PETROGRAPHY. 333 



the influence of microscopic study more revolutionary than 

 in that of the massive rocks. Zirkel, in 1866, classified the 

 massive crystalline rocks mainly upon the basis of the modi- 

 fications of felspar, and sub-divided them into five chief 

 groups orthoclase, orthoclase and oligoclase, nepheline and 

 leucite, labradorite, anorthite rocks. The orthoclase and 

 oligoclase group was sub-divided into rocks containing quartz 

 and rocks without quartz, and the members of the sub-groups 

 were further distinguished by the presence or absence of 

 hornblende or augite, or of different modifications of felspar. 

 The geological age and the structure (granitic, porphyritic, 

 glassy) afforded additional means of differentiation. 



Notwithstanding the great success that attended the micro- 

 scopic study of rocks, certain mineral elements could not be 

 identified by the finest optical methods, and it was felt 

 necessary to combine microscopic and chemical investigations. 

 Micro-chemical methods were invented for the purpose of 

 testing the composition of minute mineral grains ; excellent 

 memoirs dealing with this branch of research were published 

 by Streng, Boricky (1877), Behrens, Haushofer (1883-85), and 

 by Klement and Renard (1885). 



Cordier had in 1815 introduced a mechanical means of 

 separating the fine particles of mineral matter by reducing 

 them to powder, washing the powder with water, and allowing 

 the mineral particles to subside according to their respective 

 specific gravities. An additional device for the isolation of 

 the fine particles was communicated in 1875 by Fouque, 

 who pulverised specimens of the Santorin lava and then 

 used a strong electro-magnet to attract the mineral particles 

 .containing iron. 



A more signal improvement in mechanical means of isolation 

 had been suggested in 1862 by Count Schaffgotsch, and 

 afterwards by Church. It was proposed to introduce finely 

 powdered mineral matter into a saturated chemical solution, 

 such as the solution of iodide of mercury and potassium, 

 prepared by Thoulet, and to shake the mineral powder in 

 the solution, so that the particles which are heavier than the 

 solution will sink to the bottom while the lighter particles will 

 float. By diluting the original solution, or using other solu- 

 tions of given density, the particles can be obtained successively 

 according to their specific gravities. Since Thoulet conducted 

 his experiments, solutions of greater density have been pro- 



