Dr. E. J. Mills on certain Cobaltamines. 253 



(5) Experiment (2) was repeated, the time being* forty-five 

 hours. The free ammonia amounted to 0*02482 grm. 



(6) 0*4015 grm. chloride and 10 cubic centims. water, at 90°- 

 130° for forty-seven hours, gave 0*03485 grm. free ammonia. 



Found. 



(1) (2) (3) (4) (5) (6) Theory*. 



Chlorine .... 40*83 38*08 42*35 



Free ammonia 1008 11*43 8*27 868 13*52 



If we make the useful assumption that none of the chlorine in 

 (1) was obtained from undecomposed substance in the filtrate, it 

 appears from the operation 



(0-1003^ "g X 35 4083 )xl00 



that 10*45 is the real percentage in (3). This assumption is 

 probably sufficiently in accordance with fact to allow us to believe 

 that the whole of the ammonia which should become free does 

 not make its appearance; and the conclusion follows a fortiori 

 from a similar comparison of experiments (2) and (5) : 



(««+ ?»|^ 8 )l00 = 9- 2 per cent. 



Hydric nitrate, cobaltic chloride, and sal-ammoniac were proved 

 to exist in the "filtrate ;*" and the entire decomposition is evi- 

 dently in perfect accordance with that described in the preceding 

 section. 



III. Cobaltic OL-Pentammoniotrichloride. 

 This compound, the "roseochloride" of Gibbs and Genthf, 

 is the first of these ammoniotri chlorides in order of formation. 

 According to the account given by its discoverers, it is not easy 

 to obtain in the pure state — that is, free from the /3-chloride just 

 referred to. I have procured it without any difficulty in the fol- 

 lowing way : — 5 grms. of crystallized cobaltic chloride are dis- 

 solved in water with the aid of heat ; the solution is cooled di- 

 luted to 90 cubic centims., and mixed in rapid succession with 

 27*5 cubic centims. strong aqueous ammonia, and 2*5 grms. 

 potassic permanganate dissolved in 100 cubic centims. water. 

 The temperature of the liquids before mixing should not exceed 

 18°. After standing for twenty-four hours in a wide beaker 

 with free exposure to air, the solution is filtered from its man- 

 ganic deposit, neutralized with ordinary aqueous hydric chloride 



* Supposing the whole of the chlorine converted into sal-ammoniac and 

 the excess of ammonia liberated, as in the equation last given, 

 t Silliman's American Journal, vol. xxiii. p. 241. 



