Geology and Mineralogy. 77 



Croatia. It occurs in dense snow-white masses resembling mag- 

 nesite ; specific gravity 2-619. An analysis by the describer gave: 



CO., 32-41 CaO 37-13 MgO 23-75 H a O 6-63 = 99-92 



Although in composition near the mixtures that have been 

 called pencatite and predazzite, no evidence of essential want of 

 homogeneity of gajite could be found. It is named after the 

 Croatian patriot, L. Gaj. — Centralblattfur Min., etc., p. 312, 1911. 



16. Bismuth Ochers from California. — W. T. Schallkr shows 

 in a recent paper (J. Am. Chem. Soc., xxxiii, 162) that the yellow 

 bismuth ochers from San Diego County are either bismuth hydrox- 

 ide, or bismuth vanadate, pucherite, or a mixture of the two ; 

 Bi 2 3 was not found. The same author earlier found that molyb- 

 dic ocher, instead of having the assumed composition MoO a , was 

 in fact a hydrous ferric molybdate. 



17. The Natural Classification of Igneous Rocks ; by Whit- 

 man Cross. Quar. Jour. Geol. Soc, lxvi, 1910, pp. 470-506. — In 

 this paper the author presents a thoughtful discussion of those 

 features of igneous rocks which many petrographers consider 

 essential in the formation of a so-called natural system of classi- 

 fication, in contrast to one whex-e these features are omitted, 

 others are used, and the result is termed by them arbitrary, or 

 artificial. 



He states that chemical composition, mineral constitution and 

 texture are the properties of rocks generally considered available 

 for their classification, and discusses each of these according to 

 the different methods suggested by various authorities. In a full 

 analysis of the attempts made by Becke and Harker to divide the 

 igneous rocks geographically into two great chemical groups, the 

 Atlantic and Pacific, he points out the many discrepancies that 

 arise, and that, while the rocks of the Pacific zone of North 

 America exhibit pronounced peculiarities, they are not those 

 claimed for them by the advocates of such a grouping. He also 

 points out that the generalizations made on this basis do not apply 

 to the older rocks, which in many cases have been left out of 

 account. 



Discussing the factors of magmatic differentiation he takes up 

 the dike rock group of Rosenbusch, and shows that the magmas 

 cannot be divided exclusively into foyaitic-theralitic and granito- 

 dioritic, gabbro-peridotitic series, because these are connected by 

 intermediate magmatic series, such as the monzonitic. And if 

 this is so, then the dike rocks cannot always have the exclusive 

 associations claimed for them, and he cites numerous instances to 

 prove that the associations are sometimes other than those claimed. 



Taking up the suggestions of Becker, Vogt, and others that the 

 principle of eutectics may furnish a means of classification, the 

 author points out that if we followed Becker's ideas, in porphy- 

 ries the groundmass alone would be used, and some rocks would 

 not be susceptible of classification. Vogt's view is that magmas 

 differentiate in the same way that they tend to crystallize, into 



