The antinociceptive effects of the opioid agonists etonitazene and alfentanil, as well as the agonist/antagonists nalbuphine, [(1)-beta-2'-hydroxy-2,9-dimethyl-5-phenyl-6,7-benzomorphan (GPA 1657)] and profadol were studied in the warm water (48 degrees and 55 degrees C) tail-withdrawal assay in rhesus monkeys. Etonitazene and alfentanil produced dose-dependent increases in tail-withdrawal latency up to the maximum possible latency of 20 sec in 48 degrees and 55 degrees C water. Nalbuphine, GPA 1657 and profadol produced the maximum possible effect only at 48 degrees C, and were ineffective at 55 degrees C. The opioid antagonist quadazocine produced a dose-dependent antagonism of all agonists except profadol. In a Schild plot analysis, apparent pA2 values for quadazocine with alfentanil, etonitazene and nalbuphine were homogeneous (7.3-7.7 mol/kg), suggesting their effects were probably mediated by mu opioid receptors. The apparent pA2 value for GPA 1657 was significantly lower (6.2 mol/kg), suggesting GPA 1657 may have produced antinociception by a non mu receptor-mediated mechanism. The selective delta antagonist naltrindole (0.32-1.0 mg/kg) antagonized the antinociceptive effect of GPA 1657. The kappa-selective antagonist nor-binaltorphimine (nor-BNI, 3.2 mg/kg) caused a small rightward shift in the GPA 1657 dose-effect curve. Nalbuphine, GPA 1657 or profadol produced a rightward shift in the alfentanil dose-effect curve in 55 degrees C water, consistent with possible low-efficacy mu agonist effects of these compounds. These studies suggest agonists may be differentiated based on antinociceptive effectiveness, receptor selectivity and intrinsic efficacy in the rhesus monkey tail-withdrawal procedure.