One of the potential problems of using dextromethorphan as a neuroprotective or anticonvulsant agent is the phencyclidine-like behavioral effects that have been attributed to its major metabolite dextrorphan. Because previous behavioral studies of dextromethorphan have generally failed to consider metabolic conversion to this metabolite, the present studies were conducted to examine the effects of route of administration on dextromethorphan pharmacokinetics and locomotor activity in the rat. The bioavailability of dextromethorphan was 1.3-fold lower and the formation of dextrorphan and other metabolites was 3-fold greater after i.p. injection of 30 mg/kg of dextromethorphan as compared to the s.c. route, indicating substantial effect of first-pass metabolism. Plasma dextromethorphan was correlated with brain dextromethorphan (r = 0.84, P < .001), and the brain/plasma concentration ratio was about 6.5. Plasma-free dextrorphan, but not conjugated dextrorphan, was correlated with brain dextrorphan (r = 0.97, P < .001). Tmax of brain dextrorphan was earlier, and Cmax was higher after i.p. injection of dextromethorphan than s.c. administration (60 min vs. 120 min and 1.0 nmol/g vs. 0.2 nmol/g). Dextromethorphan (60 mg/kg i.p.) increased locomotor activity in the rat 60 min postdose, whereas the same dose of dextromethorphan administered by s.c. injection was without effect. These data demonstrate the route-specific effects on the disposition of dextromethorphan and dextrorphan in rat plasma and brain, as well as the behavioral consequence of the difference.