ON THE GENUS PITTA. 327 



Concerning Pitta cceruleitorques Dr. Meyer remarks : — 



" This species inhabits the largest island of the Sangi group ; and it is an 

 interesting one, because it is more closly allied to Pitta erythrogaster from the 

 Philippines, in the north, than to the two species from islands immediately to 

 the south (viz. Pitta palliceps from Siao, and Pitta celebensis from Celebes), 

 and therefore presents a good example of variation of species in consequence 

 of separated insular habitat. Good examples for the same point of view 

 are, amongst others. Pitta cyanonota from Ternate and Pitta rufiventris 

 from Halmahera, in their relation to the species from the neighbouring 

 islands (New Guinea, Celebes, the Sangi, and Philippine Islands). 



" Pitta palliceps on Siao is as slightly different from Pitta celebensis on 

 Celebes as Pitta cceruleitorques on Sangi is from Pitta erythrogastra on the 

 Philippines. That insular separation is a reason for such variations is not 

 to be doubted, in my opinion ; nevertheless we cannot examine this subject 

 more closely at present. Pitta celebensis^ for instance, does not show the 

 least difference over the whole extent of the island of Celebes. My 

 specimens from the neighbourhood of Makassar resemble exactly those from 

 Manado— nearly the north and south points of this long, island,— whereas, 

 when we cross over to the closely neighbouring island of Siao, immediately 

 a variation appears, in Pitta palliceps, 



" Whether this variation has specific value or not is of no importance at 

 all upon this part of the question. Authors do not agree, and never will 

 agree — at least, for some time to come : one says it has ; the other says it 

 has not. But all see that a difference exists ; and this is of value, notwith- 

 standing its smallness, because it is a constant one. 



" That insular separation does not always produce constant differences 

 is known ; and I only mention it here for this reason, that it refers to a closely 

 allied species. Pitta macMotii, I got a large series of specimens on New^ 

 Guinea in different places, viz. at Dore, Andei, Passim, Inwiorage, Rubi, and 

 on the Elephant Mountains, and some on the island of Jobi, in the north of 

 Geelvink Bay. I first thought that the Jobi specimens differed by brighter 



