402 PROCEEDINGS OF THE GEOLOaiCAt SOCIETY. [JuHG 18, 



with a much smaller and apparently insignificant brook, although in 

 reality the prime mover, the "fons et or'igo " of the whole excavation. 



This longitudinal soft band, moreover, as it is worn down by the 

 weather * more rapidly than the rocks on each side of it, may, as soon 

 as an outlet is established for the carrjing off its disintegrated par- 

 ticles, be readily conceived to extend across some of the minor lateral 

 brooks, and thus deflect their waters towards the principal one. 



This is the explanation, as I believe, of the relations between the 

 Khone and the Durance and Dranse at Martigny, and of those of the 

 Upper Ehine and the rivers that come into it at Chur, where the 

 Rhine turns at right angles towards the Lake of Constance, and of 

 aU the other valleys and rivers of the Alps, and of all other moun- 

 tain-chains, as the inspection of any good map of any one of them 

 will be sufficient to prove. 



It is not necessary to suppose that the whole mountain-chain was 

 dry land before the process commenced : it may have gone on, with 

 infinite slowness, as the mountains gradually rose above the sea ; it 

 may have been interrupted by the depression of the mountains, and 

 renewed on their re-elevation. It is, however, essentially a subaerial 

 action, and one that only commenced after the termination of the 

 marine denudation which gave to the mountains their general form 

 and outline as that form would be if all their valleys and hoUows 

 were filled up and their summits connected by gently sloping plains. 



The principal part of that marine denudation must have been 

 effected long before the commencement of the formation of the river- 

 valleys ; but it would be a very difficult task to undertake to explain 

 the mode of action of the marine denudation, and the precise rela- 

 tions either in time or effect between the marine and the atmo- 

 spheric denudation. 



I can only hope to have established this general conclusion, that, 

 while the removal of the vast masses of rock under which our present 

 mountain-chains were formerly buried may be most reasonably re- 

 fer, ed to the action of the sea {" yueya aQevos 'i^x^^'^olo "), we must 

 look chiefly to the " weather " as the producer of the glens, ravines, 

 and vaUeys by which our mountains and plains are traversed. 



Note. — In the section fig. 5. PI. XX., the line AAA is bent ; but in 

 the original section drawn on the true scale it forms one straight line 

 from the summit of the Knockmealdowns to the sea, just touching 

 the top of aU the intervening hiUs, and giving the idea of a sloping 

 plain, below which the valleys have been excavated. Professor 

 Eamsay, at the Meeting of the British Association at Oxford in 1847, 

 exhibited a precisely similar section across a part of Cardiganshire, 

 and read a paper on it, attributing the origin of the valleys to exca- 

 vation below an old plain, although he looked to the sea as the ex- 

 cavator, while I now believe it to have been the atmospheric waters, 

 (See Report of the British Association, vol. xvi. Sect. p. QQ). 



* Under the term "weather" I would include glacial action and every other 

 atmospheric agency. The Rhone valley has obviously been completed by glacier- 

 action. 



