The human acetabulofemoral joint is commonly modelled as a pure ball-and-socket joint, but there has been no quantitative assessment of this assumption in the literature. Our aim was to test the limits and validity of this hypothesis. We performed experiments on four adult cadavers. Cortical pins, each equipped with a marker cluster, were implanted in the pelvis and the femur. Movements were recorded using stereophotogrammetry while an operator rotated the cadaver's acetabulofemoral joint, exploiting the widest possible range of movement. The functional consistency of the acetabulofemoral joint as a pure spherical joint was assessed by comparing the magnitude of the translations of the hip joint centre as obtained on cadavers, with the centre of rotation of two metal segments linked through a perfectly spherical hinge. The results showed that the radii of the spheres containing 95% of the positions of the estimated centres of rotation were separated by less than 1 mm for both the acetabulofemoral joint and the mechanical spherical hinge. Therefore, the acetabulofemoral joint can be modelled as a spherical joint within the considered range of movement (flexion/extension 20 degrees to 70 degrees ; abduction/adduction 0 degrees to 45 degrees ; internal/external rotation 0 degrees to 30 degrees ).