Limbs of Vertebrate Animals. 33 



correspond to the radius and ulna. Nor is it really 

 otherwise with the scapular and pelvic arches, though 

 here the underlying con-naturality is not so conspicu- 

 ously displayed. Here, one thing seems to be tolerably 

 evident, namely this, that there is an intimate con- 

 nection between these arches and the ordinary ribs, and 

 that the limbs of vertebrate animals are to be looked 

 upon as costal processes. And certainly there is, on 

 comparing the scapular and pelvic arches, no difficulty in 

 seeing that the scapulae may answer to the ilia, the two 

 coracoid bones to the two ossa ischii, the two clavicles 

 to the two pubic bones, and the sternum to the carti- 

 laginous substance between the two pubic bones. The 

 scapulas and the ilia are more obviously related to the ribs 

 than are the clavicles and coracoid bones, or the pubic 

 and ischial bone, to the costal cartilages, but the rela- 

 tionship is not the less real on that account. And if so 

 then the chief difficulty is disposed of. The pelvic 

 arch generally is more developed than the scapular, but 

 the latter arch is also highly developed where, as in 

 birds, it is necessary to have a firm attachment for the 

 wings. Here, and also in reptiles with fully developed 

 legs, the scapulse are connected with the sternum, both 

 by the coracoid bones and by the clavicles by the 

 coracoids chiefly ; in mammals, on the contrary, with 

 the exception of the ornithorhynchus and echidna, 

 where the reptilian arrangement holds good, and in man, 

 where the clavicles are developed so as to form an 

 inverted arch in which the end of the sternum is the 

 keystone, the scapulas are free, and the coracoid and 

 clavicular bones alike are present only as insignificant 

 projections from the scapulae. Usually also the 

 pelvic arch is more complete than the scapular, but 

 not always. Indeed, in the whale and also in the 



D 



