We examined the effects of ST-91 microinjected into the medullary dorsal horn (MDH) on the ability of monkeys to detect small temperature increases in the noxious heat range. The monkeys were trained to detect temperature changes of 0.4, 0.6 and 1.O degrees C (T2) superimposed on an initial temperature shift to 46 degrees C (T1). Behavioral detection latency and the percentage of correct detections were used as indices of the monkeys perceived intensity of noxious thermal stimulation. ST-91 (1, 3, 10 and 30 [mu]g), an alpha-2 adrenergic agonist, and morphine (5 [mu]g; a dose previously found to be effective in this paradigm), an opiate agonist, were microinjected into the MDH. ST-91 in a dose- and stimulus-dependent fashion, and also morphine produced a decrement in the monkey's ability to detect noxious heat increments. There was no effect on the detection of innocuous coolings or visual stimuli, indicating that effects on the detection of noxious heat are independent of motivational, motoric and attentional factors. Systemic idazoxan (an alpha-2 receptor specific antagonist), but not prazocin (an alpha-1 receptor specific antagonist), naloxone (an opiate antagonist) or saline, significantly attenuated the effects of ST-91 on the detection of all noxious T2s. Morphine's effect were attenuated by both naloxone and idazoxan. These data demonstrate a pharmacologically-specific effect of an alpha-2 agonist on the perceived intensity of noxious heat stimuli at the MDH, the earliest central relay for noxious information. Further, they demonstrate an asymmetrical interaction of opioid and adrenergic pain control systems.