ECOLOGY OF ISLE ROYALE. 337 



ANNOTATED LIST OF THE BIRDS OF ISLE ROYALE, 



MICHIGAN. 



BY MAX MINOR PEET. 



1. Introduction. 



Our observtitions on the birds of Isle Royale extended over the period 

 between July 5 and September 22, 1905. Three parts of the island were 

 studied, namely Rock Harbor, Siskowit Bay (especially near the outlet 

 of Siskowit Lake), and Washington Harbor. The party remained at 

 Rock Harbor from July 5 to August 1; at Siskowit Bay from August 1 

 to August 17; and at Washington Harbor from August 17 to September 

 22. At Rock Harbor the observations were made by O. M'Creary, N. 

 A. Wood, and Dr. R. A. Brown. At Siskowit Bay the work for the first 

 week was carried on by M'Creary and Wood, as Brown had left the 

 island; on August 8 they were joined by the writer. On our arrival 

 at Washington Harbor M'Creary left the island, and the work was con- 

 tinued by Wood and the writer until September 1, when the former 

 was called home. However, he was forced by severe storms to remain 

 on Washington Island at the mouth of the harbor until September 5. 

 and while there made a number of observations which are included under 

 their respective heads. The observations during the remaining period 

 (September 1 to September 22) were made by the vsTiter. Before 

 joining the party at Siskowit Bay, he had spent three days, August 5 

 to 8, at Washington Harbor, the observations giving some idea of the 

 bird life at that place before migration had set in. 



Practically all the birds observed at Rock Harbor were nesting, those 

 observed within a few days after our arrival at Siskowit Bay may also 

 be considered as breeding, but after about the first of August it is not 

 safe to say whether the bird nested there or was an early migrant. As 

 an example of this we may cite the case of the Tennessee Warbler, which 

 probably did not nest on the island, and yet was first observed there 

 Auigust 2. Unless the nest was found or young unable to fly, we did 

 not consider them as breeding in that vicinity, if seen after August i. 

 In the case of the waders, the earliest migration date must be placed in 

 the latter part of July. 



Under the head of stations, the particular habitats in which the birds 

 were actually found are given with their numbers, so that a fuller 

 description of the conditions existing there can be easily obtained by re- 

 ferring to that number under the "Description of Stations." It must not 

 be supposed that the birds were limited to the station in which they are 

 recorded. In all probability the birds noted in one tamarack swamp 

 would be found in nearly every similar habitat on the island. But owing 

 44 



