460 ORIGIN OF ERUPTIVE AND PRIMARY ROCKS. 



•cently cooled, to allow the condensation of some of the constituents of 

 this primitive atmosphere, the action of these on the earth's crust must 

 have been very energetic, and must have caused the formation 

 ■of products differing considerably from the sedimentary deposits of 

 later periods. We shall return to this subject, when adverting to 

 the rocks of the so-called Primitive Slate formation. 



With regard to the fluid part of the original globe, we have 

 seen that it must have been made up, with but little exception, of 

 the inorganic constituents of the earth's crust. It is evident, 

 that in this fluid globe the heavier particles must have found 

 their way to the centre, and that then, as now, the interior of the 

 globe must have had a greater density than its surface. Indeed, 

 the fact that this is the case at the present day is another proof that 

 the globe must have been originally in a state of igneous fluidity, 

 otherwise we could not account for the accumulation of the den- 

 ser particles at the centre. In the same way as the densest par- 

 ticles were influenced by gravitation, so must also the fused sili- 

 cates of different densities, and the metallic sulphurets and arsen- 

 iurets have found their places in successive concentric zones, one 

 beneath the other, according to their increasing specific gravities. 

 Thus the theory of Sartorius von Waltershausen would appear to 

 be as fully applicable when the earth was in a fluid state, as at 

 the present time. 



There is nothing unreasonable or inconsistent with the observa- 

 tions which we are able to make at the present day, in supposing 

 the inorganic constituents of the earth to have once been in a state 

 of igneous fusion. The various layers of fused material, to judge 

 from the rocks resulting from their solidification, must have close- 

 resembled in chemical composition the scoriae produced in differ- 

 ent blast-furnaces. If we suppose the uppermost highly silicified 

 and consequently most difficultly fusible layer to be represented 

 by granite, we find many instances of slags from iron-furnaces 

 having almost as acid a composition. Many granites contain only 

 63 per cent, of silica, but those of the Hartz as high as 73.* On the 

 other hand there are instances of iron-slags containing 70 and 71 

 per cent.f silica. So far as the other more basic layers and the 

 rocks resulting from them are concerned, we can find their equi- 

 valents among the slags of iron, copper and lead furnaces, since 

 the silica contents of the latter range from 70 through every per- 



* Bischof : Chemical and Physical Geology, III. p. 414. 

 f Kerl : Handbuch der Huttenkunde, I. p, 323. 



