248 On the Cause of Chemical Proportions. [April, 



Tungstic acid 'J\'42 100 



Oxide of lead 28-58 40 



100-00 



But these 40 of oxide of lead contain 2*86 of oxygen. This 

 coincides well with the analysis of tungstate of ammonia, and 

 shows us that if the result is not quite exact, this is owing to a 

 fixed matter mixed with the acid, perhaps potash. 



Wlien we ask how many volumes of oxygen are in tungstic acid we 

 hesitate between 3 and 6. The analogy of the arsenic and chromic 

 acids, together w'vch ilie circumstance that tungstate of ammonia 

 cannot be united with more ammonia ; and that it crystallizes 

 such as I have described it, in a liquid containing a great excess of 

 amn.onia, shows us that tungstic acid must contain 8 volumes of 

 oxygen. This corresponds likewise with the great specific gravity 

 of this metal. The volume of tungsten then weighs 2424-24. 

 The brown oxide is VV + 4 O, and the acid W + 6 O. 



5. Stibium, AntimoJiy. (Sb). I have already in a preceding dis- 

 sertation, described my experiments on this metal. 1 have pointed 

 out the difiiculties which prevented me from obtaining an exact 

 result. It is only in consequence of experiments repeated accord- 

 ing to the ideas which 1 have exiilained in this paper, that I con- 

 sider myself as having gained more decisive results. I had found 

 that 100 parts of this metal combine with 57-3 of sulphur, and 

 that this sulphuret dissolves in concentrated muriatic acid, and 

 forms muriate of antimony (murias stibiosus) and snlphureted 

 hydrogen, without any excess either of sulphur or hydrogen. 

 Hence it follows, that the oxide of antimony (oxidum stibiosura) 

 ought to be composed of 100 metal and 18-6 of oxygen. The 

 next oxide, according to the notions which I entertained at 

 that time, contained 1-i- as much oxygen, that is to say, 27 9. 

 But in none of my experiments on the composition of antimonious 

 acid, could I find that the metal absorbed that quantity of oxygen. 

 This I accounted for by an imperfect oxydation, for which opinion 

 I gave reasons in the dissertation already alluded to. But when I 

 consider the ratio of the oxygen in the acids in ous to that in the 

 acids in ic, which is generally as 2 : 3 ; but never as 3 : 4, I 

 think that I endeavoured in these experiments to obtain a result 

 which could not take place. 



I resumed therefore my experiments on the antimonious acid, 

 and I found that when pure antimony is oxidated in a phial by 

 nitric acid, the mixture evaporated to dryness in a platinum cru- 

 cible, and then heated till it becomes perfectly white, we obtain 

 always the same results. J found that 100 parts of metal treated in 

 this way produce always very nearly 124-8 of antimonious acid. 

 In my former experiments I obtained a somewhat greater result, 

 because I had always employed glass phials, which could not be 



