THE BIRDS OF AUSTRALIA. 
eight males and six females, give wing measurements with a total variation 
of nine millimetres, viz. 277-286 mm., and no differences in the sexes whatever 
were observable to me. They differ from the typical form in being darker 
above and below, and in having the “ bill rose.” 
Godman notes this as being remarkable, but I simply think the colour, 
which is usually called by British collectors “flesh,” is meant. Certainly the 
French collector M. Thibault states the feet also are rose, and these we 
know in many Pufflnus are of the colour described variously as “ yellowish 
flesh-colour” (from skins), “flesh-colour,” “ fleshy- white,” or “fleshy-grey.” 
I propose for this Seychelles race the name 
Puffinus paciflcus hamiltoni, subsp. n. 
Salvin noted {Ihis, 1888, p. 352) : “ The Mascarene birds have perhaps a 
rather stouter bill, the colour of which in the skin is more of a fleshy-yellow.” 
This statement is made apparently on two specimens in the British Museum 
from Fouquet, Rodriguez, which have stout, light-coloured bills, and perhaps 
indicate another race. 
A specimen from the Society Islands has a longer bill than usual, almost 
as long as in P. c. iredali, but not as stout as in that form, and of dark colour. 
Finsch and Hartlaub {Beitr. Fauna Central Polyn., p. 245, 1867) give the 
following characters for a bird from McKean’s Islands, Phoenix group, which 
they called P. sphenurus Gould : “ Rostro rubente carneo, apice et culmine 
obscuri oribus ; pedibus flavescente-carneis. Long. c. 15'', rostr. 15 J" ; al 10 J'" ; 
caud. 4J" ; tars. 20".” 
A most perplexing factor in the study of this species is the forms grouped 
in the Monograph of the Petrels under Pufflnus cuneatus Salvin. This “ species ” 
has been ascertained to have a white-breasted phase, and Godman concludes 
that if “ P. chlororhynchus ” has also such a phase the two must be merged. 
The dark birds breeding on San Benedicto Island, off the coast of California 
certainly seem referable to Pufflnus paciflcus with subspecific rank, and there 
undoubtedly light-breasted birds are met with, though in the minority. Birds 
very similar (at present accepted as identical) to these light-breasted birds are 
met with breeding in the Marshall Group (whence P. cuneatus was described), 
in the Vulcan group, the Bonins, and the Sandwich Islands (Laysan group), 
where however the dark form only occurs as a very uncommon variation. 
The question automatically suggests itself : Can these light-coloured 
forms be due to climatic causes, or does interbreeding account for the 
San Benedicto birds ? 
Since the preceding was written, further study of the fine collection of 
these birds in the Rothschild Museum, Tring, has convinced me of the propriety 
of differentiating the forms lumped under P. cuneatus. 
82 
