696 Mr. R. C. L. Perkins o?i the 



birds, insects and plants, and have shown how in the birds 

 themselves the causes which developed the weird forms of 

 the peculiar family Drepanididae no longer exist. Nothing 

 but the severest competition for food could ever have 

 produced such birds as Fseudonestor, Heterorrhynclius and 

 Chloridops, the main food of which consists of a single 

 article of diet, to obtain which as a regular diet a very- 

 special and grotesque structure has been acquired in each 

 case. Such forms are the tips of twigs in a tree of descent 

 — and they can give rise to nothing further. It might 

 almost be said they are the tips of twigs which, having 

 produced a terminal blossom, themselves die back. A 

 comparatively easy and successful living is possible for a 

 time, but with a slight change of conditions there only 

 remains extinction. They have no chance to adapt 

 themselves to new conditions. It is, I think, noteworthy 

 how often one finds the ' finest ' things to be very rare in 

 islands, and I think this is clearly due to the fact that 

 what a systematic student calls ' fine,' is usually a form 

 peculiarly specialised in some particular way, and this 

 means a very particular mode of life. Such ' fine ' things 

 are rare, because the conditions suited to their mode of 

 life are few. They are unfortunately the first things to 

 become extinct in Oceanic Islands. 



Nov. 8, 1911. 



I ought to say I have not finished with the ' colour ' 

 question yet, because I have a still moi-e ' general ' part 

 than that which I am sending, dealing with ' species 

 formation,' ' variation,' etc., in a general and more com- 

 prehensive way, considering the whole fauna together, 

 birds, land-shells, insects and plants. 



One who has a wide systematic knowledge of the whole 

 fauna can picture a very different condition of affairs from 

 the present — when the vegetation of the islands formed no 

 true forest, but the islands were covered by a shrubby 

 growth of woody Composites, Lobeliaceae, etc., with few or 

 no trees ; when the birds were of less specialised forms 

 like Himatione and Chlo7'oclrepanis of to-day, with no 

 wonderful developments like Pseudonestw and Heteror- 

 rhynchus, and there were only a few types present, which 

 were numerous in individuals and wandered from shore 

 (where now they are absent) to the mountain tops, and 

 there was a competition for food between individuals, not 



