The effects of two general anesthetics, nitrous oxide and halothane, and oxygen on mu and kappa opioid receptor subtypes from guinea-pig brain were investigated. mu receptor binding was defined using [3H]dihydromorphine as the ligand. Nitrous oxide (100%) and halothane (2%) decreased the [3H]dihydromorphine binding affinity (Kdair = 0.87 nM, KdN2O = 1.45 nM, Kdhalothane = 2.30 nM) without affecting the density of binding sites. A decrease in the [3H]dihydromorphine binding affinity without influence on the density of binding sites was also observed in the presence of 100% oxygen (KdO2 = 1.40 nM). kappa receptor binding was defined using [3H](-)ethylketocyclazocine as the ligand, in the presence of 100 nM D-ala2-D-leu5-enkephalin and 30 nM morphine. While 100% nitrous oxide caused a slight decrease in [3H](-)ethylketocyclazocine binding affinity (Kdair = 0.24 nM, KdN2O = 0.31 nM) and a substantial decrease in the density of binding sites (Bmaxair = 115 fmol/mg protein, BmaxN2O = 84 fmol/mg protein), halothane dramatically affected both the affinity (Kdhalothane = 0.70 nM) and density (Bmaxhalothane = 38 fmol/mg protein). Oxygen (100%) reduced [3H]dihydromorphine binding affinity. Differential effects of two anesthetics on the same receptor or distinct actions of the same anesthetic on different receptors could indicate the presence of specific targets for anesthetics at the membrane level. Conversely, effects of volatile anesthetics on opioid receptors could reflect a non-specific perturbation of the lipidic and proteinaceous components of the membranes.