302 THE EVOLUTIOlSr OF PETROLOGICAL IDEAS, 



solid porouy mass. Such a separation might be eli'ected in the case of 

 a plutonic mass, if a partially solidified magma were subjected to pres- 

 sure under conditions which admitted of the escape of the still liquid 

 portions into the surrounding rocks. As a matter of fact, it has been 

 so applied by Mr. Barrow, who thus explains the relation between 

 pegmatites and certain oligoclase-biotite-gneisses in the southern High- 

 lands of Scotland. The eurite veins in granite are generally supposed 

 to owe their origin to a somewhat similar action, but in this case the 

 separation is due to the leaching out of the still liquid eutectic into 

 cracks in the nearly consolidated mass, and not to orogenetic move- 

 ments. It is comparable, therefore, to the liquation process above 

 mentioned. 



Bunsen explained the varieties of igneous rock revealed by his 

 analyses b}" assuming the independent existence of two magmas — the 

 "normal pyroxenic'' and " normal trachytic'' — and supposing' a process 

 of intermixture to account for the intermediate varieties. Von Walter- 

 shausen thought that igneous magmas were arranged in a series of 

 concentric shells, according to specific gravity. Durocher, in his cel- 

 ebrated essay on (Comparative Petrolog}', maintained "that all igneous 

 rocks, modern and ancient, were derived from two magmas which 

 coexist l)elow the solid crust of the globe, and occup}' there each a 

 definite position."" His two magmas — basic and acid — do not difler 

 materially from those of Bunsen, and his idea of their arrangement in 

 the earth's crust is practically the same as that of Von Waltershausen. 

 He compared the tAvo magmas to baths of fused metals, which separate 

 into distinct alloys on cooling. He does not give actual illustrations, 

 but we ma}^ consider one, in order to give precision to the idea. A 

 mixture of 48.6-1- per cent of bismuth and ,50.86 per cent of zinc sepa- 

 rates at a temperature between TOO- and 800" C. into two alloys, which 

 arrange themselves according to specific gravity. On cooling, the 

 heavier is found to contain S-i.S'J per cent of bisnuith and 15.18 per 

 cent of zinc; the lighter 2.47 per cent of bisnuith and l»7.o3 per cent of 

 zinc. If silver be added to the mixture, there is also a separation into 

 two alloys, so long as the amount of silver is less than about 40 per 

 cent; when it exceeds this amount, there is no longer any separation. 



Durocher speaks of eruptions which derive their supply from the 

 primary magmas as belonging to the first order, and those which draw 

 their material from more or less isolated magma basins as belonging 

 to the second order. The latter furnish rocks which depart from the 

 normal type, and this he explains, in part at least. l)v assuming a 

 process of separation analogous to that by wiiich the j)rimary magmas 

 were produced. Thus he says: 



"It is therefore probable that phonoliticand trachytic porphyry are 

 only the two opi)Osite products of a liquation which took place in 

 the midst of the fluid mass; they are, as it were, the two inverse alloys 

 into which we so often see a metallic batii divide itself,'' 



