Yesterday’s New York Times had an article by Carl Zimmer about increasing numbers of retracted papers in the biological sciences. Physics and Mathematics weren’t part of the story and I don’t know of any evidence of retractions increasing in these fields (although maybe they should, given the Bogdanov and other scandals).
There’s a blog called Retraction Watch where they follow these things, and they have come upon a mathematics example. A couple years ago the Elsevier publication Computers and Mathematics with Applications published the article “A computer application in mathematics”. It’s less than a page long, one author has a yahoo.com e-mail address, the other a budweiser.com address. Last week Elsevier finally got around to acknowledging that something was up, publishing a retraction notice that explained:
This article has been retracted at the request of the Publisher, as the article contains no scientific content and was accepted because of an administrative error. Apologies are offered to readers of the journal that this was not detected during the submission process.
I gather that this is behind a paywall, so you need to be at a place like Columbia that pays Elsevier a lot of money, otherwise you can’t read the retraction. That’s also true of the original paper, but if you want to violate all sorts of intellectual property laws, you could click here.