April 6, 1893] 



NATURE 



541 



It will be recollected that in making the extinctions, the 

 1) light of the spectrum was made equal to one amyl-ace- 

 tate lamp, and the other rays had the relative luminosity 

 to it, which they had in the spectrum before they were 

 extinguished. The luminosity curve of the spectrum is 

 shown in Fig. 5. 



Suppose we make all the luminosities of the different 

 rays equal to one A L., we should not get the same ex- 

 tinction value, as shown in the continuous lines in Fig. 3. 

 The violet would have to be much more reduced, but 

 by multiplying the extinction by the luminosity we should 

 get the curve of reduction for equal luminosities, and we 

 get the dotted curves in Fig. 3. 



It will be seen that it is the violet under such circumstan- 

 ces that would be the last to be extinguished, and that all 

 the rays at the violet end of the spectrum would be extin- 

 guished simultaneously, as would also those at the extreme 

 red. This looks like a confirmation of the Young-Helm- 



holtz theory which I have briefly explained, for we can- 

 not imagine that it can be anything but a single sensation 

 which fails to be excited. 



The violet is extinguished when it is - ^ A L, 



10,000,000 

 that is, a screen placed 817 feet away and illuminated by 

 an A L violet lamp would be invisible. The blue-green 



i) light when it is - ^J. or 770 feet away. 



10 milhonths 



The green (e) light 



35 



10 millionths 

 orange (d) light is extinguished as before. at 



-^^ or 550 feet away. The 

 350 



10 millionths 



or 1 80 feet away, whilst the red (c) light has only to be re- 

 duced to ^^,°° , or an A L lamp radiating C light 



10 milhonths ^ & & 



would have to be placed only 67 feet away, whilst the 

 radiation for an A L of the colour of the B light of the 



spectrum would have to be diminished to but 



10 millionths 

 or the screen would have to be placed 60 feet away. 



It is therefore apparent that with equal luminosities 

 the violet requires about 175 times more reduction to ex- 

 tinguish it than does the red, and probably about 25 times 

 more than the green. 



NO. I 223. VOL. 47] 



This being so, I think it will be pretty apparent that, at 

 all events from the extreme violet to the Fraunhofer line 

 D of the spectrum, the extinction is really the extinction 

 of the violet sensation, a varying amount of which is ex- 

 cited by the different colours. If then we take the recip- 

 rocals of the numbers which give extinction of the 

 spectrum, we ought to get the curve of the violet sensation 

 on the Young- Helmholtz theory. For if one violet sen- 

 sation has to be reduced to a certain degree before it is 

 unperceived, and another has to be reduced to half that 

 amount, it is evident that the violet sensation must be 

 double in one case to what it is in the other ; that is, the 

 degrees of stimulation are expressed by the reciprocal of 

 the reduction. 



Such a curve is shown in Fig. 5 (in which also are 

 drawn the curves of luminosity of the spectrum when 

 viewed with the centre of the retina and outside the yellow 

 spot). And it will be noticed that it is a mountain which 

 reaches its maximum about E. 

 Remember that the height of the curve 

 signifies the amount of stimulation 

 given to the violet sensatory apparatus 

 by the particular ray indicated in the 

 scale beneath. 



Turning once more to Fig. 3, it will 

 be noticed that if any one or two of 

 the three sensations are absent, the 

 persons so affected are, what is called, 

 colour-blind. Thus if the red sensation 

 is absent, they are red-blind ; if the 

 green, then green-blind ; if the violet, 

 then violet-blind ; if both red and 

 green. sensations are absent, then the 

 person would see every colour, includ- 

 ing white, as violet. The results of the 

 measurement of the luminosity of the 

 spectrum by persons who have this 

 last kind of monochromatic vision 

 should be that they give a curve exactly 

 or at all events very approximately, of 

 the same form as the curve given by 

 the reciprocals of the extinction curve 

 obtained by the normal eye, as the 

 violet sensation is that which is last 

 stimulated. 



It has been my good fortune to 



examine two such persons, and I find 



that this reasoning is correct, the two 



coinciding when the curves for the centre of the retina are 



employed. 



F"urther, I examined a case of violet blindness, and 

 measured the luminosity of the spectrum as apparent to 

 him. Now if the Young- Helmholtz theory be correct, 

 then in his case the violet sensation ought to be absent, 

 and the difference between his luminosity and that of the 

 normal eye ought to give the same curve as that of the 

 violet sensation. This was found to be the case. 



Again, the reciprocal of the extinction curves of the 

 red-blind and green-blind ought to be the same as those 

 of the normal eye, for the violet sensation must be present 

 with them also. This was found to be so. We have 

 still one more proof that the last sensation to disappear is 

 the violet. 



If we reduce the intensity of the spectrum till the green 

 and red disappear to a normal eye, and measure the 

 luminosity of the spectrum in this condition, we shall find 

 that it also coincides with the persistency curve. On the 

 screen we have a brilliant spectrum, but by closing the 

 slit admitting the light and placing the rotating sectors 

 in the spectrum and nearly closing the apertures, we can 

 reduce it in intensity to any degree we like. The whole 

 spectrum is now of one colour and indistinguishable in 

 hue from a faint white patch thrown above it. If the 

 luminosity of this colourless spectrum be measured we 



