Immediately over each vat is constructed a metal framework, through which are inserted a large number of carbon rods about eighteen or twenty inches long and from two to two and one-half inches in diameter. This framework is electrically insulated from the iron crucibles. The framework and the carbons are connected with the positive conductor of the electric current, and the vat or crucible with the negative. These conductors are very large, something like a foot in width and an inch in thickness, and made of some good conductor of electricity. They have to be very large because they carry a current equal to 3050 horse-power. The current is one of great volume, but very low voltage; the electromotive force at each vat or crucible being only about seven volts. As the process is electrolytic, and not simply a heating process, the direct current must be used, and therefore the current coming from the power-house must be transformed twice; first to bring it to a proper voltage and secondly to change it from an alternating to a direct current. These iron vats or crucibles are connected up in series, electrically, and then they are filled with the alumina and certain other materials, which act either as a flux or as a means of increasing the conductivity of the mixture; just what this substance is, is probably one of the secrets of the process. When all of the crucibles are filled with the mixture the current is turned on and is kept on continuously night and day seven days in the week. All of the material in the different crucibles is heated to redness, when the process of separation takes place. The oxygen of the alumina is thrown off as a gas, and other residuum floats to the top of the crucible and is skimmed off. Metallic aluminum in a melted state sinks to the bottom of the crucible, where it is dipped out from time to time with large iron ladles and poured into sand and molded into blocks similar to that of pig iron. From time to time, as the metal is dipped out, fresh alumina with the other substances are thrown in on top of the crucible, so that the process is continually going on, day and night, week in and week out. The heat in the process of reducing alumina, as we have before seen, is not the chief factor; it simply serves to reduce the compound to a fluid state so that the electrolytic action can readily take place. Therefore it is not necessary to be brought to a white heat, as it is in the case of the production of carborundum, described elsewhere.