236 C. F. Walker — Titration of Sodium 



tion, however, that although the main reaction between iodic 

 acid and sodium thiosulphate may result in the formation of 

 sodium tetrath'ionate in the proportions given, there is never- 

 theless striking evidence of some other obscure action of the 

 thiosulphate, which influences the reduction of the iodic acid 

 in such a way as to make it impossible to calculate the analyses 

 according to Riegler's reaction. Moreover, a peculiar " after- 

 coloration " which invariably follows the first formation of the 

 starch blue during the titration of one solution against the 

 other, seems to point to the possibility that the reaction between 

 the iodide and iodic acid is dependent, under these circum- 

 stances, on conditions of time and mass for its completeness. 

 It is not impossible that some third compound of iodine, 

 unstable in its nature, may be formed as an intermediate 

 product and thus delay the liberation of iodine. In considera- 

 tion of the results that have been obtained it appears that 

 Riegler's proposed process for standardizing sodium thiosul- 

 phate, as well as his related method for the analysis of iodides,* 

 must remain impracticable unless they can be modified so as 

 to do away with a number of sources of error. 



The analyses of solutions of iodic acid, during the entire 

 course of the work, was invariably performed by adding to 

 the portion of the solution to be analyzed an excess of potas- 

 sium iodide, acidifying with 5 cm3 of dilute (1 : 3) sulphuric acid,, 

 and recovering the liberated iodine by directly titrating the 

 acid solution with sodium thiosulphate, or by neutralizing with 

 potassium bicarbonate in excess, and directly titrating the alka- 

 line solution with arsenious acid. In the latter case the neu- 

 tralization was performed in a trapped Drexel washing bottle 

 such as has been described in connection with the analysis of 

 iodides. f In either case one-sixth of the iodine recovered was 

 calculated to iodic acid, according to the terms of the equation, 



5HI + HI0 3 = 3l 2 + 3H 2 0. 



It follows from these proportions that to bring the analyses 

 within the range of the decinormal solutions ordinarily 

 employed, the iodic acid 'taken for analysis must be restricted 

 to comparatively small amounts. In the present work it was 

 found convenient to analyze the iodic acid in quantities not 

 much exceeding one-tenth of a gram, in which case the varia- 

 tion in the results in the same series is found to be almost inap- 

 preciable. In both variations of the process one or two blank 

 analyses were invariably made, by performing the operation as 

 detailed, except that no iodic acid was employed, and the cor- 

 rection of one drop of iodine thereby shown to be necessary to 



* Reigler, Zeitschr far Analyt. Chem., xxxv, 305. 

 fG-ooch and Walker, this Journal, iii, 293. 



