1. To investigate postsynaptic potentials (PSPs), we made intracellular recordings from neurons of the amygdaloid central nucleus in slices from the guinea pig and rat brains maintained in vitro. The results from guinea pigs and rats were very similar. 2. In the presence of bicuculline (20-mu-M), focal electrical stimulation of the amygdaloid basal nucleus with low intensities elicited short-latency excitatory PSPs (EPSPs) followed by long-latency EPSPs. The short-latency EPSP was selectively blocked by 6-cyano-7-nitroquinoxaline-2,3-dion (CNQX; 10-20-mu-M). The long-latency EPSP was preferentially abolished by D,L-2-amino-5-phosphonovaleric acid (D,L-APV; 40-mu-M) and was augmented by removal of extracellular MG2+. The compound EPSP reversed at -4 mV, which was close to -1 mV, the reversal potential for pressure-ejected glutamate (Glu). 3. When the intensity of the focal stimulation was increased in the presence of bicuculline (20-mu-M), CNQX (20-mu-M), and D,L-APV (50-mu-M), a second EPSP with a short latency and a prolonged duration could be evoked in approximately 65% of the neurons. The EPSPs were reversibly blocked by d-tubocurarine (50-mu-M) or hexamethonium (200-mu-M) but were unaffected by atropine (1-mu-M) or a 5-hydroxytryptamine type 3 receptor antagonist, ICS-205930 (5-10-mu-M). In these neurons, acetylcholine (ACh; 1-3 mM) caused a depolarization, associated with a decreased input resistance. 4. In the presence of CNQX (20-mu-M) and D,L-APV (50-mu-M), single focal stimulation of the dorsolateral subdivision in the central nucleus with low intensities elicited a depolarizing inhibitory PSP (IPSP). The IPSP was reversibly abolished by bicuculline (20-40-mu-M). The reversal potential (-63 mV) for the IPSP was similar to the reversal potential (-61 mV) for the response to gamma-aminobutyric acid (GABA) applied by pressure ejection. 5. In the presence of bicuculline (20-40-mu-M) and CNQX (20-mu-M), a repetitive focal stimulus with high intensities delivered to the dorsolateral subdivision produced a hyperpolarizing PSP followed by a slow depolarization in most neurons. Of putative inhibitory amino acid transmitters, glycine (Gly; 3 mM) produced only a hyperpolarization, associated with a decrease in input resistance. Strychnine (1-2-mu-M) reversibly blocked both the Gly hyperpolarization and the synaptically evoked hyperpolarization. The reversal potential of -81 mV for the hyperpolarizing PSP was close to -82 mV for the Gly hyperpolarization. The reversal potential for the Gly response was shifted to less negative values by increasing the external K+ concentration or decreasing the extracellular Cl- concentration. The slope of the reversal potential was 40 mV per log unit of the external K+ concentration and 20 mV per log unit of the extracellular Cl- concentration. 6. When the strength of repetitive stimuli was further increased in the presence of bicuculline (20-mu-M), strychnine (2-mu-M), and CNQX (20-mu-M), a second hyperpolarizing PSP could be elicited in most cells. The amplitude of hyperpolarizing response was reversibly reduced by phaclofen (0.5 mM) or baclofen (0.3-0.5-mu-M). The reversal potential of -92 mV for the hyperpolarizing PSP was the same as -92 mV for the response to baclofen (3-10-mu-M). 7. Adenosine, ADP, or ATP at 0.1 mM markedly depressed the Glu-mediated EPSP (in the presence of bicuculline) and the GABA-mediated IPSP (in the presence of CNQX and D,L-APV), whereas these substances did not significantly affect responses to Glu or GABA. The order of potencies of adenosine derivatives for depressing the EPSP and the IPSP was adenosine > ADP > ATP. Phaclofen (0.5 mM) augmented but baclofen (1-mu-M) depressed the Glu-mediated EPSP, whereas these substances did not significantly affect the Glu response. 8. The results provide evidence for the existence of three distinct EPSPs mediated by non-N-methyl-D-aspartate (non-NMDA) excitatory amino acid receptors, NMDA receptors, and nicotinic receptors, respectively, and of three different IPSPs mediated by GABA(A) receptors, GABA(B) receptors, and Gly receptors, respectively, in central amygdala neurons. In addition, adenosine depressed the Glu-mediated EPSPs and the GABA(A)-mediated IPSPs by activation of presynaptic adenosine A1 receptors, and baclofen suppressed the Glu-mediated EPSPs by activation of presynaptic GABA(B) receptors.