Where Are There Still Wild Horses in the World?

Free-ranging horses roam every continent except Antarctica, but whether any of them qualify as truly “wild” depends on how you define the word. The roughly 1,800 Przewalski’s horses scattered across reintroduction sites in Mongolia, China, and a few European reserves are the closest thing to a genuinely wild horse species, yet even their status is debated. Beyond them, tens of thousands of feral horses descended from domestic stock live in places as varied as the American West, the Australian Alps, Canada’s Sable Island, Namibia’s desert, Romania’s Danube Delta, and Japan’s Cape Toi. The distinction between wild and feral is not just academic; it shapes which horses get conservation funding and which get culled.

Przewalski’s Horses in Central Asia

Przewalski’s horse is the species most people mean when they ask about wild horses. First described by Western science in the 1870s, these stocky, erect-maned horses are genetically and physically distinct from domestic breeds. Genome sequencing shows they split from the domestic horse lineage roughly 45,000 years ago, though some gene flow continued after that divergence.¹ They went extinct in the wild during the 1960s, victims of habitat loss, hunting, and competition with livestock. Every Przewalski’s horse alive today descends from just 12 individuals captured from the wild between 1898 and 1947.²

Captive breeding programs kept the species alive, and reintroduction began in the 1990s. Today, the largest free-ranging populations graze in Mongolia’s Hustai National Park, the Great Gobi B Strictly Protected Area, and several sites in China. Smaller groups have been released in Russia’s Orenburg Reserve and, somewhat unexpectedly, in the Chernobyl Exclusion Zone in Ukraine, where monitoring has confirmed the horses prefer open meadows but will also use the forested landscape that has reclaimed former farmland there.³ In France, a population managed “as wild” since 1993 has shown that allowing natural herd behavior promotes grassland diversity, giving conservation managers an additional argument for maintaining these herds.⁴

The genetic bottleneck from those 12 founders remains a serious concern. With so little founding diversity, inbreeding is a persistent threat. One creative solution involves cloning from cryopreserved cells. In 2020 and 2023, two healthy Przewalski’s horse clones were produced from cells belonging to a stallion who died in 1998 but whose genetic lineage was underrepresented in living populations.² The hope is that this technology can inject lost genetic variation back into the breeding population.

Are Przewalski’s Horses Actually Wild?

A 2018 ancient-DNA study complicated the picture considerably. Researchers compared genomes from 46 ancient and modern horses and concluded that Przewalski’s horses are the feral descendants of horses herded at Botai, a site in northern Kazakhstan where the earliest known horse domestication occurred around 5,500 years ago.⁵ If that finding holds, then Przewalski’s horses are not a never-domesticated wild lineage but instead a population that went feral thousands of years ago and has since diverged sharply from modern domestic horses.

The debate is far from settled. Other geneticists point to the deep divergence between Przewalski’s and domestic horse genomes, the clear differences in chromosome number (Przewalski’s horses carry 66 chromosomes versus 64 in domestic horses), and the behavioral and physical traits that set them apart. Whether you call them “wild” or “anciently feral” changes the narrative but not the conservation math: they remain an endangered species with irreplaceable genetic diversity, and losing them would mean losing the closest surviving relative of the horses that once ranged freely across the Eurasian steppe.

North American Mustangs

The American West is home to the best-known feral horse populations on earth. Tens of thousands of free-roaming horses, commonly called mustangs, live across ten western states on public lands managed by the Bureau of Land Management. These horses descend from Spanish stock brought to the Americas starting in the 1500s, with later mixing from British and other European breeds.⁶ Genetic analysis of nearly 8,500 horses from 235 populations, sampled over more than two decades, reveals substantial genetic diversity and disparate ancestry across herds, reflecting centuries of different founding events and varying degrees of isolation.⁷

Mustangs are legally protected under the 1971 Wild Free-Roaming Horses and Burros Act, which gives them an unusual status: not livestock, not wildlife, but something in between. That legal limbo fuels fierce political arguments. Ranchers and ecologists often view the horses as an invasive species that degrades rangeland, competes with native wildlife, and accelerates erosion. Horse advocates see them as a living symbol of the frontier, deserving of protection. The BLM sets “appropriate management levels” for each herd area and historically relied on periodic roundups to remove surplus animals, but populations tend to grow faster than removals can keep up.

Immunocontraception has emerged as one management tool. A controlled-release vaccine called PZP-22, administered to mares, has shown it can cut herd growth rates substantially. In one challenging management area, treating about 70% of the mares dropped the foaling rate to around 20% and reduced population growth by roughly three-quarters relative to baseline levels.⁸ However, follow-up research on a different population found that previously treated mares were less likely to give birth even after treatment stopped, raising questions about long-term fertility effects.⁹ Balancing population control with animal welfare and public sentiment remains one of the thorniest conservation management problems in North America.

Australian Brumbies

Australia’s feral horses, called brumbies, have been free-ranging since the early colonial period. They are found in several regions, but the most contentious populations live in the alpine and subalpine parks of New South Wales and Victoria. Here, brumbies trample fragile bog communities and threaten species found nowhere else. Research in the Australian Alps has found that even small numbers of horses, on the order of a hundred animals, cause damage to critically endangered wetland ecosystems that could take thousands of years to recover, and those impacts accumulate over time rather than reaching some stable plateau.¹⁰

Despite the ecological evidence, brumbies hold deep cultural significance for many Australians, immortalized by “The Man from Snowy River” and similar folklore. Legislative battles have played out repeatedly: New South Wales passed a law in 2018 protecting brumbies in Kosciuszko National Park, only for debate to continue over whether aerial culling or trapping and rehoming should take priority. The situation mirrors the North American standoff between conservation science and cultural attachment, though the ecosystems at stake and the management options differ.

Namibia’s Desert Horses

In the Namib Desert near the abandoned diamond-mining town of Aus, a population of feral horses has survived since at least the early 20th century, though their exact origin is disputed. Various accounts trace them to German cavalry horses released during World War I, to horses from a nearby breeding farm, or to some mix of both. What makes them scientifically interesting is their persistence in one of the most arid environments on earth. Physiological research found that these desert horses are not fundamentally different from domestic horses in their water-handling ability per unit of body mass. Instead, their smaller body size and consequently lower total water turnover appear to be the key adaptations that let them survive in a landscape where water sources can be separated by vast distances.¹¹ The population is small and vulnerable to drought, predation, and fluctuating tourism pressure, and it has no formal legal protection comparable to American mustangs.

Sable Island, the Danube Delta, and Cape Toi

Some of the world’s most isolated free-ranging horse populations live on islands or in geographically enclosed wetlands, and each tells a different story about what happens when domestic horses go feral and stay that way for centuries.

Sable Island, a narrow sandbar about 300 kilometers off the coast of Nova Scotia, supports roughly 500 feral horses that have been there since at least the 1700s. Genomic analysis shows they carry higher levels of inbreeding than typical domestic breeds, with patterns suggesting population bottlenecks in their history. Interestingly, the genomic regions that have become fixed in these horses are enriched for genes involved in metabolism and immune function, hinting at natural selection shaping the population in ways distinct from artificial selection in domestic breeds.¹² The horses are legally protected as part of Sable Island National Park Reserve, and no management interventions are carried out. Even their hooves have adapted to the sandy terrain in ways that look abnormal by domestic standards: virtually all horses examined showed long heels, contracted heels, and other conformational differences compared to shod or trimmed domestic horses.¹³

In Romania’s Danube Delta, one of Europe’s largest wetlands, feral horses have been documented in the Letea Forest area and the region between the Sulina and Chilia branches of the Danube.¹⁴ These horses live alongside domestic equines in a mosaic of marshes and forest, and they attract some ecotourism interest, but they receive far less research attention and legal protection than their North American or Australian counterparts.

At Cape Toi on the southern tip of Japan’s Kyushu island, the Misaki horse has lived in a semi-feral state for centuries. Genetic analysis of roughly 80% of the population identified low allelic diversity and evidence of a historical bottleneck, though not a severe or recent one.¹⁵ A follow-up study comparing data from 2015 and 2020 found no significant decline in genetic diversity over that five-year window, and estimated an average generation length of about 4.6 years, faster turnover than in most other native horse breeds.¹⁶ The Misaki horses are designated a natural monument in Japan, which gives them some institutional protection, though the population remains small and genetically constrained.

European Rewilding Projects

A growing number of nature reserves across Europe are introducing horses as ecosystem engineers, blurring the line between conservation and agricultural land management. In the Netherlands, Britain, Denmark, and several other countries, horses and cattle are released together into large fenced areas with the goal of recreating the kind of grazing pressure that would have existed before large herbivores were pushed out of the European landscape.¹⁷ Tracking data from Danish rewilding sites show that both horses and cattle preferentially use open habitats like dry grassland and heathland over forest, though with enough individual variation to complicate simple predictions about where they will graze.¹⁸

The breeds used vary. Some projects use Konik horses, a Polish breed that was selectively bred in the early twentieth century from a small founder population of six male lines with the goal of recreating the appearance of the extinct tarpan, Europe’s original wild horse.¹⁹ Others use Exmoor ponies, Icelandic horses, or various hardy breeds thought to tolerate minimal human management. The genetics, welfare, and ethical considerations of these programs are still being worked out. Questions include whether the horses should receive veterinary care or be left to natural population regulation, how inbred their founding groups are, and whether calling them “wild” in public communications overstates their ecological role.

The tarpan itself, which once ranged from western Europe into Russia, went extinct by the late 1800s. Attempts to “breed back” the tarpan by selecting domestic horses that looked the part produced breeds like the Konik and the Heck horse, but these are domestic horses shaped by human selection, not recovered wild animals. The strict inbreeding that went into creating the Konik from just six male lines has left the breed with elevated inbreeding levels, which may undermine the very conservation purposes for which they are deployed.¹⁹

Horses That Used to Be Everywhere

The current patchwork of small, scattered populations is a dramatic contraction from the horse’s original range. Molecular analysis of ancient horse remains from across the Americas and Eurasia has confirmed that during the late Pleistocene, a single caballine horse species ranged continuously from western Europe through Siberia and into the Americas, making it one of the most widespread large mammals on earth.²⁰ North America was also home to a separate stilt-legged horse lineage that is now entirely extinct. All native American horses disappeared at the end of the Pleistocene, around 10,000 years ago, along with many other large mammals.

When Spanish colonizers reintroduced domestic horses to the Americas in the 1500s, horses spread rapidly through Indigenous exchange networks, reaching the northern Rockies and central Great Plains by the first half of the 1600s. Archaeological, genomic, and isotopic evidence shows that Indigenous peoples were deeply integrating horses into their societies, managing herds and incorporating horses into ceremony, well before European observers arrived to document it in the 1700s.⁶ Today’s North American mustangs carry strong Iberian genetic signatures with later British admixture, a molecular record of these colonial and post-colonial waves of introduction.

Why the Wild-Versus-Feral Question Keeps Coming Up

In wildlife management and conservation law, “wild” and “feral” are not interchangeable labels. A wild species evolved in its environment without human intervention. A feral animal is a domestic species that escaped or was released and now lives independently. The distinction matters because wild species can qualify for endangered-species protections, habitat conservation funding, and international treaties, while feral animals are often classified as invasive pests and targeted for removal.

This is why the Botai question about Przewalski’s horses is not just a historical curiosity. If they are truly wild, they are the last of their kind, and their conservation is a global priority on par with saving any other endangered megafauna. If they are anciently feral, some argue they belong in the same legal and ethical category as brumbies and mustangs, animals with cultural value but no inherent claim to the landscapes they occupy. Most conservation geneticists, though, point out that after thousands of years of independent evolution, the distinction has become biologically meaningless: Przewalski’s horses carry unique adaptations, unique chromosomal structure, and genetic diversity found nowhere else.

For the feral populations, legal status varies wildly by country. American mustangs enjoy federal protection. Sable Island horses are shielded by national park status. Misaki horses are a Japanese natural monument. Australian brumbies sit in a legal gray zone that shifts with state politics. The Namibian desert horses have essentially no legal standing. And European rewilding horses often fall under agricultural regulations, classified as livestock even when the entire point of the project is to let them behave as if they were wild. These inconsistencies mean that two nearly identical populations of free-ranging horses, separated by a national border or a shift in political sentiment, can face opposite fates: one group protected, the other earmarked for removal.