May 31, 1906] 



NA TURE 



113 



IHK TARPAN AND ITS RELATIONSHIP WITH 



WILD AND DOMESTIC HORSES.' 

 CO much progress has been made during recent years in 

 ■^ worl<ing out the origin and history of domesticated 

 hiirses that the time has now come when inquiries may be 

 profitably pursued along certain definite lines. 



In the first place (assuming that horses have had a 

 imiltiple origin), inquiries should be instituted with the view 

 of ascertaining so far as possible the characteristics of 

 the post-Glacial species and varieties which have taken part 

 in forming the present domestic races and breeds ; in the 



British Museum, certain vestigial structures, known as 

 callosities, warts, or chestnuts, are of considerable taxo- 

 nomic value. Warts or chestnuts, as is well known, arc 

 present on both the fore and hind limbs of the common 

 horse, and they also occur on the hind as well as the fore 

 limbs of Prejvalsky's horse; while in the Celtic pony, as 

 in asses and zebras, the hind chestnuts are completely 

 absent. It is especially worthy of note that though the 

 hind chestnuts were not invariably present in Tarpans 

 (they were absent in a Tarpan described by Krymsch), 

 they were present In the Moscow specimen. 



It thus appears that the Moscow Tarpan agreed in its 



next place, inquiries should be instituted with the view of colour with the specimens referred to by Gmelin and Pallas, 

 ascertaining to which of the lower Pleistocene species the but differed in the mane and tail, in both of which, as in its 

 more immediate ancestors of the living horses are most callosities, it resembled the common horse, Equus caballus. 

 intimately related ; and, in the third place, an attempt Two Tarpan skeletons have been preserved. The chief 

 should be made to determine from which of the ancestral point of interest about these skeletons is fhat, as in the 

 forms the various domesticated breeds have inherited their kyang and Prejvalsky's horse, and in certain Arabs, there 

 more striking characters, i.e. to ascertain to which are only five lumbar vertebrae. 



ancestral types the Shire, Clydesdale, Percheron, and other In having only five lumbar vertebrae these Tarpans 

 heavy breeds, the Barb, Arab, thoroughbred, K.Ttti.Twar, differed from the common horse of Europe, at least from 

 and other slender-limbed breeds, are indebted for 

 thrir chief peculiarities. 



In this paper I shall not attempt to show that 

 eilher Prejvalsky's horse, the Celtic pony, or the 

 Libyan variety recently described by Prof. Ridge- 

 way - is genetically related to pre-GIacial species, 

 or entitled to be regarded as an ancestor of one 

 or more domestic breeds. 



Sufficient data for a discussion of this kind is 

 not yet available. I propose now, by way of 

 clearing the ground for the investigations men- 

 tioned above, to inquire whether the Tarpan (long 

 regarded as the wild progenitor of the common 

 horse of Europe) deserves a place amongst the 

 ancestors of living races and breeds. 



The first account of the Tarpan ' we owe to 

 Gmelin, who came across a troop near Bobrowsk 

 during his journey through Russia between 1733 

 and 1743. He describes them as mouse-coloured, 

 with a short, crisp mane; the tail always shorter 

 than in domestic horses, sometimes full, some- 

 times only furnished with short hair ; the legs 

 dark from the knees and hocks to the hoofs ; and 

 the head thick, with the ears sometimes long, 

 sometimes short. 



Since this description appeared, some Conti- 

 nental naturalists have regarded the Tarpan as a 

 true wild species ; others, like Dr. Nehring, con- 

 sidered it the last survivor of the ancient pre- 

 historic horses of Europe modified by an infusion 

 of domestic blood ; while not a few agreed with 

 Pallas that the Tarpan herds might very well be 

 the olTspring of escaped domestic horses. 



English naturalists have, as a rule, adopted the 1905. 



view of Pallas. . . •■• i , 



Notwithstanding all that has been written on the sub- j the forest variety E. caballus iypicus, m which I have 

 ject since Gmelin's time, hippologists agree with Salensky invariably found eighteen pairs of ribs and six lumbar 

 that the relationship of the Tarpan with wild and domestic ' vertebrae. . , ,, 



horses has not vet been cleared up.' [ From this striking difference in the skeleton it follows 



Durina the 'nineteenth centurv very little was done ' that, even should the Tarpan turn out to be a true wild 

 towards determining the systematic position of the Tarpan ; ' species, it cannot be regarded as the sole ancestor of the 

 but in 1S66 a Tarpan foal was captured in the Zagradoffe common horse of Europe. 

 Steppe and reared by a domestic mare. When about As to the skull of the M 

 eighteen years old this specimen was sent to the Moscow ; to the conclusion that it has, on 

 Zoological Garden, and eventually described in a paper characteristics of Oriental horses, while on the oth. 



. I.— The mouse-dun 

 Welsh pony. This \ 

 established race. It h 

 full eyes, good quaite 

 semi-erect, while the t 



thickn 



ny, though a cross, looks as if it belonged to an old- 

 -, a striking, well formed, massive head, well-placed ears, 

 , and excellent limbs. The mane is, however, short and 

 1 consists of three kinds of hair which differ in structure. 



colour, and arrangement. From a photograph taken September, 



loscow skeleton, Czerski came 



published by Schatiloff. 



This, like Gmelin's specimen, had a somewhat coarse 

 head, was of a mouse colour, with legs black below the 

 knees and hocks. The mane, however, instead of being 

 short and crisp, as in Gmelin's specimen, was 48 cm. in 

 length and hanging to one side of the neck. 



As clearly realised some years ago by Gray, of the 



1 By Prof. J. 

 Royal Society of Edii 



"Ori 

 o?V 



nd Influ 



•t, FR.S. 

 ,rgh, Sessio, 

 ence .jf th^ 



Abridged fron 

 ; Thoroughbr 



:d Ho 



edings of the 

 (Cambridge, 



By the Tarpan I mean the mouse-dun horse of R 

 Continent.-il naturalists, not the so-called " 

 Smith ("Naturalists' Library," vol. xii., t84T). 



■* The chief papers on the T.-irpan are mentioned by Sal 

 graph on Prejvalsky's Horse " (St. Petersburg, 1902). 



pan of Ha 

 ky, '■ 



approaches the Scottish breed to which belongs the pony. 



The skull of the Tarpan in the St. Petersburg Museum 

 resembles skulls of immature specimens of E. prejvalskii, 

 but the bones of the limbs and limb girdles are decidedly 

 more slender, and have less pronounced muscular ridges 

 than in the wild horse of Central Asia. 



It may here be mentioned that for more than a century 

 all the horses living in a wild state in Europe, w-hich 

 happened to be of a mouse-dun colour, seem to have been 

 regarded as Tarpans. 



Seeing that herds of mouse-dun wild horses no longer 

 occur in Europe, and have not during recent years been 

 met with in even the most remote parts of Central Asia, 

 it might perhaps be assumed that the Tarpan 's place in 

 nature must for ever remain a mvstery. 



NO. 1909, VOL. 74] 



