94 University of California Publications in Geology [Vol. 13 



recently by Max Weber, and W. D. Matthew. In view of the highly 

 debatable nature of the evidence that has been offered, I hesitate some- 

 what to offer any opinion. Only two of these theories will be here 

 considered. The rest may be put in the doubtful column, for no 

 adequate evidence has so far been adduced to place these speculations 

 on an acceptable basis. 



The first theory which will be considered is the derivation of the 

 Pinnipedia from the Oxyaenidae, a family of inadaptive creodonts. 

 These Oxyaenidae of the Eocene fauna appear to correspond with the 

 Mustelidae among- modern Carnivora. The exponent for this deriva- 

 tion of the Pinnipedia from Pairiofelis or from the Oxyaenidae is 

 AVortman. 180 He states that the primitive pinnipeds retain many 

 characters in common with the Oxyaenidae, and has adduced the 

 following points which have yet to be explained before the ursid 

 derivation can be accepted. 



1. The presence of a subungual foramen. 



2. A large astragalo-cuboid contact. 



3. An oblique cubo-caleaneal facet. 



4. The exceptionally large size of the trapezium. 



Dr. Matthew 187 has pointed out the chief objections to Wortman's 

 theory and has adduced a series of characters to show that the evidence 

 is, instead, in favor of the derivation of the pinnipeds from the arctoid 

 Fissipedia. The following is his summary. 



1. The lachrymal is large and broadly expanded upon the face in both the 

 Inadaptive Creodont groups. In the Adaptive Creodonta it is smaller, in the 

 Pissipedia still f urtlier reduced, especially in Ursidae and some Mustelidae. In the 

 Pinnipedia it has entirely disappeared. 



2. The mode of molar reduction indicated in the Pinnipedia corresponds well 

 with that generally indicated in the Adaptive Creodontia and Pissipedia, and dis- 

 agrees fundamentally with the Oxyaendae and Hyaenodontidae. In the Pinnipedia 



M ^ are. always present and of large size; a small M- is occasionally present, but 



never any trace of M;. Their more generalized ancestors must therefore have had 

 2 . 3 



M * small, early reduced and lost. This agrees with the Adaptive Creodonts 



and Pissipedia, in particular with Mustelidae and Ursidae. In the Oxyaenidae 



2 3 



M — ~ — are early reduced and lost, but Mj is the largest of the lower teeth, and 



progressively increased and specialized. If the seals were descended from 



1 1.2 1 



Oxyaenidae their formula might vary from M — — — to M - — — but not from M - to 



1.2 1.2 1.2.3 



M . In the Hyaenodontidae M - — = — ^ or M - — 5 — 5- would be the molars 



1 1 ■ 2 • o 1 ■ a ■ o 



likely to be preserved. 



186 Wortman, J. L.. Bull. Am. Mus. Nat. Hist., vol. 6, p. 157. 1894. Am. Jour. 

 Sc.i. (4), vol. 13, no. 73, pp. 115-128. 1902. Science, n.s., vol. 24, p. 89. 1906. 



187 Matthew, W. D., The Carnivora and Insectivora of the Bridger Basin, Mid- 

 dle Eocene. Mem. Am. Mus. Nat. Hist., vol. 9, pt. 6, pp. 413-417. 1909. 



