Normal knee function depends on the integrity of the collateral and cruciate ligaments. These ligaments consist of multifiber bands with complex patterns of insertion to the femur and tibia. Because of the frequency of anterior cruciate injury and the potentially serious consequences of anterior cruciate deficiency, it is often necessary to reconstruct the ruptured ligament with biologic or synthetic material. In this procedure, the surgeon seeks femoral and tibial attachment sites which allow the ligament reconstruction to remain supportive, but not excessively tight, throughout the joint's range of motion. In the proposed research we will identify potential ligament attachment sites on the femur and tibia which satisfy this requirement. We will digitize both the anatomy of cadaver knees and the motion of these knees under defined loading protocols. In each knee, we will use the quantified knee motion to characterize the length patterns of all potential ligament attachment sites on the femur and tibia. Using computational optimization techniques, we will locate all potential anterior cruciate attachment sites which have specified length patterns (for example, near-isometry) over the full range of flexion. We will characterize these insertion sites in terms of their size, shape, and relation to anatomic landmarks. We will also locate ligament attachment sites which are nearly isometric over certain subsets of the full arc of motion and which would not result in excessive ligament strain over the remaining range. We will compare these sites with the attachment sites of anatomic ligament bands. These investigations will characterize the length relationships of all potential ligament attachment sites in knees moving under defined loads and will correlate these observations with anatomic ligament insertion sites. These studies will provide insight into the structure-function relationships of knee ligaments and help surgeons understand the challenge of insertion site selection in ligament reconstruction.