204 COLOR PHOTOGRAPHY. 



XYI. — Summary and Conclusion. 



I had set before myself the task of determining the causes of the 

 color reproduction in the older processes of color photography which, 

 in their main features, were introduced by Seebeck, Becquerel, and 

 Poitevin. 



The explanation of Schults-Sellach, by disintegration colors, was, in 

 the first place, shown to be erroneous. 



A method was required for the discrimination of interference from 

 body colors which appear in substances of high refractive indices. This 

 was found in the employment of a right-angled glass prism, also of 

 high refractive index, through which the colors to be investigated 

 were observed. 



By means of the alteration in color thus produced it was shown that 

 the Becquerel picture upon an underlying silver mirror was chiefly pro- 

 duced by interference. Here, therefore, Zenker had correctly ascribed 

 the cause of the color reproduction to the formation of stationary light 

 waves. 



In the pictures of Seebeck and Poitevin there was, on the contrary, 

 no color change. They consist, therefore, of body colors, and Zenker's 

 explanation finds here no application. 



The fact that these pictures show the same colors by transmitted as 

 by reflected light leads to the same conclusion. 



It was shown that in Becquerel's pictures body colors cooperate in a 

 slight degree. 



The understanding of the formation of body colors was promoted 

 for the Seebeck and Poitevin processes by the proof for these processes, 

 respectively by Carey Lea and by Krone, that the substances present in 

 the plates are capable of yielding compounds which embrace almost 

 all the spectral colors, if not all their tones. 



The explanation was, however, lacking why the color substances 

 produced agreed in hue with the illumination-producing decomposition. 



This explanation is found, that of all colored substances capable of 

 being produced only those will be stable which agree in color most 

 nearly with the incident light, since these will best reflect and least 

 absorb it, and can therefore be least changed. Decomposition products 

 of other colors, on the other hand, absorb this light and will be again 

 decomposed. 



A test of this explanation was made by throwing a spectrum at right 

 angles on a color photograph of the spectrum. It was found, in fact, 

 that a correctly-reproducible illuminating color was capable of decom- 

 posing all colors differing from it, but similar colors remained 

 unchanged. 



It is therefore fundamentally possible that colored illumination shall, in 

 suitable substances, produce similar body colors. 



I have designated such substances as color receptive. 



