206 EOCENE PERIOD. [Ch. XIX. 



carried away by the force of running water, the cone whence 

 the lava issued escaped destruction, because it stood upon a 

 platform of gneiss several hundred feet above the level of the 

 valley in which the force of running water was exerted. 



It is needless to multiply examples^ or the Vivarais would 

 supply many others equally striking. Among many we may 

 instance the cone of Jaujac, and its lava current *, which is a 

 counterpart of that near Pranal last mentioned. 



Lavas and Alluviums of different Ages, We have seen that 

 on the flanks of Etna, since the commencement of the present 

 century, several currents of lava have flowed at the bottom of 

 the Val del Bove, at the foot of precipices formed of more 

 ancient lavas and tuff's. So we find in Auvergne that some 

 streams of melted matter have flowed in valleys, the sides of 

 which consist partly of older lavas. These are often seen 

 capping the hills in broad sheets, resting sometimes on granite, 

 sometimes on fresh-water strata. 



Many^of the earlier lavas of Auvergne have flowed out upon 

 the platform of granite before all the existing valleys had been 

 excavated ; others again spread themselves in broad sheets over 

 the horizontal lacustrine deposit, when these had been covered 

 with gravel, probably soon after the drainage of the lakes. 

 Great vicissitudes in the physical geography of the country 

 must have taken place since the flowing of these ancient lavas ; 

 and it is evident that the changes were gradual and suc- 

 cessive, caused probably by the united agency of running 

 water and subterranean movements. We frequently observe 

 one mass of lava capping a hill, and a second at a lower eleva- 

 tion, forming a terrace on the side of a valley ; or sometimes 

 occupying the bed of a river. 



It is a most interesting fact that we almost invariably find 

 in these cases beds of gravel underlying the successive currents 

 of lava, as in Catalonia before described (pp. 18 ( ,), 190). Occa- 

 sionally, when the highest platform of lava is 70i) or 800 feet 

 above the lowest, we cannot fail to be struck with the won- 

 "' See Scrupe's Central Trance, plate 14. 



