A model for testing the properties of gliding tendon grafts has been developed that allows anastomoses to be evaluated separately from the mid-portion of the graft. In addition, two different graft materials may be implanted in one sheep foreleg whilst maintaining control (not operated) tendons in both the operated leg and contralateral foreleg. The model has been used to evaluate the response of xenografts made from chemically treated kangaroo tail tendon (KTT) compared with autografts. At 3 month the mid-sections of the glutaraldehyde-fixed xenografts maintained between 57 and 82% of their initial ultimate tensile strength whereas lyophilized KTT dropped to 10% and autografts retained 91% of initial strength. Sterilization by gamma-radiation of wet xenografts did not affect the material and implant properties significantly. Longer term studies are necessary to determine the resorption behaviour of the xenografts. Anastomosis strengths were found to be about the same for all grafts, at about 25% of the strength of the original tendon. Alternatives need to be investigated to improve this strength.