Brain Research, 191 (1980) 569-571 569 © Elsevier/North-Holland Biomedical Press Locomotor activity elicited by injections of picrotoxin into the ventral tegmental area is attenuated by injections of GABA into the globus pallidus GORDON J. MOGENSON, MICHAEL WU and DOUGLAS L. JONES Department of Physiology, University of Western Ontario, London, Ont. N6A 5C1 (Canada) (Accepted February 6th, 1980) Key words: locomotion -- picrotoxin -- ventral tegmental area -- GABA -- globus pallidus Recent neuropharmacological experiments have implicated the nucleus accum- bens (NAcc) and its mesolimbic dopaminergic afferents in the initiation of locomotor responses. Injections of dopamine, or its precursor L-DOPA, into the nucleus accum- bens increase locomotor activity. Also increased locomotor activity was observed after injecting picrotoxin, a GABA antagonist, into the ventral tegmental area (VTA), the site of mesolimbic dopaminergic neurons 6. Since the microiontophoresis of picrotoxin increases the discharge rate of VTA dopaminergic neurons 10, presumably by blocking GABA inhibitory synapses, it has been suggested that these disinhibited neurons release more dopamine from their axon terminals in the NAcc. Similar experiments have provided evidence that there is a GABAergic projection from the NAcc to the globus pallidus (GP) and have implicated this projection in the initiation of locomotor activity 2,3,9. Injections of picrotoxin into the GP increased locomotor activity whereas injections of GABA into the GP attenuated locomotor activity initiated by injecting dopamine into the NAcc 3. These observations, taken together, suggest that VTA-NAcc dopaminergic pro- jections and NAcc-GP GABAergic projections contribute to the initiation of loco- motor responses. In the present study this suggestion was investigated further by injecting GABA into the GP to see whether locomotor activity initiated by picrotoxin injections into the VTA could be attenuated. The procedures were similar to those used in previous experiments3, 6. In 9 male Wistar rats (225-250 g) stainless-steel guide cannulae made of 23-gauge hypodermic tubing were positioned stereotaxically into the VTA and GP and cemented to the skull. Locomotor activity was measured by recording the number of light beam inter- ruptions in an open field test chamber (69 × 69 × 46 cm). There was a 20 min prein- jection period. The animal was removed from the test chamber, given bilateral injec- tions into the VTA and/or GP and returned to the test chamber for a 30 min post- injection period. Injections were made with a 30-gauge stainless-steel cannula connect- ed by PE-10 polyethylene tubing to a Hamilton microlitre syringe. The injection can- nula extended into the brain 1.25 mm beyond the guide cannulae and the injection