The antinociceptive effect of nicotine administered intracereboventricularly (i.c.v.) or intrathecally (i.t) in several pain models was examined in the present study. We found that i.t. treatment with nicotine (from 5 to 20 g) dose-dependently blocked pain behavior revealed during the second phase, but not during the first phase in the formalin test. In addition, i.c.v. treatment with nicotine (from 0.1 to $10;{mu}g$) dose-dependently attenuated pain behavior revealed during both the first and second phases. In addition to the formalin test, nicotine administered i.c.v. or i.t. attenuated acetic acid-induced writhing response. Furthermore, i.c.v. or i.t. administration of nicotine did not cause licking, scratching and biting responses induced by substance P, glutamate, TNF-${alpha}$(100 pg), IL-$1{eta}$(100 pg) and INF-${gamma}$ (100 pg) injectied i.t. The antinociception induced by supraspinally-administered nicotine appears to be more effective than that resulting from spinally administered nicotine. Our results suggest that nicotine administration induces antinociception by acting on the central nervous system and has differing antinociceptive profiles according to the various pain models.