A. Holmes — Classification of Igneous Rocks. 117 



It also seems to the present writer that Order 5, embracing as it does 

 rocks with normative quartz and felspar, felspar alone, or felspar and 

 felspathoid, is of too broad a character. It traverses one of the few 

 natural lines of distinction available for classification, the line between 

 the antipathetic minerals quartz and felspathoid. Certainly the 

 norm indicates which of these is present, but in the symbol it might 

 advantageously be expressed by suffixes such as the following : — 

 5g (accessory normative quartz present), 

 5 (no quartz nor felspathoid present ; in this case normative olivine would 



generally be present), 

 5; (accessory felspathoid, or lenad, present). 



The division of orders (of Classes I to III) into Rangs is based 

 on the ratio of the molecular proportions of alkalies to lime, only 

 those entering into the salic minerals being employed for comparison. 

 Mineralogically, this is practically the ratio of orthoclase plus 

 albite to anorthite, it being understood that felspathoids, where 

 present, are to be expressed in terms of the amounts of orthoclase 

 and albite to which they are equivalent. The ratio, however, tells 

 us very little about the rock to which it is applied beyond expressing 

 in a general way its alkalic or calcic character. It is suggested below 

 that a more useful method of division would be afforded by the ratio 

 of the molecular proportions of soda to lime in the felspars, or better, 

 of the direct ratio by weight of albite to anorthite. Such a ratio 

 Avould express the normative soda-lime felspar, and in many cases the 

 latter would not materially differ from the actual soda-lime felspar 

 of the rock. 



The division of rangs into Sub -rangs is based on the molecular 

 ratio of salic potash to salic soda, and is useful in distinguishing 

 rocks with minerals like orthoclase, muscovite, leucite, and biotite, 

 from those containing albite, analcime, nepheline, etc. 



It has frequently been a matter for surprise that the sub-rangs of 



the C.I.P.W. Classification (of which in the first three classes there 



are 675) do not always contain rocks of closely similar types, and that 



adjacent sub-rangs may sometimes enclose rocks that are much more 



closely related to one another than they may be to other rocks falling 



within their own sub-rang. The latter possibility is illustrated by 



the following analyses, taken from Iddings' Igneous Rocks, vol. ii : — 



A. B. C. D. E. 



Si0 2 . 75-04 75-17 74-37 73-05 64-57 



A1 2 3 . 13-12 12-66 13-12 14-67 16-80 



Fe 2 3 . 2-12 0-23 0-73 0-89 0-97 



FeO . n. d. 1-40 0-87 n. d. 3-02 



MgO . 0-34 0-05 0-35 0-26 1-69 



CaO . 0-40 0-82 1-26 0-97 3-53 



Na 2 . 2-44 2-88 2-57 3-99 3-81 



K 2 . 6-32 5-75 6-09 5-11 4-01 



H 2 . 0-76 0-82 0-30 0-91 1-28 



Inch . — 0-47 0-45 — — 



Total . 100-54 100-26 100-11 99-85 99-68 



Symbol : 1,4,1,2 1,4,1,3 1,4,2,3 1,4,2,3 1,4,2,3 



Omeose. Liparose. Toscanose. Toscanose. Toscanose. 



