The application of the quantum theory (q.v.) (introduced by Planck in 1899) to the problem of atomic structure by Niels Bohr in 1913, while it by no means solved all the difficulties, at once introduced order into chaos. A quantum theory outlook has since dominated atomic physics, and developments seem to have oc curred with unprecedented rapidity. For atomic magnetism, a fundamental feature of Bohr's theory is the fact that the angular momentum of an electron in an atom can only assume certain discrete values, and a natural unit is suggested for the associated magnetic moment, a unit roughly five times as large as the Weiss magneton. The most direct experiment to test the view suggested is that of deviating a stream of atoms in a non-homogeneous magnetic field. From the deviations observed the magnetic mo ments may be calculated. Such experiments have been carried out by W. Gerlach and 0. Stern (1921 and later). The general results confirm the predictions in a most remarkable manner. The mo ments found agree with those which may be calculated from the analysis of the normal atomic spectra and the "splitting" of the spectral lines in the Zeeman effect, that is when the emitting atoms are in a magnetic field (see ZEEMAN EFFECT). From a study of spectra, in fact, on a quantum theory basis, predictions can be made with considerable success as to magnetic suscepti bilities.
The association of angular momentum with magnetic moment suggests that magnetization should be accompanied by rotational impulse, an idea first developed by 0. W. Richardson (1908). The effect has been sought and found, but with the ratio of the magnetic to the mechanical effect twice that predicted. Of this anomaly, as well as of others connected with the analysis of spectra in relation to atomic structure, the simple quantum theory offered no explanation. Many of these difficulties can be cor related by attributing to the electron itself an angular and a magnetic moment as suggested by S. Goudsmit and G. E. Uhlen beck (1925).
The progress of the last twenty or thirty years has enabled magnetic phenomena to be traced back to the atom and even to the electron. The scheme is, indeed, far from complete. Many gaps are apparent when the attempt is made to account for mass magnetic phenomena, and there are many fundamental problems still unsolved. The rapidity of the recent advance is none the less astonishing; but this survey of the history shows that it has only been possible because the later investigators have been able to build on the foundations laid by those who preceded them.
Before 182o artificial magnets were made by stroking iron or steel with lodestone, a number of more or less efficacious methods being devised. The discovery of the fact that an electric current gave rise to a magnetic field led to the method of magnetizing iron that is now usually adopted. If a piece of iron is placed in
the interior of an elongated coil or spiral of wire (a solenoid) carrying a current, the iron becomes magnetized. If the current ceases to flow, or the iron is removed from the coil, the magneti zation decreases, but a certain amount is retained depending on the "retentivity" of the specimen. By using suitable steels arti ficial "permanent" magnets may be made, and for scientific pur poses these have replaced the natural magnet, lodestone or mag netite, which is an ore of iron, approximately of the composition Magnetic Poles.—If a bar magnet, produced by the solenoid method, is dipped into a mass of iron filings, the filings cling to it most thickly round the ends, where the attraction is greatest. These regions are the poles of the magnet. If the magnet be suspended horizontally in a stirrup by a thread of unspun silk, it will come to rest in a definite direction so that the line joining the poles of the magnet (the magnetic axis) lies in the magnetic meridian. The direction of the magnetic axis with reference to the body of a bar magnet may be determined by suspending it with first one and then the other face uppermost. The pole which points towards the north is known as the north-seeking, or more commonly the north pole, the other the south pole (convention ally the north pole is called the positive, the south the negative pole). By experiments with two magnets, one suspended, it is readily shown that like poles repel each other and unlike attract. As mentioned in the last section, Coulomb, by the use of the torsion balance, established (to within about 3%) that the force between two poles varied inversely as the square of the distance between them, and as the product of the "strengths" of the poles. The strengths may be most directly varied by building up a composite magnet. The law found, which may be confirmed by more accurate methods, may be expressed by the equation where m2 are the strengths of the poles, r the distance between them, and 1./ a constant for a particular medium. The force is one of repulsion when the poles are similar, and of attraction when they are opposite. This equation leads to a convenient way of defining the strength of a magnetic pole. The constant is trarily given the value tufty for a vacuum (the value for air will differ very slightly). With the centimetre-gram-second (c.g.s.) system of units for force and distance, the unit pole may then be defined as one which repels an equal pole at a distance of one centimetre with a force of one dyne. The units arrived at for electric and magnetic quantities by makingµ equal to unity in the above equation constitute the electromagnetic system of units.