impression; 2d, the appearance of the accidental image; 3d, the permanence for a longer or shorter period of the latter, its intensity and duration depending on the intensity and time of the direct impression, while the color gradually fades and then gives place to others; or if the eyes be successively opened and shut, or directed to light and dark surfaces in turn, a succession of appearances and disappearances of the image, usually attended with changes of its color. Any one may witness these curious results by taking a momentary glance at the sun or a candle flame, or looking steadfastly on a strongly lighted colored object, and then proceeding as already indicated. The subjective color obtained from the red sun of morning or evening will be a hue of green, passing, if the eyes be kept closed, gradually to darker hues, as blue and violet, or restored to red by looking momentarily on a white surface, and then followed by green on closing the eyes; that obtained from the yellowish-white sun of midday will be indigo or violet. In either case, if the eyes are turned at once to a white surface, the image at first appears of its darkest color, passing successively through the lighter colors to white; and whatever color appears when the eyes are closed, its complementary is seen when they are fixed on a white surface. Place, again, any small colored object upon a white surface, and look on it for some time: colors will be seen to develop themselves about the edges of the object, the color being in each case the complement of that of the object, bluish green surrounding red, and so on. The general explanation of these phenomena is, that the retina, having been once impressed with any color, gradually loses its sensibility to that color, and acquires a disposition to be affected by its complementary; and that this tendency is manifested both successively, or in time, and simultaneously, or in space. M. Scherffer considers that the continued or powerful action of certain colored rays enfeebles or fatigues the sensibility of the retina to those rays; so that when the eye afterward receives white light, it is affected for a, time by the other or complementary rays only. M. Plateau explains both the persistence of the original, and the appearance and changes of the successive accidental colors, by supposing these changes to constitute the transition in time of a portion of the retina from an excited to its normal state; while irradiation and the simultaneous accidental color constitute a similar transition in space, or from the actually excited portion of the retina to that which is in repose. When we look continuously on any color, as red, this color loses its vividness and beauty, because a color the opposite of red is excited in the eye, and blends with it; but its complementary, or any color near to this, as green, being now presented, the latter is at once improved, rendered more pure and vivid, by the acquired tendency of the eye to see that color. This is successive contrast; and it is thus shown that colors which will harmonize, or affect the eye agreeably, and be mutually improved, by being viewed in succession, are opposites or complements of each other. Colors nearly allied will be injured when thus beheld, and will affect the eye unfavorably. A purchaser who is shown in succession several pieces of bright red cloth will pronounce those last seen to be much inferior in brightness and beauty of color to the first; but if his attention be now directed to green stuffs, he will declare these extremely bright; and after them will see red stuffs quite as favorably. Again, two hues of red or blue seen side by side are not improved, because the impression made by either tends to excite an impression of green or orange in its neighborhood, which impression blends with the actual color of the other piece, and impairs it. But if blue and orange strips be viewed side by side, the blue throws orange upon the orange, and vice versa, so that the brilliancy and purity of both colors are improved. This effect constitutes simultaneous contrast; and it shows that harmony in colors viewed at once and near to each other also requires that these should be, or approach to, complernentaries of each other. If different tones be associated, the effect is always to make the tinted appear lighter, and the shaded darker, than it really is. This is easily shown by placing side by side several gray strips, passing gradually from quite light to dark; although the shade of each strip is homogeneous, yet its side toward a darker strip will appear to be the lighter, and that toward a lighter strip the darker. Chevreul's law, both for hues and tones, deduced from facts such as those now stated, may be thus expressed:
"When the eye sees at the same time two contiguous colors, they will appear as dissimilar as possible, both in optical composition and in height of tone. Guided by this principle, the juxtaposition of colors in painting, in dress, in furniture, in the planning of gardens, in bouquets, and indeed wherever colors are employed with a view to artistic effect, ceases to be a matter of accident or an ill-understood experience, and becomes a subject for the rules and predictions of science. In all chromatic arrangements, harmonies of contrast are first to be sought. But as these are limited, harmonies of analogy are also called into requisition, with less striking, but often with very pleasing results. These may be secured in three ways: by arranging different tones in a series; by associating nearly related hues of a like tone, except where these, as blue and violet, distinctly injure each other; and by viewing appropriate groupings of colors by colored light, as that from a stained window, which modifies them all in a particular direction. The effect of the contiguity of white is to deepen . all hues in whatever tone, unless it may be a light yellow; but with the deeper hues and tones, the contrast with white is generally too violent. Black accords well with almost any hue or tone, except that the deeper, as indigo or violet, render
