> This may be an area where the "blood type diet" theory has some merit.
> It is a fact that people with type A blood make less gastic HCl than
> those with type O blood.  It also appears to be true that the frequency
> of type A blood in human populations increased about 40,000 years ago,

Interesting, do you have a reference where I can learn more on this? 

> although there's no reason to suppose that it *began* then.  

Indeed, we know that it didn't begin then because type A blood is found in
other primates, such as chimps, and it predates the human race (more than
2.5 million years ago).

So that
> raises an interesting question:  What might have happened 40,000 years
> ago to make the world more hospitable to those with type A blood,
> resulting in an increase in the relative frequency of that phenotype?
> 
> Note that both type As and type Os make *enough* HCl to activate the
> proteolytic enzymes.  To my knowledge, there is no evidence for the
> claim that type As don't make enough HCl to digest much meat.  But type
> Os seemingly make more than enough, suggesting that the additional HCl
> may have provided another advantage, at least until about 40,000 years
> ago.  So one theory is that gastic HCl also serves to kill food-borne
> microorganisms (dogs and other carnivores also produce high levels of
> HCl), and the cooking of meat became common, this defense became less
> important, resulting in a proliferation of the type A phenotype that,
> prior to cooking, was held back.

That would appear to presume a very late start to cooking prevalence of
around 40k years ago, whereas the recent evidence has been pushing cooking
back further and further--more than 200k years ago.