122 THE RIDDLE OF MIGRATION 



was over, the gonads of the experimentals had 

 attained the spring maximum. They increased in 

 volume particularly rapidly towards the end (as is 

 also the case with wild juncos in the spring) the 

 minimum temperatures during the 7 days of most 

 rapid growth being respectively 21, 31, 44, 37, 36, 

 20 and 4 degrees helow zero Fahrenheit. A more 

 convincing refutation of the increasing temperature 

 conception could hardly be wished for. 



The controls, in the meantime, were receiving no 

 artificial light. They were subjected to the nor- 

 mally decreasing days of an Alberta autumn. In 

 all other respects, unlimited food supply, tempera- 

 tures, etc., their environment was identical with 

 that of the experimentals. Samples were taken at 

 regular intervals from both aviaries and the gonads 

 preserved, measured and sectioned for the micro- 

 scope. Gonads of the controls, already quite small 

 when the birds were trapped at the end of Sep- 

 tember, continued to diminish till they reached the 

 winter minimum in November where they remained 

 till the end of February. 



Various modifications of this experiment were 

 conducted. The remarkable dependence of the 

 state of development of the gonads of the junco on 

 lighting conditions was strikingly illustrated by 

 some of the birds that had been brought to spring 



