AUSTEALIAN DOVE-PRION. 
H. desolatus inacquariensis, subsp. n. ; Macquarie Islands. 
BiU broader than the preceding, and approaching more to the typical form. 
H. desolatus alter y subsp. n. ; Auckland Islands. 
Bill broader still, and agreeing very closely with H. d. hanJcsi, under which 
name it has been commonly known. The type of H. d. hanJcsi seems immature, 
and therefore the present race would have a slightly narrower bill. 
The whole six however are very close and very constant, yet sea-kiUed 
specimens are quite common which do not seem to be referable to any of the 
preceding. Thus birds killed at sea, in lat. 35° S. and from long. 9° W., long. 
OJ E., 6° E., 10° E., and then 40° E., all vary, and seem to represent more 
subspecies. I am convinced that almost each breeding-place of Prions shelters 
a distinct subspecies, and all the blundering has been simply due to confusion 
of well-defined races through study (!) of sea-killed specimens. 
YoinrQ-, Adult. 
Not able to fly. 
H. desolatus alter. s® H. desolatus mattmgleyi. 
The above cut is drawn from specimens, to show the amount of difference 
due to age ; thus the first figure is drawn from a juvenile which has just lost 
its down, but which has not left the nesting-hole ; the second is of a parent- 
bird from the same breeding-colony, collected at the same time, and these 
constitute the whole range of variation in the subspecies. The adults of the 
subspecies are quite constant in their characters, and the difference between 
this subspecies {H. desolatus alter) and the Australian one {H. desolatus 
mattingleyi) is well seen in the third figure. It should be remembered that 
the differences are more easily appreciable in nature than would appear from 
a study of these figures, which are most accurately and carefully drawn. 
231 
