i06 LiTBRATURE ON TORRENTS. 



The action of glaciers has been studied in the Alps ; results obtained 

 have been applied to the phenomena presented by and in connection with 

 the parallel roads in Glen Roy ; these have been satisfactorily shown to 

 have been produced by glacial action, exhibiting on a grand scale the same 

 phenomena as are to be seen in the valley of Bagnes, marking out what 

 must have been the banks of the ancient lake of Gi6troz. 



He quotes a paper by M. Ch. Martins, which appeared in the Bevue des 

 Deux Monies for March 1867, in which it is said, — " Throughout almost the 

 entire length of the valley of Glen Roy — that is to say, for ten miles and 

 upwards — there, may be traced on the opposite declivities three terraces or 

 parallel banks, perfectly horizontal and corresponding exactly on the two 

 sides of the valley. From a distance they are distinctly visible ; when 

 reached they are found to be a pebbly surface from 10 to 60 feet wide, the 

 slope of which is less steep than that of the mountain above and below. 

 The lowest of the terraces is 750 feet above the level of the sea, the second 

 about 210 feet higher, the third upwards of 80 feet above the level of the 

 second, — all terminating at the head of the valley on the col which separates 

 it from the valley beyond. 



"In 1840 Buckland and Agassiz visited Glen Roy, and perceived that 

 temporary barriers to the flow of water could alone account for the formation 

 of these level lines. Glaciers coming successively to close up the one or the 

 other issue of the valley, the stream which ran through it would form a 

 lake, which would flow by the col towards which the terrace inclined. 

 Agassiz recognized polished and striated rocks and the ancient moraines 

 which he had learned to distinguish in the Alps ; and subsequently Mr 

 Jamieson has given a chart and details confirming completely the view of 

 the illustrious Swiss naturalist. 



" Mr Jamieson carries back the formation of these terraces to the close 

 of the second glacial period, when it was due to an oscillation of the glaciers 

 descending from Ben Nevis and the surrounding mountains. These barriers 

 have barred up, one after another, the vaUey of Glen Roy and the neigh- 

 bouring valleys. The waters, stopped in their flow, have formed lakes at 

 different levels, determined in each case by the height of the col which 

 closed the extremity of the valley opposite to that barred up by the glacier. 

 The entiireness of the terraces prove also that subsequent to the formation 

 of these Scotland has never been submerged." 



And M. Cfeanne goes on to say, — " If this theory be coiTect, it follows 

 that we may expect to find at the dihouche of each of the great valleys of 

 the Pyrenees and of the Alps masses of loess, deposited in accordance with 

 the characteristic forms of torrential deposits ; and the dimensions and com- 

 position of which may be in accordance with the immense duration of the 

 glacial period, and so in accordance with the greatness of the phenomena 

 to which this has given rise. 



" At the foot of the Pyrenees there exist such deposits ; if not most con- 

 siderable, they are at least most perspicuous. Let us ascend some height 

 such as the Pic du Midi of Bigorre, or the Cap of the Col d'Aspin, the 

 ascent of which is very easy ; from this elevated point the observer, turning 

 to the north, can freely oast his eye from the west to the east, over a vast 

 plain which rounds itself off at the horizon, as does the sea, and the bleak 

 fawn colour of which contrasts strongly with the green and sombre wall of 

 the Pyrenees. 



"^This general effect is pretty well rendered by the chart of the Mat- 



