260 J. D. Dana on Pseudomorphism. 



bably due, by a double exchange, to the slow action of 

 nascent sulphate of iron, proceeding from the decomposition 

 of pyrites in the same rock ; but whether heat be required or 

 not, as in the process of the laboratory, we cannot say. 



2. Pseudomorphous changes requiring an Elevated Tempe- 

 rature. 



Magnesian Pseudomorphs. — The magnesian pseudomorphs 

 appear to fall within this division. The magnesian minerals 

 constituting them are talc, including the variety steatite, 

 saponite or hydrous steatite, serpentine and chloi'ite. In the 

 change of felspar to talc, magnesia, by its affinity for silica, 

 expels alumina, and, at the same time, the other elements of 

 felspar that cannot enter into the new magnesian compound 

 pass off. The same takes place in the change of mica, 

 kyanite, augite, &c., to talc. The general theory of the pro- 

 cess corresponds, then, with that already stated at top of 

 page 259. 



Magnesia, either in solution or in the state of vapour, has 

 acted upon the mineral and rocks that have undergone this 

 change. 



The idea of magnesia in vapour is an assumption, as yet 

 unauthorized by chemical science, except as a theoretical 

 possibility. It may be convenient to summon the hypothesis 

 to our aid, for the sake of an explanation ; but, as far as facts 

 have been ascertained, they stand against such a view. 

 Many of the rocks so altered to magnesian contain water / 

 evincing thus that they have not been subjected to that im- 

 measurable degree of heat, which the fancy might call up for 

 producing evaporation of this most fixed and infusible of the 

 earths. 



Magnesia compounds in solution are well known. Sea 

 water is such a solution, and one that has existed, as geolo- 

 gical facts testify, since geological changes commenced on 

 the globe. And are not these pseudomorphs some of the 

 effects I'e suiting through its agency, when under the influence 

 of volcanic heat \ The view, we would here maintain, is the 

 same already presented in a preceding volume of this Jour- 



