The data reviewed indicate that extracts of Curcuma longa exhibit anti-inflammatory activity after parenteral application in standard animal models used for testing anti-inflammatory activity. It turned out that curcumin and the volatile oil are at least in part responsible for this action. It appears that when given orally, curcumin is far less active than after i.p. administration. This may be due to poor absorption, as discussed. Data on histamine-induced ulcers are controversial, and studies on the secretory activity (HCl, pepsinogen) are still lacking. In vitro, curcumin exhibited antispasmodic activity. Since there was a protective effect of extracts of Curcuma longa on the liver and a stimulation of bile secretion in animals, Curcuma longa has been advocated for use in liver disorders. Evidence for an effect on liver disease in humans is not yet available. From the facts that after oral application only traces of curcumin were found in the blood and that, on the other hand, most of the curcumin is excreted via the faeces it may be concluded that curcumin is absorbed poorly by the gastrointestinal tract and/or underlies presystemic transformation. Systemic effects therefore seem to be questionable after oral application except that they occur at very low concentrations of curcumin. This does not exclude a local action in the gastrointestinal tract.