358 On the Proportion in which Bismuth unites [Nov. 



till the mixture detonated. The heat became in consequence so 

 high that the mass immediately became red-hot. The glass vessel 

 was now exposed to a red heat, in order to drive off any excess of 

 sulphur that might be present. 



When cold it weighed 15-2266 



J5ut the weight of the glass was 'I'WdO 



1 224/6 



Hence ten grammes of metal combined with 2'2476 of sulphur. 



To prevent the oxidizement of the bismuth in this operation, the 

 neck of the glass capsule was drawn out long and slender. The 

 colour of the sulphur which sublimed showed me that it carried no 

 bismuth along with it. I was not able to make experiments that 

 agreed with each other in such capsules. I therefore made choice 

 of small retorts with long and slender beaks, and I put the sulphur 

 into the bottom of the vessel. Many of my experiments failed, 

 because the retorts were not able to endure the sudden elevation of 

 temperature occasioned by the spontaneous combustion of the mix- 

 ture which they contained. 



Ten grammes of metal were put into a small retort, and mixed 

 with a sufficient quantity of sulphur. 



Grammes. 



The retort, after being exposed to a red heat, weighed . . . .31-8075 

 Subtracting the weight of the glass = 19-5800 



Remains for the weight of the sulphuret of bismuth formed . .12-2275 



The difference between this experiment and the preceding is 

 considerable ; but 1 found that a weight of 50 grammes, which had 

 been employed in weighing the retort, was 0-02 15 too light. This 

 deficiency must be added to our result, which will bring it to 

 12-2520. 



Subsequent experiments, made in the same way, gave the fol- 

 lowing results : — 



12-2065 



12-2230 



1 2-2465 



These results vary from each other in the second decimal place. 

 We must defer determining which of them is nearest the truth till 

 we have ascertained the quantity of oxygen with which bismuth 

 unites. Small differences in our experiments are unavoidable ; but 

 we have it in our power, notwithstanding, to determine on which 

 side the truth lies. 



2. With Oxygen. 



Into a dry and clean glass vessel, which weighed 37"! 775, there 

 were put'8'5045 parts of metal, which was boiled with nitric acid. 



