Earlier studies from this laboratory indicated that intracerebroventricular administration of physostigmine and clonidine activated both a spinal descending analgesic and antianalgesic system. It was proposed that the latter was mediated spinally by dynorphin A (1-17), because small intrathecal doses (fmol) of dynorphin A (1-17) antagonized analgesia, while intrathecal administration of naloxone and nor-binaltorphimine (at doses which had no effect on spinal mu and kappa receptors) enhanced analgesia by attenuating the antianalgesic component. In the present studies in mice, using the tail-flick response, intrathecal administration of dynorphin antibody (antiserum to dynorphin) enhanced the analgesic effect of (10 min) physostigmine and clonidine given intraventricularly. Peak effect for the antiserum was at 1 hr. Inhibition of the tail-flick response, induced by DAMGO (Tyr-D-Ala2-Gly-NMePhe4-Gly-ol5, a mu agonist), U50, 488 H (trans-3,4-dichloro-N-methyl-N-[2-(1-pyrrolidinyl)-cyclohexyl]- benzeneacetamide methanesulfonate hydrate, a kappa agonist) and morphine was also enhanced by intrathecal administration of dynorphin antiserum. Thus, a variety of analgesic agonists appear to activate a dynorphin-mediated antianalgesic system. Such a system appears not to be activated by intraventricular administration of beta-endorphin and DPDPE (D-Pen2-D-Pen5-enkephalin, a delta agonist) because neither beta-endorphin- nor DPDPE-induced analgesia was enhanced by intrathecal administration of antiserum. The results of the experiments with the antibody provide further evidence to support the role of dynorphin A (1-17), as a putative endogenous opioid, which mediates an antianalgesic descending system in the spinal cord.