The last wild horse may not be wild — 9 animals saved it

The last wild horse on earth, which may not be wild, and is descended from nine animals. · published 26 September 2026

The animal sold as the last truly wild horse may be the escaped livestock of a culture that vanished five thousand years ago - and specialists are still fighting about it. What did survive from the same work is that horse domestication did not start where everyone said it did. Meanwhile every Przewalski’s horse on earth descends from nine animals that made it through the 1940s.

The angle

Przewalski’s horse is sold as the only never-domesticated horse, but a 2018 Science study of 42 ancient genomes concluded it is the feral descendant of Botai livestock from ~5,500 years ago — a claim specialists including William Taylor still contest, while the same work’s uncontested finding is that modern domestic horses carry only about 2.7 per cent Botai ancestry, meaning domestication did not begin where the textbooks said.

The notes this episode was written from

These are the working notes, unedited. Everything the video states as fact comes from here, and nothing else may be stated as fact. The sections on limits and on what the finding does not show are part of the brief, not an afterthought — see the method.

What it is called. Przewalski’s horse is the animal presented everywhere as the only truly wild horse left - never domesticated, the last survivor of the horses that existed before people got hold of them. Mustangs and brumbies are feral, domestic animals gone loose. This one was supposed to be the real thing.

The study that upset it. Gaunitz, Fages, Outram, Orlando and colleagues, “Ancient genomes revisit the ancestry of domestic and Przewalski’s horses”, Science, 2018. They sequenced 42 ancient horse genomes, including horses from the Botai culture of the Central Asian steppes - the site that gives the earliest archaeological evidence of horse husbandry, roughly 5,500 years ago, with bridling, milking and corralling.

Two conclusions came out of it, and they are not equally secure.

The one that held. Modern domestic horses are not descended from the Botai horses. They carry only about 2.7 per cent Botai-related ancestry, which means the textbook account - that horse domestication began at Botai and everything followed from there - is wrong. A later study settled where they did come from: Librado and colleagues, Nature, 2021, using 273 ancient genomes, placed the origin of modern domestic horses in the lower Volga-Don region, spreading from about 2200-2000 BC alongside the chariot and replacing local horse populations across Eurasia within a few centuries. That part is not in dispute.

The one still being argued. The 2018 paper also concluded that Przewalski’s horses are the feral descendants of the Botai horses - not wild at all, but escaped livestock from five thousand years ago. That claim is contested. William Taylor and Christina Barron-Ortiz challenged it in 2021, and Taylor has continued to argue, including in Scientific American in 2024, that Przewalski’s horse was never domesticated. The alternative reading is that the Botai horses and Przewalski’s horses both descend separately from the same ancient wild population, which would make Przewalski’s horse wild after all. This is an open argument between specialists, not a settled fact in either direction.

The number that is not in doubt. The species came within a few animals of disappearing. Thirty-one horses survived into the end of the Second World War, and only nine of them became ancestors of the captive population. Every Przewalski’s horse alive today descends from those nine. They were bred back from that, reintroduced to the steppe, and in 2005 the species was moved from Extinct in the Wild to Endangered.

Limits

Ancient DNA is inference from a small number of recoverable samples, and forty-two genomes is not many for a continent and five millennia. Whether the Botai horses were domesticated in any full sense is itself argued over - herding, milking and corralling are not the same as domestication. And “wild” is doing heavy lifting as a word: a population that went through nine founders and a century of captivity is not wild in the sense the phrase implies, whatever its ancestry turns out to be.

What this does not show

Not that Przewalski’s horse is definitely feral - that is the disputed half. Not that the conservation effort was pointless or misdirected; bringing a species back from nine founders is remarkable regardless of its taxonomy. And not that ancient DNA is unreliable: it produced the finding about modern horses that nobody has challenged.