BY COMPRESSION. gQ^ 



Some experiments which I formerly * laid before this A low rolcaukr 

 Society and the public, combined with those mentioned in this the carbonate 

 paper, prove, that the feeblest exertions of volcanic fire, are of lime, 

 of sufficient intensity to perform the agglutination, and even 

 the entire fusion, of the carbonate of lime, when its carbonic 

 acid is effectually confined by pressure ; for though lava, after 

 its fusion, may be made, in our experiments, to congeal into a 

 glass, in a temperature of l6° or 18° of Wedgwood, in ^vhich 

 temperature the carbonate would scarcely be affected ; it must 

 be observed, that a similar congelation is not to be looked for 

 in nature ; for the mass, even of the smallest stream of lava, 

 is too great to admit of such rapid cooling. And^ in fact, the 

 external part of a lava is not vitreous, but consists of a sub- Lavas, con- 

 stance which, as my experiments have proved, must have been 22oan(i'28«^^*'* 

 congealed in a heat of melting silver, that is, in 22° of Wedg- 

 wood ; while its internal parts bear a character indicating that 

 they congealed in 27° or 28° of the same scale. It follows, 

 that no part of the lava, while it reiiiained Hquid, can have 

 been less hot than 22" of Wedgwood. Now, this happens to 

 be a heat, in which I have accomplished the entire fusion of 

 the carbonate of lime, under pressure. We must therefore 

 conclude, that the heat of a running lava is always of suffi- 

 cient intensity to perform the fusion of limestone. 



In every active volcano, a communication must exist be- The phenome- 

 tween the summit of the mountain and the unexplored region, volcanoes, 

 far below its base, where the lava has been melted, and 

 whence it has been propelled upwards ; the liquid lava rising 

 through this internal channel, so as to fill the crater to the 

 brim, and flow over it. On this occasion, the sides of the 

 mountain must undergo a violent hydrostatical pressure out- 

 wards, to which they often yield by the formation of a vast 

 retjt, through which the lava is discharged in a lateral erup- 

 tion, and flows in a continued stream sometimes during months. 

 On i^tna most of the eruptions are so performed ; few lavas 

 flowing from the summit, but generally breaking out laterally, 

 at very elevated stations. At the place of delivery, a quantity 

 of gaseous matter is propelled violently upwards, and, along 

 with it, some liquid lava ; which last, falling back again in a 

 spongy state, produces one of those coaical hills which we see 

 * Edinburgh Transactions, vol. v. part I. p. 60—66. 



i& 



