pony head

the international Equine research site

PHYLOGENETIC NETWORK PROVIDES ANSWERS TO QUESTIONS REGARDING THE  ORIGINS OF TODAY'S HORSES by Thomas Jansen und Hardy Oelke

A phylogenetic network, constructed by a German/British Team of  (Jansen et al., 2002), and based on the largest currently available data bank, provides answers to questions such as: Are domestic horses descendants of one or more postglacial primeval horses? Has there been one or more domestication events? Are genotypes geographically linked? Do the pony breeds of northern and western Europe have common ancestors? Are Arabian and Barb horses related? Is the Sorraia horse the ancestor to modern Andalusians and Lusitanos? Is the Mongolian wild horse ancestral to domestic breeds?

 

Included in the mitochondrial D-loop sequencing were 318 horses from 25 oriental and European breeds, American mustangs, and Mongolian wild horses. Together with previously published data, including such from prehistoric permafrost horses, this amounted to 652 horses, the largest data bank currently available. The phylogenetic network constructed on the basis of these sequences showed 93 different mtDNA types, which grouped into 17 distict phylogenetic clusters. The network revealed also that several genotypes correspond to geographic areas, and/or breeds, indicating geographically distant domestication events.

 

The sheer number of different mtDNA types found prove the existence of different postglacial primeval horses, which, according to zoological systematics, should be referred to as subspecies, and which evidently were sources for the domestication process. A consideration of the horse mtDNA mutation rate, and the archeological timeframe, requires that a minimum estimate of 77 mares had to be recruited from the wild for the domestication process, and must have successfully reproduced in captivity/domestication. Each of these 77 mares must have been of different genotype, which, according to the results of this study, at least in part also meant distant geographic areas. This presents a feat absolutely unrealistic for any prehistoric community to accomplish. The conclusion is that the extensive genetic diversity of these 77 ancestral mares means that several distinct horse populations were involved in the domestication of the horse.

 

Are today's wild horse populations (i.e. Mongolian wild horse) of diminished genetic diversity due to bottleneck effects? Did prehistoric wild horse populations have such great genetic diversity that different genotypes found today may derive from the domestication of horses of just one population? The answer is no: Analyses of Alaskan permafrost horses spanning 16,000 years show six of the eight ancient mtDNA samples to cluster monophyletically.

 

Of the genotypes which correspond with geographic areas, cluster C1 is the most striking one: It is geographically restricted to central Europe, the British Isles, and Scandinavia, including Iceland. 17 of 19 documented horses of the C1 type are North European ponies: Exmoor, Norwegian Fjord, Icelandic pony, and Scottish Highland. Furthermore, 14 of 27 insufficiently documented horses of C1 type were ponies, including Connemaras. Two ancient Viking horses were also found to have the C1 type. Another mtDNA type, cluster E, consisted entirely of Icelandic, Shetland, and Fjord ponies.

 

Another geographically striking cluster is D1. Its widespread distribution is no surprise, given the strong influence Iberian horses have had on most domestic breeds, but there is a clear frequency maximum in Iberian breeds (Andalusian and Lusitano) as well as in North African horses (Barbs). A high percentage of American mustangs could be expected to be also of this genotype, due to the historical Spanish presence in the Americas, which mtDNA sequencing did confirm (31 %). The results show clearly that Andalusians, Lusitanos, and Barbs originally belong to the same population, but the almost complete absence of this genotype in Arabian horses (only 5 %) prove that Arab and Barb horses are not related and of different genetic origin.

 

Sorraia horses originate from a small group of horses obtained and preserved by the late Ruy d'Andrade, after he had seen phenotypically identical wild horses in 1920 near Coruche, Portugal. Although d'Andrade considered this horse to be the ancestor of today's Andalusians and Lusitanos, the mtDNA research shows the Sorraia to have a quite different origin. All 18 Sorraias sampled for this study had either of two A1 mtDNA types, both on the same branch in the phylogenetic network (61 % A1 root type, 39 % ancestral JSO41 type), which are quite unrelated to the D1 type predominant in other Iberian horses.

 

The Mongolian wild horse (Przewalski's horse) provides another geographically linked genotype (central Asia). Two mtDNA types were found for the Mongolian wild horse in this study. Ishida et al. had published a third mtDNA type, all three are closely related. In agreement with Ishida et al., it was found that the Mongolian wild horse had no ancestral role in regard to domestic horse breeds - the mtDNA types of the Mongolian wild horse were not found in any other breed or race.

 

Reference: T. Jansen, P. Forster, M. A. Levine, H. Oelke, M. Hurles, C: Renfrew, J. Weber, K. Olek: MITOCHONDRIAL DNA AND THE ORIGINS OF THE DOMESTIC HORSE. Proc. Natl. Acad. Sci. USA 99 (16), pp. 10905-10910 (2002)

 

return to EquiScience home page