The purposes of this study were to document the dimensions of the intercandylar notch in the normal knee; to compare normal knee notches stenosis and femoral intercondylar roof angle to those knees with ACL tears; to compare sexual difference in normal knee to determine if there is a relationship between femoral intercondylar roof angle and notch stenosis and ACL tears. We analyzed 128 MRI of knees taken using the Signa 1.5T MR machine between the ages of 18 and 46 from Feb. 1995 to Feb. 1996. The Group 1 was ninty-one normal knees. The Group 1-F was the normal twenty-two knees of female. The Group 1-M was the normal sixty-nine knees of male. The Group 2 was thirty-seven knees with MRI and arthroscopically confirmed ACL tears within 2 months after injury. We measured the femoral intercondylar roof angle, open notch angle, ratio of notch width at two-thirds of the notch height to condylar width, and ratio of maximum notch width to condylar width from sigittal, transverse cut of MRI. The measurements of the two groups were compared for statistical significance using the student's t-test. Statistically significant differences were found between normal (Group 1) and ACL injured knees (Group 2) in regard to ratio of notch width at two-thirds of the notch height to condylar width, and ratio of maximum notch width to condylar width, but no significant differences were found in the femoral intercondylar roof angle, and open notch angle, suggesting a significant association between anterior outlet stenosis and ACL tears. And statistically significant difference was found only in the ratio of maximum notch width to condylar width between normal male (Group 1-M) and female (Group 1-F), suggesting more stenotic in male group. Although a stenotic femoral intercondylar notch may contribute to a torn ACL, the intercondylar roof angle and open notch angle does not correlate with an ACL tears.