I just received the latest Science News (6/28/97 issue) which contains a
very interesting article on a new genetic clock study dating the early
lineage of dogs. The information may prove of interest here in light of Ray
Audette's observation on the listgroup some weeks ago that small bands of
hunters armed only with atlatls and aided by dogs could be fearsome hunters
capable of taking down the largest of mammals. I'll just summarize the
basic points here.

Titled, "Stalking the Ancient Dog: Man's Best Friend May Go Way Back," the
results of the new mitochondrial DNA studies (widely used in genetic
dating, though still controversial) indicate that the origin of the dog
goes back much further than previous estimates of 10,000-20,000 years
ago--perhaps to well beyond 100,000 years ago. The studies weren't able to
determine the original geographic origin of dogs but they did eliminate
from consideration any parent species other than the wolf (which has long
been suspected). They also suggest that dogs may have evolved from wolves,
not once, but several different times, as well as the possibility that dogs
may have interbred with wolves from time to time.

The study, published in the 6/13/97 issue of Science (no refs given) by
Robert K. Wayne et al of UCLA, looked at DNA from 162 wolves taken from 27
populations in Europe, Asia, and North America, and compared it with DNA
from 140 modern dogs from 67 different breeds around the world. There was
such a wide range of  variability between modern dogs and wolves that
researchers close to the case say it is impossible for the genetic
variability observed to have evolved in the traditional assumed timeframe
of just 10 to 20 thousand years. (This interpretation assumes of course
that the molecular-clock dating technique, which is still controversial,
holds up.)

The more ancient timeframe for the origin of the dog can be explained, say
the researchers, because even though the fossil record for dogs is obscure
beyond about 14,000 years ago (after which dog bones are often found in
association with human bones), it has been overlooked until now that wolf
bones can be found in association with human bones going back beyond the
100,000-year horizon. Thus what were formely considered "wolves" may in
fact have been "dogs-in-process." If wolves and humans were indeed
associating with each other this early in symbiotic fashion, it suggests
that we may be looking at a different type of mutually beneficial
relationship than the traditional assumption of "domestication." In other
words, 100,000 years ago or before, wolves might have been "tamed" in some
way that did not lead to changes that would show up in the fossil record.
Then with the impending advent of agriculture, late mesolithic or early
neolithic peoples might have begun actually breeding dogs selectively to
become different kinds of hunters than before, or into herders, guards, etc.

Says Elaine Ostrander, a molecular biologist collaborating on a study of
the dog genome who works at the Fred Hutchsinson Cancer Research Center in
Seattle, "When we became an agricultural society, what we needed dogs for
changed enormously, and a further and irrevocable division occurred at that
point. That may be the point--at which dogs and wolves were noticeably
different physically--that stands out in the fossil record."

--Ward Nicholson <[log in to unmask]>