ABSORPTION OF LIGHT, ETC. 39 



a focus, but from two focal lines, as they are called. Now if 

 you combine the prism with a lens in order to project the 

 image on a screen, and allow homogeneous light to pass 

 through a hole at a sufficient distance, and to fall on the prism 

 when not in the position of minimum deviation, you will find 

 in one position of the screen a vertical line of light, and in 

 another position a horizontal line, and between the two you 

 will get a circular patch. Now suppose that the screen is 

 held in such a position that on it is formed the vertical line, 

 how is the image of a slit which you substitute for the hole 

 formed 1 It is formed by a succession of lines overlapping one 

 another in the direction of their length, which gives you in 

 fact a single straight line ; so that when white light is used 

 light of any one kind will be brought to a line on the screen, 

 or in this case on the retina of the eye, and the spectrum 

 will be seen distinctly. The particular azimuth in which the 

 prism must be held to see the spectrum distinctly depends on 

 the distance at which the slit is held from the observer, and 

 on the length of sight of the observer, and it will be different 

 from one end of the spectrum to the other. If I hold a prism 

 so as to see the red end distinctly, I must turn it a little to 

 see the violet end distinctly. Turning it in one way brings 

 the virtual image nearer to the eye, and turning it the other 

 way moves the image further off. If it is focused for the 

 red by holding the prism in a certain position, we must turn 

 it a little so as to diminish the angle of incidence, in order to 

 get the true focus for the violet. The reason for that is, that 

 there is no provision for chromatic compensation in the eye. 

 The eye in that respect is to be compared, not to an achro- 

 matic object-glass, but to a simple lens. The effect of the 

 dispersion of light as regards ordinary vision is not perceived 

 under ordinary circumstances, but it becomes very perceptible 

 indeed when you supply the eye with homogeneous light 

 of different kinds, or with light from which certain portions 

 are abstracted. I have tried the experiment of throwing a 

 pure spectrum on a printed page. On holding the page at 

 the usual distance of distinct vision, I was able to see quite 

 distinctly in the green and brighter part of the spectrum ; in 

 the red end I saw somewhat indistinctly from long-sightedness ; 

 and in the violet end very indistinctly from short-sightedness. 

 In choosing a prism it is very easy to see whether the glass 

 of which it is composed is good or not. If you look at the 



