20J: BIUD STUDIES WITH A CAMEIIA 



north and the south, prevents them from joining 

 many other birds en rovle. However, ajiparontly 

 most of the birds are warned at nearly the same 

 time by a physiological change that the season has 

 come for them to return to their nesting grounds. 

 This is evidently in January, since in March a large 

 number of the young on the island were found 

 almost ready to fly, while some, as has been said, 

 were alrea<ly on the wing. There was, it is true, a 

 great variation in the develojiment of the young 

 found, and indeed the birds were still laying, but I 

 believe that the parents of these later broods had 

 been robbed of their eggs by tourists. 



A careful count yielded a total of 845 nests, 

 which had evidently been built during the season, 

 but only '.'51 of them were occupied. Most of the 

 vacant nests were on the ground, and had been 

 deserted by their tenants, who Avere now running 

 aljout the island. 



The :.'51 occui)ied nests contained eggs or young, 

 as follows : 



55 nests with 1 egg each ; 

 G;5 " " ■' eggs " 



9!! " 



03 " " I young each ; 



4(> " " 2 " " 



1 nest "" ;] " 

 Incubation was found to be well advanced in 

 eggs whi(di were alone in their nest, showing either 

 that one egg sometimes composes the set, or that the 

 other eggs of the set had been destroyed. The fact 

 that one nest was found with three young while 

 twenty-three were found each containing three eggs, 



