of Iodine in Haloid Scdts hy Arsenic Acid. 335 



tion from 100 cm 3 to 35 cm 3 ; and further, that arsenic, re- 

 duced to the arsenious condition to an amount the exact 

 equivalent of the iodine liberated, remained in solution and 

 was determinable, after neutralization of the acid, in presence 

 of an alkaline bicarbonate, by titration against standard iodine 

 according to Mohr's classical method. We studied carefully 

 the behavior of alkaline bromides and chlorides under identical 

 conditions and determined that 0*5 grm. of potassium bromide 

 acted upon the mixture of arseniate and acid to the extent of 

 reducing arsenic equivalent to 0*0008 grm. of iodine, and that 

 0*5 grm. of sodium chloride did not reduce arsenic but 

 did cause, under the conditions, a volatilization proportional 

 to the amount of arsenious oxide present, the loss amount- 

 ing at the most — when - 56 grm. of the iodide was 

 present to exert its reducing action upon the arsenic — to 

 0"0011 grm. We showed, furthermore, that these maximum 

 errors, due to the action of bromides and chlorides, though not 

 large and tending to neutralize one another when both bromides 

 and chlorides are present, may be eliminated by the application 

 of a numerical correction to the results whenever the amounts 

 of bromide and chloride present become known. 



Recently Messrs. Friedheim and Meyer* have recognized 

 the value of our reaction and applied it to the elimination of 

 iodine from mixtures of haloid salts. They have, however, 

 taken issue with us (unadvisedly, as we think) as to matters of 

 detail. They have, in the first place, put themselves upon 

 record as being unable to titrate arsenious oxide by iodine in 

 alkaline solution under the conditions of our process. They 

 account for their failure by the wholly unsupported hypothesis 

 that the iodine reaction is unavailable in presence of the 

 amounts of salts present, and modify the treatment by distill- 

 ing, collecting the iodine in the distillate, and determining it 

 by the thiosulphate method, thus introducing complexity of 

 apparatus and manipulation, and sacrificing the simplicity and 

 rapidity which are chief advantages of our process. Had they 

 read our paper with intelligent care it must have been evident 

 that we had given special attention to the question of the in- 

 fluence of the salts present upon the iodine reaction ; for we 

 expressly stated that " due correction was made for the amount 

 of iodine necessary to develop the test-color in a solution pre- 

 pared and treated similarly in all respects to the experimental 

 solutions excepting the introduction of the iodide — the correc- 

 tion amounting to a single drop more of the decinormal iodine 

 than was required to produce the end reaction in the same vol- 

 ume of pure water containing only the starch indicator." It 



* Zeitschr. f. Anorg. Chem., i, p. 407. 



