i86 



SuppUfnent to ''Nature,'' Auj^ust 4, 1923 



Unfortunately, the ether is a very perfect, elusive, 

 liighly-endowcd substance, which makes no direct 

 appeal to any of our sense organs. Accordingly, it is 

 only investigated with some difticulty ; and its 

 properties are so differcnl Irom those of matter that 

 very elaborate and expensive arrangements have to 

 be made in order to cope with it. Even when we have 

 made those arrangements, it may decline to give an 

 answer, and the result may be negative. Still, a 

 truly negative result is something definite, and is better 

 than nothing. But every positive result is of extreme 

 value. Let us summarise the results we already 

 know. 



First of all, we know lliut the ether can transmit 

 waves at a definite and finite speed of 300,000 kilo- 

 metres a second. Next, that those waves are electro- 

 magnetic, with the electric and magnetic vectors at 

 right angles to each other, and in the same phase. 

 We also know that the superposition of related electric 

 and magnetic vectors in the same phase results in 

 propagation with the speed of light. 



Then we know that light cannot be transmitted by 

 conductors of electricity, which act like a solution 

 of optical continuity. Furthermore, inside a trans- 

 parent body light travels more slowly than in free 

 space, showing that the ether is affected somehow 

 by the neighbourhood of matter ; the amount of this 

 affection being sometimes called the refractive index, 

 which is the measure of the retardation experienced 

 by light, and sometimes, from another point of view, 

 the dielectric coefficient. 



We further know that if transparent matter is 

 moved in the direction of the light inside it, a certain 

 fraction of its velocity is added to the light — added 

 of course algebraically. Thus demonstrating, not 

 that matter has any power of conveying light — which 

 it has not — but that some influence or reaction on 

 the ether belongs to the matter and travels with it ; 

 that influence being just the one which effected the 

 retardation and is responsible for the refractive index, 

 the fraction of added velocity being, as surmised by 

 Fresnel, i - iIjj-^. 



We also know, from certain experiments conducted 

 by myself, that this property of matter does not 

 extend in the slightest degree beyond its boundary ; 

 so that however fast matter is moving, light just 

 outside it is not affected at all. Or, as we may express 

 it, matter has no power of carrying the ether with it. 

 The ether has nothing of the nature of viscosity. 

 If a fluid at all, it is a perfect fluid. Not even if the 

 matter is charged or magnetised does the influence 

 extend beyond its boundary so as to affect the stream 

 of light close to it. (See Phil. Trans. 1893 and 1897.) 



Yet inside transparent matter the phenomena of 

 light show that the ether must be modified in many 

 ways, giving rise to all manner of crystalline effects 

 — the optics of crystals — and the various phenomena 

 of polarisation ; especially that interesting one dis- 

 covered by Faraday, that the plane of vibration is 

 rotated, in one direction or another, by even non- 

 crystalline and fluid matter when immersed in a 

 longitudinal magnetic field : and those other phenomena 

 discovered by Kerr, all of which may be summed up 

 under the names elliptic and rotatory polarisation. 



It is easy enough to say that light is retarded to 



a definite extent by transparent matter, but the 



complete theory of it is not so simple. Something 

 about it will l)e found in the writings of Sir J, J. 

 Thomson, All the phenomena of dispersion and 

 anomalous dispersion must be taken into account 

 if we would understand the inter-relation between 

 matter and ether. 



New Effects. 



Not long ago the interesting phenomenon u.i 

 discovered by Prof. Richardson that the act (d 

 magnetisation rotates a piece of iron, and a quantila 

 tive investigation of this delicate effect has bet 

 made recently by Prof. Chattock and Mr. Bates, i 

 understand that a converse effect has now also l)een 

 observed by Mr. Barnett, namely, that rotating a 

 piece of iron magnetises it. I remember making an 

 attempt to discover such a phenomenon long a;.' 

 at University College, London ; but I found nothin 

 securely. Capricious and spurious effects were difllcu! 

 to avoid, and I suppose I had not sufficient perseveranc( . 

 We knew nothing in those days about electrons or 

 their orbits, though we felt that there was something 

 rotatory about magnetism ; nor was it more than 

 suspicion that electricity itself might possess a trai 

 of inertia, in addition to the recognised quasi-inert i 

 of self-induction. Modem skill may have been ablt. 

 to overcome the difficulties inherent to such an experi- 

 ment, but caution is desirable, since it is not clear 

 why rotation should develop one polarity rather than 

 another, if the atomic arrangement were truly random. 



There is more to be got out of the original discoverx 

 by Richardson than has yet appeared — and I ventui 

 to predict that we have by no means heard the la 

 of it. The Zeeman effect seemed small at one tim* 

 and if it be said that the Richardson effect could ha\ >_ 

 been anticipated, I reply that Larmor anticipated 

 the Zeeman effect ; though it is true he did not expert 

 the right magnitude, because the mass of the particl 

 responsible for radiation was not then known. Tin 

 quantitative relations of the Zeeman phenomenon 

 clearly showed, for the first time, that the radiatiiu 

 particle was one of electronic and not of atomic mas.s. 



I mention these two apparently disconnected 

 phenomena together advisedly ; for while the orienta- 

 tion or precession of electronic orbits in gases account 

 for the Zeeman effect, the orientation of electronic 

 orbits in iron accounts for the Richardson effect. 

 Both are small, but the Zeeman effect is the smaller 

 of the two ; it needs the appliances of spectrum analys: 

 for its detection. It is far bigger than it would ha\c 

 been if the atom had been the radiating element 

 instead of the thousand-times smaller mass of the 

 electron. As to the Richardson effect, it is surprising 

 that it has been observed and measured at all, for x\ 

 smallness (in mass) of the electron is no help to that : 

 and the detected reaction is not something optical 

 or etherial, but the gross movement of a mass of 

 ordinary matter. Not much movement, truly, — 

 quartz fibres must be used of course, and plenty of 

 refinements, — but still a material movement is observed 

 as the result of orientation of electronic orbits ; and 

 that is noteworthy. Reaction of radium from atomic 

 projectiles was observed before, and reaction of radio- 

 meter vanes too ; but alpha rays are atoms, and these 



