the invisible system with which they are in connection, and that A and B plus the invisible system, together form a system which obeys the ordinary laws of mechanics and whose momentum is constant. We meet in our ordinary experience cases which are in all respects analogous to the one just considered. Take for example the case of two spheres A and B moving about in a tank of water, as A moves it will displace the water around it and produce currents which will wash against B and alter its motion, thus the moving spheres will appear to exert forces on each other, these forces have been calculated by Kirchhoff and resemble in many respects the forces between moving electric charges, in particular unless the two spheres are moving with the same speed and in the same direction the forces between them are not equal and opposite so that the momentum of the two spheres is not constant, if, however, instead of confining our attention to the spheres we include the water in which they are moving we find that the spheres plus the water form a system which obeys the ordinary laws of dynamics and whose momentum is constant, the momentum lost or gained by the spheres is gained or lost by the water. The case is quite parallel to that of the moving electric charges and we may infer from it that when we have a system whose momentum does not remain constant the conclusion we should draw is not that Newton's Third Law fails, but that our system, instead of being isolated as we had supposed, is connected with another system which can store up the momentum lost by the primary, and that the motion of the complete system is in accordance with the ordinary laws of dynamics.
Returning to the case of the electrified bodies we see then that these must be connected with some invisible universe, which we may call the ether, and that this ether must possess mass and be set in motion when the electrified bodies are moved. We are thus surrounded by an invisible universe with which we can get into
