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



stances indicated, on any existing hypothesis. I should add that 

 the black precipitate was found to contain traces of nitrogen and 

 chlorine, but in far too minute proportion to estimate quanti- 

 tatively. 



The colourless liquid always contains free ammonia, sal-am- 

 moniac, and cobalt (perhaps often wholly as chloride) . 



(1) 0*3086 grm. hexammoniochloride and 10 cubic centims. 

 of water were heated in a sealed tube at 100° for forty-two hours. 

 The nitrate, on treatment with standard hydric sulphate, fur- 

 nished 005525 grm. free ammonia. 



03086 grm. substance, similarly heated, gave a nitrate from 

 which hot hydric and argentic nitrate threw down 0"4956 grm. 

 argentic chloride. 



(2) 0*29843 grm. yielded, in the manner already mentioned, 

 0*0527 grm. free ammonia. 



0*29843 grm., treated as before, gave, on evaporation of the 

 nitrate with hydric and platinic chloride, &c, 0*6265 grm. 

 platinum. 



These numbers indicate an almost complete decomposition, 

 with a slight loss of ammonia : — 



Found. 



I. II. Theory*. 



Free ammonia . . 17*90 17*66 18*99 1 



Total ammonia 36-10 37*99 V ] 



Chlorine . . . 39*73 39*66 



cent. 



In the following experiment the decomposition took place in 

 presence of a somewhat greater relative quantity of water. 



(X) 1*5204 grm. substance and 52*4 cubic centims. water were 

 heated together at 90°-100° for forty-four hours in a soda-water 

 bottle, the mouth of which was closed by a plug of caoutchouc 

 kept tightly in its place by a screw and fittings. The slightly 

 coloured filtrate was diluted with water to 500 cubic centims. 

 Of these, 50 cubic centims. gave 0*01819 grm. free ammonia, 

 100 cubic centims. gave 0*4852 grm. argentic chloride, and 

 300 cubic centims., boiled with 2 grms. pure sodic hydrate, 

 yielded a precipitate which was converted into 0*2156 grm. co- 

 baltic sulphate. Hence the following percentages : — 



Free ammonia . . 11*96 

 Chlorine .... 39*47 

 Cobalt in the filtrate 9*07 



The invariable non-liberation of the whole of the ammonia, 



* Supposing the chlorine converted into sal-ammoniac and half the am- 

 monia set free. 



