84 PROCEEDINGS OP THE AMERICAN ACADEMY. 



Kothner's paper does not make clear whether or riot fusion of the silver 

 iodide took place at the beginning of his experiments, so that it is un- 

 certain whether the gaiu in weight of the coil of glass tubing, which was 

 attached to the reaction tube, was due to volatilized silver halicles or to 

 attacking of the glass by the hot halogens. At any rate, it is hard to 

 believe that a meter and a half of glass tubing subjected to the long- 

 continued action of mixed chlorine and iodine at 150° should not have 

 altered its weight somewhat. Furthermore, Kothner himself showed 

 that the tube in which the reaction took place was slightly attacked by 

 the fused silver chloride, but the nature of the correction for this attack- 

 ing of the glass is uncertain. Finally, although the precaution was 

 taken of fusing the silver iodide before the initial weighing, so that the 

 salt must have been free from moisture, it was not fused in au atmos- 

 phere of iodine. Hence it is possible that the iodide still contained 

 traces of occluded silver nitrate (or metallic silver). This deficiency in 

 iodine, as well as the possible gain in weight of the apparatus during the 

 experiment, would have lowered the atomic weight of iodine. 



As far as Kothner's syntheses of silver iodide from weighed amount of 

 silver are concerned, it need only be said that, even assuming that occlu- 

 sion was avoided in the synthesis in the wet way, and that in the syn- 

 thesis in the dry way a single precipitation of the iodine from solution in 

 an iodide had completely removed such impurities as chlorine and bro- 

 mine, which would have accumulated in the silver iodide during the 

 experiment, the fact that Richards and Wells * have shown that silver 

 fused in the air or with borax and saltpetre, according to Stas, must con- 

 tain oxygen, makes it certain that the results of Kothner's syntheses 

 are too low. 



I am indebted to the Cyrus M. Warren Fund for Research in Harvard 

 University for platinum vessels, quartz crucibles, and balance. 



Summary. 



The results of this investigation are, then, as follows : — 



1. The atomic weight of iodine is found to be 126.985 (Ag= 107.930) 

 one one-hundredth of a unit higher than the value previously obtained. 

 If silver is assumed to be 107.920 and oxygen 16.000, iodine becomes 

 126.973. 



2. The observation of Kothner and Aeuer that under certain coudi- 



* Publications of the Carnegie Institution, No. 28, 62. 



