100 University of California Publications in Geology [Vol. 13 



occurred along with correlated changes in structure in each of the 

 three families of pinnipeds. 



An enquiry into the morphological characters of the Pinnipedia 

 leads one to the conclusion that the nature of the characters exhibited 

 by the various members indicates that the changes must have been 

 brought about by slow though gradual variations. Even if the den- 

 tition and the correlated changes in skeletal and anatomical structure 

 were modified before they sought new food or assumed new habits, 

 or, in other words, structure precedes function as held by most experi- 

 mentalists, it appears that the Pinnipedia passed through the following 

 stages of evolution : 



1. A terrestrial type, chiefly carnivorous with trituhercular dentition. The 

 skull must have possessed well developed sagittal and occipital crests. It must 

 have had an astragalus with slight or ungrooved trochlea, an astragalar foramen, 

 an entepicondylar foramen, and an alisphenoid canal. The vertebral formulae 



was: C7, D15, L5, S3-4, Ca8-12. The dentition was: I. — , C. — Pm — , M. — 



3-3' 1-1' 4-4' 2-2 



2. A subaquatic type subsisting in part on a carnivorous diet though largely 

 on turtles and fish. The teeth were greatly abraded because of the nature of their 

 food and probably exhibited an incipient retrogression of the tritubercular type 

 to the haplodont type. 



3. A pelagic type subsisting largely on cephalopods and fish. The skull pos- 

 sessed a high sagittal crest. The dentition now, in the otarids at least, was of 

 a haplodont type. 



4. A highly specialized pelagic type subsisting largely on fish and Crustacea. 

 Skull now marked by lateral expansion of parietals and reduction of sagittal crest. 

 Molariform teeth now characterized by the remarkable specialization of secondary 

 cusps. 



In conclusion it may well be stated that the origin of the Pinnipedia 

 is completely unknown as yet. From a careful study of this Temblor 

 form it appears that we must look for the ancestral otarids among the 

 Carnivora long before there were any true bears or wolves. Our evi- 

 dence for 'assuming such an early divergence from the parent stock is 

 warranted by the fact that the families Phocidae, Otariidae and 

 Odobenidae were already well differentiated from each other before the 

 end of the Miocene. These differentiations probably represented at 

 first local changes in adaptation to different feeding zones, terrestrial, 

 littoral, and pelagic. Furthermore, as specialization proceeded, each 

 of these types has diverged further and further from each other. 



With regard to the phocids, they may or may not have descended 

 from the same ancestor as the otarids. It is possible that much of the 

 confusion and uncertainty that has arisen in phylogenetic studies of 

 the pinnipeds is due to a polyphyletic origin. 



