XIII. ] ORIGIN OF CRYSTALLINE ROCKS. 307 



for producing, by diagenesis, the aluminous silicates just men- 

 tioned are to be niet with, in the niud and clay-rocks of all 

 ages, the chemically formed silicates, capable of crystallizing 

 into pyroxene, talc, serpentine, etc., have only been formed 

 under special conditions. [While the generation of various 

 crystalline silicated minerals in rocks since the Eozoic age is 

 theoretically not impossible, the accumulation of evidence goes 

 to show that although such changes have taken place locally in 

 the proximity of eruptive rocks, and by the invasion of thermal 

 waters, there has been no wide-spread alteration or regional 

 metamorphism, as it has been called, of these more recent 

 sedimentary deposits.] 



The same reasoning which led me to maintain the theory of 

 an original formation of the mineral silicates of the crystalline 

 schists, induced me to question the received notion of the epi- 

 genic origin of gypsums and magnesian limestones or dolomites. 

 The interstratification of dolomites and pure limestones, and 

 the enclosure of pebbles of the latter in a paste of crystalline 

 dolomite, are of themselves sufficient to show that in these 

 cases, at least, dolomites have not been formed by the altera- 

 tion of pure limestones. The first results of a very long series 

 of experiments and inquiries into the history of gypsum were 

 published by me in 1859, and further researches, reiterating 

 and confirming my previous conclusions, appeared in 1866. 

 (Ante, page 80.) In these two papers it will, I think, be 

 found that the following facts in the history of dolomite are 

 established : namely, first, its origin in nature by direct sedi- 

 mentation, and not by the alteration of non-magnesian lime- 

 stones ; second, its artificial production by the direct union of 

 carbonate of lime and hydrous carbonate of magnesia, at a gen- 

 tle heat, in the presence of water. As to the sources of the 

 hydrous magnesian carbonate, I have endeavored to show that 

 it is formed from the magnesian chloride or sulphate of the sea 

 or other saline waters in two ways : first, by the action of the 

 bicarbonate of soda found in many natural waters ; this, after 

 converting all soluble lime-salts into insoluble carbonate, forms 

 a comparatively soluble bicarbonate of magnesia, from which a 



