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



Sulphuretted hydrogen 24-64 



Protoxide of antimony 75-36 



100-00 



It is easy to show that this is the true composition of kermes 

 mineral. Put a quantity of it into a glass tube, and expose it to 

 a heat of between 500° and 600° ; a quantity of water is extri- 

 cated, and common sulphuret of antimony remains. Thus by 

 heat a double decomposition is produced, the oxygen of the 

 oxide and the hydrogen of the gas unite together, and form 

 water, while the metal and the sulphur combine and form 

 the sulphuret. This experiment shows us that sulphuret of 

 antimony is a compound of one atom antimony + one atom sul- 

 phur, or of 



Antimony 5*5 or 73^- 



Sulphur 2-0 26|- 



7-5 100 



The analysis of sulphuret of antimony, which 1 published in 

 the Annals of Philosophy (First Series), vol. iv. p. 99, was not 

 quite accurate. 1 obtained 



Antimony 73*77 



Sulphur ." 26-23 



100-00 



But it is very difficult to come nearer the truth when we follow 

 the method by which that analysis was conducted. In rigid 

 experiments all nitrations and transvasations must be avoided. 



IV. Mercury. 



The experiments of Sefstrom leave little doubt that the weight 

 of an atom of mercury is 25 ; and that the protoxide is a com- 

 pound of 100 mercury and 4 oxygen, and the peroxide of 100 

 mercury and 8 oxygen ; for he actually obtained 



Mercury. Oxygen. 



Protoxide composed of 100 + 3*99 



Peroxide 100 + 7-99 



In these experiments the deviation was only one-fourth and 

 one-eighth of per cent. 



I tried a great many experiments to ascertain the point with 

 precision ; but the greater number of them failed. The crystals 

 of nitrate of mercury contain water, from which I attempted to 

 free them by placing them in the exhausted receiver of an air- 

 pump with sulphuric acid; but this method was not successful. 



