}<■ 



256 Dr. E. J. Mills on certain Cobaltamines. 



or to 0*38 per cent, for the equation 



4C1 2 + NH 3 + 3H 2 = HN0 3 + 8HC1. 



It has been seen, however (continuation of Exp. X, Sect. I.), 

 that the actual yield of hydric nitrate is not by any means 

 accordant with the number (1'41) which flows from the latter 

 consideration ; neither does the corresponding quantity of am- 

 monia neutralized and oxidized in nitrification (HNO 3 : 9NH 3 : : 

 O06 : 0*14) go far to account for the loss. The same experi- 

 ment fortunately suggests an additional explanation. On com- 

 paring the percentage of free ammonia found with that indicated 

 by theory, there is a difference of 18*99 — 11'96=:7 , 03, which 

 number is to the cobalt in the filtrate (9*07) nearly as 6NH 3 to 

 Co 2 . It may be inferred that the compound 



[2CoCl 2 + 2(CoCl 3 .6NH 3 )] 



really existed in the filtrate, and that through its agency tetra- 

 cobaltic oxide was formed, 



[2CoCl 2 + 2(CoCl 3 .6NH 3 )] + (5 + rc)H 2 = Co 4 5 .rcH 2 0~\ 



+ 10(NH 3 .HC1)+2NH 3 



It is not improbable that this equation holds good also for the 

 pentammoniochloride to a certain extent, especially when we 

 remember that the hexammoniochloride is necessarily produced 

 by the ammonia which remains free throughout the reaction, 

 when the former chloride is decomposed by hot water : and I 

 need not point out that equation (b) receives, on this ground, 

 great additional strength. A special experiment showed that 

 the following mixture, viz. 



0-13425 grm. hexammoniochloride . . 2 (CoCl 3 . 6NH 3 ), 

 0-06550 grm. anhydrous cobaltic chloride (2CoCl 2 ), 

 20 cubic centims. water, 



is comparatively stable, and is far from undergoing complete 

 decomposition when heated under pressure for forty-seven hours 

 to 80°-100°; the presence of the supposed compound in any 

 one of the filtrates is therefore what might be expected under 

 the circumstances which have been described. It may, then, be 

 said that the oxidation of ammonia to hydric nitrate, and the 

 combination or mutual presence of cobaltic chloride and hexam- 

 moniotrichloride are sufficient causes to account for the non- 

 appearance of the expected ammonia : any deficiency not arising 

 from these sources must be attributed to incomplete decompo- 

 sition. 



(2) The above considerations induced me to investigate the 

 action of cobaltic chloride on the ^-pentammoniochloride and 

 hexammoniochloride. 



