Naloxone and naltrexone were compared neuropharmacologically, especially in several rodent models considered to be predictive of antipsychotic activity, since narcotic antagonists have been reported to be effective in the treatment of schizophrenia. Naltrexone was found to be a potent antagonist of amphetamine-induced aggregate toxicity in mice; this was in marked contrast to naloxone, which enhanced amphetamine's toxicity. Both naloxone and naltrexone exhibited an unusual profile of activity in the apomorphine-induced stereotypy test in mice in that they antagonized the chewing behavior but failed to antagonize the rearing; this is in contrast to the activity of standard neurolepitcs (e.g., haloperidol, clozapine) which antagonize both behaviors induced by apomorphine. Neither of the narcotic antagonists exhibited any activity in several other tests predictive of antipsychotic activity. Naltrexone was a potent inhibitor of spontaneous locomotor activity in mice; naloxone, although much weaker, also antagonized motor activity. Thus, endorphins may play a role in the modulation of exploratory or locomotor activity. The marked difference in activity between the two antagonists in the amphetamine aggregate toxicity study may indicate a significant neuropharmacological difference between these agents; i.e., naltrexone may be much more than merely a more potent naloxone. Furthermore, naltrexone's potent activity as an antagonist of amphetamine-induced aggregate toxicity in mice may be predictive of a better therapeutic effect in schizophrenia than has been observed with naloxone.