ANIMAL LOCOMOTION. 89 



When the outer border is longer and stouter than the inner, 

 the first-named parts of support are of more importance than the 

 last. The difference in this regard between the foot of man and 

 that of the bear should be the difference between the respective 

 lengths of the first and the fifth toes. 



When the inner toe is rudimental, or lost, the foot has left the 

 ground before the inner border is reached. But it is not so evident 

 why the fifth toe should be lost. One would suppose that this toe 

 would be the most persistent. The sequence of the limbs of the 

 horse from the five- to the single-toed forms would be of interest 

 to study in this connection. 



Rotation of the Limb. 



In another relation than that included in the foregoing the 

 study of the photographs has been of interest. Allusion is partic- 

 ularly made to the forward swing of the fore limb and the outward 

 rotation of botii limbs (for inward rotation see p. 59). 



The extent of the forward swing is found to correspond to the 

 extent of the trochlear surface of the humerus as it enters into 

 the composition of the elbow-joint. In addition to this fact atten- 

 tion may be called to the difference in the length of the olecranon 

 in different animals. The olecranon is greatly longer relatively 

 to the shaft of the ulna in primitive types of Mammalia than in 

 the more specialized, as the writer has been informed by Professor 

 W. B. Scott. The same difference is found to obtain between 

 Dicotyles and Sus. May it not be claimed that the difference 

 correlates with the facilities possessed of maintaining the fore limb 

 in backward strain ? The enormous leverage secured by a long 

 olecranon must enable the limb to maintain a strong position for 

 the limb when it is held in backward strain as compared to one 

 in which the olecranon is short. This must be especially marked 

 when the humerus is long, as is always the case in these types. 



The mechanism which accompanies the rotation of the limb 

 outward at the knee after it has passed the vertical line is as- 

 sociated, as has been seen, with inward rotation of the hock. 

 This form of torsion is described (see p. 45), and the springing 

 power of the limb is a resultant. The amount of springing power 

 should correlate with the torsion, and the facets by which twists 

 are effected be indices of the act. The conversion of the calcaneo- 



