The analgesic effects of a series of muscarinic agonists were investigated by use of the mouse acetic acid writhing, grid-shock, hot-plate and tail-flick tests. The compounds tested were oxotremorine, pilocarpine, arecoline, aceclidine, RS86 and four 3-3(substituted-1,2,5-thiadiazol-4-yl)-1,2,5,6-tetrahy-dro-1 -methyl pyridines (substituted TZTP), these being propoxy-TZTP, 3-Cl-propylthio-TZTP, xanomeline (hexyloxy-TZTP) and hexylthio-TZTP. These agonists were also assayed for their ability to displace [3H]oxotremorine-M and [3H]pirenz-epine binding and for their functional selectivity at pharmaco-logic M1, M2 and M3 receptors. These compounds all produced dose-dependent antinociceptive effects in all of the mouse analgesia tests. The effects of oxotremorine in the writhing test were fully antagonized by the muscarinic antagonist scopolamine (0.1 mg/kg), but only partially antagonized by methsco-polamine (10 mg/kg) and unaffected by the opioid antagonist naltrexone. 3-Cl-propylthio-TZTP and propoxy-TZTP had virtually no effect at the M1 receptor subtype as measured by the human m1 clone expressed in baby hamster kidney cells or the rabbit vas deferens assay. These compounds, however, were more potent in the analgesia tests than the selective M1 agonists xanomeline and hexylthio-TZTP. These data suggest that muscarinic analgesia is mediated by central muscarinic receptors. However, activity at the M1 receptor subtype is not a requirement for antinociceptive activity.