Because little comparative information is available concerning receptor profiles of antiparkinson drugs, affinities of 14 agents were determined at diverse receptors implicated in the etiology and/or treatment of Parkinson's disease: human (h)D 1 , hD 2S , hD 2L , hD 3 , hD 4 , and hD 5 receptors; human 5-hydroxytryptamine (5-HT) 1A , h5-HT 1B , h5-HT 1D , h5-HT 2A , h5-HT 2B , and h5-HT 2C receptors; h␣ 1A -, h␣ 1B -, h␣ 1D -, h␣ 2A -, h␣ 2B -, h␣ 2C -, rat ␣ 2D -, h 1 -, and h 2 -adrenoceptors (ARs); and native histamine 1 receptors. A correlation matrix (294 pK i values) demonstrated substantial "covariance". Correspondingly, principal components analysis revealed that axis 1, which accounted for 76% variance, was associated with the majority of receptor types: drugs displaying overall high versus modest affinities migrated at opposite extremities. Axis 2 (7% of variance) differentiated drugs with high affinity for hD 4 and H 1 receptors versus h␣ 1 -AR subtypes. Five percent of variance was attributable to axis 3, which distinguished drugs with marked affinity for h 1 -and h 2 -ARs versus hD 5 and 5-HT 2A receptors. Hierarchical (cluster) analysis of global homology generated a dendrogram differentiating two major groups possessing low versus high affinity, respectively, for multiple serotonergic and hD 5 receptors. Within the first group, quinpirole, quinerolane, ropinirole, and pramipexole interacted principally with hD 2 , hD 3 , and hD 4 receptors, whereas piribedil and talipexole recognized dopaminergic receptors and h␣ 2 -ARs. Within the second group, lisuride and terguride manifested high affinities for all sites, with roxindole/bromocriptine, cabergoline/pergolide, and 6,7-dihydroxy-N,N-dimethyl-2-ammotetralin (TL99)/apomorphine comprising three additional subclusters of closely related ligands. In conclusion, an innovative multivariate analysis revealed marked heterogeneity in binding profiles of antiparkinson agents. Actions at sites other than hD 2 receptors likely participate in their (contrasting) functional profiles.