98 



W. Gibbs and F. A. Genth 



under the microscope. It is nearly insoluble in cold and but 

 very slightly soluble in hot water. The solution is readily de- 

 composed by boiling. The insolubility of this oxalate and its 

 characteristic appearance render it of great value in detecting 

 the presence of salts of Xanthocobalt. 

 Oxalate o£ Xanthocobalt has the formula 



+ 



ITO2.5NH3.C02O3, 2C2O3+5HO 



as appear from the following analyses : 



0'3191 grs. gave 0-1762 grs. sulpliate of cobalt 

 0-2980 grs. gave 0'1650 grs. " '* 



2'2520 grs. gave O'lOYS grs. carbonic acid 

 2-2850 grs. gave 0-7260 grs, " " 



The calculated results are 



21 01 per cent cobalt. 



21-06 



25*70 ** oxalic acid. 



25-99 " " •• 



Cobalt, 

 Oxalic acid, 



Eqs. 

 2 



2 



Theory. 



21-14 

 25-81 



Found. 



2101 

 25-70 



21-06 

 25-99 



THEORETICAL CONSIDERATIONS. 



1 



The empirical constitution of tlie ammonia-cobalt bases being 

 as we belieyCj establislied, it remains to offer an exposition of 

 our views of their theoretical structure. Claudet* and Weltzienf 

 have endeavored to reduce the salts of Eoseocobalt and Luteo- 

 cobalt to the type of ammonium, while Fremj has abstained 

 from adopting any particular theory, and givCvS, withou-t com- 

 ment, the results of his analyses in the shape of empirical for^ 

 mulse. Claudet's view is necessarily erroneous, from the fact 

 that his formula for what we term the chlorid of Purpureocobalt 

 is incorrect, inasmuch as he assigns to it 16 in place of 15 equiv- 



alents of hydrogen. Our own numerous analyses, as well as 



those of Eogojski and Gregory, have clearly shown that the 

 number of equivalents of hydrogen is fifteen. 



For an exposition of Weltzien's views we must refer to his 

 paper; they appear to us wanting in simplicity, since they re- 

 quire us to admit, not merely an equivalent replacement of hy- 

 drogen by ammonium and cobalt, but even that the compound 

 ammonium thus formed may replace cobalt in its sesqui-combi- 

 nations. Thus, according to Weltzien, the formula of nitrate of 

 Luteocobalt is 



N 



N 



H 

 NH4 



Co 



n 



NH4 



Co 



O3-J-3NO5 



which is obviously reducible to the type E2 03-f SNOs. 



* Phil. Mag. (4) ii. 253. 



f Arinalen der Chemie und Pliarmacie, xcrn, 19, 



