1. Intracellular microelectrode recording techniques were performed on mouse spinal cord and cerebral hemisphere neurones grown in primary dissociated cell culture. The effects of several anxiolytics applied by local pressure ejection on responses to gamma-aminobutyric acid (GABA) evoked by iontophoresis were investigated. Responses to GABA were depolarizing since intracellular chloride ion concentration was increased by injection from potassium chloride (3M)-filled recording micropipettes and neurones were held at large negative membrane potentials (-70 to -90 mV). The agents studied were six 'non-sedative anxiolytics', CL 218,872 (3-methyl-6-(3-trifluoromethyl-phenyl)1,2,4-triazolo(4,3-b) pyridazine), PK 8165 (2-phenyl-4-(2-(4-piperidinyl)ethyl)-quinoline), PK 9084 (2-phenyl-4-(2-(3-piperidinyl)ethyl)-quinoline), CGS 9896 (2-(4-chlorophenyl)-2,5-dihydropyrazolo(4,3-c)quinoline-3(3H)-one) , ZK 91296 (ethyl 5-benzyloxy-4-methoxymethyl-beta-carboline-3-beta-carboxylate), buspirone (8-4-[4-(2-pyrimidinyl)-1-piperazinyl]butyl-8-azaspiro[4.5]decane- 7,9- dione), and two sedative anxiolytics, diazepam and zopiclone [( 6-(5-chloro-2-pyridyl)-6,7-dihydro-7-oxo-5H-pyrrolo[3,4-b]pyrazin- 5- yl]4-methyl-1-piperazinecarboxylate). 2. Direct effects on responses to GABA were studied for all drugs applied in varying concentrations. For the drugs which significantly altered responses to GABA, the effects of the benzodiazepine receptor antagonists Ro 15-1788 (ethyl-8-fluoro-5,6-dihydro-5-methyl-6-oxo-4H-imidazo(1,5a)-(1,4)benzodi azepine - 3-carboxylate) and CGS 8216 (2-phenylpyrazolo(4,3-c)-quinolin-3(5H)-one) were evaluated. For the drugs devoid of significant direct effect on responses to GABA, the influence on diazepam-induced enhancement of responses to GABA was evaluated. 3. Diazepam, zopiclone and CL 218,872 concentration-dependently and reversibly enhanced responses to GABA. Maximal enhancement was 82% for diazepam (500 nM), 64% for zopiclone (10 microM) and 20% for CL 218,872 (10 microM). PK 8165 effects varied with concentration, enhancing responses to GABA (up to 18%) at nM concentrations and reducing responses to GABA (up to 90%) at microM concentrations. CGS 9896, ZK 9126, PK 9084 and buspirone, in concentrations ranging from 1 nM to 10 microM, lacked significant direct effects on responses to GABA. 4. The enhancing effects of diazepam, zopiclone, CL 218,872 and PK 8165 were antagonized by Ro 15-1788. However, the reducing effect on responses to GABA of PK 8165 at microM concentrations was not antagonized by CGS 8216. CGS 9896 and ZK 91296 concentration-dependently blocked the diazepam-induced enhancement of responses to GABA. However, PK 9084 and buspirone did not antagonize the diazepam-induced enhancement of responses to GABA. 5. These results indicate that diazepam and zopiclone may be full agonists, CL 218,872 and PK 8165 are partial agonists, and CGS 9896 and ZK 91296 are pure antagonists at benzodiazepine receptors. On the other hand, PK 9084 and buspirone do not interact with benzodiazepine receptors.