by an article in the Sep/OCt 2009 Psychology Today magazine. "A Chip off the Best Block" by Satoshi Kanazawa, p.60-61
It discusses the Trivers-Willard hypothesis on sex selection, and has interesting numbers. For instance "Engineers and scientists have 140 boys for every 100 girls; nurses and schoolteachers have 135 girls for every 100 boys." The premise there being that brain types area heritable, so there's selection going on there. Wealth and promiscuity also select for more male babies.
Another interesting comment was that it isn't just the male sperm that determines the sex of a baby; that the mother's hormones can make things inhospitable for the 'wrong' gender, and prevent proper implantation.
Perhaps I'm behind in my reading, but that's the first I've seen mentioned of any of this; so I'm intrigued. Alas, not quite intrigued enough to investigate the science behind it, but it's an interesting conversation piece.
The man who wrote the article is on of the magazine's bloggers -
http://www.psychologytoday.com/blog/the-scientific-fundamentalist