M. Mosander's Analysis of the Oxides of Iron. 187 



the magnesia caustic. As carbonate of potash is commonly 

 used for this operation, there is no risk of being exposed to this 

 inconvenience ; but it is evident, that, whenever the liquid con- 

 tains a salt with a base of soda, even though potash is used to 

 precipitate the magnesia, the double carbonate of soda and mag- 

 nesia ought to be precipitated. 



7. M. Mosandei^s Analysis of the Oxides of Iron, formed by 



continued Heat. 



It is well known that M. Berthier analyzed the cinder 

 which is formed on pieces of iron intended for being laminat- 

 ed to form iron-plate, and that he considered it as a new de- 

 gree of oxidation of iron, which contained \\ as much oxygen 

 as the protoxide. M. Berzelius kept a piece of iron, intended 

 for iron-plate, forty-eight hours in a furnace. The oxidated 

 crust was two lines thick, and exhibited, what M. Berthier 

 has also observed, two distinct layers. M. Mosander under- 

 took the analysis and examination of it. It was at first evi- 

 dent, that M. Berthier had analyzed together two distinct 

 substances. The interior layer is more black than the other, 

 has a grained fracture of little lustre, and is very slightly 

 magnetic. The exterior layer has a brilliant metallic lustre, 

 a bright and shining fracture, and a grey metallic colour, and, 

 what is very remarkable, has a powerful action on the mag- 

 netic needle. 



M. Mosander found that the layers consisted of 



Inner Layer. Atoms. Outer Layer. 

 Peroxide of Iron, 27 1 36 



Protoxide of Iron, 73 3 64 



The formula for the first is Fe+3Fe. The analysis of the 

 outer layer is the same as M. Berthier had obtained for the 

 two layers mixed together. This approached to Fe-\-%Fe. 



But, as on this occasion, the oxidated layer was supplied 

 with oxygen from the outer layer, and with iron from the in- 

 terior layer, it may be supposed that there was an insensible 

 gradation from the most oxidated to the least oxidated side. 



M. Mosander was thus led to analyze different zones of the 

 two layers ; and he found that at the exterior layer this was 



