THE BIRDS OE AUSTRALIA. 
Atlantic bird be compared witb Sparrman’s figure, it will be found to agree 
in these details just exactly where Pacific birds do not. 
Instead therefore of Saunders’s comment “ Sparrman’s figure is very bad,” 
it will be found that Sparrman’s figure is quite accurate when an Atlantic 
bird is placed beside it. 
Examination of young birds from the Pacific show the nostril to be 
placed well forward ; they are mottled above with dark, reddish-brown tips 
to the feathers and have the primary-, etc., shafts dark. Juveniles from the 
Atlantic have the nostril placed as far forward as in the preceding, and though 
about the same age have no mottling apparent, but the primary-shafts are 
dark like those of the Pacific young ones. I have not seen juveniles of G. inicro- 
rJiyncha Saunders, but in the position of the nostril in the adult they approach 
more closely to the Atlantic forms of G. alba than to Pacific forms, as they 
also do in the general shape of the bill. 
On account of the absence of coloration in the plumage, the differen- 
tiation of subspecies is not easy. 
Firstly, having demonstrated that G. alba must be used for the Atlantic 
birds, these as a whole are easily separated, when adult, from Indian and 
Pacific Ocean forms, by the white shafts to the primary-quills and tail- 
feathers. The general shape of the bill is also different, and immature fully- 
fledged specimens may be distinguished by means of this character. I have 
studied specimens from Ascension Island, Fernando Noronha and South 
Trinidad, and it may be that more than one subspecies should be recognised 
in the Atlantic. In the meantime I would use 
Gygis alba alba (Sparrman),; Ascension Island (breeding), Fernando 
Noronha (breeding), and South Trinidad Island (breeding). 
At present Gygis crawfordi Nicoll must be cited as a synonym. 
To pass right to the Pacific Ocean, many names are available, though 
the limits of the subspecies are not yet fixed. None of these can be 
confused with G. a. alba as noted above, so they can be contrasted among 
themselves. 
The type-locality of Gygis alba Candida (Gmelin) is Christmas Island : 
through confusing Sterna Candida Gmelin with Sterna Candida Forster, I gave 
the type-locality of the latter in my “ Reference List ” {Nov. Zool., Vol. 
XVIII., p. 211, 1912), where I used this name for all the Pacific birds as 
separable from the Atlantic birds. 
In the Avifauna Laysan, p. 36, Rothschild noted that the North Pacific 
birds seemed separable as a whole from the South Pacific, and this was 
endorsed by Hartert in the extract quoted {ante^ p. 440). If no further 
442 
