When using the "amsplain" style in BibTeX, it automatically prints MathReview MR numbers in the citation. For example if the citation entry looks like:

@book {Babook,
 AUTHOR = {Baldwin, John T.},
 TITLE = {Fundamentals of stability theory},
 SERIES = {Perspectives in Mathematical Logic},
 PUBLISHER = {Springer-Verlag},
 ADDRESS = {Berlin},
 YEAR = {1988},
 PAGES = {xiv+447},
 ISBN = {3-540-15298-9},
 MRCLASS = {03-02 (03C45)},
 MRNUMBER = {918762 (89k:03002)},
 MRREVIEWER = {John B. Goode},

Then MRNUMBER is printed in the citation (but not MRCLASS or MRREVIEWER).

I would like to stop this from happening, i.e. prevent the MRNUMBER from being printed in the citation. This website suggests adding


to the beginning of the *.bib file. But this did not work for me. The other suggestion is to edit the amsplain.bst file.

Can someone tell me how to edit the amsplain.bst file so that the MRNUMBER is not printed? Alternatively, is there some other way to accomplish this?

Note: I would like to avoid individually deleting the MRNUMBER from each entry in my reference file.


the command added with @preamble is good, except that if the document class you are using already defines \MR, it won't be overwritten by what's specified by a \providecommand.

you can override all "predefinitions" by adding this to the preamble of your document, as late in the preamble as you can manage:


it could be done more subtly, of course, by first checking whether a definition for \MR already exists, and using \newcommand or \renewcommand as appropriate, but i haven't got time to test that right now. (i've also forgotten whether one # is enough, or two are needed; none of the "usual" manuals bother to mention that.)

  • Wow, that's a lot more elegant than the answer I was about to write down, which would involve editing (a copy of) the file amsplain.bst! – Mico Sep 16 '14 at 20:30
  • 1
    @Mico -- i have edited amsplain.bst, which is why i prefer a "more direct" approach if one exists. (if you want to insert the code to \if the situation, or confirm whether one or two # are needed, please feel free.) – barbara beeton Sep 16 '14 at 20:36
  • Your code is fine -- nothing for me to edit. :-) – Mico Sep 16 '14 at 21:01
  • 1
    Since the purpose of the instruction is to make sure the \MR macro does nothing, I think there's little to be gained from checking first if the macro has been defined -- and clobbering it only if it's been defined. If \MR for some reason has not been defined, I see no harm in going ahead anyway and defining it to do nothing. – Mico Sep 16 '14 at 21:19
  • 1
    ah, well. purists say that \def is a bludgeon (or a primitive, which it is). they prefer scalpels. – barbara beeton Sep 16 '14 at 21:21

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.