Aims: Kinematic and pointing procedures, are used for non-image based navigated implantation of TKA. Pointing procedures require exact knowledge about the landmarks. In this anatomical study, landmarks are defined and repeatedly referenced. The precision and the reproducibility are evaluated, by means of inter- and intra- observer study. Using the landmarks, the axes of the femur and tibia are calculated. Methods: The specific landmarks of 30 femur and 27 tibia specimens, were palpated by 3 surgeons and digitised by means of a photogrammetric system, as used intra-operatively. The recorded data are evaluated. Results: The specific landmarks can be referenced with great precision. The vectors that influence the implant position, show femoral a mean inter-observer deviation of 0,9mm and 1,0mm tibial. The repeating accuracy of every single observer was 1,5mm femoral and 1,0mm tibial. The calculated long axes at the femur and tibia, thus reach a precision of 0.1° (min-max:0°–0,9°) at the femur and 0,2° (min-max:0°–1,1°) at the tibia. The short axes at the distal femur and at the proximal tibia, exhibit an average deviation of 0,7° to 1,9° (min-max: 0°–11,3°). Conclusion: Long axes (mechanical axes) can be determined exactly, the precision of the short axes (rotational axes) is unsatisfactory, although palpation of landmarks were accurate. Therefore, palpation of more than one rotational axis at the femur and the tibia, is mandatory and should be visualized on the monitor during the operation.