596 KEPORT— 1895. 



attained. It is the rejected theories which have been the necessary steps 

 towards formulating- others nearer the truth. It would be an extremely interest- 

 ing study to consider the history of these discarded theories ; to show the part 

 they have taken in the evolution of truer conceptions, and to trace the persistence 

 and modification of typical ideas from one stratum of theories to a later. I pro- 

 pose, however, to ask your attention for a short time to one of these special 

 theories — or rather to t\vo related theories — on the constitution of matter and of 

 the ether. They are known as the vortex atom theory of matter, and the vortex 

 sponge theory of the ether. The former has been before the scientific world for 

 a quarter of a century, since its first suggestion by Lord Kelvin in 1867, the second 

 for about half that time. In what I have to say I wish to take the position not of 

 an advocate for or against, but simply as a prospector attempting to estimate what 

 return is likely to be obtained by laying down plant to develop an unknown dis- 

 trict. This is in fact the state of these two theories at present. Extremely little 

 progress has been made in their mathematical development, and until this has been 

 done more completely we cannot test them as to their powers of adequately ex- 

 plaining physical phenomena. 



The theory of the rigid atom has been a very fruitful one, especially in ex- 

 plaining the properties of matter in the gaseous state ; but it gives no explanation 

 of the apparent forces which hold atoms together, and in many other respects it 

 requires supplementing. The elastic solid ether explained much, but there are 

 difficulties connected with it — especially in connection with reflection and refraction 

 — which decide against it. The mathematical rotational ether of MacCullagh is 

 admirably adapted to meet these difficulties, but he could give no physical concep- 

 tion of its mechanism. Maxwell and Faraday proposed a special ether for 

 electrical and magnetic actions. Maxwell's identification of the latter with the 

 luminiferous ether, his deduction of the velocity of propagation of light and of 

 indices of refraction in terms of known electrical and magnetic constants, will 

 form one of the landmarks in the history of science. This ether requires the same 

 mathematical treatment as that of MacCullagh. Lord Kelvin's gyrostatic model 

 of an ether is also of the MacCullagh type. Lastly, we have Lord Kelvin's labile 

 ether, which again avoids the objections to the elastic solid ether. In MacCullagh's 

 type of ether the energy of the medium when disturbed depends only on the twists 

 produced in it. This ether has recently been mathematically discussed by Dr. 

 Larmor, who has shown that it is adequate to explain all the various phenomena 

 of light, electricity, and magnetism. To this I hope to return later. Meanwhile, 

 it may be borne in mind that the vortex sponge ether belongs to MacCullagh's 

 type. 



Already before a formal theory of a fluid ether had been attempted, Lord 

 Kelvin ^ had proposed his theory of vortex atoms. The permanence of a vortex 

 iilament with its infinite flexibility, its fundamental simplicity with its potential 

 capacity for complexity, struck the scientific imagination as the thing which was 

 wanted. Unfortunatelj^ the mathematical difficulties connected with the discus- 

 sion of these motions, especially the reactions of one on another, have retarded the 

 full development of the theory. Two objections in chief have been raised against 

 it, viz. the difficulty of accounting for the densities of various kinds of matter, and 

 the fact that in a vortex ring the velocity of translation decreases as the energy 

 increases. There are two ways of dealing with a difficulty occurring in a general 

 theory — one is to give up the theory, the other is to try to see if it can be modi- 

 fied to get over the difficulty. Such difficulties are to be welcomed as means of 

 help in arriving at greater exactness in details. It is a mistake to submit too 

 readily to crucial experiments. The very valid crucial objection of Stokes to 

 MacCullagh's ether is a case in point. It drew away attention from a theory 

 which, in the light of later developments, gives great hope of leading us to correct 

 ideas. As Larmor has pointed out, this objection vanishes when we have intrinsic 

 rotations in the ether itself. A special danger to guard against is the importation 

 into one theory of ideas which have grown out of one essentially diS"erent. This 



> Vortex Atoms,' Proc. Boi/. Soc. Edin., vi. 94; Phil. Mag. (4), 34. 



