For many years, exercise was not a recommended rehabilitation strategy for persons with a diagnosis of idiopathic Parkinson's disease (PD). Since it was believed that exercise had no measurable effect on PD, or might worsen the underlying pathology, it was to be avoided. A rich vein of bench and translational research now suggest non-pharmacological approaches, such as exercise or physiotherapy, have a far greater effect on the cardinal features of PD than previously believed. In particular, recent studies utilizing animal models of PD have begun to explore the molecular mechanisms of exercise-induced changes in the pathophysiology of PD. Yet, many clinicians and communities remain unaware of the scientific literature underlying exercise-induced brain repair or reorganization (neuroplasticity) and accompanying behavioral recovery in animal models of PD. The authors will summarize some noteworthy preliminary studies suggesting that continuous, deficit targeted, intensive training may confer neuroprotection and thereby, slow, stop or reverse the progression of the disease or promote neurorestoration through adaptation of compromised signaling pathways. While much work remains and these preliminary results await replication in larger prospective human trials, we believe a major challenge in the field of non-pharmacological, rehabilitative intervention for PD will be the extent to which healthcare providers are able to translate the science of exercise and PD to the level of the community.