THE BIHDS OF AUSTEALIA. 
buff than perplexa: its wing measures over 320 mm., but cannot be relied 
upon, as it is a young bird. Both have the upper surface a shade darker 
than the other Owls of this species from Australia. It is therefore possible, 
but — considering the variation of Barn Owls — needs to be confirmed by a 
better series, that the MelviUe Owl differs slightly and will have to be called 
melvillensis. The material on which Mr. Mathews based his new subspecies 
would not have been sufficient for us to separate it.” 
I have termed this extract “ amusing ” on account of the caution 
displayed when dealing with an Island form. I described the form because 
it showed a phase previously unknown from the North Australian mainland. 
They doubt the sexing. Eothschild and Hartert once more return to the 
variation, the subject they knew so well from the continued examination of 
their four specimens. Further, they admit the possibility of the subspecies 
being admitted, but do not suggest with what they would lump it. They 
surely did not propose to unite it with the New South Wales form. If so, 
their distribution, when plotted out, would look somewhat peculiar. Again, 
they state that they would not have separated it. I have known them do 
less wise work. Hartert separated a Cape York finch as slightly paler on one 
skin, to instance an Australian example. 
They then sum up : “ All of which a study of these forms admits of being 
done at present is to assume : 
“ (A) T. n. novcehollandice, a larger and (underneath) darker form from 
Victoria and New South Wales (synonym riordani) to be confirmed by a 
series ! 
“ (B) T. novcEhollandicB perplexa from the rest of Australia (underneath 
never brownish ? and averaging smaller, synonyms kimherli, macTcayi, 
whitei). 
“ (C) T. novcBhollandicB melvillensis from Melville Island — a doubtful 
form which rests on one adult — apparently wrongly sexed — and one juvenile 
specimen, and which requires further confirmation ! ” 
The reconsideration I have given to this matter with more material and 
having the preceding criticisms always in view, giving them their due worth, 
has enabled me to present a new view of the facts (not fancies induced by 
examination of specimens of other species). It is somewhat peculiar that 
such new material should have confirmed to a great extent the imperfect 
results submitted by me in April, 1912, and which received such scathing 
condemnation at the pens of Rothschild and Hartert. 
The first matter to be discussed is the one neglected by Rothschild 
and Hartert in their criticism and which is really the crux of the matter, 
viz. : “ Are there not two different species of Owls confused under 
390 
