A single injection of streptozocin (50 mg/kg, i.p.) led to the development of static and dynamic allodynia in the rat. The two responses were detected, respectively, by application of pressure using von Frey hairs or lightly stroking the hind paw with a cotton bud. Static allodynia was present in the majority of the animals within 10 days following streptozocin. In contrast, dynamic allodynia took almost twice as long to develop and was only present in approximately 60% of rats. Morphine (1-3 mg/kg, s.c.) and amitriptyline (0.25-2.0 mg/kg, p.o.) dose-dependently blocked static allodynia. However, neither of the compounds was effective against dynamic allodynia. In contrast, gabapentin (10-100 mg/kg, p.o.) and the related compound pregabalin (3-30 mg/kg, p.o.) dose-dependently blocked both types of allodynia. However, the corresponding R-enantiomer (10-100 mg/kg, p.o.) of pregabalin, was found to be inactive. The intrathecal administration of gabapentin dose-dependently (1-100 microg/animal) blocked both static and dynamic allodynia. In contrast, administration of similar doses of gabapentin into the hind paw failed to block these responses. It is suggested that in this model of neuropathic pain dynamic allodynia is mediated by A beta-fibres and the static type involves small diameter nociceptive fibres. These data suggest that gabapentin and pregabalin possess a superior antiallodynic profile than morphine and amitriptyline, and may represent a novel class of therapeutic agents for the treatment of neuropathic pain.