418 On Some Points in Chemical Geology. 



claseand quartz, and the pyroxenic rocks of pyroxene and labradorite, 

 or a feldspar approaching it in composition, it is evident that the 

 calculations of Bunscn will to a certain extent hold good ; but in the 

 analyses, by Dr. Strong, of the volcanic rocks of Hungary and 

 Armenia, we often find that the actual proportions of alkalies, lime, 

 and magnesia vary considerably from those deduced from calculation. 

 This will necessarily follow when feldspars like albite or anorthite 

 replace the labradorite in pyroxenic rocks. The phonolites are 

 moreover highly basic rocks, which contain but very small amounts 

 of lime, magnesia, or iron, being essentially mixtures of orthoclase 

 with hydrous silicates of alumina and alkalies. 



IV. In a recent enquiry into the probable chemical conditions of 

 a cooling globe like our earth, I have endeavoured to show that in the 

 primitive crust all the alkalies, lime and magnesia must have existed 

 in combination with silica and alumina, forming a mixture which 

 perhaps resembled dolerite, while the very dense atmosphere would 

 contain in the form of acid gases, all the carbon, chlorine and sul- 

 phur, with an excess of oxygen, nitrogen and watery vapour. The 

 first action of a hot acid rain, falling upon the yet uncooled crust, 

 would give rise to chlorids and sulphates, with separation of silica ; 

 and the accumulation of the atmospheric waters would form a sea 

 charged with salts of soda, lime, and magnesia. The subsequent 

 decomposition of the exposed portions of the crust, under the influ- 

 ence of water and carbonic acid, would transform the felspathic 

 portions into a silicate of alumina (clay) on the one hand, and alkaline 

 bicarbonates on the other ; these, decomposing the lime-salts of the 

 sea, would give rise to alkaline chlorids and bicarbonate of lime — 

 the latter to be separated by precipitation, or by organic agency, as 

 limestone. In this way we may form an idea of the generation 

 from a primitive homogeneous mass, of the siliceous, calcareous and 

 argillaceous elements which make up the earth's crust, while the 

 source of the vast amount of carbonate of lime in nature is also 

 explained.* 



When we examine the waters charged with saline matters which 

 impregnate the great mass of calcareous strata constituting in Canada 

 the base of the Silurian system, we find that only about one-half of 

 the chlorine is combined with sodium ; the remainder exists as chlor- 

 ids of calcium and magnesium, the former predominating, — while 

 sulphates are present only in small amount. If now we compare this 

 composition, which may be regarded as representing that of the 



* Am. Jour. Sci. (2) xxv. 102, and Canadian Journal for May 1858, 



