﻿498 J)E. WHITMAN CROSS ON THE NATURAL [Aug. I9IO, 



of representing a chemical analysis.' The authors consider it 

 4 chemico -mineralogical ' and as, in fact, an indispensable interpreta- 

 tion or expression of the purely chemical. Mr. Harker says that 

 the Quantitative System is based 



' upon a hypothetical mineralogical composition which is not in general that 

 of the actual rock.' (Op. cit. p. 363.) 



The norm is otherwise called * ideal,' and ' a certain artificially 

 selected " standard"'* list ' of minerals. He notes that the complex 

 micas, garnets, aluminous hornblendes and augites, melilite, spinel, 

 and others, are not in the norm, 



' while in their place we find kaliophilite, sodium and potassium metasilicates, 

 wollastonite, and akermauite, — compounds which are foreign to igneous rocks, 

 and some of which are not known in Nature.' (Op. cit. p. 365.) 



Nowhere do I find recognition of the plain fact that the norm 

 is almost wholly made up of substances which 

 Mr. Harker and many others believe to be the prin- 

 cipal simple molecules of the magmatic solution. Nor 

 is the repeated statement of the authors of the Quantitative 

 System that it is essentially a magmatic classification, 

 so far as the new divisions go, emphasized as it should be. 



I wish to return to the discussion of the norm in connexion 

 with the third desideratum of an ' ideal system ' presented by 

 Mr. Harker, which it seems well to quote in full : — 



1 III. Since rock-magmas are mixtures of minerals, and the variation met 

 with in igneous rocks is a variation in the associations and relative proportions 

 of minerals, it follows that a natural classification must be in its expression a 

 mineralogical one, not a chemical. It is true that the actual mineralogical 

 constitution of a given magma may vary to some extent in accordance with 

 temperature and other conditions, and this would seem to import some com- 

 plication where successive differentiations have been effected at higher and 

 lower temperatures. If it should be found necessary to represent complex 

 low-temperature minerals by simpler compounds, these must at least be 

 compounds which actually exist in the rock-magmas.' (Op. cit. pp. 376-77.) 



The first sentence of this statement appears to recognize that 

 one element in an ideal system is an expression of the relative 

 proportions and associations of the mineral molecules in magmatic 

 solution. The second sentence recognizes variation in molecular 

 constitution in accordance with conditions, and the last one that 

 complex molecules may be necessarily represented by simpler ones 

 in any statement. 



In his discussion of rock-magmas {op. cit. chap, vii, p. 169), 

 Mr. Harker states that 



' the constituents of a molten rock-magma exist there in the form of definite 

 compounds, mostly silicates, and in general identical with those compounds 

 which are familiar in the less complex rock-forming minerals.' 



I find direct mention of orthoclase, albite, anorthite, leucite, 

 nepheline, quartz (tridymite), and corundum, and must infer that 



