1896-97-] SPATIAL THRESHOLDS OF COLOURS. 229 



is present but much weakened. Moreover, a pigment may owe its 

 colour to a combination of other coloured rays without having any of the 

 kind after which it is named. For example, it is possible to find a 

 violet pigment which sends forth no direct violet rays at all, but 

 a combination of blue and red. Such being the case it was certainly 

 a desideratum, which, by the way, our experimental handling of the 

 problem has sought to satisfy, namely, to secure specimen colours which 

 should be tolerably free from this adulteration by other colour elements, 

 and fairly representative of the spectrally pure colour tones. In the 

 absence of any attempt to obtain spectrally pure colours to operate 

 upon, as was the case with Von Wittich, it is evident that the results 

 must embody at least, a double unavoidable defect which renders them 

 scientifically useless. On the one hand, since the colours experimented 

 upon are actually, in each case, a congeries of many colour elements it 

 cannot be determined, even approximately, what specific part each of 

 the various colour constituents contributed to the determination of the 

 Spatial Thresholds of the colour pigments employed. On the other 

 hand, since no two pigments, even representing the same colour tone, 

 will emit exactly, or even closely, the same combination of colour 

 constituents because of the fact that a variety of colour combinations 

 can produce the same aggregate effect, and since further, there are a 

 great many pigments all differing in appreciable degrees which have 

 equal claims to be taken as representative of any colour tone, it follows 

 that experiments such as Von Wittich's, based upon the examination of 

 colour pigments taken up at random, can have no general validity. 

 They cannot be considered at all as representative results for the main 

 colour tones which compose our colour system. They are, at best, only 

 indicative of the thresholds of those particular pigment papers which 

 Von Wittich used and which can never be reproduced with any surety 

 for any one else's experiments, simply because there was no attempt 

 made to secure, and hence, no record left of a diagnosis of their colour 

 constitution. 



A similar objection can be urged against his method of securing 

 black (and also white) grounds, upon which the colours were placed for 

 research. Black ground is a very indefinite quantity which may mean a 

 great variety of black surfaces, all differing in the degree of their 

 blackness. It is well-known that the surfaces which we ordinarily 

 designate black in reality reflect a considerable quantity of light and are 

 hence very far from being absolute black. Dr. Kirschmann has made 

 some investigations along this line with the purpose of finding the 

 relative brightness of different so-called blacks to white. His methods 



