ECOLOGY OF ISLE ROY ALE. 413 



hind legs were conspicuously white crossed the Wendigo road (I, '04). 

 Others were seen which had apparently not begun to change, even on 

 the ears or feet. Two were shot which had much white on the ears. 

 Tn general the adults appeared to change first. Other specimens were 

 seen at close range with white patches on the legs, especially on the 

 hind ones, while the ears were apparently unchanged. 



From the above observations it seems probable that the Isle Royale 

 hares begin their fall moult about the middle of September. This is 

 somewhat earlier than might have been expected from the observations 

 of Miller made at Peninsular Harbor, Ontario, on the north shore of 

 Lake Superior. He reports ('97, p. 8) that one specimen was taken 

 October 5, in which the white winter pelage had begun to appear upon 

 the ears and buttocks while others secured "about two weeks later 

 had nearly completed the moult." It seems likely that there may be 

 a considerable amount of individual variation in the moulting process. 

 This is very clearly shown by an examination of the upper sides of 

 the hind legs in the series secured from Isle Royale. 



G^eographic Range. — The typical form of this hare has an extensive 

 northward range from Labrador and New Brunswick westward through 

 Ontario, north of Lake Superior; Isle Royale, Michigan; northwestward 

 to Alaska and the tree limit on the north. In 1900 Miller (p. 117) re- 

 ported that "The northern varying hare occupies the wooded portions 

 of Labrador. Its southern limit is not definitely known; but the 

 animal does not reach the northern border of the United States." The 

 specimens found last season (1904) by the Museum party thus appear 

 to be the first recorded from the United States. Miller ('97, p. 8) 

 records it common on the north shore of Lake Superior and Preble 

 ('02, p. 59) states that it is "quite generally distributed throughout the 

 region between Lake Winnipeg and Hudson Bay." 



A decayed hare was found, July 6, upon the beach in a cove south of 

 the Light-house (I, 1). It was, of course, impossible to determine 

 whether the specimen came from the immediate vicinity or had been 

 washed in from a distance. The abundance of table refuse (orange 

 rinds, chicken bones?, etc.) stranded at the head of this cove suggested 

 that at least part of the material came from the open lake to the 

 northeast. This inference is further supported by the fact that the lake 

 currents, as mapped by Harrington, favor this interpretation. The 

 occurrence of the dead hare is of interest in connection with the ques- 

 tion of the direction of origin of the mammal fauna upon the island 

 and its relation to lake currents and the lake drift. 



8. Lynx canadensis Kerr. Canada Lynx. A Ijnx skull was picked 

 up at the Ransom clearing (II, 1) at Rock Harbor; its fractured condi- 

 tion suggested that it had been killed by a trapper. Victor Anderson and 

 son, John, secured 48 skins during the winter of 1903 and 1904. Most 

 of these were from about three miles southeast of the head of Rock 

 Harbor, in the vicinity of Lake Richie. Lynx tracks were seen abun- 

 dantly on the jack pine ridge on the north side ef Conglomerate Bay 

 (I, 5), also along the trail to McCargoe Cove, from the top of the 

 Greenstone Range to the end of the trail. William Garnish, of Ash- 

 land, Wis., was camping at McCargoe Cove and reported that lynx 

 tracks were abundant in the clearings about the old mines. Tracks 



