Specially designed load-transducers that measured the resultant forces exerted by the posterior and anterior cruciate ligaments on their respective femoral and tibial insertions were applied to eighteen fresh-frozen cadaveric knees for a series of controlled loading experiments. The mean force in the posterior cruciate ligament at 5 degrees of forced hyperextension of the knee was 23 per cent of the mean force in the anterior cruciate ligament. When the knee was hyperflexed by application of 10.0 newton-meters of bending moment to the tibia, the mean force in the posterior cruciate ligament was 55 per cent of that in the anterior cruciate ligament. Quadriceps tendon pull increased the force in the posterior cruciate ligament in twelve of the fourteen specimens to which it had been applied, at 80 and 90 degrees of flexion only. The force generated in the posterior cruciate ligament by applied internal tibial torque was greatest when the knee was in 90 degrees of flexion; the force in the anterior cruciate ligament was greatest when the knee was fully extended. External tibial torque generated force in the posterior cruciate ligament in only eight specimens, and only at 80 and 90 degrees of flexion. The levels of force that were generated in the posterior cruciate ligament by applied varus and valgus bending moment were greatest at 90 degrees of flexion of the knee; the levels of force in the anterior cruciate ligament were greatest with the knee in full extension. With the knee flexed 90 degrees and the tibia in neutral rotation, fifty newtons of applied posterior tibial force increased the mean force in the posterior cruciate ligament by 58.4 newtons; at full extension, no increase in the force in the ligament was recorded, indicating that tensed capsular structures were absorbing the applied load. When the tibia was internally or externally rotated by applied tibial torque, the increases in the force in the ligament from applied posterior tibial force were sharply diminished.