The author holds that the phi-gamma hypothesis is theoretically unsound in that it demands a constant limen for all stimulus magnitudes, and that although it may yield, in some cases, a good approximate fit for the psychometric curve "it should be thought of not as a psychophysical hypothesis but rather as any other empirical equation which is used to draw a smooth curve through a set of experimental observations." For two-category judgments the psychometric curve is positively skewed if the absolute limen increases with the stimulus, the amount of skew depending upon the coarseness of the perceptual proces. "The phi-gamma hypothesis is denied if the absolute limen increases with the stimulus magnitude even if Weber's law is not exactly verified." Approximation of fit increases, theoretically, as Weber's ratio becomes smaller. (PsycINFO Database Record (c) 2006 APA, all rights reserved)