ANNUAL MIGRATIONS 121 



going out nightly at 11 till January 9 (1928) when 

 the experiment terminated. 



The experimental were thus artificially provided 

 with approximately spring conditions as far as 

 illumination was concerned. In the matter of heat- 

 ing they received what the fates might contribute. 

 The aviaries were unheated, removed from all extra- 

 neous sources of warmth and wide open on two sides 

 to the weather. Eleven days of November regis- 

 tered temperatures below zero (with a minimum of 

 -23°F.); December, the severest recorded in 32 

 years, produced 23 days with the thermometer 

 below zero (minimum — 44°F.). 



Keeping the birds in unheated cages in the open 

 served two purposes. It demonstrated their re- 

 sistance to extreme cold and disproved the generally 

 accepted view that the spring recrudescence of the 

 gonads is attributable to rising temperatures. This 

 opinion is universially adopted in popular literature 

 and has been expressed by several scientific investi- 

 gators. In the hypothesis outlined above it was 

 assumed that light, not temperature, controlled the 

 changes. By exposing the birds to the extreme tem- 

 peratures of an Alberta winter it was obviously put- 

 ting this viewpoint to the severest possible test. In 

 this regard the results of the 1927 experiment were 

 very striking. By January 9, when the experiment 



