Notes 247 



and violet; and, further, as violets and purples, with all their 

 varieties, may be imitated by mixing blue and red in different 

 proportions, and as green also may be compounded in a sim- 

 ilar way by mixing blue and yellow, and orange by mixing 

 red and yellow, we need not attend to more than the three 

 primary colours, red, yellow, and blue ; for, in fact, it is found 

 that by mixing these three colours in certain proportions, a sort 

 of white, or any colour may be formed ; and there is reason 

 to believe that if we had colours equal in brilliancy to the 

 prismatic colours, the white so formed would be perfect. 



11. This last observation shews us that white may be 

 considered as made up even of two colours only, and we shall 

 find it very convenient, in the explanation of the phenomena 

 in question, to consider white as so made up, namely, of red 

 and green, of yellow and purple, or of blue and orange. These 

 colours are called contrasts to each other respectively : their 

 apparent brilliancy, when they are placed contiguous to each 

 other, is promoted in a remarkable manner, but they cannot 

 be mixed together without mutual destruction to their natural 

 properties, and an approach to a white or a grey colour. 



1 2. To understand the experiment above represented on the 

 paper, we are first to consider the nature of the shadow d v 

 green, as it is in appearance ; that is, we are to consider what 

 kind of light or lights can possibly come to this portion of the 

 paper which we call the shadow d v ; and here it is plain that 

 this space d v is illumined only by the white light ' (I will 

 call it) which comes from the small taper a, directly, and also 

 by a small quantity of white light from b, not directly, but by 

 reflection from the sides of the room or from other objects. 

 The direct red light coming from b, through the red glass c, 

 is intercepted by d ; and the small quantity of this red light 

 which can arrive at the space d v by reflection is not worth 

 mentioning ; the green shadow d v, therefore, is illumined by 



' I call it white light because it is nearly so, and because it answers all the purposes 

 of perfectly white light in such an experiment, supposed to be made in a room without 

 daylight. When actually compared with daylight, it is found to be yellowish, or even 

 orange-coloured. 



