An in-vitro model based on the semi-automated microdilution technique has been developed for selecting compounds that might be used clinically for the reversal of chloroquine resistance. This was used initially to test the susceptibility of Plasmodium falciparum clone W2 to chloroquine (CQ). The model was then employed to investigate the effects of each of four resistance-reversing agents (verapamil, desipramine, chlorpheniramine and promethazine, at 1 microM) on this parasite's susceptibility to CQ, with and without alpha(1)-acid glycoprotein (AGP), at a patho-physiological concentration (1.25 g/litre), in the culture medium. In the absence of AGP, each of the resistance-reversing agents reduced the median inhibitory concentrations of CQ by 82%-97%, from a baseline value of about 94 ng/ml. In the presence of AGP, however, most of the resistance-reversing agents had much less effect. There appears to be competitive interaction between CQ, the resistance-reversing agents and AGP. The binding kinetics between CQ, resistance-reversing agents, AGP and other plasma proteins will clearly need to elucidated if clinically effective resistance-reversing agents are to be selected in vitro.