Hyatt.] 238 [May 17, 



spheniscidje. 



The general affinities of the genera and species of the larger part 

 of this family come to a focus in Splieniscus minor. This, while 

 holding a strictly intermediate position, presents a nearer approach 

 to the lowest members of the genera Pygoscelis, Eudyptes and Ap- 

 tenodytes than any other existing form. Though not very closely 

 related to Aptenodytes, it is quite as near this as any other known 

 species, for, as I shall show in the generic analyses, the resemblance 

 of Pygoscelis papua to Aptenodytes is counterbalanced by character- 

 istics which really ally it closely with Splieniscus minor. Eudyptes 

 chrysolophus, which stands precisely intermediate between Sphenis- 

 us and Eudyptes, affords an opportunity of making a new genus 

 founded upon a combination of the short tail and coloration of the 

 former with the bill and head plumes of the latter. 



If we consider the three modifications which presumably take 

 place upon the basis of the organization of Splieniscus, we find that 

 they cannot be associated in any system but one of radiating straight 

 lines. 



Every series, except Aptenodytes, which has only two very closely 

 allied species, exhibits a decided change in each species. This 

 change in Pygoscelis and Eudyptes carries each species farther from 

 the lowest member, which in each case is more or less like the ances- 

 tral form which was probably closely allied to Splieniscus minor. It 

 is entirely conjectural, whether Aptenodytes came originally from 

 Splieniscus or vice versa, or whether they both sprang from a common 

 ancestor. Such a question can of course only be settled by refer- 

 ence to fossils, and these, with a single Australian tertiary example, 

 are at present wanting. 



The family may be divided into two groups; those with truncated 

 mandible and hooked maxilla, and those with both jaws curved and 

 the spindle-shaped areas of the sides of the mandible bare and 

 highly colored. If it is divided according to the length of the tail 

 feathers, the genus Eudyptes must be dismembered, if, according to 

 the greater or less feathering of the bill, Pygoscelis must be separa- 

 ted into as many groups as it has species, and in either case the 

 genetic connection of Spheniscus with Pygoscelis and Eudyptes is 

 violated. Such a connection may be legitimately inferred in any se- 

 ries of species from the affinities of the adults, when they present in 

 regular succession or gradation a progressive series of changes 01 



