374 Allen and Crenshaw — Stokes Method for the 



p = the percentage of sulphur in the decomposed sulphide 

 which is converted into sulphuric acid ; b is the ferrous iron at 

 the end of the experiment originating both from the reduction 

 of the ferric salt and from the decomposition of the iron 

 disulphide ; c is the total iron, ferrous and ferric,* at the end of 

 the experiment ; while a is the iron originally present in the 

 standard solution. Our experiments soon showed that the 

 reaction normally proceeds to the complete reduction of the 



o.qoi 



ferric salt: b and c are therefore identical. Hence p=., 25, 



* b — a 



or better written, since b is variable, (2) p=— 25. 



Wjf x—a 



B. Relation of the iron dissolved to the composition of the 

 mixture. — A simpler relation than this and one more satisfactory 

 in estimating the quantities of pyrite and marcasite in a mix- 

 ture of the two is obtained by comparing directly x, the total 

 quantity of iron in solution at the end of the test, with y, the 

 percentage of pyrite in the sulphide mixture. Some results 

 obtained by us before any such relation was suspected, indi- 

 cated, when plotted in this way, a rectilinear relation ; only 

 the mixtures lowest in pyrite deviated much from a straight 

 line (see fig. 1). Stokes concerned himself only with the rela- 

 tion between the composition of the sulphide mixture and the 

 percentage of sulphur oxidized, but when his results are put 

 into the same form they confirm our own. Table 1 gives the 

 values of x, the total iron in solution at the end of the test, 

 calculated by means of equation (2) from Stokes' published 

 values of p. Our own value for a was used, a is the concen- 

 tration in iron of the standard ferric solution. Both x and a 

 are given in terms of a standard permanganate solution, 1 g. 

 of which contained -001544 g. KMn0 4 ; and 100 cc of the ferric 

 solution, measured at 25°, required 36'71 g. of this solution to 

 oxidize it after reduction by sulphur dioxide. In tig. 2, 

 Stokes' values for (x — a), the iron dissolved by 100 cc of the 

 standard ferric solution from pyrite-marcasite mixtures, are 

 plotted against ?/, the percentage of pyrite which they con- 

 tained. Here also the points fall into a nearly straight line, 

 excepting those which stand for mixtures low in pyrite. 



1. Cause of the deviation of the values of xfrom a straight 

 line. — It seems decidedly probable that the low values for x 

 in mixtures which contained 75^-100$ marcasite were due to 

 some disturbance in the reaction. This hypothesis would sim- 

 ply mean that the chemical action of the ferric solution on each 

 sulphide in the mixture was proportional to its surface; that 

 each sulphide acted independently of the other; and that the 



* Stokes did not get a complete reduction of the standard ferric solution. 



