238 



several volcanoes of the Andes, we must not con- 

 clude from this isolated fact, that the whole of the 

 Archipelago of the Canaries is the production of 

 submarine fires. The island of Gomera contains 

 mountains of granite and mica-slate * and it is 

 undoubtedly in these very ancient rocks^ that 

 we must here seek, as well as on all other parts 

 of the Globe the centre of the volcanic action. 



mounted by columnar basalts, which serve, in their turn, as 

 a basis to the feldsparry lavas. The last alone appear to be 

 owing* to the present volcano. The basalts and the porphy- 

 ries belong to a system of older mountains, which cover a 

 great part of Sicily. The porphyries of Etna are volcanic 

 without doubt j but every rock which Owes it's composition 

 and it's form to the action of fire and vapors, has not made 

 part of a current of lavas. These observations appeared to 

 me so much the more necessary, as some very distinguished 

 mineralogists have recently affirmed, that the Peak of Tene- 

 riffe and Vesuvius are mountains of porphyry of Neptunian 

 origin, and undermined by subterranean fires. The lava of 

 la Scala has been described without hesitation as a particular 

 rock, under the name of graustein, though it issued from the 

 crater at a well known epocha, in 1631 : some have even 

 gone farther; they have supposed, that Somma exhibits the 

 untouched nucleus of Vesuvius, though it's stratified mass, 

 traversed by veins filled with more recent Java, is identical 

 with the rock constituting the actual crater, which has evi- 

 dently been in a state of fusion. Somma exhibits the same 

 leueites as abound in the greater part of the lavas of Vesuvius, 

 and their crystals are included in a phonolite resembling that 

 of the top of the Peak of Teneriffe. 



* Note manuscrite de M. Broussonet. 



f Dolomieu, in the Journ. cle Phys. 1798, p. 414. 



