Val, you make a number of good points about the distribution of IQ, especially those having to do with genetic disorders and cultural expectations. It's still a fairly small difference, though, and mostly measurable only at the tails. And the cultural aspect appears to be lessening, at least in industrialized nations.
Even in the historical data you mention from 1932 Scotland, the gender difference is quite small at +2 SD, which is where we generally draw the GT line (though some use +1.7 SD). And it certainly isn't a ratio of 1:2. (More like 5:7.)
More recent meta-analyses have confirmed the use of combined norms for males and females on cognitive measures.
There have been some interesting speculative discussions about lower variance among females than among males, possibly having to do with two X chromosomes vs one X chromosome (though the discussions I've seen don't fully incorporate X inactivation).
