36 RESEARCHES UPON ATOMIC WEIGHTS. 



cipitate discarded. In a similar way third and fourth fractions of sulphide were 

 removed. Next the solution was thrice fractionated with small portions of 

 potassiimi hydroxide, the precipitate being rejected in each case. Then the 

 manganese was twice precipitated as carbonate by means of ammonium car- 

 bonate and the manganous carbonate was converted into bromide exactly as in 

 the case of Sample B. The first crop of thrice recrystallized bromide is desig- 

 nated Sample Di, a second similar crop obtained from the mother-liquors is 

 Sample D2. 



HYDROBROMIC ACID. 



Commercial bromine was freed from chlorine by twice converting the bro- 

 mine into hydrobromic acid by means of thoroughly washed hydrogen sulphide 

 and water, and heating the hydrobromic acid, after distillation, with recrystal- 

 lized potassium permanganate. The bromine was thus twice distilled from a 

 bromide, the bromide in the second distillation being almost free from chloride. 

 Iodine was eliminated by boiling the hydrobromic acid in each case with a small 

 quantity of permanganate and rejecting the bromine set free. A portion of the 

 final product, when converted into ammonium bromide by means of ammonia, 

 and added to a solution of 3.46875 gm. (in vacuum) of pure silver, yielded 

 6.03855 gm. (in vacuum) of fused silver bromide, whence the ratio of silver to 

 silver bromide is 57.443, while 57.445 is the value to be expected.^ 



By treating this bromine, covered with water, with washed hydrogen sulphide, 

 hydrobromic acid was again produced. The solution was boiled, after mechani- 

 cal separation of the greater part of the free sulphur and bromide of sulphur, 

 and was then filtered. In order to remove the sulphuric acid produced during 

 the action of the bromine upon the hydrogen sulphide, the hydrobromic acid 

 was first distilled. Then it was diluted, and a small quantity of recrystallized 

 barium hydroxide was added to precipitate last traces of sulphuric acid. The 

 slight precipitate of barium sulphate was collected upon a filter, and the acid 

 was three times distilled with rejection of the first and last portions, with a 

 glass retort and condenser. Finally the acid was once distilled with the use of a 

 quartz condenser. The product of the final distillation was collected in quartz 

 vessels and was used immediately for dissolving the manganous carbonate. That 

 this acid was free from solid impurities, such as alkalies and silica, was shown 

 by evaporating 30 c.c. in a weighed platinum crucible. No weighable residue 

 remained after the crucible had been heated to very dull redness. 



NITRIC ACID. 



This acid was twice distilled, all but the last third of the distUlate being 

 rejected in each distillation. This acid gave no test for chloride in a nephel- 

 ometer. 



* Baxter: Proc. Amer. Acad., 42, 210 (1906); Jour. Amer. Chem. Soc, 28, 1332; Zeit. 

 anorg. Chem., 50, 398. (See page 59.) 



