# Alignment within align environment

My minimum working example:

\documentclass{report}

\usepackage{amsmath}

\begin{document}

\begin{align}
x_1 = \alpha,\, x_2 = q,\, u = \delta \\
f_1(x_1) &= C_1 \left [ \phi_{z1}(x_1) + \phi_{z2}(x_1)M \right ] \\
f_2(x_1) &= C_2 \left [ \phi_{m1}(x_1) + \phi_{m2}(x_1)M \right ] \\
g_1 = C_1 b_z,\, g_2 = C_2 b_m \\
C_1 = \frac{\bar{q}S}{mV_T},\, C_2 = \frac{\bar{q}Sd}{I_{yy}}
\end{align}

\end{document}


The result:

What I want:

As far as I understand, you want to obtain C_1 in the last line moved to the left. (If no, simply remove \!). However, I would suggest replacing ,\,'s by ,\quad's.

\documentclass{report}

\usepackage{amsmath}

\begin{document}

\begin{align}
&x_1 = \alpha,\, x_2 = q,\, u = \delta \\
&f_1(x_1) = C_1 \left [ \phi_{z1}(x_1) + \phi_{z2}(x_1)M \right ] \\
&f_2(x_1) = C_2 \left [ \phi_{m1}(x_1) + \phi_{m2}(x_1)M \right ] \\
&g_1= C_1 b_z,\, g_2 = C_2 b_m \\
&\!C_1 = \frac{\bar{q}S}{mV_T},\, C_2 = \frac{\bar{q}Sd}{I_{yy}}
\end{align}

\end{document}


Przemyslaw's answer gives the output you want, but in the case that your set of equations becomes more complex (e.g. you want to align two sets of two equations with different alignment points), you could use aligned inside gather for just the two you want aligned:

\documentclass{report}

\usepackage{amsmath}

\begin{document}

\begin{gather}
x_1 = \alpha,\, x_2 = q,\, u = \delta \\
\begin{aligned}
f_1(x_1) &= C_1 \left [ \phi_{z1}(x_1) + \phi_{z2}(x_1)M \right ] \\
f_2(x_1) &= C_2 \left [ \phi_{m1}(x_1) + \phi_{m2}(x_1)M \right ] \\
\end{aligned}\\
g_1 = C_1 b_z,\, g_2 = C_2 b_m \\
C_1 = \frac{\bar{q}S}{mV_T},\, C_2 = \frac{\bar{q}Sd}{I_{yy}}
\end{gather}

\end{document}

• unfortunately, the two aligned lines end up with only one equation number, and they also start farther left than the other three lines. (thinking about this situation.) – barbara beeton May 8 '14 at 13:23
• @barbarabeeton I don't get them starting further left, but you're right I hadn't noticed the missing equation number. I'm always trying to remove equation numbers in my own code! – FionaSmith May 8 '14 at 13:33
• @barbarabeeton why does it do that? I have never used aligned/alignedat before, I just read about them on the wikibooks page where it says "similar to align" and doesn't say anything about equation numbering being different. – FionaSmith May 8 '14 at 13:42
• treating the "subsidiary" environments as a unit was a design decision made with amstex, and not changed when the math processing was converted to the amsmath package for latex. no reason is given, but i surmise that the "style" of writing math for publication was less "free form" in 1980 than it is now (certainly, editorial controls were much stricter). aligned and alignedat were expected to be used with "broken" single formulas, not with "lists" of equivalent formulas. but i can't find written "justification" in the archives. – barbara beeton May 8 '14 at 14:06
• addendum: i haven't read the wikibooks pages on math closely, but i've taken a quick glance at parts of them. i don't think they're up to the standards i expect for such documentation. someday, in my "voluminous free time" (= nonexistent), i intend to read and edit the material; "misguidance" actually causes the ams editorial staff quite a few problems, so maybe i can get clearance to do this as an "official project". (after all pending tasks are cleaned up, of course.) – barbara beeton May 8 '14 at 14:10