APPENDIX 



53 



more radiant flux is required to achieve a match 

 with the smaller test patch, although in every 

 other set of curves less radiant energy is required 

 for the smaller test patch. 



Fig. 43 shows still another state of affairs. 

 Here matches were made with the same size of test 

 field, 4° 49', but the matches were made with the 

 observer light-adapted (crosses) or dark-adapted 

 (dots). Although the external conditions of the 

 stimulus situation were exactly the same, entirely 

 different psychophysical relationships were ob- 

 tained with different states of adaptation of 

 the eye. 



The point of this discussion is that the ordinary 

 photometric procedures which work well at high 

 luminance levels do not work at very low lumi- 

 nance levels. The relationships between radiant 

 flux and luminous flux vary with the size of the 

 stimulus field, the state of adaptation of the eye 

 and the area of the eye being stimulated. The 

 problem became especially serious during the 

 last war in connection with the photometry of 

 luminescent and phosphorescent materials. Ten- 

 tative procedures were worked out at the time in 

 order that some degree of consistency could be 

 achieved between different laboratories and 

 between laboratories and manufacturers. The 

 procedures were admittedly tentative and the 

 problem of low luminance photometry still needs 

 to be studied thoroughly. Examples of inconsis- 

 tencies in visual data arising from this source 

 have been pointed out in the chapter above. 



Colorimetry 



The measurement of the chromatic aspects of 

 light is termed colorimetry. Here again this 

 evaluation is made in terms of the sensation 

 aroused in an average eye. 



The Eye is an Integrator. The eye cannot 

 resolve different wavelengths when they are 

 combined in a ray of light. This is another way 

 of saying that it is an integrating mechanism 

 rather than an analyzing one. Fig. 44 shows five 

 different combinations of wavelengths which look 

 exactly the same to an average eye. The upper- 

 most section, for example, shows a spectrum 

 which contains equal amounts of all the visible 

 wavelengths, i.e., an equal-energy spectrum. 

 The second chart shows two single wavelengths 

 which, when combined in the ratio of 1 to 0.78, 

 produce exactly the same sensation as the equal - 

 energy spectrum. The same is true for all the 

 other combinations shown in Fig. 44. Even 

 though the combination of wavelengths differs 

 markedly in all cases, the color is exactly the 

 same — very nearly white. 



Color Mixtures. The integrating behavior of 



the eye provides the basis for the psychophysical 

 specification of color. It was discovered a long 

 time ago that every color can be matched by a 

 mixture of three colored lights, red, green, and 

 blue. This is very convenient because it means 

 that any color can be specified in terms of how 

 much of a standard red, green, and blue must be 

 mixed together to match the unknown color. 

 This information can then be charted so that any 

 color can be compared with any other. 



Chromaticity Diagram. The chart used for this 

 purpose, Fig. 45, is called a chromaticity diagram. 

 This one was standardized by the International 

 Commission on Illumination in 1931. The red, 

 green, and blue lights which were standardized 

 are mathematically defined and are hypothetical 

 colors. But the reader will not be far off if he 

 regards them as being real colors. 



Only two dimensions are needed on the chro- 

 maticity diagram because the amounts of green, 

 blue, and red required to match a color are ex- 

 pressed in percentage terms, i.e., they add up to 

 100 percent. This means that if the amount of 

 red and green are known, the amount of blue can 

 be found by subtraction from one. It would take 

 us far beyond the scope of this discussion to 

 consider precisely how the position of a color can 

 be computed on this chart. A summary of these 

 techniques is contained in another report of the 

 OSA Committee (19) . 



The location of the spectrum colors in the 

 chromaticity diagram is shown by the roughly 

 triangular curve in this figure. The numbers 

 along the curve are wavelengths. Since the 

 spectrum colors are the purest one can find, all 

 other colors fall somewhere inside this boundary. 



White. In a general kind of way, white results 

 from a mixture of all wavelengths. It can be 

 matched by very nearly equal proportions of red, 

 green and blue and should be located somewhere 

 near the center of this chart. Actually white is a 

 hard color to define. Most people say that a 

 Mazda lamp or fluorescent lamp gives off white 

 light, although the lights are yellowish and 

 bluish, respectively. Even the color of dajdight 

 varies considerably^ It is more reddish in the 

 morning and late afternoon, and its color depends 

 greatly on the number of clouds in the air, the 

 time of year, and so on. The International 

 Commission on Illumination defined several 

 whites and the one shown in Fig. 45 is lUuminant 

 C. This is about the kind of light that comes 

 from a summer sky when the air is clear, there are 

 no clouds, and the sun is high. The point E, 

 incidentally, is the color of all those different 

 wavelength combinations shown in Fig. 44. It is a 

 little more yellowish than lUuminant C. 



