Central to the use of fluorescein in vascular diagnosis is the requirement that the intensity of evoked fluorescence be proportional to blood flow. With the introduction of the digital dermofluorometer, a device that quantitates cutaneous fluorescence, establishment of this relationship has become possible. After experimentally producing measured reductions in the distal aortic flow of eight rabbits, the ratio of fluorescence in the flow-restricted and unrestricted areas was obtained by measuring hind- and forelimb fluorescence. At any time between 20 and 60 minutes following a bolus injection of sodium fluorescein (1 mg/kg body weight), there was a significant linear relationship (p less than 0.05, r greater than 0.75) between residual aortic flow and the ratio of hind-/forelimb fluorescence. Simultaneously obtained plasma fluorescein concentrations decayed rapidly by first-order kinetics with a half-life of 12.5 minutes, regardless of the degree of distal aortic occlusion. The time course of the rise and fall of cutaneous fluorescence was slower than that of the plasma fluorescein concentration, proving that interstitial rather than intravascular fluorescein was responsible for the measured fluorescence. We conclude that the intensity of tissue fluorescence is linearly related to blood flow and that conclusions regarding perfusion may be drawn from relative fluorescence at any time between 20 and 60 minutes following a bolus injection of fluorescein. Furthermore, the passage of fluorescein into the interstitium is dependent on a time-limited diffusion process, which along with flow, establishes the time to peak and the absolute amplitude of the tissue fluorescence curve.