306 ORIGIN OF CRYSTALLINE ROCKS. [XIII. 



At the early periods in which the materials of the ancient 

 crystalline schists were accumulated, it cannot be doubted that 

 the chemical processes which generated silicates were much 

 more active than in more recent times. The heat of the earth's 

 crust was probably then far greater than at present, while a 

 high temperature prevailed at comparatively small depths, and 

 thermal waters abounded. A denser atmosphere, charged with 

 carbonic-acid gas, must also have contributed to maintain, at 

 the earth's surface, a greater degree of heat, though one not 

 incompatible with the existence of organic life. (Ante, page 

 46.) These conditions must have favored many chemical 

 processes, which, in later times, have nearly ceased to operate. 

 Hence we find that subsequently to the eozoic times, silicated 

 rocks of clearly marked chemical origin are comparatively 

 rare. In the mechanical sediments of later periods certain 

 crystalline minerals may be developed by a process of mo- 

 lecular rearrangement, di agenesis. These are, in the f eld- 

 spathic and aluminous sediments, orthoclase, muscovite, garnet, 

 staurolite, cyanite, and chiastolite, and in the more basic sedi- 

 ments, hornblendic minerals. It is possible that these latter 

 and similar silicates may sometimes be generated by reactions 

 between silica on the one hand and carbonates and oxides on 

 the other, as already pointed out in some cases of local altera- 

 tion. Such a case may apply to more or less hornblendic 

 gneisses, for example, but no sediments, not of direct chemical 

 origin, are pure enough to have given rise to the great beds of 

 serpentine, pyroxene, steatite, labradorite, etc., which abound 

 in the ancient crystalline schists. Thus, while the materials 



were once pastes, magmas or muds, and so were the primitive rocks at flu- 

 time of their origin, but during these first ages of the earth the consolidating 

 and crystallizing forces (differing in degree only from those of tin- ; 

 time, and aided by a higher temperature) allowed the magma to assume the 

 form of mineral species, more or less distinct. If we choose to call this 

 change metamorphism, then the rocks thus formed are metamorphic; but so 

 also are the limestones of later periods. The primitive rocks originated l>y 

 way of sedimentation, the one after the other, constituting distinct forma- 

 tions, and there are no eruptive gneisses." See, in this connection, the Intro- 

 duction to Essay III. of the present volume, and the statements of Favre in 

 the Appendix to Essay XIV.] 



