Vor. II, Pt. 11] LOOMIS—A REVIEW OF THE TUBINARES Al 
Under the second method the separations rest largely on the 
shifting sands of individual opinion. It is obvious that the 
subspecies theory has complicated, not simplified, the study of 
birds. Nevertheless, the theory has served a highly useful 
purpose; it has revealed to ornithologists geographic varia- 
tion, which is a variation within the limits of the species. In 
the present paper geographic variation is considered in con- 
nection with the other variations of species, the subspecies 
theory being discarded as a theory that has outlived its use- 
fulness.* 
SYSTEMATIC ACCOUNT 
In the succeeding pages the sequence of the species has 
been considerably changed from that followed by Mr. Salvin 
and Dr. Godman in their monographs, but it still remains 
largely provisional. Departures have also been made from 
the nomenclature employed by these authors, particularly in 
names where the law of priority was not enforced. Further- 
more, numerous names of supposed species recognized by Dr. 
Godman have been consigned to synonymy. In certain doubt- 
ful cases, names have been put in limbo with a query, as the 
surest means of hastening a final decision as to their standing. 
In the analytical keys to the species I have employed struc- 
tural characters wherever available, thus avoiding repetition 
by eliminating color variations. I am reminded by long ex- 
perience as a student of ornithology that artificial keys to 
birds often fail to unlock. Much care has been bestowed on 
the present keys, but in some of the species I have been ham- 
pered in the determination of distinguishing characters by 
lack of specimens; in consequence the keys must be regarded, 
in a measure, as tentative. 
In part VI will be found the synonymy and comments on 
nomenclature of the species that are mentioned only by name 
in the present part. The synonymy relates to the monographs 
of Dr. Coues, Mr. Salvin, and Dr. Godman. 
With the exception of the few diving petrels, all of the 
specimens were measured by Mr. Edward Winslow Gifford. 
The method of measurement is described in detail in the fore- 
word of part VI. 
1Cf. Auk, 1903, v. 20, pp. 294-299. 
