179 



more strongly supported than that of the ingroup relationships. However, reasonable 

 resolution was still found within the phocids, perhaps due to an increased number of more 

 discriminating, non-traditional and/or multistate characters. Increased resolution is still 

 possible, especially within the Phocini (plus Erignathus), where even more finely 

 discriminating data are required to dissect the evolutionary pattern out of the rapid adaptive 

 radiation of the group (see below). 



Finally, one limitation of cladistic analysis as it applies here, and specifically as it applies 

 to the determination of homologous features, needs to be addressed. The increasing 

 evidence for, and consequent acceptance of pinniped monophyly (e.g., Wyss 1987; Flynn 

 1988; Flynn et al. 1988; Berta 1991; Cozzuol 1992; Novacek 1992; Wyss & Flynn 1993; 

 Vrana et al. 1994; Arnason et al. 1995; Lento et al. 1995; this study), together with the 

 interpretation that the aquatically related features of the group are homologous, and not 

 convergent (Wyss 1988b), needs to be reconciled with available fossil evidence. Currently, 

 the fossil record of the pinnipeds strongly indicates distinct, if not diphyletic, origins of 

 the otarioids and the phocids, whose first appearances in the fossil record are separated 

 by the North American continent and some seven million years (Repenning et al. 1979). 

 However, assuming for the moment that the known fossil evidence does provide an 

 accurate picture of pinniped origins (but see below for alternative possibilities), note that 

 it does not automatically imply the diphyly of the pinnipeds (as has been held in the past). 

 Granting that the modern pinnipeds all originated from the same ancestral lineage, they 

 would still fit under the strict cladistic definition of a monophyletic group - "a group of 

 species that includes an ancestral species and all of its descendants" (Wiley et al. 1991: 

 3) - so long as their common ancestor did not also give rise to some other lineage that 

 we would not classify as a pinniped. 



Of more concern, however, is that the acceptance of the scenario given above by 

 Repenning et al. (1979) would require us to re-interpret some otherwise apparently 

 homologous features. Even if the pinnipeds are truly (strictly) monophyletic, their separate 

 origins from a presumably terrestrial (or, at best, only partially aquatic) ancestor (as would 

 likely be the case given their geographic separation) would mean that most of their 

 aquatically related characters identified here and elsewhere (e.g., Wyss 1988b) as 

 homologies, would have to have evolved in parallel. The common appearance of such 

 features (and possibly of some non-aquatically related ones as well) would then be a 

 consequence of the adaptational limitations imposed by the inheritance of a common, 

 ancestral body plan (i.e., developmental constraints; see Maynard Smith et al. 1985), 

 possibly based on some key innovation (sensu Liem & Osse 1975; see discussion in 

 Russell 1979), combined with the necessity of adapting to an aquatic environment. 

 (Naturally the case for convergent characters due to developmental and functional 

 constraints becomes even more widespread if the Pinnipedia are diphyletic.) An interesting 

 test of this scenario involves the lutrines and the polar bear [Thalarctos maritimus; 

 Corbet & Hill (1991)], species becoming increasingly adapted to an aquatic environment. 

 Given their derivation from a reasonably similar arctoid body plan, and a continued 

 tendency towards an increasingly aquatic existence, these species may become 

 indistinguishable from the pinnipeds in a few million years, despite their clearly 



