The technique of Anjou and Krakau for measuring flare and Krakau's method for estimating aqueous humor flow in humans was compared to the fluorophotometric procedure of Jones and Maurice.The precision of flare was measured in 20 normal human volunteers, and the precision of flow was measured with fluorophotometry in 24 volunteers. In addition, the circadian rhythms of both flow and flare were measured simultaneously in a separate group of 20 subjects. This simultaneous measurement allowed us to calculate the entry rate of scattering substances into the aqueous humor as a function of time of day. This entry rate is critical for measuring flow by the method of Krakau.The mean coefficient of variation (±SD) of repeated measurements of flare at the same time of day was 23.2%±10.3%. The mean coefficient of variation of the measurement of aqueous flow by fluorophotometry was 15.9%±8.2%, significantly better than the measurement of flare (p=0.01).The daytime entry rate of scattering substances into the aqueous humor varied from 1.39±0.86μg/min (calculated as albumin) to 1.05±0.58μg/min. The rates in daytime hours were not significantly different from each other. At night during sleep, the entry rate was 0.67±0.48μg/min, significantly slower (p between 0.002 and<0.001). These data indicate that the flare technique cannot be used alone to calculate the circadian rhythm of flow and that direct measurement of flow by fluorescein clearance is likely to be the most repeatable and accurate of the two.