THE BIKDS OF AUSTEAUA. 
Faho longipennis murchisonianus Mathews (East Murchison, West Australia). 
Distribution : South and Mid- west Australia. 
Falco longi'pennis afsleyi Mathews (Melville Island). 
Distribution : Northem Territory, North-west Australia. 
F. 1. murchisonianus I introduced in the Nov. Zool., Vol. XVIII., 
p. 252, 1912, with the diagnosis : “ Differs from F. 1. lunulatus in its much 
paler coloration on the upper and under-surface,” and F. 1. apsleyi in the 
Austral Avian Rec., Vol. I., p. 33, 1912, as follows : “ Differs from 
F. 1. lunulatus in its blue-grey upper-surface and paler under-surface ; and 
from F. 1. murchisonianus in its larger size.” 
Later, White and Mellor described Falco melanotus from Flinders Island, 
Fumeaux Group, as follows : 
Adult female. — Fore-head, crown of the head, cheeks, ear-coverts, and all 
the upper-surface glossy black ; primaries dull black, with exception of tips, 
and marked on the internal webs with oval spots of brown ; two centre tail- 
feathers almost black, with very indistinct markings of bluish-grey : remaining 
tail-feathers black, marked with many spots (in some cases almond-shaped) 
of reddish-brown, all slightly tipped with buff ; throat and head yellowish- 
white, some of the feathers of the latter slightly marked with black. All the 
under-surface a rich rufous colour, many of the feathers marked down the 
centre with black ; feathers on flank and thighs blotched strongly with black, 
upper tail-coverts bluish-black, tipped with bluish-grey ; centre of tail- 
feathers marked with black. Iris brown, bill horn colour, black at tip, feet 
grey, nails black. 
“ Total length 13 inches ; wing 10 inches.” 
No remarks were given indicating how this bird differed, if at all, 
from any other Hawk, and the only comparison I know of is the one here 
published, forwarded to me by Captain S. A. White. The t 5 rpe-locality of 
Swainson’s Falco longipennis is Tasmania, and that of Gould’s F. jrontatus is 
New South Wales. If F. melanotus were separable, F. frontatus should be ; 
further, the name selected by White and Mellor is invalid, but if the bird 
be separable it could be called Falco longipennis samueli. 
Since I separated my forms, the acquisition of more material and more 
prolonged study causes me to discard all subspecies for the present. None 
of the characters utilised have been found constant, and I am compelled 
to refer all the variations seen to individualistic causes. Notwithstanding 
this conclusion, it may be that still longer and more perfect series wiU show 
later the possibility of differentiating subspecies. The large number of 
leracidea I have now got together has given more light than the smaller 
number (though not smaU) I had previously. 
232 
