470 MR. O. SALVIN ON THE AVIFAUNA 
gable Island three, Bindloe three, and Abingdon four. Thus we see that in Chatham 
Island every gradation in size is represented. 
Then as to coloration. The assumption of the black plumage by the old males seems 
to be a slow process, the fully black feathering being only arrived at in probably the 
third year; and it is probable, from the comparatively scanty numbers of the black 
individuals, that the cock birds breed in the mottled plumage long before the fully 
black dress is assumed. Observations on paired and nesting birds would here be 
invaluable. 
It would seem, then, that with these singular birds the sexual selection displayed 
amongst them is such that it is almost a matter of indifference whether the cock birds 
are mottled or black, and also that the nature of their food and the general conditions 
of life are such that birds with huge bills as large as a Grosbeak’s, as well as those 
with bills no bigger than a Bunting’s, can equally find sustenance, variation as regards 
the length of wing, tail, and tarsi being equally unimportant. The members of this 
genus present a field where natural selection has acted with far less rigidity than is 
usually observable. 
The gap between Geospiza and Cactornis is fairly defined—not but what we see in 
some specimens of the former more elongated bills than in others, showing a tendency 
in Geospiza to develop in the direction of Cactornis. But it would seem that the con- 
necting links are gone; hence our ability to define the differences between these 
genera. 
Camarhynchus belongs to a somewhat different type; and I am inclined to believe 
that it is not to be traced to the same origin as Geospiza and Cactornis, but to a 
common ancestor with, perhaps, Spermophila, the bill of the somewhat abnormal 
ally of that genus (Neorhynchus nasesus) recalling to some extent the peculiarities of 
the bill of Camarhynchus. So much for structural characters. In coloration Cama- 
rhynchus resembles both Geospiza and Cactornis; but the males of none of the species 
are so black as those of the adults of the other genera. 
It seems worthy of notice that, though the different species of Finches are not 
restricted to any one particular island, there appears to be a prevalence of some one 
species in each. Dr. Habel’s collection shows that in Indefatigable Island two dominant 
species occur—Geospiza fortis and G. fuliginosa. In Bindloe Island G. strenua abounds, 
whilst in Abingdon we find G. parvula. It may thus be argued that the tendency of 
each of these islands is to produce the form most prevalent in it; but, on the other 
hand, I am not able to say what attention Dr. Habel devoted to each species. 
