33fi SIR W. DE W. ABXEY: MODIFIED APPARATUS FOR MEASUREMENT OF COLOUR 



person matched without any difficulty the blue lithium line* with the violet near H, 

 though he described the former as rather paler than the latter, a description which the 

 colour- vision theory indicates as probable. Using my two eyes, one fatigued and the 

 other not, I endeavoured to obtain a measure of the amount of red sensation 

 destroyed, but owing to the mixture of white in the blue the match was never 

 perfect, as the fatigue passed away before the match was made, and when white was 

 added to the blue, it too had lost part of its red, and the " fatigued " violet appeared 



too green. 



These experiments and others went to prove the absence of the 4th sensation, and 

 if further proof were required, it would be found in the ease with which the violet, 

 when white is added to it, can be matched with a mixture of red and blue near the 

 blue lithium line. 1 have therefore felt justified in using the violet as a temporary 

 sensation in all my measures, reducing it to its components of red and blue in my 

 final results. 



(4.) Fi, ''ft! f'niiifs in flic Spectrum. 



Several points in the spectrum could be readily f'ouzid. Thus the complementary 

 colour to the red in the blue-green is a fixed point, as is the complementary 

 colour to the violet. The complementary colour to the blue (near the blue lithium 

 line) is also known. For other preliminary details a reference should be made to my 

 previous paper. 



(5.) DeteriuiiMtioti of the \Vhitc in f/ic Colour which only excites the Green 



Sensation u-ith \\'hite. 



In my previous investigations 1 was unable to match spectral orange to which 

 white could be added with mixtures of red and the green, but had to use the light 

 transmitted through a solution of bichromate of potash placed in the path of the 

 white (reflected) beam as representing an orange. By a suitable arrangement white 

 could be added to it, till the mixtures were of the same colour. A small quantity of 

 white had then to be added to the spectral orange to match the colour and the 

 bichromate solution. From the two amounts of white added, the amount of white 

 necessary to add to the spectral orange in order for it to match the mixture of red 

 and green was deduced. With the apparatus now employed the determination of the 

 amount of white to be added to the orange was made direct. There was also an 

 advantage in these direct determinations with the spectrum colour, as more than one 

 shade of orange could be used as checks to one another. 



The results of the many measures made show that a slight correction has now to 

 be made to my previous determination. 



* It will be noted further ou that the blue lithium excites only the blue sensation and that of white. 



