AKT. 15 AISJ'ATOMY OF THE EARED AND EARLESS SEALS HOWELL 23 



apart transversely, the metapophyses are also farther apart. In both 

 there are hypapophyseal keels to all the lumbars. 



As previously mentioned the outstanding character of the vertebral 

 column as a whole is the looseness of the articulations. The cat is 

 looked upon as being rather a limber mammal, but its vertebrae are 

 far more securely interlocked than are those of the pinnipeds. 



Sacral vertebrae. — The sacral series in the pinnipeds is almost 

 always three, although Flower and Lydekker (1891) say four. In 

 all of the 47 individuals of divers sorts examined by Thomson (1909) 

 there were three, as there is in my Zalophus skeleton, but in the 

 Phoca hispida^ a Phoca groenlandica in the National collection, and 

 apparently always in Odohervus, they number four. In Zalophus they 

 constitute 11 and in the Phoca 14 per cent of the body length (8 in a 

 cat), but this detail is of but slight value because of the difference in 

 the two in the number of the sacrals. In the Phoca but not the 

 Zalophus the postzygapophyses of the last lumbar and prezygapo- 

 physes of the first sacral are so shaped as to allow the whole sacrum, 

 and thus the pelvis, to be elevated above the general vertebral axis, 

 bowing the back concavely at this point. The specimens available 

 had the articulation of the pelvis with the sacrum confined to the 

 first sacral except in Odotenus^ in which the first three sacrals were 

 involved. 



The only lateral processes in addition to the " transverse processes " 

 are poorly developed metapophyses. In the Zalophus the width of 

 the vertebrae decreases regularly from the first to the third. In the 

 Phoca the greatest width of the first is considerably greater than in 

 Zalophus^ to allow for the broadei" apaxial musculature, and there is 

 then a rather sharp constriction in width of the second. The trans- 

 verse processes of the third and fourth, fused into a single plate, are 

 again much broader, and the variation in this item must be due to 

 some detail of the sacrospinal musculature that was not detected, 

 for there is no difference in the origin of any of the hip or thigh 

 muscles sufficient to account for it. 



Caudal vertebrae. — Flower and Lydekker (1891) give the num- 

 ber of caudals as from 9 to 15, while Thomson (1909) says that in 

 47 individuals of 14 sorts of pinnipeds there were between 10 and 12. 

 In a single mounted Monachus tropicalis there are apparently 13, 

 and in a Phoca fasciata of the National collection, at least 14, with 

 the possibility that the terminal ossicle of the tail has been lost. 

 Although possibly unusual, it is therefore not startling to find that 

 there are 14 caudals in both Zalophus and my Phoca^ these bones 

 constituting respectively 24 and 35 per cent of the body length. In 

 the Phoca the transverse processes of the more cranial vertebrae are 

 broader, and the spinous processes are lower and broader cranio- 

 caudad. 



