126 Some Recent American Petrological Literature. 



" The System Anorthite-Forsterite-Silica," by 0. Andersen. Amer- 

 Journ. Sci., vol. xxxix, pp. 407-54, 1915. 

 A detailed experimental and theoretical discussion of this system,, 

 which must he treated as one of four components, in order to account 

 for the formation of spinel in some of the ternary mixtures. The 

 results are applied to actual rocks, especially to varieties that 

 contain olivine. 



" Crystallization-Differentiation in Silicate Liquids," by N. L. 

 Bowen. Amer. Journ. Sci., vol. xxxix, pp. 175-91, 1915. 

 Experiments were undertaken with artificial melts to determine 

 whether sinking or floating of crystals could be obtained. Olivine 

 and pyroxene were found to sink and tridymite to float: the rate of 

 sinking indicates a progressive increase of viscosity with increase of 

 silica. The results obtained are applied to the observations of Lewis- 

 on the Palisade sill, and it is concluded that sinking of crystals is of 

 importance even in acid magmas. 



"The Crystallization of Haplobasaltic, Haplodioritic, and related 



Magmas," by N. L. Bowen. Amer. Journ. Sci., vol. xl,, 



pp. 161-85, 1915. 



The relations of diopside to the plagioclase series are studied by 



the quenching method of thermal analysis. The facts determined 



for artificial melts are applied to their natural analogues, and it is- 



concluded that there can be little reason to doubt that crystallization 



controls the differentiation of the sub-alkaline series of igneous rocks. 



"The Later Stages of the Evolution of the Igneous Rocks," by 



1ST. L. Bowen. Supplement to the Journal of Geology, vol. xxiii, 



No. 8, 1915. 91 pp. 



In this long paper Dr. Bowen gives a summary of his conclusions 



based on an extensive discussion of the whole problem of the 



evolution of the igneous rocks, taking into account the results of 



investigations of the course of crystallization in artificial melts and 



the theoretical conclusions arising from them. Assimilation and 



direct refusion of sediments are regarded as unimportant, since they 



would lead to rock-types such as are never found. The decision is- 



reached that differentiation is controlled entirely by crystallization, 



mainly by sinking of crystals and squeezing out of residual liquids, 



and it is shown that with slow cooling typical rock-series could be- 



formed from basaltic magma, which is probably the primitive type. 



"Differentiation in Intercrustal Magma Basins," by A. Harker. 

 Journ. Geol., vol. xxiv, pp. 554-8, 1916. 

 This is mainly a criticism and review of Dr. Bowen's advocacy of 

 differentiation in situ as opposed to differentiation before intrusion. 

 It is pointed out that increase of viscosity in a cooling body of 

 moderate size would soon stop sinking of crystals, and the frequency 

 of small, separate, but obviously related intrusions is insisted on. It 

 is also shown that the great majority of the crystalline schists of 

 igneous origin belong to the calcic branch, indicating the connexion 

 of calcic magmas with tangential thrusting movements. 



