The isolation of ligands for large numbers of proteins is an important goal in proteomics. Whereas peptide libraries are rich sources of protein-binding molecules, native peptides have certain undesirable properties, such as sensitivity to proteases that make them less than ideal for some applications. We report here the construction and characterization of large, chemically diverse combinatorial libraries of peptoids (N-substituted oligoglycines). A protocol for the isolation of specific protein-binding molecules from these libraries is described. These data suggest that peptoid libraries will prove to be inexpensive and convenient sources of protein ligands.