Birds of Celebes: Alcedinidae. 
295 
the last few years and it has on several occasions received close attention from 
ns. The result is, as already indicated, that we are unable to point to a single 
local race or subspecies ; if local variation exists, it is entirely swamped by the 
individual variation, and the racial development could only be ascertained by long 
investigation on hundreds of specimens and by so determining the average 
individual of this or that spot. Some of the more prominent points of variation 
in the Celebesian area are: the bill-length from 30 mm (young — but not appearing 
so) to 47 mm, its width noticeably different, the culmen sometimes slightly re- 
curved ; the white nuchal spot, sometimes large, sometimes present only on the 
basal part of the feathers and concealed, the tips being blue; the colour of 
the ear-coverts — blackish, or strongly washed with blue or green; the colour 
of the upper surface varying, as already pointed out, from bright China -blue to 
verdigris-green. At one time we thought the Talaut race ran larger in size 
and in size of bill than in Celebes, a larger series has, however, shown that 
this is not the case; at another time we thought that Talaut possessed two races 
— a large and a small one, but later investigations simply go to prove that the 
small one is young and not full-grown. Dr. Sharpe has broken up H. Moris 
into many species and subspecies: were his views to be adopted, not only H. 
Moris^ but also H. armstrongi Sh., H. for stem Bp., H. solomonis Salvad., H. 
humii ^h.., H. megeri Sh. and perhaps others would apparently have to be ad- 
mitted into this work — all from the Celebesian region. H. humii we hold 
simply for young, H. armstrongi, H. solomonis, and H. forsteni for individual 
variations. 
In some of the Moluccas Dr. Sharpe finds that specimens with black ear- 
coverts joining a very broad black nape-band are prevalent, but does not find 
it possible to draw a line of division between them and others from the same 
or neighbouring islands. Most of the Moluccan examples in the Dresden 
Museum have a slight wash of blue on the ear-coverts ; on the contrary those 
of Timorlaut are without colour on this part, being simply black. 
In Malacca, Sumatra, and Borneo there can be no doubt that the typical 
chloris intergrades with H. armstrongi of the Burmese countries; Sharpe records 
both forms from the two latter localities, and there is no reason to believe that 
they separate themselves at a given season by migration. 
Again, in Acheen, Sumatra, Dr. Sharpe records both the typical chloris and 
H. humii of the Malay Peninsula and Siam at the same date. 
Count Salvadori questions the identity of sj)ecimens from the Pelew Islands 
with H. chloris] a specimen from there in the Dresden Museum is remarkably 
green on the head, mantle and scapulars, but not more so than one from Timor- 
laut, where much bluer specimens also occur. 
We unite Halcyon forsteni with this form of H. chloris, believing, like Count 
Salvadori (k6), that it is only a melanotic variety of it. The feathers of the 
under surface of the type of forsteni which we have seen, are muddled with 
