In Parkinson's disease the neurones of the subthalamic nucleus show increased synchrony and oscillatory burst discharge, thought to reflect a breakdown of parallel processing in basal ganglia circuitry. To understand better the mechanisms underlying this transition, we sought to mimic this change in firing pattern within sagittal slices of rat midbrain. The firing patterns of up to four simultaneously extracellularly recorded subthalamic nucleus (STN) neurones were analysed using burst and oscillation detection programs, and correlated activity between pairs of neurones assessed. In control conditions all but 11 of 488 (2%) neurones fired in a predominantly tonic pattern (with mean oscillation frequency >3 Hz), with no significantly cross-correlated activity in any of 393 pairs of neurones. The glutamate antagonists DL-2-amino-phosphonopentanoic acid (APV), 6-cyano-7-nitroquinoxaline-2,3-dione (CNQX) and 6-methyl-2-(phenylethynyl)pyridine (MPEP) did not change the firing rate or pattern of these cells, providing no evidence for a role of glutamatergic collaterals within the STN under these conditions. The GABA(A) receptor antagonist bicuculline and GABA(B) receptor antagonist (2S)-3-[[(1S)-1-(3,4-dichlorophenyl)ethyl]amino-2-hydroxypropyl]phenylmethyl phosphinic acid (CGP 55845) were also without effect on firing rate or pattern in these cells, suggesting that there was no active input from other GABAergic basal ganglia nuclei in this slice. The dopamine receptor antagonist haloperidol caused no significant change to firing rate or pattern of firing in these cells, suggesting that there was no active dopaminergic input in this slice. Excitations of STN neurones by muscarine, (+)-1-aminocyclopentane-trans-1,3-dicarboxylic acid (ACPD), N-methyl-D-aspartic acid (NMDA) or dopamine were all unaccompanied by a change in firing pattern or any significant correlated activity between STN neurone pairs. Burst firing could be induced in STN neurones with either the potassium channel blocker tetraethylammonium (TEA; 10 mM; in 100/138 [72%] of cells) or with a combination of NMDA and the calcium-activated potassium channel blocker apamin (in 101/216 [47%] of cells). Burst firing in TEA was unchanged by CNOX and APV, MPEP, CGP55845, haloperidol, dopamine, and ACPD, although muscarine produced a significant increase in oscillation frequency. Burst firing in NMDA and apamin was unchanged by CNQX and APV, dopamine, muscarine and ACPD, although bicuculline caused a significant increase in oscillation frequency. Such burst firing was not accompanied by synchrony in any condition, either alone, or during application of excitatory agents or glutamate or GABA antagonists. As the bursting seen here was unaccompanied by the synchronous activity that has often been observed (pathologically) in vivo, it probably reflects solely intrinsic STN neuronal properties, rather than network activity. No functional role was found for glutamatergic collaterals within the STN, either when cells are firing tonically or burst firing. The circuitry needed to produce synchrony in the STN is most likely not intrinsic to the STN itself, but requires connections with other basal ganglia nuclei, and/or the cortex, which are not present in this preparation.