BLOODSTAINED COCKATOO. 
but the lack of colour makes the accurate discrimination of subspecies difficult. 
I would rearrange the forms thus : 
Ducorpsius sanguineus sanguineus (Gould) 
Northern Territory (Port Essington) 
and Melville Island. 
As synonyms C. gymnopis Sclater, and C. s. apsleyi Mathews may be noted. 
This form is of medium size, wing measurement dd 270-275 mm. and. 
263-8 in the females. 
Ducorpsius sanguineus distinctus (Mathews) 
Northern Territory (Inland West) and 
North-west Australia. 
As a synonym I quote D. s. subdistinctus (Mathews) or perhaps it would 
be better to call it a secondary subspecies. Under the scale employed by 
authorities in connection with Palsearctic birds, it would rank as a valid 
subspecies, but conservatism seems here better employed. The Northern 
Territory (Daly River and Alligator River inland) specimens measure in the 
wing d 310, $ 280-292 mm. The Parry’s Creek, North-west Australia, birds 
measure in the wing, d 294, $ 273-280 mm. It will be noted that these 
measurements are in agreement with that given by Ramsay for his specimen 
from Derby, North-west Australia. These are all larger than the Melville 
Island series and the West Australian birds. For these latter I must 
propose 
Ducorpsius sanguineus westralensis , subsp. n. 
Type from Murchison, Mid-west Australia. 
The specimens measure d 272-283, $ 2554 — 270 mm. These therefore 
approach the typical series in measurements and cannot be lumped with the 
preceding form. It will be noted that the North-west species are intermediate 
between these and those from the West interior Northern Territory. 
Consequently, were the North-west form not regarded as sub-specific, yet 
it was desirable to indicate its peculiarities it could be referred to as 
Ducorpsius sanguineus distinctus — westralensis. 
The simple trinomial seems preferable and may later be reverted to. 
The series from Normanton, Queensland, are the most perplexing as 
they all measure very small and are constant, thus 3 del’s, 245, 245, 247 ; 3 $$’s 
238, 240, 247 mm. These do not correlate with any other form and though 
I am unwilling it seems necessary to designate these by name. I therefore 
introduce for them the name 
Ducorpsius sanguineus normantoni , subsp. n. 
Specimens from Cloncurry, Queensland, agree with those from the interior 
of New South Wales and South Australia in size, there apparently beingjpo 
211 
