I always put numbers for multi-line equations as follows (e.g., using equation and gathered): enter image description here

My co-author wants to do (e.g., using gather and nonumber).

enter image description here

Is there a commonly accepted convention for such instances?

  • 1
    This is really off-topic. How to do it would be on-topic; whether to do it not. Could you ask on another site specific to your discipline?
    – cfr
    Commented May 30 at 4:04
  • 1
    @cfr I wouldn't call it off-topic. The question is about typing LaTeX documents. Commented May 30 at 4:20
  • 1
    @Michael_1812 By that logic, keyboards are on-topic, as are questions about the conventions used in recipe books, nursery timetables and guides for ramblers. Typographic conventions etc. are off-topic. How to realise them is on-topic. This isn't my personal view. I'm just reporting the scope of the site.
    – cfr
    Commented May 30 at 4:27
  • @Michael_1812 See e.g. tex.meta.stackexchange.com/questions/6842/…, tex.meta.stackexchange.com/questions/679/…, tex.meta.stackexchange.com/questions/6647/…, or, generally, tex.meta.stackexchange.com/search?q=typography. This one would be more plausibly on-topic imho if it included the code used to implement each option rather than applying as much to Word as TeX.
    – cfr
    Commented May 30 at 4:36
  • It would have been helpful if you had provided some information about the relative importance of the two equations in question. Are they roughly equally important, and are you likely to be referring to both equations in a cross-reference to this group of equations? If so, use your positioning approach. Conversely, is the first equation more important (in whatever sense of the term "important" that may be applicable here) than the second, and can a cross-reference to the group of equations only meaningfully be interpreted as referring to the first equation? If so, use your colleague's approach.
    – Mico
    Commented May 30 at 5:32

2 Answers 2


There is no right or wrong for this case. What you can do, in order to please yourself as well as your coauthor, is to define a specific environment. You can eventually fight with your coauthor and the winner will decide what option to use.

Of course, the formula doesn't make sense, but I'll ignore the fact.


\usepackage{kantlipsum}% for mock text



H=\sum_{J\in S}H_{J} \\
J=\{j_{1},\dots,j_{k}\}\in S,\quad j_{i}\in\Lambda


enter image description here

If you just remove [t] from the definition, you get

enter image description here

Much better, if the main formula is not too long, would be using a single line.

enter image description here


I, personally, always place equation numbers the way you do, and I cannot recall a case when a technical editor of journal changed it.

As a reader, I also would prefer this manner of numbering, and my guess is that most readers will share my opinion.

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .