Rats (220-250 g) were bilaterally implanted with cannulae in the amygdala, trained on an inhibitory avoidance response and two weeks later, on a Y-maze discrimination response. Immediately following the training on each task, they were injected intraperitoneally (i.p.) or intra-amygdally. Retention was tested one week after training for each task. Retention of the Y-maze task was assessed by discrimination reversal training. Naloxone administered i.p. (3.0 mg/kg) significantly facilitated retention of both tasks in unoperated control rats as well as in rats implanted bilaterally with amygdala cannulae. The memory-enhancing effect of naloxone i.p. was blocked by propranolol (0.3 or 1.0 microgram) injected in the amygdala, but not when this beta-noradrenergic antagonist was injected (0.3 micrograms) into either the caudate or the cortex dorsal to the amygdala. Further, intra-amygdala injections of the beta 1-adrenoceptor blocker atenolol (0.3 or 1.0 microgram) and the beta 2-adrenoceptor blocker zinterol (0.3 or 1.0 microgram), in doses which were ineffective when administered alone, blocked naloxone-induced (3.0 mg/kg, i.p.) memory facilitation. In contrast, posttraining intra-amygdala administration (1.0 micrograms) of the alpha-antagonists prazosin (alpha 1) or yohimbine (alpha 2) did not attenuate the memory-enhancing effects of systemically administered naloxone. These findings support the view that naloxone-induced enhancement of memory is mediated by the activation of beta- but not alpha-noradrenergic receptors located within the amygdaloid complex.