42C SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 54 



tained some free oxalic acid, and that the cohalt salt was possibly con- 

 taminated with carbonate or with basic compounds. In bis later papers 

 Schneider rejects these suggestions as unfounded, and in tui'o criticises 

 Marignac. The purity of anhydrous NiS04 is not easy to guarantee, and, 

 according to Schneider, the anhydrous chlorides of cobalt and nickel are 

 liable to be contaminated with oxides. This is the case even when the 

 chlorides are heated in chlorine, unless the gas is carefully freed from 

 all traces of air and moisture. 



Dumas' ^ determinations of the two atomic weights were made with 

 the chlorides of nickel and coljalt. The pure metals were dissolved in 

 aqua regia, the solutions were repeatedly evaporated to dryness, and the 

 residual chlorides were ignited in dry hydrochloric acid gas. The last 

 two estimations in the nickel series were made upon NiCL formed hy 

 heating tbe spongy metal in pure chlorine. In the third column I give 

 the XiCL or CoCL equivalent to 100 parts of silver : 



.9123 grm. NiCl, = 1.515 grrn. Ag. 60.218 



2.295 " ""^^" " ''''"■"' 



3.290 

 1.830 

 3.001 



.0062 

 Hence Ni = 58.97. 



2.3.52 grm. CoCL = 3.9035 grm. Ag. 60.254 



60.229 

 60.268 

 60.186 

 60.202 



Mean, 60.2278, db .011 



Hence Co = 59.03. 



These values for Co and Ni differ by less than a tenth of a unit; 

 here, as elsewhere, the fig-ure for Ni being a trifle the lower. 



Combining these data for nickel with Marignac's series, we have — 



2Ag:yiCL::100:r. 



Marignac 60.139, ± .0310 



Dumas 60.199, ± .0062 



General mean 60.194, ± .0061 



The cobalt figures will bo combined witli others later. 



1 Ann. Chem. Pharm., 113, 2r,. ISGO. 



