Skip to main content



A canine companion so nice it (maybe) evolved twice

NEW YORK — For years, one of the most confounding questions in science — alongside what is dark matter and why do we sleep — has been one that many pet owners may have found themselves casually pondering: Where did dogs come from?
Scientists generally agree that humanity’s best friend descended from gray wolves, scampering into our lives at least 15,000 years ago.
Scientists generally agree that humanity’s best friend descended from gray wolves, scampering into our lives at least 15,000 years ago.

NEW YORK — For years, one of the most confounding questions in science — alongside what is dark matter and why do we sleep — has been one that many pet owners may have found themselves casually pondering: Where did dogs come from?

Scientists generally agree that humanity’s best friend descended from gray wolves, scampering into our lives at least 15,000 years ago. Virtually everything else is a matter of debate.

“When and where did this happen and with whom — with what human group?” said Dr Pontus Skoglund, a paleogeneticist at the Francis Crick Institute in London. “It’s really a mystery.”

Studies have turned up widely divergent answers, variously concluding that dogs were first domesticated somewhere in Asia or Europe or the Middle East or perhaps in multiple locations.

Now, a new analysis of 72 ancient wolf genomes spanning the last 100,000 years suggested one possible explanation for some of the seemingly contradictory results: Two different ancient wolf populations, one in Asia and another in the Middle East or surrounding area, contributed DNA to modern dogs.

Precisely how that happened remains unclear. It is possible that dogs were domesticated twice, in two different locations, and that the populations subsequently mixed. An alternate scenario is that dogs emerged just once, somewhere in Asia, and later bred with a more western wolf population, picking up additional wolf DNA.

“We can’t tell the scenarios apart,” said Mr Anders Bergstrom, an expert in evolutionary genomics at the Francis Crick Institute and an author of the study, which was published Wednesday in Nature. “But we can say that there were at least two source populations of wolves.”

The research also offers insight into the evolutionary history of gray wolves, providing hints about how the species managed to escape the fate of other, long extinct ice age mammals.

“Ancient DNA is the only way to untangle this history,” said Professor Elisabetta Cilli, a paleogeneticist at the University of Bologna in Italy, who was not involved in the new research. “This study raises the bar of our knowledge of dog domestication and wolf population dynamics.”


In the new study, an international team of researchers set out to learn more about the origins of dogs by painting a detailed picture of the gray wolf’s past.

The goal was to “build a big map of wolf ancestry, wolf genetics, around the time of domestication,” said Dr Skoglund, the study’s senior author. “And then we take the dog puzzle piece and put it in the big wolf jigsaw puzzle and see where it fits.”

Archaeologists from 38 institutions in 16 countries donated bones and teeth from ancient wolf specimens that had been previously unearthed in Siberia, Europe and North America. Then, scientists at nine ancient DNA labs set to work. They drilled tiny holes in the specimens, collected samples of bone powder and extracted and sequenced the DNA. Then, the scientists compared the wolf genomes they assembled to genomes from both ancient and modern dogs.

The researchers found that as a whole, dogs were more closely related to ancient wolves in Asia than those in Europe.

“This reinforces the idea that Central Asia was likely to be the center of origins for dogs,” said Dr Adam Boyko, a canine geneticist at the Cornell University College of Veterinary Medicine, who was not involved in the research. “Of course we want more data to pin that down a bit better. That’s a big area.”

But there were differences between different dog specimens. Early dogs in Siberia, East Asia, the Americas and northeastern Europe appeared to derive nearly all of their genetic material from an ancient wolf population in Asia.

But early dogs from the Middle East and Africa also had significant ancestry from a second, unidentified wolf population. This genetic material was most closely related to DNA from modern Middle Eastern wolves, the scientists found.

Modern village dogs in Africa and the Middle East — as well as breeds that originated in those regions, such as the Basenji — still have considerable amounts of ancestry from this second wolf population, the researchers found. But this genetic legacy persists in nearly all modern dogs; globally, most dog breeds today can trace 5 per cent to 30 per cent of their ancestry to this second wolf population, Mr Bergstrom said.

“It looks like there is a smoking gun of a second population,” said Professor Greger Larson, a paleogenomicist at the University of Oxford and an author of the new study.

Prof Larson had previously proposed that dogs might have been domesticated twice, before changing his mind when subsequent evidence suggested a single origin. Now, he said he was “on the fence” about whether dogs truly emerged twice or simply bred with wolves after migrating. “We’re back to that square one again,” he said.

Dr Boyko said the simplest explanation, and the scenario he favored, was that dogs were domesticated just once, in Central Asia, and later mated with another population of wolves. “It just seems more parsimonious to believe — because we already know that dogs and wolves exchange genes and have done it at other times — that that’s what went on, and not a secondary domestication event.”

None of the ancient wolf specimens the scientists analyzed were a close enough match to have been part of the actual population that gave rise to our canine companions, suggesting that dogs’ ancestors may have come from locations not covered by the study.

Pinpointing the specific ancestral populations and determining how the second lineage got into dogs will require more ancient wolf and early dog specimens, especially from more southern regions that were not included in this study, scientists say.

“We only have the northern half of the map,” Dr Skoglund said.


The researchers also found that ancient wolves that lived in far-flung locations were highly genetically similar — more closely related than wolf populations today — indicating that there was extensive movement and interbreeding between wolves across the globe.

This fluid, global population might explain how the species survived the end of the last ice age, the scientists posit.

“Cave lions, hyenas and bears and other mammals went extinct, but wolves survived,” Mr Bergstrom said. “And perhaps it was this great connectivity and mobility of wolves that allowed the population to thrive.”

The researchers also traced natural selection in wolf populations over time, identifying multiple genetic variants that rapidly spread through ancient wolves, likely because they conferred a significant survival advantage.

For instance, new mutations in a gene called IFT88, which appeared to be involved in the development of the skull and face, emerged about 40,000 years ago, the researchers found. Over the next 10,000 years, these mutations spread to all wolves — and are still present in both wolves and dogs today.

“It is a very dramatic episode of natural selection,” Mr Bergstrom said.

What drove it is unknown, but the mutations might have helped ancient wolves adapt to changing diets after some of their preferred prey species went extinct, the researchers speculated.

The scientists also found that certain genetic variants affecting wolves’ olfactory receptor genes became much more common in wolf populations over time, suggesting that the animals’ sense of smell also benefited from natural selection.

Although many questions remain, the new data will be a valuable jumping-off point for scientists who are still trying to solve the many mysteries of wolf evolution and dog domestication, scientists said.

“Having this many ancient wolf genomes is a huge advance in the field,” Boyko said. “I’m sure other researchers are going to love to get their hands on it and explore some of their own pet theories.”

So to speak.

This article originally appeared in The New York Times.

Related topics

dogs canine science research

Read more of the latest in




Stay in the know. Anytime. Anywhere.

Subscribe to get daily news updates, insights and must reads delivered straight to your inbox.

By clicking subscribe, I agree for my personal data to be used to send me TODAY newsletters, promotional offers and for research and analysis.