Composition of White Light. 99 



There is an argument in favour of a limited number of 

 degrees of refrangibility, if arguments were necessary where 

 facts are so abundant, which deserves consideration. The 

 rays of light of different colours, I have already observed, 

 are independent of each other ; and this independence is so 

 complete, that in every part of the retina there is a distinct 

 channel of communication for each colour, which, when 

 the ray it is designed for is absent, remains unemployed.* 



* This view of the subject will satisfactorily account for the production of a 

 black accidental image, by the alternate action of complementary colours, viewed 

 in the manner proposed by M. Plateau, [6.*] whose method of producing this 

 effect I was unacquainted with at the time it was written ; if we view crimson 

 and green, or any other complementary colours alternately, the effect, to be con- 

 sistent with these observations, ought to be the same as when we look upon both 

 colours together, as they are reflected from a white object. 



I cannot see, in this experiment, any thing opposed to the theory of accidental 

 colours most generally admitted, as it is, I think correctly, stated by M. Plateau 

 at the commencement of the paper [1.] 



I admit that if what follows [2 .] be really a fact, " that accidental colours may be 

 seen in perfect darkness," this theory is unsatisfactory ; but I am not acquainted 

 with any clearlv established case of the kind. I confess, however, that from the 

 defective state of my eyes, I have been unable to make some experiments which 

 would enable me to speak more decidedly on this point. 



White light, so weak as to be imperceptible in this state, is frequently rendered 

 visible, and, in some cases even brilliant, by withdrawing either one, or two, of 

 its constituent colours. This may be proved by several experiments that are 

 familiar to us. 



The production of colours, on the pressure of the eye-ball, and in several cases 

 of the disease of the organ, as well as those cases where its appearance is altogether 

 the work of the imagination, ought, in my opinion, to be excluded from this class 

 of phenomena ; we otherwise confound appearances which have a totally different 

 origin. These colours are generally uncertain ; whereas, those which are, I think 

 improperly, termed accidental colours, are always complementary to the primary 

 colours, with reference to the light with which they are seen ; and, of course, when 

 this light is white, the accidental are truly complementary colours ; generally, 

 however, of less intensity than the primary colours ; because the sensibility of the 

 eye to the latter, though lessened by its previous action, is seldom, in these expe- 

 riments, entirely removed, and the accidental colour is, consequently, diluted 

 with white light. , 



In cases where the accidental colour is seen with light of a much lower inten- 

 sity, as, for instance, when the eyes are closed after being impressed with the 

 primary colour, the eye appears to be totally insensible to the latter, in its reduced 

 state, and the accidental colour is thus rendered very brilliant. 



Black accidental images may be accounted for on the same principle : [4] the 

 eye is rendered less sensible to all the colours by the contemplation of white light, 

 and insensible to light of a much lower intensity ; when any part of it, therefore, 

 is impressed with strong white light, and the eye is then exposed to weaker light, 

 that part of it, thus impressed, is insensible to it, and forms a corresponding black 



• Records, vol. ii. p. 281. 



