^{page 97 note *} “Contributions to the Molecular Theory of Induced Magnetism,” Proc. Roy. Soc., xlviii, p. 342; Phil. Mag., Sept. 1890.

^{page 97 note †} Proc. Roy. Soc., Feb. 1922; Phil. Mag., March 1922.

^{page 98 note *} In a paper by Honda, and Okubo, on “Ferromagnetic Substances and Crystals in the light of Ewing's Theory of Molecular Magnetism,” Science Reports of the Tôhoku University, vol. v, Aug. 1916Google Scholar, also Phys. Rev., vol. x, p. 705, this point seems to have been overlooked. The distance between centres is taken there as probably more than twice the length of each magnet, which gives a wider range of stable deflection than is admissible.

^{page 105 note *} When i is made indefinitely small the ratio is 1 to or 2·598. As i increases the ratio is somewhat diminished, becoming 2·54 for i=1·05, 2·48 for i = 1·1, and so on.

^{page 107 note *} A description of these coils will be found in Proc. R.S.E., vol. xiii, 1885, p. 523.

^{page 107 note †} Searle, , Proc. Camb. Phil. Soc., vol. xii, 1902, p. 27.Google Scholar

^{page 113 note *} Yensen, “Magnetic and other Properties of Electrolytic Iron melted in vacuo,” Bulletins of the Engineering Station of the University of Illinois, Nos. 72, 83, 95.

^{page 117 note *} Phys. Rev., ix, p. 84, 1917; x, p. 661. This is confirmed by Westgren, A., Jour. Iron and Steel Inst., ciii, p. 303, 1921.Google Scholar

^{page 118 note *} Phys. Rev., ix, p. 85, 1917.

^{page 118 note †} Ibid., xvi, p. 464, 1920.

^{page 120 note *} For a summary of effects of stress see chapter ix of the author's book on Magnetic Induction in Iron and other Metals.

^{page 120 note †} Hull (Phys. Rev., May 1921) finds that the space-lattice of nickel is the face-centred cube. This gives each atom twelve nearest neighbours. If the shell electrons assume a corresponding grouping, the nickel model should have twelve “fixed” magnets, set in lines inclined to one another at 60°, 120°, and 180°.

^{page 121 note *} Jour. de Phys., iv, p. 469, 1905.

^{page 124 note *} Loc. cit., p. 315 et seq. “The so-called β iron has the same lattice as the α iron…. No difference has been found in the structure of iron below and above A2. At A3, however, the atoms of iron are completely rearranged, and the iron passes from one crystal class into another.” The magnetic change occurs at the lower arrest point A2.

^{page 124 note †} From known data as to the space-lattice of the iron crystal and the saturation limit of magnetisation in iron, it is easy to show that the moment of the Weber element in an atom of iron is very nearly 2 × 10^-20 c.g.s. units. (See Proc. Roy. Soc., February 1922, vol. 100, A, p. 453.)

^{page 125 note *} Industries, Sept. 19, 1890.

^{page 125 note †} Phil. Trans., vol. clxxxvii, A, p. 715, 1896.

^{page 126 note *} The words “inner” and “outer” are retained only as a convenient means of fixing the ideas. We are concerned here simply with relative motion between parts of the atom, with elastic relative displacements through a limited range, with the breaking away from a stable configuration, and with the dissipation of energy that attends the settling again into a stable configuration.

^{page 127 note *} Franck, and Hertz, , Verh. d. deutsch. Phys. Gesells., 1914, p. 457Google Scholar and p. 512; M'Lennan (and pupils), Proc. Roy. Soc., vol. 91, A, p. 485; 92, A, p. 305 and p. 574; Phil. Mag., Dec. 1918. See also Davis and Goncher, Phys. Rev., Aug. 1917 and Jan. 1919; Goncher, Proc. Phys. Soc., vol. xxxiii, p. 13; Horton and Davies, Proc. Roy. Soc., 95, A, p. 408; Phil. Mag., June 1921; Horton and Bailey, Phil. Mag., Oct. 1920; Mohler, Foote, and Stimson, Phil. Mag., July 1920.