37^ 



Transactions of the Canadian Institute. 



[Vol. VII. 



Even this process does not give us a direct photograph in colours, 

 taken in an ordinary camera, Hke an ordinary monotone such as the 

 world is looking for, and still less does the three-colour process, which is 

 an indirect and composite method, fulfil such an ideal. It is, however, 

 the only really practical method at present available, and is the one I 

 shall attempt to describe in this paper. It was, in its inception, based 

 on the Young-Helmholtz or three-colour theory of vision, and, although 

 the principles of the three-colour process are independent of any visual 

 theory, yet a short statement of the essential points of this theory ma)' 

 be of service in simplifying the succeeding explanations. 



The facts of colour vision are accounted for in the Young-Helmholtz 

 theory by assuming that there are three fundamental colour sensations, 

 a red, a green, and a blue-violet; and that all colours, except deep 

 spectrum red and the extreme violet, according to Abney's latest 

 researches, are compound sensations, produced by the excitation 

 simultaneously of two, or sometimes even of the three colour sensations. 

 Although this theory has been very generally discredited by 

 physiologists and psychologists, it still possesses many strong advocates 

 on the physical side, and will always retain considerable interest on 

 account of the historical associations connected with it. It has the 

 merit of giving a simple and direct explanation of the main facts of 

 colour vision, while those not explainable on this hypothesis meet with 

 little better fate at the hands of the other theories advanced. 



Maxwell supported this theory and, by means of a modified form of 

 spectroscope which he called a colour box, made measurements to 

 determine the ranges of these colour sensations. These measurements 

 placed in the form of curves, can be projected upon the screen (Fig. i). 

 They are of great interest to all students of the three-colour process, 

 not only for their historical value, but principally by reason of the fact 



Ao<BC 



Fig. I. — Maxwell's Colour-Sensation or Colour-Mixture Curves. 



