Someone doing work on bioinformatics with a [Google Code project](http://code.google.com/p/iphylo/) has a nice collection of notes about the way JSTOR interacts with the DOI infrastructure:
> JSTOR uses a mixture of identifiers, namely stable URLs, SICIs, and DOIs.
> As far as I can establish, all JSTOR DOI’s start with the prefix 10.2307. CrossRef resolves a JSTOR DOI to a URL with a SICI. For example, this is what happens to the DOI `doi:10.2307/3760372`:
curl -I http://dx.doi.org/10.2307/3760372
HTTP/1.1 302 Moved Temporarily
Date: Thu, 22 May 2008 10:43:56 GMT
> In the example above, the SICI is 0027-5514(199209/10)84:5<615:AROC(S>2.0.CO;2-O
> SICI’s have a rather baroque syntax. See the Wikipedia article for an introduction.
> On April 4, 2008 JSTOR released a new platform that uses stable URLs to identify articles (see note on Facebook). These URLs arhave the prefix `http://www.jstor.org/pss/`. The rest of the URL is the identifier part of the DOI. Hence, the article `doi:10.2307/3760372` has the stable URL `http://www.jstor.org/pss/3760372` (note the shared `3760372`).