300 Ohervations vpoT< the 



the bra aiid its substance; or are they foreign to it, and 

 formed anteriorlv, by the humid way, in the substances or 

 layers which the volcanic fires, have reduced into fusion i* 

 From, this inquiry, deduced from the true state of things, 

 and carried bv facts to a degree of evidence beyond all un- 

 certaintv, a question will be decided of important conse- 

 quences to ceologVj by presenting a jilst idea of volcanoes 

 and their phaenomena. 



The principal argument of M. Fleuriau de Bellevue is 

 drawn from the connection which he finds between the 

 formation of crystals' contained in lavas and the species of 

 cryslallization which has been called crystallites, and which 

 are formed in the glass-house pots when in a state of fusion, 

 and afterwards slowly cooled. 



We shall now examine what these glass crystallites are. 

 The entire mass of cooled glass presents a confused crvstal- 

 lization all of the same colour, in which we see small com- 

 pact bars confusedly interwoven, some of them slightly stri- 

 ated and others disposed in stars equally confused. At other 

 times there is formed at the bottom of the crucible a multi- 

 tude of threads v.hich cross each other, and also present 

 Starry forms. 



In the first case, these crystallites compose even the mass 

 of the glass, and are only distinguished in some places; in 

 the second, there are seen, through the transparency of the 

 ghiss, these heaps of threads and star-like forms, which 

 have some connection of form with the small stars of ice 

 which fall along wi*.h the snow in a strong frost. Perhaps 

 ■w-e see some examples of vitreous crystallizations more de- 

 cided ; but this case, which is rare, only proves that there 

 mav be such a circumstance as favours this crystallization iii 

 a very small space. 



■ M. Fleuriau de Bellevue finds that these crystallite forms 

 sin^u-larly resemole the treuiolite. This opinion, that there 

 exists a singular resembiap.ee between two substances, of 

 \vhich the one is a production of a vitreous nature, and the 

 6ther of a mineral layer, astonishes me, I must confess ; for 

 in that case there is noiubstiince which we might not assi- 

 milate to another, it it only has a rLLilioa to ii in point of 



form. 



