244- ON THE CRYSTALS IN LAVAS. 



in its incandescent paste retain their form : no change takes 

 place, the fire of volcanoes not being sufficiently intense, to 

 fuse them or alter their nature. I have adduced a great 

 many instances of this. 

 Volcanic fires On this occasion I shall recal to the reader's mind the 

 soUcl rocks "*^ ^^^^ ^ suggested respecting the probable state of the sub- 

 terranean strata, from which the lavas issue. We see, that 

 to reduce stones or minerals to a state of fusion, they must 

 be broken into very small pieces : but there are neither pes« 

 ties nor stampers in the strata from which the lava originate ; 

 and volcanic fires are as incapable as those of our furnaces, 

 to fuse rocks in a solid mass. These strata then must be 

 in a pulverulent and muddy state, to be capable of being; 

 fused. In such a state we can easily conceive chemical af- 

 finities may exert themselves, and form crystals either soli- 

 ta,ry or in groupes, that would remain enveloped in the 

 matter in fusion. How is this fusion effected ? whence 

 In these fires arise the fires that occasion it? We perceive from its ema- 

 sulphur IS the ^^^jJQjjg^ that sulphur is the principal ingredient, that iron 

 gredient, with enters into the mixture, and that muriatic acid and sal 



iron, muriatic ammoniac likewise form a part of it. But what circum- 



acid, and am- ... 



monia. stance, what combination is necessary, to excite the fer- 



mentationsj that produce the fires, the fusion, and all the 

 phenomena of volcanoes.'* On this we shall never be able 

 to do more than form conjectures, some of which may ap- 

 proximate to the truth, and others be very wide of it. But 

 as no means we are capable of employing can prevent any 

 of them, it is of little importance, whether our conjectures 

 on the origin of these fires, and the manner in which they 

 act, be just or erroneous. All that is essential is not to 

 ascribe to them a greater extent, more activity, and a wider 

 influence, than they really have ; that we may not be led to 

 form syt-tems on mistakes or exaggerations. 

 Sea-water said Mr. Fl. de Bellevue does not admit, that sea-water is ab- 

 Bot to -i ne solutely necessary to produce volcanoes ; and he quotes in 

 canic fires, opposition to this opinion, which at first appeared to him 

 as an eruption very plausible, a volcanic eruption mentioned by Messrs. 

 miles^from the '^'^^ Humboldt and Bonpland, which took place in 1759, 

 sea. " in a plain in Mexico, forty leagues from the sea in a di- 



rect line; an eruption that in one night threw up a volcano 



