This is exceedingly pretty to watch. We will suppose that the article to be operated upon is the tyre of a wheel. The bar of iron has already been bent by rollers into the correct curve and the two ends are touching. Brought to the machine, it is gripped, each side of the junction, in the jaws of an insulated vice and the current is turned on. In a few seconds the place where the two ends are just touching begins to glow. Rapidly it increases in brightness until in about half-a-minute it is at welding heat. Then one vice, which is movable, is forced along a little by a screw, so that the ends are pressed firmly together, a little judicious hammering meanwhile helping to complete the job. Then the current is switched off and the complete tyre taken out of the machine. The current used has a force comparable with that which operates domestic electric bells, but in volume it is thousands of amperes. Alternating current is used, and it is obtained from a transformer or induction coil. In such a case the primary part of the coil is made of many turns of fine wire, so that little current passes through it, while the secondary part is but one or two turns of thick bar. Thus the voltage generated in the secondary is very little, but since the secondary has an almost negligible resistance the current caused by that small voltage is enormous. Such an arrangement is in industrial realms generally called a transformer, the term induction coil being employed more for those things of a similar nature intended for the laboratory. The one just described is, moreover, a "step-down" transformer, since it lowers the voltage, to distinguish it from "step-up" transformers, which raise the voltage. And the "resistance" principle is also applied in another way to large furnaces, such as those for refining iron. In these the resistance of the iron itself is utilised to generate the heat. Of course, it should be well understood, heat is always generated in everything through which current flows. There is no perfect conductor, and so every conductor is more or less heated by the passage of current through it. Some energy needs to be expended to drive current, even along large copper wires, and that energy must be turned into heat in the wires. If the same volume of current be forced along iron wires of the same size, the heat will be greater, since iron is but a poor conductor compared with copper, the relation being about as one to six. And if the iron be hot the resistance will be still more, for it stands to reason that when heated the molecules, being farther apart, will be the less easily able to exchange corpuscles. We have the best reasons for believing, as has been suggested already, that a current of electricity is but a flow of corpuscles, and so we are not surprised to hear that, as a general rule, the hotter a thing is the less does it conduct electricity. [Illustration: _By permission of Cambridge Scientific Inst. Co., Ltd., Cambridge, Eng._