R. H. Lee—Atomic Weights of Cobalt and Nickel, 47 
The mean of all the results is 785926, from which we find 
for the equivalent of cobalt on the old system 29°37. 
I the subject was taken up by Sommaruga,* who de- 
termined the amount of metallic cobalt in chloride of purpureo- 
cobalt by reduction in a bulb tube with hydrogen. His results 
were as follows: 
Salt taken. Cobalt found. Equivalent. 
1 0°665 0°1588 0°002 
1:0918 2600 29°929 
3 0°9058 2160 29°982 
4 1°5895 3785 29°926 
5 2°9167 6957 29°992 
6 1°8390 "4378 29°916 
7 275010 5968 9 
: Mean, 29°965 
Winkler,+ in 1867, determined the atomic weight of cobalt by 
heating a known weight of the metal with perfectly neutral so- 
lution of double chloride of gold and sodium. The cobalt was 
obtained by the reduction of chloride of purpureocobalt. 
Salt taken. Gold found. Equivalent of €o(/Au=196.) 
i, 0°589 1°3045 29°497 
2 0°3147 0°6981 29°451 
3 0°5829 12913 29°492 
4, 05111 1:1312 29°518 
5 0°5821 1°2848 29°522 
Mean, 29-496 
Finally, in 1869, esos 6: made a new determination of 
the atomic weight of cobalt by finding the quantity of metallic 
cobalt in weighed quantities of cobalticyanide of aniline-am- 
monium and of ammonium. The results were as follows: 
Cobalt salt. Cobalt found. Equivalent. 
1. 08529¢r, ——01010 29°44 
rae OCit: — 0°0723 29°38 
3. 0°7140 “ 0°0850 29°59 
4 0°9420 “ 0°1120 29°54 
‘io 0°7575 “ 0°1160 29°46 
2. 0°5143 071130 29°55 
3 Mean, 29°48 
The first four analyses were made with the anilin, the last 
two with the ammonium salt. These are the only determina- 
tions of the atomic weight of cobalt which I have been able to 
find, and I will therefore pass to my own analyses. 
* Sitzungsberichte der Wiener Akad., vol. liv, p. 50. 
Zeitschrift fir Anal. . 1B. 
¢ Berichte der Deutcshen Chem. Gesellschaft, 1870. 
