This pen that I hold in my hand seems, as I pass my fingers over it, to be continuously extended. It does not appear to present an alternation of filled spaces and empty spaces. I am told that it is composed of molecules in rapid motion and at considerable distances from one another. I am further told that each molecule is composed of atoms, and is, in its turn, not a continuous thing, but, so to speak, a group of little things. If I accept this doctrine, as it seems I must, am I not forced to conclude that the reality which is given in my experience, the reality with which I have contrasted appearances and to which I have referred them, is, after all, itself only an appearance? The touch things which I have hitherto regarded as the real things that make up the external world, the touch things for which all my visual experiences have served as signs, are, then, not themselves real external things, but only the appearances under which real external things, themselves imperceptible, manifest themselves to me. It seems, then, that I do not directly perceive any real thing, or, at least, anything that can be regarded as more than an appearance. What, then, is the external world? What are things really like? Can we give any true account of them, or are we forced to say with the skeptics that we only know how things seem to us, and must abandon the attempt to tell what they are really like? Now, before one sets out to answer a question it is well to find out whether it is a sensible question to ask and a sensible question to try to answer. He who asks: Where is the middle of an infinite line? When did all time begin? Where is space as a whole? does not deserve a serious answer to his questions. And it is well to remember that he who asks: What is the external world like? must keep his question a significant one, if he is to retain his right to look for an answer at all. He has manifestly no right to ask us: How does the external world look when no one is looking? How do things feel when no one feels them? How shall I think of things, not as I think of them, but as they are? If we are to give an account of the external world at all, it must evidently be _an account_ of the external world; _i.e._ it must be given in terms of our experience of things. The only legitimate problem is to give a true account instead of a false one, to distinguish between what only appears and is not real and what both appears and is real.