An analysis of the patellofemoral joint based on experimental determinations of pressure distributions on the patellar cartilage and vectorial calculations is presented. The extension torque of all quadriceps muscles is similar. The force transmitted to the patella by the retinacula is small in all knee positions. Because the rigid patella has its own lever arms and tendon insertions at different heights, the forces on patellar and quadriceps tendons are unequal. The small patellar contact areas change according to flexion and are dependent on quadriceps force. Retropatellar pressure is thereby limited and becomes independent of flexion angle. By virtue of the direct force transmission of the quadriceps tendon to the femur, the increase in the retropatellar force is strongly limited beyond 70 degrees. The lateral-to-medial ratio of contact areas, pressing forces, cartilage areas, and bone mass is always 1.6:1, whereas the mean pressure is the same on both facets. The osseous form of the patella does not allow identification of dysplasia because cartilage compensates for apparent incongruence, especially about the medial ridge. This permits the rigid patella to adapt itself to very different abutments. Its individual form results from functional adaptation and does not explain chondromalacia. The ventralization of the tibial tuberosity does not alter retropatellar loading. The mediodistal transfer results in a high increase of pressure on both facets.