Recent evidence suggests that ligaments are subject to repetitive loads in vivo. Hence, the creep behaviour (increase in strain under constant or repetitive stress) of ligament scars is of significance, since healing ligaments may elongate permanently over time. A rabbit medial collateral ligament model was used to assess the creep behaviour of healing ligaments at stresses corresponding to 30% of the scar failure strength at 3 (n = 6), 6 (n = 6), and 14 (n = 5) weeks of healing. The stresses for the creep tests of scars (and contralateral controls) were 2.2, 4.1, and 7.1 MPa for the 3, 6, and 14-week healing intervals, respectively. Normal medial collateral ligaments from comparable rabbits were tested at two of the corresponding stresses: 4.1 (n = 7) and 7.1 (n = 6) MPa. Total creep strain-the cumulative increase in strain resulting from serial cyclic and static creep testing-was independent of the order of testing and was compared between scars and controls. Water contents after testing were also quantified. Water contents before testing were assessed for additional animals: six normal animals and three from each healing interval. At 3 weeks of healing, the total creep strain of medial collateral ligament scars was four times greater than that for contralateral controls tested at the same stress. Although there was improvement from 3 to 14 weeks, the total creep strain of scars remained more than two times greater than that of controls at 14 weeks. Scar water content decreased with healing from elevated initial values, possibly contributing to the marginally improved creep response. Comparisons of this deficiency in scar creep with previously published scar abnormalities in the same model suggest that collagen crosslink density, proteoglycan content, soft-tissue flaws, and the combined effect of collagen fibre changes may be mechanistic factors involved in scar creep.