In this article we have tried to unify several different lines of evidence bearing on how neuromuscular mechanisms might contribute to the etiopathogenesis of OA. Several key points bear emphasis. (1) The neuromuscular system is both the greatest threat that the joint faces and the principal means by which joints are protected, because this system exercises the definitive influence on the spectrum of forces to which the joint is exposed. If muscle contraction is not properly coordinated, the joint will exceed its normal extreme of excursion and the loading of its cartilage will be excessive. This trauma will be manifest pathologically as OA. It follows that if the neuromuscular system cannot control the mechanical environment of the joint, the articular and periarticular tissues will break down even in the face of treatment with "chondroprotective" drugs. (2) Speculation about the role of sensation in joint protection is pointless unless sensory ablation influences the development of OA. (3) Only a small minority of patients with severe peripheral sensory neuropathy develop a Charcot joint, and a clear correlation between the development of a Charcot joint and the severity of the sensory neuropathy or the activity level of the patient has yet to be established. Until proved otherwise, it should not be assumed that ipsilateral sensation plays a role in protecting the stable joint from breakdown. (4) There have been surprisingly few animal experiments in which the effect of sensory nerve ablation on synovial joints has been examined. Although the results of the early studies are not entirely consistent, they suggest that extensive deafferentation of a limb does not necessarily lead to joint pathology in the absence of exogenous trauma or infection. Such studies should be repeated under carefully controlled conditions. (5) Recent work in dogs suggests that ipsilateral sensation is not important in protecting the stable joint from OA but is necessary to protect the unstable joint from rapid deterioration. Moreover, there is evidence that sensation is temporarily important in protecting the unstable joint but that the CNS eventually acquires the ability to protect the unstable joint even in the absence of ipsilateral sensory input.