The different varieties are readily convertible, for the grey iron when melted and suddenly cooled becomes white, when cooled more slowly is mottled, and when carefully preserved from rapid loss of heat, retains its colour. On the other hand, experiments on a small scale have shown, that white cast iron, subjected to a heat equal to that at which the grey melts, and allowed to cool slowly, becomes grey. Hence their difference can hardly be ascribed to chemical constitution. Neither can the presence of a greater or less quantity of oxygen, as is sometimes supposed, produce the difference, for under circumstances in all other respects similar, except the rate at which they are cooled, iron of the three different varieties may be produced, We therefore feel warranted in rejecting the usual theory, particularly as the reception of it has rather impeded than advanced the manufacture of iron. The theory of Karsten is far more consistent with the facts, and is directly applicable to the practical purposes of the iron master. We shall endeavour to give a succinct exposition of this theory, introducing all that is necessary for its full explanation. The ores of iron, which are all oxides, are reduced by exposing them to the action of carbonaceous matter, at a high temperature. The carbon first separates the oxygen from the ore, which becomes metallic, but as it has for the carbon a high affinity, that substance tends to combine with it. The iron combined with carbon is rendered far more fusible than it is when pure, and thus readily melts; when the heat of the furnace is little more than is sufficient for effecting this fusion, the two substances are uniformly mixed, and probably form a compound analogous to a metallic alloy; this is the white cast iron. When the compound is exposed to a heat higher than is sufficient to melt it, a separation appears again to take place, the carbon tending to assume in part the form of plumbago, the iron to retain no more of carbon than is sufficient to keep it liquid at the new temperature, and thus passes from the state of cast iron to that of steel, and finally approaches to that of malleable iron. If the cooling take place slowly, the carbon, obeying its own law of crystallization, arranges itself in thin plates, and the iron, consolidating afterwards, fills up all the interstices with grains or imperfect crystals; and thus the mass assumes a dark grey colour, partly owing to the natural colour of the iron, but in a greater degree to the plumbago. When the cooling is rapid, the carbon still disseminated throughout the mass, does not crystallize separately, but the two substances again form an uniform compound.