The author reviews several phenomena occurring in sine wave flicker experiments, classifying them into low-frequency (below 10 cps) and highfrequency (above 20 cps) effects, and suggests that two different, kinds of mechanisms are needed to account for the behavior of amplitude sensitivity curves in these two frequency ranges. At low frequencies, the threshold modulation ratio is independent of the time-average luminance or adaptation level of the flickering stimulus (but decreases significantly with increasing field size); whereas at high frequencies, the threshold amplitude of the sinusoidal component alone (measured in retinal illuminance units) is independent of adaptation level. In the intermediate frequency band, a smooth transition occurs from one type of behavior to the other. These facts simplify the well-known calculation of classical flicker-fusion thresholds from sine-wave data (first demonstrated by de Lange). The present paper discusses such calculations in some detail, developing the concept of the sinusoidal equivalent stimulus and explicitly stating the reducibility hypothesis for various flicker-fusion waveforms. Additional equivalence tests are provided by periodic-flash experiments conducted for this purpose and reported here.