# Inline equation number

I would like to be able to give inline equations using $...$ a label such that I can refer to it later in my thesis.

My goal in pseudocode:

Some text $mc^2(eqn number)\label{some eqn}$ some more text

And than be able to use

some text \ref{some eqn} some more text

Obviously the above does not work but I hope it gets along what I am hoping to achieve.

I will also sketch the reason for this inquiry because I am definitely open for other solutions than my idea above.

I frequently come across small formula that are not really worth a full:

$$...$$

or

\begin{align} ... \end{align}

and are better mentioned in line. The problem is that it also happens that I need to refer to these small formula later in the work.

• It is not a good idea to put eq numbers in the text as your reader will probably have a hard time keeping track of them. Remember you can refer to page number to any label. – daleif Apr 2 '17 at 16:59
• welcome to tex.sx. although the equations you mention are small, think of your reader -- if the reference isn't quite close by, it will be very hard to find what is being referred to if the number is buried in the text. – barbara beeton Apr 2 '17 at 16:59
• Are you OK with loading the hyperref package, or does something in your document's setup indicate that hyperref must not be used? Please advise. – Mico Apr 2 '17 at 17:00
• I do indeed realize that my idea is not the most preferable option. As I said other solutions are very welcome. I just don't want to clutter my text with small equations left right and center. @Mico, the hyperref package may be used. (it is already as a matter of fact) – gertian Apr 2 '17 at 17:18
• How about adding the reference into a \marginpar? This way the reader can spot them easily (at least the numbering). – Skillmon Apr 2 '17 at 17:35

In case my \marginpar idea is ok for you, this would allow it:

\documentclass{scrartcl}

\usepackage{lipsum}
\usepackage{marginnote} % better placement than marginpar

%CODEBLOCK
\newcounter{inlinelabel}[section]
\let\theinlinelabelbak\theinlinelabel
\renewcommand{\theinlinelabel}{(\thesection.\theinlinelabelbak)}
\newcommand{\inlinelabel}[1]{%
\marginnote{%
\refstepcounter{inlinelabel}\label{#1}%
\theinlinelabel%
}%
}
%CODEBLOCK

\begin{document}
\section{first section}
$mc^2$\inlinelabel{ineq:mc2}
\lipsum[1]
also see equation \ref{ineq:mc2}.
\end{document}

To avoid confusion, perhaps you shouldn't use a custom counter but the equation counter. For this change the above contents of %CODEBLOCK to:

\newcommand{\inlinelabel}[1]{%
\marginnote{%
\refstepcounter{equation}\label{#1}%
\theequation%
}%
}

EDIT: For the alignment of the other equationlabels, try (it works for align of the amsmath-package, but I didn't test any others):

\makeatletter
\def\tagform@#1{\maketag@@@{\marginnote{#1}}}
\makeatother

On the other issue: You might also put the label next to the equation, like in Werner's answer. Other than that I have no idea at the moment.

EDIT2: I just noticed, that \marginnote places two notes invoked on the same line above each other. Because of this I changed the command \inlinelabel:

\newcommand{\inlinelabel}[2][0]{%
\refstepcounter{equation}\label{#2}%
\marginnote{%
\ifnum#1>0%
\newcount\inlinecount%
\inlinecount=0%
\loop\ifnum\inlinecount<#1%
\setbox0\hbox{\theequation}%
\hskip\wd0,% This % was missing resulting in more space than 1ex
\hskip1ex%
\stepcounter{equation}%
\repeat%
\fi%
\theequation%
}%
}

Now you can call the \inlinelabel with an optional argument which is the number of prior equations in that line. The result looks like this:

And this is the content of \begin{document}...\end{document} in the above image:

$mc^2$\inlinelabel{ineq:mc2}$E=mgh$\inlinelabel[1]{ineq:emgh}$A=\pi r^2$\inlinelabel[2]{ineq:apir}
\lipsum[1]
also see equation \ref{ineq:mc2}.
\begin{align}
E = hf
\label{eqn:ehf}
\end{align}

EDIT3: I've changed the function again (now the commas in the marginnotes are only printed once not multiple times). The new complete MWE shows some limitations of this code.

\documentclass{scrartcl}

\usepackage{lipsum}
\usepackage{marginnote} % better placement than marginpar
\usepackage{amsmath}

\makeatletter
\setbox0\hbox{,}%
\hskip\wd0\relax%
}
\newcommand{\@labelskip}{%
\setbox0\hbox{\theequation}%
\hskip\wd0%
\stepcounter{equation}%
}

\newcommand{\inlinelabel}[2][0]{%
\refstepcounter{equation}\label{#2}%
\marginnote{%
\ifnum#1>0%
\newcount\@inlinecount%
\@inlinecount=0%
\@labelskip%
\loop\ifnum\@inlinecount<\numexpr#1-1%
\@labelskip%
\repeat%
\fi%
\theequation%
}%
}

\def\tagform@#1{\maketag@@@{\marginnote{#1}}}
\makeatother

\begin{document}
\section{first section}
$E=mc^2$\inlinelabel{ineq:mc2}
nice
$E=mgh$\inlinelabel[1]{ineq:emgh}
equations
$A=\pi r^2$\inlinelabel[2]{ineq:apir}
being
$U=2\pi r$\inlinelabel[3]{i}
nice
$E=hf$\inlinelabel[4]{ii}. $H=ot$\inlinelabel[5]{iii}.
\lipsum[1]
also see equation \ref{ineq:mc2}.
\begin{align}
E = hf
\label{eqn:ehf}
\end{align}
Note that \verb|\eqref| doesn't work with this setup: \eqref{eqn:ehf}. This is
because of the changes made to \verb|\tagform@|
\end{document}
• Hey thank you for this great suggestion, I still have some questions though: Is there a way to modify the usual labels of equations such that they are aligned with this new type of equation ? And if I may ask some more advice, what would you suggest in case of two equations on the same line. How do I clarify which one I refer to ? – gertian Apr 2 '17 at 18:03
• @gertian see my edit for this. – Skillmon Apr 2 '17 at 18:12
• @gertian if you're still interested in this, I made an edit improving the code (not the behavior, sadly) and showing some limitations. – Skillmon Apr 5 '17 at 15:11
• Thank you very much ! Sorry for my late respond I was on a field trip and failed to find WiFi ! – gertian Apr 5 '17 at 17:46

\documentclass{article}

\usepackage{amsmath}
\newcommand{\inlineeqnum}{\refstepcounter{equation}~~\mbox{(\theequation)}}

\begin{document}

Some text $E = mc^2 \inlineeqnum\label{eqn:einstein}$ some more text.

Some text \eqref{eqn:einstein} some more text.

\end{document}

Consider the possible confusion when viewing the equation number as a possible element of the equation.

• Thanks for your solution. It is exactly what I asked for. But after seeing it in person it turns out that I like Skillmons suggestion more. – gertian Apr 5 '17 at 17:47