For our present purpose, suffice it to quote the account given in a popular treatise on metallurgy, published at the time when Bessemer was in the midst of his experiments. "Wrought iron," it says, "or soft iron, may contain no carbon; and if perfectly pure, would contain none, nor indeed any other impurity. This is a state to be desired and aimed at, but it has never yet been perfectly attained in practice. The best as well as the commonest foreign irons always contain more or less carbon.... Carbon may exist in iron in the ratio of 65 parts to 10,000 without assuming the properties of steel. If the proportion be greater than that, and anywhere between the limits of 65 parts of carbon to 10,000 parts of iron and 2 parts of carbon to 100 of iron, the alloy assumes the properties of steel. In cast iron the carbon exceeds 2 per cent, but in appearance and properties it differs widely from the hardest steel. These properties, although we quote them, are somewhat doubtful; and the chemical constitution of these three substances may, perhaps, be regarded as still undetermined." Now, in the Bessemer converter the carbon was almost entirely consumed. In the small gun just described,[25] there were only 14 parts of carbon for 1,000,000 parts of iron. Bessemer's next difficulty was to carburize his pure iron, and thus to make it into steel. "The wrought iron," says Mr. I. L. Bell, "as well as the steel made according to Sir Henry Bessemer's original plan, though a purer specimen of metal was never heard of except in the laboratory, was simply worthless. In this difficulty, a ray of scientific truth, brought to light one hundred years before, came to the rescue. Bergmann was one of the earliest philosophers who discarded all theory, and introduced into chemistry that process of analysis which is the indispensable antecedent of scientific system. This Swedish experimenter had ascertained the existence of manganese in the iron of that country, and connected its presence with suitability for steel purposes." Manganese is a kind of iron exceptionally rich in carbon, and also exceptionally free from other impurities. Berzelius, Rinman, Karsten, Berthier, and other metallurgists had before now discussed its effect when combined with ordinary iron; and the French were so well aware that ferro-manganese ores were superior for steel-making purposes that they gave them the name of _mines d'acier_. So Bessemer, after many experiments, discovered a method whereby, with the use of ferro-manganese, he could make what is known as mild steel. The process of manufacture, when described by Sir Henry Bessemer at Cheltenham in 1856,[26] was so nearly complete, that only two important additions were made afterwards. One was the introduction of the ferro-manganese for the purpose of imparting to his pure liquid iron the properties of "mild steel." The other was an improvement in the mechanical apparatus. He