86 



(e.g., character #24 is a derivation of character #23). Typically, this amounted to detailing 

 the various manifestations of a "present" feature that was initially offered in present/absent 

 form only. Finally, although some character states were not represented at the species level 

 (e.g., state 2 for character #51), they did characterize individual specimens and so were 

 retained. 



Unfortunately, in attempting to assess the relative size of several characters, an admittedly 

 arbitrary scheme often had to be employed (e.g., small, medium, large, or something 

 equivalent). In such cases, size was determined in relation to the size of the surrounding 

 bone (e.g., the skull as a whole for cranial characters, or the bone in question for post- 

 cranial material), bearing the range of variation observed over the phocids in general in 

 mind. 



The use of tendencies for traits (e.g., tendency towards single-rooted postcanines; 

 characters #143 and 144) has occasionally been criticized in cladistic analysis on the basis 

 that characters should be more discretely discernible. However, characters examining 

 tendencies were still employed here as they may be the only way to summarize highly 

 variable morphologies or taxa that show differences between the two optimization criteria 

 used here (accelerated transformation, ACCTRAN; and delayed transformation, 

 DELTRAN). 



Unless otherwise noted, anatomical terminology is standardized according to Miller (1962) 

 and/or Davis (1964). Synonyms are given wherever possible. Finally, no polarity is implied 

 by the sequence of character state coding (e.g., zero does not necessarily indicate the 

 plesiomorphic state). In all cases, the polarity of each included character is explicitly stated 

 in the text detailing its phylogenetic reconstruction. 



Snout (21 characters) 



Clearly the most important feature of the snout deals with the nasal processes of the 

 premaxilla. The morphology of these processes and their relationships to neighbouring 

 elements contain a good deal of useful, and largely untapped, systematic information, as 

 is generally true for the remaining characters as well. 



*1) relative position of external nares on snout: 0 = relatively dorsal ("high"); 1 = relatively 

 ventral ("low") (Ridgway 1972). 



The dorsal situation of the external nares on the snout has been proposed as a 

 synapomorphy of the Monachinae (Ridgway 1972). However, the examination of both 

 study skins and photographs revealed no appreciable distinction in the placement of the 

 external nares between monachines and phocines. Due to this lack of variation, the 

 character was deleted from the analysis. 



*2) relative orientation of external nares on snout: 0 = vertical; 1 = horizontal (Ridgway 

 1972). 



Ridgway (1972) described states 0 and 1 as being synapomorphic for the subfamilies 

 Cystophorinae and Monachinae respectively. However, with the accepted paraphyly of the 

 Cystophorinae (see King 1966), vertically oriented external nares would be better classified 

 as another convergent feature between Cystophora and Mirounga. Or, as tentatively 

 suggested herein, they might be another feature retained from the primitive phocid ancestor 



