244 Mr Adie, On Bismuth. 



and the gas was passed until no further trace of chloride distilled 

 over. 



On cooling, the tube was cut on both sides of the glass wool 

 plug and the products separated. 



By this method 50 gm. of the precipitate could be easily 

 handled without any leakage of chlorine into the room. 



9. The first fraction, R^ yielded a colourless chloride and left 

 very little residue, also colourless, the last fraction, R 4 , yielded a 

 yellow chloride and a bulky brownish residue. 



The bismuth chloride from R x was found to leave no residue 

 on redistillation in chlorine. It was dissolved in hydrochloric 

 acid, precipitated by water to free it from any traces of iron, 

 washed, dried, and reduced to the metal by fusion with potassium 

 cyanide. 



This pure bismuth showed no traces of impurity in quantity 

 likely to affect its properties either on chemical analysis or in 

 its spectrum. The oxide prepared from it was lemon-yellow, 

 much less fusible than before purification, did not become coloured 

 red so rapidly and did not change in weight on prolonged heating ; 

 it had also little or no action on a porcelain crucible in which it 

 was fused. 



The metal thus obtained was reserved for the atomic weight 

 determination. 



10. It appeared likely that any foreign element would be 

 found concentrated in the residue. This resisted the common 

 reagents except that it gave off hydrochloric acid when dropped 

 into concentrated sulphuric acid, and was partly dissolved by 

 caustic alkali. 



On reduction by fusion with potassium cyanide a dark grey 

 metallic powder was obtained, with a few small metallic beads if 

 all traces of bismuth had not been distilled off. 



When reduced in a current of hydrogen, hydrochloric acid 

 and steam were given off but only in small quantity, thus in one 

 case the residue gave on analysis 



Chlorine 1'96 



Oxygen 4'57 



Residue 93'47 



10000 



It hardly seemed likely that this substance would be a definite 

 compound, so some of its physical constants were determined. 



Thus it was found to have a density of 4*78 gm. per c.c. and a 

 specific heat of 01, while these constants for bismuth are 9"8 gm. 

 per c.c. and 0"03 respectively. Both of these values point to a 

 mixture. 



