For all practical purposes, then, it seems reasonable to define a "fast color" as one which will not be materially affected by those influences to which, in the natural course of things, it will be submitted. Hence, in speaking of a fast color, it becomes necessary to refer specially to the particular influences which it resists before the term acquires a definite meaning. To be precise, one should say that a color is "fast to light," or "fast to washing," or "fast to light and washing," and so on. Further, it is necessary, as we shall see afterward, to give always the name of the fiber to which the color is applied. All that I have said with respect to the term "fast" may be applied with equal propriety to the term "fugitive." This, too, has no very definite meaning until a qualifying statement, such as I have referred to, gives it precision. The most important question to be considered is THE ACTION OF LIGHT ON DYED COLORS. That light can effect radical changes in many substances was known to the ancients. Its destructive action on artists' pigments, e.g., the blackening of vermilion, was recorded 2,000 years ago by Vitruvius. Since that time it has been well established, by numerous observations and experiments, that light possesses, in a high degree, the power of exerting chemical action, i.e., causing the combination or decomposition of a large number of substances. The union of chlorine with hydrogen gas, the blackening of silver salts, the reduction of bichromate of potash and of certain ferric salts in contact with organic substances, are all familiar instances of the action of light. In illustration of this, I show here some calico prints produced by first preparing the calico with a solution of potassium bichromate, then exposing the dried calico under a photographic negative, and, after washing, dyeing with alizarin or some similar coloring matter. During the exposure under the negative, the light has reduced and fixed the chromium salt upon certain parts of the fiber as insoluble chromate of chromium (Cr_{2}O_{3}CrO_{3}) in the more protected portions, the bichromate remains unchanged, and is subsequently removed by washing. During the dyeing process, the coloring matter combines with the chromium fixed on the fiber, and thus develops the colored photograph. The prints in Prussian blue are produced in a similar manner, the sensitive salt with which the calico is prepared being ammonium ferricitrate, and the developer potassium ferricyanide. Investigation has shown that the most chemically active rays are those situated at the blue end of the solar spectrum; and although all the rays absorbed by a sensitive colored body affect its change, it is doubtless the blue rays which are the chief cause of the fading of colors. Experiments are on record, indeed, which prove this.