The basal ganglia have been implicated in a number of important motor functions, in particular in the initiation of motor responses. According to the current model of basal ganglia functions, motor initiation is supposed to be associated with an inhibition of basal ganglia output structures (substantia nigra pars reticulata/entopeduncular nucleus) which, in turn, might be brought about by corresponding striatal activity changes conveyed via direct and indirect intrinsic pathways to the substantia nigra pars reticulata and the entopeduncular nucleus. Rodent studies using neuropharmacological manipulations of basal ganglia transmitter systems by neurotoxins or drugs provide converging evidence that dopamine within the caudate-putamen, but also within extrastriatal basal ganglia nuclei, is involved in motor initiation by modulating the activity of direct and indirect intrinsic pathways. However, the striatal segregation of dopamine D1 and D2 receptors in control of the direct and indirect projection neurons seems not to be maintained throughout the basal ganglia. In dopamine intact animals, striatal glutamate plays a major role in response initiation probably through actions on striatopallidal neurons involving N-methyl-D-aspartate, but not alpha-amino-3-hydroxy-5-methylisoxazole-4-propionic acid (AMPA) receptors. Striatal adenosine might also contribute to movement initiation by acting on adenosine A2A receptors located on striatopallidal neurons. Analysis of two integral parts of the indirect pathway revealed that inactivation of the subthalamic nucleus was found to facilitate response initiation, while inactivation of the globus pallidus resulted in facilitation as well as inhibition of response initiation indicating a complex contribution of this latter nucleus. Glutamate and gamma-amino-butyric acid (GABA) controlling the activity of the substantia nigra pars reticulata could be involved in control of response initiation in a way predicted by the simplified model of basal ganglia functions. In contrast, the role of the entopeduncular nucleus in response initiation and its control through GABA and glutamate is at variance with this hypothesis, suggesting functional differences of the output structures. Taken together, neurochemical systems of the basal ganglia significantly contribute to intact response initiation by mechanisms which are only partly consistent with predictions of the current functional scheme of the basal ganglia. This suggests that a more complex model is required to account for these disparate findings.