464 ADAPTATIONS TO PHOTIC QUALITY 



has yet been clearly demonstrated to have color vison. The tieup of cones 

 and color vision is entirely to be expected and is not accidental. Rod 

 vision, in the vast majority of nocturnal and twenty-four-hour animals, is 

 even much more diffuse and unclear than in ourselves. Where visual 

 acuity is so low as to be little more than movement- and silhouette-per- 

 ception, contours and contrasts are so ill-defined to begin with that per- 

 ceived color-differences could add nothing to visibility. There were 

 almost certainly cones before there were rods, but there was probably no 

 color vision in the vertebrate world until retinal and general ocular struc- 

 ture had progressed to the point where an optical basis for decent visual 

 acuity had been laid. So, when color vision did arrive, it was only logical 

 that it be installed in the cone mechanism. Even if the acuity of rod vision 

 were always equal to that of cone vision (which seems to be true only of 

 the frog), the operation of Weber's law (p. 534) would still lead to fatal 

 reductions of contrast in the intensities of illumination suited to rod 

 activity. Color vision could be of value only in the photopic visual mech- 

 anism of animals with diurnal activities. 



The retinal mechanism of human color vision may be much simpler 

 than we are fond of imagining; but on the assumption that the human 

 retinal process is complex — so much so that it must have evolved step by 

 step over a painfully long period of vertebrate history — ^biologists have 

 long been interested in the question of where, and in what degree of com- 

 pleteness, color perceptions first appeared, like Christmas decorations, 

 upon the phylogenetic tree. Psychologists have hoped, firstly, that by 

 working out the color-vision systems of a series of vertebrate types, they 

 might be able to identify simpler systems than the three-primary or tri- 

 chromatic human one, which would then represent stages through which 

 the human system evolved. The various dichromatic or other reduction 

 systems of occasional humans might then be interpretable as atavisms. 

 And, secondly — ^holding the quite unwarranted conviction that the 

 chromatic photochemical system must have differentiated from rhodopsin 

 — they have hoped to find, in the retinae of lower animals, chemical way- 

 stations which would justify the assumptions of one or another of the 

 metabolic or genetic theories of human color vision, such as the elaborate 

 ones of Hering and Ladd-Franklin. 



Before the discovery of visual-cell transmutation in 1934, the first of 

 these hopes was quite reasonable. It may still be, and the color-vision 

 tenet of the Duplicity Theory — that only cones can mediate color vision 

 and that no rods do so — may hold quite strialy for all vertebrates, for 



