ECOLOGY OF ISLE ROYALE. 103 



way out. The Sharp-shinned Hawks proved to be their worst enemy. 

 The first flocks of Savannah Sparrows to arrive remained for several 

 d'ays, their numbers constantly increasing. On September 5 many 

 of the Chipping, White-throated, and Savannah Sparrows left the island, 

 and for the next two days only a comparatively few were seen, then 

 others came in from the north and the flocks were rapidly increased. 



It was noticeable that most of the birds which migrated from the 

 island on September 5 were adults, the young remaining until a later 

 time. The Savannahs showed the least fear of man during migration 

 of any of the sparrows. On September 12, ■with the great bird wave, 

 came the Lincoln Sparrows. Throughout this and the next three days 

 hundreds of these birds were seen. As a rule they kept secreted in 

 the burned brush and weeds of the first clearing, but individuals Avere 

 met with all along the road, where they were seen hunting among the 

 fallen logs and underbrush for insects. 



Chipping. Sparrows remained throughout the entire period during 

 which observations were made, but probably none of the individuals 

 first seen remained throughout that time. This seems the more likely 

 as on several dates the bulk of the sparrows of all species left, while 

 more came in later from the north. 



c. Haioks. During a few days spent at Washington Harbor early 

 in August (Aug. 5 to 8) only a few Sparrow and Sharp-shinned Hawks 

 were seen, but by August 18, many individuals of both species had ar- 

 rived. These remained here to feed on the swarms of grasshoppers in- 

 fecting the meadows, and on the small birds, as warblers and sparrows, 

 which were easily caught in the exposed clearing. The first few days 

 the Sparrow Hawks outnumbered the Sharp-shinned about 10 to 1, but 

 as the season advanced their numbers became more equal and toward 

 the last the Sharp^shinned outnumbered the Sparrow Hawks, both, 

 because of a steady increase of the former and because many of the 

 latter left the island for the south. When the observations were first 

 made the adult Sparrow Hawks were as numerous as the immature, 

 but toward the last of September the adults had nearly all left and 

 many more immature had taken their places. Some idea of their 

 number may be gained from the statement that more than thirty were 

 counted at one time, sailing over the first clearing. Until the middle 

 of September the immature Sharp-shinned greatly outnumbered the 

 adults. These immature were full size, but did not have the spotted 

 plumage of the adult. The females of both species greatly predominated. 

 Toward the end of September many male Sharp-shinned, both im- 

 mature and adult, appeared. These two species of hawks fed on grass- 

 hoppers to a considerable extent, but many crops of both species were 

 found filled with the remains of Tennessee, Palm and Blackpoll War- 

 blers, Savannah Sparrows and other species not determined. As a 

 rule the older hawks were the ones which destroyed the birds, and this 

 may account for their migration from the island at the same time that 

 the large warbler and sparrow wave passed, while the immature hawks 

 remained. Pigeon Hawks were recorded from time to time, but not 

 until September 12, when the lower end of the island was suddenly 

 flooded with bird life, did they appear in any numbers. On this date 

 several flocks of C or 8 were seen in different parts of the clearing. 



