In a paper read before the British Society of Arts, by Charles Sanderson, upon the sub, ject of Iron, he says, as regards the construction of furnaces, that generally speaking, they should have an internal form favorable to the gradual reduction and diminution of the volume of materials charged, which it is important should be so mixed that the earthy matter of the ore and the flux may readily unite j the descent of the materials into the furnace is regulated by the inclination and width of the boshes, and necessarily this inclination or width varies according to the nature of the ore which is to be melted— those which are most easily reduced require the boshes to be most inclined, while those which are difficult of reduction, and consequently require to be subjected to the action of the carbon of the fuel and the gases which are generated, for a longer period, are retained in this part of the furnace by greatly increasing its width, and giving the b08hes a greater inclination. In explaining this action of the blast-furnace, he showed how the metal is reduced and carbonized, from which it will appear how difficult it is to obtain a pure meLI, became, as it'hecomes developed in the lower regions of the furnace, it is necessarily mixed with substances formiug a variety of metalloidsj besides which, it is mechanically associated with the slag, which protects itin the dam from the oxydizing influence of the blast, through which it descends, carrying also with it a mixture of unreduced matter, which, from its gravity, becomes more or less mixed with the metal when it is in a state of pig-iron.