>> The nature vs. nurture debate reminds me of something I've observed in the
>> offspring of well trained dogs...
I would be careful claiming this unless you have some good records and more than anecdotal evidence to back it up. I have not seen any evidence that leaning can affect the DNA of a creature in such a way that they somehow "pass" the learning down to their offspring, biologically. There are anecdotes about this, but I've never seen a study that shows it, and believe me, people have tried.
There was once a study done on dogs. There was an ancient belief that docking the tail of a female dog would result in any litters she had after the operation would have deformed tails. Much to the astonishment of many, even in the scientific community, it was experimentally shown that cutting off the mother's tail did not, in any way, affect the tail growth of her pups.
Now, it can be the case that a trait that a trait that is learned by a species effects the environment in such a way that having that trait specified innately would make members of the species more fit. In this case, natural selection would lead toward that trait becoming innately specified so that the members of the species wouldn't have to learn it.
This is known as the "Baldwin Effect", here is a link to a site about it:
Now, the important thing to keep in mind here, is that the Baldwin Effect can, indeed, speed up the evolution of some particular innately specified trait, BUT, it's not going to happen in the course of a single generation, resulting in well-trained dogs have innately smart puppies.
Now, having said that, your theory about stress causing chemical changes that effect DNA might be possible, but I would be severely skeptical of it unless someone provided some evidence. Also, considering that females are born with all the eggs they'll ever have through life, I think such a "stress-induced chemical/biological evolution" would probably have to take place by changing male gametes...though changes in female mitochondria could have some effect, as well...an interesting theory that I would definitely like to see further explored experimentally.
Cheers,
Sask