ON THE FOEMATION OF CRYSTALLINE ROCKS. 269 



Btnicture, or, in other words, the himellar structure, which does not 

 owe its origin to stratification by deposition, although it is of frequent 

 occurrence in the primitive formations, is not always found there, nor 

 is it peculiar to them. On the one hand, we do not find real phyl- 

 lades in the silurian formations of Sweden, of Russia, or of the United 

 States, which have remained horizontal, and which have been men- 

 tioned above as in general not being metamorphosed. On the other 

 hand, the schists suitable for being quarried as slates are found in the 

 more recent formations which have been dislocated, as in the creta- 

 ceous formation of the Pyrenees and of Terra del Fuego,* and in the 

 nummulite formation of Switzerland, in the environs of Glarus. Thus 

 the origin of the lamellar structure, as of the metamorphic state, ap- 

 pears to be essentially connected with the existence of dislocations. 



CHAPTER IV. 



DOLOMITE, GYPSUM, ROCK-SALT, SULPHUR, AND BITUMINOUS DEPOSITS IN 

 THEIR RELATIONS WITH METAMORPHISM. ! 



We know that certain dolomites result from the transformation of 

 limestone.t This epigeny may be explained, as the synthetical experi- 

 ments which have been produced upon this subject tend to show, by 

 the action of combinations of magnesia on carbonate of lime. There 

 is, however, nothing to prove that this transformation into dolomite 

 has always been produced by the same agents, and that the dolomite 

 of Campo-Longo, for instance, with its tourmalines, its corundums, 

 and its various minerals, is to be assimilated with the dolomite of the 

 other parts of the Alps| and of Nice, or those which are near the de- 

 posits of calamine in Belgium. 



But there are dolomites, and this is the case with the greatest 

 number, situated in regular beds, which are often horizontal, consti- 

 tuting very extensive geognostic formations. When they contain 

 remains of testaceous molusca the shell has disappeared ; they are 

 often crystalline and riddled with holes in such a way as to suggest 

 a substitution. It is possible that the principal part of these last 

 dolomites was directly precipitated. § But on account of the disap- 

 pearance of the shells we must admit, with Elie de Beaumont, 

 that this second case allies itself with the first, by the reaction which 

 the medium in which the precipitation has taken place, has exerted 

 on the matter precipitated, a reaction of such kind that the carbonate 



• According to Mr. Darwin. 



•j- To tiie simple exposition of facts I will add in these three chapters a few theoretical 

 consideration.s. This will exempt me from recurring to these phenomena in the third part. 



J See especially the memoirs of Studer and his Fhysikalische Geographie, vol. i, p. 14G. 



§ Annales de Chiinie el de Physique, vol. xxviii, p. 710 ; Erdmann Jahrb. far pr(ju:t. Chem., vol. 

 ilix, p. 52, 1850. 



