# Tombstone misplaced with multline* and \qedhere

It looks like the end-of-proof symbol is misaligned if multline* and \qedhere are used. Or am I just using them wrong?

\documentclass{article}
\usepackage{amsthm,amsmath}
\begin{document}
\begin{proof}
Only words here.
\end{proof}
\begin{proof}
\begin{equation*}
A=form(u)la. \qedhere
\end{equation*}
\end{proof}
\begin{proof}
\begin{multline*}
A=form(u)la.\\ with/many*lines \qedhere
\end{multline*}
\end{proof}
\end{document}


Is there a way to fix this?

-
this is definitely a known issue, but you have turned up a previously unknown problem as well -- with equation* a spurious warning is issued. that has been added to our list. – barbara beeton Mar 12 '13 at 12:53

This seems to be a known issue; the log file for your example contains the following.

Package amsthm Warning: The \qedhere command may not work correctly here on input line 15.

I'm not aware of a fix, but there are others here who know much more about ams packages than I do. A workaround is

\tag*{$\qed$}


in place of \qedhere.

-
I checked, and I get that warning once for line 9 (where the \qedhere inside the equation* is) and twice for line 15 (where the one inside multline* is). Strange. Anyway, thanks for pointing this out; I use kile, and it did not display the warning for some reason. – Federico Poloni Mar 12 '13 at 10:31
this workaround will do the right thing only when equation numbers are positioned on the right. also, it's necessary to suppress the non-"here" box with \renewcommand{\qedsymbol}{}; since a proof is a local group, this definition will not persist after \end{proof}. – barbara beeton Mar 12 '13 at 13:02
@barbarabeeton Another workaround is to "guess" the correct shift and do \let\myqedsymbol\qedsymbol \def\qedsymbol{\myqedsymbol\hspace*{-10pt}} \qedhere inside minipage*. It's a dirty solution, but it works; one just have to make sure that the output is correct. – yo' Jan 26 '15 at 21:05