Arsenic associated with Antimony and Tin. 383 



the adaptability of Koehler's treatment to the detection of 

 arsenic in the ordinary course of analysis is limited by the 

 necessity of so constituting the solution to be tested that hydro- 

 gen sulphide shall occasion no deposit of free sulphur to con- 

 ceal or be mistaken for a precipitation of arsenious sulphide. 

 In the course of analysis the mixed sulphides of arsenic, anti- 

 mony, and tin, remaining after the removal of the sulphides 

 insoluble in alkaline sulphides and recovered from solution by 

 the action of hydrochloric acid, require for their complete 

 solution the action of an oxidizing agent which must, of 

 course, interfere Math the immediate use of Koehler's method. 

 If simple means can be found for the destruction of the ex- 

 cess of the oxidizing agent and the simultaneous reduction of 

 the arsenic and antimony to the lower condition of oxidation, 

 it is plain that the test for arsenic by passing hydrogen sul- 

 phide into the solution of antimony and tin in hot hydrochloric 

 acid of half-strength should be sure and easy. In a former 

 paper from this laboratory* a method was described for the 

 quantitative separation of arsenic from antimony, based upon 

 the reduction and volatilization of salts of arsenic by the 

 action of a current of gaseous hydrochloric acid upon the 

 solution containing potassium iodide. It is this reaction — the 

 reduction of arsenic and antimony and the volatilization of the 

 former by the simultaneous action of potassium iodide and 

 hydrochloric acid — which we have now endeavored to apply 

 in simple form to the rapid detection of small amounts of 

 arsenic associated with antimony and tin. We have studied 

 the effect of repeated distillations of small portions of concen- 

 trated hydrochloric acid upon mixtures of the salts with potas- 

 sium iodide. The apparatus which we employ is essentially 

 the distillation apparatus of Mohr, and consists of a 25cm 3 

 flask fitted by means of a rubber stopper to a pipette bent, 

 drawn out at the lower end, and dipped into a test tube 

 which is at the same time supported and cooled in a flask 

 partly filled with water. The pipette tube is wide enough 

 (about 0*7 cm in diameter) to prevent the formation of bubbles 

 within it, and the bulb, holding about 20cm 3 , is sufficiently 

 large to retain any liquid which may be momentarily forced 

 back by the accidental cooling of the flask during the distilla- 

 tion. 



In the test experiments recorded below the arsenic was intro- 

 duced into the flask in the form of arsenic acid dissolved with 

 3 grm. of potassium iodide in 5cm 3 of water, an equal volume 

 of the strongest hydrochloric acid (sp. gr. 1-20) was added, the 

 distillation was carried nearly to dryness, and the distillate was 



* Gooch and Banner, this Journal, xlii, 308. 



