Progesterone was administered to postmenopausal women in a form of vaginal suppositories containing 100 and 200 mg active substance in Butyrum cacao (BC) and Massa estarinum (ME), a base with emulsifying properties. In the case of single doses, blood samples were taken at 2, 4, 6, 24, 48 and 72 h. Another group of patients received vaginal suppositories (100 mg progesterone) once a day for a 6 day period, with blood samples taken 12 h after each administration. The plasma levels of progesterone were evaluated by radioimmunoassay. The time of maximum concentration (tmax) was 4 h in most cases, and 6 h in the others. The plasma levels were not dose-proportional. Peak plasma concentrations were in the range of 10-15 ng/ml with a mean of 10.5 ng/ml for the 100 mg and 12 ng/ml for the 200 mg doses. The ratio of the mean area under the curve (AUC) for 200 mg and the mean AUC for the 100 mg dose was found to be 1.37. Replacing BC with ME resulted in the lowering of cmax and AUC, and an increase in tmax following a reducing in the rate and extent of adsorption. In the case of ME suppositories, the variability in AUC, cmax and tmax was greater compared to that observed with the BC suppositories. Elimination half-time was in the range of 9-10 h for BC and 14 h for ME suppositories. In vitro assessment of the release kinetics from a hydrophobic and an emulsion type base confirmed previous findings: the latter base assured better pharmaceutical availability. The repeated doses did not seem to produce an accumulation of progesterone in the plasma. On the contrary, a small decrease in plasma levels over time appeared during the 6 day period. Numerical analysis revealed an excellent goodness of fit for the in vivo experimental data via biexponential curves, i.e. a pseudomonocompartmental model.