A. W. Jackson — Nomenclature of Crystalline Rocks. 121 



composition would be provided with different names according 

 as they were formed by one process or another. When actual 

 differences in composition result from different modes of form- 

 ation, such differences will find expression in a mineralogi -al 

 nomenclature, and when no differences exist, it would seem far 

 more rational to use but a single name. I would quote here. 

 with additions of my own in italics, from Prof. J. I). Dana's 

 Manual of Geology, 3d edit., p. 76. "Further, rocks, as ob- 

 jects in science, should b."h fm. land nam d according to their 

 kinds,— not according to the era of formation nor (he method of 

 format/, „~ — ince the sunc things are the same whenever made 

 and however Foi a lately there is no pronounced ten- 



dency to transgress in this direction. 



3. Mineralogical. We come now to an examination of the 

 mineralogical I - r < i disposal, in the light of the three 

 principles previously laid down. I'ndcr this head I propose to 

 include both the minerals themselves and the manner in which 

 they are combined together, in other words rock-texture. 



First, with respect to the minerals themselves. They are 

 susceptible of exact determination, there can be no differences 

 of opinion as to what bh ots of a rock really 



are. The use of the microscope, the application of polarized 

 light whereby the positions of optical planes can be accurately 

 determined, Sorby's method for determining the indices of re- 

 fraction of minerals in rock -sections, Thoulet's method of me- 

 chanically ly be quan- 



titatively analysed if necessary, have all reduced the process of 

 mineralogical determination of the constituents even of compact 

 rocks to great exactness. 



It may be conceived that the difficulty, rarely perhaps im- 

 possibility, of distinguishing between the different kinds of 

 would interpose a serious obstacle in the way of a 

 purely mint be remem- 



bered on the one hand that Descloizeaux' optical distinctions 

 and Thoulet's method of mechanical isolation, together enable 

 us, in most cases, to make the distinction with the greatest 

 accuracy and, on the other hand, that the practical distinction 

 between the different forms of plagioclase that occur as rock 

 constituents is of secondarv importance compared with the 

 fundamental distinction between orthoclase and pi; 

 and this latter we have always been able to make with the 

 greatest ease. 



While it cannot be regarded as proved that albite and anor- 

 thite, the soda-plagioclase and lime-plagioclase, do not widely 

 occur as rock < 'i-titaents still the i -ults ot investigations up 

 to the present point in this direction. It is the intermediate 



