Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss

No italics in \newtheorem environment?

6,041 views
Skip to first unread message

WJ Bland

unread,
Jan 25, 2001, 2:59:26 AM1/25/01
to
Hello,
I'm currently writing up my thesis and I'm having problems
with getting LaTeX to do what I want. I think it's probably easy to do -
when you know how! At the start of my file is the following:

\newtheorem{theorem}{Theorem}[chapter]
\newtheorem{lemma}[theorem]{Lemma}
\newtheorem{corollary}[theorem]{Corollary}
\newtheorem{definition}[theorem]{Definition}
\newtheorem{example}[theorem]{Example}

so I can write things like:

\begin{lemma}
If something is true then something else follows.
\end{lemma}

\begin{definition}
Let $A$ be a something. We say $A$ is a \emph{something else} if
some conditions hold.
\end{definition}

These work fine (note that the numbering for all theorems, lemmas,
definitions, etc. is in the same sequence, which is what I want).

Unfortunately the italics in definitions are the opposite way
around to how I want them, because (it seems) any text in an environment
created by a \newtheorem statement is in italics.
I've tried writing \begin{definition}\emph{...}\end{definition}
which has the disadvantages that (1) I'd have to change every definition
(there are well over 100) and (2) any maths symbols in "...", don't
appear in italics any more!

I'd be greatful if anyone could show me how to correct this.
Thanks very much for your help,
William Bland.

Pieter Edelman

unread,
Jan 25, 2001, 3:50:41 AM1/25/01
to
> Unfortunately the italics in definitions are the opposite way around to
> how I want them, because (it seems) any text in an environment created
> by a \newtheorem statement is in italics. I've tried writing
> \begin{definition}\emph{...}\end{definition} which has the disadvantages
> that (1) I'd have to change every definition
> (there are well over 100) and (2) any maths symbols in "...", don't
> appear in italics any more!

Use the theorem package, which allows you to set the font for your
theorems:

\theorembodyfont{\rm}% Or which font you use
\newtheorem{Foo}{Bar:}

\begin{document}
\begin{Foo}Hello \emph{World}!\end{Foo}
\end{document}

This produces the desired result.

WJ Bland

unread,
Jan 25, 2001, 7:44:04 AM1/25/01
to Pieter Edelman
On Thu, 25 Jan 2001, Pieter Edelman wrote:

[snipped]

> Use the theorem package, which allows you to set the font for your
> theorems:
>
> \theorembodyfont{\rm}% Or which font you use

Brilliant! Thanks very much for your help.
Cheers,
Bill.

dlue...@my-deja.com

unread,
Jan 25, 2001, 11:28:48 AM1/25/01
to
In article
<Pine.GSO.4.21.010125...@granby.ccc.nottingham.ac.uk>,

WJ Bland <pmy...@unix.ccc.nottingham.ac.uk> wrote:
> Hello,
> I'm currently writing up my thesis and I'm having problems
> with getting LaTeX to do what I want. I think it's probably easy to
do -
> when you know how!
>
[...]
> ...the italics in definitions are the opposite way

> around to how I want them, because (it seems) any text in an
environment
> created by a \newtheorem statement is in italics.
> I've tried writing \begin{definition}\emph{...}\end{definition}
> which has the disadvantages that (1) I'd have to change every
definition
> (there are well over 100) and (2) any maths symbols in "...", don't
> appear in italics any more!

Your problem has already been solved in another message. I just wanted
to point out that math symbols are not affected by \emph unless you
forget to put them in math mode. Try

\textit{Let $x$ be positive.}
versus
\textit{\emph{Let $x$ be positive.}}

--
Dan Luecking
University of Arkansas


Sent via Deja.com
http://www.deja.com/

Michele Dondi

unread,
Jan 26, 2001, 4:15:37 AM1/26/01
to
On Thu, 25 Jan 2001 07:59:26 +0000, WJ Bland
<pmy...@unix.ccc.nottingham.ac.uk> wrote:
>Unfortunately the italics in definitions are the opposite way
>around to how I want them, because (it seems) any text in an environment
>created by a \newtheorem statement is in italics.

Check the documentation for amsthm. You can create very fine-tuned
theorem environments. More generally, you'd better use ams packages if
you write a lot of math!


Michele

0 new messages