I 



163 



extreme, at the point where the. interorbital region merges with the braincase (pers. obs.). 

 This posterior shift arises somewhere within the lutrines and is retained throughout the 

 pinnipeds. The lutrines and basal monachines generally possess the intermediate condition 

 (state 1), while the remaining pinnipeds are characterized by the more extreme 

 morphology. This shift to the anterior end of the braincase may be a synapomorphy of 

 the pinnipeds, with state 1 being derived ancestrally in the monachines (ACCTRAN 

 optimization), or the ancestral state for the pinnipeds and phocids may be equivocal 

 between states 4 and 2 (DELTRAN optimization). 



191) relative position of vertebrarterial (= intervertebral) foramen of atlas: 0 = visible in 

 dorsal view; 1 = visible in posterior view (King 1966; Wyss 1988a). 



Another distinction between the phocid subfamilies concerns the position of the 

 vertebrarterial foramen of the atlas. In phocines, as in most carnivores, the foramen is only 

 visible in posterior view. Only among canids and monachines does the foramen become 

 visible in dorsal view (King 1966; Wyss 1988a). Some problems do exist with this 

 character. Monachus spp. more closely approach the typical carnivoran pattern (King 1966; 

 Wyss 1988a). As well, the unusually large size of the foramen in Odobenus and Monachus 

 tropicalis makes it at least partially visible in both dorsal and posterior views. Here, a 

 dorsally visible vertebrarterial foramen was only present for Hyclrurga among the 

 monachines, along with a convergent appearance in Canis. Although many monachines 

 were polymorphic for this character, this distribution renders both the polarity of this 

 character and its utility for elucidating phocid relationships questionable. 



192) claw morphology in cross-section, I: 0 = semicircular; 1 = triangular (Doutt 1942; 

 Ridgway 1972). 



Together with the following character, Doutt (1942) used the shape of the claws in cross- 

 section to distinguish between Phoca vitulina (state 0) and Histriophoca, Pagophilus, and 

 Pusa hispida (all state 1). Ridgway (1972) applied the same character to distinguish 

 between the genera Phoca spp. (state 0) and Pusa spp. (state 1). However, observations 

 for both this and the following character were hindered by the general paucity of suitable 

 material. Observations could not be made for Mirounga angustirostris, Phoca largha, and 

 Pusa caspica, and the condition for a number of other specimens could only be estimated 

 from the ungual processes of the terminal phalanges. Bearing this in mind, only Mirounga 

 leonina consistently possessed the apomorphic triangular morphology, possibly as a 

 synapomorphy with Mirounga angustirostris (ACCTRAN optimization). A fair number of 

 other pinnipeds were polymorphic for this trait. 



193) claw morphology in cross-section, II: 0 = dorsal ridge or annuli absent; 1 = dorsal 

 ridge or annuli present (Doutt 1942; Ridgway 1972). 



In combination with the previous character, Doutt (1942) used the presence of a dorsal 

 ridge on the claw to distinguish between Pusa hispida (state 1) and Histriophoca, 

 Pagophilus, and Phoca vitulina (all state 0). Again, Ridgway (1972) applied this character 

 to distinguish Phoca spp. (state 0) from Pusa spp. (state 1). Heeding the problems 

 mentioned under the previous character, an apomorphic dorsal ridge was present only in 

 Pusa sibirica and then only polymorphically. 



