A similar question has been posted here about the align
environment. However, the solution does not seem to work for align*
.
Preceding paragraph.
{\small
\begin{align*}
a &< b
\end{align*}
}%
Following paragraph.
A similar question has been posted here about the align
environment. However, the solution does not seem to work for align*
.
Preceding paragraph.
{\small
\begin{align*}
a &< b
\end{align*}
}%
Following paragraph.
As we have no minimal example, an example when it works. (If you have some special combination of packages, please show the minimal example, when it doesn't work for you.)
\documentclass{article}
\usepackage{amsmath}
\begin{document}
Preceding paragraph.
{\small
\begin{align}
a &< b
\end{align}
}
Following paragraph.
Preceding paragraph.
{\small
\begin{align*}
a &< b
\end{align*}
}
Following paragraph.
\end{document}
(When we replace \small
by, say, \tiny
, the result is awful, but the same sizes are easy visible.)
Sorry, but this never works. Consider this example, where \scriptsize
is used to make the effect more apparent:
\documentclass{article}
\usepackage{amsmath}
\usepackage{lipsum} % for context
\begin{document}
\lipsum[1][1-4]
\lipsum[1][1-4]
{\scriptsize\begin{align*}
a&<b \\
c&<d
\end{align*}}
\lipsum[2][1-4]
\end{document}
With \small
instead of \scriptsize
the bad typesetting is still present, albeit less evident:
Here's a way to make it work without bad effects. Again, \scriptsize
is used to better illustrate that the result is as wanted.
\documentclass{article}
\usepackage{amsmath}
\usepackage{lipsum} % for context
\newenvironment{sizeddisplay}[1]
{\par\nopagebreak#1\noindent\ignorespaces}
{\nopagebreak\ignorespacesafterend}
\begin{document}
\lipsum[1][1-4]
\lipsum[1][1-4]
\begin{sizeddisplay}{\scriptsize}
\begin{align*}
a&<b \\
c&<d
\end{align*}
\end{sizeddisplay}
\lipsum[2][1-4]
\end{document}
{\small ... }
, but the pernicious spacing badness it causes in the preceding paragraph is a sad side-effect since it might go unnoticed. Thanks for this.
a &< b
appear smaller.align
andalign*
work the same way, so there is no issue.