[38] 



ster says of the North Carolina mountains " The boundaries be- 

 tween the different faunal areas are sharply marked in places, in 

 others only faintly so, one set of birds often overlapping^ and 

 mingling with another throughout a belt of neutral ground. 



The line of separation between the Canadian and Alleghanian 

 divisions so far as I have observed is better defined than that be- 

 tween the Alleghanian and Carolinian. The Canadian fauna is 

 also purer than either of the other two.* * * It is evident that 

 altitude plays only a secondary part, various local conditions — 

 such as the presence or absence of certain trees or shrubs — 

 having clearly more influence. Dcndroica virens^ for example, 

 was seen only where spruces and balsams predominated over 

 other trees and D. ccerulescens invariably in or near extensive 

 tracts of rhododendrons. 



For the rest it will not do to draw the lines too closely in a 

 region where a bird can easily fly in a few minutes from a valley 

 filled with southern trees and shrubs to a mountain summit clothed 

 with northern Coniferae. Indeed it is chiefly surprising that faunal 

 lines can be drawn at all under such conditions." These obser- 

 vations of Mr. Brewster are very just and, as it seems to me, 

 apply very well to the Virginia mountains. 



The question as to the effect of altitude upon bird distribution 

 is a complex one for the climate and vegetation necessarily vary 

 with the elevation and it is perhaps the combined effect of the 

 general conditions rather than any single factor taken by itself, 

 which renders a locality attractive to birds. No one who has 

 visited the respective regions can have failed to observe the 

 resemblance between the climate, fauna and flora of the summits 

 of the higher southern mountains and the localities of average 

 elevation in northern New York and New England. The most 

 sharply defined faunal line in the Virginias as In North Carolina, 

 appears to be the lower limit of the Canadian fauna, its charac- 

 teristic birds not being ordinarily found below a considerable 

 altitude. That this rule however is subject to exceptions is 

 shown by the discovery of Mr. L. M. Loomis that D. virens is 

 common at a low elevation in the mountains of South Carolina. | 



*An ornithological reconnolsance in Western North Carolina. Bj' William 

 Brewster. The Auk, Vol. Ill, No. 1. Jan. 1886. p. 100. 



tThe Auk, Vol. VII — 1890. pp. 33 and 128. This species, however, is not al- 

 together typically Canadian. 



