124 RESEARCHES UPON ATOMIC WEIGHTS. 



nitric acid diluted with an equal volume of water as described on page 12. 

 After the silver was completely dissolved, the solution was diluted with an 

 equal volume of water and heated until free from nitrous acid. This latter 

 precaution was especially necessary, since nitrous acid readily liberates iodine 

 from iodides, while nitric acid has this eflfect only when hot or rather concen- 

 trated. Finally the solution was diluted until at least as dilute as thirtieth 

 normal and was added with constant agitation to the solution of the hydriodic 

 acid which had been diluted to a similar concentration in a large glass-stoppered 

 precipitating bottle. The whole was then thoroughly shaken and allowed to 

 stand until clear. In order to determine whether silver or iodide was in excess, 

 30 c.c. portions of the solution were tested with silver nitrate and hydriodic acid 

 in the nephelometer. The results of these earlier analyses were extremely un- 

 satisfactory, the end-point changing continuously with time, and an unreason- 

 ably large amount of silver was required. This could not have been due to 

 occlusion of silver nitrate by the silver iodide, since it has already been shown 

 by one of us that at concentrations less than thirtieth normal, if very nearly 

 equivalent amounts of silver and iodide are used, the latter difficulty is too 

 small to have an appreciable effect.^ 



The cause of the trouble was finally traced to occlusion of silver sulphate by 

 the silver iodide. Richards and Jones "^ found that silver chloride occludes silver 

 sulphate very markedly and tenaciously, hence it is not in the least surprising 

 to find the same difficulty here. It is to be noted that the concentration of sul- 

 phuric acid in these experiments is relatively large, three molecules of sulphuric 

 acid being produced in the reduction for every atom of iodine. That this diffi- 

 culty did not appear in the above mentioned researches upon the ratio of iodine 

 to silver, where iodine was reduced with sulphurous acid and the hydriodic acid 

 titrated against silver, is undoubtedly due to the fact that the concentration of 

 sulphuric acid was relatively much lower, only one molecule of sulphuric acid 

 being formed in the reduction of one molecule of iodine. 



A search for a more satisfactory reducing agent than sulphurous acid failed 

 to reveal any substance more promising than hydrazine, for the reduction of 

 iodic acid with this substance yields only hydriodic acid, nitrogen and water.' 

 In order to avoid the introduction of undesirable acids, the hydrazine was used 

 in the form of a solution of the hydroxide instead of as a salt. The hydrazine 

 hydroxide was made by distilling either the chloride or the sulphate with a 

 concentrated solution of a considerable excess of sodium hydroxide in a plat- 

 inum still. The product was then redistilled in the platinum still to eliminate 



» Baxter: Proc. Amer. Acad., 41, 77 (1905); Jour. Amer. Ckem. Soc, 27, 880; Zeit. anorg. 

 Chem., 46^41. (See page 107.) 



2 Pub. Carnegie Institution, ft^^ 73 (1907); Jour. Amer. Chem. Soc, 291837; Zeit. anorg. 

 Chem., 55, 84. 



^ Brown and Shetterly have shown that no hydronitric acid is formed by the action of 

 hydrazine on iodatcs or iodine. Jour. Amer. Chem. Soc, 30, 53 (1908) 



