1 835 .] Composition of White Light. 29 



in every respect within the nearest limits, to the least re- 

 frangible ray of the colour adjoining ; and, consequently, 

 that there was no distinct line of separation between dif- 

 ferent colours. 



If this opinion be well founded it is of little consequence 

 into what number of colours the spectrum is divided ; it 

 being, in fact, if this principle be admitted, formed of innu- 

 merable circles of colours, each differing from the other, 

 though imperceptibly, in colour as well as in refrangibility. 



But there does not appear to me to be any good ground 

 for such a conclusion : it is directly in opposition to the 

 well known experiment of Newton, in which, by making a 

 hole in the screen upon which the spectrum was received, 

 he suffered any one of the colours to pass and fall upon a 

 second prism, which now formed a correct image of the 

 hole by which it was admitted, and not an elongated 

 image, which it would have done if light of the same colour 

 differed in refrangibility ; and, at a future time, I shall 

 produce other experiments which, in my opinion, will not 

 leave a doubt on the subject. 



Newton's erroneous views arose, I apprehend, from his 

 making his experiments with light admitted through small 

 apertures, by which it was inflected into an innumerable 

 variety of different directions ; and as a subsequent refrac- 

 tion, upon well known optical principles, bends the rays so 

 that a difference in their previous direction is retained in 

 their further progress, every colour after being refracted 

 by the prism, instead of being formed of parallel rays, must 

 have had an almost infinite number of different directions. 



Diffracted light may be formed either by admitting it 

 through a small aperture, or through a lens of very short 

 focus ; and the light thus admitted appears to have similar 

 properties. Now, we know that in the latter case, the rays 

 are so refracted, that at a certain distance from the lens they 

 meet in one point, and, consequently, that every circle of 

 rays proceeding from the lens, at different distances from 

 the centre, must be refracted so as to form different angles 

 with the central ray. The light, then, thus prepared, must 

 meet the face of the prism at different angles of incidence ; 

 some of the rays at a larger and some at a smaller angle than 

 the central ray; and the previous difference of direction, 

 must thus be considerably increased by the second refraction. 



