EFFECTS ATTRIBUTABLE TO EROSIOIS. 179 



porcelain tubes. We also know, from experiment, and many effects ob- 

 served in manufactories, that the peroxide, of iron, the oxides of chrome, &c., 

 are vola'tilized, and penetrate the substance of bodies that envelope them. 

 The experiments of M. Gaudin, with a blow-pipe on a de'tonating mixture, 

 show that silex, magnesia, and lime, are also volatile oxides ; the first after 

 fusion, the others belbre being melted. These facts evidently lead to an ex- 

 planation of all the phenomena of metamorphisrn, and the intrusion of 

 'oreign substances into sedimentary deposits, either in veins or in a state of 

 dissemination. 



EFFECTS ATTRIBUTABLE TO EROSION. 



We have seen that waters act by the carbonic acid they contain ; by tht ir 

 Weight; by their dissolving power; by their transporting power; by their 

 shock, as in waves of the sea, and thus denude continents. We have also 

 pointed out, that in arena'ceous formations, valleys are produced by erosion, 

 precisely as ravines are formed in sandy soils, by the action of rain-water. 

 Hence we may infer that, in every revolution that movements of the soil 

 must have necessarily determined, the waters, thrown forcibly sometimes on 

 one side and sometimes on the other, must, as in our time during earth- 

 quakes, have ravaged, divided, and modified pre-existing deposits in various 

 ways. Many circumstances may be explained by erosion of waters, and the 

 denudations it occasions. 



25. At first, when we see more or less numerous hillocks of 

 sedimentary matter in a country (Jig. 288), whose summits are 



Fig. 288. Hills produced by denudation. 



nearly on the same level, and whose strata correspond with each 

 other, we are naturally led to consider them as evidence of great 

 removals effected by the waters, at certain epochs, the relative 

 dates of which remain to be ascertained. In this way we explain, 

 according to appearance, all the sections which the sandstones pre- 

 sent on the eastern slope of Vosges ; that remarkable assemblage 

 of peaks of every form seen at Aldersbach, in Bohemia; the nu- 

 merous hills that cover Ross-shire, in Scotland ; the gypseous hills 

 in the neighbourhood of Paris, all composed of the same beds 

 placed at the same height ; and the division of the basa'ltic tables 

 that crown the hills, in certain localities, as well as the rupture of 

 certain lava-floods that had barricaded valleys, &c., &c. 



Valleys which intersect moveable formations are evidently produced in the 

 same way ; and there is no doubt that most of those existing in solid forma, 

 tions, have been modified by erosion of water after the rupture which gave 

 origin to them. In this way we may explain the smoothing of all their 

 parietes, in a great many localities, and the widening of their upper parts. 

 The great lakes sometimes found at the extremity of valleys, as on the two 

 slopes of the Alpr>, in Switzerland and Piedmont, may be attributed to the 

 afflux of waters which rushed through them, at the period of some great ca- 

 tastrophe, and emptied with violence on the plain in which they terminated 



25. What forms of surface are attributable to erosion and denudation T 

 36* 



