THEORY AND MEASUREMENT OF VISUAL MECHANISMS : XII. ON VISUAL DUPLEXITY

Flicker contours from vertebrates (fishes to man) show that the slope parameter σ'(log I) in the efficiently descriptive probability summation 100 F/F(max.) = ∫(–inf;) (log I) e (–(log I/I(i))(–(log I/I(i))2)/2(σ')(/2(σ')2)) ·d log I is distributed bimodally (simple fields, "white" light), from 0.60 to 2.3, with well defined peaks at 0.80 and 1.75. This parameter is independent of F(max.), log I(i), temperature, light-time fraction, and in general not greatly influenced by λ. "Rod" components of known visually duplex contours, without exception, and some "cone" contours, are in the first group; an equal number of "cone" curves are in the second group, together with one simplex "rod" contour; purely cone contours are in each group, as well as cone segments of duplex curves. No firm zoological grouping of the "cone" curves can be made, on present evidence,—although the 5 fishes used give high-slope curves, 2 amphibians low slopes, reptiles (5) either high or low, birds (2) and anthropoids (2) low-slope "cone" curves. By subdivision of the visual image and by change of wave-length, under certain conditions, in man, and by use of the "pecten effect" in birds (and man), cone contours of the low-slope class can be transformed into curves of the high-slope group. These procedures do not fundamentally change the "rod" slopes. Consequently, although under simple conditions they are specifically determined, the forms of the F - log I contour cannot be used as diagnostic for rod or cone functioning. It is reinforced, by new data on Anolis (lizard) and Trionyx (turtle), that an obviously duplex retina is specifically correlated with a duplex performance contour, a simplex retina with a simplex one. But no support is given to the view that the shapes of these curves are diagnostic of differences in rod or cone fundamental excitabilities, or that they describe properties of these units. In visual duplexity we have to do simply with the fact that two groups of neural effects are available; it is with their properties that we deal in measurements of duplex visual excitability.