20 RES 



EARCHI 



IS ON THE 





The formula requires 











Eqs. 



Calculated. 



Found. 



Cobalt and iron . 



. 4 



30.02 



30.05 



Carbon 



. 12 



18.79 



18.69 



There can be no reasonable doubt that this salt is isornorphous with the corre- 

 sponding cobalt salt. Like the latter it is an extremely valuable test for the salts 

 of Roseocobalt and for certain salts of Purpureocobalt. 



OXIDE OF ROSEOCOBALT. 



The oxide of Roseocobalt exists only in solution. It is obtained either by 

 decomposing the chloride by oxide of silver, or by adding baryta-water to a cold 

 solution of the sulphate ; the latter method is the better one, because the chloride 

 of silver is soluble in solutions of the chloride of Roseocobalt. The solution as 

 thus obtained is red, has an alkaline non-metallic taste and reaction, and is very 

 easily decomposed. By standing in the air it absorbs carbonic acid and forms a 

 carbonate. 



MAGNETIC OXIDE OF COBALT. 



In connection with the " salts of Roseocobalt we may perhaps with propriety 

 describe the peculiar oxide of cobalt, which is, in some cases at least, one of the 

 products of their decomposition. The hydrated oxide which is precipitated by 

 boiling the neutral salts of Roseocobalt with an alkaline solution is considered by 

 Fremy as a hydrate of the sesquioxide, and he attributes to it the formula Co,0 3 ,HO. 

 According to the same chemist, all the other ammonia-cobalt bases give this hydrate 

 by boiling with solutions of the alkalies. It does not appear probable that the 

 oxide obtained by boiling the neutral solution should have a different constitution. 

 We have however found that the dark-brown oxide obtained by boiling a solution 

 of sulphate of Roseocobalt, and afterward washing and drying the precipitate in 

 the air, has the formula 



CoA + 3IIO 

 as the following analyses show : 



I. 0.4430 grs. gave 0.6990 grs. sulphate of cobalt = 60.05 per cent, of cobalt. 

 II. 0.9269 grs. gave 0.1672 grs. water (ignited with chromate of lead) = 18.03 per cent. 

 III. 0.7150 grs. ignited in hydrogen gas gave 0.3010 grs. water, which in connection with the 

 2d analysis gives 21.38 per cent, oxygen in the oxide of cobalt. 



The formula requires 



Eqs. 

 Cobalt ... 3 

 Oxygen . . .4 

 Water ... 3 



147.5 100.00 99.46 



Fremy's formula requires 64 per cent, of cobalt. Claudet gives also the formula 

 Co 3 4 + 3HQ as probable, but without analyses. On the other hand, it is possible 





Calculated. 



Found. 



88.5 



60.00 



60.05 



32.0 



21.69 



21.38 



27.0 



18.30 



18.03 



