THE BIRDS OF AUSTRALIA. 
I do not propose to criticise this classification as regards extra- Australian 
groups, but had intended to suggest some emendations in connection with 
Australian Parrots. After sketching out my own groupings, I referred to a 
paper entitled, “ On Characteristic Points in the Cranial Osteology of the 
Parrots,” by D’Arcy Thompson ( Proc . Zool. Soc. (Lond.) 1899, pp. 9-46). 
That investigator wrote : “To discover anatomical characters such as might 
yield or help to yield a natural classification of the Parrots has been the desire 
of many ornithologists, but the search has availed little. Garrod’s abundant 
work has told us many facts in regard to the presence or absence of 
an ambiens, of an oilgland, of one carotid or two, and other varying characters 
in a multitude of species ; but when we come to put these data together the 
result is unsatisfactory, and one is left with the impression that the several 
series of facts are incoordinate and cannot be linked together in a single 
system. When we find, for instance, that the collation of these facts places 
in a single group Ara, Psittacus, Poeocephalus , and Nestor, and in another 
Stringops, Melopsittacus, and Agapornis, one is tempted to think that the 
only thing proved is that the data are invalid or antagonistic — in other words, 
that the several structures had really followed diverse or parallel or 
convergent lines of modification and evolution. While such internal structures 
seem to me to lead to confusion by indiscriminate variability, the characters 
of the skeleton are generally deemed too monotonously alike to present 
features of significance. Even in Stringops, the osteological peculiarites of 
which are greater than those of any other form (except perhaps Nestor), they 
are yet not conspicuous enough to have prevented certain recent writers 
from remarking that the divergence of Stringops from the other Parrots is 
not so great as it had been supposed to be.” 
Thompson then gave the results of the critical examination of skulls, 
and the paper appears to furnish a concrete example of the value of 
specialisation. The close study given to these skulls has shown features 
overlooked by casual osteological examination, and by means of such 
characters Thompson has indicated errors in the grouping proposed by 
Salvadori. 
Thompson wrote : “ The Cockatoos possess certain cranial characters 
in common and their skulls are easily to be recognised, but there are many 
variations within the family and even within the restricted genus Cacatua. 
. . . It is clear that the skull of Calopsittacus, though at first sight very 
similar to, is different in several respects from, the true Cacatuine. It is 
possible that these differences involve resemblances to the Plutycercini, and 
this question will be further discussed below.” 
I had previously dissociated the species classed by Salvadori in Cacatua 
6 
