MINERALOGY 21 



lines for both the ordinary and extraordinary rays. Since 

 gehlenite is optically negative and ^kermanite positive, the 

 two curves intersect, a mixture approximating to 45 per cent, 

 gehlenite and 55 per cent, akermanite having zero bire- 

 fringence for sodium light. The density curves for both the 

 crystalline and the glassy series of mixtures are likewise 

 straight lines, but akermanite itself is abnormal in that the 

 glass has a higher density than the crystalline form. 



As a continuation of his well-known work on the equilibria 

 of such complicated systems as those occurring in igneous 

 rock magmas and metallurgical slags, J. H. L. Vogt has pub- 

 lished a memoir " Die Sulfid-Silikatschmelzlosungen " {Norsk 

 Geol. Tidsskr, 191 7; Vidensk. Selsk. Skrift., 191 8 ; cf. Geol. 

 Mag., 58, 87, 1 921). From the petrographic standpoint, the 

 most important part of the memoir is the discussion of the 

 sulphide segregations which occur in conjunction with norites 

 and peridotites. In this connection, it is notable that sulphide 

 ore-bodies high in nickel are commonly associated with rocks 

 rich in the orthorhombic pyroxenes, such as the norites ; less 

 commonly with the peridotites, which are mainly composed of 

 orthosilicates of iron and magnesium ; and very rarely with 

 such calcic types as the gabbros. From a physico-chemical 

 study of the norites, Vogt comes to the conclusion that while 

 many of the examples of the latter rock are " anchieutectic," 

 the ore-bodies tend to be associated with those types in which 

 the pyroxene is in excess of the eutectic proportions. The 

 segregation of the sulphides is probably to be explained by 

 their comparative immiscibility in the molten condition with 

 silicate magmas, and by their low viscosity. The bearing of 

 the investigations on the problems arising in the smelting of 

 copper matte is also discussed at some length. 



In a paper on " Crystallisation Differentiation in Magmas " 

 {Jour. Geol., 27, 393, 1919 ; Proc. Nat. Acad. Set., 6, 159, 

 1920), N. L. Bowen replies to some of the criticisms of the 

 theory of the origin of rocks by the gravity separation of the 

 early formed crystals {Jour. Geol., 23, Suppl. 191 5). Discon- 

 tinuous variations in rock masses have been ascribed by various 

 authors {e.g. R. A. Daly, ibid., 26, 117, 1918 ; F. C. Grout, 

 ibid., 26, 656, 191 8) to immiscibility of the two component 

 parts in the liquid state, despite the fact that all the experi- 

 mental evidence available is in favour of the complete misci- 

 bility of all molten silicates. Bowen points out that under- 

 cooling cannot be postulated as the reason for the formation 

 of non-consolute magmas on account of the probably limited 

 amount of undercooling which is liable to occur, for example, 

 in batholiths. Further, in the case of abrupt transitions from 

 one rock type to another, such phenomena are even more 



