Noxious stimulation-induced sensitization of the central nervous system has been proposed as a key element in the development of subsequent protracted pain. Accordingly, the authors used the formalin model of pain to test the hypothesis that general anesthesia can produce preemptive analgesia and thereby interfere with noxious stimulation-induced central sensitization. Rats received 0.9% or 1.8% halothane, 30% or 75% nitrous oxide (N2O), or 75% N2O plus 0.9% halothane (n = 4 or 5 per group). Control rats (n = 5) received only 100% oxygen. Fifteen minutes after the induction of anesthesia, formalin was injected subcutaneously into a hind paw of each rat, and anesthesia was maintained for 5 more min. Because the behavioral pain response to formalin (i.e., flinching of the injected paw) is biphasic, these treatment groups were anesthetized only during phase 1 (acute phase). Another group (n = 5) received 75% N2O only during phase 2 (delayed phase). Reversibility of the N2O effect was tested by the administration of naloxone before phase 1 or naltrexone during phase 2 (n = 5 per group). Finally, additional rats anesthetized as described above (n = 4 or 5 per group) underwent tail-flick testing during anesthesia. All anesthetics reduced phase 1 pain behavior, but only N2O produced antinociception on tail-flick testing. Thirty percent and 75% N2O, administered during phase 1, suppressed phase 2 flinching 29% and 49%, respectively, whereas nitrous oxide administered after phase 1 did not suppress phase 2 pain behavior. This effect of nitrous oxide was reversed by an opioid antagonist given during phase 1 but not phase 2. Halothane administered during phase 1 had no effect on phase 2 flinching, and it antagonized the effect of 75% N2O. Nitrous oxide induces dose-dependent preemptive analgesia in this model that is reversed partially by naloxone, thus suggesting the involvement of endogenous opioids in this action. In contrast, halothane has no preemptive analgesic properties and even antagonizes the analgesic effect of nitrous oxide. Hence, the hypnotic potency of an anesthetic is a poor indication of its preemptive analgesic potential.