by a number of ill-founded conjectures, and sometimes even by “ figments and falsehoods, which in the earliest times, no less than nowadays, used to be put forth by raw scatterers and copyists to be swallowed of men.”1 Thus it was taught that “ if a lodestone be anointed with garlic, or if a diamond be near, it does not attract iron, ” and that “ if pickled in the salt of a sucking fish, there is power to pick up gold which has fallen into the deepest wells.” There were said to be “ various kinds of magnets, some of which attract gold, others silver, brass, lead; even some which attract flesh, water, f1shes;” and stories were told about “ mountains in the north of such great powers of attraction that ships are built with wooden pegs, lest the iron nails should be drawn from the timber.” Certain occult powers were also attributed to the stone. It was “of use to thieves by its fume and sheen, being a stone born, as it were, to aid theft, ” and even opening bars and locks; it was effective as a love potion, and possessed “ the power to reconcile husbands to their wives, and to recall brides to their husbands.” And much more of the same kind, which, as Gilbert says, had come down “ even to [his] own day through the writings of a host of men, who, to fill out their volumes to a proper bulk, write and copy out pages upon pages on this, that and the other subject, of which they know almost nothing for certain of their own experience.” Gilbert himself absolutely disregarded authority, and accepted nothing at second-hand. His title to be honoured as the “ Father of Magnetic Philosophy ” is based even more largely upon the scientific method which he was the first to inculcate and practise than upon the importance of his actual discoveries. Careful experiment and observation, not the inner consciousness, are, he insists, the only foundations of true science. Nothing has been set down in his book “ which hath not been explored and many times performed and repeated ” by himself. “ It is very easy for men of acute intellect, apart from experiment and practice, to slip and err.” The greatest of Gilbert's discoveries was that the globe of the earth was magnetic and a magnet; the evidence by which he supported this view was derived chiefly from ingenious experiments made with a spherical lodestone or terrella, as he termed it, and from his original observation that an iron bar could be magnetized by the earth's force. He also carried out some new experiments on the effects of heat, and of screening by magnetic substances, and investigated the influence of shape upon the magnetization of iron. But the bulk of his work consisted in imparting scientific definiteness to what was already vaguely known, and in demolishing the errors of his predecessors.
N o material advance upon the knowledge recorded in Gilbert's book was made until the establishment by Coulomb in 178 5 of the law of magnetic action. The difficulties attending the experimental investigation of the forces acting between magnetic poles have already been referred to, and indeed a rigorously exact determination of the mutual action could only be made under conditions which are in practice unattainable. Coulomb, ” however, by using long and thin steel rods, symmetrically magnetized, and so arranged that disturbing influences became negligibly small, was enabled to deduce from his experiments with reasonable certainty the law that the force of attraction or repulsion between two poles varies inversely as the square of the distance between them. Several previous attempts had been made to discover the law of force, with various results, some of which correctly indicated the inverse square; in particular the German astronomer, I. Tobias Mayer (Giitt. Anzeiger, 1760), and the Alsatian mathematician, ]. Heinrich Lambert (Hist. de l'Acad. Roy. Berlin, 1766, p. 22), may fairly be credited with having anticipated the law which was afterwards more satisfactorily established by Coulomb. The accuracy of this law was in 1832 confirmed by Gauss, ” who employed an indirect but more perfect method than that of Coulomb, and also, as Maxwell remarks, The quotations are from the translation published by the Gilbert Club, London, 1900.
2 C. A. Coulomb, Mem. Acad. Roy. Paris, 1785, p. 578. °Intensitas vis magneticae, § 21, C. F. Gauss's Werke, 5, 79. See also I. ]. Thomson, Electricity and Magnetism, § 132. by all observers in magnetic observatories, who are every day making measurements of magnetic quantities, and who obtain results which would be inconsistent with each other if the law of force had been erroneously assumed.
Coulomb's researches provided data for the development of a mathematical theory of magnetism, which was indeed initiated by himself, but was first treated in a complete form by Poisson in a series of memoirs published in 1821 and later! Poisson assumed the existence of two dissimilar magnetic fluids, any element of which acted upon any other distant element in accordance with Coulomb's law of the inverse square, like repelling and unlike attracting one another. A magnetizable substance was supposed to consist of an indefinite number of spherical particles, each containing equivalent quantities of the two fluids, which could move freely within a particle, but could never pass from one particle to another. When the fluids inside a particle were mixed together, the particle was neutral; when they were more or less completely separated, the particle became magnetized to an intensity depending upon the magnetic force applied; the whole body therefore consisted of a number of little spheres having north and south poles, each of which exerted an elementary action at a distance. On this hypothesis Poisson investigated the forces due to bodies magnetized in any manner, and also originated the mathematical theory of magnetic induction. The general confirmation by experiment of Poisson's theoretical results created a tendency to regard his hypothetical magnetic fluids as having a real existence; but it was pointed out by W. Thomson (afterwards Lord Kelvin) in 1849 that while no physical evidence could be adduced in support of the hypothesis, certain discoveries, especially in electromagnetism, rendered it extremely improbable (Reprint, p. 344). Regardingit as important that all reasoning with reference to magnetism should be conducted without any uncertain assumptions, he worked out a mathematical theory upon the sole foundation of a few well known facts and principles. The results were substantially the same as those given by Poisson's theory, so far as the latter went, the principal additions including a fuller investigation of magnetic distribution, and the theory of magnetic induction in aeolotropic or crystalline substances. The mathematical theory which was constructed by Poisson, and extended and freed from doubtful hypotheses by Kelvin, has been elaborated by other investigators, notably F. E. Neumann, G. R. Kirchhoff, and Maxwell. The valuable work of Gauss on magnetic theory and measurements, especially in relation to terrestrial magnetism, was published in his Intensitas vis magneticae terrestris, 1833, and in memoirs communicated to the Resnltate aus den Beobachtungen des rnagnetisehen Vereins, 1838 and 1839, which, with others, are contained in vol. 5 of the collected Werke. Weber's molecular theory, which has already been referred to, appeared in 1852.5
An event of the first importance was the discovery made in 1819 by H. C. Oersted 6 that a magnet placed near a wire carrying an electric current tended to set itself at right angles to the wire, a phenomenon which indicated that the current was surrounded by a magnetic field. This discovery constituted the foundation of electromagnetism, and its publication in 1820 was immediately followed by A. M. Ampére's experimental and theoretical investigation of the mutual action of electric currents] and of the equivalence of a closed circuit to a polar magnet, the latter suggesting his celebrated hypothesis that molecular currents were the cause of magnetism. In the same year D. F. Aragossucceeded in magnetizing a piece of iron by the electric current, and in 1825 W. Sturgeon 9 publicly exhibited an apparatus “acting 45. D. Poisson, Mém. de l'In.vtitut, 1821 and 1822, 5, 247, 488; 1823, 6, 441; 1838, 16, 479.
For outlines of the mathematical theory of magnetism and references see H. du Bois, Magnetic Circuit, chs. iii. and iv. ° Gilbert's Ann. d. phys., 1820, 6, 295.
7Ann. de shim. et de iphys., 1820, 15, 59, 170; Recueil d'observa.tions électrodynarniques, 1822; T héories des phénomenes électrodynarniques, 1826.
8 Ann. de chim. et de phys., 1820, 15, 93.
° Trans. Soc. Arts, 1825, 43, 38.