OF THE PRODUCTION OF COLOURS. 



637 



planation of the dispersion of colours by re- 

 fraction, perhaps more simple and satisfactory 

 than that which I advanced in the last Bakerian 

 lecture. We may suppose that every refrac- 

 tive medium transmits the unduhitions coii- 

 stitutin*^ hght in two separate portions, one 

 passing tlirough its ultimate particles, and 

 ihe other through its pores : and that these 

 portions reunite continually, after each suc- 

 cessive separation, the one having preceded 

 the other by a very minute but constant in- 

 terval, depending on the regular arrange- 

 ment of the particles of a ho:.iogeneous me- 

 dium. Now, if these two portions were al- 

 ways equal, each point of the undulations, re- 

 sulting from their reunion, would always be 

 fourid lialF way between tlie places of the 

 corresponding point in the separate portions; 

 but, supposing the preceding portion to be 

 the smaller, the newly combined undulation 

 will be less advanced than if both had been 

 equal, and the difference of its place will 

 depend, not only on the difference of the- 

 lengths of the two routes, which will be con- 

 stant for all the undulations, but also on the 

 law and magnitude of those undulations ; so 

 that the larger undulations will be somewhat 

 further advanced after each reunion than the 

 smaller ones, and, the same operation re- 

 curring at every particle of the medium, the 

 whole progress of the larger undulations vvill 

 be more rapid than thatol'the smaller; hence 

 the deviation, in consequence of the retard- 

 ation of the motion of light in a denser me- 

 dium,, will of course be greater for the 

 smaller than for the larger undulations. As- 

 suming the law of the harmonic curve for 

 the motions of the particles, we might with- 

 out much difficulty reduce this conjecture to 

 a comparison with experiment ; but it would 



he necessary, in order lo warrant our con- 

 clusions, to be provided with accurate mea- 

 sures of the refractive and dispersive jjowers 

 of various substatices, for rays of all de- 

 scriptions. 



Dr. Wollaston's very interesting observa^ 

 tions would furnish very great assistance in 

 this inquiry, when compared with the sepa- 

 ration of colours by thin plates. I have re- 

 peated his experiments on the spectrum will* 

 perfect success, and have made some at- 

 tem[)ts to procure comparative measures 

 from thin plates; and I have found that, as 

 Sir Isaac Newton lias already observed, the 

 blue and violet light is more dispersed by re- 

 fraction, than in proportion to the differ- 

 ence of the appropriate dimensions deduced 

 froui the phenomena of thin plates. Hence 

 it happens, that when a Une of the light, pro- 

 ceeding to form an image of the rings of co- 

 lours of thin plates, is intercepted by a prism, 

 and an actual picture is formed, resembling 

 the scale delineated by Newton from theory, 

 for estimating the colours of particles of 

 given dimensions, the oblique spectrums, 

 formed by the different colours of each series, 

 are not straight, but curved, the lateral re- 

 fraction of the prism separating the violet end 

 more widely than the red. The thicknesse?, 

 corresponding to the extreme red, the line of 

 yellow, bright green, bright blue, aiid ex- 

 treme violet, I found to be inversely as the 

 numbers 27, .30, 35, 40, and 45, lespectively. 

 Ir» consequence of Dr. Wollaston's correc- 

 tion of the description of the prismatic spec- 

 trum, compared with these observations, it 

 beconses necessary to modify the supposition 

 that 1 advanced in the last Bakerian lecture, 

 respecting the proportions of the sympathe- 

 tic fibres of the retina ; substituting red, 



