The effect of riluzole administration, an antiglutamatergic compound, on the electrophysiological activity of the pallidal complex of 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP)-treated monkeys is compared with those induced by two dosages of levodopa (L-DOPA), the first affording the best clinical alleviation, the second sufficient to induce dyskinesias. Both dosages of L-DOPA reduced sharply the firing frequency of globus pallidus pars internalis (GPi) neurons (respectively, 43.8+/-23.0 and 27.4+/-20.2 vs. 111. 2+/-31.4 Hz), decreased the percentage of bursting cells (respectively, 60.7 and 50.0 vs. 80.3%) and augmented the number of regular cells (respectively, 6.5 and 33.0 vs. 4.8%). Riluzole restored the firing frequency (75.0+/-26.9 Hz) and the firing pattern of the GPi (39.7% bursting, 9.5% regular and 50.8% irregular). These results suggest that the emergence of dyskinesia may well be due to a modification of the neuronal messages transmitted from the GPi to the motor nuclei of the thalamus. Riluzole would represent an interesting alternative to dopamine therapy in Parkinson's disease since it regularizes firing but does not cause dyskinesia.