"In order to ascertain whether this elevation of temperature was really alone owing to the absorption of the oxygen, a case containing twelve _grammes_ of this iron was placed under a receiver, which contained two hundred millimetres, inverted over a tube of mercury; and it was observed, in the course of an hour, that this air had diminished by forty millimetres, or one-fifth of its volume. Examining afterwards that which remained in the receiver, it was discovered that it had no effect whatever either on lime-water or the tincture of _tournesol_,--that it was not inflammable,--that it extinguished a candle; in a word, that it presented all the negative qualities of azotic gas, strongly infected with the smell before stated. "It must be evident that the oxygen which was absorbed in this experiment was employed solely in burning the iron, which was already in a state of _protoxide_, as was indicated by its little degree of cohesion, by the avidity with which it seized this principle, and by its dissolving in sulphuric acid, which operated without effervescence, and without disengaging hydrogen gas. "Five _grammes_ of this oxidized iron being reduced to an impalpable powder, and then made red-hot in a platina crucible, and mixed with three parts of _potasse à l'alcool_, were reduced to a clammy mass, coloured on its edges with a clear beautiful green, and with a greenish yellow on the other parts; which at once indicated the presence of a small portion of manganese, and that of a little _chrôme_; metals which are found united in almost all sorts of iron. Treated in the usual way, this mass exhibited-- "First, Traces, scarcely sensible, of these two metals. "Secondly, One _gramme_ of brilliant black powder, soft to the touch, staining paper, insoluble in muriatic acid when applied boiling: it was therefore a true percarburet of iron. "Thirdly, Three _grammes_ and ten _decigrammes_ of peroxide of iron. "On being subjected to the action of boiling water, five grammes of this pulverized iron gave out three _decigrammes_ of soluble matter, composed, for the greater part, of _hydrochlorate_ of iron, and a trace of hydrochlorate of magnesia, together with a little organic matter, the combination of which with the iron will account for the insufferable smell which it gave out when the iron was heated. This saline solution sensibly reddened the litmus paper: an effect which was owing to the muriatic acid, which, in uniting with oxidized iron, and with most other metallic oxides, never forms combinations which are perfectly neutral, but which are always more or less acid. "It has in vain been attempted to discover in this oxidized iron the presence of silex, of alumina, and of the sulphuric and carbonic acids, either free, or in combination. "It results from this analysis, that the fragments of the protectors, which have been the object of it, are composed, in a hundred parts, of about 64 oxidized iron,