222 ' ON ACCIDENTAL OR SUBJECTIVE COLORS. 



moderately illumiuated may disappear completely, as is shown by the results of 

 M. Brewster's observations, when their image is portrayed on the retina, in the 

 neighborhood of a part of the organ vividly excited by the presence of a bril- 

 liant object. Of this we may easily be convinced by regarding objects placed 

 nearly behind the flame of a candle. These objects appear of a duller hue in pro- 

 portion as their image more closely approaches that of the flame, and when the 

 dis lance of the two images is very small they disappear altogether. 



In order to complete our statement of the principal facts relating to accidental 

 colors, we add to the experiments already cited the following, which we owe to 

 Dr. Smith: Place a hghted candle very near one of the eyes, (the right eye, 

 for instance,) but out of the line of the optical axis; then hold in front of the 

 eyes a small strip of white paper, and direct your sight toward a remote point, 

 so as to see the small slip double; the image seen by the right eye will appear 

 green, and that seen by the left eye will appear of a reddish tint. 



General law of simultaneous subjective colors. — Reflection on the assemblage 

 of experiments which we have been reporting will show that they lead to the 

 following conclusions : 



When we look directly or indirectly at a colored space, there is manifested, 

 beyond the outline of that space and to quite a considerable distance, the appear- 

 ance, more or less decided, of a color complementary to its own, a color which 

 continues to decrease in intensity in proportion as the distance augments. White 

 and black, brightness and obscurity, are assimilated, in this case, to two colors 

 complementary one to the other. If two colored spaces or objects are near one 

 another, the effect is then reciprocal , regard being had to the extent and bright- 

 ness of each of the two objects. 



This development of the complementary color is especially distinct when the 

 space on which we observe it is small and surrounded by a much more consider- 

 able extent of the color which is to produce this complementary color, as ia the 

 experiments of Rumford, Meusnier, and Prieur. The intensity of the effect is 

 also much augmented if we contrive that the small space shall be feebly illumi- 

 nated relatively to the colored ground on which it is projected. This serves to 

 account for the colored appearances displayed in a great number of circum- 

 stances. Thus, as M. Brewster remarks, when an apartment whose walls are 

 of a bright color is illuminated by the sun, those parts of the furniture on which 

 the light does not directly fall seem to be tinged with a color complementary to 

 that of the walls. So, again, according to the observation of M. Chevreul, 

 when designs are imprinted on colored stuff's or papers, the color of these designs 

 is usually modified by the complementary of that of the ground- Fault has been 

 sometimes found with the manufacturer from this circumstance, because the de- 

 signs do not appear of the color which had been prescribed. In this case, it can 

 easily be determined whether the difference is owing to an eff'ect of subjective 

 color. It suffices to cut and apply some white paper in such manner that it 

 shall cover the ground and allow only the design to be seen. 



It is known that the shadows produced on a white wall at the rising or setting 

 of the sun appear blue or green. It is because the light of that body is then 

 of an orange or red tint, and the shadows are tinged with the complementary 

 color. When an apartment in which candles are lit is still feebly illuminated 

 by the light of departing day, the shadows produced by bodies interposed be- 

 tween the candles and white objects appear blue for the same reason. It is in'this 

 way, also, that M. Necker de Saussure accounts for the remarkable phenomena 

 presented by the summit of Mont Blanc at the setting of the sun. M. Arago 

 thought that to the same cause might be attributed the blue or green color 

 under which the sun appeared through the extraordinary mist of 1831, assuming 

 that the mists or clouds near the sun may have been colored red by reflection. 



There is no one who has not observed that in a place partially illuminated by 

 the sun, the parts which are in shadow appear much darker than if the sun 



