possible value. The coercive force of steel is much greater than that of any other substance; the intensity of magnetization which it can retain is, therefore, relatively very great, and it is hence used for permanent magnets. The coercive force depends upon the quality and temper of the steel.
It was found by Ewing that the intensity of magnetization of a mass of iron in a magnetic field of given intensity is not dependent upon that intensity alone, but depends also upon the previous history of the magnetic body. In general the intensity of magnetization lags behind the magnetizing force; that is, if the magnetizing force be increasing, the intensity of magnetization is less for a given value of the force than it is for the same value if the force be diminishing. The relations between these two quantities are exhibited in Fig. 78, in which the magnetizing force is measured along the axis the intensity of magnetization along the axis The curve represents the relation of these quantities as the magnetizing force of the field changes from a high negative to a high positive value. The area between these curves may be shown to measure the work done on the magnet during the cycle this work is almost wholly expended in heating the magnet. The phenomenon here described is called magnetic hysteresis.
Changes of temperature cause corresponding changes in the magnetization of a magnet. If the temperature of a magnet be gradually raised, its magnetization diminishes by an amount which, for small temperature changes, is nearly proportional to the change of temperature. The magnet recovers its original magnetization when cooled again to the initial temperature, provided that the temperature to which it was raised was never very high. If it be raised, however, to a red heat, all traces of its original magnetization permanently disappear. Trowbridge has shown that, if the temperature of a magnet be carried below the tempera-
