# Labeling Multiple Equations Per Row of a Table

I am trying to create a table with two columns of independently labelled equations (1-6). Unfortunately, everything I have found so far either has not worked or would number them as sub-equations (1a, 1b, etc).

My current attempt is:

\documentclass{article}
\usepackage{graphicx, verbatim, tabularx}
\usepackage{amsfonts}
\usepackage{amsmath, amssymb}
\newcommand{\prtl}[2]{\frac{\partial #1}{\partial #2}}
\begin{document}
\begin{tabular}{p{5cm}p{5cm}}
$\curl\vec{E} = -i\omega\mu\vec{H}$ &
$\curl\vec{H} = i\omega\mu\vec{E}$ \\
$$\label{eq:curlHx} \prtl{E_z}{y} +\gamma E_y= -i\omega\mu H_x$$
&
$$\label{eq:curlEx} \prtl{H_z}{y} +\gamma H_y= i\omega\mu E_x$$\\
$$\label{eq:curlHy} -\prtl{E_z}{x} -\gamma E_x= -i\omega\mu H_y$$
&
$$\label{eq:curlEy} -\prtl{H_z}{x} -\gamma H_x= -i\omega\mu E_y$$
\\
$$\label{eq:curlHz} \prtl{E_y}{x}-\prtl{E_x}{y} = -i\omega\mu H_z$$
&
$$\label{eq:curlEz} \prtl{H_y}{x}-\prtl{H_x}{y} = -i\omega\mu E_z$$
\end{tabular}

\end{document}


Is there an environment other than tabular that would make this work? Tabularx came close (got the numbering correct), but the formatting/spacing was horrific.

• equations are vertical display environments so need to be in vertical (par) mode not horizontal (LR) mode. So you columntypes need to be p{3cm}p{3cm} not cc – David Carlisle Mar 21 '14 at 21:06
• Welcome to TeX.SX! Please make your code compilable (if possible), or at least complete it with \documentclass{...}, the required \usepackage's, \begin{document}, and \end{document}. That may seem tedious to you, but think of the extra work it represents for TeX.SX users willing to give you a hand. Help them help you: remove that one hurdle between you and a solution to your problem. – jub0bs Mar 21 '14 at 21:11
• Thank you, @Jujobs. This version is not compilable, but I do believe it has all the other pieces. Is there anything else I should do to be more helpful to those helping me? – LucidLunatic Mar 21 '14 at 21:34
• @user48407 providing the definitions (or packages containing them) for \curl and \prtl would be helpful. I had to made some faked definitions in my answer. – Gonzalo Medina Mar 21 '14 at 21:36
• @GonzaloMedina Added! – LucidLunatic Mar 21 '14 at 21:38

Your code won't work since c columns expect material in horizontal mode and displayed math needs vertical mode.

Below I present three options; the first two using a tabular approach and the third one using minipages and align, thus allowing alignment points at the equal signs. In my opinion, the third option is the one that produces a more pleasing result.

One option would be to use p{...} columns (I used the maximum available width for both columns and provided some faked definitions for \curl and \prtl):

\documentclass{article}
\usepackage{amsmath}

\DeclareMathOperator{\curl}{Curl}
\newcommand\prtl[2]{\dfrac{\partial#1}{\partial#2}}

\begin{document}

\noindent\begin{tabular}{@{}*{2}{p{\dimexpr.5\textwidth-\tabcolsep\relax}}@{}}
$\curl\vec{E} = -i\omega\mu\vec{H}$ &
$\curl\vec{H} = i\omega\mu\vec{E}$ \\
$$\label{eq:curlHx} \prtl{E_z}{y} +\gamma E_y= -i\omega\mu H_x$$
&
$$\label{eq:curlEx} \prtl{H_z}{y} +\gamma H_y= i\omega\mu E_x$$ \\
$$\label{eq:curlHy} -\prtl{E_z}{x} -\gamma E_x= -i\omega\mu H_y$$
&
$$\label{eq:curlEy} -\prtl{H_z}{x} -\gamma H_x= -i\omega\mu E_y$$
\\
$$\label{eq:curlHz} \prtl{E_y}{x}-\prtl{E_x}{y} = -i\omega\mu H_z$$
&
$$\label{eq:curlEz} \prtl{H_y}{x}-\prtl{H_x}{y} = -i\omega\mu E_z$$
\end{tabular}

\end{document}


If the unnumbered equations in the first row should be centered, one could load array in the preaamble:

\usepackage{array}


and then use

\begin{tabular}{
@{}
*{2}{>{\centering\arraybackslash}%
p{\dimexpr.5\textwidth-\tabcolsep\relax}}
@{}
}
...
\end{tabular}


for the table format specification. Extra vertical spacing can be avoided using the optional argument for \\:

\documentclass{article}
\usepackage{amsmath}
\usepackage{amssymb}
\usepackage{array}

\newcommand\prtl[2]{\frac{\partial#1}{\partial#2}}

\begin{document}

\noindent\begin{tabular}{
@{}*{2}%
{>{\centering\arraybackslash}p{\dimexpr.5\textwidth-\tabcolsep\relax}}
@{}
}
$\curl\vec{E} = -i\omega\mu\vec{H}$ &
$\curl\vec{H} = i\omega\mu\vec{E}$
\\[-3ex]
$$\label{eq:curlHx} \prtl{E_z}{y} +\gamma E_y= -i\omega\mu H_x$$
&
$$\label{eq:curlEx} \prtl{H_z}{y} +\gamma H_y= i\omega\mu E_x$$
\\[-6ex]
$$\label{eq:curlHy} -\prtl{E_z}{x} -\gamma E_x= -i\omega\mu H_y$$
&
$$\label{eq:curlEy} -\prtl{H_z}{x} -\gamma H_x= -i\omega\mu E_y$$
\\[-6ex]
$$\label{eq:curlHz} \prtl{E_y}{x}-\prtl{E_x}{y} = -i\omega\mu H_z$$
&
$$\label{eq:curlEz} \prtl{H_y}{x}-\prtl{H_x}{y} = -i\omega\mu E_z$$
\end{tabular}

\end{document}


A different approach using two side-by side minipages and align to produce alignment for each column; in this case, \tag was used to retain the numbering schema of the previous solutions:

\documentclass{article}
\usepackage{amsmath}
\usepackage{amssymb}
\usepackage{array}

\newcommand\prtl[2]{\frac{\partial#1}{\partial#2}}

\begin{document}

\noindent
\begin{minipage}[t]{.5\textwidth}
\begin{align}
\curl\vec{E} &= -i\omega\mu\vec{H} \notag \\
\label{eq:curlHx}
\prtl{E_z}{y} +\gamma E_y &= -i\omega\mu H_x \\
\label{eq:curlHy}
-\prtl{E_z}{x} -\gamma E_x &= -i\omega\mu H_y\tag{3} \\
\label{eq:curlHz}
\prtl{E_y}{x}-\prtl{E_x}{y} &= -i\omega\mu H_z\tag{5}
\end{align}
\end{minipage}%
\begin{minipage}[t]{.5\textwidth}
\begin{align}
\curl\vec{H} &= i\omega\mu\vec{E} \notag \\
\label{eq:curlEx}
\prtl{H_z}{y} +\gamma H_y &= i\omega\mu E_x\tag{2} \\
\label{eq:curlEy}
-\prtl{H_z}{x} -\gamma H_x &= -i\omega\mu E_y\tag{4} \\
\label{eq:curlEz}
\prtl{H_y}{x}-\prtl{H_x}{y} &= -i\omega\mu E_z\tag{6}
\end{align}
\end{minipage}

\end{document}


• I found \curl in package mandi. Also I would add >{\centering} and >{\centering\arraybackslash} to the column specifications to center the first unnumbered equations. – Heiko Oberdiek Mar 21 '14 at 21:29
• @HeikoOberdiek Thanks. I decided to keep my "faqed" \curl definition since I am not sure if the OP is using a package or a customized definition that might then conflict with the one in the package. I added a remark about the column specifications. – Gonzalo Medina Mar 21 '14 at 21:38
• Thank you! @GonzaloMedina, is there a way I could alter the spacing between the rows? On my machine there is nearly an inch of spacing between the equations in the table. – LucidLunatic Mar 21 '14 at 21:42
• @user48407 Please see the second code in my updated answer. – Gonzalo Medina Mar 21 '14 at 21:57
• @user48407 I added another option to my answer that might be of interest for you. – Gonzalo Medina Mar 21 '14 at 22:07