RESEARCHES IN ALASKA 



545 



ice resting where one stood to take a pho- 

 tograph the day before, or to find some 

 great tree, 100 years old, prone on the 

 ground with the butt beneath the glacier, 

 where the day before the tree was up- 

 right with the ice just touching it or with 

 room to go between the glacier and tree. 



The rate of movement in midglacier 

 during- the summer was not determined, 

 but it would not be improbable, judging 

 from the rate at the north margin and 

 from rates at intervals across other gla- 

 ciers that have been measured, that it 

 was at least six times the rate in 1909, or 

 30 feet a day, and even more at times. 



But the middle of the glacier did not 

 advance as far during- the summer as the 

 north margin did, and between June 10 

 and August 11 it retreated 450 feet. 

 From August 11 to August 17 it re- 

 treated 65 feet more. This retreat was 

 occurring while the north margin was 

 advancing most strongly, and it seemed 

 to the National Geographic Society's 

 party that this retreat was not due to a 

 cessation of advance, but wholly to un- 

 dercutting by the Copper River, which 

 rose over six feet between June and Au- 

 gust because of summer melting of snows 

 and glaciers. 



We accordingly predicted in August 

 that the ice front in the river would re- 

 advance the latter part of September, 

 when the river was lower. This predic- 

 tion proved correct, the ice front advanc- 

 ing 390 feet up to October 5, plus the 

 amount of additional retreat between 

 August 17 and the date when the ad- 

 vance commenced, the level of the river 

 having fallen nine feet meantime. As 

 the middle of the glacier advanced 390 

 feet in the 49 days between August 17 

 and October 5, the minimum rate of 

 actual advance was fully eight feet a day, 

 if there was advance on each of these 

 days. As (a ) icebergs were discharging 

 all the time and melting was in progress ; 

 as (b) the advance was surely over 300 

 feet; and as (c) the forward movement 

 began not on August 17, but probably 

 after the middle of September, when the 

 river was much lower, the rate of ad- 

 vance was much faster and may easily 

 have been 30 feet a day, as previously 

 estimated. 



