A Chihuahua wears a U.S. flagMario Anzuoni / Reuters

Between 14,000 and 18,000 years ago, the ancestors of today’s Native Americans first entered the land where they now live. They came from Asia, walking east across a broad land bridge that connected the two continents, snaking south past a stretch of retreating glaciers, and eventually spreading across a new land. A few millennia later, dogs followed them.

The origin of those indigenous American dogs is unclear—as is their fate. Some say they were wiped out after European colonizers arrived in the 15th century, bringing their own dogs with them. Others believe their genes still exist in modern-day Chihuahuas and Xolos.

A team of researchers, led by Laurent Frantz, Greger Larson, and Elizabeth Murchison, has now settled the debate. By analyzing DNA from 71 archeological dog remains and comparing them to the genomes of modern breeds, the team showed that the indigenous dogs all but died out. There are tiny traces of their DNA in modern dogs, but that genetic legacy is so faint that it might not be real.

And that’s a shame, because those canines represented a unique lineage of dogs “that were completely different from anything that exists anywhere else, and that doesn’t exist anymore,” says Larson, who works at the University of Oxford.

“This study has answered some of the big questions about the arrival of dogs in the Americas,” says Sarah Brown from the Washington Department of Fish and Wildlife. “It’s an excellent contribution that has been much needed.”

Dogs are domesticated descendants of gray wolves, which first arose in Eurasia on perhaps more than one occasion. There have long been wolves in the Americas too, but genetic studies show that those wild animals were not the ancestors of the continent’s indigenous dogs. Instead, the dogs arose from already-tame ancestors that lived in Asia around 14,600 years ago. Those ancestors were most closely related to prehistoric sled-pulling dogs from Siberia’s Zhokhov Island, whose residents were among the first people to deliberately breed dogs for a purpose.

Humans likely introduced dogs into the Americas—but not immediately. The oldest dog bones in the continent are 4,500 years younger than the oldest human ones, and genetic evidence supports a similar time lag. But once dogs entered the continent, they spread widely, and “evolved along a trajectory completely of their own,” says Larson. “If you look at their bone structure, they are very clearly dogs. They’re not from outer space with a third ear or a fifth leg. But they were also slightly different.”

But when Europeans arrived with dogs of their own, the native dogs started disappearing. Perhaps they were directly persecuted. Perhaps they were killed by introduced diseases, much like Native Americans themselves were. Today, they are almost entirely gone.

When the team analyzed the DNA of 5,000 modern American dogs, they found only tiny traces of indigenous DNA in modern Arctic breeds, like Alaskan huskies, Alaskan malamutes, and Greenland dogs. And they found no traces at all in most other groups, including street dogs, which are the least likely to be affected by introduced European breeds, and in supposedly “native” breeds like Catahoulas and Xolos. “Some people have said that these represent ancient lineages that haven’t been diluted by the introduction of European dogs,” says Larson, “but that’s all bullshit.”

Amazingly, the organism that is most closely related to the vanished American dogs isn’t even a dog. It is, however, of a dog.

Canine transmissible venereal tumor, or CTVT, is a type of contagious cancer. It arose in the genitals of a single dog some 8,000 years ago, and somehow evolved into an independent parasite that can spread sexually from one individual to another. Through millions of doggy-style liaisons, CTVT has traveled all over the world—an immortal pandemic tumor that afflicts dogs today, long after its original owner’s death. And it still retains that progenitor’s genes.

By analyzing those genes, Elizabeth Murchison, from the University of Cambridge, showed that CTVT is the closest living relative of the indigenous American dogs. It didn’t arise from one of them, but from a very close cousin. “It felt too amazing to be true,” she says. “There are many issues when you’re working with both ancient DNA and CTVT. But every way we cleaned up the data, they pointed to the same conclusion.”

Murchison’s work suggests that CTVT arose from a dog that lived in Asia, and was likely related to those that eventually spread into the Americas.

Picture a lineage of Asian dogs, who found two ways of spreading their genetic material around the world. One group of dogs went into the Americas on foot. Another took a more unorthodox route of passing on their DNA, spawning a tumor that hitchhiked into the Americas on the genitals of other dogs. The former is now gone; the latter still thrives.

There is, however, one confusing twist to this already incredible narrative: CTVT seems to have small traces of coyote DNA. Coyotes are only found in the Americas, so how could the tumor have arisen in an Asian dog, as Murchison’s other evidence suggests? Did coyotes once roam back and forth between the two continents, leaving traces of their genes in Asian dogs? Did the tumor actually arise in the Americas first, before spreading into Asia, and then making its way back?

Whatever the reason, it’s likely complicated—like the rest of the story of dogs in America. After the first dogs arrived, they moved around the continent, dispersing, reconnecting, mixing, and interbreeding. Meanwhile, more waves of dogs were being introduced. The Thule people brought more Asian dogs to the American Arctic a thousand years ago, Europeans brought their dogs from Europe, and gold miners brought Siberian huskies over during the 1920s gold rush. “Up in Alaska, a lot of shit is going down,” says Larson. “There’s a lot of back-and-forth that mixes everything up into a blender.”

In recent years, studies involving ancient DNA have greatly complicated the old picture of the peopling of the Americas. “I expect that the more ancient genomes we get from North America, the more complex it’s going to become,” Jennifer Raff told me in May. Now, it seems that the history of American dogs will eventually prove to be just as convoluted as the history of their human companions.

We want to hear what you think. Submit a letter to the editor or write to letters@theatlantic.com.