Prof. Chandler Roberts has shown that in soft, tempered, and hardened steels respectively, the carbon present has distinct modes of existence, as is indicated by the widely different action of solvents on the metal in these 3 states. The question whether the carbon in steel is combined chemically or merely dissolved mechanically has been much discussed. Roberts inclines to the opinion that in white cast - iron and steel the carbon is merely dissolved. The phenomena of hardening and tempering steel witnessed in the workshop are briefly these: a piece of steel (by steel in this connection is understood a piece of the special metal which is sold as tool steel, shear steel, crucible steel, etc.) is forged into the shape of a tool, heated to redness, or as high a temperature as can be reached without running a risk of "burning" it, and, while still red, dipped into water and held there until cold. The metal is now found to be changed in its mechanical properties: it is technically known as "dead - hard "; that is, it is as hard as it is possible to make it - brittle, and proof against a file or drill.

If the piece is now placed on a hot plate, the surface will first assume a light straw colour, and as the temperature increases, the tint will change, the order being deep yellow, red, brown, purple, blue; at which last - named colour the metal, instead of being brittle, is springy and resilient. The first process is known as hardening; the second as tempering, or drawing down to a yellow, brown, or blue temper, the exact temper being practically obtained by dipping the metal in water as soon as the desired colour is reached. These are the phenomena in brief; in the workshop different brands of steel are treated in various ways, and the peculiar dips and tempering liquids are looked upon as trade secrets; but in principle the processes are as described. The problem to be solved is, why steel heated to redness and suddenly cooled should become so very hard, and when subsequently reheated should become softer and springy. So far as known, the properties of the steel depend upon the quantity of carbon, and upon its condition. The action of slowly - oxidizing agents upon cold rolled steel, made by the cementation process, appear to indicate that the carbon in such steel exists in the form of iron carbide, and is not mechanically diffused in the mass as simple carbon.

The iron carbide resists the action of the oxidizing agents, which readily attack the iron through which it is distributed; but whether all kinds of steel possess this alloy of iron and carbon remains to be proved, although for about half a century it has been suggested that tempered steel is a true alloy of carbon and iron, the carbon being present in varying proportions according to the temperature, and being rendered permanent by the sudden cooling. The case of chilled cast - iron is analogous, although in that the hardening is effected with liquid metal, and not with solid, as in the case of steel. The surface of cast - iron is case - hardened by heating it while closely surrounded with carbonaceous substances, the sudden cooling shutting up the carbon, and converting the skin into a kind of steel. The presence of carbon is visible to the eye in cast - iron, but the chilled portion is uniform in structure. Soft steel shows a high degree of molecular rigidity as compared with the hardest iron, but far less than that of tempered steel; which fact, Prof. Hughes thinks, indicates that in its softest state steel is still an alloy, though only feeble quantities of carbon may be held in that condition.

Hard - drawn iron shows that mechanical hardening has not diminished the freedom of the molecules in any great degree; but if the iron is even slightly alloyed with tungsten, sulphur, or other substances, it loses its molecular freedom. (Eng. Mech.)