( 345 ) 

 quantity and fize of the bubbles they con^ 

 t.aincd. This is confequeiuly another an4 

 a mofl convincing proof, that a greater 

 gafificaiion of flony fubPtances may be ob- 

 tained ill a very firong heat than in a 

 weaker. 



From thefe ten experiments I conceive 

 niyfelf fufficiently warranted to conclude, 

 that thofe bubbles and inflations of various 

 lizes, which v^'e fo frequently obferve in 

 the products of volcanos, are hj no means 

 canfed by the aalon of any permanent gas, 

 but by that of an aeriform fluid produced 

 by the exceffive attenuation of thofe fame 

 produdis in confequence of heat. 



I do not know that fnr/ilar experiment^ 

 have been made by any other perfon ex- 

 cepting Dr. PrieRley, Vv'ho defcribes three, 

 whicl:^ I thinl>: proper to be mentioned here, 

 as they differ fomewhat from mine. This 

 liaturaiift, having fufcd in a fand-fcone re- 

 tort 4-^ ounces of Iceland lava, obtained 20 

 jnpafures of air, the half of wbich, about the 



beginning 



