1874.] MR. R. B. SHARPE ON CERTAIN ACCIPITRES. 583 



the darker and more slaty blue of the head and tail in ft rupicola 

 compared with the paler and more greyish blue of those parts in 

 ft tinnunculus. Below, the differences are very striking, the breast in 

 ft rupicola being of a deep chestnut-fawn-colour with black spots, 

 while in ft tinnunculus the under surface is fulvous fawn-colour 

 with a slight vinous tinge. But the best character which separates 

 the two species exists in the sides of the face of ft rupicola being 

 entirely blue like the crown, with no fulvescent ear-coverts. The 

 females of both species differ from the males in having rufous instead 

 of blue heads ; and whereas the hen of ft rupicola always has a blue 

 tail banded with black, the ordinary plumage of the female ft tin- 

 nunculus is a rufous tail banded with black, excepting when a bluish 

 shade is apparent on the tail of the dark form which inhabits the 

 localities mentioned by me above. The specimen obtained by Mr. 

 Wharton endangers the validity of the blue tail as a good character ; 

 but, for all this, the bird is a regular Common Kestrel with the pale 

 under surface, and with greyish cheeks and ear-coverts, the sides 

 of the neck being also light fulvous with blackish streaks ; whereas 

 in the female ft rupicola these parts are all deep rufous, with a 

 very slight greyish shade on the upper ear-coverts. 



While on the subject of Kestrels, I would draw attention to the 

 fact that the Moluccan species includes two forms, as has been 

 already pointed out by Professor Schlegel in his ' Valkvogels van 

 Nederlandsch Indie ' (p. 48). He says there, " the Kestrel of 

 the Moluccas has been observed by our travellers in Java, Borneo, 

 Celebes, in the Halmahe'ra group, Ceram, Amboina, Timor, and 

 Flores. With the exception of Java, we possess specimens to the 

 number of twenty-five collected in the localities above cited. On 

 comparing them inter se, one sees that in all those killed in the Hal- 

 mahera group the brownish-red tinge is darker and extends over all 

 the sides of the head, whilst the throat, as well as the forehead, 

 borders more plainly on rufous. These individuals have been killed 

 in different parts of Halmahera and in the islands of Morotai, Ter- 

 nate, March, Tidore, and Batchian. We only possess three exam- 

 ples from the Ceram group ; of which two, killed in Ceram and 

 Amboina, are indistinguishable from those of the Halmahera group, 

 whilst the third one, by means of its clear tints, approaches more 

 those from the other parts of the archipelago. Those from Borneo 

 and Celebes have the colours less pronounced, and that of the region 

 of the ear passes more or less perceptibly to whitish. It is the same 

 in our example from Timor and in that from Flores ; the latter is 

 remarkable for the restricted number and minuteness of the dark 

 spots on the back and wings." An excellent plate with three figures 

 (pi. 1. figs. 3-5) illustrates Prof. Schlegel's remarks; and when I 

 was in Leiden last year I saw the original specimens. Fig. 3 re- 

 presents a Batchian specimen, fig. 4 the Flores bird referred to 

 above, and fig. 5 the light-cheeked Ceram bird. 



When examining the Leiden series of ft moluccensis I found that 

 there was certainly a recognizable difference, as Professor Schlegel 

 has jiointed out, in birds from various localities ; but I could not sepa- 



