Species 0/ Pachycephala. 97 



Dr. Gadow lias described this species as having a tail aS) 

 long or longer than the wing. This is a clerical error. The 

 dimensions of the tail given as 3'85, 3'9, 3'8 inches should 

 read 2-85, 2*9, 2*8 inches. 



Male. No yellow supraloral patch ; rectrices nearly black ; 

 underparts orange-yellow ; black pectoral collar very broad 

 and quite perfect. 



In the synonymy of this species {op. cit. p. 201) Dr. Gadow 

 includes Pachycephala vitiensis, Finsch and Hartlaub, Fauna 

 Centralpolynesiens^ p. 74 (part, 5 tantum). It can scarcely 

 be a matter of doubt that the female referred to belongs to 

 P. neglecta, since it comes from Ovalau, and the male to 

 P. viiiensis, since the description is a translation of Gray^s 

 original diagnosis of that species. Under no circumstances 

 can it be said that any part of this reference relates to 

 P . torquata, which is only known from Taviuni. 



Pachycephala kandavensis. 



The Kandavu Bush Shrike was described from examples 

 obtained on Kandavu (Ramsay, Proc. Linn. Soc. New South 

 Walesj 1877, p. 65), and the types are in the Macleay Col- 

 lection in Sydney. 



Dr. Gadow does not regard the Bush Shrike from Kandavu 

 as distinguishable from that found on the island of Ngau, nor 

 is he able to reconcile Mr. Ramsay's description of Pachy- 

 cephala kandavensis with the examples from the island of 

 Kandavu in the British Museum, of which there were five 

 when the Catalogue was printed, now increased to seven by 

 the addition of two obtained during the cruise of the ' Chal- 

 lenger.^ I do not think that there can be any doubt that 

 Dr. Gadow is mistaken on both these points. Species which 

 are isolated on islands cannot be judged from quite the same 

 standpoint as those which occupy large areas on great con- 

 tinents. It does sometimes happen that one island-race varies 

 to such a degree that the variations overlap those of another 

 island-race, whilst the extreme forms of the two races may be 

 very distinct. In such a case the two races intergrade, though 

 they do not interbreed, and can only be regarded as subspe- 



SER. VI. VOL. III. H 



