1891.] 



Colour of Cases of Tobacco Scotoma, 



495 



his area of luminosity should be 1100. As it is 1400, the ordinates 

 of the above carve should be multiplied by 0'8, to compare with that 

 of the normal eye. 



It should be mentioned that his matches of luminosity were ma.de 

 without any hesitation, and were concordant for each observation, 

 which is not to be wondered at, as the matches, except at the blue 

 end, were practically matching shades of black to white. 



From the foregoing, it will be seen that the white which C. sees as 

 white is the same as the D sodium light, and that the red he says is 

 yellowish. The mixture of this yellowish-white with the blue 

 apparent makes white at X 5430. He sees a little blue in the 

 spectrum colour at A, 5720, so it must be taken that at that point of 

 the spectrum he begins to see colour, a point which is considerably 

 lower than that given by his preliminary examination of the spectrum 

 colour, and due, no doubt, to the fact that the white light used by 

 the comparison light was that of the positive pole of the electric 

 light. It seems probable that what C. called yellowish was really a 

 sensation of white mixed with a very small quantity of red sensation 

 (as he saw no yellow in the orange, in which that colour would be 

 most easily distinguished on account of its luminosity), and red light, 

 when strongly diluted with white light, to the normal eye appears 

 slightly orange. 



Subsequently C. was tested for the illuminating value of white 

 light compared with my own and that of Mr. Nettleship. The appa- 

 ratus used in this case (fig. 2) was, I believe, somewhat on the prin- 

 ciple of Dr. Forster's photometer, with which I was unacquainted 

 before I made the instrument. It is made as follows : 



FIG. 2. 



