VIII.] CHEMISTRY OF DOLOMITES AND GYPSUMS. 89 



chemical precipitate of nearly pure carbonate of lime, in which 

 are often imbedded silicified shells and corals.* It is not per- 

 haps easy in all cases to distinguish between such precipitates, 

 which may assume a concretionary structure (see on this ques- 

 tion Bischof, Chemical Geology, I. 428), and those deposits 

 which, like travertines, have been formed from subterranean 

 springs. In neither case, however, should they be confounded 

 with the tufaceous limestones mentioned above. 



The union of the mingled carbonates of lime and magnesia 

 to form dolomite is attended with contraction, which, in case 

 the sediment was already somewhat consolidated, would give 

 rise to fissures and cavities in the mass. Should the dolomitic 

 strata be afterwards exposed to the action of infiltrating carbon- 

 ated waters, the excess of carbonate of lime and any calcareous 

 fossils would be removed, leaving the mass still more porous, 

 and with only the moulds of the fossils. Insoluble, however, 



* The large proportion of dissolved silica which many river-waters contain 

 appears in sedimentary deposits, not only replacing fossils and fanning 

 concretions and even beds of flint, chert, and jasper, but also in a crystal- 

 line state, as is seen in the crystallized quartz often associated with these 

 amorphous varieties, and in some beds of sandstone which are made up en- 

 tirely of small crystals of quartz. Elie de Beaumont long since called atten- 

 tion to the crystalline nature of certain sandstones which, as Daubree has 

 remarked, could not have been derived from the disintegration of any known 

 rock ; and Mr. J. Brainard, at the meeting of the American Association for 

 the Advancement of Science, held at Cleveland in 1860, insisted upon the 

 crystalline character of the grains composing certain sandstones in Ohio, as 

 evidence that these were chemical deposits. He however fell into the error of 

 supposing that all sandstones and even quartzose conglomerates have had a 

 like origin, while the latter and the greater part of the former are undoubtedly 

 mechanical deposits from the ruins of pre-existing quartzose and granitic 

 rocks. 



These crystallized sands, according to Daubree, are met with in beds in the 

 sandstone of the Vosges, the variegated sandstone (Triassic and Permian), 

 and in the tertiary of the Paris basin and elsewhere. Other sands are made 

 up of globules of chalcedony, apparently, like the crystallized sands, a chemical 

 deposit, and associated with oolitic iron ores in the lias, and with glauconite 

 grains in the green-sand. (Daubree, Recherches sur le Striage cles Roches, 

 etc., Ann. des Mines, 1857.) We may here mention the so-called gaize from 

 the green-sand of the Ardennes, which gave to Sauvage 56.0 p. e. of amor- 

 phous soluble silica mixed with quartz-sand and glauconite. (Bischof, Lehr- 

 buch, I. 768-811.) 



