The electronic structure and magnetic properties of atoms and ions can be fairly satisfactorily correlated and brought into rela tion with the magnetic properties of simple and complex molecules, of gases and liquids made up of these, and of polar salts. In most paramagnetic salts, however, it is necessary to assume the exist ence of a molecular field whose precise nature is still very obscure. The difficulties arising in the interpretation of the magnetic prop erties of metals are largely due to the fact that little is known as to the way in which the electronic configuration of the atoms is modified when the atoms unite to form crystals. The silver atom, for example, definitely has a magnetic moment (as shown by the Gerlach and Stern experiments) ; but solid silver is dia magnetic. The valency electrons may become free electrons, with a very small paramagnetic effect (as suggested by the Pauli theory referred to in the last section) which is more than counter balanced by the diamagnetism of the ionized atoms ; but the role of the valency electrons in metallic crystals is not yet clear.
Among metals, the most surprising magnetic properties are those shown by iron, nickel and cobalt. The maximum magnetic intensity is not greater than that which might be anticipated from the presence of ions of the metals, whose magnetic moments are known; indeed, as has been discussed under ferromagnetism, a satisfactory explanation of the magnetization observed can be given in terms of the properties of groups of atoms. Why these particular metals, however, should be so strongly ferromagnetic and should acquire a strong magnetization in a weak field is an outstanding problem. In terms of the Weiss theory, the question is why the positive molecular field, due in some way to the mutual action of the atoms, is so large. The ferromagnetism is probably connected with the way in which the atoms are united in the crystal structure, as is suggested by the fact that non-ferromag netic nickel films may be prepared by sputtering; but ferromag netism is not associated with any one particular type of crystal structure. It is interesting that manganese is an essential con stituent of the ferromagnetic Heusler alloys, for, as the neutral atom contains only one electron less than iron, by gaining or losing electrons it may acquire an electronic configuration similar to that of the neutral atom, or of the ions, of iron itself. Besides the general problem of ferromagnetism, there are many problems of detail in connection with the modification of the magnetic proper ties of ferromagnetics by the presence of impurities and by par ticular thermal and mechanical treatment, problems whose solution is of technical importance as well as of purely scientific interest.
The science of magnetism has a long history, but even in the last few years enormous advances have been made. Investiga tions, apparently remotely connected with the problems, have helped to solve magnetic questions, while purely magnetic investi gations have been of service in shedding light on wider problems of the structure of atoms and of matter generally.
BiBuocRAPHY.—General: S. G. Starling, Electricity and Magnetism (1924) general text-book; Bulletin of the National Research Council, vol. iii., "Theories of Magnetism" (1922) ; P. Debye, "Theorie der Elek trischen und Magnetischen Molekulareigenschaften" in Handbuch der Radiologie, vol. vi. (1925) ; E. C. Stoner, Magnetism and Atomic Struc ture (1926 bibl.) ; P. Weiss et G. Foex, Le Magnetisme (1926). See also J. H. Jeans, Electricity and Magnetism (1925 ; mathematical theory) ; A Dictionary of Applied Physics, edit. Sir Richard Glaze brook, vol. ii., "Electricity" (1922) ; Handbuch der Physik, vol. xv., "Magnetismus; Elektromagnetisches Feld" (Springer, 1927).
Magnetic Measurements: J. A. Ewing, Magnetic Induction in Iron and Other Metals (1900) ; A. Gray, Absolute Measurements in Elec tricity and Magnetism (1921) ; D. W. Dye, "Magnetic Measurements" in A Dictionary of Applied Physics (edit. Sir Richard Glazebrook, 1922) ; E. C. Stoner, T. F. Wall, Applied Magnetism (1926) ; T. Spooner, Properties and Testing of Magnetic Materials (1927). See also Handbuch der Physik, vol. xvi., "Apparate und Messmethoden fiir Elektrizitat und Magnetismus" (Springer, 1927). References to the original papers are given in these books. References to recent papers and summaries may be found in Science Abstracts, published by the Institute of Electrical Engineers. (E. C. S.)