72 S. L. Penfield — Crystalline form of Sperrylite. 



"were detected ; this is the only case in which a dodecahedral 

 (110) face was found. In a few cases the characteristic combi- 

 nation of octahedron and pyritohedron was detected, but the 

 latter faces were always very small. These results are most 

 satisfactory and from the number of crystals which have been 

 examined and measured, in all of which the pyritohedral faces 

 occur with their proper order and arrangement, the hemihedral 

 nature of the mineral can not be doubted. Some of the crys- 

 tals are somewhat rounded and probably other isometric forms 

 are present but none of them were determined. The faces on 

 the crystals are usually very flat and must be very highly pol- 

 ished to give such satisfactory measurements. It may also be 

 noted that the cubic faces are not usually striated parallel to 

 their intersection with the pyritohedron as is common in pyrite, 

 although it was a slightly striated cube which first called my at- 

 tention to the pyritohedral nature of the crystals. 



The first attempts to determine the pyritohedral faces of the 

 mineral yielded results which were very perplexing but which 

 are not without interest. Cubes with pyritohedral faces were 

 found and measured giving repeatedly the angle of cube on 

 pyritohedron between k 29-J° and 30£°, the pyritohedral faces 

 always giving poor reflections. The calculated value of i-i on 

 i-%, 001 /s 407 being 29° 45'. The pyritohedral arrangement of 

 the faces was perfect but I always failed to find the common 

 pyrite form $(«-2) 7r(210). On talking this over with Professor 

 Wells he stated that all of the material which he had given to me 

 had been cleaned, as for analysis, with aqua-regia and that per- 

 haps the acid had had some action on the faces, as the mineral 

 was not wholly insoluble. He therefore gave me some mate- 

 rial wmich had not been cleaned with acid and the results which 

 were given earlier in this article were obtained from it. The 

 aqua-regia seems to have no effect on the cubic and octahedral 

 faces, at least not enough to diminish their power of reflecting 

 light, for the first measurements given in the article of cube 

 on the octahedron were obtained from a crystal which had been 

 cleaned with acids ; the acids have, however, a very decided ac- 

 tion on the pyritohedral faces, nearly destroying . their power 

 of reflecting light and perceptibly changing their angle. 



To sum up the cry stall ographic observations, the crystals 

 usually show the combination of cube 100, i-i; octahedron 

 111, 1 ; pyritohedron >r(210), i(fc-2) find very rarely dodecahe- 

 dron 110, i. Taken in connection with the chemical results the 

 mineral takes a place in our classification in the pyrite group 

 where an atom of a metal, usually Fe, Co or Ni is united with 

 two atoms of either S, As or rarely Sb, or an isomorphous 

 mixture of them. As this is the first time that platinum has 

 been found in combination as a mineral it mav be noted that 



