The role of the serotonergic system in the antinociceptive effect of centrally administered salmon calcitonin (sCT) was studied in rats. The animals were given sCT either intracerebroventricularly (i.c.v.) or intrathecally (i.t.). I.c.v. administration of sCT (2,5 micrograms/rat) to animals depleted in CNS serotonin (5-HT) either by treatment with 25 mg/kg desmethylimipramine (DMI) i.p. plus 100 micrograms/rat 5,7 dihydroxytryptamine (5,7 DHT) i.c.v., ten days before or by 150 mg/kg p-chlorophenylalanine (pCPA) i.p., 72 and 24 h before, still significantly increased the hot-plate latencies comparable to those of non-depleted animals. The same result was obtained when the 5-HT receptors were blocked with metergoline. The i.t. administration of sCT (2 micrograms/rat) to animals with spinal cord 5-HT depleted by treatment with DMI plus 5,7 DHT, i.t., delayed but did not abolish the antinociceptive activity of i.t. injected sCT, which was of the same intensity as in non depleted animals. When 5,7 DHT was administered alone, either i.c.v. or i.t., without protection of the catecholaminergic neurons so that noradrenaline (NA) was greatly reduced, the antinociceptive effect of sCT was completely abolished even when NA had been depleted only in the spinal cord. We conclude that it is the catecholaminergic system, not the serotonergic, that plays a fundamental role in the anti-nociceptive effect of centrally administered sCT.