Rhesus monkeys undergoing opioid withdrawal either due to withholding morphine administration for 14 h or due to administration of naloxone, were treated with or due to administration of naloxone, were treated with either morphine or clonidine. Morphine eliminated all of the withdrawal signs that developed when morphine was withheld for 14 h. Clonidine also eliminated some but not all signs that developed when morphine was withheld. The frequencies of individual signs prior to drug administration were directly related to the minimal doses necessary to eliminate signs for morphine but not for clonidine. Morphine also eliminated most of the signs precipitated by naloxone, whereas clonidine did not eliminate as many of the naloxone-precipitated signs. Additionally, some of the naloxone-precipitated signs that were eliminated by clonidine were not eliminated by morphine. The present results are consistent with clinical findings indicating an efficacy of clonidine in the treatment of opioid withdrawal through a non-opioid mechanism.