PERCEIVE DIFFEHENOES IN COLOK. 225 



as in the preceding set. Sinre, however, a certain 

 quantity of light was transmitted through the blue, the 

 result may indicate a want of sensitiveness to the blue 

 rays. 



Ill the red the numbers were 1928 as against 3072. 



As regards the yellow, the results were very different, 

 the numbers being, under the yellow, 3096 ; in the 

 uncovered part, 1904. Here, therefore, we see a very 

 distinct prefereDce, all the more remarkable because 

 the amount of light was really less than in the un- 

 covered part. 



In the green the numbers were mui-h more equal, 

 namely, 2406 against 2594. Here also the love for green 

 neutralized the preference for light. I do not, however, 

 wish for the moment to draw any conclu.sion from these 

 last figures, though I give them for what they are worth. 

 The coloured medium was, I believe, somewhat too 

 opaque. With a more transparent green, as will be 

 seen subsequently, the result would have been very 

 different. 



At any rate, the above observations seemed to show a 

 marked preference for yellow. Still, I thought it might 

 be objected that, though the Daphuias obviously pre- 

 ferred the uncovered to the shaded half of the vessel, 

 and the yellow to the uncovered half of the vessel, 

 perhaps in the former the uncovered water was rather 

 too bright, and in the latter the shaded part was rather 

 too dark, and that after all the yellow was chosen, not 

 because it was yellow, but because it hit off the happy 

 medium of intensity. The suggestion is very improb- 

 able, because the observations were made on several 

 successive, and very different, days, and at very 

 different hours. I also thought that the green was 



