The factors influencing the range of motion of the knee after total condylar knee replacement were analyzed in fifty-five consecutive patients who had seventy-one replacements and a minimum follow-up of two years. Statistically significant improvement in flexion occurred between the first postoperative evaluation in the hospital, the evaluation at discharge, and the evaluations at three months and one year after replacement. Postoperative range of motion was not influenced by the patient's gender, by the extent of the disease (monoarticular in comparison with polyarticular), or by treatment with a unilateral or bilateral procedure. The knees of patients with rheumatoid arthritis gained extension at the expense of flexion. The average flexion arcs of the knees with rheumatoid arthritis and of those with osteoarthritis were the same two years after operation as they had been preoperatively. Knees with preoperative flexion of more than 100 degrees had lost flexion at follow-up, whereas those with preoperative flexion of less than 100 degrees had gained flexion. Significant improvement in extension occurred only before discharge. For the patients whose preoperative flexion contracture was 10 degrees or more, virtually all improvement in the contracture occurred at the time of surgery. Patients who had a simultaneous bilateral total knee replacement did as well as those with a single knee replacement.(ABSTRACT TRUNCATED AT 250 WORDS)