42 Bulletin No. 15. 



south-west, passed over Ladies' Grove, followed the course of the creek 

 downward, and so reached their roosting place. 



The grackles first occupied the campus trees in June, 1893, when a 

 small company was in the habit of spending the night there. The num- 

 ber was never large, probably less than 500 at any one time during the 

 summer. The following summer there was a marked increase in the 

 number occupying the trees, but still not a company large enough to 

 attract the attention of the casual observer. Absence from town during 

 the greater part of the summer of 1895 prevented any systematic study 

 of the roost and its nightly occupants during that year. An effort to 

 determine the number of individuals was made late in September, which 

 resulted in an estimate of 3000. It is almost certain that the number at 

 this time was smaller than a little earlier, because every day noted a de- 

 crease in the birds occupying the trees. It was learned that at least 

 nine-tenths of all of the birds came from the east-south-east, apparently 

 from the region of Black river, which lies about three miles distant in 

 that direction. A few small companies came in at a considerable eleva- 

 tion, but the great majority skimmed along just above the tree tops. By 

 the end of the second week in October all had dispersed. 



Practically the entire summer of 1896 was devoted to the study of this 

 roost and the habits of the birds occupying it. Spear Library proved an 

 excellent point for observation, the highest part of its roof being almost 

 on a level with the tree-tops, and its north end removed only a few rods 

 from the southernmost row of trees of the roost. But the point of ob- 

 servation varied with the habits of the birds and the time of day, as will 

 be seen later. Several unsuccessful attempts were made to determine 

 the exact feeding grounds of the birds. Neither horse nor bicycle is in 

 the same class as the grackle a-wing, and during the day they could not 

 be found. From some high vantage point their approach at night could 

 be watched, and their general direction in that way determined, but that 

 was all. 



This habit of collecting in such large companies to pass the night 

 together is so interwoven with the other habits of the birds that the 

 whole life history of the grackle must be known before any correct ex- 

 planation can be. hoped for. The gregarious instinct seems to pervade 

 his whole being and finds expression in every phase of his life. He 

 winters in communities, migrates in communities, nests in communities 

 and even molts in communities, as we shall see presently. It does not 

 seem incongruous, therefore, if he becomes partial to communities of 

 men instead of clinging to the old habit of nesting and roosting in woods 



