[83] C.R., 1893, 116, 997.
[84] Journ. de Phys., 1896, 5, 53.
[85] Phil. Trans., 1887, 177, 373.
[86] Proc. Roy. Soc., 1885, 39, 513.
[87] Phys. Rev., 1902, 15, 321. The sign of the thermo-electric effect for nickel, as given by Rhoads, is incorrect.
[88] Proc. Roy. Soc., 1904, 73, 413.
[89] C.R., 1903, 136, 1131.
[90] Journ. Coll. Sci. Tokyo, 1906, 21, art. 4. The paper contains 40 tables and 85 figures.
[91] This deduction from Ewing’s theory appears to have been first suggested by J. Swinburne. See Industries, 1890, 289.
[92] R. Beattie (Phil. Mag., 1901, 1, 642) has found similar effects in nickel and cobalt.