68 



NATURE 



[January 19, 1922 



except perhaps Larmor and I.orentz — thought of 

 an electron as of something attached to an atom, 

 inaking it an ion, in accordance with Faraday's 

 ielectrolytic ideas; and the notion of a free satellite 

 electron, inside the boundary of an atom, was of 

 later growth. In fact, it was a development largely 

 brought about by Zeeman's discovery. 



Parenthetically I may remark that there is some 

 risk of Dr. Stoney's contributions to science being 

 overlooked, partly because the Transactions of the 

 Royal Dublin Society are not so readily accessible 

 as some other publications, and partly because he 

 expressed himself in terms and ways not always in 

 accordance with ordinary custom. Let me put on 

 record here, therefore, that, at that early date, 

 1 89 1, he examined dynamically the problem of 

 satellite electrons perturbed from a simple orbit 

 by unknown forces. He deals with elliptic, apsidal, 

 and precessional motions, with periodic changes in 

 each, and clearly depicts the double and treble and 

 quadruple lines which would result. 



He is not dealing with perturbations excited by 

 some definite outside physical cause, such as a mag- 

 netic field applied to the source, but \yith the normal 

 series of lines observed by spectroscopists — Balmer, 

 Kayser and Runge, etc. ; and the inference he 

 draws is that many of the known groups can be 

 accounted for on the analogy of astronomical per- 

 turbations. The problem he set himself is thus 

 worded (p. 569) : — 



" We shall accordingly, for the present, regard 

 certain points in the molecules of the gas as acting 

 dynamically on an aether capable of receiving and 

 transmitting only transverse vibrations, and we have 

 to inquire what motions of these points within the 

 rnolecules would impart to the medium the oscilla- 

 tions which correspond to the observed lines in the 

 spectrum." 



To return from this digression. Whether on 

 account of my article in the Electrician, or because 

 I had written direct to Prof. Zeeman (probably for 

 the latter reason), he sent me the MS. of a finished 

 paper of his, giving the experimental details and 

 also his version of Ix)rentz's theory developed on 

 equations like those of the Foucault pendulum; 

 and this paper I at once communicated to the 

 Phil. Mag. for March, 1897 (vol. 43, p. 226), 

 adding a brief footnote to say that I had verified 

 the author's results .so far as related to emission 

 spectra and their polarisation. This memoir is now 

 reproduced as the first in the volume of Zeeman's 

 collected papers just issued, and it is printed in 

 four languages— Dutch, English, French, and 

 German ; but the English and other versions contain 

 an appendix, not in the original Dutch, giving an 

 account of the attempts made long ago by Faraday, 

 and likewise a theoretical anticipation by Prof. Tait 

 NO. 2725, VOL. 109] 



in 1875 (an anticipation based on Kelvin's general 

 theory of magneto-optic rotation), together with 

 the record of a contemporary failure experimentally 

 to detect any such effect. This appendix also re- 

 moves from competition some apparently similar 

 but not identical observations made by a M. Fievez. 



It is of interest to find that in this remarkable 

 and fundamental paper by Prof. Zeeman the pos- 

 sible effect of solar magnetism on the sun's radia- 

 tion is indicated as a subject for inquiry — a develop- 

 ment afterwards so brilliantly followed up by 

 Prof. Hale. 



In May, 1897, I communicated another note to 

 the Royal Society (Proc. Roy. Soc, vol. 61, 

 p. 413), in which details of the appearance of the 

 lines are given, and the curious complexity of some 

 of them; also, which surprised me, a difference 

 between the behaviour of the components of the pair 

 of sodium lines. The red cadmium line was also 

 examined, and other spark spectra. The substance 

 of this paper is reported in Nature, vol. 56, 

 p. 237. And in the same month (on May 19) I ex- 

 hibited the effect at the Royal Society soiree, as 

 appears from the following entry in the Year-book 

 for 1897, p. 119 : — 



" Demonstration of Zeeman' s Discovery of the 

 Broadening of Spectrum Lines by the Action of a 

 Magnetic Field on the Source of Light. Exhibited 

 by Prof. Oliver Lodge, F.R.S. 



" Sodium lines produced by an oxyhydrogen flame 

 between the poles of a powerful magnet are exam- 

 ined by means of a Rowland concave grating (the 

 one with which Mr. George Higgs photographed the 

 solar spectrum), and can be seen to broaden when- 

 ever the magnet is excited. A nicol or other 

 analyser shows that the light of changed refrangi- 

 bility is polarised, as it would be if the source of 

 radiation consisted of revolving electrified particles 

 whose motion is accelerated or retarded by magnetic 

 lines of force through the plane of motion. 



" Recent Observations. — By reason of reversals, 

 the usual appearance of each sodium line is as if it 

 were doubled; the magnetic field makes it appear 

 triple, or even quadruple. A nicol properly oriented 

 removes the magnetic effect. Dj shows it more 

 sharply than D.2. The new lines intrude into the 

 middle, after the fashion of Newton's rings.'' 



It may seem from this that the observation of 

 doublet^ and triplets, as indicated by the theory, 

 was made by me; but that I disclaim, as appears 

 in the volume under review, p. loi, since, though 

 I saw something like the real effect, I did not 

 apprehend it clearly as a pure precessional effect 

 (akin to that which Dr. Johnstone Stoney 

 had worked out long ago), and was inclined 

 to suppose that the magnetic acceleration and 

 retardation of frequency, acting on a random 

 collection of molecules, would be likely to 



