I'm afraid you are attacking straw men. This specific case, this specific correlation, is not really ready to go into the popular press as headlines for the ge

I'm afraid you are attacking straw men. This specific case, this specific correlation, is not really ready to go into the popular press as headlines for the general public to take home. It is easy to do controlled experiments regarding gravity which are not completely loused up with confounds. Nobody needed to go data-fishing in order to find gravity - it's a very clear and strong phenomenon. Are you sure that there are no possibly relevant systematic differences OTHER than age, between men who have children early and ones who have children late? A 'proposed' mechanism cannot substitute for understanding of what is actually happening, in the context of popular publication of data suggesting that autism is caused by having babies too late... this is a month away from the women's magazines and two months from parents being blamed for their children's autism. It would be more responsible to actually demonstrate a causal link before running to the presses.