GEOLOGY. 309 



origin.* In this way the deposits of emery, which is a mixture of crystal- 

 lized alumina with oxide of iron, have doubtless been formed. 



Waters deficient in organic matters may remove soda, lime, and magnesia, 

 from sediments, and leave the granitic elements mingled Avith oxide of iron; 

 while, on the other hand, by the admixture of organic materials, the whole 

 of the iron may be removed from strata which will retain the lime and soda 

 necessary for the formation of basic feldspars. The fact that bicarbonate of 

 magnesia is much more soluble than bicarbonate of lime, is also to be taken 

 into account in considering these reactions. 



The study of the chemistry of mineral waters, in connection with that of 

 sedimentary rocks, shows us that the result of processes continually going 

 on in nature is to divide the silico-argillaceous rocks into two great classes, 

 the one characterized by an excess of silica, by the predominance of pot- 

 ash, and by the small amounts of lime, magnesia, and soda, and represented 

 by the granites and trachytes, while in the other class silica and potash are 

 less abundant, and soda, lime, and magnesia, prevail, giving rise to pyrox- 

 enes and triclinic feldspars. The metamorphism and displacement of sedi- 

 ments may thus enable us to explain the origin of the different varieties of 

 plutonic rocks without calling to our aid the ejections of the central fire. 



YI. The most ancient sediments, like those of modem times, Avere doubt- 

 lessly composed of sands, clays, and limestones, although from the prin- 

 ciples already defined in IV. and V., it is evident that the chemical com- 

 position of these sediments in different geologic periods must have been 

 gradually changing. It is from a too hasty generalization that an eminent ge- 

 ologist has concluded that limestones Avere rare in earlier times, for in Canada 

 the Laurcntian system an immense series of stratified crystalline rocks, 

 Avhich underlie unconformably both the Silurian and the old Cambrian or 

 Huronian systems contains a limestone formation (interstratified with dol- 

 omites), the thickness of which Sir \V. E. Logan has estimated at not less 

 than 1000 feet. Associated with this, besides great volumes of quartzite 

 and gneiss, there is a formation of vast but unknown thickness, the predom- 

 inant element of Avhich is a triclinic feldspar, varying in composition bctAveen 

 anorthite and andesine, and containing lime and much soda, with but a 

 small proportion of potash. These feldspars are often mixed with hypers- 

 thene, or proxene ; but great masses of the rock are sometimes nearly pure 

 feldspar. These fcldspathic rocks, as well as the limestones, are associated 

 with beds of hematitic and magnetic iron-ores, the latter often mixed with 

 graphite. Ancient as are these Laurentian rocks, AVC have no reason to sup- 

 pose that they mark the commencement of sedimentary deposits; they Avere 

 doubtlessly derived from the ruins of other rocks, in which the proportion of 

 soda was still greater; and the detritus of these Laurentian feldspars, mak- 

 ing up our paleozoic strata, is now the source of alkaline Avaters, by which 

 the soda of the silicates, rendered soluble, is carried down to the sea in the 

 form of carbonate, to be transformed into chloride of sodium. The lime of 

 the feldspars being at the same time removed as carbonate, these sediment- 

 ary strata, in the course of ages, become less basic, poorer in soda and lime, 

 and comparatively richer in alumina, silica, and potash. Hence, in more 



* Hydrated alumina, in the form of gibbsite, is however met with in incrusting 

 Jimonite, and the existence of compounds like pigotite, in which alumina is united 

 AA-ith an organic substance allied to crenic acid, seems to show that this base inay, 

 under certain conditions, be taken into solution by organic acids. 



