Alpine Valleys and Alpine Lakes. 89 



from the glacier of the Tessior over the pass of Monte Cenere. 

 The relatively slender streams which flowed over these barriers 

 might suffice under given climatal conditions to fill the lake- 

 basin with ice, which would spread, as really occurred, beyond 

 its present limits ; but the result must have been to produce an 

 ice lake rather than an ice river; and the boldest speculator in 

 glacial theories must hesitate to assert that any agency was there 

 present that could even tend to excavate a trench which, according 

 to the latest measures, is 919 feet below the present level of the 

 lake. 



The difficulty of applying this theory, even if it were other- 

 wise tenable, to such a lake as that of Zug, 1279 feet in depth, 

 or to others that might be cited, need not be discussed in detail. 

 The facts simply show that water, whether in the solid or liquid 

 state, tends, under the action of gravity, to seek the lowest level. 



"With reference to that part of Professor Ramsay's memoir 

 which tends to clear the way for the admission of glacier-action 

 by denying the validity of other explanations of the origin of 

 lakes, it would be rash in me to enter into controversy with so 

 accomplished a geologist ; yet I own to the belief that the causes 

 may be various, and that, until we have acquired more accurate 

 knowledge of the processes by which mountain chains have been 

 uplifted, it is premature to declare ourselves incompetent to 

 explain the depressions of the surface with which the mountain 

 ridges are correlated. In any case we are not entitled to argue 

 from our own ignorance to the admission of a new agent, until 

 its competency shall be proved by direct evidence. 



I am thus led to examine the point upon which the new theory 

 really turns, and which, as it seems to me, has been taken for 

 granted rather than cautiously investigated. The assumption, 

 which is common to Professor Uamsay and to M. Mortillet, 

 that the excavating power of a glacier in hollowing out a plain 

 surface, or in deepening an existing basin, is proportioned to the 

 weight of the mass pressing on its bed, seems to me to rest unon 

 a superficial view of the mechanical conditions of glacier-motion. 

 It has long been demonstrated that, even when lying on a slope 

 considerably inclined, the friction of the rocky bed against the 

 under surface of the glacier suffices to retard the motion of the 

 lower portion of the ice-stream so that this bears but a small 

 ratio to the velocity of the upper surface. It is impossible to 

 doubt that this retardation would be increased very largely in 

 the case of a glacier lying upon a level though irregular surface, 

 though, if this were of slight extent, the weight and momentum 

 of the glacier behind might impress upon it some slight pro- 

 gressive movement. But if we suppose a glacier of great thick- 

 ness to lie on a level surface many miles in extent, it is easy to 



