every state introduces a new one, which is of the same kind as that by which it was itself brought about, but individually is new. Only here the matter is complicated, because the chain no longer consists of links of two kinds, but of many kinds, so that a link of the same name only recurs after several others have intervened. But we always see before us only an application of the single and simple law of causality which gives the rule to the sequence of states, but never anything which must be comprehended by means of a new and special function of the understanding. Or is it perhaps advanced in support of the conception of reciprocity that action and reaction are equal? But the reason of this is what I urge so strongly and have fully explained in the essay on the principle of sufficient reason, that the cause and the effect are not two bodies, but two successive states of bodies, consequently each of the two states implicates all bodies concerned; thus the effect, _i.e._, the newly appearing state, for example, in the case of an impulse, extends to both bodies in the same proportion; therefore the body impelled produces just as great a change in the body impelling as it itself sustains (each in proportion to its mass and velocity). If one pleases to call this reciprocity, then absolutely every effect is a reciprocal effect, and no new conception is introduced on this account, still less does it require a new function of the understanding, but we only have a superfluous synonym for causality. But Kant himself, in a moment of thoughtlessness, exactly expressed this view in the “Metaphysical First Principles of Natural Science” at the beginning of the proof of the fourth principle of mechanics: “All external effect in the world is reciprocal effect.” How then should different functions lie _a priori_ in the understanding for simple causality and for reciprocity, and, indeed, how should the real succession of things only be possible and knowable by means of the first, and their co-existence by means of the second? According to this, if all effect is reciprocal effect, succession and simultaneity would be the same thing, and therefore everything in the world would take place at the same moment. If there were true reciprocity, then perpetual motion would also be possible, and indeed _a priori_ certain; but it is rather the case that the _a priori_ conviction that there is no true reciprocity, and no corresponding form of the understanding, is the ground of the assertion that perpetual motion is impossible.