12 • THE HORSE FAMILY. 



terminate inferiorly in thin scale-like extremities. In some 

 instances, however (as in the fore limb of Stockwell, exhibited in a 

 wall-case), they have a small flat expansion at the lower end, and 

 from the evidence of the Shire Horse bones these expansions maj'^ 

 be definitely identified with the lateral toe-bones of the three-toed 

 ancestors of the Horse. In a certain sense, therefore, a con- 

 siderable number of existing Horses are really three-toed animals. 



Whether the highly developed condition of the splint-bones and 

 rudimentary toe-bones of the Shire Horse indicates an ancestry 

 for that breed distinct from either the Arab thoroughbred stock 

 on the one hand, and the ordinary North European stock on the 

 other, or whether it is merely a kind of reversionary redundancy 

 due to the large size and vigour of the Shire, is difficult to decide. 



Variation of another type has formed the subject of a paper by 

 Mr. 0. C. Bradley in the Proceedings of the Edinburgh Physical 

 Society. The trapezium of the carpus of the Horse is the structure 

 discussed ; and it is shown that this bone is present, either in one 

 or both hmbs, in about 50 per cent, of the skeletons examined, 

 while if each carpus be taken separately (that is, without reference 

 to the condition in its fellow) the percentage is a little more than 

 40. This, in conjunction with its minute size, leads to the 

 conclusion that in the evolution of the single-toed foot of the 

 Horse the bone in question is following in the steps of the lateral 

 metacarpal (splint-bone) with which it was originally connected. 



Another point of interest is the existence of a 



variation in me considerable amount of difference in the structure 



onape 01 me ^|? ^j^^ hooi in the various members of the Horse 



_ .. tribe ; this being illustrated by a series of 



^' specimens (N.H. 81-89) in the large case on the 



north side of the North Hall. In the Horse [Equus caballus) the 



" frog," or central cushion is reduced to a narrow ridge, dee()ly 



grooved posteriorly, which does not extend behind the case of the 



hoof, and is not applied to the ground. In Grevy's Zebra {E. 



grevyi), of North-east Africa, the frog becomes much broader, 



with scarcely any trace of the groove, and its hind part touches 



the ground. In the Kiang [E. hcmionus kiang), of Tibet, the 



posterior develo])ment of the frog becomes more marked, so that a 



considerable portion projects behind the case of the hoof and 



