February i6, 1893] 



NATURE 



2>^7 



such authorities as Goethe and Brewster, both of whom 

 believed that they saw in green the blue and yellow, of 

 which, being misled by experience with pigments, they 

 believed it to be composed. 



He goes on, " Helmholtz says, quite correctly, ' so far as 

 I see, no way has been found of determining one of 

 the elementary colours except the investigation of colour 

 blindness.' This investigation has notoriously not con- 

 firmed Young's theory." 



This would, even if it were true, be in itself no argu- 

 ment against the admissibility of the theory. The theory 

 of colour-blindness seems, as we shall shortly see, to be 

 a particularly hard crux for Hering's theory ; while the 

 hitherto well-established facts of red-blindness and green- 

 blindness admit of comparatively easy and perfect 

 explanation by Young's theory. 



He adds, " And the three sets of fibres, which, how- 

 ever, as Helmholtz remarks, are not essential to the 

 theory, have hitherto been sought for in vain." 



This objection applies to Hering's theory as much as 

 to Young's. 



The reader will easily convince himself that these 

 objections are of no weight whatever. He follows them 

 up by an enumeration of contradictions and inaccuracies 

 which he professes to have found in Grassmann's and my 

 own explanation of Newton's law of colour-mixture, and 

 partly also in that of Kries, errors which, even if they 

 existed, would in no way tell against Young's theory, but 

 only against its interpreters. Here, however, the 

 obscurity seems to me to lie on the side of our 

 opponent. 



These objections arise out of the fact that, in mi.xtures 

 of a saturated colour with white, the tint of the mixture 

 sometimes seems changed (pale red for example 

 approaches more to rose, and pale blue to violet) ; and 

 that, on the other hand, with increase of intensity, the 

 colours of the spectrum appear sometimes paler, some- 

 times yellower. But if we speak of those elementary 

 excitations which, from the point of view of Newton's 

 law, are alone entitled with certainty to the name of 

 elements, as being able to coexist without mutual dis- 

 turbance, then the only sensation which can with cer- 

 tainty be regarded as corresponding to the coexistence of 

 a white and a red elementary sensation is that which 

 comes into existence under the simultaneous influence of 

 the corresponding white and red lights. The term " ele- 

 mentary sensation " is in this connection to be taken, of 

 course, not in the narrow sense of Young's hypothesis, 

 but in the wider sense above explained — the sense in 

 which we speak of linear relations between colour- 

 sensations and linear superposition of elementary- 

 sensations. In the domain of colour-mixture we know 

 nothing of any elements but these superposable ones ; 

 md if we would preserve a constant meaning for our 

 colour-equations we must interpret them in this sense, 

 is I have explained above. This is what H. Grassmann 

 md myself have always done. 



Moreover, erroneous estimates of the difference be- 

 tween a pale and a more saturated colour are liable to be 

 made, and hence those colours which are really most 

 diluted with white do not always appear the palest. If, 

 without sufficient experience of colour-mixture, we only 

 :,^uideourjudgmentsby similarity of sensations,we are liable 

 10 make mistakes as to which colour contains white. The 

 c[uestion of the power of perceiving differences will there- 

 tore arise. Further, it is found that colours of very 

 strong luminosity do not differ so much from one another 

 in the sensations they produce as colours of moderate 

 luminosity, — a fact which finds its explanation in 

 Young's theory, of which it is a natural consequence. 

 Colours when highly luminous appear more similar 

 to one another and more similar to white. We 

 express this by calling them pale as compared with 

 colours of feebler luminosity. I have, however, already 



NO. I 216, VOL. 47] 



mentioned that the law of superposability ceases to be 

 applicable when the luminosity is excessive. 



Nevertheless, in view of the fact that simple colours of 

 high luminosity are always as saturated as colours of 

 such luminosity can be, it is not necessary, or rather it is 

 not correct, to designate them as less saturated. The 

 true statement is that differences of tint become more 

 uncertain at high intensity — an uncertainty^ which at- 

 taches also to the estimation of the intensity itself, as 

 has long been known. 



If Hering's sensation of white and opposed-colour- 

 sensations are truly to deserve the name of elements or 

 constituent parts of sensation (as he plainly intends, 

 since he assigns to them special visual-substances), either 

 he must acknowledge them as the elements deducible 

 from the law of addition, or else they are purely hypo- 

 thetical processes of whose existence and superposability 

 no one knows anything. His polemic against Grassmann 

 and me then amounts to this — that at a time when his 

 hypothesis had not been propounded we did not speak 

 in the sense of it. 



Hering seems to regard as the chief point of superiority 

 of his own hypothesis its closer conformity with the 

 names which have established themselves in language — 

 names which, as I have explained above, relate rather 

 to the colours of material bodies than to the colours of 

 light. To this circumstance it is, in fact, indebted for a 

 certain amount of popularity and facility of apprehension. 

 He himself assumes that these names have sprung from 

 an immediate perception of the simple elements of sensa- 

 tion by a kind of inner consciousness, and thinks that 

 he has thus very certain and immediate knowledge of the 

 pure red-sensation, the pure white-sensation, and so on. 



In his publication of 1887 he has discussed the possi- 

 bility of assuming, instead of three or six simple pro- 

 cesses of sensation, a larger and perhaps indefinitely 

 great number, and a corresponding number of " elemen- 

 tary powers " for the several kinds of objective light. He, 

 however, gives the geometrical representations of such 

 actions in such a manner that practically these powers all 

 depend on three independent variables. On the other 

 hand, as regards these independent variables, which are 

 the most important factors in the problem, he gives as 

 good as no clue to them ; he only seeks to remove them 

 as far as possible from the sphere of physiology. For 

 my own part I am able to understand this whole series 

 of descriptions only as meaning that an arbitrary number 

 of visual substances can be assumed to exist in the brain, 

 and that their respective strengths of excitation are 

 different functions of the same three independent vari- 

 ables, each visual substance being unaffected by the 

 excitations of the rest, and the excitation of each being 

 susceptible of direct apprehension in consciousness. I 

 do not think it is necessary, in this book, to go further 

 into such hypothetical views. 



Hering especially claims the credit of opening up the 

 way to understanding colour-blindness. He makes all 

 dichromasy depend upon a single cause, namely want of 

 sensibility in the red-green visual-substance. The differ- 

 ence between red-blindness and green-blindness is, 

 according to him, attributable to different colourations of 

 the media of the eye ; partly of the yellow spot of the 

 retina, partly of the crystalline lens. 



These colourations are chiefly met with in the sick or 

 the very old, and, when occurring in otherwise useful eyes, 

 are not of such strength that they could bring out con- 

 spicuous deficiency of brightness in different parts of 

 the spectrum. 



The colouration of the yellow spot of the retina takes 

 effect in a very limited but very important part of the 

 field of view, and in only a narrow band of the spectrum. 

 The most trustworthy observations on the influence 

 of the wave-length of the incident light upon the strength 

 of the red and green excitations, have been made with 



