Plasticity in the nervous system underlies the adaptive, non-dependable relationship between the intensity of a noxious stimulus and the perception of pain. Sensitisation of the nervous system can contribute to the development and maintenance of some long-term pain states and may occur at peripheral, spinal and supraspinal levels. With respect to the latter, complex networks of pathways integrate and project from various brain structures to influence the spinal processing of nociceptive input in a top-down fashion. The rostroventromedial medulla (RVM) is a major endogenous modulatory system implicated in the supraspinal control of spinal sensory information. We determined the extent to which facilitatory and inhibitory neurones in the RVM shape the responses of spinal cord neurones to primary afferent input. Single unit extracellular recordings were made in vivo from L4–L5 deep dorsal horn neurones in anaesthetised rats (0.7% halothane in 66% N₂O and 33% O₂) both before and after the stereotaxic injection of 0.8μl lidocaine into the RVM. The effects of this local anaesthetic on electrical and natural (brush, von Frey filaments and heat) evoked responses in each animal were measured against pre-drug control responses. Experiments were performed on normal rats (n=16) and rats that had undergone spinal-nerve-ligation surgery under 1% halothane anaesthesia 2 weeks previously and who accordingly displayed behavioural signs of mechanical and cold hypersensitivity (n=17). Results from these studies show that in normal and neuropathic rats alike, intra-RVM lidocaine differentially (and significantly) inhibited or facilitated the responses of dorsal horn neurones to the range of stimuli tested (comparison of pre- and post-drug responses using a Student’s paired t test with significance set at p<0.05). In both sets of animals, descending facilitatory influences predominated over inhibitory influences, evidenced by the fact that in the majority of experiments, neuronal responses decreased following the histologically verified injection of lidocaine into the RVM. This inhibition was particularly pronounced for input, C-fibre responses and post-discharge, whilst the reduction in dorsal horn responses to higher threshold mechanical and thermal stimuli (vF 30g, 75g & 45°C, 48°, 50°C, respectively) was significantly greater than the reduction in responses to lower threshold stimuli (vF 5g, 9g, 15g & 35°C, 40°C). Moreover, the proportion of neurones that showed an overall inhibitory action of lidocaine increased in the neuropathic condition; in the normal state, 65% of recorded cells had reduced responses post-lidocaine injection whereas in the neuropathic state this increased to 88%. Thus, results from this study have shown that descending facilitations on dorsal horn neurones dominate in normal rats and support previous data suggesting their increase after nerve injury.