I'm submitting a paper to the journal Elsevier journal "Nonlinear Analysis: Theory, Methods & Applications" (http://www.sciencedirect.com/science/journal/0362546X).

They told me to use the elsearticle document to prepare my submission. The problem is the margins are huge and the line spacing is huge, so my document has doubled in size under this class. So I thought: ok, I need to cut down on my calculations etc. But then I looked at some published articles under that journal, and they do not resemble the elsearticle class at all! The margins are much smaller and the line spacing is much nicer, so much more text fits in.

What is the point of them telling us to use that awful style file? Why should I spend hours reducing my paper when it won't be published in that style and so the changes I make may be entirely unnecessary!

  • 1
    Not sure what you mean by "reducing your paper". – DCTLib Jul 21 '15 at 11:48
  • Is there a page limit? – Federico Poloni Jul 21 '15 at 12:19
  • 3
    Submission and published articles are VERY different in style. Submission style is made to improve correction by reviewer, based on the time they were doing it on paper. Some journals will refuse to review article not prepare according to submission style (at least in my field). Your choice, but deal with the consequences. – Emilie Jul 21 '15 at 13:13
  • @FedericoPoloni I did not get a page limit. – C_Al Jul 21 '15 at 14:18
  • @DCTLib I just meant removing some calculations and going from Step 1 to Step 4 instead of through Steps 2 and 3, etc. – C_Al Jul 21 '15 at 14:21

It is very rare (in mathematics) that a journal really makes using their style necessary for submission, even for the journal making that claim in their guide to authors. My advice is to never adjust the style of a paper for submission, and simply submit the version you originally produced. The reason is that formatting for each submission turns out to be a waste of time, all the more so that journal styles are rarely that much more readable for the referee than any standard style. (Note that unless you really know about typesetting you should not depart from standard style too much, e.g. do not use the fullpage package. Large margins have a purpose.)

In fact, most of the time this is not even mandatory to use the journal's style after acceptance, because the journal will do the job of adapting the article to their class. I try to do it for reasonably cheap journals (e.g. MSP journals), but I never do that for Springer and Elsevier.

Also, unless strictly mandatory (e.g. page limit) you should never change the content of an article in function of the formatting style (of course you can resolve overfulls and other warnings): the content is driven by what you have to say, period.

  • Regarding your last sentence, this is true, but one can always reduce detail to a degree. – C_Al Jul 21 '15 at 14:20
  • 1
    one can always reduce detail to a degree — Of course. But you should reduce detail if and only if the reduction actually improves the paper. – JeffE Jul 21 '15 at 18:56

The elsevier LaTeX class file has a couple of options that influence the appearance of the paper.

If you see very large margins and double spacing, then you are using the class with a setting that is supposed to make reviewing the paper easier. It allows the reviewers to add notes to the paper "in situ", i.e., directly where they belong. This is especially helpful when correcting grammar and spelling.

The style of a LaTeX document can be changed by simply replacing a few words in the first few lines of the document's source code. For example, an option to the elsevier document class will remove the double-spacing. So it is not unreasonable by the journal to ask you to use that style, even if you think that it is ugly. When changing the look of the paper, it may be necessary in a few cases to add a few hyphenation patterns or break formulas a bit differently when changing the appearance, though. Also note that the publisher has an own version of the class file that is not publicly available for final versions of the paper. If you use their provided style, then it will be easy for the publisher to switch your paper to this alternative style file, so that the final result looks just as the other papers in the journal.

  • 1
    Thanks for the info. my main issue was that I didn't know how many pages I would get in the final version. But fortunately there is a preprint option which looks quite like the published versions. – C_Al Jul 21 '15 at 14:20
  • I didn't know how many pages I would get in the final version — One page for every page of content. Math journals don't tend to have page limits. (And if they do impose page limits, those limits should be clearly specified in the Instructions for Authors.) – JeffE Jul 21 '15 at 18:56

They told me to use the elsearticle document to prepare my submission.

Who is "they"? If your paper has not been accepted yet, I recommend contacting someone on the editorial board (i.e. a scientist, not someone who works for Elsevier) and asking them if this is in fact required.

It might be that Elsevier merely "encourages" you to incorporate their style file so that their own employees don't have to do it themselves. In this case, you might remember that Elsevier makes over $1B of profit every year by selling scientists' own research back to them. My reaction would be that they can damn well afford to hire professional copy editors; others' reactions may vary.


I've refereed papers for Elsevier journals, and the format of the paper that they provided to me was quite different from any published version. Not only did it have lots of white space (in the margins and between lines) but it also had line numbers in the margin. (The latter were useful, so that I could easily refer to specific lines in my report.) So you should not assume that any particular pre-publication format will resemble the final published version. And you should should certainly not base any revisions of the paper on its appearance in an early format.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.