ON ACCIDENTAL OR SUBJECTIVE COLORS. 221 



terposed between it and tb(? screen, an intensely colored shadow will be exhib- 

 ited. By varying the color of the glass-plate or that of the light of the candle, 

 shadows of different colors and of different intensity will be produced 



J. H.] 



Effect of contrast. — In the preceding experiments, the space whose color ap- 

 pears modified is supposed to be of small extent relatively to the colored surface 

 which surrounds it, and the color of this last does not appear sensibly altered : 

 yet M. Chevreul has proved in a general manner that this modification of colors 

 is reciprocal ; that is to say, that when we see simultaneous!}' two colored ob- 

 jects placed in tin neighborhood of one another, their two colors seem to react 

 simultaneously in such a way that to each of them is added the comphmcntary 

 color of the other. Thus, when I place beside one another a red and a yellow 

 object, the first will seem to tend more or less to violet, and the second to green. 

 This reciprocal modification of the two colors is commonly too feeble to be per- 

 ceived without having recourse to a particular expedient. The following is the 

 ingenious process of M. Chevreul : two strips of paper or cloth of the two colors 

 which it is proposed to submit to observation, one, for example, red, the other 

 yellow, are pasted close to one another on a card; the strips should be J 2 milli- 

 metres in breadth and o centimetres in length. Parallel to one of the strips is 

 now pasted, at the distance of a millimetre, a second strip which is identical 

 with the former in dimensions and color, and designed to serve as a term of com- 

 parison ; the same operation is practiced in relation to the strip tinted Avith the 

 other color, so that there are finally four colored strips, two of one color and two 

 of the other. The two inner ones are in contact, and it remains to observe the 

 modifications which they mutually produce on one another. With this view 

 the card is to be looked at in a particular direction and for some seconds. The 

 reciprocal effect of the two contiguous colors, which would escape us under or- 

 dinary circumstances, becomes in this way almost always sensible, by help of 

 the exterior strips which serve for comparison. Thus, in the example we have 

 chosen of red and yelloio strips, it will be seen that the interior red strip will 

 tend to a violet color, and the contiguous yellow strip to a green. M. Chevreul 

 reports a great number of other examples which follow the same law ; it is ne- 

 cessary to mention here only the following facts : 



1. If the two colors employed are complementary one of the other, as red 

 and green, they enliven one another by their juxtaposition, acquiring a remark- 

 able brightness and purity. 



2. If we place in juxtaposition witb white any color whatever, the former is 

 slightly tinted with the complementary color, and the color employed becomes 

 brighter and deeper. Thus, by the contact of white and red, the first becomes 

 greenish, the second deeper and more brilliant. 



3. If any color be placed in contact with black, the latter takes, in a greater 

 or less degree, the complementary tint of the color employed ; and this last ap- 

 pears, in general, brighter and clearer. 



4. Black and white undergo, likewise, by their juxtaposition, a reciprocal 

 modification ; the first becomes brighter, and the second deeper. 



5. The mutual modifications of colors are not limited to cases in which the 

 colored objects that modify one another are in contact. M. Chevreul proved, 

 by experiment, that they may be rendered sensible, even when the objects are 

 five centimetres apart, though the intensity of the effect is less as the distance 

 is greater. 



6. To these observations, derived from the memoir of ]\[. Chevreul, ^.[ Pla- 

 teau adds another, which is verified every moment, and which may be deduced, 

 in some sort, from 2, 3, and 4 preceding. When two neighboring objects differ 

 in brightness, this difference appears, in general, augmented by their neighbor- 

 hood; one appears brighter, the other duller, than if they were seen separately 

 or surrounded with objects of a brightness equal to their own. Further, objects 



