It might be, of course, that the mechanical conceptions were inadequate to deal with this new region of phenomena. It had been brought within the realm of mathematics; it had not been brought within the realm of mechanics. But to many physicists of the nineteenth century a mathematical description which established relations between concepts which were not mechanical was thereby incomplete. They could not feel satisfied with such descriptions. Lord Kelvin, a great and representative member of this school, went so far as to say that he could understand nothing of which he could not make a mechanical model. This passion for mechanical explanations was doubtless, to some extent, the result merely of mental inertia. Men, when their minds are mature, are reluctant to think except in terms to which they are accustomed. No a priori reason could be given for supposing that the Newtonian conceptions must be adequate for the description of the whole material universe. But the attitude, even if, as a psychological fact, it was largely based on prejudice, could be justified by the axiom that entities are not to be multiplied unnecessarily. Men were certainly justified in seeing whether or no electromagnetism could be reduced to mechanics. But they were hardly justified in rejecting Maxwell's theory, as Lord Kelvin did, because they could not give a mechanical explanation of it.

The attempt to reduce electromagnetism to mechanics took the form of designing ether models. We have already seen that the ether theory had never given a completely satisfactory account of all the

phenomena of light. Now that all radiation was regarded as electromagnetic, the ether was required to have such mechanical properties as would issue in Maxwell's equations. An extraordinary number and variety of ethers arose in response to this demand. Some were designed to exhibit electricity as a linear, and magnetism as a rotatory phenomenon. Others inverted the rôles, making magnetism linear and electricity rotatory. Some regarded the ether as an elastic solid; others abandoned the elastic solid analogy. Ethers were then invented which were not continuous media at all, but were complicated conglomerations of vortices. A refinement on these was invented by Kelvin, the so-called "vortexsponge "ether. All these ethers were successful in representing certain features of the phenomena to be explained, but none of them were completely successful. The one impression they all concur in giving to the reader is that they are incredibly complicated. This fact led some mathematicians, particularly in France, to wonder what their authors intended by these models. Suppose, for instance, that an ethermodel consisting of wheels, each wheel geared to four neighbours by india-rubber bands, the bands being of varying elasticity and also capable of slipping, suppose that such a model is shown to be capable of transmitting vibrations analogous to those of light, what precisely is proved? Is it to be supposed that there is a vast medium, filling interstellar space, constituted in this way? Or is it to be taken merely as a tour de force, an exhibition of human ingenuity, but corresponding to nothing in natural processes ?

It seems that the question will be answered differently according to whether one has a mystical conviction that the physical universe is a machine that could be reproduced on a small scale by a nineteenth century engineer in his workshop, or whether one believes that the concepts of mechanics are merely the first step in the endeavour to isolate abstractions that shall be sufficient for the mathematical description of nature. In any case, the attempt to give a mechanical description of electromagnetic phenomena has not hitherto been a success. As a consequence of this failure the inverse question has been propounded. Can an electrodynamic explanation be given of mechanics? Or must we regard the electromagnetic concepts as additional to those of mechanics, the complete science of physics using both groups? We shall see later what answer may be given to this question.

CHAPTER VII

THE ATOM OF ELECTRICITY

THE Faraday-Maxwell way of regarding electromagnetic phenomena laid stress upon the processes going on in the field. On the basis of that theory Poynting, as we have seen, obtained the energy accompanying an electric current from the surrounding space. Instead of the old action at a distance, the new theory rested upon equations that traced the electromagnetic actions from one instant of time to the next and from one point of space to the next. This way of regarding phenomena made the problem of the ether acute. The centre of interest, as it were, was transferred from matter to space. One consequence of this way of regarding phenomena, as we shall see, was Einstein's theory of relativity. But, side by side with the development of Maxwell's theory, an entirely different aspect of electrical phenomena was being investigated that was to lead to generalizations of equal importance.

We have already seen, from the experiments of Nicholson and Carlisle and of Davy, that when an electric current passes through a liquid conductor the liquid conductor is decomposed. This pheno

menon was investigated, by Faraday, amongst others, and, besides determining certain very important quantitative laws, he invented, for this class of phenomena, a terminology that has lasted to this day. Thus the metallic plates through which the current enters and leaves the liquid conductor are called electrodes. The liquid conductor is called an electrolyte, and the process of decomposition is called electrolysis.

Since the very beginning of the nineteenth century attempts were made to frame theories that should satisfactorily account for the phenomena of electrolysis. The earliest of these was advanced by Grothuss and by Davy. They supposed that the electric force applied to the liquid conductor dissociated its molecules into their constituent atoms, these atoms carrying opposite electrical charges.

This decomposing force is strongest in the neighbourhood of the electrodes. The positive electrode attracts the negatively charged atoms and repels those having a positive charge. The negative electrode behaves similarly, attracting positive atoms and repelling negative ones. In each case the repelled atoms attack their neighbours and decompose them. The atoms freed in this way attack, in turn, their neighbours, and so on. In this way a chain of decompositions and recompositions travels from each electrode to the other, the atoms that reach the electrodes being given off there. These free atoms are known as ions.

There are various difficulties in the way of this theory. Faraday established the law that the