The relative affinity and selectivity of the stereoisomers of 3-quinuclidinyl atrolactate (QNA) and the enantiomers of 3-quinuclidinyl xanthene-9-carboxylate (QNX) for the pharmacologically defined muscarinic receptor subtypes was determined using functional responses of rabbit vas deferens (M1), guinea pig atria (M2) and bladder detrusor muscle (M3). All the stereoisomers behaved as competitive antagonists yielding the same rank order of potency at each receptor subtype: (RR)-QNA greater than (RS)-QNA greater than (SR)-QNA greater than (SS)-QNA and (R)-QNX greater than (S)-QNX. Moreover, the eudismic ratios relative to (RR)-QNA for (RS)-, (SR)- and (SS)-QNA, respectively, ranged from 4 to 308 at all three subtypes. Stereoselective effects were also observed for QNX; (S)-QNX/(R)-QNX ratios ranged from 76 to 248. In contrast, there was a distinct lack of receptor selectivity among the isomers of QNA and QNX for either the M1, M2 or M3 muscarinic receptor subtypes. Stereoselective effects were also evident in vivo in the guinea pig cystometrogram, where the rank order of potency of the isomers of QNA and QNX was similar to that observed in vitro. (RR)-QNA and (R)-QNX equipotently depressed intravesical bladder pressure (PvesP) (ID50 = 0.06 mg/kg i.v.). Other parameters (bladder capacity, threshold pressure) were unaltered by the stereoisomers. The data demonstrate that despite the high affinity of the eutomers of QNA and QNX for muscarinic receptors, they discriminate poorly among muscarinic subpopulations, thus limiting their utility to subclassify muscarinic receptors.