Spectrum and the Primary Colour Sensations. 



69 



region, the wave-lengths X/480ftand A '460/x were successively 

 used as fatigriing colours, for which the results in Table VIII. 



CT> o " 



-and Table IX. were obtained. 



Table VEIL 



X. 



Normal 



Eye fatigued 



i A - 



Normal 



E 5 



e fatigued 



Persistence. 



with X -480 //. 



Persistence. 



with X -480 ft 



•735 n 



•0230 sec. 



•0234 sec. 



•522 [x 







•0150 sec. 



•690 



•0171 





•495 



0171 sec. 





•0168 



•647 



•0147 



•0147 



i -453 



0246 





0244 



-593 



•0130 



0133 



j -423 



•0312 





0309 



•553 





•0131 



1 -410 



•0368 





•0361 



•535 



•0138 



•0141 



•400 



•0432 





•0423 



Table IX. 



X. 



Normal 



Eye fatigued 



X. 



Normal 



Eye fatigued 



Persistence. 



with X'460/i. 



Persistence. 



with X-468f< 



735 ft 



•0230 sec 





•495 ft 



•0171 sec. 



•0176 sec. 



690 



•0171 





•453 



•0246 



•0249 



647 



•0147 





•423 



•0312 



•0311 



593 



•0130 



•0124 sec. 



•417 





•0341 



•553 





•0132 



•410 



0368 



•0378 



•535 



•dibs 



•0137 



•405 





0415 



•522 





0150 



•400 



•0432 



0426 



These curves are exhibited in fig. 14 and fig. 15 respectively. 

 They show no elevations at all, but are coincident with the 

 normal curves. The conclusion appears to be that this transi- 

 tion region is broader than those previously described, probably 

 because the dispersion o£ the blue in the prismatic spectrum 

 is greater than that of the red. The wave-lengths X *480 fi 

 -and X "460 fi may be regarded, therefore, as marking the 

 edges of the transition band, the centre being the wave-length 

 X *470 fi. For this colour no persistency curve was considered 

 necessary. The curve with the wave-length X 'ISO //,, fig. 14, 

 appears to retain a slight vestige of the green elevation which 

 I have indicated by altering the curve at " ft." 



The fourth and last transition point obviously must lie 

 'between the wave-lengths \°44G/6 and V400/U, since the curve 

 for the former, fig. 16, has two elevations in the green and 

 the violet regions, and that for the latter, fig. 17, only one 

 in the violet. The colour X'±20 jj, was selected for the first 

 trial. This proved to be the desired point, and no other 



