Consider, for example, an extreme case—say a proposition of Euclid. Here we have a belief logically inferred from well-assured premises—so, at least, we were accustomed to suppose before mathematicians became so very fastidious in the matter of proof. Can we not say that in this case the elements of the two series are in a sense identical, that all the causes for our belief are also reasons for it? Certainly we are not moved by prejudice, or affection, or authority. It is neither self-interest nor party passion that induces us to believe, for example, that the three angles of a triangle are equal to two right angles. Has our thought, then, in this case freed itself from the dominion of a-logical conditions? Is our belief the child of uncontaminated reason? I answer—No. Though the argument, _qua_ argument, is doubtless independent of time, the argumentative process by which we are in fact convinced occurs in time, and, like all psychological processes, is somehow associated with physiological changes in the brain, These, again, are part of the general stream of physical happenings, which in themselves have nothing rational about them. Follow up this stream but a little further and every trace, not only of mind but of life, is completely lost; and we are left face to face with unthinking matter and its purposeless movements. Logical inference is thus no more than the reasoned termination of an unreasoning process. Scratch an argument, and you find a cause. If this be admitted, the question at once arises whether we can treat the two kinds of series thus intimately connected as separable when we are estimating the values of the beliefs with which they are both associated. Is it permissible, is it even possible, to ignore the genesis of knowledge when we are considering its validity? Do not origins qualify values? In many cases they notoriously do. A distinguished agnostic once observed that in these days Christianity was not refuted, it was explained. Doubtless the difference between the two operations was, in his view, a matter rather of form than of substance. That which was once explained needed, he thought, no further refutation. And certainly we are all made happy when a belief, which seems to us obviously absurd, is shown nevertheless to be natural in those who hold it. But we must be careful. True beliefs are effects no less than false. In this respect magic and mathematics are on a level. Both demand scientific explanation; both are susceptible of it. Manifestly, then, we cannot admit that explanation may be treated as a kind of refutation. For, if so, the more successfully science carried out its explanatory task, the more completely would it shatter its own principles. This way lies universal scepticism. Thus would all intellectual values be utterly destroyed.