116 THE RIDDLE OF MIGRATION 



to man. Nearly all animal activities are related in 

 one way or another, directly or indirectly, to the 

 influence of the sun. If a species leaves Alberta 

 habitually in the first week of September it may 

 leave (in different years) with the barometer either 

 high or low; in depressingly warm weather or with 

 the smaller lakes frozen over; the earth may be sun- 

 baked or the very gopher holes may be spouting 

 water; the leaves may be golden or they may have 

 fallen weeks before ; the food supply may be abun- 

 dant or it may have failed or be completely covered 

 by a pall of snow. Only one factor of the environ- 

 ment would be certainly constant — the length of 

 day. Its dependability suggests it as the inaugurat- 

 ing principle. 



That many species of migratory birds when in 

 captivity exhibit unusual excitement and restless- 

 ness during their normal migratory period is well 

 known to all aviculturists. Only two factors of the 

 outside environment can effect them indoors — baro- 

 metric pressure and day-length. The former alone 

 cannot be held to elicit the symptoms. Day-length 

 again suggests itself as the probable key. More- 

 over, recent investigations by different workers on 

 the roosting, waking and singing periods of various 

 species have demonstrated remarkable sensitivity 

 to variations in light intensity. 



