The constants of the magnetic dispersion of light

1. The rotation of the plane of polarisation of light by a magnetic field provides perhaps one of the easiest approaches to a study of the spectroscopic behaviour of ordinary substances. The present work is an analysis of the available measures of the dispersion of this magnetic gyration. Several formulæ have been proposed for it, and these will be reviewed below, but it has proved most convenient to take one of them and use that; afterwards testing, to the rather limited extent possible, how far others would fit the facts. The test is made with the formula given by Becquerel in 1897, V =e /2mc 2 λdn/d λ , (1.1) wheren is the refractive index, λ the wave-length and V is Verdet’s constant, the rotation of the plane of polarisation per centimetre per gauss. If this formula fits, and we shall find that it does, it should give the value ofe/m . A few values were worked out by Becquerel himself on the rather meagre data available at that time, and later Siertsema obtained more accurate values for some other substances. All these gave e/m roughly of the right order of magnitude, and this fact was duly noted in the text-books and has been copied from one to another ever since, but usually without giving any numerical values at all. Since Larmor’s theorem fails to hold for molecules we should hardly expect to find the ordinary value, but nevertheless it seemed useful to analyse all the experimental measures available, so as to discover if any regularity would emerge. In the physical journals there are several results of this type for particular substances, but they are very much scattered, and it should prove convenient to collect together an analysis of all the substances for which the gyration has been measured. We are not attempting any deep theory of the matter, but merely a convenient summary which may prove useful when the time comes for a proper theory of the spectroscopic behaviour of ordinary substances of the type that has been so successful for monatomic gases and vapours. It is outside the scope of the present work to discuss the behaviour of the gyration of light of frequency very close to opaque bands ; this has been the subject of many experiments, but they are not by any means concordant, and take us deeper into the unknown theory than it is possible at present to go. To avoid this trouble we have limited ourselves to transparent substances, that is, to regions of the spectrum far from the bands which cause the optical effects. The data have been extracted from Landolt and Bornstein’s tables (edition 1921), in some cases supplemented from the original sources, and it has not, of course, been possible to assess the merit of each of the individual measures recorded. In a few cases there are measures of the gyration but not of the refraction; and we are greatly indebted to Dr. I. C. Somerville of the Chemical Department of the University of Edinburgh for measuring some of these refractive indices for us.