778 Mr. W. P. Pycraft on the 



power of long-sustained flight, wliich is commonly wanting 

 in birds whose lives are passed in circumscribed areas. 



The Avifauna of Central and South America bears out 

 tliese contentions : for the stable population of these 

 regions, as distinct from the utterly difi'erent migratory 

 species, is non-migratory. Such species as have passed 

 from South into Central America have passed as a conse- 

 quence of the gradual exteusion of their range, and not a 

 process of colonization by migrants placing a sudden and a 

 wide space between themselves and their ancestral home. 



These facts have a more important bearing on the subject 

 of the geographical distribution of birds tban is hitherto 

 supposed, for they demonstrate, in no uncertain way, the 

 important part played by temperature, as distinct from 

 climate. For temperature, in regard to some species, may 

 serve as a barrier as effectively as would a desert or a broad 

 belt of water ; while with others it may serve no less 

 efficiently as a bridge. It is temperature, not elevation, 

 which makes a barrier of a mountain- chain : temperature, 

 due to altitude, enables birds of temperate regions to enter 

 the zone of the tropics. By such njcans the Paridae, Ampe- 

 lidse, and many of the Friugillidae — for example, birds of 

 northern latitudes — have been able to penetrate through 

 Mexico and into Central America, travelling by way of 

 the oak and pine forests of the Alpine regions, at an 

 altitude of from 5000 to 10,000 feet and even higher. 



Among the Fringillidse a special interest in this connec- 

 tion attaches to the genus Junco, which has spread from 

 North America southward, along the highlands of Mexico, 

 to Costa Rica, giving rise to new species all along the 

 route. Thus /. dorsalis ranges from New Mexico to 

 Central America, /. ph(£onotas inhabits the mountains of 

 central and south Mexico, /. fidvescens occurs only in the 

 mountains of Chiapas, in south-eastern Mexico, J. alticula is 

 met with only in the mountains of Guatemala, and J. vulcani 

 on the volcano of Irazu in Costa Rica, How these several 

 P-iountains became stocked is by no means apparent, but 

 o^e very important inference is obvious ; these several 



