CHEMICAL SCIENCE. 233 



of concentrated, boiling hydrochloric acid. It appears to me evident that 

 siliciura exists in aluminum castings in the same state as carbon in gray 

 cast iron, a state as yet but little understood, but which my researches in 

 aluminum will enable me, I hope, to elucidate in some degree. 



This silicium is in brilliant metallic plates, similar to platinum, and 

 under this form it differs considerably from the silicium of Berzelius. 

 However, I do not think that this silicium is the real metal ; on the con- 

 trary, I think that this new form of silicium is to ordinary silicium what 

 graphite is to charcoal. This body possesses, with a more complete unal- > 

 terability , all the chemical properties which Berzelius attributes to the residue 

 of the incomplete combustion of ordinary silicium. Thus, to give an idea of 

 this indifference to the action of the most powerful re- agents, I will men- 

 tion, that the new silicium has been heated to whiteness, without changing 

 its weight, in a current of pure oxygen ; that it resisted the action of 

 hydrofluoric acid, and only dissolved in a sort of aqua regia, formed of 

 hydrofluoric acid and nitric acid. Potassa fusa transformed it into silicia, 

 but the operation took a very long time. It conducts electricity like 

 graphite. 



The aluminum castings from which I extracted the silicium contained 

 more than 10 per cent. It appears that, for the preparation of this casting, 

 the silicium must be in a nascent state at the moment of combination ; for 

 aluminum, melted in an earthern crucible, attacks the sides, frees the sili- 

 cium, but does not unite with it : * the metal preserves all its malleability, 

 and a chocolate-colored powder is found in the crucible, which is nearly 

 identical with the silicium of Berzelius. "SVe shall see, farther on, that 

 this casting is the first product which results from the action of the bat- 

 tery on chloride of aluminum and chloride of silicium, which always 

 exist together in the impure matters which are subjected to decomposition. 



I shall only give in this communication two methods of operating ; the 

 two which I know well, and have often employed. 



1st. Process with Sodium. Take a large glass tube of three or four cen- 

 timeters in diameter, introduce into it 200 or 300 grms. of chloride of alu- 

 minum, which are well isolated between two plugs of asbestos. By one 

 extremity of the tube, dry, pure hydrogen is introduced. Heat the chlo- 

 ride of aluminum in this current of gas with charcoal, so as to drive out 

 the hydrochloric acid, chloride of silicium and chloride of sulphur, with 

 which it is always impregnated. Then introduce into the tube some boats 

 as large as possible, containing each a few grms. of sodium crushed 

 between two sheets of very dry blotting paper. The tube being full of 

 hydrogen, the sodium is melted and the chloride of aluminum is heated ; 

 it distils and decomposes with an incandescence which can be moderated, 

 and even prevented, if desired. The operation is terminated when all the 

 sodium has disappeared, and the chloride of sodium formed has absorbed 

 sufficient chloride of aluminum to be saturated with it. The aluminum 



*I now prepare infusible and unattackable crucibles with calcined alumina, rendered 

 plastic by means of gelatinous alumina. 



