288 BREEDING 



colts under observation with as natural surroundings as circumstances 

 permitted. 



Having fixed on the size of the horses to be studied, it was next neces- 

 sary to consider how the rate of growth before and after birth could be 

 best determined. 



The difference between a tall and an undersized man is mainly a differ- 

 ence in the length of the legs; but in the case of the horse the height, as 

 commonly understood, instead of bearing, as in man, an intimate relation 

 with the length of the hind-limbs, is intimately related to the length of the 

 fore-limbs. 



The height of a horse, it is hardly necessary to state, depends mainly 

 on (1) the length from the elbow to the ground; (2) the length and 

 obliquity of the arm-bone (humerus)', and (3) the length of certain spines 

 of the dorsal vertebrae, the spines which give rise to the more or less arched 

 ridge known as the withers. 1 



In the living animal it is impossible to measure the length of the 

 vertebral spines, and only possible to estimate roughly the length and 

 obliquity of the humerus, and hence it will be necessary in studying the 

 rate of growth in the horse to trust chiefly to the length of the fore-limb 

 as measured from the elbow to the ground. In man the limbs belong to 

 the common or ordinary vertebrate type, but in the horse they have 

 departed as far from the general plan as highly useful structures well 

 could, for instead of five digits, as in man, there is but one complete 

 digit, and in their hard parts the limbs are infinitely more highly special- 

 ized than is the case in any other mammal, and more profoundly altered 

 than even the wing of a bat. 



Influenced by the doctrine of recapitulation (the belief that each animal 

 climbs its own ancestral tree), not a few were wont to believe that when a 

 sufficiently young horse embryo was examined, the fore-limbs at least, as 

 in the early Eocene " fossil horses ", would be pentadactylous, i.e. have 

 rudiments of five digits. This, however, is not the case; at no stage in 

 the development (in the life -history as distinguished from the ancestral 

 history) of the horse are there any visible rudiments or vestiges of the first 

 and fifth digits. In other words, the horse is at the most tridactylous, and 

 only one of the digits the one corresponding to the human middle finger 

 in front and the human middle toe behind ever comes into use. 



In the case of the horse, the first rudiments of limbs appear in the form 

 of short bud-like outgrowths between the twenty-first and twenty-eighth 



1 Than the height at the withers it would be difficult to find a less trustworthy index of the size of a horse. 

 The height at the elbow is a safer guide, or, seeing that a horse (like a man) propels itself by the hind-limbs, 

 the height at the croup should be taken into consideration. 



