Shards of pottery with traces of mare's milk, mass gravesites for horses, and drawings of horses with plows and chariots: These are some of the signs left by ancient people hinting at the importance of horses to their lives. But putting a place and date on the domestication of horses has been a challenge for archaeologists. Now, a team of geneticists studying modern breeds of the animal has assembled an evolutionary picture of its storied past. Horses, the scientists conclude, were first domesticated 6000 years ago in the western part of the Eurasian Steppe, modern-day Ukraine and West Kazakhstan. And as the animals were domesticated, they were regularly interbred with wild horses, the researchers say.
"This is a very good paper," says biologist Michael Hofreiter of the University of York in the United Kingdom. "Nobody has applied this method of population modeling to horses before."
Throughout their history, horses have been interbred, traded between populations of people, and moved across continents. All of this makes their genetic history hard to follow. Moreover, the wild ancestor of horses, Equus ferus, is extinct, complicating researchers' efforts to compare the genetics of domestic animals with wild ones. Previous research nailed down a broad area—the Eurasian Steppe, which stretches from Hungary and Romania through Mongolia—as the region where horses originated and were domesticated. But earlier genetic studies relied mostly on mitochondrial DNA, which is inherited only from a mother, to try to understand horses' evolutionary history.
"The problem was that there was a lot of diversity in the mitochondrial DNA," says biologist Vera Warmuth of the University of Cambridge in the United Kingdom, the first author of the new study. And the diversity didn't group the horses into their breed or place of origin. "Every horse breed has almost all the mitochondrial lineages represented," she says.
Warmuth instead studied sequences of horse DNA inherited from both parents and known to be diverse between horse populations. She and her colleagues collected genetic samples from more than 300 horses at 12 different sites across the steppe. Data were collected for only working animals bred within a local area, not those bred for show or appearance, to minimize any human-guided selection that would make some genes more common. Then, the researchers used computer programs designed to model the spread of a population to simulate how different locations of horse domestication and spread throughout the steppe would influence modern genetic diversity. They compared each model with the real data they had collected to see which fit best.
The best-fit model, the team reports today in the Proceedings of the National Academy of Sciences, showed the wild ancestor of domestic horses originating in eastern Eurasia 160,000 years ago and being domesticated in the western part of the Eurasian Steppe around 6000 years ago. The model also helped explain why there had been so many female lineages when previous studies had tried to rely on mitochondrial DNA. "We think that as domestic horses spread out of the western steppe, local wild females were continuously incorporated into the spreading herds," says Warmuth. The constant addition of new females made the genetic patterns—in particular, the female lineages—more complex than if the domestic population had been totally isolated.
Hofreiter is impressed. "They have still only narrowed down the domestication region to a fairly big area," he says, "but they did have enough genetic data to get a signal out of the noise."
Not all researchers are convinced, however. Archaeologist Marsha Levine of the University of Cambridge thinks using modern genetic samples to retrace horses' evolution is a dead end. "There's been mixing of cultures and mixing of horses in this region for many thousands of years," she says. "And so when you're looking at any modern horse, you just don't know where it's from."
Bringing together many kinds of evidence is what will ultimately answer the whens and wheres of horse domestication, Levine says. "What we need to be doing is using material from excavations, sequencing ancient genes, and combining that with what we know from archaeological evidence about how animals were used in the past."
Ultimately, says Hofreiter, getting to the bottom of horse domestication will reveal more than just the history of these animals. "Horse domestication has changed human cultures a lot. It has changed warfare, it has changed transportation," he says. "Studying the past of horses can tell us a lot about our own past."