56 Bulletin No. 15. 



of the birds would seem unfavorable for a long journey, but, as we 

 know, not impossible. 



The study of this roost has shown that during courtship and nesting, 

 each occupying about two weeks under normal conditions, none of the 

 grackles flock together to pass the night, whatever they may have done 

 previously ; but as soon as incubation has well begun the old males seek 

 the shelter of some convenient grove and pass the night there with others 

 of their kind. As soon as the most forward young are able to fly they 

 are escorted to the common roost by the old male, or if the whole brood 

 should develop at the same time, by both parents ; and where there is 

 any marked difference in the development of the young of the same 

 brood, the later ones, accompanied by the old female, bring up the rear. 

 The young are fed for some days after they begin to roost with the old 

 ones. All of the birds eat fruit while it is sufficiently abundant, then 

 green and later ripening corn until the roost is deserted. The complete 

 summer molt is performed while the roost is occupied. The fruit and 

 green corn diet is coincident with the molting season, either from neces- 

 sity or from choice. In general, the birds depart from the roost with 

 the rising sun, and return to it at sunset. Singing and calling begin with 

 the break of day, and continue until the birds depart for their feeding 

 grounds. At night there is comparatively little singing, and all noise 

 and shifting about cease as darkness falls. Early in the season the birds 

 arrive and depart independent of each other, but with the advancing 

 summer flocking increases, until finally all move as one individual. 



During his sojourn in the north the grackle has a period distinctly bene- 

 ficial to agriculture — the breeding season — and a period distinctly injuri- 

 ous to agriculture — the roosting season. The little injury done by robbing 

 other birds' nests during the beneficial season is not worth taking into 

 account. It is not probable that the berries eaten are any loss to any- 

 body, since there are no complaints from fruit growers against this species. 

 Hence, the only real damage done by the grackles is when they feed upon 

 growing grain and upon that which has not been put out of their reach. 

 This reduces the injurious season to a small fraction of the whole year, 

 and is far too little damage done for the death sentence to be pronounced 

 as a penalty. 



Scarcely a beginning has been made in the study of this roosting habit. 

 Before we can understand it we need to know more about the influence of 

 migration, nesting, molting, food and food supply, disposition, enemies, 

 and the role of man upon the life of the grackle. In short, we need to 

 know far more about the grackle than we do now. His is an interesting 

 character which will amply repay persistent study. 



