As we are thus incessantly, and thus intensely, occupied with cases of constant conjunction, while cases of casual conjunction pass slightly over the mind, and {365} quickly vanish from our consciousness, every event calls up the idea of a constant antecedent. The association is so strong, that the combination is necessary and irresistible. It often enough, indeed, happens, that we do not know the constant antecedent of an event. But never does it fail to call up the idea of such an antecedent; and so inseparably, that we can as little have and not have the idea of an event, as we can have the idea of it, and not have the idea of an inseparable antecedent along with it.--Ignorant, sometimes, of the constant antecedents of such and such events, we find them out by subsequent inquiry. Those cases of successful investigation still further strengthen the association. All that we call good, and all that we call evil, depend so entirely upon those constant conjunctions, that we are necessarily under the strongest stimulus to find them out, and to trace them with greater and greater accuracy. Thus we very often find a constancy of sequence, in which we acquiesce for a while; but after a time discover, that though constant, indeed, it is not immediate; for, that between the event and supposed antecedent, several antecedents intervene. At first we regard the ignition of the gunpowder, as the immediate antecedent of the motion of the ball. Better instructed, we find that a curious process intervenes. The constancy of the sequence is always more certain, the more nearly immediate the antecedent is. And so frequent is our detection of antecedents, more immediate than those which we have just observed, that an association is formed between the idea of every antecedent, and that of another antecedent, as yet unknown, intermediate between it and the consequent which we {366} know. In no sequence do we ever feel satisfied that we have discovered all. We see a spark ignite the gunpowder, we see one billiard-ball impel another. Though we consider these as constant antecedents and consequents, the idea of something intermediate is irresistibly conjoined. To this, though wholly unknown, we annex a name, that we may be able to speak of it. The name we have invented for this purpose is POWER. Thus, we conceive that it is not the spark which ignites the gunpowder, but the _power_ of the spark; it is not one billiard-ball that moves the other, but the _power_ of the ball. The Power, in this case, is a _supposed_ consequent of the moving ball, and antecedent of the moved; and so in all other cases.