534 GEOGRAPHICAL VARIATION IN NORTH AMERICAN BIRDS. 
we are perhaps warranted in concluding that the ancestral typeof 
lepidopterous larvæ was provided with two pairs of thoracic spira- 
cles. ie 
GEOGRAPHICAL VARIATION IN NORTH 
AMERICAN BIRDS.* 
BY J. A. ALLEN. 
ae 
_ Propasty the birds of no equal area of the earth’s surface a 
better known than those of North America north of Mexico, or of 4 
the whole continent southward even to the Isthmus of Panama. : 
No museums in the world, probably, possess so large suites of : 
specimens of single species as there are of North American birds 
in the Museum of the Smithsonian Institution and in the Museum 
of Comparative Zoology, nor from so many localities. many 
f specimens 
collected at frequent intervals throughout their known range. 
Those contained in the Smithsonian Institution have : 
carefully elaborated by Prof. Baird and others, whose reports 
them have justly acquired a world-wide reputation for their Wer 
oughness and accuracy. Those in the Museum of Compa 
of material? What are the allowable inferences, and whats 
principles have been apparently established? To answer thes 
tions as briefly as may be is the object of the present re: 
premising, however, that the formerly current opinions Te 
the rank of a certain class of forms heretofore generally T° 
as specific have been radically modified. Intergradation 1 
frequently traced between widely different forms, à 8™ | 
cence in scores of instances having been positively €s 
and rendered extremely probable in a large number of oth 
In North America geographical variation exhi 
phases : — (1) a differentiation with differences 
elevation, and (2) differentiation with differences Ha 
which, for convenience, may be termed, respectively _ 
* From the Proc. Bost. Soc. Nat. Hist., vol. XV, P» 212. 
