184 N. L. Bowen — Crystallization of Haplobasaltic, 



this is some frozen aqueous solution. The presence of volatile 

 components and the existence of various equilibria between 

 them and silicates removes the necessity for the existence of 

 an end-point similar to that in the ' haplo' series. 



There appears then to be no room for reasonable doubt 

 that the differentiation of the sub-alkaline series of igneous 

 rocks is controlled entirely by crystallization. Moreover, the 

 systematic decrease in the amount of colored constituent with 

 increasing alkalinity of the feldspar indicates that the more 

 acid types are not original magmas,* for in this case there 

 would be no reason for such a balance in the proportion of 

 these contrasted constituents, whereas this finds a natural ex- 

 planation if the more acid magmas are regarded as derived 

 from basic material, being, as it were, successive mother liquors 

 from the crystallization of the basic magmas. 



In this connection it seems worth while to point out the 

 considerable degree of similarity in composition (when the 

 absence of iron and potash and the fact that other constituents, 

 principally lime, must make up this shortage are taken into 

 account) between a haplodiorite magma from the investigated 

 system and the average diorite as computed by Daly (Table VI). 

 This haplodiorite is not any random one but is one which could 

 be deprived as a mother liquor from the crystallization of 

 haplobasaltic magma, f The similarity is suggestive in con- 

 nection with the idea that diorites are related to basaltic 

 magma in the same way. 



J. H. L. Vogt believes that differentiation is accomplished in 

 the liquid state,:]: but when he proceeds to discuss the problem 

 he bases the discussion entirely on the progress of crystalliza- 

 tion. Indeed, Yogt's paper, though professedly the opposite, 

 is one of the best essays we have in favor of crystallization- 

 differentiation. 



mols. 





f 



rABLE VI. 







I 



mols. 



II 



Si0 2 



57-56 





57-8 



TiO, 



■85 







aia 



16-90 





17-0 



Fe,0 3 



3-20 







FeO 



4-46 







MgO 



4-23 





6-5 



CaO 



6-83 





14-0 



Na 2 



3-44 



- -077 



4-7 



K,0 



2-15 





•076 



I Average diorite as calculated by Dal}-. 

 II A haplodiorite represented by a point on the boundary curve. 



*In the sense that they always existed as such, 

 f I. e., it lies on the boundary line. 



j Uber anchi-monomineralische und anchi-eutectische Eruptivgesteine. 

 Videnskabs-Selsk. Skr. I., Math. Naturv. Kl. 1908, No. 10. 



