That mathematics would be particularly strongly represented should not be much of a surprise. This site is the product of its users, and the user base here is quite mathematical. Among those who committed on Area 51, almost a third of users are active on Mathematics SE. Over 10% are also active on MathOverflow. The only science site to show up in the top 6 sites listed there is Physics, with about 1/6 of all committers. Other science sites don't even show up.
There are, of course, essentially only two things we can do to "improve" the ratio of mathematics to other topics.
- Ask more questions about these other topics.
- Ask fewer questions about mathematics.
Of the two, the first is a good idea. We need people to begin asking questions about sciences. The second is not a good idea unless we're getting a lot of questions about mathematics of relatively poor quality. I don't currently think that is the case. So really, our only good option is to ask more questions about topics other than mathematics.
This puts much of the onus of asking such questions on the users with interests which are currently underrepresented. That's not really avoidable though. But we can (and should) invite others to this beta site to try to get better representation. If you know some biologists, chemists, or other scientists who have an interest in history, you can send them an invitation through the widget on the right of the front page of the main site.
However, we can also take steps to ensure that even if mathematics is over-represented, this does not stifle the development of other communities on this site. The most important step here is making sure we have a fairly robust tag system. Users should easily be able to ignore subjects they don't care about by ignoring the tags. This means every question should have a top-level discipline tag (e.g. mathematics, experimental-physics, etc. as described in Danu's proposal).
It's also potentially a good idea during this private beta period to actively seek out a couple questions for major disciplines, even if they aren't personally of great interest. It would be something of an embarrassment to enter public beta with only two questions tagged biology. I think we should wait a few days and see what happens, but if it looks like that's the direction we're going I think we should try to come up with at least a few good questions for the major scientific disciplines before public beta begins, just so that future users will have somewhere to start.
In all honesty, so long as we reach public beta with at least a few high quality questions and answers from all major disciplines, I think this site will eventually grow to one which has reasonable (if not equitable) representation of all disciplines. Having a lot of mathematics questions won't be a problem for someone whose interests are elsewhere any more than the converse so long as our tag system effectively classifies these.