New Study Places First Dogs In Europe

Each new study seems to contradict the others. What's a poor dog to do?

Posted Nov 17, 2013

A new front in the battle over bragging rights as homeland to the first canines opened on November 15, 2013 with publication in the journal Science of a paper by an international group of geneticists naming Europe as the dog’s site of origin. Dogs appeared between 18,000 and 32,000 years ago, a period that encompasses the Last Glacial Maximum, when glaciers had advanced to their farthest extent and begun their retreat, wrote the paper’s principal author, Olaf Thalmann of the University ofTuku, Finland.  The study is the first to compare DNA from modern wolves and dogs to that from ancient canids.

Among the study’s most intriguing results is the suggestion that the conversion of wolves into what I call dogwolves, or doglike wolves, , probably occurred multiple times in multiple places without producing the direct ancestors of modern dogs.  That finding fits with my notion that wherever they met on the trail of the big game they were hunting, wolves and humans wound up traveling together.

The team’s paper directly refutes the two most popular current theories of dogs evolution:  that they emerged as much as 50,000 years ago in the Middle East or that they diverged from gray wolves some 16,000 years ago south of the Yang-tze river in China.   These theories have remained dominant in discussions of dog origins despite the fact that the oldest fossils of dogs and doglike canids have come from Europe and Russian Siberia. 

The new study places the origins of the dog in the camps of hunters and gatherers preying on large grazing animals that roamed Europe in the Late Pleistocene rather than in the garbage dumps of the semi-permanent villages of proto-agriculturalists. This finding alone should force a long overdue revision of the oft-cited theory that dogs emerged from a population of obsequious, self-domesticating dump-diving wolves who renounced hunting to savor the waste  from human villages.  Unfortunately, Thalmann fails to present a lucid alternative model, although at least one exists, as I discussed in my posting of October 2, 2013.

The geneticists reached their conclusions after comparing the full mitochondrial genomes of eighteen ancient canids—dogs, doglike canids, and wolves—with those of seventy-seven modern dogs and forty-nine contemporary wolves. Modern dogs sorted into four clades based on their genetic similarities and had more in common genetically with the long dead European canids and with modern European wolves and dogs than with those from any other region.  Curiously, the oldest specimens from Belgium and the Altai Mountains of Russia appear to represent separate lineages of dogwolves or proto-dogs that died out without making any apparent contribution to early dogs.  These canids clearly need additional study.

The possibility of European origins of dogs have been largely dismissed in recent years, primarily because geneticists studying breed formation routinely reported that nearly all European breeds were created during the past 200 years through the mixing of different types of dog and because humans moved into Europe only after expanding through the Middle East into Asia.

But I have long placed parts of Europe among the locations where dogs emerged in large measure because of the continent’s long, deep, and active involvement with wolves and dogs.

 Robert K. Wayne, evolutionary biologist UCLA and senior author on this paper, agreed in a recent email exchange about the article that Europe’s deep involvement with dogs seemed to make it a logical candidate for early dogs, but only in recent years have researchers been able to find supporting evidence in old bones and their DNA. 

The difficult task of extracting and analyzing DNA from ancient fossils was conducted in Wayne’s UCLA Lab and in the evolutionary anthropology lab at the Max Planck Institute in Leipzig, Germany.  Their success in obtaining full mitochondrial sequences from  eighteen  ancient specimens is itself a noteworthy achievement.   Geneticists have long known that to obtain a more complete understanding of canine evolution they would have to compare ancient and modern genomes. The more complete the genomes they compare, the more detailed and accurate the view they obtain.

But obtaining that DNA from ancient samples is difficult because it is often corrupted—degraded with age or contaminated with DNA from other animals.  Then, too, those institutions and individuals in possession of ancient fossils are often reluctant to let them be used for such studies given that the extraction involves destruction of a small amount of already scarce material.

Intriguing though this paper’s findings are, they are hardly the last word on the subject, as even the geneticists have recognized.  Asia and the Middle East are not represented in the survey.   The researchers say that is because there are no ancient wolf or bone remains from southeastern China to test.  That is true, but my feeling has always been that researchers will find more fruitful material from northwestern China into Mongolia and the Altai Mountains. 

The oldest dog remains from the Middle East date to around 13,000 years ago; these are unlikely to be remnants of the oldest dogs ever, but are old enough to test for comparison, I would think.  Moreover, humans migrating out of Africa lingered in the Middle East where they must have encountered wolves.  What they did with them has implications for the dog’s origins, as we shall see. 

Wayne, in fact, says that the team failed to obtain a clean genome from one Israeli sample and others are just now being studied. 

Useful though mitochondrial DNA is for showing evolutionary relationships it provides a snapshot only of maternal descent.  Nuclear DNA from both parents yields a much more complete view, but it is also much more difficult to extract from fossils. 

Another problem arises from the dates the researchers have derived from their analysis. By most accounts Homo sapiens entered Europe around 40,000 to 50,000 years ago from the south and the east.  Neanderthal was still alive at the time.  If the humans lacked canine companions when entering the landmass what transpired to make them seek out the company of wolves some 8,000 years later? Did they experience a sudden change of consciousness unlike any that has occurred since but would amount to a species wide conversion experience?  Did they inherit or steal from Neanderthals animals already predisposed to seek human company?  Or did they arrive with dogwolves already among them?

Another challenge to the date comes from the researchers’ own confirmation of previous findings that dogs came to the New World with humans crossing the Bering Land Bridge perhaps 18,000 or more years ago. Unless one believes that traits arose independently in that population that were identical to those in their Old World cousins—smallness, black coat, and the like—then the people making that migration must have had those types of dogs living among them.  Those people were probably in the vicinity of the Altai Mountains weathering the Last Glacial Maximum. The ancient Altai Mountain canid, dated to 36,000 years ago, appears unrelated to modern dogs, but its presence in the area suggests that the people were actively engaged with canids.

There are ways to get people and dogs with the necessary variability in the Altai Mountain region in time for a major migration but they require either that the evolution of dog and those varieties arose there, for which there is no evidence to date, or that the people arrived from somewhere in the Middle East or region of the Caucasus or Caspian Sea, their dogs mixing with other wolves or dogwolves along the way.  That scenario is not one the current study supports.

Complicating the issue is a rather strong amount of data supporting the emergence in the Middle East around 50,000 years ago of a genetic mutation associated with small size in dogs and another associated with black coats.  Dogs are thought to have passed the mutation for black coat color to New World wolves and coyotes.

When I suggested to Robert Wayne that all of these threads pointed to an earlier date for the dog’s emergence than the 18,000 to 32,000 years ago proposed in the paper, he agreed.

In her posting on this study, my fellow Psychology Today blogger Pat Shipman says that it raises the question of when a wolf is not a wolf and suggests the answer is "When it's a dog," meaning it has cast its lot with humans rather than its fellow wolves--no matter what its genes or appearance seem to say.  

It is becoming increasingly apparent --at least to these tired eyes that the dog was born on the move with the people with whom its forebears had cast their lot.  Wolves are remarkable beings much akin in family structure and temperament to a big-brained biped.  They are sociable and fiercely protective of their young. They are cooperative hunters capable of bringing down game considerably larger than themselves.  Becoming dogs involved expanding their capacity for sociability with another species, and the same can be said of the humans with whom they joined forces.