0'2 Mr. Cooper on the Colours, that enter into the 



If we look to the spectrum itself, in the different stages 

 of its developement, every appearance indicates one degree 

 of refrangibility for each colour, and an uniformity in it 

 throughout. Where, and in which of these stages, do we 

 find that blending of colours which must necessarily arise 

 from the superposition of innumerable circles of colours 

 gradually differing from each other. So far from there being 

 the slightest indication of this, every appearance is opposed 

 to it ; the two extreme colours, violet and red, preserve the 

 same colour from their first appearance to their full deve- 

 lopement ; the blue and yellow are equally unchangeable 

 throughout their different stages to their final disappear- 

 ance ; and the first line of green light, exhibited by the 

 separation of the violet and red images, is of the same colour 

 as when the image is fully developed. 



Now, all these appearances, would be very different if 

 the spectrum were formed of innumerable circles differing 

 gradually in refrangibility and colour; the extension of the 

 spectrum would withdraw these circles in succession, from 

 the extreme colours ; but beyond these, on both sides, the 

 combinations would be of the most complicated character, 

 and the developement of any distinct colours, except at the 

 very extremities of the spectrum, would be impossible. 



If we direct our attention to the coloured fringes, which 

 formed one of the subjects of my last paper, we shall find 

 the most uniform appearance in the different colours ; and 

 as we have it in our power to produce a great breadth of 

 these colours, by looking at distant objects through a prism, 

 we cannot easily be mistaken in their appearance. 



It may be said, admitting there are only three colours, 

 these colours, though uniform in other respects, may have 

 different degrees of refrangibility. It must be confessed, 

 that this hypothesis would be much more consistent with 

 the appearance of + he colours in the spectrum, and it would 

 equally well account for its termination by two rectilinear 

 and parallel sides ; but the experiments of Newton, before 

 quoted, are totally inconsistent with it. 



Newton has demonstrated mathematically, as well as 

 experimentally, that circular bodies, whether luminous or 

 illuminated, preserve their round appearance when viewed 

 through a prism, provided the light be homogenious ; and 



