xvii. 5 MIGRATION 495 



experience they can do more than just steer a course — they can navi- 

 gate, that is fix position, calculate the course to steer, and follow it. 



Powers of 'homing' are remarkable in birds, altogether apart from 

 migration. It is probable that these feats are performed by the use of 

 visual clues, combined with a tendency to follow coastal outlines and 

 other conspicuous geographical features. Pigeons are trained to 'home' 

 by release at progressively increasing distances and can acquire the 

 ability to return from more than 500 miles. 



It has now been satisfactorily proved that pigeons return home after 

 release from a distant point even if they have never been there before, 

 nor have had any previous training in returning from situations out of 

 sight from the loft. Matthews (1955) and others have shown that upon 

 release the birds fly off towards home provided that they can see the 

 sun. This capacity to navigate by the sun must depend upon deter- 

 mination of position on two coordinates by observations of the sun's 

 altitude, azimuth, and/or movement. This implies the use of a very 

 accurate chronometer as a means of determining the difference be- 

 tween home and local time. It is hard to believe that all this could be 

 achieved in the few seconds following release, but the facts demand 

 some such hypothesis. Moreover, experiments designed to upset the 

 'chronometer' by altering the period of daylight have been claimed to 

 be effective in altering the direction of flight of birds upon release. 



Other birds are also able to return home from spectacular distances, 

 an instance being the Manx shearwater removed from its nest (burrow r ) 

 on Skokholm Island off the Welsh coast and sent to Boston by air: it 

 returned in 12 days, the distance being 3,067 miles across the Atlantic. 

 In another experiment three out of ten untrained terns returned to 

 their nests from a distance of 855 miles, in about 6 days. There are 

 many peculiar and unexplained features about such long journeys. 



5. The stimulus to migration 



It is probable that the north-to-south migrations of birds in the 

 northern hemisphere take place under some stimulus provided by the 

 internal condition of the gonads, these being themselves affected by 

 the seasonal change. Rowan has made extensive experiments with 

 juncos, birds that are summer visitors to Alberta, Canada. If the birds 

 are caged in the autumn and illuminated to compensate for the shorten- 

 ing day the gonads do not regress as they normally do at that season, 

 so that full breeding song continues on into the middle of winter. 

 Rowan had the interesting idea of releasing these birds in the winter 

 and found that they immediately moved away, perhaps northwards, 



