ECOLOGY OF ISLE ROYALE. 101 



consisted only of the parent birds and young, but as these wandered 

 aboiit they were joined by other families and, impelled by the gregarious 

 instinct which is so strong after the breeding season, kept together and 

 formed one large flock. Beginning to feel the migratory impulse they 

 were restless and wandered about over considerable territory, probably 

 being joined from time to time by other families and often by other 

 forms, for a flock of migrating warblers is rarely composed of a single 

 species, as are the flocks of so many birds. Small bands of Myrtle 

 Warblers were seen feeding in the balsam trees on August 18, and on 

 August 19 the first flocks of Tennessee Warblers appeared. However, 

 these were scattered and composed of only a few birds, mostly adults. 

 On August 20 I saw the first Eedstarts, and from then on different 

 species were constantly making their appearance. On the 26 the bulk 

 of the Black Poll Warblers began to arrive, only a few adults being 

 seen among the hundreds which came to the clearing. It is a significant 

 fact that, in all cases where the young were not in company with the 

 adults, the latter and not the former, as some have reported, preceded. 

 In tlie case of the Tennessee Warblers three days elapsed before there 

 was any noticeable number of young, while toward the last of 

 the migration the young greatly outnumbered the adults. Throughout 

 the entire migration, however, the immature Blackpolls outnumbered 

 the old birds, in fact the latter were very rarely seen. Only two Black 

 and White Warblers were observed, and only four small flocks of 

 Black-throated Green Warblers. 



The principal feeding grounds were among the alders, birches and 

 balsams which lined the more open parts of the road. In the narrow 

 strips where the high conifers bordered the path, the bird life was 

 scanty, and when these portions were encountered by the migrating 

 warblers they were quickly passed, often in a single flight. The Myrtle 

 Warblers were the only ones observed to linger among these large trees. 



The food of the warblers consisted lai'gely of insects, most of it being 

 gleaned from the leaves and twigs of the bushes, but some was taken on 

 the ^\■ing. The open area here afforded a greater supply of insects than 

 the forest, and this may possibly have played a part in the choice of 

 this jjarticular habitat. 



On cold mornings, when the thermometer registered about 45° F. or 

 below, the warblers would remain hidden in the dense underbrush, 

 not appearing until about nine o'clock, when the sun would be quite 

 warm and the usual morning fog be dispelled. This fog hung over the 

 harbor nearly every morning and frequently was so dense that Beaver 

 Island, in the harbor, was almost invisible. It Avas often blown back over 

 the clearings, and until it raised, the majority of the birds remained 

 quiet. The height of the migrating movement seemed to be fi'om the 

 middle of the afternoon until nightfall; how far into the night it ex- 

 tended I was unable to ascertain, but the cries of innumerable birds 

 could be heard until nearly morning. These cries, usually of a single 

 faint syllable, were possibly uttered to help keep the birds together. 

 When the migrating flock had to cross the clearing it was a noticeable 

 fact that they rarely flew directly across, thus exposing themselves to 

 the attack of the numerous hawks, but instead kept near to the ground, 

 making short flights from bush to bush, and where these were scattered 



