30 P. C, on the Colours that enter into the [July 



It is not surprising, therefore, that light admitted to the 

 prism through a small aperture should, after refraction, 

 have some of its rays of different colours inseparably 

 blended. The difference of refrangibility is in this case 

 compensated by a different degree of refraction, and the 

 rays being thus brought to a state of parallelism, no dis- 

 tance, however great, can effect their separation. 



But the most convincing proof is, that if we suffer the 

 direct and unobstructed rays of the sun to fall on a prism, 

 a spectrum is formed in which the colours, with the excep- 

 tion of the violet, do not greatly exceed the breadth of the 

 face of the prism ; each colour being, in fact, when fully 

 developed, a distinct image of it ; or of the breadth of light 

 which falls on its surface. There is neither in this case the 

 extraordinary elongation, which in some of Newton's ex- 

 periments, were in the proportion of seventy to one, nor 

 the variety, nor the regular gradation of colours, which 

 were produced in these experiments. 



If we observe with attention the developement of the 

 spectrum, thus produced, in different stages of its progress, 

 we shall be enabled to form some idea of the number and 

 character of the primitive colours which it exhibits ; and 

 we shall then be prepared to confirm or correct these 

 theoretical views, by submitting them to the test of ex- 

 periment. 



If we receive the spectrum on a screen placed at a few 

 inches from the prism, we shall observe a breadth of white 

 light in the centre of the spectrum, fringed with violet and 

 blue on one side, and yellow and red on the other ; as the 

 screen is withdrawn to a greater distance from the prism, 

 the fringes increase in breadth at the expense of the white 

 light, which at length wholly disappears ; and the spectrum 

 then appears to be formed of four colours, red, yellow, 

 blue, and violet ; by withdrawing the screen still further, 

 green makes its appearance between the yellow and the 

 blue; and, finally, the two latter colours entirely disap- 

 pear, apparently absorbed in the green ; the colours being 

 now reduced to three, red, green, and violet. 



In this experiment the white light is evidently formed by 

 the intersection of all the different colours, which, although 

 they have received different directions by refraction, are still 

 superposed in the centre, in consequence of the succession 



