Six knees from cadavers were tested for change in stability after release of the medial collateral ligament with posterior cruciate-retaining and substituting total knee replacements. Load deformation curves of the joint were recorded in full extension and 30°, 60°, and 90° flexion under a 10 N-m varus and valgus torque, 1.5 N-m internal and external rotational torque, and a 35 N anterior and posterior force to test stability in each knee. The intact specimen and posterior cruciate ligament-retaining total joint replacement were tested for baseline comparisons. The superficial medial collateral ligament was released, followed by release of the posterior cruciate ligament. The knee then was converted to a posterior-stabilized implant. After medial collateral ligament release, valgus laxity was statistically significantly greater at 30°, 60°, and 90° flexion after posterior cruciate ligament sacrifice than it was when the posterior cruciate ligament was retained. The posterior-stabilizing post added little to varus and valgus stability. Small, but significant, differences were seen in internal and external rotation before and after posterior cruciate ligament sacrifice. The posterior-stabilized total knee arthroplasty was even more rotationally constrained in full extension than the knee with intact medial collateral ligament and posterior cruciate ligament.