The risk of incorporating inapparent recirculation under the extrapolated downslope of dye dilution curves is greater with exponential than gamma function extrapolation since the latter makes use of ascending and early descending limbs of the curve. Extrapolation of a true gamma function is not affected by the level at which extrapolation begins. With curves obtained in eight normal subjects, cardiac output (phi) was comparable by exponential and gamma extrapolation if the latter began between 75 and 55% of the peak concentration (Cmax.). phi was underestimated or overestimated according to whether extrapolation began higher or lower than the above limits. Therefore, experimental curves were not true gamma functions. However, the level at which extrapolation began had little effect on pulmonary mean transit time and pulmonary blood volume (PBV) calculated by use of the double injection-single sampling method. Similar effects on phi of the level at which extrapolation by a gamma function begins were found with 24 chronic bronchitics. PBV calculated with a gamma function extrapolated from 0.75 X Cmax. downwards and with an exponential function averaged approximately 340 ml and did not differ from one another. The data suggest that the low values of PBV in chronic bronchitis are not an artifact due to the method of extrapolation.