temperature of the interior of the earth. 125 



ries and a half. It is clear that, in all cases, a very minute ac- 

 tion suffices to produce the phenomena. 



" it will be remarked that this action, if it be real, is necessa- 

 rily connected with the whole contraction which the g^lobe un- 

 dergoes from the effect of secular cooling. It furnishes a basis 

 for calculating the very weak influence which this total contrac- 

 tion exercises in accelerating the velocity of rotation. 



" JVothing less than this enormous power which I have de- 

 scribed, is required to raise lava. In the particular case where 

 lavas come from a depth of twenty leagues, it is easy to prove 

 from their mean specific gravity, that they would be pressed with 

 a force equal to twenty-eight thousand atmospheres. We know 

 moreover that they overflow after an eruption of gaseous mat- 

 ters, which may well be the case on my theory of the subject. 



" This is not the place to develope the hypothesis purely ther- 

 mometric, which I propose in explanation of volcanic phenom- 

 ena ; and to shew how well it applies to all their details. It will 

 suflice to observe that it assigns a reason for the identity of the 

 circumstances which characterize volcanic phenomena every 

 where, of the prodigious reduction in the number of volcanos 

 since the origin of things, of the diminution in the quantity of 

 ejected matters at each eruption, their nearly similar composi- 

 tion at each geological epocha, and the small difference that ap- 

 pears between ejected lavas of different epochs. In short, in this 

 hypothesis, the most usual direction of earthquakes announces the 

 zones where the crust of the earth is thinnest; and the volcanic 

 centers, ancient and modern, point out the thinnest and least re- 

 sisting portions of this crust. 



"In my preceding remarks, I have left uncalculated the gase- 

 ous matters which are produced at each eruption; for supposing 

 them reduced to their primitive state of liquidity while in the 

 mixture from whence they have been disengaged, they would 

 occupy but little bulk ; and the medium I have adopted, of a cu- 

 bic kilometre, is much beyond the actual volume of ejected lava." 



The suppositions of M. Cordier in this paragraph seem too 

 gratuitous. We have as yet nothing that approaches to 

 proof of the primitive formations (Hmestone, talcose rocks, 

 clay slate, mica schist, hornblende rock, the gneiss to which 

 all these seem subordinate, and the older small grained gran- 

 ite whereon the gneiss usually reposes,) being from forty to 

 sixty miles thick. No accurate measure of their edges and 

 angles has yet been taken. As yet, all is conjecture. That 

 a lava should require for its ejection the force of twenty- 

 eight thousand atmospheres, is neither probable from any 



