255 



seems to have a relation to the height of the mountains [the lakes of greatest 

 elevation being in the valleys running down from the highest mountains.] 

 The lakes terminate just where their valleys begin to widen out into plains. 

 Along their sides, and at their southern ends, there ai-e invariably vast 

 collections of shingle and large blocks of rock. 



From these similai'ities it is evident that some great natural cause or law 

 has had a uniform action over this 300 miles of country in producing the 

 lakes. 



An observer might have these facts at his command, and yet be sorely 

 puzzled to account for the origin of the lakes, if he had no access to the records 

 of Arctic or Antarctic discovery, or to the Alpine researches of such philosophers 

 as Agassiz and Forbes. But with these as guides, it is plain that the glaciers 

 which now lie far up the valleys and ravines of the mountains ax^e, compara- 

 tively speaking, the puny descendants of glaciers that formerly filled valley 

 and lake beds with their vast dimensions, and in their slow but ii-resistible 

 march, carried forward the spoil of the mountains, and deposited it as lateral 

 and terminal moraines. 



Accepting this explanation of the glaciers at one time filling the present 

 lake basins, the question arises : Did the glaciers excavate the basins, or did 

 they simply occupy thein for a time, as the lakes now do 1 



In how far a glacier could excavate a valley out of rock, is necessarily a 

 question very much of speculation. It will be of interest, however, to 

 endeavour a general demonstration of the elFect. In a paper read before this 

 Society by Mr. Beal,* attention was directed to the wearing power of 

 ice in motion, and to the rounded outline of the hills so operated on. As 

 an illustration of this action, Peninsula Hill, near Queenstown, was mentioned. 

 This hill is 1700 feet above the level of the lake surface, or about 3000 feet 

 above its bottom. If, then, we suppose that the glacier did not scoop out the 

 lake bed, but simply smoothed its surfaces, it follows that there would be a 

 glacier of from 3000 to 4000 feet in action. Now, can it be conceived that 

 this vast mass slid over the bed of the present lake for a geological era, without 

 working its bed deeper and deeper 1 



We find, on examining the beds of siich rivers as the Kawai-au and Shot- 

 over, that running water, probably never even in the highest floods more than 

 forty feet deep, can cut or wear channels through hard schist rock, of 200 and 

 300 feet deep. If running water, and the sand and gravel which it carries 

 along, can produce such effect, it seems easy to arrive at the inference that a 

 glacier, say 1000 feet thick, would, with chips of rock adhering to its under 

 surface, plane down its bed to a depth only limited by the duration of the 

 process. 



See Art. XLIX. 



