270 M. Arfwedson on [Aphil, 



of 100 parts of protoxide, when converted into black oxide, to 

 be 10*4 ; so that black oxide is composed of 



Manganese 100-00 



Oxygen 41*35 



I conceive that I may conclude from these experiments, that 

 we always find the weight of oxygen in the black oxide of man- 

 ganese prepared by means of nitric acid, a circumstance which 

 is often of considerable importance in the analysis of minerals, 

 in which the question often is to determine the composition of a 

 given quantity of oxide. Perhaps the best method of proceed- 

 ing may be to expose the oxide to a strong red heat till it is 

 converted into brown oxide, the composition of which is always 

 uniform, and of course the quantity of oxygen in it may be 

 known with great accuracy. 



According to the preceding experiments, the quantity of oxy- 

 gen in the protoxide is to that in the black oxide as 1 to \\ ; for 

 100 parts of the metal combine in the protoxide with 28*107 

 parts of oxygen, and this quantity multiplied by l-i makes 

 42*160. The brown oxide, on the other hand, appears from the 

 quantity of its oxygen tro be intermediate between the protoxide 

 and black oxide ; but the oxygen constitutes no multiple either 

 of the oxygen in the protoxide or black oxide. It must, there- 

 fore, be a combination of these two oxides. If this compound, 

 like the oxydum ferroso-ferricum be supposed so constituted 

 that the oxide contains twice as much metal, and three times as 

 much oxygen, as the protoxide, then it will be a compound of 



Manganese 72-82 



Oxygen 27*18 



100*00 

 Direct experiments have given us 



Manganese 72*784 



Oxygen 27*216 



100*000 



numbers which correspond as nearly with the calculated quan- 

 tities as could be expected. On the other hand, we find that 

 this compound by calculation may be composed of 



Black oxide of manganese 68-932 



Protoxide of manganese 31*068 



100-000 



I have called this oxide oxidum manganoso-manganicum, on 

 account of its resemblance to the oxidum ferroso-ferricum, the 

 constitution of which has been explained by Prof. Berzelius in 



