100 Transactions. — Zoology. 



Then, again, the Kermadec Islands possess a Green Parra- 

 keet which Mr. Rothschild declares to be distinct ; and, 

 although many naturalists will insist that this and the other 

 island forms are, for the most part, local varieties of the well- 

 known Platycercus nova-zealandice, their very existence as 

 such is the best evidence of the constant operation of the law 

 of development by variation and the survival of the fittest. 



Even the sea-birds, whose range is practically unrestricted, 

 furnish additional and, indeed, very important evidence. Mr. 

 Rothschild, with the aid of Mr. Salvin, our great authority on 

 the Petrel family, has lately been investigating the Albatroses 

 of the Southern Hemisphere. Talking over the result with 

 me, he said, " Why, every group of islands seems to have its 

 own species of Albatros ! " And, in a sense, this is true. 

 Here we have birds enjoying the freedom of the wide ocean — 

 commingling daily on their great hunting-fields on the face of 

 the deep ; and then, on the approach of the reproductive 

 season, separating themselves, according to their species, and 

 repairing to their own island-nurseries to breed. As far as 

 our information at present goes, Campbell Island is held ex- 

 clusively by my Diomedea regia, the noblest member of the 

 group. The Auckland Islands are occupied by thousands of 

 Diomedea exulans, with the exception of a small colony of the 

 former breeding in a remote corner of the main island, and at 

 a somewhat earlier season — according to Captain Fairchild's 

 observations, four or five weeks earlier. On the Snares Dio- 

 medea salvini (hitherto known here as Diomedea canta) reigns 

 supreme. The Albatros breeding on the Sisters, some outlying 

 islands in the Chatham group, on which the Maoris are said 

 to have collected as many as a thousand young birds in one 

 season, is probably Diomedea melanophrys, which is plentiful 

 in that latitude ; but I have not yet been able to obtain any 

 specimens from that locality for identification. The breeding- 

 place of Diomedea bulleri, Rothschild (hitherto confounded with 

 Diomedea, culminata), I have not yet discovered, all the speci- 

 mens of that form known to us (some twenty in number) 

 having been captured off the Otago coast. 



Many of the smaller species of Petrel, it may be observed, 

 confine themselves to particular islands : for example, Pnffmus 

 carncipes is the commonest of birds on the Island of Karewa, 

 in the Bay of Plenty, but, so far as I am aware, has never 

 been found breeding on any other island off our coast. 



As with the Petrels, so in a limited sense with the Shags 

 and Penguins, many of the species of both having their par- 

 ticular island group, which they resort to, for breeding pur- 

 poses, to the exclusion of all others. 



Finally I may refer to the Snipes, the local distribution 

 of which is very remarkable indeed. My Gallinago imsilla r 



