that shows you what is needed and what can be done. It is noteworthy
that this is still just scratching the surface of what can be done. Please visit TEXsample.net for
numerous examples of Tikz output.
The first example shows a flow diagram for data collection at Tarfala Research Station.:
\tikzstyle{block} = [rectangle , draw , fill=blue !20,
text width =5em , text centered , rounded corners , minimum height =4em]
\tikzstyle{rblock} = [rectangle , draw , fill=red!20,
text width =5em , text centered , rounded corners , minimum height =4em]
\tikzstyle{line} = [draw , -latex ']

\begin{tikzpicture }[ node distance = 2cm , autoscale =0.7,

every node /. style ={ scale =0.7}]

\node [rblock] (logger) {logger in the field };

\node [block , below of=logger] (pc) {PC in field or office };
\node [block , below of=pc] (infolder) {local folder };
\node [block , below of=infolder] (rename) {rename file };
\node [block , below of=rename] (copy) {copy to Dropbox };
\node [block , below of=copy] (sync) {sync Dropbox };
\node [block , below of=sync] (toserver) {Dropbox to server };
\node [rblock , below of=toserver] (server) {server \\ (= backup )};

\node [block , left of=pc , node distance =3cm] (shuttle) {shuttle download };
\path [line] (logger) -| (shuttle );
\path [line] (shuttle) -- (pc);

26
 \node [block , left of=sync , node distance =3cm] (ftp) {ftp to server };
\path [line] (rename) -| (ftp);
\path [line] (ftp) |- (server );

\path [line] (logger) -- (pc);

\path [line] (pc) -- (infolder );
\path [line] (infolder) -- (rename );
\path [line] (rename) -- (copy );
\path [line] (copy) -- (sync );
\path [line] (sync) -- (toserver );
\path [line] (toserver) -- (server );
\end{tikzpicture}

logger in
the field

shuttle PC in field
download or office

local folder

rename file

copy to
Dropbox

ftp to sync
server Dropbox

Dropbox
to server

server
(=
backup)

The second example is a figure I made for describing the energy fluxes at the ice surface on
a glacier.
\begin{tikzpicture }[>=latex ,scale =0.8, every node /. style ={ scale =0.8}]
\fill[top color=cyan !20, bottom color=black !30]
(0,-1) -- (0,0) -- (10 ,0) -- (10 , -1);

\draw[thick] (0,0) -- (10 ,0);% Ground

\draw[<->] (5,-.9) -- (5 ,0);% Geothermal heat

\node at (5 , -0.57) [right] {$G$};

\draw[decorate ,decoration=snake ,segment length=11,->,blue !50! green]

(0 ,2.9) -- (0.8 ,0); % Incoming SW radiation
\node at (0.15 ,2.5) [right] {$I_S\ downarrow $};
\draw[dashed ,decorate ,decoration=snake ,segment length =11,->,blue !50! green]
(0.8 ,0) -- (1.7 ,2.9);
\draw[dotted ,decorate ,decoration=snake ,segment length =11,->,blue !50! green]
(1.2 ,0) -- (2 ,2.9); % Outgoing SW radiation
\node at (1.9 ,2.5) [right] {$I_S\ uparrow $};
\draw[dotted ,blue !70! white] (0.8 ,0) -- (1 , -0.4) -- (1.2 ,0);

\draw[decorate ,decoration=snake ,segment length=18,->,red !70! black]

(3 ,2.9) -- (3 ,0); % Incoming LW radiation

27
 \node at (3 ,2.5) [right] {$I_L\ downarrow $};
\draw[decorate ,decoration=snake ,segment length=18,->,red !70! black]
(4,0) -- (4 ,2.9); % Outgoing LW radiation
\node at (4 ,2.5) [right] {$I_L\ uparrow $};

\draw[decorate ,decoration=coil ,->] (5.5 ,2.9) -- (5.5 ,0); % Latent heat

\node at (5.5 ,2.5) [right] {$L$};

\draw [->] (6.75 ,.15) arc (280:0:.1 cm) -- +(283:0.05 cm); % Turbulence
\draw [->] (6.75 ,1.15) arc (280:0:.15 cm) -- +(283:0.05 cm);
\draw [->] (6.75 ,2.15) arc (280:0:.2 cm) -- +(283:0.05 cm);
\node at (6.75 ,2.8) [gray !50! black] {Turbulence };

\draw[decorate ,decoration=coil ,->] (8 ,2.9) -- (8 ,0); % Sensible heat

\node at (8 ,2.5) [right] {$H$};

\draw[->,blue] (9 ,2.9) -- (9 ,0); % Precipitation heat

\node at (9 ,2.5) [right] {$P$};

\node at (10 , -.6) [left ,gray !50! black] {Ice};

\node at (10 ,1.5) [left ,gray !50! black] {Air};

\end{tikzpicture}

Turbulence
IS ↓ IS ↑ IL ↓ IL ↑ L H P

Air

G Ice

The examples above require certain tikz libraries to be loaded in the pre-amble. Which are
required depends on what sort of graphic elements are required. The manual details this.
\usetikzlibrary{arrows ,shadows ,positioning}
\usepackage{pgfplots}
\usetikzlibrary{decorations.pathmorphing}
\usetikzlibrary{decorations.shapes}
\usetikzlibrary{patterns}

Although these examples may be daunting, they indicate the flexibility of LATEX.

Do it
• Visit the TEXsample.net site and try out a few examples. If possible try to make changes
to the code to familiarize yourself with the tikz programming.

7.11 xcolor
The xcolor package introduces colour to LATEX. In the hyperref and tikz packages we could
see colour already and and xcolor is in fact loaded already by those packages. The package is
otherwise loaded by
\usepackage{xcolor}

With this package you can access a large number of predefined colours. These are described
in the xcolor package documentation. In addition you can define your own colours using the
\definecolor command. The command takes three arguments, first the name you wish to give
to the colour, second the color space (cmyk, rgb etc.) and third the colour combination that
makes up the colour. The following define the official Stockholm University colour palette in
cmyk space

28
 \definecolor{SUBlue }{ cmyk }{1.00 ,0.70 ,0.00 ,0.60}
\definecolor{SUOlive }{ cmyk }{0.25 ,0.10 ,0.60 ,0.20}
\definecolor{SUSky }{ cmyk }{0.35 ,0.00 ,0.10 ,0.00}
\definecolor{SUWater }{ cmyk }{0.40 ,0.15 ,0.00 ,0.05}
\definecolor{SUFire }{ cmyk }{0.00 ,0.65 ,1.00 ,0.00}
\definecolor{SUSilver }{ cmyk }{0.12 ,0.08 ,0.08 ,0.23}
\definecolor{SUGold }{ cmyk }{0.30 ,0.40 ,0.80 ,0.15}

There are also ways to create hues out of the colours by mixing colours using the command
\color{}. To make up a hue we specify how much of each to mix
\color{green !40! red}

In the example the mix will be 40% green and thus 60% red (since they must sum up to
100%). Table 1 shows the official Stockholm University palette defined in CMYK through the
combinations given on the SU web site.

Table 1. The official Stockholm University colours in CMYK.

Colour name Colour CMYK

SU LATEX
Blue SUBlue 1.00 0.70 0.00 0.60
Olive SUOlive 0.25 0.10 0.60 0.20
Sky SUSky 0.35 0.00 0.10 0.00
Water SUWater 0.40 0.15 0.00 0.05
Fire SUFire 0.00 0.65 1.00 0.00
Silver SUSilver 0.12 0.08 0.08 0.23
Gold SUGold 0.30 0.40 0.80 0.15

Do it
• Define and use some of your own colours.

7.12 caption
The caption package provides easy tools for changing how the figure and table captions are
formatted but also extends on the native functionality. The package is loaded by
\usepackage{caption}

In the standard layout the the captions start with ‘Figure 1:’ and ‘Table 1:’ Since this part
of the caption is created automatically, you cannot manually change the colon to a period,
which is the format required by most journals, in the text. The following commands changes
the formatting of the caption start:
\renewcommand {\ figurename }{ Figure .}
\renewcommand {\ tablename }{ Table .}

Note that you can enter whatever formatting you want in the second argument. A perhaps
quicker way is to load the package with options for fonts
\usepackage[labelsep=period ,font ={small ,it},justification=justified ]{ caption}

With this you will receive a caption written in ‘small’ size italics font, and with the words
Figure and Table followed by a period. This is shown in ‘Figure’ 1 (which is just a caption).
Please refer to the caption package documentation for more details.

Figure 1. This is a trial caption to see how the different changes will look in the case of a caption in a
float

29
Do it
• Use the caption package to change the look of your captions.

7.13 float
The float package adds a placement option H to the figure and table environment (see section
5). The option is a stronger placement directive the the native options. The package is loaded
by
\usepackage{float}

7.14 amslatex
The amslatex package provides many improvements and extension to LATEX native mathematics
capabilities. The package is loaded by
\usepackage{amslatex}

There is no point to try to describe the package here. You need to carefully look at the
AMSLATEX information at the American Mathematical Society.

7.15 babel
The babel package contains ability to switch many built in features that are language dependent
to a language other than English. If you wish to write a document n Swedish you simply load
the package as
\usepackage[swedish ]{ babel}

The document will now be shown with many small differences to the English version. Certain
headings such as ‘Contents’ and ‘References’ will be in your chosen language, dates will be
written in the standard of your language, etc. Please refer to the babel package documentation
for details.
With babel it is also possible to write multilingual documents and switch between languages
mid-document.

7.16 lipsum
The lipsum package produces a body of dummy text in your document. The text is commonly
used in type setting to assess the look of the layout design and consists of a text body in
improper Latin. By adding a number or range of numbers as an optional argument (1–4 in
the example below) it is possible to display a specific number of paragraphs in your text. the
benefit of this is that you can quickly add some text to see effects of your type setting design
decisions without having to type anything into your document. The package is loaded by
\usepackage{lipsum}

and to add text to your document you add, for example,

\lipsum [1-4]

to where you want the text to appear.

Lorem ipsum dolor sit amet, consectetuer adipiscing elit. Ut purus elit, vestibulum ut,
placerat ac, adipiscing vitae, felis. Curabitur dictum gravida mauris. Nam arcu libero, non-
ummy eget, consectetuer id, vulputate a, magna. Donec vehicula augue eu neque. Pellentesque
habitant morbi tristique senectus et netus et malesuada fames ac turpis egestas. Mauris ut leo.
Cras viverra metus rhoncus sem. Nulla et lectus vestibulum urna fringilla ultrices. Phasellus
eu tellus sit amet tortor gravida placerat. Integer sapien est, iaculis in, pretium quis, viverra
ac, nunc. Praesent eget sem vel leo ultrices bibendum. Aenean faucibus. Morbi dolor nulla,
malesuada eu, pulvinar at, mollis ac, nulla. Curabitur auctor semper nulla. Donec varius orci
eget risus. Duis nibh mi, congue eu, accumsan eleifend, sagittis quis, diam. Duis eget orci sit
amet orci dignissim rutrum.

30
8 Managing references
LATEX has the capability to handle literature references and generating reference lists in your
documents. To accomplish there a few points we need to cover.
You need to have a data base of the literature you wish to cite in your documents. This
means that you should start building a data base of references. There are many reference
managers available such as EndNote, RefWorks, Zotero, Mendeley and JabRef to mention a
few. The Comparison of reference management software page provides an overview of available
tools. What you should look for is that the tool can export BibTEX format, can provide a
unique BibTEX ‘cite key’ (see below) for each entry, is continually supported, supports your
platform and, I would suggest, is free and open.
I use and recommend the Open Source Zotero for your reference handling. With this tool
you can save reference information directly into Zotero and quickly build your reference data
base. Zotero works with a cloud based storage that is free up to 300 Mb. You need to install
Zotero and a connector for your web-browser (Chrome/Firefox/Safari). You should also install
the Better-BibTeX for Zotero plugin. Just follow instructions to do so.
Zotero allows you to also store pdf files with your references. this will, however, quickly fill
your free data base space. I personally save only the reference information and at some point
had 2974 references in Zotero which used up 0.3% of my 300 Mb. So the free space will be
sufficient for quite some time as long as you store only the reference information and not pdf
files.
To work with your data base in your document, you also need a package that can handle
author-year referencing, the standard format in Physical geography and related fields, and also
a reference format so that your reference list is correctly formatted.
The goal is to be able to relate any article you reference in your manuscript with an entry
in your data base. We also want all references we added in the document to correspond to a
reference in the reference list. We of course also want the reference list to correspond exactly to
what we want to reference in the text. All this will be accomplished by LATEX with the natbib
package.

8.1 The natbib package

When you want to use author-year style referencing you need to use the natbib package (section
7.4). This package adds many new commands dealing with different types of references.
When you make references in the text you basically need two forms for the reference.

active where the author name is part of the text and the year is within parenthesis as in ‘Smith
(2013)’
passive where both the author name and the year are within parenthesis as in ‘(Smith 2013)’
The natbib package provides two commands for this, \citet{} and \citep{}, respectively.
As with other cross-referencing the referencing works through labels or cite keys. Each
reference must therefore have a unique cite key. With Zotero the cite key will by default be
created from the first authors last name, the first word in the title (Ignoring words such as
‘The’ and ‘A’) and the year. A reference such as
Smith AZ, 2013. Bad science – surviving in the jungle. Journal of Long-winded Pseudoscience,
23, 214–523.
will be assigned a cite key Smith_Bad_2013. this means you can reference this article in the text as
either \citet{Smith_Bad_2013} or \citep{Smith_Bad_2013}. Either one of these cite commands
will enable LATEX to add the reference to the reference list. Hence you simply add references
and the reference list is built for you.
The natbib package has several commands for referencing but the most commonly used are
shown here

31
 Command Result
\citet{} ‘active’ citation ‘Smith (2013)’
\citep{} ‘passive’ citation (Smith, 2013)
\citeauthor{} cites only author name, not year
\citeyear{} cites only year, not author name
In all cases the BibTEX key should be entered as argument to the commands.
If you need to reference many authors in a passive reference, you simply add them all
as argument separated by a comma. \citep{Smith_Bad_2013,Day_Two_2014,Knight_Three_2015}
will yield (Smith 2013; Day 2014; Knight 2015).
The \citet{} and \citep{} commands can also take optional arguments. If we want
to have a reference that looks like (e.g., Smith 2013; and references therein) we can write
\citep[e.g.,][and references therein]{Smith_Bad_2013}. What we want added before the ref-
erence goes as the first optional argument and what should be placed after the reference goes
in the second. If you do not want one or the other, you just leave the argument empty.
If you make a spelling mistake in the BibTEX key in your document, you will see this as two
question marks in the text (??). This is how LATEX signals such errors. It will also appear as a
warning in your log file.
With these commands you should be able to add references to the text but in order for the
system to work, you need to provide somewhere from which BibTEX can obtain its information
and this is where we have two ways to go, the fully and the semi automated way. I will cover
both but there is little point in making the semi automated system your standard.

8.2 The fully automated reference system

The fully automated system handles both the cross-referencing and builds the reference list for
you. In order to do this you need to have several things in place. First you need to have your
data base as a BibTEX .bib file in your project, including all references you need to reference
in your document. You also need a file that tells BibTEX how to format your references. This
is called a BibTEX style file (.bst). It is possible to custom make your own style but that is
not something that is easily described here. Instead you can browse CTAN’s repository for
bibliographic styles and try them.
Journals that take LATEX manuscripts usually have their own styles that can be downloaded
or come with their template files for the journal. Otherwise, I would otherwise recommend
going with the Council of Science Editors (CSE) style. It is a neutral international standard
that is simple.
When you have your data base file and your style file, you can start using it in your document.
You need to add the following code where you want your reference list to appear.
\bibliography{foo}
\bibliographystyle {CSE}

The first command \bibliography{} tells BibTEX which .bib file to look for, your bibliography
file. The file name does not need to be written out with the extension .bib and you can add
a path if your bibliography file is not located with your LATEX document file. The second
command tells BibTEX which bibliographic style that should be uses, the name of your .bst
file here exemplified by the CSE.bst (Council of Science Editors).
With all this in place your reference list should be up to date whenever you add a reference
to a new publication in your document with the different cite commands and you should not
need to worry about any discrepancies in the document. As mentioned earlier, a prerequisite
is that the reference you you reference must be in the data base file and all entries in the data
base must have unique BibTEX styles.
Note that if you add references in your data base you will need to export the database again
and replace the earlier .bib file with the new updated version to be able to make use of new
references added to the data base.

32
Do it
• Add a few references to a .bib file and cite them (using natbib citing commands) using
the fully automated method described above.

8.3 Integration of Overleaf and Zotero

When using Overleaf you can connect Zotero to Overleaf. This is done by coupling your Zotero
account to your Overleaf account in the account settings in Overleaf. Once you have completed
this you will be able to transfer data directly from Zotero into Overleaf documents.
When you need to reference articles in your data base in a document you first need to
obtain a copy of the data base in .bib format to reside in your Overleaf project. When you
press ‘Upload’ in your Overleaf project view to add new files, you will see several options in
the pop-up Window that appears. One option will be ‘From Zotero’. This will allow you to
transfer your online Zotero data base into you Overleaf project. When you make the selection
you will also be asked to name the file. The file name you provide is what needs to go in the
\bibliography{} command. You then also need to make sure you have the .bst file of your
choice and the correct file name entered in the \bibliographystyle{} command.
When you transfer the data base from Zotero to Overleaf, the process may take a little while
depending on the size of your data base.
Whenever you update your data base in Zotero, you need to update the file in your project(s)
to access these new references. This is very easy to do. You simply click on the .bib file name in
your project view. This displays the data base file in Overleaf and you then have the possibility
to ‘refresh’ the data base. This transfers the latest version of the data base to your project.
You then have access to any new items added to your data base.
Sometimes the updated .bib file does not seem to be immediately recognised by Overleaf.
The problem sorts itself out given a little time.
When you work with the integration between Zotero and Overleaf, the data base will be
provided Zotero default cite keys. These have the format [auth]_[veryshorttitle]_[year]. If
you use a different format in Zotero, the cite keys will be changed to this format in the transfer.
At least at the time of writing this there is no way to work around this with the direct transfer.

8.4 The semi-automated reference system

The semi-automatic system does not make use of a reference manager or an external bibliogra-
phy file as discussed above. In the semi automated referencing system, LATEX keeps track of the
cross-referencing but you need to manually type in all references the way they should appear.
This means that you need to create a reference list at the end of your document following a
specific format.
\begin{thebibliography }{}

\bibitem[Smith , 2013]{ Smith_Bad _2013}

Smith , A.Z., 2013. Bad science -- surviving in the jungle.
\textit{Journal of Long -winded Pseudoscience}, 23, 214 - -523.

\end{bibliography}

As you can see the bibliography has its own environment within which you put all your refer-
ences. each reference must start with the command \bibitem[]{} which takes two arguments.
the first argument is the in-text reference, the information you want LATEX/BibTEX to place
in the text. Note that the comma is not a style, it is for LATEX to separate author and year.
The second argument is the BibTEX key that is unique for all references. This is what natbib
uses to link a unique reference in the list with a natbib cite command in the text. After the
bibitem command follows the reference formatted the way it should look according to any
author instructions that may be given for the reference list.
Using the system this way is very simple but has the draw back that while all references in the
text will be matched by a reference in the reference list, there is nothing that prevents you from
adding references in the list that are not matched in the text. This means the cross-referencing
will always be correct but LATEXcannot do anything about the content in the reference list.

33
That will be up to you. You also have to manually type in the references and check that each
gets its unique cite key.
So to sum up, this way of using references in LATEX is useful to know but the fully automated
system, described below, is what you need to aim for.

Do it
• Add some references using the semi-automated method described above and cite them
(using natbib citing commands) in your document.

9 Automating LATEX or creating your own commands

Since LATEX is programmable you can accomplish almost anything. The problem is that it can
become complicated. There is, however, a simple way to make life easier through the command
\newcommand. This allows you to create your own command from the very simple to the very
complicated. In its simplest form you can use the command to create a new command that will
type something complicated for you in an easier way. Take δ 18 O, for example. To type this you
need to provide
$\ delta ^{18}$O

So if you were writing a manuscript on stable isotopes, you may get really tired of repeating
this over and over. We can then create a new command that types this out for us
\newcommand {\dO }{$\ delta ^{18}$O}

In this case all you need to type in your manuscript i \dO and you will receive δ 18 O.
When you create a new command, it is important not to use an existing command name.
LATEX commands are always in lower case and LATEX is case sensitive. I therefore suggest you
write yours with some capital letters, so called camel case or Pascal case (after the programming
language Pascal). If you, for example, wanted to create a command called ‘do it now’ this would
be written as \DoItNow, that is the first letter in each word is capitalized.
The \newcommand can be used with additional input. Let us look at the new command \Nada
defined by
\newcommand {\ Nada }[3]{\ textsc {#1} \textit {#2}\\ \textbf {#3}}

The first argument defines the command name \Nada. The second tells LATEX that there will
be three arguments. The third argument shows LATEX what to do with the three arguments.
In our case the first given by #1 will be set as small caps, the second as italics and the third on
a new line as bold. With input like
\Nada{What On Earth }{ kind of joke typesetting }{is this ?}

we get
What On Earth kind of joke typesetting
is this?
Even though the example is pointless in content the point that you can automate repetitive
and boring tasks is there. The newcommand can contain up to 9 arguments.
There is also a related command \renewcommand that allows you to use an existing command
name and redefine it. This can be very useful but also quite ‘dangerous’ if you do not know
what you are doing.

Do it
• Create your own new command. Try to replicate
this using the
marginpar (section
3.12) as a starting
point

34
10 Copying documents from Word to LATEX
Why have a section on moving Word content to LATEX? As you probably have understood there
are some significant differences between how Word works and how LATEX works. Most evident
is the fact that while Word does all formatting in the hidden, most such work is upfront in
LATEX. This can cause much grief when trying to copy text originally written and formatted
in Word to LATEX. Since you are likely to encounter persons who do not work with LATEX, you
may find yourself in a situation where most of a document is prepared in Word and left for
you to, for example, transfer into a journal class template. Or, you may have an old favourite
document you wish to move to LATEX. In this section I will provide some hands-on guidelines
for transferring material between the two tools as smoothly as possible.
Before continuing I need to mention that there are tools that translate Word files into LATEX.
The tools are usually good at replicating the look and feel of the Word document in LATEX,
but since that is rarely what we want, we end up with a code that mimics Word documents
well but with a LATEX code that is unnecessarily complicated. I recommend you to try such
tools just to get a sense of what they can do but in the following we will work on the principle
of preparing the word document so that the content can be copied using ‘copy–paste’ between
Word and your LATEX editor.
The first thing we need to realize is that no Word-based formatting will carry over into the
LATEX editor, only the text and the returns indicating end-of-paragraph. What order we make
the following changes is immaterial, each find their own praxis. In a way, one can say that
anything that has required some form of formatting in Word needs to be changed.
The basic strategy I recommend is to ‘LATEXify’ your Word document already in Word before
cutting and pasting it into the LATEX editor.
Paragraph breaks LATEX uses empty lines to identify paragraph breaks, Word does not. You
need to go through and add empty lines between all paragraphs otherwise, you will have
a hard time finding them in the LATEX editor.
Italics and bold face The italics and bold face typesetting will not be carried over. It is
much simpler to go through the Word document and look for your italics and bold text
and use the commands \textit{} and \textbf{} to mark those words in Word than in
your LATEX editor where the formatting will be lost.
Section headings The section headings formatting (levels) will not carry through to LATEX
(unless they are numbered so that you at least can identify them). This means it is easier
to insert the commands \section{}, \subsection{}, and \subsubsection{} in the word file
prior to copying.
Mathematics If you use equations and you have formatted variables with super and subscripts
in the text, you should rewrite these using the mathematical mode already in Word.
Scientific units Mixing text and mathematical mode to produce super and subscripts when
writing scientific units can be quite a mess. I strongly suggest you retype units using the
commands in the siunitx package.
Dashes You need to replace en-dashes in Word with the -- (double dash) format used by
LATEX.

Tables There is no easy way to prepare the tables for LATEX in Word. Just remember that any
fancy formatting will be lost so it is best to keep the table in as simple form as possible.
Complex tables are hard to typeset anywhere so this is where you probably need to spend
some time.
A useful tool to try is excel2latex, a plugin for Excel that allows you to save LATEX code
from a section of an Excel spreadsheet. If you take your tables from Word into Excel then
this tool will yield a good basis for your table, quick and easy. At the time of writing
the macro should work with Excel 2012–2016 (both 32 and 64-bit). In fact if you are
uncertain about the coding of tables, this macro may be a good way to learn by making
a table in excel and then saving it for LATEX.

35
Figures Since you cannot copy graphics from Word into LATEX the only thing that will be
copied is the figure caption. If you have your figures in Word and cannot produce originals
for LATEX you can right-click on the figure and save it as a picture. Use PNG format for
line graphics and JPG for photographs if you do so.
References If you have used EndNote or some other reference manager to produce references
and reference list in your Word document then all such couplings disappear when you copy
the text. Your reference should, however be correct unless you decide to do some serious
editing in LATEX after copying. You basically have two ways to deal with the references.
One is to stick to the manual way and simply manually check for correct cross-referencing.
The other is to load the natbib package and either enter/import references into BibTEX
and use that system or go halfway and use the \bibitem and different \cite{} commands
to produce the cross referencing. Either way this could cost you some time.
Extended type faces Depending on what system you use for compiling your LATEX document,
extended type faces may be come an issue. This stems from the introduction of type faces
that contain a much larger number of characters than used by older computers. XeLATEX
uses these modern type faces and may compile a text containing such extended character
sets correctly while pdfLATEX generally cannot. Switching to XeLATEX may solve issues
and for pdfLATEX there are work arounds (see the pdfLATEX documentation). However,
care should be taken when moving text containing, e.g. greek letters or mathematical
signs to ensure they become type set correctly.
There may be additional details in the Word document that either do not translate well or are
simply lost but the list above should normally cover 99% of a normal scientific document in
Word.

11 Making presentation slides in LATEX

LATEX can be used for much more than creating regular documents. There are several different
classes for creating presentation slides in LATEX. The most popular is called ‘Beamer’. Beamer
is the German ‘term’ for a computer projector and has been so named by its German author.
Creating a simple Beamer presentation is not difficult. A single slide is made by the following
code
\documentclass{beamer}
\begin{document}

\begin{frame}
\(a=b\)
\end{frame}

\end{document}

What you see is that you need to make sure you use the beamer class. A slide is created by
an new environment called frame. anything you place within the frame environment will be on
your slide. Adding new slides is done by simply adding a new frame environment before or after
the first.
There are of course many things you can add to your basic beamer presentation. The
command \framettitle{} can be placed in a frame environment and will then produce title for
that slide. As with a regular document you can add a title page by providing the commands
\title{} and \author{} in combination with \titlepage (note not \maketitle). A two page
presentation with a title page can thus look as
\documentclass{beamer}

\begin{document}

\begin{frame}
\title{Presentation title}
\author{Your name}
\titlepage
\end{frame}

36
 \begin{frame}
\frametitle{The title of the frame}
\(a=b\)
\end{frame}

\end{document}

Beamer comes preloaded with a series of colourful layouts. These can be invoked by using
the \usetheme{} and \usecolortheme{} commands. The themes and colour schemes have specific
names which are described in detail at the Beamer theme gallery. You can also create your
won layouts but that requires a little bit of involvement.
In the end you should consult the Beamer class user’s guide to learn more about all the
possibilities available in Beamer.

Do it
• Make a short Beamer presentation consisting of a title slide and a couple of text and
figure slides.
• Add some colour by using a few ‘beamer themes’.

12 Reinventing the wheel? …or not

As with all programming languages, there will be bits and pieces of code that you will need over
and over. You will soon find several such pieces as you develop your own ‘style’, good examples
are all the packages that you will soon learn not to be able to live without. To prevent having
to re-enter all this information in every new document you use, you can start to build up your
own ‘style’ file.
LATEX allows you to read in external files into the file you are authoring. This could literally
be anything from a file containing some plain text to a file containing more or less a complete
document. There are several commands for entering files such as \input{} and \include{},
where both commands take a file name as argument. These two commands are quite similar
but differ in that the include command adds the content on a new pages whereas input simply
adds the input text right where the command occurs. But a third possibility is to input a file
of commands, to replace much of the preamble. This is accomplished with the now familiar
\usepackage{} command and where your file should be named with the extension .sty.
What you need to do to start building your own style file is to add all the preamble infor-
mation you think you will use repeatedly and unaltered into a file fname.sty. This file must not
contain the \documentclass and \begin--\end{document} commands and environments or any
text. It can only contain material you would normally put in the preamble. If we assume you
have a file you are working on and you take everything in the preamble and paste that into a
file called MyStyle.sty then you bare document will look like the following
\documentclass{beamer}

\usepackage{MyStyle}

\begin{document}

\end{document}

The \usepackage command will then take your file and use its content in order to format the
current document.
It is important to realize a couple of things here. First, you will always need to place a copy
of your style file along with whatever document you are working on. LATEX will not keep track
of it for you. Second, if you enter additional instructions in the preamble, you need to make sure
they do not interfere with your style file. If you add something before the \usepackage{MyStyle},
you run the risk of having whatever you want to do over-ruled by the content in your style and
if you add something after the call the new material may over-rule something you have decided
will be part of your style.

37
 It is possible to add some additional safety to the style file so that it will only be used with
appropriate versions of LATEX etc. We can start the style file by adding the following commands
\NeedsTeXFormat{LaTeX2e}
\ProvidesPackage {MyStyle }[2016/04/01 my LaTeX style]

This tells LATEX that the latest format of LATEX (LATEX2e) is needed and that this is ‘my’ style
file of such and such date. At the end of the file we can add
\ endinput

which tells LATEX that the style file content is ended. This command means you can add more
material to the file after the \endinput command but which will be ignored. This can be useful
if you add material you are unsure you want to use but do not want to lose.
So a complete style file should look like the following example
\NeedsTeXFormat{LaTeX2e}
\ProvidesPackage {MyStyle }[2016/04/01 my LaTeX style]

% your preamble material :

\ endinput

It is possible to take the style file much further and essentially build your own class. This
is beyond the scope of this text but the document LATEX2e for and package writers provides
much more in depth information on the subject.

13 How does it really work?

For those who wonder how the LATEX system really works, I will provide a brief overview of
what goes on when a document is being typeset by the LATEX engine. First we should recognize
that beneath LATEX is the original typesetting language TEX created by Donaldh Knuth in 1977.
LATEX is essentially a, albeit huge, set of macros that facilitate type setting of documents of
various kinds. The name LATEX is short for Lamport TEX after Leslie Lamport who wrote the
LATEX system.
In its most basic form a LATEX text document is compiled by feeding a TEX compiler the
document and the LATEX format macros to generate a so-called Device Independent file (DVI).
This file can then be used by different drivers to show the document on screen or to print it on
a printer or create a pdf. To respond to different needs variants of the LATEX compiler has been
created. One such is pdfLATEX which directly generates a pdf-file as output without creating an
intermediate DVI file. pdfLATEX is probably the most widely used way to generate documents
in LATEX (Figure 2).

LATEX .dvi .ps

LATEX dvi2ps ps2pdf PDF file
document

LATEX
pdfLATEX PDF file
document

Figure 2. The basic way to generate a pdf file from a LATEX file. (upper work flow) Generic work flow
running LATEX to generate a DVI file followed by running dvi2ps to produce a postscript file from the
DVI and then ps2pdf to generate the pdf from the postscript file. (lower work flow) Work flow running
pdfLATEX to directly generate the pdf file.

Since LATEX in its generic form cannot easily handle system fonts a variant called XeLATEX
has been created. XeLATEX can access system fonts in a more direct way but because there
are internal differences between pdfLATEX and XeLATEX we must be careful when trying to use
certain packages in documents. Some (most) work with pdfLATEX while some require XeLATEX.

38
This is the price paid by maintaining backwards compatibility. However, any problem is easily
handled with some understanding of the two systems and by not overusing packages that we
do not know anything about.
The way LATEX handles information that requires recursive treatment is to store results in
external files (Table 2). This applies to, for example, the table of contents, figure and table
references, reference citations. The first run of LATEX generates lists of all such cross-referencing
information while LATEX type-sets the document. This way the location of each cross-referenced
object becomes known. LATEX then needs a second run to find all references to the objects and
replace these locations with the number of the figure or table or reference of the article. The
result is that the document is compiled, not once, but in fact several times in order for all
cross-referencing information to be properly located. Once this is accomplished you will find
several files generated by this process that contains information from the process.

Table 2. Some of the more common files (extensions) generated when compiling a LATEX document.
Note that some files are generated only when specific implementations are used.

Extension Description
.aux Generic information, mostly cross-referencing
.bib BibTEX reference data base file
.bbl bibliography produced by BibTEX
.bst BibTEX bibliography style file
.blg bibliography (BibTEX) log file
.cls LATEX class file
.dvi device independent file
.lof list of figures
.log complete compilation diagnostics reported by LATEX
.lot list of tables
.out hyperref PDF bookmarks
.sty LATEX package file
.tex LATEX or TEX document file
.toc table of contents

Depending on what you include in your document LATEX will need to be run several times
and maybe also involve other supplementary programs such as BibTEX. As an example, we
can view the series of runs necessary in order to produce a table of contents and a reference list
using the fully automated bibliography (Figure 3).

LATEX .toc .bbl

pdfLATEX BibTeX pdfLATEX pdfLATEX PDF file
document .aux .blg

.bib .bst

BibTEX
data base
ref. style

Figure 3. The pdfLATEX work flow for a document containing a table of contents (generating the .toc
and a reference system based on a a BibTEX data base. Note that pdfLATEX needs to be run twice
after the BibTEX run in order for the cross referencing and reference list to be inserted correctly in the
document. Many, if not all, LATEX editors take care of this work flow automatically.

14 Saving the worst to last

Errors, the one thing we all do not want to see. It is common to get errors in LATEX, some are
simple to fix others may seem inexplicable. The one thing they have in common is that we have

39
done something wrong. It is therefore useful to lay down a few ground rules to, first, reduce
the numbers of errors and, second, to see how we can solve them. The best way to solve errors
is to make sure they do not happen.
The first rule is thus, be meticulous when you write your document. The by far most common
error comes from having forgotten to pair { and }. The second most common error is that you
misspell a command. A third, concerns mathematical mode; you have either forgotten to switch
back to text mode or you use a mathematical command in text mode. In all cases, the errors
should be quite easy to spot and correct.
The second rule is to write slowly when you are doing more advanced LATEXing. When
you have some complex code you need to enter, do not rush and put a lot in at once. Instead
complete the code bit by bit and make sure each bit works before you continue with the following
parts. A useful tip is to maintain an empty document, I usually call test.tex, where I can build
and test code before pasting it into the main document. This has two advantages, one is that
you are working only on one piece of code in the test document and the other is that a small
document compiles much faster than if you work in your main and most likely much longer
document.
Sometimes, you will come across errors that may be ridiculously difficult to spot and to solve.
Such errors can come about for a multitude of reasons but, remember, you entered something
into the document which did not agree with LATEX, directly or indirectly. Therefore, go back
to what it was you just entered and try to figure out why this may have caused an error. Is
something incompatible? Have you misunderstood how a certain command works? There is no
easy answer as to how to solve these problems, you need to figure out what you did to cause
it. If it is possible, try to remove the part in the document that appears to cause the problem
and put it into a separate file (the test file) and see if you can replicate the error there.
Occasionally, you end up having a recurring problem that just does not seem to go away
and you cannot find a single problem. One detail to remember is the set of files generated
during a LATEX run (Table 2). It sometimes happens that one of these files contain material
that causes the problem. If you have run LATEX and then made some changes that will also
change the content of these files, there may be conflicting information left from earlier runs. It
is therefore useful to make a clean compile of the document by first removing all the temporary
or intermediate files. You need to be careful, however, since they all will have the same main file
name and just differ in their extension. What you should remove are typically the .aux, .bbl,
.blg, .dvi, .lof, .lot, .log, .out, and .toc files; if they exist. You should at all costs avoid to
remove any .tex, .bib, .bst, .cls and .sty files, of course. Many editors can do this for you
safely. In Overleaf this is called a recompile. If the problem persists after such a recompile,
then you unfortunately still have a problem in your document.
With time, and as you have created enough errors to solve, you will improve your problem
solving ability. This may seem as a cumbersome way, but it is the price we pay for the flex-
ibility and power of LATEX, and something quite familiar to those who do programming. As
stated earlier, the best way to avoid errors is to not make them so write slowly and carefully,
particularly when you enter commands., and compile your document frequently to that you see
when an error appears.

Do it
• It is tempting to write here that you should create your own error, but most likely you
have already done so in the process of trying things out. So, try to solve your errors
instead.

40
Appendix
A Running LaTeX on your own computer
This section outlines the basic installations necessary to run LATEX on your own computer. In
order to run LATEX you need an implementation of LATEX, a distribution, and a LATEX-compatible
editor.

A.1 LATEX
Which LATEX-distribution you chose depends on which operating system platform you are on.
For Linux, you most likely already have LATEX already installed. For Mac and Windows, you
definitely need to install it. There is one implementation call TEXLive which can be installed on
all platforms; on Macs using its Mac version MacTEX. The web page provides good instructions
on how to download and install the distribution. For Windows there is also a distribution called
MikTEX. I would recommend installing MikTEX on Windows platforms. MikTEX has utilities
to easily update LATEX. It is also possible to create a portable LATEX-environment by installing
on a memory stick.
A complete distribution of LATEX can be large (1–2 Gb). I would nevertheless recommend
to make a full installation rather than a partial. The reason for this is that LATEX contains a
huge number of so-called packages that simplifies tasks and provides new features to the basic
functionality. If you install the entire distribution, you are sure to have them within reach
if you need them. It is, however, also possible to make minimal installations and have the
LATEX-implementation download packages as necessary (applies at least to MikTEX).
LATEX in its basic form is a command-line driven software. It is thus possible to use it in
command line mode. This is not a very smooth way to use it and so a front-end editor is a
useful complement.

A.2 LATEX editors

Since a LATEX file is a simple text file any editor can be used to write a LATEX document as long as
you remember to save it as a plain text file with the extension .tex. Since LATEX contains many
specially formatted commands, it is very handy to have a designated LATEX editor. Fortunately
there are many such editors and most of them freeware. I can recommend two similar editors
called TEXstudio and TEXmaker. These editors are implemented on all three major platforms.
In fact, TEXstudio is a separate version of the original TEXmaker and they therefore have
very similar characteristics. TEXstudio is completely open source while TEXmaker is free but
maintained by a closed group of persons. The choice therefore largely depends on which version
you like the most. TEXstudio has the advantage of being more customizable than TEXmaker.
Another popular editor is TEXnicCenter. The MikTEX distribution also contains its own simple
editor which is installed along MikTEX.. A good shareware editor is WinEdt. All editors should
work with whatever implementation of LATEX you use. WinEdt is not exclusively intended for
LATEX and can also be used for other programming languages. For Linux users, there are good
ways to use the Emacs editor.
TEXstudio allows you to use spell checking (and thesaurus) from Openoffice. You can
download spell checking dictionaries for (currently) 101 languages (not that you need them)
from the OpenOffice extensions page page. Simply search on the word ‘dictionary’ and the
language you are looking for. Please note that spell checking files for British and US English
as well as French and German come with the TEXstudio installation. The files you download
will have the extension .oxt but are in fact Zip-files. You can therefore rename the file to .zip
and unzip the files into the ’dictionaries’ folder in your TEXstudio installation and, presto, you
have a new language installed. Please check the TEXstudio manual for details.

A.3 Other software that may be required

For those who chose MikTEX and WinEdt you must also download the free PostScript inter-
preters GhostScript and Ghostview. Check GhostScript for the latest versions. These allow
you to handle documents with postscript graphics.

41
A.4 Reference management
LATEX contains its own reference management system called BibTEX. BibTEX is a database
system for references that consists of a standardized database format for reference material.
Many other systems can rad and export to BibTEX, such as EndNote, and may databases
including Web of Science. To manage your database you will need some form of software.
There are two ways forward, either you chose an on-line service such as Zotero or Mendeley, or
you go for a local service installed on your computer.
The online services Zotero and Mendeley provide means to build reference databases by
accessing material directly from the web and is probably the best choice if you have not begun
building such a database with any particular software. These services can provide the necessary
information to LATEX to enable you to reference data from your data base directly in your
document. The data is stored on-line but you also have the option of installing a standalone
version that can be synchronized with the on-line version. This is of course most valuable if
you wish or must to work off-line.
If you want to keep your reference database only on your computer I strongly recommend
JabRef. JabRef is a Java-based reference manager. This also means it works equally well on
all platforms and it is free! JabRef installs through the downloaded installer. Since JabRef is
Java-based, it will also require Java to be installed on the computer. In Windows, you will be
prompted to download it if you try to start your already installed JabRef without Java. Java
installs easily through an interactive installer.

A.5 Yet more software that might be of use

Converting documents from Word to LATEX or vice versa is not difficult in principle but can be
really tedious. There are some tools that can be helpful for such conversions. One of the most
advanced is the GrindEQ suite of applications. GrindEQ is not free and whether or not you
find it useful is a matter of preference. You should try it out and see what you think. My own
experience is that it is useful to a point. The LATEX created from a word file still requires a bit
of editing to be usable. Most often manual editing is the best option since it gives you a chance
to go through the document more carefully.
If you need to create tables, you can make use of a excel2latex Excel macro. This installs
in Excel and allows you to generate table output for inclusion in LATEX. Currently the macro
is supported up to Office 2012–2016 so use in later versions will have to be tested.

B Install LATEX on your computer

B.1 Windows
Installing LATEX and the editor of your choice should be a breeze. In the case of MikTEX
and TEXmaker you basically run automatic installers. MikTEX requires you to download an
installer (be sure to chose 32 or 64-bit depending on your system) from the web page and to
run it. You can chose to either download LATEX or to install it directly. Depending on how
fast or steady your internet connection is, you could chose either way but I recommend to use
the download option. If you chose this option, make sure you download a full version and also
note where the installer saves the file. You will find it in a folder called MiKTeX 2.9 Setup
(depending on which is the current version you download). Once the download is complete
you open the folder containing the downloaded files and start the installer in that folder (e.g.
basic-miktex-2.9.7152-x64.exe, again, depending on which is the current version). Running
this installer will install LATEX.
The editors install without problems by just running the installers.

B.2 Mac
The simplest way to install LATEX on the Mac is via MacTEX . The site provides the MacTex.pkg
file to install everything needed for running and maintaining your LATEX installation. There is
also an optional MacTeXtras.zip which provides some additional tools for editing etc. Once

42
installed you will find a Preference Pane for the selection of your TEX distribution (i.e. ver-
sion) but you can ignore this unless you have an older installation on your Mac. The basic
MacTex.pkg installer provides BibDesk, LATEXiT, TEXShop, TEXWorks and Excalibur as well
TEXLive Utility. The TEXLive Utility is used for updating and installing new packages (func-
tions or tools) and can be found under /Applications/TeX/TeX Live Utility.app.
The other tools installed are useful but there are alternative applications available that may
be more to your liking. TextWrangler is a very useful, freeware text editor that can parse TEX
using a set of scripts. Alternatively, Latexian (available through the App Store) is a very clean,
’Mac-like’ program that provides live previews of the compiled text as well as a good navigation
interface.

B.3 Linux
A basic TEX-environment is included in most Linux distributions.

C Learning more on LATEX

As with all high quality and hence popular freeware, the internet is a very good source for
more information. There are several sources for information that deserves mention: The LATEX
Project home (see additional links on the site), the TEX StackExchange, and perhaps the LATEX
Community. The TEX StackExchange is highly recommended because you can easily search for
older asked questions, pose new ones (if searching the site did not yield an answer) and take
part in discussions. In fact if you do a search with the keyword ‘latex’ and something you are
looking for, StackExchange entries will inevitably be high on the search.
In addition to the web-based sources above, the Comprehensive TEX Archive Network
(CTAN) repository contains both files and manuals for thousands of packages that solves spe-
cific tasks in LATEX. The two main distributions MikTEX and TEXLive both use CTAN as their
source for packages so if you have installed a complete set-up of these you also have most of
what is on CTAN installed and ready to use.
Getting to know the details of LATEX may be a little difficult because of the wealth of
material available on the internet. There are, however a few books that may be of use such
as Mittelbach et al. (2008), Voss (2010), and van Dongen (2012). Mittelbach et al. (2008) is
a comprehensive book about the core features and packages of LATEX. It works well both for
learning but also to be used as a dictionary. Voss (2010) is also mostly a reference handbook
of different commands. It does not provide as much explanations as Mittelbach et al. (2008)
but is well organized and much more condensed. van Dongen (2012) includes discussion of
a very useful subset of packages in LATEX. It is a very good start and overview that is hard
to find elsewhere but does not go into depth. An additional book worth mentioning is Voss
(2011) which is devoted to typesetting tables. One book that also deserves mentioning but for
a different reason is Kottwitz (2011). This book is aimed to be an introduction to LATEX and
will only work as such. It is not very useful as a reference after its introductory values. You
will also find many introductions to LATEX on the web, the most used is probably the so-called
lshort, ’The Not So Short Introduction to LATEX’. There is also a Wiki book on LATEX which
is under development.
… and then there is this document.

References
Kottwitz, S., 2011. LATEX Beginner’s Guide. Packt Publishing, Birmingham. 336 p.
Mittelbach, F., Goossens, M., Braams, J., Carlisle, D., and Rowley, C., 2008. The LATEX
companion. Second edition. Addison-Wesley, New York. 1090 p.
van Dongen, M.R.C., 2012. LATEX and Friends. Springer, Berlin. 299 p.
Voss, H., 2010. LATEX Quick Reference. UIT Cambridge, Cambridge. 229 p.
Voss, H., 2011. Typesetting tables with LATEX. UIT Cambridge, Cambridge. 208 p.

43
D Revision history
v.3. 15 Oct., 2019 The part on referencing is completely rewritten with notes on Zotero and
Overleaf 2 integration included. Many minor typos and errors in v.2 have also been
corrected throughout the document.
v.2. 19 Sep., 2018 Added appendix on how to install LATEX on a personal computer. Minor
typos in v.1 corrected. Adjusted for Overleaf v2.
v.1. 29 Apr., 2016 First public version

44