390 



SECTIONAL TRANSACTIONS.— 1. 



Monday, September 8 



Discussion (Sections I, J) on the question, In what Sense can we speak of 

 Primary Colours ? (Opened by Dr. J. Drever. Prof. H. E. Roaf, 

 Dr. J. H. Shaxby, Dr. F. W. Edridge-Green, C.B.E.) 



Dr. J. Drever. — Though the employment of the term ' primary ' with respect 

 to colours m^ay be regarded at the present time as characteristic mainly of the 

 physicist's approach to the investigation of the phenomena of colour, there are really 

 four distinct points of view from which it is apparently possible to speak of ' primary ' 

 colours — those of the artist, the physicist, the physiologist, and the psychologist re- 

 spectively. The artist's primaries are red, yellow, and blue, the physicist's red, green, 

 and blue, the physiologist's red, green, yellow, and blue, and the psychologist's also 

 red, green, yellow, and blue, though a different red and green from those of the 

 physiologist. 



Artist and physicist apply what is essentially the same logical principle of classifi- 

 cation. The artist finds that all the known bright colours can be derived from a 

 mixture of red, yellow, and blue pigments, while the physicist claims that all the colours 

 of the spectrum and of nature can be produced by an admixture of red, green, 

 and blue lights. The psychologist and the physiologist in turn challenge the right of 

 the physicist to dogmatize concerning colour at all, which, they point out, is a physio- 

 logical, and ultimately a psychological, not a physical, phenomenon. Moreover, the 

 psychologist claims that yellow is as much a primary colour as red, green, or blue, 

 and the physiologist's investigation in perimetry would seem to confirm this claim. 

 The phenomena of perimetry, indeed, as well as the phenomena which appear with 

 increase of luminosity of the spectral series up to the limit at which the individual 

 colours disappear, suggest that red and green are relatively secondary to blue and 

 yellow, the latter being perhaps the true physiological primaries. ' 



Attempts have been made to determine more precisely — that is by wave-length — the 

 primary colours, and the general results may be summarised as follows : 



The agreements no less than the discrepancies are significant. The red of both 

 physicist and physiologist is outside the spectrum and appears bluish. The green of 

 the physiologist is its complementary and also appears bluish. Physicist and psycholo- 

 gist agree as regards green, and physiologist and psychologist as regards yellow, while 

 all three are more or less in agreement as regards blue. The discrepancies, however, 

 are such that it would appear desirable to discontinue the practice of speaking of 

 ' primary colours,' at any rate without specification of the sense in which the term 

 ' primary ' is employed. 



Prof. H. E. Roaf. — Colour being a sensation, a primary colour is one that appears 

 to be a single entity. 



The sensation of colour can be produced by stimulation by radiations of certain 

 wave-lengths ; the physiological problem is to determine what receptors are necessary 

 in order that the whole range of the visible spectrum and the extra spectral colours 

 can be recognised. 



There is experimental evidence that three separate kinds of receptors are necessary, 

 whilst histologically two different kinds, namely rods and cones, are recognised. In 

 many birds, however, there are three sets of cones which must be stimulated by dif- 

 ferent parts of the visible spectrum. Experimental evidence is in favour of colour 

 vision in man being of the same nature as the histological evidence suggests that it 

 would be in birds. 



