cunning steps, putting 27 for 50, 30 for 70, and reserving some force still for the last 90 to 100. So Rembrandt modifies his system on the other side, putting his 40 for 100, his 30 for 90, his 20 for 80; then going subtly downwards, 10 for 50, 5 for 30; nearly everything between 30 and zero being lost in gloom, yet so as still to reserve his zero for zero. The systems expressed in tabular form will stand thus:-- NATURE. REMBRANDT. TURNER. VERONESE. 0 0 0 0 10 1 10 10 20 3 20 20 30 5 24 30 40 7 26 32 50 10 27 34 60 13 28 36 70 17 30 37 80 20 32 38 90 30 36 39 100 40 40 40 Sec. 12. Now it is evident that in Rembrandt's system, while the _contrasts_ are not more right than with Veronese, the _colors_ are all wrong, from beginning to end. With Turner and Veronese, Nature's 10 is their 10, and Nature's 20 their 20; enabling them to give pure truth up to a certain point. But with Rembrandt _not one color_ is absolutely true, from one side of the scale to the other; only the contrasts are true at the top of the scale. Of course, this supposes Rembrandt's system applied to a subject which shall try it to the utmost, such as landscape. Rembrandt generally chose subjects in which the real colors were very nearly imitable,--as single heads with dark backgrounds, in which Nature's highest light was little above his own; her 40 being then truly representable by his 40, his picture became nearly an absolute truth. But his system is only right when applied to such subjects: clearly, when we have the full scale of natural light to deal with, Turner's and Veronese's convey the greatest sum of truth. But not the most complete deception, for people are so much more easily and instinctively impressed by force of light than truth of color, that they instantly miss the relative power of the sky, and the upper tones; and all the true local coloring looks strange to them, separated from its adjuncts of high light; whereas, give them the true contrast of light, and they will not observe the false local color. Thus all Gaspar Poussin's and Salvator's pictures, and all effects obtained by leaving high lights in the midst of exaggerated darkness, catch the eye, and are received for true, while the pure truth of Veronese and Turner is rejected as unnatural; only not so much in Veronese's case as in Turner's, because Veronese confines himself to more imitable things, as draperies, figures, and architecture, in which his exquisite truth at the bottom of the scale tells on the eye at once; but Turner works a good deal also (see the table) at the _top_ of the natural scale,