IFN-gamma production is prominent in some models of autoimmune disease, including adjuvant arthritis (AA), but the role of IFN-gamma in the pathogenesis of these diseases is uncertain. Experimental manipulation (administration of cytokine, blocking cytokine action with specific antibody, disruption of genes encoding the cytokine or its receptor) has revealed both pro- and anti-inflammatory effects, depending on the nature of the manipulation and the timing of the treatment. We examined separately the effects of cytokine blockade during the afferent and efferent phases of AA in Dark Agouti rats, using an adoptive transfer system. Effects of IFN-gamma on the efferent phase were investigated by treating recipients with mAb DB-1, which blocks the activity of rat IFN-gamma. When treatment was commenced before cell transfer, the resulting polyarthritis was more severe than in controls treated with normal IgG. Commencing treatment after the adoptively transferred disease had become established caused neither amelioration nor exacerbation, but did cause some delay in resolution. In contrast, treatment of donors did not appear to affect the generation of arthritogenic cells. The main effect of IFN-gamma appears to be modulation of the arthritogenicity of the migratory effector T cells that can transfer AA.