The doctrine of Hume's which next confronts us is his famous one concerning Cause and Effect. He commences it by explaining that all objects of human enquiry are divisible into two kinds--1. Relations of Ideas, like those of which geometry, algebra, and arithmetic treat, and which are either intuitively certain, or 'discoverable by the mere operation of thought, without dependence on what is anywhere existent in the universe,' as, for example, the truths demonstrated by Euclid, which would be equally incontestable even 'though there were never a circle or a triangle in nature:' 2. Matters of Fact, as, for example, the sun's rising and setting, or the emission of light and heat by fire, which are never discoverable by unassisted reason, because of no one of them would the opposite imply a contradiction or be consequently inconceivable; and in our knowledge of any one of which we can never 'go beyond the evidence of our memory and senses,' except by means of reasons derived from experience of some fact or facts connected in some way or other with the particular matter of fact we are considering. So far, all is comparatively plain sailing, but Hume now propounds a difficulty which he at first presents as seemingly insurmountable, but which I cannot help thinking to be mainly of his own creation, and which he himself, almost immediately afterwards, suggests a mode, though a very inadequate mode, of overcoming. His language here is not marked by his usual perspicuity, or rather--to speak without respect of persons--it contradicts itself in most astounding fashion; but his meaning is not the less certainly the following, for there is no other construction which his words will bear. 'What,' he asks, 'is the foundation of all conclusions from experience?' Why is it that, having found that such an object has always been attended with such an effect, we infer that similar objects will always be attended with similar effects? The proposition that a certain antecedent has always been followed by a certain consequent, and the proposition that the same antecedent will be followed by the same consequent, are not identical. What, then, is the connexion between them which causes one to be inferred from the other? The connexion is unhesitatingly pronounced by him to be neither intuitively perceived, nor yet to be 'founded on any process of the understanding.' If you insist that the inference is made by a chain of reasoning, he challenges you to produce that reasoning, and taking for granted that you have none to produce, he proceeds to indicate what principle it is which, in his opinion, does determine us to form the inference. That principle he declares to be custom or habit, by which alone, he asserts, we are, after the constant conjunction of two objects, determined to expect the one from the appearance of the other; adding that all inferences from experience are effects of custom, not of reasoning.