THE BIRDS OE AUSTRALIA. 
Regarding the small variation in the bills I have found, I now note that 
Salvin, reporting on the Procellariidce of the “ Challenger ” Expedition {Proc. 
Zool. Soc, (Lond.) 1878, p. 738), recorded Prion desolatus from Kerguelen 
Island, and noted that he had eight specimens, five of which were males, and 
three females, and stated : “In these specimens there is no tangible difference 
between the bills of the sexes.” 
I would also note that the early history of this species is not quite as given 
in the Monograph (p. 294), thus : “ This species was discovered during Captain 
Cook’s first voyage, and a specimen captured in lat. 59° was figured by 
Sydney Parkinson. On this drawing, which is only a pencilled outline . . . 
the name of Procellaria turtur was founded by Kuhl. The oldest name, 
however, is that of P. desolata of Gmelin, founded [on a bird] . . . procured 
on Desolation or Kerguelen Island ; it is doubtless the specimen obtained 
by Captain Cook.” Without discussing the identity of P. turtur Kuhl and 
P. desolata Gmelin, I would point out that P. turtur Kuhl was not founded on 
the pencilled outline of Sydney Parkinson ; P. desolata Gmelin has been 
already shown to be the Latin name given to Latham’s Brown-banded 
Petrel, which was in the collection of Sir Joseph Banks ; whether the bird 
itself was there or simply the painting of it made by Ellis, is at present 
uncertain, but in any case Cook did not call at Kerguelen Island until on 
his third voyage. Parkinson’s drawing was made from a specimen killed on 
February 1st, 1769, in lat. 59 S., long, direct south of Tierra del Fuego W., 
while Ellis’s painting was made at Kerguelen Island in 1776, and lat. 50'^ S. 
long. 70° E. 
The wording in the Monograph insinuates that these two were the same 
bird. 
This is one of the species of Heteroprion that induces one to cast doubt 
upon the amount of variation I have admitted in the case of Prion vittatus. 
Good series are available from many localities and such show the amount of 
variation to be inconsiderable, and that the species, as a whole, shows very 
little variation. Thus I would separate six subspecies, as follows : — 
H. desolatus desolatus (Gmelin) ; Kerguelen Island. 
E. desolatus hanksi (Smith) ; Cape seas. 
Bill broader than the preceding ; breeding-place unknown. 
H. desolatus peringueyi^ subsp. n. ; Cape seas ; Pondoland coast. 
BiU narrower than H. d. hanksi Smith and approaching H. d. desolatus, but 
slightly longer than the bill of the typical form ; breeding-place unknown. 
H. desolatus inattingleyi, subsp. n. ; East Australian seas. 
Narrower bill than any other subspecies of H, desolatus. 
230 
