AND ITS APPLICATION TO THE DETERMINATION OF COLOUR SENSATIONS. 337 



(6.) Amount of Blue Sensation in Yelloiv and Yellow-yreen. 



In reviewing my previous measures of the amount of blue sensation in the yellow 

 and yellow-green of the spectrum, I was struck with the variation of results 

 obtained on different occasions, and though every care was taken at the time, I am 

 led to think that the amount of this sensation was under estimated, though at the 

 most the quantity is but small. This part of the spectrum has occupied my 

 attention for a considerable time, and the determination of the blue sensation in this 

 region has been conducted on perfectly different lines to that formerly employed, 

 which was to make an equation by mixing the colour under consideration with red 

 and violet in sufficient quantities to form white, and then to equate it with the 

 standard equation. The equations were formed, but uo great stress is laid on the 

 correctness of the blue sensation found, but only on the correct proportion of red and 

 green sensation. The corrected value of the blue sensation was found by the following- 

 plan : 



Slits were placed in the red and green at the standard positions red Li and 

 SSN 37 '5 (standard scale number) in the green and as good a match as possible was 

 made by mixtures of the two colours with the intermediate spectral colours, to which 

 a little white was added. The amount of white added was not considered, but only 

 the white inherent in the green. This last was deducted from the green and the 

 percentage of red and green sensations in each colour calciilated without taking into 

 account the white which was due to the presence of the blue sensation. From the 

 equations were obtained the percentage of red and green when the white present in 

 each colour -was included, and by the last measiires the percentage of red and green 

 when such white was excluded. From these different percentages it was easy to 

 calculate the amount of blue sensation present, for it only exists in the '' inherent 

 white." On subsequently considering the sensation curves of equal stimulation as 

 given by KCENIG and myself, my attention was called to the fact that at the place 

 where the red and blue curves cut a large and very sudden increase of white inherent 

 in the colour should be seen, so large indeed that it would never escape notice. The 

 colour at that point ought to be much paler than colours close to it, but such is not 

 the case. The new measures show that there is no sudden rise in the amount ot 

 white present in any colour, and that the maximum of white is at SSN (standard 

 scale number) 43 (X 5427) and not at 37'5 (X 5150). This point will be referred to 

 later. 



(7.) Measures from the Blue-green to the Violet. 



The measures taken from the blue-green to the violet were made by the same 

 method as described in the paper above mentioned, but the process was much 

 simplified. The colour whose percentage composition had to be measured was isolated 

 by a slit placed in one spectrum, and a slit in the other spectrum moved till a 



VOL. ccv. A. 2 x 



