22 



THE NATIONAL GEOGRAPHIC MAGAZINE 



THl- AUT0A10BIJ.1-: IN ]?RONT OJ? VAI.DKZ GLACIER (SEE PAGE 9) 



through the moraine, and bare, steep, 

 ice cHffs were developed in place of the 

 former morainic slopes. The ice cliff 

 ■was broken, and blocks of ice, dislodged, 

 were falling to its base, while the 

 moraine, in the grasp of running water 

 hberated from the newly exposed ice, 

 -was being deposited on the mud flat at the 

 "base of the glacier, burying willow bushes 

 and annual plants that had previously 

 "been growing there. From the sea cliff 

 many icebergs were being discharged, 

 -proving still further that this margin of 

 the glacier had begun to advance ; but we 

 ■could not be certain that the main part 

 ■of the glacier was also moving forward 

 more rapidly than before, although it 

 seemed to us that the cliff was higher and 

 there was rnore ice in the fiord, and more 

 frequent discharge of icebergs than in 

 1905 and 1906. 



From the condition of Hubbard glacier 

 we conclude that an advance under the 

 -flood impulse had just begun in the sum- 

 mer of 1909, certainly along the eastern 

 ■margin, and perhaps throughout the en- 



tire glacier. It would, therefore, be of 

 great interest to return to it in 1910 to 

 see exactly what change follows, for in 

 this case we seem to have discovered a 

 glacier showing the very first signs of ad- 

 vance. It will also be interesting to ob- 

 serve whether all or only a part of the 

 glacier feels the flood impulse, for Hub- 

 bard glacier is made by the union of two 

 very large tributaries, and each of these 

 has_ large feeders. By the flood expla- 

 nation it is not to be expected that gla- 

 ciers of different lengths, with different 

 numbers of tributaries, and with different 

 conditions in their reservoirs, will ad- 

 vance at the same time. 



It is quite possible that the eastern side 

 of Hubbard glacier will respond to the 

 glacier flood at a different time and in a 

 different amount than the western mar- 

 gin, which is supplied from an entirely 

 different source. Indeed, even the same 

 side may advance under the influence of 

 the flood from some of its larger feeders, 

 then halt, and later advance again under 

 the flood impulse supplied from more re- 



