The present experiment examined the role of the central nucleus and basolateral complex in the retention of inhibitory avoidance training by reversibly inactivating these regions with lidocaine immediately following training. Male Sprague-Dawley rats were surgically implanted bilaterally with cannulae aimed at the central nucleus or the basolateral complex. One week later, they received one trial inhibitory avoidance training (0.45 mA; 1 s), followed immediately by infusions of lidocaine hydrochloride or buffer (10 micrograms/0.25 microliters). Retention was tested 2 days after training. Immediate posttraining infusions of lidocaine into the central nucleus did not affect retention performance; in contrast, immediate posttraining infusions of lidocaine into the basolateral complex significantly impaired retention performance. In addition, the effect of posttraining infusions of lidocaine into the basolateral complex was time-dependent: infusions administered 6 h after training also impaired memory, but infusions administered 24 h after training had no effect. Immediate posttraining infusions of lidocaine also impaired the retention performance of rats trained with a more intense footshock (0.75 mA). However, at the higher footshock intensity, administration of lidocaine 6 h after training had no effect on retention performance. The time- and footshock-dependent retrograde impairment of memory produced by posttraining reversible inactivation of the basolateral complex suggests that this region of the amygdala is involved in the consolidation of memory for inhibitory avoidance training.