Seasotml Periodicity 229 



recognized as one environmental factor that changed regularly year 

 in and year out and that could account for the exact timing of the 

 migration regardless of the weather conditions. Present evidence 

 indicates that the effect of the daylength is brought about through 

 controlling the amount of light reaching a sensitive tissue in the 

 bird's body (the pituitary gland). Under natural conditions the 

 larger amount of light received during the long days of spring causes 

 an increase in the size of the gonads and also, in some way as yet un- 

 known, sets off the northward migration. Following the breeding 

 season the diminishing daylengths of late summer and autumn cause 

 a decrease in gonad size and provide the stimulus for the southward 

 migration. Evidence on the mode of action of light and other factors 

 on bird migration has been summarized by Farner ( 1950 ) . 



Although photoperiodism thus appears to be basically involved in 

 the annual stimulus for the migration of birds, it does not explain the 

 evolutionary origin of the migratory habit nor, most mysterious of 

 all, how the birds find their way. Once established, however, migra- 

 tion would tend to be perpetuated by its advantages. At higher lati- 

 tudes the longer days make more time available for nest building 

 and for feeding the young. The young grow more rapidly, and hence 

 the whole process of rearing the family, with its attendant dangers, 

 is completed more quickly than would be the case nearer the equator. 

 By leaving the north during the winter, on the other hand, the birds 

 avoid the hazards of low temperatures and shortage of food. 



The migration of birds is evidently a secondary effect of the in- 

 fluence of light. The primary effect is in the control of the breeding 

 cycle since this can be demonstrated in species of birds that do not 

 migrate. The occurrence of breeding in the English sparrow at dif- 

 ferent latitudes over the face of the earth is indicated in Fig. 6.24. 

 Near the equator some breeding takes place in every month of the 

 year. In areas farther north and farther south reproduction tends to 

 be confined to the months of the year with longer days. North of 

 50° N latitude breeding is strongly centered in the month of May with 

 no breeding at all during the months of short daylength, and south 

 of 50° S latitude most breeding takes place in December. Outside of 

 the tropics, therefore, the light factor plays an important role in con- 

 fining the breeding period to the months of the year when reproduc- 

 tion is most successfully carried forward. Since short daylength pre- 

 vents birds from entering the breeding condition too early and the 

 long days of midspring act to speed up the laggards, the whole popu- 

 lation tends to breed within the same favorable period ( Bartholomew, 

 1949). 



