122 Dr. Thomson's Experiments to determine the [Aug. 



found that I could obtain results agreeing more nearly with each 

 other than those which he describes. 



I. Bismuth. 



The salts of bismuth are all decomposed when they are placed 

 in contact with water. This prevents the possibility of deter- 

 mining' the weight of the oxide of bismuth in the same wav as I 

 ascertained that of the metals whose atomic weights have been 

 given in the former essays. It seems needless to relate the great 

 variety of unsuccessful attempts which I made with the nitrate 

 and sulphate of bismuth. I was not more successful with the 

 chloride. 



After a very great number of experiments, I gave up the salts 

 and the chloride in despair. I then dissolved a quantity of the 

 bismuth of commerce in nitric acid, decomposed the nitrate by 

 means of water, edulcorated the oxide, and reduced it to the 

 metallic state by heating it in a covered crucible with black 

 flux. Upon bismuth thus obtained (which I consider as pure), 

 I made the following experiment: Nine grains of it were put 

 into a platinum crucible, and dissolved in pure nitric acid. The 

 solution was evaporated to dryness by a gentle heat, which was 

 gradually raised to 600° ; and, finally, to redness. The oxide of 

 bismuth which remained weighed exactly 10 grains. Two repe- 

 titions of the experiment gave exactly the same result. Hence 

 I conclude that an atom of bismuth weighs exactly 9, and that 

 the weight of oxide of bismuth is 10. Hence it is obviously a 

 compound of one atom bismuth + one atom oxygen ; and not 

 two atoms, as Berzelius supposes. Indeed it is not difficult to 

 show, that 10 is the equivalent number for oxide of bismuth. 

 Weigh 10 grains of oxide of bismuth in a platinum crucible, pour 

 on it a quantity of pure sulphuric acid (the experiment will not 

 succeed with the acid of commerce), digest the whole upon a 

 sand-bath for 24 hours, then raise the heat gradually and slowly 

 till the excess of sulphuric acid is evaporated. The weight of 

 the sulphate of bismuth thus formed is exactly 15 grs. Now 5 

 is the weight of an atom of sulphuric acid. We see that 5 of 

 sulphuric acid are just neutralized by 10 of oxide of bismuth; 

 consequently 10 is the atomic weight of oxide of bismuth. 



To leave no ambiguity with respect to the weight of oxide of 

 bismuth, I put into a retort 15 grains of sulphate of bismuth, and 

 13 - 25 grains of chloride of barium with a sufficient quantity of 

 distilled water. The retort was put upon the sand-bath, and 

 kept for several days at the boiling temperature. It was then 

 placed in a cool part of the laboratc ry, and the fluid allowed to 

 settle. The clear liquid was neither precipitated by muriate of 

 barytes, nor by sulphate of soda. Hence it contained neither 

 sulphuric acid nor barytes. Now 13'25 of chloride of barium 

 yield one atom of barytes, just capable of saturating one atom of 

 sulphuric acid, which we have seen already was contained in 15 



