( 20 ) 



BiiuHigka, and Toinia are entirety similar. The proportion of peculiar forms on the 

 Tukang-Besi Islands is, in proportion to the number of species, fairly large, being : 



1. Pistn-himt manadensis Icaliihipae: Kalidupa. 



2. Tanyijnathus mer/alorhynchos viridipennis: Kaliduiia, Binongka, Tomia. 



3. Dicaeum kii'ini: Kali<lupa, Binongka, Tomia. 



4. Cinnyrls infrenata : Wantjee, Kalidupa, Tomia, Binongka. 

 o. Zosteropsjiavissinia: Wantjee, Kalidupa, Tomia, Binongka. 



(i. (h-iolus broderipi oscillwns : Wantjee, Kalidupa, Tomia, Binongka. 



7. Hi/potuenidi<(. kaeh)ti : Kalidupa, Binongka. 



It is quite possible, and even probable, that on account of the thick population 

 and the destruction of forests some interesting local forms have disapi)eared. 



The systematical arrangement of the following list is, for the sake of con- 

 venience for those who wish to compare the lists, that of Meyer & Wiglesworth's 

 liirds of Cdehe-i. Though the system of giving full references to the original 

 description and habitat has met with universal approval, and though I should like to 

 adhere to it generally, at least in the more important instances, I could not carry it 

 through in every case, for want of time. 



1. Astur torquatus wallacii Sharpe. 



'[Fulco torquatus Temminck, PL Gol. 43 (1821 : ex Cuvier : Australia, Timor, 

 .Java, etc. ! — Australia, errore ! I accept Timor as the typical habitat, because the 

 plate and description agree best with the Timor form).] 



Astin- imUiicii .Sharpe, Gat. B. Brit. Miid. i. p. 128 PI. V. (1874: Lombok, 

 Burn. Lombok is the typical locality, the Buru example being a young bird, 

 probably belonging to a totally different bird). 



Tomia Island; (Jc? ad., 20. xii. 1901. "Iris orange, feet ochreo-chromeous, 

 bill black, greyish at base below, about the nostrils and eyelids sulphur-yellow." 

 ? juv., 23. xii. 1902. 



Kalidupa: ? ? ad., 2. 10. i. 1902. (Xos. 4403, 4404, 4608, 4609, 4611.) 



I have called these birds as above not without consideration. Restricting the 

 habitat of typical Astur torquatus to the Timor group of islands, we have the 

 following forms : — 



Astu.r torquatus toi'quatus : Underside white, sharply barred with a more or 

 less jiale rufous, under tail-coverts often pm-e white, breast more or less washed with 

 pale asii-grey. Timor, Savu, Alor.* 



,l«^(,)' torquatus ivallacii: Underside much more rufous, the ground-colour 

 generally much less pure white, more tinged with jjale grey or rufous-grey ; the bars 

 generally wider, often less sharply defined ; the chest much more rufous, less greyish ; 

 tlie barring less distinct on chest and breast. Lesser Sunda islands : Lombok, Flores, 

 Java to Jam pea, Kalao, and Tukang-Besi islands. 



It is true that the Tomia and Kalidupa specimens are lighter grey on bead and 

 cheeks, but I do not venture to separate them without more e\idence. 



Astiir torquatus cruentus: Much like .1. t. tvallacli, but the under wing-coverts 

 much more distinctly and regularly barred. W. Australia and Southern New- 

 Guinea. 



* Everett sent an ivlult pair from Alor, but in my Alor list (Nov. ZooL. 1S9S) this species has 

 iii.iilvertcntly been omitted. On Alor As/itr si/hvstrix, a. species wliicli is quite different and lias no 

 rufous collar above, also occur.-* (Nov. ZooL. IK'.tS. p. -162). 



