Nicholas Wade / New York Times / September 7, 2009
The dog has so many fine qualities it is hard to know which it was prized and bred for by the early people who first domesticated its noble ancestor, the wolf. Was it the dog's valor in the hunt, perhaps, or its role as night watchman, or its strength in pulling a sled, or its companionable warmth on cold nights?
A new study of dogs worldwide, the largest of its kind, suggests a different answer, one that any dog owner is bound to find repulsive: wolves may have first been domesticated for their meat. That is the proposal of a team of geneticists led by Peter Savolainen of the Royal Institute of Technology in Stockholm.
Sampling the mitochondrial DNA of dogs worldwide, the team found that in every region of the world all dogs seem to belong to one lineage. That indicates a single domestication event. If wolves had been domesticated in many places, there would be more than one lineage, each leading back to a local population of wolves.
The single domestication event seems to have occurred in southern China, where the dogs have greater genetic diversity than those elsewhere. The region of highest diversity is usually the place of origin because a species tends to lose diversity as it spreads.
Dr. Savolainen sampled a part of the dog genome, the mitochondrial DNA, and was able to estimate the time of the domestication - probably around the period that hunter-gatherers first settled down in fixed communities in China, about 11,000 to 14,000 years ago. Those people would have had an organized culture that enabled them to make muzzles, and possibly cages, that would have been needed to handle wolves.
There is a long tradition of eating dogs in southern China, where dog bones with cut marks on them have been found at archaeological sites.
Dr. Savolainen said wolves probably domesticated themselves when they began scavenging around the garbage dumps at the first human settlements, a theory advocated by Ray Coppinger, a dog biologist at Hampshire College in Massachusetts. As the wolves became tamer, they would have been captured and bred. Given local traditions, Dr. Savolainen suggests, the wolves may have been bred for the table.
Thus, dogs may have thus insinuated themselves into human life by means of garbage and dog meat, but they quickly assumed less demeaning roles. Once domesticated, they rapidly spread west from the eastern end of the Eurasian continent.
Most people do not eat dogs, so they must have spread so quickly for other reasons, perhaps because of their use as guard dogs or in pulling sleds, Dr. Savolainen said.
His report was written with Jun-Feng Pang of the Kunming Institute of Zoology in China, who analyzed the DNA of the many Chinese dogs in the study. It was published last week in the journal Molecular Biology and Evolution.
In 2002, Dr. Savolainen wrote that dogs had been domesticated from wolves in East Asia, a conclusion that was challenged last month by a team at Cornell University. The Cornell team said genetic diversity was as high in African village dogs as in those in China.
Dr. Savolainen disputed the Cornell calculation in his new report, contending that diversity was, in fact, higher in Chinese dogs.
Adam Boyko, a member of the Cornell team, said that Dr. Savolainen's team had now built a plausible hypothesis from detailed genetic data but that other explanations might still be possible, including that dogs had been domesticated at a second site, outside China, and had spread everywhere but China.
Stephen O'Brien, an expert on the genetics of domestication at the National Cancer Institute, said Dr. Savolainen's argument for a single domestication event in southern China was "a pretty good conclusion" but one that could be strengthened by a more thorough sampling of wolves throughout the world.
A team of American researchers is examining the genetics of dogs and wolves with a so-called dog chip, a device that is programmed to recognize thousands of different sites on the dog and wolf genome, not just the mitochondrial DNA studied by Dr. Savolainen. The data have not yet been published, but some of it "doesn't agree completely" with an East Asian origin of dogs, Dr. O'Brien said.
The disputes about the origins of dogs arise because researchers are just cutting their teeth on what Dr. O'Brien called "genomic archaeology."
"It's a brand new field," he said. "We're just learning how to do it."
Domestication of the dog and other animals is both of intrinsic interest and of relevance to the human past. "Domestication was really the lever by which civilization was able to organize into communities larger than those of foraging families," Dr. O'Brien said.
Dogs were evidently so useful to early people that they spread like wildfire. On the basis of current evidence, they were the first species to be domesticated