iSgo.l Brewster on Robin Roosts. 3^CJ 



nearly certain that during August practically all our Robins visit 

 some roost nightly. 



It is by no means equally clear that individual birds or flocks 

 go always to the same roost. If this were so the number at any 

 given roost should remain uniform for a time after it has reached 

 its maximum and before the migration begins. But it not only 

 varies from week to week, but from night to night. Thus 

 at the Norton roost, where, owing to the small size of the 

 colony and to the fact that most of its members enter at one side 

 over a wide opening, it is possible to count the birds with close 

 approach to accuracy, Mr. Batchelder noted S6i on July 23, and 

 on the next evening 1062, an increase of about 23 per cent ! On the 

 evening of Aug. 15 following, standing at precisely the same 

 point, I counted only 51S; on that of Sept. 4, 1251. Mr. Faxon's 

 counts at Belmont and Mr. Torrey's at Melrose Highlands show 

 similar variations. As already stated, Mr. Faxon's largest count 

 was made Sept. 2, 1SS9, the next largest Aug. 2S of the same 

 year. The Melrose roost, during the last two seasons, was appar- 

 ently most populous in the latter part of July. More observa- 

 tions are perhaps necessary before the cause of these fluctuations 

 can be definitely ascertained; but as Robins, when not tied 

 down by family cares, are addicted to wandering more or less 

 widely in pursuit of food or recreation, it seems more than 

 likely that during August and early September they sometimes 

 pass outside the limits of the region — often, be it remembered, 

 only a few miles in extent — tributary to their chosen roosts. 

 If this is true, — and it can scarcely be doubted — what more nat- 

 ural than that they should, join at nightfall the general exodus 

 from the surrounding fields and woods, even though it leads in a 

 direction contrary to that which they have been accustomed to 

 take? In support of this assumption is the fact that, as far as I 

 have seen, the evening flights over any given place are invariably 

 in one direction, that is there are never two sets of Robins pass- 

 ing in different directions at the same time, as would be the case 

 were such rovers to return at evening straight to their own roosts. 

 This theory, it will be observed, is only necessary to account for 

 the acquisition of the habit by the young, for after it has become 

 established a knowledge of the existence and position of several 

 roosts would certainly be taken advantage of more or less fre- 

 quently. Another factor which possibly has some influence on 



