Presynaptic inhibition in the mammalian central nervous system is associated with a depolarization of the primary afferent terminals 11, 23. Since presynaptic inhibition and primary afferent depolarization (PAD) are both reduced by picrotoxin and bicuculline3, 4, 1~, 14, 1n, 22, it has been suggested that~-aminobutyric acid (GABA) might be the mediator of the inhibitory processlZ, 22. It has also been reported that GABA can increase the excitability of afferent terminals2, 6, 14 such as occurs during PAD triggered by stimulation of afferent nerves 24. Moreover it has been demonstrated that GABA directly depolarizes primary afferents in the isolated amphibian spinal cord 2, 3; however, this sort of evidence is still~ acking in a mammalian species. It is technically difficult to study the direct effects of GABA on primary afferent terminals in a mammalian preparation; an alternative approach is to investigate in vivo the depolarization induced by GABA on dorsal root ganglion cell somata.

In rats and cats, GABA has already been shown to depolarize the somata of dorsal root ganglion cells9, 13 and this depolarization, which is blocked by picrotoxin and bicuculline, is accompanied by an increase of the neural membrane conductancelL In amphibians, the depolarization induced by GABA on primary afferent terminals and dorsal root ganglion cells somata is known to be CI-dependent2, 3, 20 and C1-dependent PAD in the cat cuneate nucleus has recently been reported by Davidson and SimpsonL