We have previously defined the concept of functional partial agonism as the partial agonist responses recorded in brain slices after administration of full ionotropic glutamate receptor agonists and competitive antagonists at fixed ratios. Functional partial agonism can be established at any level of maximal response, depending on the molar ratio of agonist and antagonist used. Using recombinant human muscarinic acetylcholine receptors (m1 and m5) and the functional assay, receptor selection and amplification technology (R-SAT), we have now shown that co-administration of the full agonist, carbachol, and a competitive antagonist, atropine or pirenzepine, at fixed ratios display functional partial agonism. The levels of apparent intrinsic activity of the functional partial agonist responses were shown to be dependent of the receptor density and G-protein concentration in the same manner as that determined for the true partial muscarinic agonist, 4-[N-(3-chlorophenyl)carbamoyloxy]-2-butynyltrimethylammonium chloride (McN A-343). Thus, functional as well as true partial agonist responses became more efficacious and potent with increasing receptor and G-protein levels. The level of maximal functional partial agonist response, which is dependent on the agonist/antagonist ratio, is predictable from the Waud equation, describing competitive receptor/ligand interactions. In agreement with the relative antagonist potencies of pirenzepine at m1 and m5, a 10:1 ratio of carbachol and pirenzepine produced very low-efficacy functional partial agonism, approaching full antagonism, at m1 but virtually full agonism at the m5 subtype.