200 ON THE ORIGIN, MATERIALS, COMPOSITION, 



granite on the other, apply a common mathematical 

 axiom to the conclusion. 



If it be said that volcanoes do not produce perfect 

 granite, it must still be recollected that they produce 

 compounds of an analogous nature in every respect. 

 Fanjas indeed had said that lavas never contained 

 (juartz, but Breislak has produced numerous instances 

 of this ; while Dolomieu^mentions quartz, felspar, and 

 mica, as forming the white lavas of Ischia, and de- 

 scribes some of them as being " almost granitic." It 

 was also shown that the trap rocks often assumed the 

 characters of perfect granite ; so that, by this inter- 

 mediate step, the several products which are most 

 distant are again associated. Even admitting that the 

 volcanic rocks stood exclusively at one extremity of a 

 scale of chemical compounds, and the granites at the 

 other, the trap rocks, containing examples of both, 

 form the common link by which they are united. 

 This view of the chemical origin of granite is con- 

 firmed by the same set of appearances which confirm 

 it in the case of the trap family, and which are fully 

 described in other parts of this work. 



It is not difficult to assign probable reasons for the 

 differences in the chemical appearances of the rocks 

 in these three distant productions. They have how- 

 ever already been sufficiently pointed out; and it was 

 shown that they probably consisted, in a great mea- 

 sure, in differences of the time through which the 

 fused materials had cooled: circumstances confirmed 

 by a great number of collateral appearances already 

 mentioned; though in many cases, there can be no 

 doubt that great differences have resulted from the 

 different proportions of the several earths in the fused 

 compounds. 



It is unnecessary to repeat, that the production of 



