A REVISION OF THE ATOMIC WEIGHT OF IODINE. 99 



RATIO OF SILVER TO IODINE. 



The ratio of silver to iodine was next investigated. The problem of obtaining 

 iodine in a dry state presented most difficulties in this portion of the research, 

 and was finally solved as follows: Iodine, which had been prepared in the same 

 way as Sample I, was first freed from the greater part of the water which it con- 

 tained by exposure to concentrated sulphuric acid, which had been boiled to 

 remove every trace of hydrochloric acid. It was then sublimed from a crystalliz- 

 ing dish to the bottom of a glass dish half filled with water, which covered the 

 crystallizing dish. In this way the greater part of the "included" moisture 

 must have been eliminated. The sublimed crystals were finally again sublimed 

 in a current of pure dry air from a porcelain boat contained in a hard-glass tube 

 into the weighing tube. This weighing tube was about 10 cm. long and 15 mm. 

 diameter, and was drawn down to less than half this diameter at both ends. 

 Glass stoppers were ground into both ends of the tube. The air was purified and 

 dried by passing over beads moistened with a solution of silver nitrate, then over 

 sodic carbonate, and finally over 3 feet of beads moistened with concentrated 

 sulphuric acid, all in an apparatus made entirely of glass and connected with 

 the sublimation tube by means of a ground glass Joint. During the final sub- 

 limation of the iodine, the end of the hard-glass tube, which had been drawn 

 down to small diameter, was inserted into one end of the weighing tube. After 

 the weighing tube had been filled, the glass stoppers were inserted, the tube was 

 carefully wiped with a slightly damp "chemically clean" cloth, and was 

 allowed to stand in a desiccator for some time. It was then weighed. Need- 

 less to say, the weighing tube had been originally treated as above before 

 being weighed empty. Both weighings were made with a counterpoise exactly 

 similar to the weighing tube. 



In all but one of the experiments the weighing tube was broken during the 

 solution of the iodine, so that it was impossible to weigh the tube after the ex- 

 periment, and determine how much the glass was attacked by the warm iodine. 

 In one experiment, however, a loss in weight of 0.00032 gm. was found. Whether 

 or not this change was accompanied by the evolution of a gas, the loss in weight 

 must have been at least partially due to combination of the iodine with the 

 alkaline metals of the glass, with the formation of soluble iodides which were 

 dissolved by the solution of sulphurous acid. Hence the error must have been 

 considerably less than 0.3 mg., if, as was necessarily the case owing to breakage 

 of the tubes, the first weight of the tube was used in determining the weight of 

 the iodine. 



Immediately after being weighed, in order to avoid loss by volatilization, the 

 iodine was converted into hydriodic acid by means of pure sulphurous acid. 

 This acid was made by heating sulphuric acid with metallic copper and collect- 

 ing the sulphurous oxide in water, and then distilling the sulphurous oxide from 

 the solution into the purest water. During this distillation any trace of halogen 





