Cannabinoids have been demonstrated to be effective antinociceptive agents when given intravenously. In order to determine whether spinal antinociception can be achieved while minimizing psychotomimetic properties, the pharmacological activity of delta 9-tetrahydrocannabinol (THC) was evaluated after intrathecal injection in male ICR mice. Although delta 9-THC produced potent antinociception, it also caused hypoactivity, hypothermia, and catalepsy. Intrathecal administration of delta 9-THC in mice which had their spinal cord transected at T12 also produced potent antinociception suggesting a spinal component to the antinociceptive effect. Biodispositional studies of [3H]delta 9-THC demonstrated that brain levels of the drug following intrathecal injection in spinally transected animals were not sufficient to produce the antinociceptive effect. These studies suggest the involvement of a spinal component in the antinociceptive action of the cannabinoids.