196 Gooch and Browning — Method for the 



It appears, therefore, from the results of Series I, that, when 

 the amount of sodium chloride present is restricted to 0*5 grm., 

 the liquid may be boiled until the sulphuric acid amounts to 

 33*3 per cent of it, with a loss, on the average, of 0-0008 grm. 

 of arsenious oxide, aud to 25 per cent with a loss of 0'0004 grm. 

 With either of these proportions the results are sufficiently 

 favorable to warrant quantitative testing of a method for sepa- 

 rating chlorine and iodine based upon the volatility of iodine 

 and the non-volatility of arsenious oxide under the conditions. 

 The case is even more favorable for the similar separation of 

 bromine from iodine — the loss of arsenic amounting on the 

 average to 0*0001 grm., under similar conditions. The experi- 

 ments of Series I) to H indicated that the degree of concen- 

 tration at which all iodine vanishes from the liquid corresponds 

 to the presence of 28*6 per cent of the acid — a point midway 

 between the points of concentration indicated above. We fixed 

 this proportion, therefore, for the following experiments, as 

 being rather more favorable as regards the fixity of the arsenic 

 and perfect as to the elimination of iodine. The smallest ab- 

 solute amount of sulphuric acid which we have used — 10 cm 3 

 of the [1 : 1] mixture — demands evaporation of the liquid to a 

 bulk too small to be easily determined with certainty in flasks 

 of the shape and dimensions which we found convenient for 

 this work. The larger amounts of acid are uncomfortably large 

 in the»subsequent neutralization. We took, therefore, 20 cm 3 

 of the [1 : 1] mixture of sulphuric acid and water as the amount 

 best adapted to our purpose and set the limit of concentration 

 at 35 cm 3 . 



The potassium iodide which we used was prepared with 

 great care, by acting with re-sublimed iodine upon iron wire, 

 three-fourths of the iodine being added to an excess of iron 

 covered with distilled water, decanting the solution from the 

 excess of iron when the color of iodine had vanished, adding 

 the remainder of the iodine, pouring the filtered solution 

 slowly into boiling water, to which the exact amount of acid 

 potassium carbonate necessary to combine with the iodine had 

 been added, and filtering off the magnetic oxide of iron thus 

 precipitated. The solution of potassium iodide thus made, 

 very faintly alkaline and entirely free from chlorine and bro- 

 mine, was made up to a suitable volume and standardized by 

 precipitating the iodine of aliquot portions by weight as silver 

 iodide, which was heated and weighed upon asbestos. Weighed 

 portions of this solution, containing approximately 0*5 gm. "of 

 potassium iodide to 30 cm 3 , were taken for the tests in which 

 the larger quantities of iodine were introduced ; measured 

 portions of a solution made by diluting the former were em- 

 ployed in the tests involving the smaller amounts. 



