is a kind of assumed truth which cannot be reduced to its pragmatic meaning, because it must be true literally in order that the pragmatic meaning of other beliefs may be conceived or tested at all. Now, if it be admitted that the pragmatic theory of truth does not touch our knowledge either of matters of fact or of the necessary implications of ideas, the question arises: What sort of knowledge remains for pragmatic theory to apply to? Simply, Mr. Russell answers, those "working hypotheses" to which "prudent people give only a low degree of belief." For "we hold different beliefs with very different degrees of conviction. Some--such as the belief that I am sitting in a chair, or that 2+2=4--can be doubted by few except those who have had a long training in philosophy. Such beliefs are held so firmly that non-philosophers who deny them are put into lunatic asylums. Other beliefs, such as the facts of history, are held rather less firmly.... Beliefs about the future, as that the sun will rise to-morrow and that the trains will run approximately as in Bradshaw, may be held with almost as great conviction as beliefs about the past. Scientific laws are generally believed less firmly.... Philosophical beliefs, finally, will, with most people, take a still lower place, since the opposite beliefs of others can hardly fail to induce doubt. Belief, therefore, is a matter of degree. To speak of belief, disbelief, doubt, and suspense of judgment as the only possibilities is as if, from the writing on the thermometer, we were to suppose that blood heat, summer heat, temperate, and freezing were the only temperatures." Beliefs which require to be confirmed by future experience, or which actually refer to it, are evidently only presumptions; it is merely the truth of presumptions that empirical logic applies to, and only so long as they remain presumptions. Presumptions may be held with very different degrees of assurance, and yet be acted upon, in the absence of any strong counter-suggestion; as the confidence of lovers or of religious enthusiasts may be at blood heat at one moment and freezing at the next, without a change in anything save in the will to believe. The truth of such presumptions, whatever may be the ground of them, depends in fact on whether they are to lead (or, rather, whether the general course of events is to lead) to the further things presumed; for these things are what presumptions refer to explicitly.