1893.] On the Metallurgy of Bismuth. 471 



metal, but appeared to rise as from a boiling centre, and this 

 although the metal was at a temperature very little above its melting 

 point. A portion of the film or layer was removed and tested in 

 order to ascertain its nature, and it was found to contain a very 

 appreciable proportion of antimony. I, therefore, caused the opera- 

 tion to be continued, stirring the metal from time to time with a 

 dried wood stirrer. In the course of three or four hours, removing 

 the film from time to time, the surface of the melted metal assumed a 

 much brighter appearance, and on carefully testing it at this point, 

 the metal was found to be absolutely free from antimony. To confirm 

 this and to ascertain more exactly the conditions under which this 

 separation takes place, a further quantity of impure bismuth was 

 operated upon in a similar manner. This second quantity contained 

 other impurities besides antimony, its analysis being as follows : 



Bismuth, by difference 96'20 



Antimony 0'80 



Tellurium 0'40 



Lead 2-10 



Copper 0-50 



Arsenic traces 



100-00 



The same simple process of fusion and stirring was again adopted 

 the quantity being about 350 kilogrammes and when the same 

 oily film commenced to rise to the surface the temperature of the 

 molten mass of the alloy was taken by means of the Le Chatelier 

 pyrometer. A portion of the film removed showed, on being tested, 

 a percentage of over 30 per cent, of antimony. A slightly perceptible 

 fume of arsenic was apparent as volatilising, so proving what I found 

 to be the case in the separation of arsenic by simple fusion. (See 

 ante.) 



The point at which this separation of antimony occurs was found 

 to be 350 C., and at this temperature the metal was maintained for 

 about five hours. 



The evidence of an oxidising action became now much less, and, 

 although a very small amount of antimony was present, there was still 

 a little remaining in the alloy ; the temperature, therefore, was slightly 

 raised and maintained at 458 C., as shown by the pyrometer, for 

 about four hours, at the end of which time the bismuth became 

 absolutely free from antimony. 



The form in which the antimony separated was peculiar a trans- 

 parent glass, consisting of antimony oxide containing about 10 per 

 cent, of bismuth, but of course in the removal of the antimony oxide 

 a small proportion of the bismuth was mechanically carried with it, 



