430 PROCEEDINGS OF THE AMERICAN ACADEMY. 



ammonia was added. When iodine is dissolved in a large excess of sul- 

 phurous acid the solution becomes colored yellow, owing to the formation 

 of an iodide of sulphur,* and upon standing, this solution may deposit 

 sulphur. But since the solution was made alkaline with ammonia as 

 soon as the iodine was dissolved, with the complete disappearance of the 

 color and without the. formation of the slightest trace of a precipitate, uo 

 danger was to be feared from this source. 



From the weight of the iodine, the weight of silver necessary exactly 

 to combine with it was calculated. This silver was weighed out and dis- 

 solved in nitric acid as described previously in this paper. The solution 

 was diluted until not stronger than one per cent and was added slowly to 

 the already dilute solution of ammonic iodide in the precipitating flask. 

 The flask was shaken for some time, and was then made acid with a con- 

 siderable excess of nitric acid. Long-continued shaking, followed by 

 several days' standing, yielded a clear supernatant solution. Twenty-five 

 cubic centimeter portions of this solution were pipetted into nephelometer 

 tubes and were tested with cubic centimeter portions of hundredth normal 

 silver nitrate and hydriodic acid solutions in a nephelometer.f If an 

 excess of either iodine or silver was present, the deficiency of the other 

 was made up in the remaining solution by addition of standard silver 

 nitrate or hydriodic acid until the exact end-point was reached. It was 

 never necessary to add more than one tenth of a milligram of either 

 iodine or silver, so that the liquid removed from the fiask for the tests 

 could be neglected. This end-point is very sharp in the case of silver 

 and pure iodine; for so little silver iodide is dissolved that the two nephel- 

 ometer tubes remain almost absolutely clear. An excess of a tenth 

 of a milligram of silver in a liter of solution is easily detected. This 

 almost complete lack of opalescence in the nepiieloraeter tubes is strong 

 evidence of the absence of even a trace of either chlorine or bromine in 

 the iodine, for silver chloride or bromide, on account of tlieir greater 

 solubility, would have produced much more marked precipitates. 



The specific gravity of solid iodine was assumed to be 4.933, J hence 

 a vacuum correction of -f0.000102 was applied to every apparent gram 

 of iodine. 



As was to be expected, the average of this series is slightly higher 

 than that of the previous one, and undoubtedly represents very closely 



* Dammer, Ilandb. d. anorg. Cliem., 1, 557. 



t Ricliards and Wells, Anicr. Clicm. Jour., 31,235. 



I Ladenburg, Ber. d. d. Clieni., Gcscll., 35, 1250. 



