Spinal nerve (SN) lesion models of neuropathic pain, including the L5 SN axotomy (SNA) model, cause allodynia and hyperalgesia, but spontaneous foot lifting (SFL) thought to indicate spontaneous pain (e.g. Bennett & Xie, 1988) is rarely reported. The causes of spontaneous pain, an important aspect of human neuropathic pain, are poorly understood. To test the hypothesis that spontaneous activity (SA) in dorsal root ganglion (DRG) neurones may cause SFL in a modified SNA model (mSNA) that shows significant SFL (Koutiskou & Lawson, 2002), we examined SFL and SA in both this (mSNA) and SNA models. The mSNA model involves SNA plus loose-ligation of the adjacent L4 SN with chromic gut which causes inflammation. For electrophysiology, female rats (150-180g) were anaesthetised with pentobarbitone (60 mg kg^-1, i.p.), neuromuscularly blocked with pancuronium (0.5 mg kg^-1, i.a.) and artificially ventilated. The end-tidal CO₂ was monitored and maintained between 3-4% by adjusting the volume and stroke of the respiratory pump and blood pressure was monitored and maintained at >80 mmHg. Axotomy was carried out under deep anaesthesia (sodium pentobarbitone 50 mg kg^-1, i.p.). Intracellular recordings of somatic action potentials evoked by dorsal root stimulation and any spontaneously occurring spikes were made from: a) normal L4/L5 DRG neurones in control rats (n=48), b) axotomised L5 neurones in mSNA rats (n=18), c) adjacent L4 neurones in both mSNA and SNA rats (n=10) and d) L4/L5 neurones in rats (n=38) 4 days after hindlimb inflammation induced by intradermal injections of Complete Freunds Adjuvant (100 μl). Neurones were classified from their dorsal root conduction velocities as C, Aδ or Aα/Β units and according to their responses to natural stimuli as non-nociceptive or nociceptive-type units. One week after SN injury, significantly increased percentages of nociceptive-type C-fibre neurones showed SA (from 7% to about 35%) in L4 but not L5 DRGs in both SNA and mSNA rats (Fisher’s test, P<0.01). Thus this increased percentage of SA could not have been the cause of SFL seen only in mSNA rats. It may result from inflammation (perhaps from degeneration of L5 fibres) since a similar increase in percentages of C-nociceptors with SA was seen 4 days after hindlimb inflammation (Fisher's test, P<0.05). Interestingly, L4 nociceptive-type C-fibre neurones showed faster SA rates in mSNA (1.8 Hz) than SNA (0.02 Hz) rats (Mann Whitney test, P<0.001) implicating the L4 in SFL. Thus mSNA rats had both greater SA rates and SFL (Fisher's test, P<0.001). We propose that faster firing rates in intact C-nociceptive neurones are required for SFL/spontaneous neuropathic pain and that these result from added inflammation/damage, caused by chromic gut in mSNA rats.