464 



Dr. F. W. Edridge-Green. The Relatioii of [June 3, 



to my notice with Eayleigh's colour mixing apparatus, in addition to other 

 methods. In this apparatus a match is made between A, 589 (sodium yellow) 

 and a mixture of X.670 (lithium red) and X 535 (thallium green). The results 

 obtained, however, seem to be quite independent of the power possessed by 

 the individuals examined to appreciate differences in colour. In fact, by 

 this apparatus we may obtain abnormal values in persons who have 

 apparently no defect in hue perception, while others who are undoubtedly 

 colour-blind may give results which correspond to that given by the average 

 individual. The results seem to be determined not by varying power of 

 appreciation of hue, but by slight variations in the luminosity curve of 

 different parts of the spectrum in different individuals. Another important 

 factor, probably the chief one, is the state of adaptation of the eye at the 

 time of the observation. 



All who have had practical acquaintance with colour-blindness are aware 

 that all dichromics are not equally colour-blind ; one will be readily detected 

 by almost any test, whilst another re({uires the greatest care to detect at all. 

 I find that the cases of dichromic vision vary fi-om almost total colour- 

 blindness to cases bordering on those I have called the trichromic. I give 

 below the monochromatic patches marked out by four dichromics with my 

 spectrometer in similar conditions. The light was 180 meter-candies, the 

 slit the same size in each, and the eyes light-adapted. The size of the patches 

 is measured in micro-milliiuetres {[ifi). 



Xo. 1 cannot see any difference between the cactus flower and its leaf, and 

 does not see any difference between the red and green signal lights except at 

 the moment when there is a change from one to the other. It is obvious 

 that this case is much more colour-blind than the oitlinary dichromic. On 

 the other hand, a case boixlering on the trichromic will see about eight 

 definite patches in the spectrum ; make a normal matcli with liayleigh's 

 apparatus ; pass the Holmgren test with ease ; and play pool regularly 

 without on any occasion mistaking the coloured balls. Such a case, when 

 examined with the spectrometer, has eiglit monochromatic patches instead of 

 the normal (jighteen ; says tliat there are only two colours in tlie spectrum, 

 red and violet, and has a very small neutral patch. It is (juite easy to show 

 that such a case is dichromic, as he will match spectral Ijlue-green with a 

 mixture of H])ectral red and violet : X. G70 -f- X, 470 = \ 490. How could a defect 

 ill a light-])erceiving substance have such varying results ? If wo assume 

 that in these cases there is a varying perception of colour difference, the facts 

 are explained. 



It is stated in favour of the Young-Helmholt/, tlieory tliat with three 

 variables we can ]ir(idu(;(! iiimuiicrabh! variations, tlie fiict that on trial we 



