56 STUAKT-SUTHERLAXI). Ronud the Lamp, ^f juTy" 



situation is out of the line of any migratory streams, for the ap- 

 pearances are very erratic. 



The following is a complete list of the birds ohserved arcnind 

 the lighthouse whilst the light has been burning during the period 

 March, 1919, to January, 1922. 



Names according to Sir W. L. Ruller's "Sup[)lement to the 

 Birds of New Zealand," 1^'05, in bracket^, following the 

 R.A.O.U. Checklist name where it differs : — 



1. Prion vittatus. Broad-billed Dove-Petrel. Whiroia. — Sometimes 

 in hundreds, and observed during every month of the year. 



2. Prion banksi. Banks Dove-Petrel. — Rather uncommon. 



3. Prion desolatus. Fairy Dove-Petrel. Titi wainui. — Plentiful, 

 but not quite so common as the Broad-billed. Three only of the four 

 species of Dove-Petrels have been noted, P. ariel not having; so far 

 been taken. 



Professor Mills Loomis lists all the four under one species {Pachyptila 

 vittata) but after having carefully examined numerous specimens, I 

 can hardly agree with this finding. He says: "I have searched the 

 literature in vain for a description of constant characters s^-parating 

 'P. banksi. ' P. desolata, ' and 'P. brcvirostris, ' from 'P. vittata. ' So far as 

 the evidence shows, the variations upon which these supposed species 

 rest are inconstant, and therefore are not of specific significance; an 

 absence of intergradation, not degree of variation, gives specific value 

 to characters. The series before me is a meagre one, but it sti'ength- 

 ens the conviction that the variations are all within the limits of a 

 single species. Ample series from breeding stations alone will de- 

 termine the precise nature of the variations." 



The professor may be right, but all the birds I have examined were 

 readily distinguishable — that is to say, there was no doubt as to 

 which species a specimen belonged, for I find the sizes of the bills of 

 the different species to be constant, and moreover the species differ 

 as to general dimensions. 



Huttcn and Drummond are two authorities who cannot be over- 

 looked, and they give, 



Key to the Species. 



(1). Wing from fle.xuie, 9 to 10 inches. P. rittattis. 

 Wing from flexure, 8 to i) inches. 2 



Wing from flexure, 6 to 7 inches. P. arid. 



(2). Upper mandible convex on the margin. P. banksi. 



Upper mandible straight on the margin. /'. desolatus. 



The professor gives a series of nieasuiemeiits, ami his largest and 

 smallest specimens give the following: — 



Upper mandible. 

 Wing. Tail. Culmen. Depth. Width. Tarsus. 

 Sex .^ 210. 103. 35.9 14.5 20.1 33.2 



Sex I 169. 86. 23.3 7.8 9.5 29. 



