710 REPRODUCTION AND MIGRATION IN BIRDS 



taken within 60 miles of the equator and shipped to the San Francisco 

 area, where they arrived in April. The normal breeding period in the 

 Galapagos Islands extends from mid-December to April and is cor- 

 related with the rainy season, but breeding, to a limited extent, may 

 occur in almost every month of the year. The captive birds in San 

 Francisco bred from March to November inclusive, but the nesting 

 period for most of the captive birds was confined to spring and sum- 

 mer. In this case, as in others cited by Baker, it is probable that day 

 length is the factor responsible for the change in breeding season. 

 Observations on the molt cycle also suggest that it was influenced by 

 day length. When some of the finches were slightly over a year old, 

 they were in a plumage which in the Tropics takes about three years 

 to attain. 



A number of other correlations point to a relationship between day 

 length and reproduction in the Tropics. ( 1 ) In many tropical species 

 which occur on both sides of the equator the breeding periods in the 

 northern and southern populations are correlated with the seasons 

 and, hence, occur at opposite times of the year (Baker, 1938; Chapin, 

 1932). Frequently this is associated with the wet or dry season, but 

 often it is not. (2) In many groups of birds, and even in a species with 

 wide distribution, clutch size tends to be smaller in the Tropics than 

 in the temperate latitudes (see Lack, 1947, for review). In the 

 domestic fowl, egg laying (which is not homologous with clutch size) 

 is greatly influenced by latitude (Whetham, 1933; Romanoff and 

 Romanoff, 1949). (3) The maximum size of active testes in tropical 

 species is only 8 to 67 times the size of minimum inactive testes, 

 whereas in temperate species it ranges from 267 to 2096 times the 

 minimum size (Moreau et al., 1947). 



The gonadal response in experimental juncos and white-throated 

 sparrows simulated some of the correlations between latitude and re- 

 productive activity. Birds subjected to 20-hr photoperiods achieved 

 breeding condition rapidly, showed a tendency toward larger and 

 more active gonads, and exhibited maximal activity for only a short 

 period. Juncos held at 12-hr photoperiods achieved breeding condi- 

 tion at a slower rate, showed a tendency toward smaller and less ac- 

 tive gonads, remained sexually active for long periods of time, and 



