48 ATOMIC WEIGHT DETERMINATIONS. 



mean 29.996. The extreme difference is 0.093. The salt 

 was prepared by solution of the carbonate in chlorhydric 

 acid, addition of ammonia in excess, exposure to the air, 

 washing of the precipitate with acidulated, then with pure 

 water and drying at 110°. A special examination showed 

 it free from other metals. Sommaruga took CI =: 35.5 ; 

 ]Sr = 14. [Erdmann's Journ. fiir Prak. Chem.,100, 1867, 113 ; 

 Sitz.-Bericht dcr k. k. Akad., 1866.) 



C. Winkler : 59 (0 = 16). 



This value is derived from the mean of five experiments 

 on the precipitation of gold from a solution of neutral 

 crystallized chloride of gold and sodium. The metallic 

 cobalt employed was prepared by the reduction of purpureo- 

 cobaltic chloride. The latter was made from oxide, and was 

 purified by recrystallization. Gold was assumed at 196. 

 The mean of the results was 29.496 ; extreme difference, 

 0.071. {Fresenius' Zeitschr. fiir Anal. Chem., 6, 1867, 22.) 



P. 'Welesky : 58.98 (0 = 16). 



Determined from the analysis of cobalti-cyanides, per- 

 formed by drying the salt at 100°, and heating to redness, 

 first in a current of oxygen then of hydrogen. Four experi- 

 ments with phenylammonium-cobalti-cyanide gave cobalt at 

 from 29.38 to 29.59. Two experiments with ammonium- 

 cobalti-cyanide gave from 29.46 to 29.55. Mean, 29.48; 

 extreme difference, 0.21. A single experiment by Winkler's 

 method gave 29.42. {Berlin, Beriehtder Chem. Ges., 2, 1869, 

 592.) 



W. J. RussEL : 58.76 (O = 16). 



Determined by the amount of hydrogen set free by the 

 solution of cobalt in hydrochloric acid. The value is the 

 mean of 2 (or 4 ?) trials. The cobalt employed was that 

 reduced by Russel in his former experiments on the same 

 atomic weight. {Chem. News, W, 1869, 20.) 



R. H.Lee: 59.10 (0 = 16). 



Determined by analysis of cobalti-cyanide salts. They 

 were decomposed in a crucible by heating from above. The 

 carbon separated was burned oft" in air and then in oxygen, 

 and the metallic oxide reduced in hydrogen. Six experi- 

 ments on the strychnine salt gave a mean of 59.05. Six 

 experiments on the brucine salt gave 59.15. Six experi- 



