Concerning the latter inference we shall by and by have something to say. Meanwhile this whole speculation as to the final cessation of cosmical work seems to me--as it does to my friend, Professor Clifford [3]--by no means trustworthy. The conditions of the problem so far transcend our grasp that any such speculation must remain an unverifiable guess. I do not go with Professor Clifford in doubting whether the laws of mechanics are absolutely the same throughout eternity; I cannot quite reconcile such a doubt with faith in the principle of continuity. But it does seem to me needful, before we conclude that radiated energy is absolutely and forever wasted, that we should find out what becomes of it. What we call radiant heat is simply transverse wave-motion, propagated with enormous velocity through an ocean of subtle ethereal matter which bathes the atoms of all visible or palpable bodies and fills the whole of space, extending beyond the remotest star which the telescope can reach. Whether there are any bounds at all to this ethereal ocean, or whether it is as infinite as space itself, we cannot surmise. If it be limited, the possible dispersion of radiant energy is limited by its extent. Heat and light cannot travel through emptiness. If the ether is bounded by surrounding emptiness, then a ray of heat, on arriving at this limiting emptiness, would be reflected back as surely as a ball is sent back when thrown against a solid wall. If this be the case, it will not affect our conclusions concerning such a tiny region of space as is occupied by the solar system, but it will seriously modify Sir William Thomson's suggestion as to the fate of the universe as a whole. The radiance thrown away by the sun is indeed lost so far as the future of our system is concerned, but not a single unit of it is lost from the universe. Sooner or later, reflected back in all directions, it must do work in one quarter or another, so that ultimate stagnation be comes impossible. It is true that no such return of radiant energy has been detected in our corner of the world; but we have not yet so far disentangled all the force-relations of the universe that we are entitled to regard such a return as impossible. This is one way of escape from the consummation of things depicted by our authors. Another way of escape is equally available, if we suppose that while the ether is without bounds the stellar universe also extends to infinity. For in this case the reproduction of nebulous masses fit for generating new systems of worlds must go on through space that is endless, and consequently the process can never come to an end and can never have had a beginning. We have, therefore, three alternatives: either the visible universe is finite, while the ether is infinite; or both are finite; or both are infinite. Only on the first supposition, I think, do we get a universe which began in time and must end in time. Between such stupendous alternatives we