AUSTRALIAN GOSHAWK. 
in which case we should distinguish a darker subspecies in Victoria.” He 
then added Astur cruentus Gould” for a bird from West Kimberley 
explaining : “ I have named this specimen cruentus, as Sharpe, Robinson and 
Australian authorities have named similar specimens cruentus. They are 
underneath light rusty-cinnamon with narrow white bars. The rufous band 
on the hind-neck is wide and uninterrupted. The wing of this bird 
(marked (J) measures 270 mm., that of a female from Point Cloates 310 mm. 
It is possible that these birds do not belong to A. cruentus, though it is 
by no means certain. Mr. Rothschild thinks they must be the same, since 
the variation of A. fasciatus ( = approximans) and that of torquatus from 
Timor is considerable.” 
As recently as December 21, 1912, Rothschild and Hartert had recorded 
{Nov. Zool., Vol. XlX.j, p. 190) a bird from the Kumusi River, North-eastern 
British New Guinea as ? Astur cruentus with a note : “ Besides this specimen 
we have another male from Nicura, in British New Guinea, which is similar. 
Unfortunately we have no Australian specimens for comparison, but Gould’s 
figure shows a bird with a more whitish under-side and wider bars. Astur 
cruentus ? may be only a sub-species of Astur torquatusj^ 
Compare this with the preceding account written seven years previously 
and with Hartert’s statement of 1898 above quoted, and it must be conceded 
that, whether the birds be considered puzzling or not, the writers who have 
dealt with them have manufactured puzzles enough. Again, note there is 
no mention of Ramsay’s A. sharpei, which was described from British 
New Guinea. 
In the Nov. Zool., Vol. X., p. 20, 1903, Hartert discussed the races of 
Astur torquatus, admitting : 
Astur torquatus torquatus. Timor, Savu, Alor. 
wallacii. Lesser Sunda Islands. 
cruentus. W. Australia and Southern New Guinea. 
sumhaensis. Sumba. 
The recent nomination of A. fasciatus polycryptus would imply the 
rejection of torquatus as the specific name, but would also indicate the extreme 
closeness of torquatus and fasciatus. The preceding accounts will show that 
“ polycryptus was referred to torquatus by Hartert above^ also by Salvador!, 
and tentatively accepted by North as also quoted. Typical torquatus differs 
slightly from typical fasciatus, but inasmuch as cruentus has been referred to 
both, it will be admitted that the differences are not particularly striking. 
The immature of torquatus, however;, differs rather noticeably from the 
immature of fasciatus, and for that reason alone I at present keep them 
separate. When series of plumage changes are studied it may be that only 
67 
