D-Penicillamine (D-PA), 80-100 mg/kg/day enhanced the early phase of adjuvant arthritis (AA) in rats when it was administered orally for about 4 weeks after the adjuvant injection day, whereas dexamethasone (1 mg/kg/day, p.o.) and chloroquine diphosphate (25 mg/kg/day, p.o.) inhibited AA in the same dosing regimen. On the other hand, subcutaneous injection of sodium aurothiomalate (12.5 mg/kg/day) enhanced AA initially, but inhibited it later. The enhancing effect of D-PA on the early phase of AA was observed also at doses of 50 mg/kg/day and 200 mg/kg/day, but, in the case of 200 mg/kg/day, inhibited the later phase of AA. When the administration of D-PA was started before the adjuvant injection, it showed a tendency to suppress AA on proportion to the dosing period. The effect of D-PA, however, was not observed in the model when the drug administration was started after the establishment of arthritis. The co-administration pyridoxine HCl did not influence the effect of D-PA on AA. A good correlation was not obtained between the inflammatory score and the PPD induced skin reaction, serum metals level and histopathological changes of lymph node in the AA rats treated with D-PA. Thus the effect of D-PA on AA was related to dose, timing and duration of the administration. It was suggested that the enhancing and inhibitory effects of D-PA on AA were not based on vitamin B6, depletion, but rather were caused by inhibition of T1 and T2 lymphocytes which may be regulating this arthritis process.