NATURE 



265 



THURSDAY, JULY 19, 1900. 



THE RELATIONS BETWEEN ETHER AND 

 MA TTER. 

 'titer and Matter. By Joseph Larmor, Pp. xxviii + 

 365. (Cambridge University Press, 1900.) 



THIS work is essentially the same as an essay to 

 which an Adams Prize was awarded by the Uni- 

 versity of Cambridge. The subject for which the prize 

 was offered was Aberration, and as this phenomenon, 

 together with the Doppler effect on the frequency of 

 light vibrations are the only ones known due to the 

 motion of matter through the ether (the spelling aether 

 is disagreeably cumbrous),'it naturally led to a discussion 

 of the connection between ether and matter, and the 

 effect of their relative motion on the phenomena of 

 electro-magnetism. 



There is a good deal of similarity between the develop- 

 ment of this work and that of Maxwell's treatise. If one 

 reads Maxwell's papers, it is pretty evident that he began 

 with a somewhat definite hypothesis as to the nature of 

 the strains in the ether to which he attributed electro- 

 magnetic phenomena ; but in his treatise on electricity 

 and magnetism there is hardly a trace of all this 

 except in the reference to molecular vortices by 

 which he justifies his equations for wave-propagation 

 in magnetised media. In a similar way, Mr. Larmor 

 has published papers in which the relations between 

 ether and matter are developed in connection with 

 a suggestion as to the nature of an electron 

 which is only hinted at in the body of the present work 

 and is relegated to an appendix with some deprecatory 

 remarks as to its being merely an analogy to show that 

 the properties of an electron are not impossible. The 

 hypothetical structure attributed to an electron requires 

 the medium to possess a very remarkable property which 

 we do not find in matter, namely, an elastic reaction 

 against absolute rotation of its elements ; and although 

 Mr. Larmor shows that gyrostats connected with these 

 elements might confer such a property on them, he does 

 not go so far as to develop any very definite structure 

 for the medium, contenting himself with having shown 

 that such properties as he assumes are not necessarily 

 anti-dynamical. No structure for the medium that 

 depends on gyrostats supported by a rigid framework 

 can possibly be more than a rough working hypothesis, 

 being, in fact, very little better than the brass wheel and 

 india-rubber band, or tubes full of liquid with circulating 

 pumps, that have been suggested as models to show 

 that Maxwell's equations do not necessarily postulate 

 impossible or adynamical properties for the ether. 



But just as Maxwell's treatise is really independent of 

 the dynamical analogies from which it grew, so Mr. 

 Larmor's work is really independent of his suggested 

 working analogy as to the structure of an electron. The 

 whole work is based upon the hypothesis that electricity 

 is atomic in its nature, there being only two kinds of 

 atoms, positive and negative electrons. These electrons 

 are, he supposes, essentially centres of strain in the ether, 

 and move from place to place in much the same way 

 as a drop of water might move through ice, melting in 

 NO. 1603, VOL. 62] 



front and freezing up behind. Mr. Larmor leaves it for 

 future investigation to determine whether there is any 

 core, like that of a vortex ring, that accompanies this 

 complex strain wave as it moves through the ether. As 

 to the nature of matter, the only suggestion is that it 

 consists of clusters of electrons in orbital motion round 

 one another ; but as the dynamics of such a system has 

 never been worked out, it is impossible either to assert 

 or deny the possibility of a permanent existence of such 

 clusters. If this be the structure of matter, it certainly 

 makes it probable that the transmutation of the elements, 

 is a possible development of chemistry, while a structure 

 such as that of knotted vortices would make it improbable 

 that we would ever be able to untie them and thus 

 transmute one atom into another. There is the alterna- 

 tive possibility that we may find means of transmuting 

 elements within any one of their related groups, but that 

 we may find ourselves unable ever to transmute one 

 group into another. Of course, if we ever found out 

 some means of manufacturing electrons and matter we 

 could probably transmute one kind of matter into another, 

 though this latter might be possible according to Mr. 

 Larmor's hypothesis, while the manufacture of either 

 electrons or matter would be impossible. 



All theories that explain electric currents by the 

 motion of electrons are really based upon Rowland's 

 classical experiment, that a moving electric charge pro- 

 duces the same magnetic effects as an electric current, 

 and its converse that the electric force due to changing 

 magnetic induction produces the same effect in moving 

 an electric charge that the electric force due to another 

 electric charge would produce, i.e. that electric force due 

 to these two causes is the same. Our whole treatment 

 of electro-magnetism is practically based upon these 

 same assumptions, but it is remarkable that so few 

 attempts have been made to repeat this fundamental 

 experiment of Rowland's, and no successful attempt 

 seems to have been made to directly verify its converse. 

 In a recent number of the Cot)iptes rendus there is an 

 account of a most interesting attempt to measure the 

 electric current that one would expect to be produced in 

 a surrounding coil when a convection current such as 

 Rowland studied is being started and stopped. M. 

 Cremieu has carried out an experiment on this with 

 great care, and in a form in which one would certainly 

 expect that the changing magnetic induction due to the 

 magnetic force Rowland observed should produce an 

 induced current in a coil of wire. He observed no such 

 effect, and concludes that there is no magnetic force 

 such as Rowland observed due to a moving electric 

 charge. These moving electric charges are, however, 

 in some respects, so imperfectly known that there 

 may yet be some difference between driving a current 

 by mechanical and electrical forces, and that it is 

 still possible there may be some other explanation 

 than that drawn by M. Cremieu as the result 

 of his interesting and important experiment. Mr. 

 Larmor's investigations in this treatise of the effects 

 of moving matter hardly touch the question raised by 

 M. Cremieu, for his investigation is concerned with 

 steady states, while M. Cremieu's experiment is essenti- 

 ally concerned with variable ones. If he is right and 

 there is really no magnetic force due to a moving electric 



N 



