682 REPRODUCTION AND MIGRATION IN BIRDS 



on the basis of his studies of the gonadal cycle of the starling {Stiirnus 

 vulgaris), postulated that there was an inherent rhythm of activity 

 of the pituitary gland; the gonads, therefore, were cyclical in their 

 activity, since they were under the control of the pituitary. In the fall, 

 birds migrated southward owing to regression of pituitary activity and 

 did not breed on the wintering grounds because the pituitary and 

 gonads were in a refractory phase. On the spontaneous recovery of 

 the pituitary in late winter or spring, the birds were stimulated to 

 migrate before or while the gonads recrudesced. Bissonnette's theory 

 precluded any regulatory effect by day length in the Tropics and 

 Southern Hemisphere. 



Although there was ample experimental evidence for the existence 

 of a "refractory period" in the fall when increasing day lengths would 

 not stimulate gonadal growth — and Bissonnette made this very impor- 

 tant contribution to the problem — it seemed that, if day length 

 played a regulatory role in North Temperate species, it might play 

 a similar role in species that winter in the Tropics and Southern 

 Hemisphere. Since the days were not increasing in length in those 

 areas, then perhaps some other aspect of day length was operating. 

 It occurred to me that perhaps it was the total amount of light which 

 the birds received in a given period of time that was important and 

 not whether the day lengths were increasing, decreasing, or relatively 

 constant. This idea was put forth as the "summation of day lengths" 

 hypothesis (Wolfson, 1947). I shall discuss first at some length the 

 experiments and data that pertain to this theory. Then I shall consider 

 the refractory period. 



DAY LENGTH AND INITIATION OF MIGRATORY 

 AND REPRODUCTIVE RESPONSES 



The first test of the summation hypothesis was the determination 

 of the total amount of light that birds wintering at various latitudes 

 received under natural conditions (Wolfson, 1952a). Experimental 

 work notwithstanding, conditions in nature could well preclude the 

 possibility that summation is a regulatory factor. These data showed 

 only small differences in the summations from December 21 until the 

 onset of migration in March or April. Hence, experimental studies 



