THE 



AMERICAN JOURNAL OF SCIENCE 



[FOURTH SERIES.] 



Art. XXXII. — The Stokes Method for the Determination of 

 Pyrite and Marcasite ; by E. T. Allen and J. L. 

 Crenshaw. 



Syllabus. 



I. Demand for the Method. 

 II. The Stokes Method. 



A. Formula for the percentage of sulphur oxidized. 



B. Eelation of tht- iron dissolved to the composition of the mixture. 



1. Cause of the deviation of the values of x from a straight line. 



(a) Explanation of the low results with pure marcasite. 



(b) The necessary excess of the sulphide. 



2. Eesults on mixtures of the natural minerals. 



3. Other sources of error in the Stokes method. 



(a) The ratio of the surface of the two minerals in weight 



percentage mixtures. 

 (&) Loss of water during the oxidation process. 



(c) Reduction of the ferric solution during the blank test. 



(d) Errors in the determination of the iron. 



(e) Oxidation of either ferrous solution or finely divided 



sulphides by atmospheric oxygen. 



4. Accuracy of the method. 



5. Linear relation between any pyrite and any marcasite. 



(a) Independence of the chemical action of the two 



sulphides. 



(b) Electrolytic effect in the Stokes reaction. 



(c) Stokes' curve. 



Summary. 



I. Demand for the Method. 



The question whether a given specimen is pyrite or mar- 

 casite, in the case of well-crystallized natural material, will 

 seldom arise; color and crystal form are generally sufficient 

 criteria. For doubtful cases, which are more liable to be met 

 with in synthetic products, few reliable tests exist, and for 

 estimating either in mixtures with the other, the Stokes method 

 is the only one known. 



Am. Jour. Sci.— Fourth Series, Vol. XXXVIII, No. 227.— November, 1914. 

 26 



