IV.] THE CHEMISTRY OF THE PRIMEVAL EARTH. 43 



always now found associated with it ; but we have indirect 

 evidence that it was formed and subsequently carried away, in 

 the case of many gypsum deposits, whose thickness indicates a 

 long continuance of the process under conditions much more 

 perfect and complete than we can attain under our present 

 atmosphere. While studying this reaction I was led to inquire 

 whether the carbonic acid of the earlier periods might not have 

 favored the formation of gypsum ; % and I found, by repeating 

 the experiments in an artificial atmosphere impregnated with 

 carbonic acid, that such was really the case.* We may thence 

 conclude that the peculiar composition of the primeval atmos- 

 phere was the essential condition under which the great deposits 

 of gypsum, generally associated with niagnesian limestones, 

 were formed. 



The reactions of the atmosphere, which we have considered, 

 would have the effect of breaking down and disintegrating the 

 surface of the primeval globe, covering it everywhere with beds 

 of stratified rock of mechanical or of chemical origin. These 

 now so deeply cover the partially cooled surface that the amount 

 of heat escaping from below is inconsiderable, although in 

 earlier times it was very much greater, and the increase of tem- 

 perature met with in descending into the earth must then have 

 been many times more rapid than now. The effect of this 

 heat upon the buried sediments would be to soften them, pro- 

 ducing new chemical reactions between their elements, and 

 converting them into what are known as crystalline or meta- 

 morphic rocks, such as gneiss, greenstone, granite, etc. We are 

 often told that granite is the primitive rock or substratum of the 

 earth ; but this is not only unproved, but extremely improbable. 

 As I endeavored to show in the early part of this discourse, 

 the composition of this primitive rock, now everywhere hidden, 

 must have been very much like that of a slag or lava ; and 

 there are excellent chemical reasons for maintaining that granite 

 is in every case a rock of sedimentary origin, that is to say, 

 it is made up of materials which were deposited from water, 

 like beds of modern sand and gravel, and includes in its com- 

 * See Paper VIII. 



