NATURE 



?>^z 



THURSDAY, AUGUST 6, 1891. 



A PHYSICIST ON COLOUR-VISION. 

 Colour-Measure7netit and Mixture. By Captain Abney. 

 • (London : The Society for the Promotion of Christian 

 Knowledge, 1891). 



THIS interesting little book extends over only 200 

 pages, but is full of careful and important observa- 

 tions. It is, in fact, a summary of the results arrived at 

 by the author during his careful and laborious investiga- 

 tion of the properties of the spectrum. It forms one of the 

 " Romance of Science " series published by the S.P.C.K., 

 a series intended " to show that science has for the masses 

 as great an interest as, and more edification than, the 

 romances of the day." Now, though the earlier portion of 

 this book could be understood by anyone, we venture to 

 think that the second half is for the most part so tech- 

 nical that the full meaning and value could only be 

 appreciated by those who are more or less conversant 

 with the methods of experimenting on colour. To those 

 who are familiar to even a slight extent with the techni- 

 calities of colour experiments, the characteristic of the 

 book is its extreme lucidity. We are carried on from 

 point to point, until, when we look back on the closed 

 book, we find we have travelled over the greater number 

 of the problems of colour-vision almost without effort. It 

 is a book which will not appeal to the masses, but should 

 be read by every physiologist and physicist interested in 

 colour-vision. 



There is yet another reason for the interest which 

 attaches to this work, necessitating a fuller notice than if 

 it were simply a popular disquisition on colour. It is the 

 record of a careful series of experiments by an eminent 

 physicist, firmly convinced of the truth of the Young- 

 Helmholtz theory of colour- vision. The voluminous work 

 of Hering and his pupils is not once mentioned through- 

 out the whole book, although Konig's later publications 

 receive due notice. In fact, if space permitted, we can- 

 not imagine a book more calculated to form the basis of 

 a fruitful discussion on the merits of the rival theories 

 than that now before us. For both the problems of colour- 

 vision, and their solution according to the Young-Helm- 

 holtz theory, are definitely and clearly stated. 



The book opens with a description of the methods used 

 to obtain a spectrum, and a consideration of its properties 

 with especial reference to the ultra-red and ultra-violet 

 rays. The apparatus used by the author to investigate 

 the three fundamental properties of colour — hue, lumino- 

 sity, and purity — are described in detail. Absorption 

 and interference are then touched upon in their relation 

 to colour, and experiments are given to show that the 

 colour of a body is due to its refusal to transmit or reflect 

 certain rays of the spectrum. This is followed by an 

 interesting chapter on scattered light, with especial 

 reference to atmospheric effects, and a pretty lecture- 

 room experiment is described to show that the change 

 in the colour of the sun when on the horizon is produced 

 by small particles in the air. 



The author then passes on to consider the second 



property of colour— luminosity ; and the luminosity of the 



spectral colours is measured as follows. The light from a 



NO. I 136, VOL. 44] 



certain portion of the spectrum passes through a slit 

 which cuts off the remainder of the spectrum. A portion 

 of the same white beam which was decomposed by the 

 prisms is reflected on to the same screen as the mono- 

 chromatic beam, and an upright rod is interposed. This 

 rod throws two shadows, whose intensity is compared 

 after the manner of a photometer. The luminosity of 

 the whole reflected beam is greater than that of the 

 coloured beam, and a rotating diaphragm, with variable 

 sectors, is therefore interposed in its course. By altering 

 the size of the sectors, the intensity of the white light is 

 diminished, until the luminosity of the shadow it casts is 

 equal to that cast by the monochromatic beam. The 

 luminosity is then read off in terms of the segment of 

 the circular diaphragm which remains open when the 

 luminosity of the two shadows is equalized. The lumino- 

 sity of all the principal points in the spectrum is measured 

 on this plan. Subsequently the luminosity of a combina- 

 tion of red and green is shown to be equal to the sum of 

 the luminosities of the same red and green determined 

 separately. Three colours, A, B, and C, are chosen, 

 which, when combined, make white of a certain intensity, 

 W ; and the author shows that if the luminosity of the 

 combined light A -f B be subtracted from the luminosity 

 of the white light, W, the remainder exactly equals the 

 luminosity of the third factor, C. 



A curve of luminosity can be constructed in this 

 manner for the whole spectrum, and its maximum is 

 found to lie on the yellow side of the D line. A similar 

 luminosity curve is given for an observer who was what 

 is ordinarily called red-blind. On this curve the red end 

 of the spectrum is shortened, and the maximum lumino- 

 sity falls nearer the green than on the curve constructed 

 for a person with normal colour-vision. These facts are 

 explained as follows. To the red-blind observer red is 

 invisible, and therefore the luminosity of red is abolished ; 

 the luminosity of yellow, which is composed of red and 

 green, is also diminished, and thus the maximum of the 

 curve moves towards the green. 



This question of luminosity is intimately associated 

 with the theory of the value of white in the system of 

 colour. The author discusses later on in the book the 

 abolition of colour by white light, and examines the 

 extent to which white light can be added to a colour 

 without being perceived. He finds that both depend on 

 the luminosity of the colour, and formulates the law that 

 "the extinction of every colour is effected by white light 

 that is 75 times brighter than the colour." Again, he 

 finds that a large proportion of white light can be mixed 

 with yellow without being perceived, whilst a very small 

 proportion of white added to blue is at once apparent. 



An attempt is made to explain these facts on the 

 Young-Helmholtz theory ; but the work done by Hille- 

 brand,^ under Hering's guidance, makes the explanation 

 offered very improbable. Hillebrand used an apparatus 

 in which one half of the field could be illuminated by a 

 monochromatic spectral colour, whilst the other half was 

 illuminated by white light. The observer shielded one 

 eye from the light for a considerable time, so that it was 

 ultimately brought into a condition of complete rest. 

 Now if he looked at a field filled with monochromatic 



' "Ueberdie specifische Helligkeit der Farben," .JjVzi. d. k. Akad. d. 

 Wissenschaft. in IVien, February 1889. 



