34 PROCEEDINGS OF THE NATIONAL MUSEUM TOL. 73 



carpale 3 is markedly pyramidal, in palmar view being large and 

 dorsad appearing as a restricted bony point. In the Phoca also the 

 articulation of digit 5 with the unciform is upon the lateral side of 

 that bone and considerably proximad of the level of articulation of 

 the three middle digits. There is this difference, however ; whereas in 

 Zalophus the articular surface of metacarpus 5 is mediad, allowing 

 the digit to point straight distad, in the Phoca it is proximad as in 

 the other digits, and the fifth therefore is inclined to point laterad, 

 making this digit, in fact, very much more " opposable " than is the 

 pollex. The explanation for the narrow carpus in Phoca^ and for the 

 movement distad of the first digit and both proximad and laterad of 

 the fifth, is probably that in the phocid the motions performed by the 

 manus during swimming are of a pivotal, paddling type, much as are 

 those of man when he is maintaining a static position in the water 

 by means of his hands alone. The explanation for the movement 

 distad of the pollex and both proximad and laterad of the fifth lies 

 probably in the extreme amount of abduction of the manus in relation 

 to the antibrachium, of which this pinniped is capable. In an articu- 

 lated manus of Monachus tropicalis there is even more pronounced 

 abduction in the normal position of the fifth digit, the first carpale is 

 several times the size of the second, but the latter does not completely 

 separate, in dorsal view, the first from the scapholunar. 



In both animals the pollex is the longest digit and there is sequen- 

 tial reduction in length to the fifth, although this is much more 

 pronounced in the Zalophus. In the latter the first metacarpal is 

 much the longest and very much more robust than the others, and it 

 is sharply flattened upon its medial border, possibly to assist a stream- 

 line form of the external surface of the flipper. In this animal it 

 is slightly longer than the first phalanx of the same digit, while in 

 the phocid the reverse is the case. In the otariid the fifth digit is 

 much less robust than the first, but more so than the other three, 

 while the third is a trifle the most slender. There is present a decided 

 tendency toward the flattening of the digital elements which in the 

 Phoca is encountered only to very slight degree in the pollex alone? 

 Kukenthal (1890) reported indications of double epiphyses in the 

 phalanges of the manus of several pinnipeds, both otariid and 

 phocid. Such may be the case, but thej^ are not convincing in the 

 material at my disposal. In Zalophus the terminal phalanges are 

 flattened and distorted in characteristic fashion and contain pits for 

 the rudimentary nails. In the Phoca the broad, well-developed nails 

 are retained in the cleaned skeleton. 



Pelvic girdle. — There are fundamental differences of a very in- 

 teresting character existing in the innominate bones of otariids and 

 phocids. The practical comparison of their measurements is ren- 

 dered difficult, however, by the fact that in the earless seal the ilium 



