R O CK CLA SSIFICA TION 1 1 



these limits in which alkalis are in excess of the alumina, when 

 socialite minerals abound; 1 and others in which alumina is in 

 excess, when muscovite forms. From the foregoing it appears 

 that if differentiation has taken place to produce these magmas 

 there has been no considerable accumulation of alkalis inde- 

 pendently of all other constituents. And from the approach to 

 the pure nephelite molecules by sodium magmas, and from the 

 limitations of the potassium magmas in the neighborhood of the 

 leucite molecule, there is the strongest indication that the differ- 

 entiation of the magmas affected molecules more complex than 

 simple oxides of potassium and sodium, and probably as com- 

 plex as silicates of aluminium and sodium or potassium, corre- 

 sponding to known mineral molecules. However, as already 

 stated, there are proofs deducible from the laws of crystalliza- 

 tion that the molecules in molten magmas are not stable or fixed, 

 and the aluminium and potassium, for example, may enter 

 orthoclase, leucite or biotite according to circumstances in the 

 form of ortho-, meta- or polysilicate. 2 



In the direction of the maximum limit of silica, there are 

 rocks consisting of alkali-feldspars and abundant quartz, which 

 may grade into pure quartz, as suggested by Lehmann 3 and 

 Howitt; 4 certain quartz veins being eruptive and extreme forms 

 of aplitic intrusions. Thus the oxide molecule, Si0 2 appears to 

 be capable of separation by processes of differentiation from 

 other silicate molecules. If analyses had been made from these 

 rocks, the diagrams would have shown a distribution of analyses 

 as far as ioo per cent, of silica. The minimum limit of silica 

 should occur in connection with differentiation products free 



1 Weed, W. H. and Pirsson, L. V., The Bearpaw Mountains of Montana, Am. 

 Jour. Sci., Vol. II, No. 9, 1896, p. 197. 



2 Iddings, J. P., The Origin of Igneous Rocks, Bull. Phil. Soc, Washington, 

 Vol. XII, 1892, p. 176. 



3 Untersuchungen uber die Entstehung der altkrystallinischen Schiefergesteine, 

 etc., Bonn., 1884, p. 55. 



4 Notes on the Area of Intrusive Rocks at Dargo, Trans. Royal Soc, Victoria, 

 1887, Vol. XIII, p. 152. Also Notes on Certain Metamorphic and Plutonic Rocks 

 at Omeo, p. 10. 



