Functional models of Parkinson's disease (PD) have led to effective treatment for the motor symptoms. Toxin-based models, such as the 6-hydroxydopamine-lesioned rat and 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine-treated primate, have resulted in novel dopaminergic therapies and new therapeutic strategies. They have also been used to study processes underlying motor complications, particularly dyskinesia, and for developing pharmacological approaches to dyskinesia avoidance and suppression. Symptomatic models of PD based on nigrostriatal degeneration have a high degree of predictability of clinical effect of dopaminergic drugs on motor symptoms in humans. However, the effects of nondopaminergic drugs in these models do not translate effectively into clinical efficacy. Newer experimental models of PD have attempted to reproduce the pathogenic process and to involve all areas of the brain pathologically affected in humans. In addition, models showing progressive neuronal death have been sought but so far unsuccessfully. Pathogenic modeling has been attempted using a range of toxins, as well as through the use of transgenic models of gene defects in familial PD and mutant rodent strains. However, there are still no accepted progressive models of PD that mimic the processes known to occur during cell death and that result in the motor deficits, pathology, biochemistry, and drug responsiveness as seen in humans. Nevertheless, functional models of PD have led to many advances in treating the motor symptoms of the disorder, and we have been fortunate to have them available. They are an important reason the treatment of PD is so much better compared with treatments for related illnesses.