OPTICAL IMAGES. 



by a simple and homogeneous red light placed before a convex 

 lens A c, fig. 31, and that an image of it is produced at r r'. Let 



Fig. 31. 



the same object be now supposed to be illuminated by simple and 

 homogeneous orange light. This light being more refrangible 

 than red light, the lens A c will produce an image o o' of the 

 object, a little nearer to it than rr'. If the object be next 

 illuminated with simple and homogeneous yellow light which is 

 still more refrangible, the image y y' will be produced at a still 

 less distance from the lens, and in fine if the object be successively 

 illuminated with simple green, blue, indigo and violet lights, the 

 images will be produced successively at g g', b &', i i', and v v r y 

 nearer and nearer to the lens as the light is more refrangible. 



57. If the object, instead of being illuminated as we have here 

 supposed it to be by a simple homogeneous light, be illuminated 

 by any light compounded of two or more simple lights, then so 

 many distinct images of it will be produced at different distances 

 from the lens, as there are simple lights in the compound, and 

 these images will differ in colour from the object and from each 

 other. Thus, for example, if the object be illuminated by a 

 compound light of a green tint, composed of simple yellow and 

 blue lights, two images of it will be produced, the nearer blue, 

 and the more distant yellow. 



A like consequence will follow if the object be illuminated by 

 a compound light made up of three simple lights, when three 

 images will be formed, and so on. 



If then an object reflect from its surface the white solar 

 light, which is a compound of all the colours, it will follow that 

 all the coloured images which have been here produced in succes- 

 sion, will be produced at one and the same time, and will be 

 placed one before the other in a regular series at unequal distances 

 from the lens, as already described. 



58. It has been shown* that the colours of natural objects 

 generally are more or less compounded. It is only in very rare 



110 



* See Tract on " Colour." 



