214 Tran taction*. 



Art. XXIII. — The Geographic Relationships of the Birds of Lord Howe, 

 Norfolk, and the Kermadec Islands. 



By W. R. B. Oliver. 



[Read before the Auckland Institute, 28th November, 1911.] 



Whether the main islands of New Zealand, together with certain out- 

 lying islands, be considered entitled to the full rank of biological 

 " region," or only that of " subregion," will not affect the contention 

 which the evidence assembled in this paper is held to support — namely, 

 that the three groups of islands lying to the north of New Zealand (Lord 

 Howe, Norfolk, and the Kermadecs) should be included within that 

 region. It can be said that conclusions drawn from the study of one 

 class of animals may not be trustworthy, and should be checked by con- 

 clusions deduced from a consideration of other classes. There can be no 

 questioning the truth of this statement, and I would go even further, 

 and say that the claims of a district on the border-line of two biological 

 regions to be included in any one of such regions should be decided upon 

 an examination of the whole of the fauna and flora of the district in 

 question, together with a consideration of its geological history. In most 

 cases this is not practicable, and in this paper I will deal principally 

 with the avifauna of Lord Howe, Norfolk, and the Kermadec Islands, 

 only referring incidentally to other sections of the fauna or to the flora, 

 and endeavour to point out its true relationship to those of the adjacent 

 biological regions. 



In treating of islands, the real test for deciding to which region 

 they should be attached is to consider the evidence in favour of a land 

 connection with a part of the region within the bounds of which it is 

 claimed they should be included. If the probabilities are that the islands 

 have never been actually joined to a land-mass, then the character of the 

 fauna and flora must decide. But here certain life-forms, such as pelagic 

 species, need not be taken into account, except as characterizing a pro- 

 vince or subregion. Again, the presence of such a group as " accidental 

 visitors " may be due merely to climatic conditions, and cannot be held 

 to ally one fauna to another any more than the occasional occurrence of 

 a stray royal spoonbill or pelican in New Zealand allies the avifauna of 

 that country to that of Australia, for species of birds cannot often be 

 dispersed by such accidents. 



From a geographical standpoint, the birds of a district may be ar- 

 ranged according to their life-forms or manner of occurrence into groups. 

 The ecological groups into which I have divided the birds of Lord Howe, 

 Norfolk, and the Kermadec Islands, and whose numerical strength in 

 these islands is shown in the following table, would not necessarilv be 

 the most useful to adopt Avhen treating the avifauna of a large area, but 

 are, I think, the most convenient for the purpose of this paper. 



Ecological Group. Lor J Howe. Norfolk. Kerma lees. 



Resident land-birds — Breeding 

 Sea-birds — Breeding 

 Visitors — 



Sea-birds 



Migrants 



Occasional 



Accidental 



Totals ... ... .. 61 53 43 



