> 
Vot. II, Pr. II] LOOMIS—A REVIEW OF THE TUBINARES 35 
breeding stations can not be relied upon implicitly to furnish 
typical examples of geographic or other variations in color. 
Bill_—There is considerable variation in the size and pro- 
portions of the bill in certain species of Tubinares. In part it 
appears to be caused by age, as in the Black-footed Albatross 
and Short-tailed Albatross. Whether the bill variations in the 
breeding specimens of the Galapagos Albatross, described in 
part VI, are due to age or to individual variation is not clear. 
The color of the bill, too, varies with age in some spécies. Our 
knowledge on this point, however, is very deficient, and in- 
formation regarding it is greatly to be desired, especially in the 
genus Thalassarche. 
SEXUAL AND INDIVIDUAL VARIATION 
The sexes in the Tubinares are alike in the color of their 
plumage, so far as I have ascertained, but differ in certain 
species in size; for example, the dimensions of the females aver- 
age somewhat less than those of the males in the Black-footed 
Albatross, Short-tailed Albatross, and Fulmar. 
In parts V and VI numerous instances of apparent individ- 
ual variation in the color, form, and size of the Tubinares are 
commented upon. The variation in size is so great that many 
species overlap in their dimensions. A specially notable exam- 
ple of variation in form occurs in Cooper’s Shearwater; the 
tarsus is not invariably compressed and sharp in front, an 
alleged generic character thus proving inconstant. A common 
variation in color is the intrusion of white in dark areas and 
of a dark color in white areas. 
In some cases, owing to lack of proper specimens, I am un- 
able to distinguish individual from dichromatic and age varia- 
tion, as in the extension of the color of the sides of the neck 
across the jugulum in numerous specimens of the Dark-rumped 
Petrel and in the variability in the relative length of the first 
primary in the Christmas Island Shearwater. Because of the 
difficulty in determining maturity, it is a question whether age 
variation in birds is not sometimes mistaken for individual 
variation. 
As shown in the subjoined table, supernumerary rectrices 
are perhaps not so exceptional as to justify their classification 
under abnormal instead of individual variation. 
