58 THE BEHAVIOUR OF SOLUTIONS OF HYDRIODIC ACID 



allowed to act for a couple of hours ; portions of the residue from 

 the solution were heated with potassium cyanide a'nd powdered 

 charcoal, and with powdered magnesium ; and lastly some of the 

 residue was heated in a current of hydrogen to such a high tem- 

 perature, that the sodium sulphate was reduced to sulphide. In 

 neither of these instances was the slightest trace of iodine 

 obtained. 



As it was thought that an analysis of the solution might 

 throw some light on the problem, the following determinations 

 were undertaken. 



80 cc. of the colourless solution that had been acted 

 on by pure oxygen, and which, therefore, must have con- 

 tained all the* iodine originally in it, were exactly neutralized 

 with potassium carbonate and evaporated to dryness. The 

 residue after being dried at 110 for an hour weighed 8.0530 

 grams. This residue was used for the analysis, and the only 

 substances that it could contain besides the iodine, were potas- 

 sium, and sulphuric acid in the form of sulphate, (it was 

 proved that there was no carbonate present). 



The amount of sulphuric acid (SO'^ ) was determined by 



precipitating with barium chloride. The following <\;e the 

 results obtained : 



( 1 ; 1.4914 grams residue yielded 1.9813 grams Ba SO 4 

 (2) 1.4914 " " ." 1.9842 " " 



mean 1.9827 



This weight of barium sulphate corresponds to 0.8168 gram of 

 S0 4 , or to 54.76 per cent, of the residue used. 



The potassium was estimated by precipitating as double 

 potassium-platinum chloride. This precipitate after being 

 thoroughly dried, was heated with oxalic acid and reduced 

 to a mixture of metallic platinum and potassium chloride, 



