17 



Miss Paula Jenkins, Richard Sabin, Mary Sheridan, Daphne Hills, and Richard Harbord 

 [British Museum (Natural History)]; Dr. Maria E. Rutzmoser and Judy Chupasko (Museum 

 of Comparative Zoology, Harvard); Dr. Harold N. Bryant (Provincial Museum of Alberta); 

 Dr. Wayne Roberts (University of Alberta Museum of Zoology); Warren Fitch (University 

 of Calgary Museum of Zoology); Drs. James G. Mead, Charles W. Potter, Clayton E. Ray, 

 and Dave J. Bohaska (United States National Museum, Washington, D.C.); and Mr. John 

 Rozdilsky (University of Washington Burke Museum). Funding for this project was 

 provided by an NSERC postgraduate A scholarship, a University of Calgary Thesis 

 Research Grant, the Jake Duerksen Memorial Scholarship from the Department of 

 Biological Sciences (OBE), and by an NSERC operating grant (APR). 



METHODS AND MATERIALS 

 Data sources 



Specimens 



For this study, all extant species of phocid seal plus extant representatives of all major 

 caniform lineages, with an emphasis on putative phocid or pinniped sister groups, were 

 examined (see Appendix A). Although the Caribbean monk seal, Monachus tropicalis, is 

 believed to have been extinct since the early 1950s (Kenyon 1977), its persistence well 

 into historical times, its potentially critical role in the resolution of the systematic status 

 of the monk seals (genus Monachus) as a whole, and that fact that it is as well-represented 

 in museum collections as any other extant phocid species caused it to be included in this 

 study. Specimens were examined either while they were on loan from or in their respective 

 institutions. 



A conscious decision was made to exclude fossil taxa from this study. This was largely 

 due to a lack of available specimens, with most being on loan to other institutions at the 

 time of the museum visits. This is unfortunate as the inclusion of fossil specimens can 

 serve to bridge large gaps between highly divergent extant taxa (Gauthier et al. 1988) such 

 as exist here between the pinnipeds and other arctoid carnivores. Selected fossil forms 

 may also reveal much concerning phocid and/or pinniped ancestry. For example, the 

 advocacy of the lutrine-like fossil Potamotherium as an intermediate between the mustelids 

 and phocids is a key argument supporting the hypothesis of a diphyletic Pinnipedia (Ray 

 1976a; Tedford 1976), or at least a mustelid affinity for all pinnipeds (Wolsan 1993). 

 Likewise, the previously regarded otarioid-like fossil desmatophocids (= Allodesmus, 

 Desmatophoca, and Pinnarctidion) are now regarded as the putative phocid sister group 

 within a monophyletic Pinnipedia (Wyss 1987; Berta 1991; Wyss & Flynn 1993; Berta & 

 Wyss 1994). Finally, the exclusion of any taxa, whether extant or fossil, from such a low 

 level analysis may have deleterious effects on the resulting cladogram (Arnold 1981). 

 These points are countered somewhat by the admittedly poor fossil record of pinnipeds 

 (Davies 1958b; Hendey 1972; Hendey & Repenning 1972; Ray 1976a), and the generally 

 high preponderance of missing features (and hence data) in fossil specimens. As well, the 

 inclusion of fossil pinnipeds does not seem to alter the phylogenetic relationships of the 

 pinnipeds as determined from the analysis of extant forms alone (Flynn et al. 1988; Berta & 



