Knees of mature dogs were immobilized for 6 weeks by long-leg casts allowing 8°–15° of motion, a model studied by others, or with external fixators, a new, more severe model that kept the joints rigid. Some animals were allowed to recover for 1 week after the immobilization period. Articular cartilage was examined histologically and biochemically. After 6 weeks of immobilization, water increased 7% in both casted and fixator-immobilized joints compared with normal knee cartilage, while hexuronic acid was 23 and 28% lower, respectively. The limited motion permitted by the casts resulted in a smaller depression of proteoglycan synthesis and less proteoglycan loss during immobilization than occurred in the rigid external fixator group. The protective effect of limited motion was shown clearly during the recovery period:​ as measured by hexuronic acid content. cartilage from the casted joints had almost recovered within 1 week, whereas the external fixator group experienced little or no recovery during the week after treatment. In contrast to previous studies by others with casted joints, both newly synthesized [³⁹S]sulfate-labeled and accumulated unlabeled proteoglycans from both casted and fixator-immobilized cartilages were able to form complexes with exogenous hyaluronic acid to the same extent as those from control cartilage. Thus, in immobilized cartilage, failure of the newly synthesized proteoglycan to bind to hyaluronate is not a mechanism of accelerated proteoglycan loss. The accelerated proteoglycan turnover appears to be caused by a combination of decreased synthesis and increased proteolysis of the secreted proteoglycans. Clinically external fixators used to immobilize joint injuries should be applied for the shortest time possible and if possible should be mechanically modified to allow up to 10° of joint motion.