The most usual of the older processes of steel-making, still carried on at Sheffield and elsewhere, is known as the _cementation process_: it consists in heating bars of the best wrought iron in contact with charcoal, at a high temperature, for three or four weeks. At Sheffield the iron bars and charcoal are packed in alternate layers into troughs 14 ft. long by 3½ ft. deep and wide, constructed of slabs of siliceous sandstone 6 in. thick. The last layer of charcoal at the top is covered to a certain depth with a layer of refractory matter, and the flames from a furnace beneath are made to envelop the stone troughs or _pots_, as they are technically called, for a period of a week or more according to the thickness of the bars operated upon. These are generally 3 in. broad and from five- to six-eighths of an inch thick. When it is found by withdrawing a test bar for examination that the operation is complete, the fire is gradually diminished and the whole allowed to cool slowly, which requires about a fortnight. Instead of only charcoal, a mixture of powdered charcoal or soot with a little salt has been used by some makers—which mixture, technically called _cement powder_, has given its name to the process. In some works 16 tons or more of iron are treated in one operation. The bars are found unchanged in form, but increased in weight by perhaps 27 lbs. per ton, for carbon has combined with the iron, being apparently transferred in the iron from one particle to another. The surface of the bars becomes rough and uneven from a multitude of blebs or blisters, and hence they are called _blister bars_, and the steel of which they now consist is named _blister steel_. In this conversion we may suppose that the iron at its outer surface first enters into combination with carbon taken from the carbonic oxide gas, which would be produced by combustion of the charcoal with the limited quantity of air in its interstices, and the oxygen thus set free would immediately seize again on the surrounding charcoal, and by repeated changes of this kind in which the oxygen acts as a carrier of carbon to the iron, in which it is transferred inwards from particle to particle. The cause of the blisters has been much discussed: probably the cause is the formation and escape of a volatile compound of carbon and sulphur at the surface of the soft metal; for it is known that nearly the whole of the little sulphur in the wrought iron disappears in the cementation process. Blister steel is never homogeneous, for near the surface it always contains more carbon than within; the bars are therefore broken up into short lengths which are carefully assorted, bound together with wire, heated, welded together under a hammer or by rolling, and finally formed into a bar, which is stamped with the outline of a pair of shears, and is then known as _shear steel_, because this product was generally found the most suitable for making the shears used in dressing cloth.