[20] Across time and space, wild horses show considerable variability in body size and limb dimensions, likely as adaptations to local environmental and climatic conditions.

[25][26][22]ST2 Earlier Middle Pleistocene horses (such as those assigned to E. (ferus) mosbachensis) differ from later ones in some aspects of the anatomy of the limbs and teeth.

[27] Wild horses show considerable environmental tolerances, having historically inhabited environments ranging from temperate forest to steppe[27] (including the prehistoric mammoth steppe biome[28]) and tundra,[29] though in general they tend to show a preference for open environments.

[30] They may have seasonal food preferences, as seen in the Przewalski's horse,[31] which historically consumed browse like shrubs during the winter months due to being forced into suboptimal habitat by human pressure.

[32] Some extinct Pleistocene wild horse populations that inhabited forested environments show dental wear suggesting them to have been mixed feeders or even predominantly browsers during certain times of the year, though this may be reflecting the consumption of low growing forbs rather than shrubs.

The exact categorization of Equus remains into species or subspecies is a complex matter and the subject of ongoing work.

[27] The horse family Equidae evolved in North America, with the genus Equus appearing on the continent during the Pliocene (5.3-2.6 million years ago).

[44] Around 900-800,000 years ago, at the Early-Middle Pleistocene boundary, the ancestors of Eurasian wild horses crossed over the Bering Land Bridge from North America.

[27] North American caballine horses (which are variously referred to species like Equus scotti and Equus lambei though the true number of species is uncertain) which genetic evidence has confirmed are closely related to Eurasian horses (with some authors treating all North American horses as part of E. ferus) would persist on the continent until they became extinct as part of the end-Pleistocene extinction event along with most other large mammals in the Americas (including other equines like Haringtonhippus and Hippidion) around 12,000 years ago.

[10][11][42] A 2015 study determined that the Przewalski and domesticated horse lineages diverged from a common ancestor about 45,000 years ago.

[57] In some sources including MSW 3 (2005), the domesticated and wild horses were considered a single species, with the valid scientific name for such a single horse species being Equus ferus,[58] although MSW erroneously used E. caballus for this (enlarged) taxon on account of a mis-interpretation of the then-recent ICZN ruling on the matter,[59] refer Groves & Grubb, 2011.

[75][76] As of 2005, a cooperative venture between the Zoological Society of London and Mongolian scientists has resulted in a population of 248 animals in the wild.

[80] In 1995, British and French explorers encountered a new population of horses in the Riwoche Valley of Tibet, unknown to the rest of the world, but apparently used by the local Khamba people.

Examples of sites demonstrating horse butchery by archaic humans include the Boxgrove site in southern England dating to around 500,000 years ago, where horse bones with cut marks (with a horse scapula possibly exhibiting a spear wound[89]) are associated with Acheulean stone tools made by Homo heidelbergensis,[90][91] the Schöningen site in Germany (also thought to have been created by Homo heidelbergensis) dating to around 300,000 years ago, where butchered horses are associated with wooden spears (the Schöningen spears, amongst the oldest known wooden spears),[92][93] as well as the Lingjing site in Henan, China dating to 125-90,000 years ago.

[96] Early Paleoindians in North America hunted the continent's native horses shortly prior to their extinction.