Van Nes and Bouman made measurements similar to Patel and van Meeteren and Vos, but using monochromatic light at three wavelengths, and spanning across six decades of retinal illuminance. The salient features of their experiments are reported, which is structurally similar to. They used transparent sinusoidal gratings retroilluminated by the output light from a double-prism monochromator, choosing green (525 nm), red (650 nm), and blue (450 nm) wavelengths. Vision by the observer was monocular through an artificial pupil of 2 mm. Modulation threshold was determined by the method of adjustment, yet the authors varied the modulation in fixed steps of one tenth of a decade (0.1 log unit). In this way, they achieved both an upper and lower limit for the detection threshold. Doubts regarding the accuracy of Van Nes and Bouman's published data were raised in Barten's book, where he had to adopt a value of quantum efficiency 0 ten times greater than his mean value to obtain a satisfactory fitting. He did not try to compose the discrepancy in any way, only observing that for the vNB experiments, "the quantum efficiency is unlikely high compared with the results of other investigations," and to match them "the measured retinal illuminance should have been a factor 10 higher."
Online access to SPIE eBooks is limited to subscribing institutions.