This Sunday's New York Times Magazine had an excellent article on autism in girls. Boys are four times as likely to be diagnosed autistic as girls, and the ratio for Asperger's syndrome is even higher - perhaps 10:1 (boys:girls). Since girls are diagnosed so much less frequently, there are less data on them - leading to less frequent diagnosis, less research, etc...- an unfortunate catch-22.
The article talks about the fact that boys with autism and Asperger's often manifest their social deficits by isolating themselves from other people, whereas girls seem to wish to connect socially but lack the tools, which often leads to severe anxiety and depression. Depression is co-morbid with autism in girls at a rate much higher than that of boys.
Boys with Asperger's, while exhibiting social deficits generally are seen to excel in subjects like math and engineering, but girls exhibit their strong abilities in more in realms such as writing and reading.
One of the conundrums in researching autism is the many ways symptoms can be manifest are not consistent. It is not the same as studying polio, where the presence or absence of a single biological market, i.e. the virus, determines a diagnosis. Leading researchers agree that one single cause for what we now call autism is very unlikely. That complicates an already complicated puzzle because the target is moving. Are girls not diagnosed as often because not as many of them have autism or because girls manifest the condition with different symptoms? If not as many cases are seen in girls should autism be redefined or is that because fewer girls have it? Since we haven't completely figured out what "it" is, that's not a question we can answer right now.