130 THE RIDDLE OF MIGRATION 



eluded ready access to food (deep snow and heavy 

 rime) were both effective in inhibiting departure. 

 From what has already been said about the effects 

 of low temperatures on juncos and other birds in 

 captivity (p. 55) the former case is probably self- 

 explanatory. That birds that have been kept in 

 captivity for two or three months with an unlimited 

 quantity of food supplied them, should promptly 

 return to this known source on discovering nothing 

 available outside, is also comprehensible. All birds 

 liberated have shown themselves to be somewhat 

 bewildered upon first regaining their freedom and 

 for that reason the traps were always closed for 

 several hours after releases, just long enough to 

 force the birds to take time to orient themselves and 

 regain their confidence in the outside world. Weeks 

 of captivity must surely dull a bird's ability to 

 forage for itself. Under the extreme conditions 

 prevailing at the time of some of the releases, 

 inability to find food immediately would no doubt 

 exercise a restraining influence and induce a speedy 

 return to the one known source of supply. 



Under favorable weather conditions over 80 per 

 cent of the individuals of a single release have dis- 

 appeared. They have included birds with gonads 

 in a state of recrudescence or in regression. The 

 controls invariably stayed. The experimentals 



