398 Allen, Crenshaw, and Merwin — 



D. Analysis of the products. — The molecular ratio between 

 the two products of the interaction of hydrogen sulphide, sul- 

 phur and a ferrous salt is strong evidence of the composition 

 of the precipitate. This had already been directly proved by 

 chemical analysis to be the disulphide,* while the color, luster 

 and chemical behavior agreed with pyrite or marcasite. The 

 purposes of the present investigation necessitated a method 

 for determining whether we had one or the other, or both. 

 In the preceding paperf we have shown how the Stokes 

 method may be utilized for this purpose. The method is 

 founded on the difference in the chemical behavior of the two 

 crystalline forms toward oxidizing agents. A boiling standard 

 solution of ferric alum dissolves a characteristic and constant 

 quantity of iron from each of the minerals pyrite and mar- 

 casite, and these constants are additive for mixtures of the 

 two ; i. e., the quantity of iron dissolved by ferric alum is a 

 linear function of the composition of the mixture. We may 

 anticipate the results so far as to say that the Stokes method 

 indicates in all cases, with the possible exception of the prod- 

 ucts formed at 100°, that the sulphides are either pure mar- 

 casite or mixtures of marcasite and pyrite. 



E. Other evidence on the nature of the products. — Certain 

 objections to exclusive reliance on the Stokes method will 

 occur to the chemist. The products are all crystallized from 

 solutions containing an excess of ferrous salt as well as hydro- 

 gen sulphide. Sulphur dioxide also is a final product in 

 many instances. Any of these impurities if occluded by the 

 precipitate would disturb the Stokes test; the ferrous salt by 

 directly increasing the iron in solution ; the sulphur dioxide 

 or hydrogen sulphide by reducing the standard solution and 

 thus decreasing the iron which should normally dissolve. 



Many instances are known of occlusions which are impos- 

 sible to remove by ordinary washing. But the purification of 

 these sulphides was out of the ordinary in that the substance was 

 ground in a mortar for two hours before boiling with hydro- 

 chloric acid and washing with water. Such drastic treatment 

 might be expected to remove these occlusions if any are pres- 

 ent. Still we have positive evidence that the products which 

 we infer to be mixtures of pyrite and marcasite from the 

 Stokes tests are really such, rather than marcasite containing 

 reducing impurities. 



1. Microscopic evidence. — The identification of opaque 

 minerals as compared with the transparent minerals is much 

 less fully worked out. In default of optical constants, one 

 must rely on crystal form, color, cleavage and on microchem- 



*This Journal (4), xxxiii, 173, 1912 ; Zs. anorg. Ckein., lxvi, 206, 1912. 

 fThis number, p. 371. 



