Proprioceptive function of the knee was quantified and compared in three groups of patients: those with anterior cruciate ligament deficiency, with hamstring tendons-ligament augmentation device anterior cruciate ligament reconstructions, and with bone-patellar tendon-bone anterior cruciate ligament reconstructions. A total of 32 subjects, including 6 uninjured control subjects, were tested for threshold to perception of passive motion of the knee. All other sensory input was neutralized and testing occurred in the 30 degrees to 40 degrees range of knee flexion. The noninvolved contralateral knee served as a control for each subject. Each leg was moved at 0.5 deg/sec into flexion or extension in a random sequence. The variables of age, KT-1000 arthrometer scores, injury-to-surgery interval, injury-to-followup interval, and patient satisfaction were statistically analyzed for correlation with threshold to perception of passive motion of the knee. Control subjects showed no statistically significant differences in threshold between their two knees. The three test groups all showed significantly higher values in the involved knee compared with the noninvolved knee (P < 0.01). However, no statistically significant differences were found between the groups, including controls, with respect to mean threshold to perception of passive motion. According to these results, anterior cruciate ligament reconstruction did not improve proprioception in the patients in this study.