Graded dose of the histamine H1 antagonists tripelennamine (0.1-3.0 mg/kg), promethazine (0.1-3.0 mg/kg), diphenhydramine (0.3-10.0 mg/kg) and chlorpheniramine (0.3-1.7 mg/kg) increased rates of nonsuppressed responding under a second-order schedule of food presentation to a maximum beyond which responding was increased less or decreased. In contrast, the H2 antagonist, cimetidine, had no effect or decreased responding at the highest doses studied (56.0-100.0 mg/kg). Intermediate doses of tripelennamine, diphenhydramine and promethazine also increased rates of food-maintained responding that were suppressed by electric shock. Maximal increases in rates of suppressed responding were comparable to those produced by effective doses of chlordiazepoxide (3.0-10.0 mg/kg). Under identical conditions, clozapine (0.1-1.0 mg/kg) increased responding to a lesser extent, and d-amphetamine (0.01-0.3 mg/kg) and cimetidine (3.0-100.0 mg/kg) either did not increase or, at the highest doses, only decreased rates of suppressed responding. Doses of tripelennamine and diphenhydramine that increased rates of nonsuppressed and suppressed responding also maintained self-administration in cocaine-trained squirrel monkeys under a second-order schedule of i.v. injection. Rates and patterns of responding maintained by tripelennamine and diphenhydramine were comparable to those maintained by cocaine and d-amphetamine under identical conditions. The results show that histamine H1 antagonists can have pronounced rate-increasing effects on nonsuppressed and suppressed behavior, and that they can serve as reinforcers in monkeys. These effects occur at doses that probably are greater than those required to saturate H1 sites of action in central nervous system and may not be mediated solely through histaminic mechanisms.