,400 PROCEEDINGS OF THE GEOLOGICAL SOCIETY. [April 9, 



were merely a few local ice-centres, from which, glacier-erosion 

 might radiate. "When there was any thing approaching to a general 

 ice-coating, the erosion must have been enormous, especially at 

 spots where there were already preexisting valleys, in which, of 

 course, the ice would be thicker than elsewhere, and where, con- 

 sequently, its grinding powers would be far greater, and add 

 much to the work of the more ancient rivers. He inquired of 

 Mr. Bonney what must take place at the termination of any 

 glacier. There could be no erosion by the glacier anywhere beyond 

 the point to which it extended ; but where it existed some erosion 

 must take place, and a basin of greater or less depth must thus 

 be ground out by the alternating advance and retreat of the glacier. 

 A river flowing out might, it is true, in some cases cut a gorge so 

 as to drain the basin, but this would not prevent a basin being 

 formed. In the western Alps, where the height was greatest, there 

 also were the largest and, in some instances, the deepest lakes; 

 though the size and depth were often connected with the nature of 

 the rock. Some lakes, however, had been partially filled by detritus. 

 As to the cliffs, he perfectly agreed with Mr. Bonney in attributing 

 them to meteoric causes ; and in many cases he thqught they were 

 of preglacial origin, though subsequently much modified. He repu- 

 diated any idea of the larger valleys having, even in glacial times, 

 been always completely filled with ice, and deprecated judging of 

 these questions from diagrams in which the horizontal and vertical 

 scales were different. He repeated that the original large valleys 

 were, in his mind, of great antiquity, but all rather due to erosion 

 in some form or other than to any internal disturbances of the 

 strata. They were therefore to a great extent cut out before the 

 last glacial period. He had been much gratified to find that, after 

 so many years of opposition, the result of Prof. Gastaldi's mapping 

 of the rocks of the country on which he had founded his own 

 opinions, had been to bring the Professor round to the adoption of 

 the same conclusions. 



Mr. Blaneokd added that in tropical mountains (for instance, 

 the Neilgherries in India, arid in Abyssinia) he had not been able 

 to trace any ice-action whatever, and there also he had seen no 

 cirques, nor any lake-basins similar to those of Northern Europe. 

 He was therefore compelled to connect both lake-basins and cirques 

 with the existence of glaciers. 



Mr. Campbell, of Islay, referring to that portion of Mr. Bonney's 

 paper in which the author alluded to partial subsidence and flexure 

 to account in some degree for the basins of the alpine lakes, described 

 the Thingvalla lake in Iceland. That hollow is over six miles wide, 

 and, for a length of more than twenty miles, is clearly the result of 

 fracture and subsidence of the crust formed upon a great lava flood. 

 Mr. Campbell pointed out that the alpine hollows were not like this 

 Icelandic subsidence, and attributed them to erosion, not to sub- 

 sidence and flexure. Subsequently the author said that he attri- 

 buted the valleys to erosion. 



The Duke of Argyll thought that Mr. Bonney's paper as to the 



