286 



OPTICAL PROJECTION 



a coloured one. This also illustrates the important fact, that 

 prism analysis will always give us the correct composition of 

 whatever light there is passing through the prism. 



If instead of the card, a thin prism ground to a fine edge 

 is used, the intercepted portion of the rays will be simply de- 

 flected, instead of suppressed, and a second image of the slit, 

 complementary in colour to the other, will appear by its side 

 upon the screen. 



A more obvious and mechanically simple method is to 

 provide seven small pieces of looking-glass, each about 2 inches 

 long by J inch wide, mounting them with wax on small 



wooden feet. These 

 are to be arranged on 

 a piece of blackened 

 board A, supported by a 

 stand, and so arranged 

 that the row together, 

 at some little distance 

 from the prism p, about 

 covers the whole width 

 of the spectrum-band 

 at that spot. Each 

 mirror is then adjusted 



to throw its strip of colour on the same spot on the screen s. 

 On taking away one or more mirrors, and so suppressing any 

 of the colours, again we get colour, whereas all the colours 

 gave white. 



A better method of recompounding the colours is to use a 

 cylindrical lens, which can easily be so adjusted between the 

 prism and screen as to focus the slit again in an image 

 which appears sharp and white. The cylindrical lens should 

 be rather long in focus, from 10 to 15 inches, to produce a 

 good effect ; the usual short-focus cylindrical lenses make the 

 slit appear much too broad. A large cylindrical jar full of 

 warm water (else moisture will condense upon it), may be 



FIG. 155 



