290 THE ANIMALS OP NEW ZEALAND 



The specific characters are chiefly the cliflPerences iu the colour 

 of the phimage or of the eyes, and some of these characters are 

 probably due to that part of sexual selection which has been 

 called preferential selection. The long plumes of the tufted 

 penguin and the yellow bands on the king and emperor penguins 

 may be thus accounted for, while the differences in the species of 

 Sphemscus may possibly be recognition marks. But there are 

 some exceptions. The black throat of four different species of 

 CatarrJiactes and two of Pygoscelis cannot be due to sexual 

 selection, because we cannot suppose that six different species 

 belonging to two different genera all wished to change their white 

 throats to black ones at the same time. This seems to be due 

 simply to reversion ; but it is a ver,y interesting case. It is 

 possible that the red ey&s of the tufted penguin may be a 

 recognition mark to distinguish it from the broAvn-eyed crested 

 penguin. It is possible that the earliest members of the tufted 

 penguins were driven from the rooker^'' on account of their red 

 eyes, and in this way they have been forced to keep together; 

 but it does not seem likely that the white marks on the wings 

 of the big crested penguin and the white-flippered penguin are 

 recognition-marks, for they breed in different localities from 

 the crested penguin and the blue penguin, from which they were 

 derived. 



The effects of isolation are not so well marked in the penguins 

 as in most birds, owing, probably, to their wandering habits, and 

 to the difficulty they must experience in returning after a long 

 voyage to the place from which they started. Still we find that 

 the royal, the crested, the big crested, and the white-flippered 

 penguins, and the four species of Splieniscus, all inliabit separate 

 localities. It might be expected that the tufted penguin, being 

 so widely spread, would show more decided variation than it does. 

 Differences, however, do exist. ]Mr. Watson has shown that the 

 skull of the birds inhabiting the Falkland Islands is larger than 

 those of birds found on Tristan d'Acunha, and that these again 

 are larger than those of birds from Kerguelen Island, while the 

 Falkland Island birds have a smaller bill than that of am' of the 

 others. The birds inhabiting St. Paul and Amsterdam are bluer 



