removed from the hypo, was washed and dried as usual; but, when I examined it by reflected light, it turned out that the green tree was colored dark green just as nicely as if it bad been a camera image, and the red house was not a bit behindhand in truth and delicacy of hue. A photograph in colors, sure enough! But an examination with a lens, and a little turning and twisting of the plate, caused the illusion to vanish: the colors were those of "thin plates," soap-bubble colors, caused by the interference of light. Wet collodion often shows them in patches when it is somewhat rotten, and this sample was very rotten. The interference effect had nothing to do with the color of the light, but was controlled by its intensity. Once I bad a chance to examine some photographs in color of gayly dressed dolls made by Niepce de Saint Victor, and it seemed to me that the pale colors they presented were produced by a species of interference, acting by means of the presence of more or less finely divided particles. The details of my examination I do not recollect, but merely the conclusion that the appearances presented were due to causes analogous to those that were effective in the case of my glass negative. Photographs in color, such as they are, can be obtained with sufficient patience; but, in order to give this fact the slightest value, it is necessary to prove that a corresponding amount of patience would not be rewarded by the production of colored photographs of objects which were gray, light gray, dark gray, etc. When we think we have made a discovery, our first duty is to destroy it mercilessly if possible, and the reproduction of the same effects with white or gray objects is the proper mode of administering justice in this case. It is barely possible that some one may ask why a process that renders the colors correctly is a failure merely because corresponding colors can be obtained when the natural objects are tinted gray. The question answers itself; white and gray objects will be colored in the photograph, and, worse than that, the same color in the natural object will vary in the photograph with its brightness or luminosity.
Let us now examine this subject from a theoretical point of view, and ask ourselves why we should hope that photographs in color could ever be produced. We see the rich red rays of the spectrum falling on the plate, and we imagine that a substance which is sensitive to light will somehow be acted on by them, and arrange itself so that ever afterward it will better be able to reflect red light than any other kind of light. Why? Why should a substance that has been acted on by long waves be better able to reflect long waves than those that are shorter? Why should a sea-beach that has been acted on by long waves be on that account better able to reflect and redirect to the ocean long waves rather than mere ripples? The waves of light produce in sensitive substances chemical changes; new compounds are formed; why should the long waves of red light produce compounds that are especially capable of reflecting long waves, or red light?
