ORIGIN OF VALLEY-LAKES. 81 



essential point in connexion with diluvial accumulations is the fact, 

 which can be verified over and over again, that the lakes at the 

 outflow of which such accumulations are found, lie above the point 

 of convergence of two valleys. It is for the advocates of the theory 

 of glacial excavation to show why the work of excavation was not 

 done where the union of two glaciers should have been capable of 

 doing the greatest amount of work. 



We may define a lake as a hollow in the surface of the earth, which 

 is so placed in relation both to the superficial and underground 

 drainage of the district in which it is situated, that it is filled with 

 water, either to the lowest point or points of overflow, or to such a 

 mean level as corresponds with the excess of the mean annual precipi- 

 tation over the mean annual evaporation of the district in which it 

 lies. This is about as much as can be asserted generally of lakes ; 

 and it is impossible to give any general rule as to the formation of 

 their basins. Each basin must be worked out as it is, in relation to 

 all the geological, cosmical, hydrographical, and other conditions 

 which bear upon its existence. The hypothesis of the glacial exca- 

 vation of lake-basins is not only irreconcilable with the known 

 physical properties of ice ; it is, besides, unnecessary, since lake- 

 basins can be accounted for without it. 



Discussion-. 



Mr. W. R. Browne suggested that a lake would be emptied by a 

 slight tilt in the general slope of the country ; and a corresponding 

 diminution in the slope would form the lake. Wastwater would 

 require as much as one in twelve. This change, which would 

 only be one in fifty for the Lake of Como, and one in ninety for the 

 Lake of Geneva, might well have taken place in the various alterations 

 which have happened in times since denudation had commenced. 

 The elevation of the central part of the chain would tend also to 

 increase the glaciers. 



Prof. Blake stated that there were many causes for lakes, and 

 that those mentioned by Mr. Irving were familiar to geologists. 

 The main question is, Can a glacier ever excavate a lake-basin ? If 

 that could be made clear, some progress would be made. He did not 

 think it was proved that glaciers could not excavate lakes. Prom 

 the shape of the sides of glacier-valleys, there must be most erosion 

 at the bottom. If the medial moraines are brought up from the 

 bottom, so might the excavated material be. The lakes of Neu- 

 chatel and Constance were difficult to explain as eroded by glaciers. 

 It was singular that, as the Lake of Geneva had been filled with 

 ice, Miocene fishes should still be found in it. 



Mr. A. Tylor had jDroved by experiment that ice has erosive 

 power. The velocity increases as the cube-root of the mass ; and the 

 erosive power augments in a high ratio with the velocity. Multi- 

 plying the present mass of a glacier by 64 would increase its velo- 

 city by 4, and its erosive power by at least 200. This he had 



Q. J. G. S. No. 153. g 



