Computational chemistry and molecular modeling procedures allow us to define and compute ad hoc size and shape descriptors on the different prototropic forms assumed by drugs in biotest solutions. Together with experimental data measured on a well-identified target receptor, these descriptors are essential elements for obtaining simple, consistent, comparable, and easily interpretable theoretical quantitative structure-activity relationship (QSAR) models based on the ligand similarity-target receptor complementarity paradigm. In this context, quantitative size and shape affinity/subtype selectivity relationships have been modeled for a large set of very heterogeneous alpha 1a-, alpha 1b-, and alpha 1c-adrenergic receptor antagonists. The linear QSAR models generated have been validated by predicting both binding affinity and selectivity of a test set of noncongeneric antagonists. The satisfactory results obtained highlight both the simplicity and the versatility of the approach presented.