352 Becker — Determination of Plagioclase Feldspars. 



ning parallel to one pair of sides, while in more numerous 

 cases no twinning is visible. They also in some cases exhibit 

 a truncated corner. It appears probable that these microlites 

 are elongated in the direction of the edge between the base 

 and the brachypinacoid, and that they are cut nearly at right 

 angles to this edge, being occasionally truncated by a hemi- 

 domal face. This suspicion is confirmed by comparison 

 between the extinctions of such square sections and those of 

 microlites in the same slides which show both albite and Carls- 

 bad twinnings. An elongation in the direction of this edge is 

 also to be expected from JBravais's theory of crystallization ; 

 the two faces, base and brachypinacoid, being those of perfect 

 cleavage. Where such microlites present themselves it is easy 

 to see whether they are cut perpendicularly to their axes by 

 following the microlite through the slide ; for if the little 

 prism is inclined, the image in focus will shift laterally as the 

 objective moves. 



Now it happens that prisms bounded by these faces, and in 

 a vertical position, are very favorably situated for discrimina- 

 tion. This will appear by examining Mr. Michel-Levy's stereo- 

 graphic projections of the various feldspars at 90° to the pole 

 p. In this neighborhood there is a saddle in the extinction 

 surfaces of the feldspars (the central point of the saddle answer- 

 ing to Mr. Michel-Levy's " most frequent extinction "), and the 

 consequence is that an inclination of even 10° affects the extinc- 

 tion of a square microlite section very little. Furthermore, the 

 difference between the behavior of different feldspars is great, 

 so that a confusion between the different species is almost im- 

 possible, as may be seen from the little table below.* 



In favorable cases the sign of the extinction can be made 

 useful in the determination of these microlites, which are not 

 really square. In albite the faces c (001) and b (010) make an 

 angle of 86° 24/, while in anorthite this angle becomes 85° 50'. 

 The divergence from rectangularity is thus sensible, and the 

 proper position of the crystal is then with an acute angle in the 

 upper left-hand quadrant (fig. 1). When the hemi-domal faces 

 appear, n (021) truncates the acute angle, making sensibly equal 

 augles of between 46° and 47° with the adjoining faces. The 

 other corresponding hemidome, e (021), truncates the obtuse 

 angle and makes angles of from 42° to 43° with the adjoining 

 faces. Andesine and albite can thus be discriminated when 

 orientation is practicable. This discrimination can be confirmed 

 by testing the index of refraction. Andesine has about the 



* In this table the percentages of anorthite are given as they appear in Michel- 

 Levy's work. As is mentioned above, these do not answer precisely to the molec- 

 ular formulas, but whether the formulas or the percentages need correction I am 

 not certain. 



