Modern Mongolian herders depend on horses much the way their ancestors did.

Ludovic Orlando

Ancient DNA reveals two lost lineages of horses—but not their elusive origins

When Ludovic Orlando made up his mind to uncover the origins of domestic horses, he didn’t horse around. With 120 other researchers, the molecular archaeologist from France’s CNRS research agency in Toulouse combed stables and dig sites from across Europe and Asia to amass the world’s largest collection of horse DNA—some of it as old as 42,000 years. Now, after several years of intensive analysis, he still doesn’t know when and where modern horses got their start. But he and his colleagues have a much clearer understanding of how humans shaped equine evolution, and they’ve uncovered two previously unknown lineages of horses.

“This is something of an ancient genomics tour de force,” says Daniel Bradley, an evolutionary geneticist at Trinity College Dublin who wasn’t involved in the work. “The scale of sampling makes these data an important and durable legacy.”

To find out where and when humans first began to domesticate horses, Orlando’s team first looked to Kazakhstan, where excavations of ancient Botai settlements had suggested these herders were among the earliest to harness horses. But the DNA evidence suggested these animals were not the modern horse's ancestors, as Botai horses were on a different branch of the horse family tree than modern horses, they reported last year in Science.

The researchers then reached out to field archaeologists, geneticists, and museum curators and obtained extensive DNA data from 278 ancient horses and their relatives from throughout Eurasia. They compared those genomes to the genomes of 30 modern horses and reconstructed 5000 years of equine history. First, they assessed which ancient DNA samples were similar enough to modern horse DNA that they could have been the wild horse ancestor. No ancient samples made that cut.

That doesn’t surprise Greger Larson, an evolutionary geneticist at the University of Oxford in the United Kingdom who was not involved in the study. “Domestication is complex, and the only way to even begin to understand it is by comprehensively assessing lots of samples … from a wide range of contexts and cultures,” he says. Even though the team did its best, he still thinks there must be as-yet-undiscovered cultures that had the first modern horses.

But even though the new work does not show where domesticated horses came from, it does reveal the existence of two new horse lineages: an ancient equine that roamed what is now Portugal and Spain some 4000 years ago, and another that lived in Siberia in Russia around the same time. Since then, both lineages have gone extinct, and there are no traces of them left in modern horse DNA, the team reports today in Cell. Those results could tank an earlier theory suggesting domesticated horses arose in the Iberian Peninsula, Orlando says.

The study also reveals that a lot of the attributes of modern horses appeared much more recently. For example, there are “major genetic turnovers,” Orlando says, after the Arabs expanded into Europe in the seventh century. At that time, Arabian stallions outproduced males from other breeds, leading to their Y chromosome being present in all modern horses today. “It was really cool to see when that loss of male diversity happened,” says Molly McCue, a geneticist at the University of Minnesota in St. Paul who was not involved in the study.

The genome data also reveal that selective breeding greatly intensified about 200 years ago—with positive and negative consequences. Genetic diversity in horses severely declined, allowing more potentially deleterious mutations to accumulate and lead to a higher risk of genetic disease. But that intense breeding also led to faster, stronger horses with greater staying power. The work “really illustrates that horses some 1000 years ago and horses now are two different creatures,” Orlando says.

The new research “is significantly filling in the gaps in our knowledge and fleshing out the background information at a remarkable pace,” says Sandra Olsen, a zooarchaeologist at the University of Kansas in Lawrence. She thinks there may be many more undiscovered lineages of horses just waiting to be found, and that the wild ancestor of modern horses might hail from Ukraine, western Russia, or Hungary. And although no one really knows, Larson is optimistic: “I’m sure they’ll find it,” he says. “It’s got to be out there somewhere.”