A principal test of permanency in pigments is the impunity with which they bear exposure to light and air, an artistic proof of their stability the mere chemist is apt to neglect. Provided the colour remain unaffected by sulphuretted hydrogen, &c., he seldom hesitates to pronounce it safe. But a pigment may be fast in one sense and fugitive in another, believed in by the laboratory, and found wanting by the studio. It has happened before now that the same colour has been dubbed durable and the reverse, by the man of science and the man of art. The former, we take it, looks upon a pigment as a coloured substance of a certain composition, possessing maybe an acid and a base, either, or neither, or both of which, gases and other reagents may injure or destroy. The latter views a colour chiefly as part and parcel of his picture--that picture which _may_ meet with foul exhalations, but _must_ be exposed to light and air. And he too often thinks as little of the effects of an impure atmosphere or injudicious admixture, as the chemist considers the action of air and light. With the exception of madder, those colours mostly affected by _light and air_ are of organic origin, such as gallstone, Indian yellow, and the yellow dye-wood lakes; the red and purple lakes of cochineal; indigo; and sap green. To these may be added the semi-organic Prussian blue; and the inorganic yellows and orange of arsenic. The pigments liable to injury from _sulphuretted hydrogen_, &c., are notably those obtained from lead and copper; and that treacherous compound of iodine and mercury, known as pure scarlet. Many colours are apt to change from the action of white _lead_ and other lead pigments, &c., principally those which are altered by light and air. Many, too, cannot safely come into contact with iron, or ferruginous pigments; especially the yellows of arsenic, the lakes of cochineal, and the blues and greens of copper. With these an iron palette knife is best avoided, one of ivory or horn being used instead. The latter, indeed, is preferable in all cases, several pigments being slightly affected by iron, cadmium yellow among the number. Numerous colours are likewise injured by _lime_ and _fire_, and cannot therefore be employed in fresco, or enamel painting. Of substances which may act deleteriously on colours, there remain the _vehicles and varnishes_ with which they are mixed. Many of these have been blamed, and often with justice, for their injurious effects on pigments. The reputation of the most permanent colour may be ruined, if the vehicle, &c., employed with it be untrustworthy. The presence of lead, for instance, in such materials renders them liable to be blackened by foul air, and by consequence the pigments used therewith.