216 Prof. Nihal Karan Sethi on Talbot's Bands and 



To perform the calculation it is necessary first of all to find 

 the intensity of the various wave-lengths at the different 

 points in the spectrum. This was done with the aid of the 

 fuller expression 



I-J-^n^Ll + oosO.-aV)], . . (2) 



which does not omit X in the coefficient, by first drawing 

 graphs for a number of wave-lengths, so chosen for con- 

 venience that the retardation for them increased successively 



by —, then changing the origin of each according to the law 



of dispersion assumed and reading off the ordinate at the 

 point in question. Table I. shows the wave-lengths used in 

 calculating the colour. As will be seen, the values chosen 

 are sufficiently representative of all parts of the spectrum. 





T. 



A.BLE 



I. 





7042 



6037 





5282 



4695 



6816 



5870 





5140 



4593 



6603 



5710 





5030 



4494 



6401 



5559 





4914 



4402 



6215 



5418 





4801 





Now it is well known from the work of Maxwell * that all 

 the colours in the spectrum can be formed by the com- 

 bination in proper proportions of three primary colours — 

 red, green, and blue. An extended table was compiled by 

 Lord Rayleigh f from the more trustworthy set of the 

 observations given in Maxwell's paper, in which the propor- 

 tions of these components are given for a number of wave- 

 lengths in the spectrum as observed by an eye with normal 

 vision. Using that table and graphical interpolations, the 

 red, green, and blue components for the wave-lengths in 

 question were determined. Multiplying the intensity at a 

 point due to a particular wave-length (as described above) 

 by these numbers, the contributions of that wave-length to 

 the red, green, and blue components of the colour at that 



* Phil. Trans. 1860, and Scientific Papers, vol. i. p. 410. 

 f Scientific Papers, vol. ii. p. 503. 



