to the Knowledge of Chemical Bodies. 323 



Manganese, brown oxide .,,... 62^ 4*7264 



Nickel, black oxide 62 4-846 



Cobalt, black oxide 62 5'322 



Tungstic acid 62 5*274 



The firsj; fact I noticed in taking those densities, was that 

 the oxides were nearly, if not all, hygrometric, the quantity of 

 water they abstract from the atmosphere being regulated by 

 certain laws, which I shall endeavour to detail and explain in 

 a future paper. I conceive the overlooking this hygrometric 

 property of the oxides to be one of the principal causes why 

 the results of different chemists are so discordant, that instead 

 of calculating the equivalent weights of metals from their 

 oxides, they are obliged to have recourse to their salts for 

 that purpose. Let us take an example by way of illustration. 



I converted 100 grains of tin into gray oxide,, by solution 

 in muriatic acid, precipitating by potassa while recent: on 

 washing, drying, and weighing the oxide produced, it amount- 

 ed to 124-92 ; in a few minutes it became 126 grains; and in 48 

 hours it had absorbed as much moistui-e from the atmosphere 

 as increased its weight to 127*32. Now if we suppose two ex- 

 perimenters at work upon the same metal, the one weighing 

 his oxide hot, the other leaving it to a convenient opportunity, 

 their results would differ nearly 3 per cent. ; or if the second 

 only deferred weighing a few minutes, they would not agree by 

 1 per cent. 



There appears to be an intimate connexion between the 

 density of a metal and the quantity of oxygen it unites with ; 

 so much so, that I think when the course of accurate experi- 

 ments is complete, which I have explained above, the follow- 

 ing will turn out to be a general law. — All metals combine with 

 oxygen in the inverse ratio of their densities*. 



In order to show the extent of evidence which has pro- 

 duced this opinion, I shall give a table, where the metals 

 which are best known are arranged in the order of their den- 

 sities, from the heaviest down to the lightest, with a column 

 containing the smallest quantity of oxygen with which 100 

 parts of those metals are known to unite (protoxides); where 

 it will be perceived, that as the metals become lighter, the 

 quantity of oxygen they combine with increases. 



• We would here remark, that there are several instances of combina- 

 tion with oxygen among the metals, that appear to throw considerable doubt 

 upon the existence of this law: 100 parts of potassium, for instance, the 

 density of which is 085, take 20 parts of oxygen to form potash ; and the 

 same quantity of sodium, density 0!)7, requires 33*33 of oxygen to produce 

 soda ; both alkalies being protoxides. — Ed. 



S s 2 Table 



