314 



NA TURE 



[August 6, 1891 



light of moderate intensity with the rested eye, it appeared 

 to him colourless ; and by suitable adjustment he could 

 make an absolute match between the half of the field 

 illuminated by monochromatic light and the other half 

 illuminated by white light from the same source. Thus, 

 as the whole spectrum appeared colourless, he was able 

 to construct a curve of luminosity for the spectrum by 

 matching it with the white light in the other half of the 

 field. The maximum of this curve lay in the green. A 

 glance with the unshaded eye at once brought the colour 

 into view, although the field was unaltered. But as soon 

 as the colour came into view, he noticed that the lum.ino- 

 sity of the coloured half no longer matched that of the 

 colourless half of the field. If yellow or red were the 

 colour chosen, the luminosity of the coloured half of the 

 field appeared to exceed that of the colourless hair, whilst 

 if green or blue were selected the exact opposite was 

 observed. Moreover, as soon as the colours of the spec- 

 trum were appreciated, the maximum luminosity shifted 

 into the yellow, and the curve he then obtained closely 

 resembled that constructed by Captain Abney and other 

 observers. Thus we must conclude that every part of 

 the spectrum is capable of exciting the sensation of white 

 apart from its specific colour, and that the maximum 

 sensation is produced by a certain point in the green. As 

 soon, however, as the colour becomes apparent, this 

 sensation of white is either increased or decreased by the 

 specific luminosity of the colour. The luminosity of the 

 spectrum, as determined by Captain Abney, is the alge- 

 braic sum of two factors. Firstly, the power which every 

 part of the spectrum possesses of exciting the sensation 

 of white ; and secondly, the specific luminosity of the 

 colour sensation itself, which is a positive quantity on the 

 red and yellow side and a negative quantity in the blue 

 and green. 



If this explanation for the difference in the two curves 

 be correct, a person who was completely deficient in 

 colour-sense would construct a luminosity curve for the 

 spectrum differing considerably in the position of its 

 maximum from that given by Captain Abney in his book. 

 The curve obtained by Konig^ from a man to whom 

 yellow, blue, green, and red were invisible, to whom the 

 whole spectrum appeared in varying shades of white, 

 shows that this is the case. The maximum luminosity 

 lies in the green, over the line b. A comparison of this 

 curve with that given by Hillebrand for the normal eye at 

 rest reveals their almost absolute identity. The existence 

 of this form of colourblindness can only be explained 

 with extreme difficulty on the Young-Helmholtz theory ; 

 whereas Hering's hypothesis, that white and black form a 

 colour pair analogous to red and green, yellow and blue, 

 not only renders the existence of such a condition prob- 

 able, but also easily explains Hiilebrand's results. 



The author passes on to show that white can be pro- 

 duced from the mixture of three spectral colours, and 

 ultimately defines a primary colour as one which cannot 

 be formed by the mixture of any other colours. The 

 three primary colours he selects are red, green, and 



' "Die Grundempfindungen u. ihre Intensiiats-Vertheilung im Spec- 

 trum," Sitzb. d. k. preus. Akad. d. Wissenschaft. zu Berlin, xxxix , t886. 

 Hering has since shown, by investigating a similar case of total colour- 

 blindness, how closely the curve ot luminosity agrees with that given 

 by Hillebrand. The account of this interesting case has not yet been 

 published. 



NO. I I 36, VOL. 44] 



violet ; for yellow is formed by a mixture of red and 

 green, blue by a mixture of green and violet. But he 

 warns us from assuming that the three primary colour 

 sensations " are of necessity the same sensations as are 

 given by the three primary colours " (p. 138). On p. 150, 

 red (between C and the lithium line), violet (close to G), 

 are selected as furnishing two primary sensations, whilst 

 "all three fundamental sensations" are excited by the 

 green, except at a point where the green is mixed with 

 white only. 



Now, to say that spectral green excites the sensations 

 of red and violet seems to us radically false. For when 

 speaking of sensations we leave the realm of physics, and 

 the sole test of the sensations excited by a portion of the 

 spectrum is the colour which we perceive when light from 

 that part impinges on the retina. No one who examines 

 spectral green will say that it gives him the sensation of 

 red or violet, but rather that the greater part of spec- 

 tral green appears to be mixed with either yellow or blue. 

 Again, a primary sensation must be one which gives us 

 the sensation of one colour only. Now every eye sees in 

 violet both blue and red. Thus, whether violet be a pri- 

 mary colour from the physical point of view, physiologic- 

 ally speaking it is anything but a primary sensation. 



Though violet fails to answer the test of a primary 

 colour sensation, a point can be found both in the yellow 

 and the blue of the spectrum, from which the sensation 

 of one colour only is obtained. But throughout the 

 book we find repeated mention of the formation of yellow 

 by the mixture of spectral red and spectral green. How 

 can this be reconciled with the acceptance of yellow as a 

 primary sensation ? 



To most eyes, the red of the spectrum yields to a greater 

 or less extent the secondary sensation of yellow. Take 

 such a red, and gradually add minute quantities of spec- 

 tral blue. The yellow will gradually disappear, and a 

 red will be produced, which yields the sensation of red 

 only, untinged with either yellow or blue. Take a spec- 

 tral green, which is also slightly yellow, and treat it in the 

 same way. If we now mix the absolutely pure red with 

 the absolutely pure green, white is produced, not yellow. 

 And now we can understand why spectral red and spectral 

 green can be made to form yellow. For both the red and 

 the green, which, when mixed, form yellow, when separate 

 give the secondary sensation of yellow in addition to that 

 of their principal colour. Thus, when mixed, the pure 

 red annihilates the pure green, and yellow only remains. 

 Measured by this standard, the primary colour sensations 

 fall into two groups, in which each colour is complement- 

 ary to the other. Firstly, red and green, from which all 

 secondary sensations of yellow and blue are absent ; and 

 secondly, yellow and blue, which do not give the second- 

 ary sensations of either red or green. 



Colour-bhndness is brought in to support the Young- 

 Helmholtz theory, but the author has obviously not had 

 the opportunity of investigating many cases of this affec- 

 tion. He speaks of green-blindness, in which the sensa- 

 tions of red and violet are present, but not that of green ; 

 and of red-blindness, in which the sensations of green and 

 violet are present, but not that of red ; and gives measure- 

 ments to show that in the latter class of cases the spectrum 

 is shortened. 



