I hesitate to comment as I haven't read the paper but skimmed comments and abstract...

Wouldn't marrying/reproducing in smaller circles historically have been higher throughout history as transportation and availability would have been less prevalent?

I would imagine that people tended to marry within their social circle at a much higher frequency in say the 1800s than they do now.

Also, it's always curious when I read studies about the increase in Autism rates. I am in the middle of reading Neurotribes and have read elsewhere that much of that increase can be explained in the broadening of the definition and increase in social awareness. Perhaps that doesn't account for the entirety of the increase, but I would be curious as to what percent we're attributing in this paper.