385 



tlie larger species in the soutliorn peninsula of New 

 (ïuinea, and an intelligent Cerainese trader I mei at 

 Aru assured me that, in places he had visited more 

 eastward than the range of the Macassar traders , 

 the same kind was found as at Aru. It is therefore 

 clear that the Paradisea apoda is confined to the soii- 

 thern peninsula of New Guinea and the Aru Islands, 

 while the Paradisea papuana inhabits only the nor- 

 thern peninsula, with one or two of the islands (most 

 ])robably) near its northern extremitv. It is interes- 

 ting to observe, that though the Ké Islands and Go- 

 ram approach nearer to New Guinea than Aru, no 

 species of the Paradise birds are found upon them , — 

 pretty clearly showing that these birds have not migra- 

 ted to the islands bevond New Guinea in which thev 

 are now found. I have, in fact, strong reasons for 

 believiug from geographical , geological, and zoological 

 evidence, that Aru is but an outlving portion of New 

 Guinea, from which it has been separated at no verv 

 distant epoch. 



In examining my series of specimens, I find four 

 such wellmarked states of the male bird , as to lead me 

 to suppose that there moults are required before it ar- 

 rivés at perfection. In the first condition it is of a 

 nearly uniform cofFee-brown colour, darker on the head 

 and paler on the belly, but entirely without markings 

 or variety of colour. The two middle tail-feathers are 

 exactly equal in length to the others , from which they 

 only difFer in having a narrower web. In the next 

 series of specimens , the head has acquired the pale 

 yellow colour, and the throat and foreliead the rich 

 metallic green of the old birds ; the two middle tail- 



