One hundred fifty-six patients with a total rupture of the anterior cruciate ligament (ACL) were reexamined 41 to 80 months after injury. They were randomized to three treatment groups: (1) repair and augmentation of the ACL with an iliotibial strip, (2) repair without augmentation, and (3) nonsurgical ACL treatment. Associated injuries of menisci and other ligaments were treated in the same way for the three groups. Two-thirds of the patients in the nonsurgically treated group complained of instability and 17% had had a subsequent reconstruction of the ACL at the follow-up examination. The group treated with an augmented repair had a less abnormal laxity measured by a laxity-testing device. Sixty-three percent returned to competitive sports, as compared with 27% in the nonsurgical group and 32% in the only repair group. Relative strength of the quadriceps and hamstrings muscles were similar for all groups. The augmented-repair group had better hop tests, reflecting a superior stability, whereas running was not affected by treatment but was correlated with the activity level.