Definition of anatomical reference frames is necessary both for in vitro biomechanical testing, and for in vivo human movement analyses. Different reference frames have been proposed in the literature for the lower limb, and in particular for the tibia-fibula complex. The scope of this work was to compare the three most commonly referred proposals (proposed by [Ruff, C.B., Hayes, W.C., 1983. Cross-sectional geometry at Pecos Pueblo femora and tibiae -A biomechanical investigation: I. method and general patterns of variation. American Journal of Physical Anthropology 60, pp. 359-381.], by [Cappozzo, A., Catani, F., Della Croce, U., Leardini, A., 1995. Position and orientation in space of bones during movement: anatomical frame definition and determination. Clinical Biomechanics (Bristol, Avon) 10, pp. 171-178.], and by the Standardization and Terminology Committee of the International Society of Biomechanics, [Wu, G., Siegler, S., Allard, P., Kirtley, C., Leardini, A., Rosenbaum, D., Whittle, M., D'Lima, D.D., Cristofolini, L., Witte, H., Schmid, O., Stokes, I., 2002. ISB recommendation on definitions of joint coordinate system of various joints for reporting of human joint motion-part I: ankle, hip and spine. International Society of Biomechanics. Journal of Biomechanics 35, pp. 543-548.]). These three frames were identified on six cadaveric tibia-fibula specimens based on the relevant anatomical landmarks, using a high-precision digitizer. The intra-operator (ten repetitions) and inter-operator (three operators) repeatability were investigated in terms of reference frame orientation. The three frames had similar intra-operator repeatability. The reference frame proposed by Ruff et al. had a better inter-operator repeatability (this must be put in relation with the original context of interest, i.e. in vitro measurements on dissected bones). The reference frames proposed by Ruff et al. and by ISB had a similar alignment; the frame proposed by Cappozzo et al. was considerably externally rotated and flexed with respect to the other two. Thus, the reference frame proposed by Ruff et al. is preferable when the full bone surface is accessible (typically during in vitro tests). Conversely, no advantage in terms of repeatability seems to exist between the reference frames proposed by Cappozzo et al. and ISB.