Pharmacological blockade of N-methyl-D-aspartate (NMDA) receptors can modulate morphine analgesia in experimental animals and humans. However, this literature is highly inconsistent, with NMDA receptor antagonists variously shown to potentiate, attenuate or produce no effect on morphine analgesic magnitude. A number of factors influencing this modulation have been proposed, but no one has examined such factors simultaneously, and all existing studies in mice were conducted exclusively in male subjects. Thus, the influence of systemic administration of site-specific NMDA receptor antagonists-including dextromethorphan, dextrorphan, MK-801, LY235959, L-701,324, and Ro 25-6981-on morphine analgesia (15-45 mg/kg; 15, 30 and 60 min post-injection) was studied in male and female mice using the 49 degrees C tail-withdrawal test. We found that oral and intraperitoneal dextromethorphan, a low-affinity non-competitive antagonist, dose-dependently potentiated low-dose morphine analgesia but attenuated high-dose morphine analgesia. Dextrorphan and MK-801 were found to potentiate low- but not high-dose morphine analgesia. The competitive glutamate-site antagonist, LY235959, and glycine-site antagonist, L-701,324, potentiated morphine analgesia at all doses. In contrast, the polyamine (NR2B) site antagonist, Ro 25-6981, attenuated morphine analgesia at all doses. Strikingly, the non-competitive antagonists produced no modulation of morphine analgesia whatsoever in female mice, whereas no sex differences were observed using competitive or NR2B antagonists. These findings indicate that NMDA modulation of morphine analgesia is critically influenced by sex, site of antagonism, morphine dose and time after injection. Our data suggest that NMDA antagonism via competitive or glycine site antagonism might result in more reliable clinical effects on morphine analgesia in both sexes.