Why should we speak of it? The universe is an assemblage of solar systems which we have every reason to believe analogous to our own. No doubt they are not absolutely independent of one another. Our sun radiates heat and light beyond the farthest planet, and, on the other hand, our entire solar system is moving in a definite direction as if it were drawn. There is, then, a bond between the worlds. But this bond may be regarded as infinitely loose in comparison with the mutual dependence which unites the parts of the same world among themselves; so that it is not artificially, for reasons of mere convenience, that we isolate our solar system: nature itself invites us to isolate it. As living beings, we depend on the planet on which we are, and on the sun that provides for it, but on nothing else. As thinking beings, we may apply the laws of our physics to our own world, and extend them to each of the worlds taken separately; but nothing tells us that they apply to the entire universe, nor even that such an affirmation has any meaning; for the universe is not made, but is being made continually. It is growing, perhaps indefinitely, by the addition of new worlds. Let us extend, then, to the whole of our solar system the two most general laws of our science, the principle of conservation of energy and that of its degradation--limiting them, however, to this relatively closed system and to other systems relatively closed. Let us see what will follow. We must remark, first of all, that these two principles have not the same metaphysical scope. The first is a quantitative law, and consequently relative, in part, to our methods of measurement. It says that, in a system presumed to be closed, the total energy, that is to say the sum of its kinetic and potential energy, remains constant. Now, if there were only kinetic energy in the world, or even if there were, besides kinetic energy, only one single kind of potential energy, but no more, the artifice of measurement would not make the law artificial. The law of the conservation of energy would express indeed that _something_ is preserved in constant quantity. But there are, in fact, energies of various kinds,[88] and the measurement of each of them has evidently been so chosen as to justify the principle of conservation of energy. Convention, therefore, plays a large part in this principle, although there is undoubtedly, between the variations of the different energies composing one and the same system, a mutual dependence which is just what has made the extension of the principle possible by measurements suitably chosen. If, therefore, the philosopher applies this principle to the solar system complete, he must at least soften its outlines. The law of the conservation of energy cannot here express the objective permanence of a certain quantity of a certain thing, but rather the necessity for every change that is brought about to be counterbalanced in some way by a change in an opposite direction.