[From tiie American Journal of Science, Vol. XX, August, 1880. | 
ON THE 
EFFECTS PRODUCED BY MIXING WHITE 
WITH COLORED LIGHT. 
By Professor 0. N. Rood, 
Columbia College. 
It was noticed several years ago that when white light was 
mixed by the method of rotating discs with light of an ultra- 
marine (artificial) hue, the result was not what one would 
naturally have expected, viz: instead of obtaining a lighter or 
paler tint of violet-blue the color inclined decidedly toward 
violet, passing, when much white was added, into a pale violet 
hue. Two attempts have been made to account for this curi- 
ous fact: Briicke supposes that the light which we call white 
is really to a considerable extent red, and that the mixture of 
this reddish white light with the blue causes it to change to 
violet. Aubert, on the other hand, following a suggestion of 
Helmholtz, reaches the conclusion that violet is really only a 
lighter shade of ultramarine-blue. He starts with the assump- 
tion that we obtain our idea of blue mixed with white from 
the sky, which, according to him, is of a greenish-blue color. We 
then apply, as he thinks, this idea to the case of a blue which 
is not greenish, namely, to ultramarine-blue, and are surprised 
to find that the result is different. 
It will be shown in the present paper that these explanations 
are hardly correct, since they fail to account for the changes, 
which, according to my experiments, are produced in other 
colors by an admixture of white. I prepared a set of brill- 
iantly colored circular discs which represented all the principal 
colors of the spectrum and also purple; these discs were then 
successively combined in various proportions with a white disc 
and the effects of rapid rotation noted, a smaller duplicate 
colored disc uncombined with white being used for comparison. 0. N. Rood—Effect of mixing White and Colored Light. 
82 
Under these circumstances it was found that the addition of 
white produced the changes indicated in the following table: 
Cyan-blue became less greenish, more 
bluish. 
Cobalt-blue became more of a violet 
blue. 
Ultramarine (artificial) became more 
violet. 
Yermilion beoame somewhat purplish. 
Orange became more red. 
Yellow became more orange. 
Greenish yellow was unchanged. 
Yellowish green became more green. 
Green became more blue-green. 
Purple became less red, more violet. 
Exactly these same effects can be produced by mixing violet 
with the above mentioned colors. Let R, Gr, V represent the 
three angles of Maxwell’s color-triangle, W being the position 
of white. Now, according to the received theory, as we mix 
white with different colors we advance in straight lines from the 
angles or sides of the triangle toward W; in point of fact, 
however, I find, as a result of the above-mentioned experi- 
ments, that we advance in curves 
toward W, these curves being similar 
to those roughly indicated in the fig- 
ure. The only advance in straight 
lines is along the line joining violet 
with its complement greenish yellow. 
The other lines are disposed symmet- 
rically about this line as an axis. 
These experiments serve to explain 
the singular circumstance that when 
complementary colors are produced by the aid of polarized 
light, it is difficult or impossible to obtain a red which is 
entirely free from a purplish hue, a quantity of white light 
being always necessarily mingled with the colored light. In 
the case of the red, orange, yellow, ultramarine, and purple 
discs, I succeeded in measuring the amount of violet light 
which different proportions of the white disc virtually added to 
the mixture, and found that it is not directly proportional to 
the amount of white light added, but increased in a slower 
ratio, which at present has not been accurately determined. 
For the explanation of the above mentioned phenomena, 
Brticke’s suggestion that white light contains a certain amount 
of unneutralized red light is evidently inapplicable, since the 
effects are such as would be produced by adding a quantity 
not of red but of violet light, and for the present I am not dis- 
posed to assume that white light contains an excess of violet 
light. The explanation offered by Aubert does not undertake 
to account for the changes produced in colors other than ultra- 
marine, and even in this case seems to me arbitrary ; neither 
have I succeeded in framing any explanation in accordance 
with the theory of Young and Helmholtz which seems plausible.