Penfield and Forbes — Chrysolite-Fay alite Group, etc. 133 



For our own investigation chrysolites were selected from 

 specimens in the Brush collection. The percentages of ferrous 

 iron were determined by means of potassium permanganate, 

 but unfortunately in the available material the range in the 

 replacement of magnesia by iron was not great, leaving a wide 

 gap in the series between the one richest in iron and hortono- 

 lite. The varieties examined are arranged below according to 

 the decrease in the percentage of iron. The divergences of 

 the optical axes, 2H, were all measured with yellow light, E~a, 

 in a-mono-bromnaphthalene at a mean temperature of 25° C. 



Auvergne, France. — Loose crystals were chosen of a dark 

 green color and well developed ; similar to fig. 4, page 452, 

 Dana's Mineralogy, 6th Ed. The plate for optical examination 

 was made parallel to 010. FeO = 13*02 per cent, /9 = 1-6916, 

 2H = 92° 1'; hence 2Y a =89° 36'. 



Vesuvius, Italy. — Small loose crystals were selected of dark 

 olive green color, showing distinct faces with 010 prominent, 

 parallel to which the plate was cut. FeO = 12*6 per cent, 

 2H = 92° 9 r , and considering /9 = 1*6916 as in the Auvergne 

 crystals, 2Y a = 89° 42'. 



Hawaii, Sandwich Islands. — The material of olive green 

 color was collected by Prof. J. D. Dana on the southeastern 

 shore, south of Hilo. It consisted of waterworn grains, show- 

 ing occasionally a cleavage face parallel to 010 which served for 

 orientation, and parallel to which the plate was cut. FeO = 

 10-3 per cent, 2H = 92° 30', and using /3 = 1*678 as given by 

 Des Cloizeaux, 2Y ? = 91° 2'. 



Egypt or the Orient. — A fragmentary crystal was used, of 

 light olive green color and with crystal faces, which served for 

 orientation. The plate was cut parallel to 100. FeO = 9*16 

 per cent, 2H = 89° 39', /? = 1*679 ; hence 2Y a = 88° 41'. 



Northwestern Corner of New Mexico. — A rounded frag- 

 ment was selected from material collected by the Indians and 

 sold to lapidaries for gem purposes. The fragment showed no 

 crystal faces, but a cleavage parallel to 010 served for orienta- 

 tation, and parallel to this the plate was cut. FeO = 8*6 per 

 cent, 2H = 92° 57', and using /? = 1*678, 2Y =91 / 24 / . 



The result of this investigation can be seen best from the 

 following table, where 2V is given for yellow E~a, measured 

 over a, or the crystallographic axis b. The dispersion in all 

 cases when measured in this way is p ) u. The last three are 

 the determinations of Des Cloizeaux and Zimanyi already 

 referred to. 



