The dose-response relationships for morphine analgesia were studied in morphine-tolerant and non-tolerant rats using two pain tests: the tail-flick test which measures the threshold for an escape response, and the formalin test which assesses the behavioral response to continuous pain generated in injured tissue. The effects of prior experience with both pain tests on tolerance were also examined. In the formalin test, effective analgesia was obtained in non-tolerant rats at doses that produce minimal depression of locomotor behavior. Morphine tolerance was produced by 20 daily injections of morphine with increments that reached 16 mg/kg, a dose over the LD100 for barrier sustained Long Evans rats. This dose regimen produced a 1.8-fold increase in the ED50 in the tail-flick test and a 2.7-fold increase in the formalin test. Daily experience of the pain test, as well as the morphine regimen produced a 4.8-fold increase in the ED50 in the tail-flick test but did not affect the potency of morphine in the formalin test. The magnitude of tolerance in the absence of daily behavioral testing is consistent with recent clinical reports that little tolerance occurs after prolonged administration of morphine in cancer patients and that tolerance is not an important consideration in the management of pain.