RICHARDS AND BAXTER. ATOMIC WEIGHT OF COBALT. 123 



The Method of Analysis. 



The most insidious of all impurities in accurate quantitative work is 

 water. Its insidiousness is due to the difficulty of detecting it in small 

 amounts, as well as to its invariable occurrence in the atmosphere and its 

 common use as a solvent and medium for crystallization. Our first 

 analytical problem, in this case just as in the case of nickel, was to weigh 

 the salt to be analyzed in such a fashion as to exclude this ever watchful 

 enemy. The fact that cobaltous bromide is less easily decomposed at 

 high temperatures than is nickelous bromide, tended to make our present 

 problem the easier of the two ; but the much greater hygroscopicity of 

 the cobalt salt had the opposite tendency. The apparatus so useful 

 in the cases of magnesium,* zinc, and nickel, proved to be equally ser- 

 viceable here ; cobaltous bromide, after being ignited in a stream of mixed 



Fig. 2. Bottling Apparatus, Hokizontal Section. 



A =: weighing bottle. B = stopper of bottle. CC = hard glass tube. 

 D = Platinum boat containing cobaltous bromide. 



nitrogen and hydrobromic acid, cooled in dry nitrogen, and bottled auto- 

 matically in dry air, may be weighed with perfect certainty. The 

 bottling apparatus has been described so often that further details are 

 unnecessary here ; but the apparatus for supplying the desired gases has 

 not yet been explained in all the complication of its present form. 



In the first place, nitrogen was prepared by passing air through strong 

 ammonia water, and then over red hot copper. The nitroijen, after 

 passing through several fiasks of sulphuric acid, to remove the excess of 

 ammonia, was conducted into column 1 (see Figure 3), containing beads 

 soaked with silver nitrate, in order to free the gas from the possible 

 admixture of sulphur compounds taken from the rubber connections 

 which were used in the apparatus for making nitrogen. In the columns 

 numbered 2 were beads drenched with strong sulphuric acid, which 

 thoroughly dried the nitrogen. With stop-cock 12 open, the nitrogen 



* Richards and Parker, These Proceedings, XXXII. 59. Richards and Cush- 

 man (on Nickel), ante, p. 95. 



