A REVIEW OF IGNEOUS ROCK CLASSIFICATION 75 



material at present available to the petrographer it is inevitable 

 that chemical analysis will have to be resorted to in order to 

 determine the mode of an aphanitic rock. In any modal classifi- 

 cation on quantitative lines it seems necessary to assume that 

 the plutonic noncrystalline rocks are those affording the primary 

 units for classification, and to treat the aphanitic rocks as 

 textural modifications of these. In the case of porphyritic rocks 

 the proportions of different kinds of phenocrysts to one another 

 and to the groundmass could easily be estimated by the Rosiwal 

 method. There is need for the chemical investigation of a large 

 series of typical groundmasses with a view to the determination 

 of their average modes. These could be used in calculation in 

 much the same way as the analyses of pyroxenes, amphiboles, 

 and micas collected by the authors of the Quantitative Classifi- 

 cation are used for the purpose of modal calculation. 1 



It is admitted that the treatment of the aphanitic rocks on 

 modal lines constitutes a serious difficulty in the way of modal 

 classification. It is a difficulty, however, which may be largely 

 removed by future work. Furthermore, it is believed that an 

 estimation of the mode of many aphanitic volcanic and hypabyssal 

 rocks can be made with sufficient accuracy for classificatory 

 purposes. For example, generally with the aid of a Rosiwal 

 measurement supplemented by calculation from the chemical 

 analysis, Washington has estimated the modes of many fine- 

 grained leucitic and other lavas from Italy. 2 In the description 

 of aphanitic rocks even an approximate estimation of the mode 

 by a careful examination of the thin section would be better 

 than no estimation at all, and would help to fit the rock 

 into one of the larger compartments of the quantitative modal 

 classification. 



Notwithstanding the opinion of Williams, 3 the Rosiwal 

 method of estimation is capable of giving extremely accurate 

 results. 4 The writer has recently used it for the investigation 

 of teschenites and related rocks of Central Scotland, and has 

 found the chemical composition calculated from the results of 

 the Rosiwal measurement to be strikingly accordant with that 

 obtained by ordinary chemical analysis. 



1 C.I.P.W., Quantitative Classification of Igneous Rocks. Tables XI I. -XIV. 



2 The Roman Comagmatic Region, Publications Carnegie Inst., No. 57. 



3 American Geologist, xxxv. 1905, pp. 34-46. 



4 F. C. Lincoln and H. L.. Rietz, Economic Geology, viii. 191 3, pp. 120-39. 



