ii6 THE ART OF PROJECTING. 



a weak solution and narrow slit a series of black bands 

 are produced in the green part of the spectrum ; but 

 with a stronger solution the green and yellow are com- 

 pletely cut out, allowing only the red and deep blue 

 lights to pass. On widening the slit these bands of 

 colored light of course increase in width also, gradually 

 approaching each other until they overlap, producing a 

 fine purple by their admixture. 



If the experiment be repeated, substituting for the 

 permanganate an alkaline mixture of litmus and potas- 

 sium chromate in certain proportions, only the red and 

 green light are transmitted, the blue, and especially 

 the yellow^ being completely absorbed. 



On widening the slit as before, the red and green 

 bands overlap and produce by their union a very fine 

 compound yellow, while the constituent red and green 

 are still visible on each side. The effect is most strik- 

 ing when by the widening of the slit a round hole is 

 exposed in its place, when then appear on the, screen 

 two circles, respectively green and red, producing 

 bright yellow by their mixture. This experiment is the 

 more striking as it immediately follows the process 

 of absorbing the simple yellow. The mixture above 

 described (suggested by Mr. Strull) answers better than 

 a solution of chromic chloride. 



Of course, it is a well-known fact that all natural 

 yellows give a spectrum of red, yellow, and green, and 

 a common effect illustrating the compound nature of 

 yellow is noticed when exhibiting a continuous spec- 

 trum on a screen. When the slit is narrow the green 

 is very fully developed and only separated from the 

 red by a very narrow strip of yellow, while on gradually 

 increasing the width of the slit the red and green are 

 sure to overlap, producing the brilliant yellow we 



