I will only mention Mulready. It is generally stated that in his advanced age he painted too purple. A careful examination shows that the peculiarity of the colors of his later pictures is produced by an addition of blue. Thus, for instance, the shadows on the flesh are painted in pure ultramarine. Blue drapery he painted most unnaturally blue. Red of course became purple. If you look at these pictures through a yellow glass, all these faults disappear: what formerly appeared unnatural and displeasing is at once corrected; the violet color of the face shows a natural red; the blue shades become gray; the unnatural glaring blue of the drapery is softened. To make the correction perfect, the glass must not be of a bright gold-color, but rather of the color of pale sherry. It must be gradually darkened in accordance with the advancing age of the painter, and will then correspond exactly with the color of his lens. The best proof of the correctness of this statement is, that the yellow glass not only modifies the blue in Mulready's pictures, but gives truthfulness to all the other colors he employed. To make the proof complete, it would be necessary to show that by the aid of yellow glass we saw Mulready's pictures as he saw them with the naked eye; and this can be proved. It happens that Mulready has painted the same subject twice—first in 1836, when he was 50 years of age and his lens was in a normal state, and again in 1857, when he was 71, and the yellow discoloration had considerably advanced. The first picture was called "Brother and Sister; or Pinching the Ear;" the second was called "The Young Brother." In both pictures a girl, whose back only is visible, is carrying a little child. A young peasant, in a blue smock-frock, stands to the right and seizes the ear of the child. The background is formed by a cloudy sky and part of a tree. Both pictures are in the Kensington Museum. The identity of the composition makes the difference in the coloring more striking. If we look at the second picture through a yellow glass, the difference between the two almost entirely disappears, as the glass corrects the faults of the picture. The smock-frock of the boy no longer appears of that intense blue which we may see in a lady's silk dress, but never in the smock-frock of a peasant. It changes into the natural tint which we find in the first picture. The purple face of the boy also becomes of a natural color. The shades on the neck of the girl and the arms of the child, which are painted in a pure blue, look now gray, and so do the blue shadows in the clouds. The gray trunk of the tree becomes brown. Surprising is the effect upon the yellowish-green foliage, which, instead of appearing still more yellow, is restored to its natural color, and shows the same tone of color as the foliage in the earlier picture. This last fact is most important to prove the correctness of my supposition. My endeavor to explain it became the starting-point of a series of investigations to ascertain the optical qualities of the pigments used in painting, and thus to enable us to recognize them by optical contrivances, when the
