The history of horse domestication has been subject to much debate, with various competing hypotheses over time about how domestication of the horse occurred. The main point of contention was whether the domestication of the horse occurred once in a single domestication event, or that the horse was domesticated independently multiple times. The debate was resolved at the beginning of the 21st century using DNA evidence that favored a mixed model in which domestication of the stallion most likely occurred only once, while wild mares of various regions were included in local domesticated herds.
In the 20th century, various ideas were postulated. One set hypothesized multiple ancestral body types of the single species Equus ferus or the original wild horse, each adapted to a given environment. Another hypothesis held that prototypes originated from a single wild species and that all different body types were entirely a result of selective breeding after domestication. Yet another proposed that the theoretical prototypes were each separate species or subspecies. These theories were all based on body types and conformation, prior to the availability of DNA for research, and have since been superseded by modern studies.
Theories
In the past, several theories were proposed about the origin of the domesticated horse and how the variety in horse breeds developed. They generally can be subdivided in two categories, single origin versus multiple origins.
Single origin
The single origin theory holds that domestication occurred once, after which all breeds arose through selective breeding.
Multiple primitive types
A theory associated with James Cossar Ewart in Scotland and Johann Ulrich Duerst in Germany postulated three primitive horse types, considered subspecies of Equus caballus, as ancestors of modern breeds. They were: "Forest Horse", Equus caballus germanicus, descendant of a "Diluvial Horse", Equus caballus silvaticus Asiatic Wild Horse or Przewalski horse, then considered Equus caballus przewalskii Tarpan, then considered Equus caballus gmelini. To these Elwyn Hartley Edwards adds a fourth, the "Tundra Horse", supposedly ancestor of the Yakut pony, and "largely unconsidered by hippologists".
A later theory associated with European scholars such as Jimmy Speed, Ruy d'Andrade, Hermann Ebhardt and Edward Skorkowski, postulated four basic body types, which were not considered to be named species. They were: Pony Type 1, in northwestern Europe, resistant to cold and wet, similar to the modern Exmoor pony Pony Type 2, in northern Eurasia, larger than type 1, resistant to cold, similar to the modern Highland pony and Fjord horse Horse Type 1, in central Asia, resistant to heat and drought, similar to the modern Sorraia and Akhal-Teke Horse Type 2, in western Asia, small and fine-boned, resistant to heat, similar to the modern Caspian horse.
American paleontologist Deb Bennett postulated that the early form of E. caballus developed into seven subspecies, of which four supposedly contributed most to the ancestry of the domesticated horse, both directly and via assorted crossbred lineages between them. These were: "Warmblood subspecies", Equus caballus mosbachensis, the oldest hypothetical subspecies, supposedly ancestor of the Latvian horse, Groningen horse and some warmblood breeds. "Draft subspecies", Equus caballus caballus, ancestor of the Exmoor Pony, Shetland pony, Suffolk Punch and Belgian horse. "Oriental subspecies", Equus caballus pumpelli, adapted to arid climates, thought to be the progenitor of the modern Arabian horse, Plateau Persian and Marwari horse. "Tarpan", Equus caballus gmelini or Equus caballus ferus, supposed ancestor of Przewalski's Horse as well as the Konik, Vyatka horse, Hucul and most Mongolian horses.
The other three proposed subspecies were: the Przewalski Horse, Equus caballus przewalskii the "Lamut Horse", Equus caballus alaskae and the "American Glacial Horse", Equus caballus laurentius or Equus caballus midlandensis.
Current theory
Modern genetic evidence now points at a single domestication event for a limited number of stallions, combined with repeated restocking of wild mares into domesticated herds. This suggests that different body types might be a combination of both selective breeding and semi-feral landrace traits.