464 Scientific Notices — Chemistry. XDiRC, 



The soluble salts were found to consist of 



Carbonate of potash 0-63 



Sulphate of potash 0*37 



Silica a trace 



Too 



There was no muriatic nor phosphoric acid. The insoluble 

 matter gave by analysis : 



Silica 0-343 



Protoxide of iron 0-260 



Alumina 0-040 



Lime 0-052 



Potash ' 0-205 



Mamellated scoria 0*100 



1-000 



The alkali is undoubtedly derived from the earthy matter, 

 with which the carbonate of iron occurring with coal is always 

 intimately mixed. — (Annales des Mines.) 



4. Spontaneously/ Injiammahle Metallic Powders. 



M. Magnus has observed that when the pure oxides of iron, 

 cobalt, or nickel, are reduced by hydrogen gas at temperatures 

 but very little above that of boiling mercury, metals are ob- 

 tained, which, when allowed to cool in the hydrogen gas, inflame 

 spontaneously by exposure to the atmosphere. If the reduction 

 has been effected at a red heat, this does not take place. 



When the same oxides are mixed with a little alumine, the 

 metals obtained as before inflame spontaneously in the atmo- 

 sphere, even though the heat used has been that of redness, and 

 yet from the quantity of oxygen disengaged, it has been evident 

 that the alumine has not been de-oxidized. 



When a metal, thus competent to inflame in the air, is heated 

 in carbonic acid gas> it loses its pecuhar property, but re-assumes 

 it upon being re-heated in hydrogen gas, and allowed to cool as 

 before. 



Nevertheless, the hydrogen is not the cause of this inflamma- 

 bility ; for when oxalate of iron is heated in a vessel with a 

 narrow neck, so that the acid may be decomposed and the 

 whole allowed to cool, the metallic iron-powder obtained in- 

 flames spontaneously in the atmosphere. No other metal but 

 the three mentioned have presented this phenomena. 



It results from these experiments, that when the difficulty- 

 fusible metals are in a state of extreme division, and have not 

 aggregated either from adhesion or softness, they have the pro- 

 perty of inflaming in the air. This efiect is probably due to a 



