90 GEOLOGY OF THE HIGH PLATEAUS. 



quantity of this constituent is second only to that of sihca, it varies less 

 than any other. It rarely falls below 14 per cent, and rarely exceeds 19 

 per cent, of the entire rock. There is a tendency to a slight excess of 

 alumina above the quantity required to form feldspar in the acid rocks and 

 a tendency to a slight deficiency for the formation of feldspar in the basic 

 rocks.* Hence the slight excess of alumina of the acid rocks may readily 

 be taken up by the aluminous micas and aluminous hornblende ; and in the 

 basic rocks, on account of the deficiency of alumina, the lime cannot all 

 take the form of feldspar, and a considerable portion of it appears in the 

 very abundant augite. 



Thus we find that basic rocks have basic minerals and acid rocks have 

 acid minerals, and that the mineral ingredients stand in correlation to the 

 chemical composition of the magma, and that the nature of the latter is a 

 determinant of the former. Perhaps the most striking example is to be found 

 in the varying conditions which determine the formation of augite and 

 hornblende. These two minerals differ but little in chemical constitution, 

 and yet their slight differences are distinctly correlated to differences in the 

 composition of the magmas from which they crystallize. In augite, lime 

 and iron are found in greater quantity and alumina in less quantity than in 

 hornblende. Although the differences in these respects ai-e rather small, 

 they appear to be strictly proportional to correlative differences in the gen- 

 eral groundmass in which they respectively occur. 



Correlation between chemical composition and specific gravity. — The exist- 

 ence of such a correlation is perhaps too well known and too obvious to 

 require any discussion. In general the density holds an inverse ratio to 

 the acidity. 



Correlation between the chemical composition and fasihility. — The fusibility 

 of volcanic rocks has not been investigated so fully as other pi-operties, and 

 neither lithologists nor geologists appear to have attached any very great 



*Tho percentage of alumina, however, is less in the acid than in the basic rocks, and yet the 

 excess above the quantity required to form soda and potash feldspars is usually greater in the former 

 rocks than in the latter, on account of the great acidity of the alkali feldspars ; indeed, there is rarely 

 any notable excess of alumina in the basic rocks above what is required for the basic lime-feldspar. 

 Thus the rocks which have the smaller percentage of alumina curiously enough have an excess above 

 the ro<iuirenients of feldspar, and it appears in the accessory minerals, while the rocks which have the 

 higher percentage are rather deficient in it. 



