364 



THE VOYAGE OF H.M.S. CHALLENGER. 



Pal&eudyptes for the reception of the bird of whose skeleton it formed a part. It 

 apparently belonged to a bird closely allied to the genus Eudyptes of the present day, but 

 evidently of much larger size than any living species of that genus. If the nature of 

 the deposit (Eocene) from which this fragment was excavated has been correctly inter- 

 preted, it shows that the family of Spheniscidaa is one of great antiquity, and that it had 

 even at that time deviated so far from the primitive avian stem as to present those modi- 

 fications in structure which have remained unaltered down to the present time. This fact 

 goes far to explain the difficulty which every one must acknowledge in attempting to allot 

 to the Spheniscidse their proper place in any classification of recent birds, a difficulty 

 which will only disappear as the geological record is more fully deciphered, and the in- 

 termediate forms which at one time undoubtedly connected the Penguins with the primi- 

 tive avian stem are brought to light. I have examined the following species of Penguins : — 



Family. 



SPHENISCID.E, 



Genus. 



Spheniscus 



Eudyptes 



Species. 

 ( demersus 

 < mendiculus 

 { minor 



( chrysocome 



( chrysolophus ] 



Variety. 



magellanicus. 



( Eudyptes chrysocome, from Tristan da Cunha. 

 < Eudyptes chrysocome, from the Falkland Islands. 

 ( Eudyptes chrysocome, from Kerguelen Island. 



, , , . [ lonairostns 

 Aptenodytes {._*.. 

 c J i tceniatus 



" Of the various members of the genus Spheniscus enumerated above, it appears to 

 me that Spheniscus demersus and Spheniscus magellanicus ought to be regarded as 

 two varieties of one and the same species, while Spheniscus mendiculus and Spheniscus 

 minor are undoubtedly distinct species. Sphenisms minor is moreover possessed of 

 several cranial characters which approximate it to Eudyptes. 



" Of the so-called species associated together by ornithologists under the genus 

 Eudyptes, I have examined two, Eudyptes chrysocome and Eudyptes chrysolophus. Of 

 these two species, Eudyptes chrysocome presents three varieties, which are met with at the 

 Tristan da Cunha Group, the Falkland Islands, and Kerguelen Island respectively. 

 That Eudyptes chrysolophus ought to be regarded as a species distinct from Eudyptes 

 chrysocome is not doubted by any ornithologist, but an examination of the entire 

 anatomy both of Eudyptes chrysolophus and of Eudyptes chrysocome appears to me rather 

 to lend support to the view that they are simply two well-marked varieties of one and 

 the same species of Eudyptes. The decision of this point must depend on the relative 

 value attached by various ornithologists to difference in size and similarity of anatomical 

 structure as elements in the determination of species as distinguished from variety. 1 



1 " To diseuss the question of the comparative values of external appearance and anatomical structure as elements 

 in the determination of species as distinguished from variety, would extend this abstract beyond reasonable dimensions. 

 I would merely wish to direct the attention of naturalists to the fact that, as it seems to me, sufficient weight has not 

 hitherto been allowed to structure in the determination of species, 



