II. Atomic Theories. If matter is not continuous we

must suppose that every body is composed of aggregates of molecules. If so, it seems probable that each such molecule is built up by the association of two or more atoms, that the number of kinds of atoms is finite, and that the atoms of any particular kind are alike. As to the nature of the atoms the following hypotheses have been made.
(i) Popular Atomic Hypothesis. The popular view is that every atom of any particular kind is a minute indivisible article possessing definite qualities, everlasting in its form and properties, and infinitely hard.
This statement is plausible, but the difficulties to which it leads appear to be insuperable. In fact we have reason to think that the atoms which form a molecule are composite systems in incessant vibration at a rate characteristic of the molecule, and it is most probable that they are elastic.
Newton seems to have hazarded a conjecture of this kind when he suggested f that the difficulties, connected with the fact that the velocity of sound was one-ninth greater than that required by theory, might be overcome if the particles of air
* Descartes, Principia, vol. ii, pp. 18, 23. t Newton, Principia, bk. n, prop. 50.
CH. XXl] MATTER AND ETHER THEORIES 461
were little rigid spheres whose distance from one another under normal conditions was nine times the diameter of any one of them. This was ingenious, but obviously the view is untenable, because, if such a structure of air existed, the air could not be compressed beyond a certain limit, namely, about 1/1 021st part of its original volume, which has been often exceeded. The true explanation of the difficulty noticed by Newton was given by Laplace.
(ii) Boscovich's Hypothesis. In 1759 Boscovich suggested* that the facts might be explained by supposing that an atom was an infinitely small indivisible mass which was a centre of force — the law of force being attractive for sensible distances, alternately attractive and repulsive for minute distances, and repulsive for infinitely small distances. In this theory all action between bodies is action at a distance.
He explained the apparent extension of bodies by sapng that two parts are consecutive (or similarly that two bodies are in contact) when the nearest pair of atoms in them are so close to one another that the repulsion at any point between them is sufficiently great to prevent any other atom coming between them. It is essential to the theory that the atom shall have a mass but shall not have dimensions.
This hypothesis is not inconsistent with any known facts, but it has been described, perhaps not unjustly, as a mere mathematical fiction, and certainly it is opposed to the apparent indications of our senses. At any rate it is artificial, though it may be a prejudice to regard that as an argument against its adoption. To some extent this view was accepted by Faraday.
Sir William Thomson, afterwards Lord Kelvin, showed f that, if we assume the existence of gravitation, then each of the above hypotheses will account for cohesion.
* Philosophiae NaUtralis Theoria Redacta ad Unicam Legem Firnim, Vienna,
1759.
t Proceedings of the Eoyal Society of Edinburgh, April 21, 1SG2, vol. iv, pp. 60i— 606.
462 MATTER AND ETHER THEORIES [CH. XXI
(iii) Hypothesis of an Elastic Solid Ether. Some physicists tried to explain the known phenomena by properties of the medium through which our impressions are derived. By postu- lating that all space is filled with a medium possessed of many of the characteristics of an elastic solid, it was shown by Fresnel, Green, Cauchy, Neumann, MacCullagh, and others that a large number of the properties of light and electricity may be explained. In spite of the difficulties to which this hypothesis necessarily leads, and of its inherent improbability, it has been discussed by Stokes, Lam^, Boussinesq, Sarrau, Lorentz, Lord Rayleigh, and Kirchhoff.
This hypothesis was modified and rendered somewhat more plausible by von Helmholtz, Lommel, Ketteler*, and Voigt, who based their researches on the assumption of a mutual reaction between the ether and the material molecules located in it : on this view the problems connected with refraction and dispersion have been simplified. Finally, Sir William Thomson in his Baltimore Lectures, 1885, suggested a mechanical ana- logue to represent the relations between matter and this ether, by which a possible constitution of the ether can be realized. He also suggested later a form of labile ether, from whose properties most of the more familiar physical phenomena can be deduced, provided the arrangement can be considered stable ; a labile ether is an elastic solid, and its properties in two dimensions may be compared with those of a soap-bubble film, in three dimensions.
It is, however, difficult to criticize any of these hypotheses as a theory of the constitution of matter until the arrangement of the atoms or their nature is more definitely expressed.