(1) He may believe that what corresponds to his idea of a material object, for example, an apple, is in very many respects like the idea in his mind. Thus, he may believe that the odor, taste, color, hardness, etc., that he perceives directly, or as ideas, have corresponding to them real external odor, taste, color, hardness, etc. It is not easy for a man to hold to this position, for a very little reflection seems to make it untenable; but it is theoretically possible for one to take it, and probably many persons have inclined to the view when they have first been tempted to believe that the mind perceives directly only its ideas. (2) He may believe that such things as colors, tastes, and odors cannot be qualities of external bodies at all, but are only effects, produced upon our minds by something very different in kind. We seem to perceive bodies, he may argue, to be colored, to have taste, and to be odorous; but what we thus perceive is not the external thing; the external thing that produces these appearances cannot be regarded as having anything more than "solidity, extension, figure, motion or rest, and number." Thus did Locke reason. To him the external world as it really exists, is, so to speak, a paler copy of the external world as we seem to perceive it. It is a world with fewer qualities, but, still, a world with qualities of some kind. (3) But one may go farther than this. One may say: How can I know that even the extension, number, and motion of the things which I directly perceive have corresponding to them extension, number, and motion, in an outer world? If what is not colored can cause me to perceive color, why may not that which is not extended cause me to perceive extension? And, moved by such reflections, one may maintain that there exists outside of us that which we can only characterize as an Unknown Cause, a Reality which we cannot more nearly define. This last position resembles very closely one side of Spencer's doctrine--that represented in the last of the two citations, as the reader can easily see. It is the position of the follower of Immanuel Kant who has not yet repudiated the noumenon or thing-in-itself discussed in the last chapter (section 51). I am not concerned to defend any one of the varieties of Direct or of Hypothetical Realism portrayed above. But I wish to point out that they all have some sort of claim to the title _Realism_, and to remind the reader that, when we call a man a realist, we do not do very much in the way of defining his position. I may add that the account of the external world contained in Chapter IV is a sort of realism also. If this last variety, which I advocate, _must_ be classified, let it be placed in the first broad class, for it teaches that we know the external world directly. But I sincerely hope that it will not be judged wholly by the company it keeps, and that no one will assign to it either virtues or defects to which it can lay no just claim.