A randomized and blind experimental design was used to study the effects of naloxone and nalorphine on the development of morphine dependence in monkeys. The results suggest: (a) that significant dose-related differences existed for combined numbers of withdrawal symptoms times frequency of occurrences; (b) that naloxone and nalorphine were qualitatively similar; (c) antagonists are more effective as dependence develops; (d) naloxone is approximately 10 times more potent than nalorphine, and (e) vomiting was the only withdrawal sign with which EDsos could be calculated. Dependence on morphine still increases up to 9 months after the commonly accepted 90-day stabilization period as measured by the ED50 for vomiting for naloxone.