EEPOET ON THE SPHENISCID^. 227 



of the entire anatomy of Eudyptes chrysolophns, as compared with that of Eudyptes chry- 

 socome, leads me to the conclusion that in the former we are dealing with one of those 

 extreme and exceptional varieties between which and distinct species it is almost impos- 

 sible to draw a hard and fast line of demarcation. The most reliable test of distinct 

 species is to be found in the fertility of the offspring derived from the union of two ^ 



parents. But at present we, unfortunately, have no information as to whether Eudyptes 

 chrysocome and Eudyptes chrysolophus interbreed with one another so as to produce 

 fertile offspring. The facts, however, related by Murray^ and Moseley^ with regard to- 

 these birds show that while in some localities these two birds have distinct rookeries, at 

 others their nests are intermixed, and thus afford facilities for intermarriage. I am 

 therefore, inclined to regard Eudyptes chrysolophus as the most aberrant variety of 

 Eudyptes chrysocome, and one which is apparently about to cross the boundary line and 

 to become a distinct species. 



Coming now to the consideration of the genus Spheniscus, we find that the generic 

 characteristics of this group are to be found in the form of the skull as a whole, in the 

 great development of the transverse temporal crest, and in the coalescence of its up^jer 

 end with the cerebellar portion of the slcuU, all of which features serve to distinguish 

 the skull of Spheniscus on the one hand from that of Aptenodytes on the other ; in 

 the presence of a deep fossa on the lateral surface of the skull, which fossa is bounded 

 posteriorly by the transverse temporal crest ; in the relatively great breadth of the 

 central as compared with that of the lateral bars of the upper jaw ; in the fact that the 

 central bar completely fills up the interval between the lateral bars ; in the small size of 

 the anterior narial apertures, and in the transference of their posterior extremities 

 to a point altogether in front of the lachrymo-nasal fossae ; ^ in the breadth of the 

 supra-orbital grooves, which are broader than in A^itenodytes, Ijut narrower than in 

 Eudyptes ; in the great size and backward obliquity of the post-orbital processes ; 

 in the relatively slight curvature of the zygomatic arch, which at once distinguishes 

 the skull of Spjlieniscus both from that of Eudyptes and of Aj^tenodytes ; in the form of 

 the rami of the lower jaw bone, which are more slender than in Eudyptes but. less so 

 than in Aptenodtjtes ; in the form of the scapula ; in the presence of a comj^lete 

 coracoidal foramen, in which respect Spheniscus agrees with Eudyptes but differs from 

 A2-)tenodytes ; in the relatively greater length of the metatarsus, which at once distin- 

 guishes Sjjheniscus from both the other genera ; in the more complete separation of the 

 individual metatarsal bones from one another than in either of the other genera ; in the 

 form of the tongue, which is intermediate in form between that of Aptenodytes and 

 of Eudyptes ; in the form of the proventricular gland, which may be either crescentic in 



' Challenger Report, Zoology, part viii. p. 128. 

 - Challenger Report, Zoology, part viii. p. 127. 

 ^ Eudyptes minor forms an exception to this ari-angement. 



