SUMMARY1. Atypical (blue cone) monochromats show two action spectra when tested by the increment threshold method of Stiles with' central' fixation. One spectrum peaks near 450 nm and has the spectral characteristics of normal blue cones. The other resembles rhodopsin (r0) modified slightly by photostable macular pigment.2. Under some circumstances such observers are dichromats. There is a neutral point (matched to Illuminant 'C') in the neighbourhood of 460-470 nm.3. The spectral colour matching functions using two primaries have been measured on three such subjects. They can be fit reasonably well, although imperfectly, by linear combinations of oro and nr1. The chromaticity co-ordinates have been calculated according to the convention of W. D. Wright and compared to the results predicted from no and or1. The comparison suggests that part of the imperfections of the colour matching function fit is due to prereceptor differences (e.g. macular pigment) between the blue-cone monochromats and the hypothetical sTOp, iT observer.4. Colour matches made at high light levels continue to hold when the intensity of the field is reduced below the cone threshold.5. Therefore one of the visual pigments participating in the colour matches has an action spectrum which is not measurably different from that of the rod pigment rhodopsin.6. Increment threshold measurements show that the mechanism which has the rhodopsin action spectrum has the directional sensitivity of cones, not rods.7. Blue test threshold during dark adaptation after a full bleach follow a bipartite exponential recovery curve. The first exponential has a time constant of 1 min, the second 2 min. By comparing these curves to the increment thresholds it is possible to relate the first to the oTr, the second to the 7T0 cones.8. Using a broad band blue gelatin filter in the measuring light of the M. ALPERN AND OTHERS retinal densitometer and studying the same retinal region tested in 7 it is possible to follow the regeneration of a pigment after a full bleach which has an exponential recovery with a time constant of 10 min. With a yellow green filter in the measuring light the exponential recovery observed after a full bleach has a time constant of 2-0 min. 9. One of the two visual pigments participating in the colour matches resides in receptors which have the action spectrum, the directional sensitivity and probably the dark adaptation curve of normal blue cones.10. The other resides in receptors which have the action spectrum of normal rods but the directional sensitivity and the dark adaptation curve of normal red and green cones.

THE KINETICS of the normal human red and green cone pigment chlorolabe and erythrolabe have been extensively studied by Rushton in three important papers (RUSHTON, 1958, 1963, 1965). With his retinal densitometer he measured the fraction p of fovea1 pigment left unbleached by a 10 set flash of intensity I,. This was done by suddenly substituting for I, at the end of the 10 set a smaller value I, chosen so that further photolysis was exactly balanced by regeneration and p was seen to remain steady indefinitely. This gave plenty of time for a good measurement of (l-p), the fraction bleached by the 10 set exposure to I,. Assuming that in uivo (as in vitro) the photolysis rate is proportional to the quantum catch Z.p, and that during the 10 set exposure regeneration is negligible (as was confirmed experimentally),

2. The action spectrum of the response obtained in this way agrees reasonably well with the observer's psychophysical foveal luminosity curve.3. For the peripheral retina, the action spectrum is similar to that of the fovea when allowance is made for differences in screening macular pigment.4. Such responses diminish when the test stimulus is focused on to the peripheral retina and disappear when the test light is focused on the blind spot.5. Therefore, the response to the test light fixated centrally is the result of the excitation only of cones mainly, if not exclusively, in the fovea.6. When the intensity of the background is reduced by a factor of 10, the action spectrum shows evidence of the effect of excitation of rods in the blue part of the spectrum and of cones in the red. These red and blue responses add linearly when combined together, provided they are adjusted to coincide in phase.

No abstract

No abstract