Sec. 2. Now, leaning a little over the window sill, bring the edge of the paper at _p_ against the sky, rather low down on the horizon (I suppose you choose a fine day for the experiment, that the sun is high, and the sky clear blue, down to the horizon). The moment you bring your white paper against the sky you will be startled to find this bright white paper suddenly appear in shade. You will draw it back, thinking you have changed its position. But no; the paper is not in shade. It is as bright as ever it was; brighter than under ordinary circumstances it ever can be. But, behold, the blue sky of the horizon is far brighter. The one is indeed blue, and the other white, but the _white_ is _darkest_,[16] and by a great deal. And you will, though perhaps not for the first time in your life, perceive that though black is not easily proved to be white, white, may, under certain circumstances, be very nearly proved black, or at all events brown. Sec. 3. When this fact is first show to them, the general feeling with most people is, that, by being brought against the sky, the white paper is somehow or other brought "into shade." But this is not so; the paper remains exactly as it was; it is only compared with an actually brighter hue, and looks darker by comparison. The circumstances are precisely like those which affect our sensations of heat and cold. If, when by chance we have one hand warm, and another cold, we feel, with each hand, water warmed to an intermediate degree, we shall first declare the water to be cold, and then to be warm; but the water has a definite heat wholly independent of our sensations, and accurately ascertainable by a thermometer. So it is with light and shade. Looking from the bright sky to the white paper, we affirm the white paper to be "in shade,"--that is, it produces on us a sensation of darkness, by comparison. But the hue of the paper, and that of the sky, are just as fixed as temperatures are; and the sky is actually a brighter thing than white paper, by a certain number of degrees of light, scientifically determinable. In the same way, every other color, or force of color, is a fixed thing, not dependent on sensation, but numerically representable with as much exactitude as a degree of heat by a thermometer. And of these hues, that of open sky is one not producible by human art. The sky is not blue _color_ merely,--it is blue _fire_, and cannot be painted. Sec. 4. Next, observe, this blue fire has in it _white_ fire; that is, it has white clouds, as much brighter than itself as _it_ is brighter than the white paper. So, then, above this azure light, we have another equally exalted step of white light. Supposing the value of the light of the pure white paper represented by the number 10, then that of the blue sky will be (approximately) about 20, and of the white clouds 30.