Ranitidine produces a blood concentration curve with a pronounced secondary peak when administered orally and parenterally. A pharmacokinetic model is proposed to describe this reabsorption phenomenon. The choice of a discontinuous cyclic transfer was justified on the basis of physiological considerations and the good agreement with data from oral and intravenous administration. It is proposed that ranitidine accumulates mainly from the systemic circulation into a depot from which drug and bioreversible drug are spontaneously released in response to food intake. The evaluation of the extent of the first-pass effect and the evaluation of bioequivalency are complicated by the model-independent AUC approach because the area under the concentration versus time curve (AUC) is dependent on the extent of recycling and thus does not properly reflect the extent of primary absorption. By using intravenous administration as a reference dosage form and the integrated form of the regression equations to calculate the AUC values, the bioavailability of the oral dose was found to be 0.56, which corresponds well with the value of 0.58 obtained by linear-log-linear integration. The least-squares parameter estimate of the primary absorption is 0.43.