356 METAMORPHOSIS OF SOME SEDIMENTABT BOCKS. 



exposed to the air, the former hase alone is at first deposited, and the 

 magnesian carbonate is only separated by evaporation. "When car- 

 bonate of soda is added in small quantities to a liquid, such as sea 

 water, containing chlorides. of these two bases, the precipitate formed in 

 the cold consists chiefly of carbonate of lime, and the liquid, by evapora- 

 tion, deposits a large quantity of carbonate of magnesia with a little 

 lime, and then contains only chloride of magnesia (with soda salts) in 

 solution. It is well known that the precipitate formed by carbonate 

 of soda in a solution of chloride of magnesium is soluble in an excess 

 of either of these salts. We have in Canada a great number of saline 

 springs, which rise from Lower Silurian rocks, and appear to be formed 

 by the mingling of the bittern-like waters, destitute of earthy carbon- 

 ates, and derived from the lower limestones, with the carbonated alka- 

 line waters of some of the associated strata. These saline springs, 

 such as Plantagenet, St. Leon, and Sainte-Genevieve, deposit by eva- 

 poration, at a gentle heat, large quantities of earthy carbonates, of 

 which the carbonate of magnesia forms from 50 to 95 per cent. ; they 

 sometimes contain but a trace of carbonate of lime. The spontane- 

 ous evaporation of basins of similar waters would give rise to the 

 formation of dolomites or magnesites which would assume the form of 

 detached or interrupted beds or lenticular masses among the pure 

 limestones and other non-magnesian deposits of the region. Such 

 are precisely the conditions in which the magnesian rocks occur in the 

 Hudson River Group. Many of them may, however, be the result of 

 a direct precipitation which may take^place in deep sea water, from the 

 infusion of alkaline carbonates. 



Mingled as these magnesian deposits naturally are with sand and 

 clay, we have in the silica, magnesia, lime, alumina, and oxide of iron 

 of the sediment, the elements of serpentine, talc, pyroxene, hornblende 

 and chlorite. For the production of these minerals it is necessary to 

 dissolve the silica, and cause it to unite with the bases present, expel- 

 ling the carbonic acid. The agent in this reaction has doubtless been 

 an alkali. A solution of carbonate^of soda at 212° F. will slowly dis- 

 solve silica, even in the form of quartz, and the silicate of soda thus 

 formed, is at once decomposed at this temperature by the carbonates 

 of lime, magnesia, or iron, with the production of a silicate of these 

 bases, and the regeneration of the alkaline carbonate, which is then 

 free to operate upon a new portion"fof silica. In this way a small 

 amount of alkali may serve as tbe medium'for the silicification of a large 

 amount of carbonates. I have verified all these reactions by experi- 

 ment, and have found that a silicate of magnesia is formed when quartz 

 is boiled with carbonate of magnesia, and a solution of carbonate of soda. 



