When steel is heated red and plunged into cold water, it is _hardened_; that is, it becomes much harder than iron or than steel without this operation. Hardened steel is brittle, and cannot be extended by the hammer or corroded by a file till it is again softened by being heated and then gradually cooled. One of the most remarkable properties of hardened steel is that of being _tempered_, as it is called; by which it is adapted to the different purposes of the manufacturing artists. Tempering consists in heating the hardened steel till it acquires a straw colour for edge tools, a blue colour for watch springs, and elastic articles in general; &c. &c. Hardened steel is qualified to acquire magnetism, and to retain it so as to become a permanent magnet. This power of retaining magnetism distinguishes steel from pure iron. From the above account of steel, it is evident there is an essential difference between it and pure iron. That difference consists, according to the common opinion, in steel being a _carburet_ of iron, or carbone and iron united. The fact of the union of carbone and iron in the formation of steel does not seem to me satisfactorily proved. Mr. Collier asserts that iron gains about ¹/₁₈₀th of its weight by being converted into steel.[18] But Mr. Mushet found that though steel gains weight upon the iron when copiously imbedded in charcoal, yet it loses weight if the charcoal is only ¹/₉₀ or ¹/₁₀₀ of the weight of the iron.[19] The same ingenious gentleman seems to estimate the carbone in cast steel, from synthetic experiments, to be ¹/₁₀₀th of its weight. [Footnote 18: Manchester Memoirs, Vol. v. page 120.] [Footnote 19: Philos. Mag. Vol. xiii.] From analytic experiments, however, there does not appear reason to believe that steel contains so much, if any charcoal. Pure steel dissolved in dilute sulphuric acid gives hydrogen gas containing no carbonic acid nor oxide, neither is there any appreciable residuum of any kind in general.