In vivo effects of fixed-proportion agonist/antagonist mixtures in a warm water tail-withdrawal assay of thermal nociception in mice. Mice were injected with varying doses of fentanyl or naltrexone alone (A); CP55,940 or rimonabant alone (C); or the indicated fixed-proportion mixtures of fentanyl + naltrexone (B) or CP55,940 + rimonabant (D). The x-axes in (A and C) show dose of either the agonist or the antagonist administered alone. The x-axes of (B and D) show the dose of the agonist only, and the antagonist dose can be calculated as agonist dose ÷ fixed-proportion dose ratio of agonist to antagonist. Fixed-proportion dose ratios are expressed as X:1 agonist/antagonist, which are fentanyl/naltrexone (F/N) and CP55,940/rimonabant (C/R). Data represent %MPE with a 10-second cutoff in 56°C water. All values are means ± S.E.M. (n = 8–10). Fixed-proportion mixtures generally showed increasing antinociception as a function of increasing agonist proportion in the mixtures.