In the last decade, an increasing number of online journals have begun publishing issues where the articles are all paginated independently. PLoS One is probably the best known, but several paleontological journals and journals by MDPI also have moved in that direction.

An example: https://www.mdpi.com/2075-4450/12/8/723/htm. Note that this is ''not'' a case of a journal where each article constitutes an issue. This journal's articles are formally arranged into issues.

Page numbers are completely useless for citing these articles (though I have seen it done). Usually there is an article number, but a format such as "12(8): 723" not only looks malformed, but is potentially misleading. I have not found any guideline in citation guides (I work mostly with Vancouver/CSE style) that cover this issue. Usually I plop an "a" in front of the article number ("12(8): a723") if there is not already a mean of telling it is such a number, but I'd rather have some sort of actual style guide advice...

Has any recent guide covered this corner case?

  • I don't know about any style guide covering it, but what I'd do is notice that, within the (PDF version of the) article, there are sequential page numbers from 1 to 16, and set the BibTeX field pages="723-1--723-16" (making use of LaTeX's distinction between - for a hyphen and -- for an en dash). I think this may have originally been inspired by the way APS journals with article numbers formatted the page numbers in footers c. 2000. Sep 19, 2021 at 15:37
  • 1
    Maybe it's because I'm used to physics papers often not including the page range (just opening page number, or article number for some journals) in citations, but that format doesn't look malformed to me.
    – Anyon
    Sep 19, 2021 at 16:16

1 Answer 1


This has become extremely common in journals that are read mostly or entirely online. The Physical Review journals (even the ones that still feature little-read paper versions) made the transition to using article numbers instead of pagination over twenty years ago. To deal with this, all you really need to do is to adopt a reference format that just lists the article number instead of the initial page number for an article; e.g. (for article number 062901 of volume 103):

G. H. B. Martins, W. A. M. Morgado, S. M. Duarte Queirós, and A. P. F. Atman, "Large-deviation quantification of boundary conditions on the Brazil nut effect," Phys. Rev. E 103, 062901 (2021).

For journals that still have sequential pagination, you can just list the first page number rather that the whole range, if you want a consistent look for your references. What value does the inclusive page listing serve, after all?

  • 5
    What value does the inclusive page listing serve, after all? --- For me it indicates how long the paper is, which can often be helpful to have a sense of when scanning over the bibliography for related papers. Incidentally, my style for such articles is simply to be explicit, such as: "... Article #32, May 2018, 17 pages". Sep 19, 2021 at 18:03
  • Once upon a time the inclusive page listing would have helped long-suffering ILL librarians figure out which pages to photocopy from a physical journal and put in the mail. But it does seem like something of a vestige now. Sep 19, 2021 at 19:10
  • An important context is that I deal in taxonomy, so my referencing must be able to handle material from the late 18th century all the way to last week. In addition to what Dave said, full page ranges are also necessary for old articles that were published in parts, and thus comprise multiple page ranges.
    – Circeus
    Sep 19, 2021 at 19:50
  • @Circeus: multiple page ranges --- My personal method for handling this, for example (the following is from [6] in this answer):"... Bulletin des Sciences Mathématiques et Astronomiques (1) (1873), 20-48 & 79-96". Incidentally, (n) means the n'th series (unless something like (N.S.) is more appropriate), and for journals with more than one series I use (1) for the first series, even when the actual issues/volumes give no such indication, to provide additional redundancy for the reader's benefit (i.e. helps avoid oversight/typo concerns). Sep 20, 2021 at 13:17
  • @Circeus: FYI, I just noticed in [6] in the answer I cited (and copied an excerpt into my previous comment) is incomplete. It should have been "(1) 5 (1873), 20-48 & 79-96", where '5' is the volume number. And while I'm here (again), I suppose I'll mention that sometimes the multiple page ranges extend over more than one volume, in which case I do something like "(1) 5 & 6 (1873 & 1874), 20-48, 79-96 & 14-23". If I know the issue numbers, then those get incorporated also, but at this point (if not before), things possibly start becoming a bit too personally idiocentric. Sep 20, 2021 at 13:38

You must log in to answer this question.

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