>  Of course, selection pressure doesn't *create* adaptation; it
 >  only favors it.  For adaptation to happen there must be the right
 >  sorts of variations, occurring with sufficient frequency.

I can't recall any details on this, but in a book I read there was a report
of an experiment that showed striking variation among individuals in the way
they metabolized a certain chemical. The chemical was administered orally,
and the concentrations of various metabolites appearing in the subjects'
urine were plotted. The relative concentrations of metabolites in the urine
specimens were dramatically different. Some people converted the test
chemical largely to one metabolite, others produced a lot of different
metabolites, and the patterns shown in the book looked as individual as
fingerprints.

Also, consider that different people react very differently to medications
such as antidepressants. It's common for depressed people to try many
antidepressants until they find one that works for them and does not cause
intolerable side effects. I don't think there is yet any way to predict
which drug will work for a given patient.

Thus, it would appear that there is already a lot of biochemical diversity
in the population. It would then not be surprising to find that a major
shift in diet could result in fairly rapid adaptation.

Interesting..., very interesting...