1. The pharmacological characteristics of muscarinic receptors in the rabbit iris sphincter muscle were studied and compared to M3 receptors in rabbit urinary bladder smooth muscle. 2. (+/-)-Cis-dioxolane induced concentration-dependent contractions of the iris sphincter muscle (pEC50 = 6.41+/-0.10, Emax = 181+/-17 mg, n = 38) and urinary bladder smooth muscle (pEC50 = 6.97+/-0.04, Emax = 4.28+/-0.25 g, n = 54). These contractions were competitively antagonized by a range of muscarinic receptor antagonists (pK(B) values are given for the iris sphincter muscle and the bladder smooth muscle, respectively): atropine (9.30+/-0.07 and 9.40+/-0.04), AQ-RA 741 (6.35+/-0.04 and 6.88+/-0.03), darifenacin (9.56+/-0.05 and 9.12+/-0.05), methoctramine (5.75+/-0.07 and 5.81+/-0.06), oxybutynin (8.10+/-0.09 and 8.59+/-0.06), pirenzepine (6.79+/-0.05 and 6.89+/-0.04), secoverine (7.54+/-0.05 and 7.66+/-0.05), p-F-HHSiD (7.55+/-0.09 and 7.50+/-0.05) and zamifenacin (8.69+/-0.10 and 8.36+/-0.06). A significant correlation between the pK(B) values in the bladder and the pK(B) values in the iris was obtained. 3. In both tissues, the pK(B) values correlated most favorably with pKi values for these compounds at human recombinant muscarinic m3 receptors. A reasonable correlation was also noted at human recombinant muscarinic m5 receptors given the poor discriminative ability of ligands between m3 and m5 receptors. 4. Overall, the data from this study suggest that the muscarinic receptors mediating contraction of the rabbit iris sphincter muscle and urinary bladder smooth muscle are similar and equate most closely with the pharmacologically-defined muscarinic M3 receptor.