Comparative bioavailability studies should be designed and the resulting data evaluated based on estimates of both intersubject and intrasubject variances in the kinetic parameters for the particular drug products(s) being studied. This report presents the results of two comparative bioavailability studies. In the first study, three production lots of macrocrystalline nitrofurantoin capsules (Macrodantin) were compared in 21 subjects, and in the second study, capsules from one production lot were administered to 21 different subjects on three occasions. Both model-independent kinetic parameters for urinary excretion and a one-compartment model with zero-order absorption were used to evaluate both the rate and the extent of bioavailability. Overall the results showed a very low variance between and within production lots and a relatively large intersubject variance in the rate and extent of absorption.