Dopamine, a retinal neurotransmitter, is known to affect electrical measures of retinal pigment epithelial (RPE) function: the standing potential and the DC ERG. To locate the origin of these effects, studies were performed on in vitro preparations of chick retina-RPE-choroid, which were separately perfused on the retinal and choroidal tissue surfaces. Dopamine (250 micrograms) in the retinal bath depolarized the RPE basal membrane, decreased the apparent basal membrane resistance (Rba) and increased the ERG c-wave. At concentrations less than or equal to 100 microM, retinal dopamine often caused a transient basal membrane hyperpolarization, accompanied by an apparent increase in Rba and decrease in c-wave. Surprisingly, 20-100 microM choroidal dopamine induced similar changes in basal membrane potential, resistance and c-wave amplitude, and the transient hyperpolarization and increase in Rba were often more pronounced than at comparable concentrations of retinal dopamine. Experiments in RPE-choroid preparations suggested that the effects of retinal dopamine were not secondary to effects on the neural retina. The effects of retinal and choroidal dopamine in the same tissue often were distinct, suggesting separate receptor populations on the apical and basolateral membranes of the RPE. The c-wave changes could be explained by the changes in Rba, and not by an effect on the light-evoked decrease in subretinal [K+]0. Choroidal perfusion with 50 microM 4,4'-diisothiocyanostilbene-2,2'-disulfonate (DIDS), which appears to block a Cl- conductance in chick RPE, blocked the effects of dopamine perfusion on either side of the tissue. These results suggest that perfusion with either retinal or choroidal dopamine leads to electrical effects on the RPE basal membrane, possibly via a second-messenger system affecting a basal membrane Cl- conductance. Dopamine could suppress the "light-peak" depolarization of the RPE basal membrane. When either retinal or choroidal dopamine induced a large net change in trans-tissue potential (originating as a change in basal membrane potential), the light peak was severely depressed, while smaller changes produced correspondingly smaller decreases in light-peak amplitude. We found, however, that light-peak amplitude was not significantly reduced when there was little net change in the trans-tissue potential, even though dopamine may have produced sizable transient effects. Thus, despite apparent occupation of dopamine receptors on the RPE, the light peak persisted under these conditions. Similar relations between light-peak amplitude and net change in trans-tissue potential have been observed for a variety of different conditions, suggesting that the effect of dopamine on the light peak is nonspecific.