The continuous increase in strains of the human malaria parasite Plasmodium falciparum resistant to most front-line antimalarial compounds is reason for grave clinical concern. The search for new drugs led us to investigate a number of membrane active polyene macrolide antibiotics, such as amphotericin B, nystatin, filipin and natamycin. The interaction of these compounds with sterols in bilayer cell membranes can lead to cell damage and ultimately cell lysis. The malaria parasite modifies the host erythrocyte membrane by changing the protein and lipid composition and thus the infected cell could be a selective target for membrane active compounds. We found that erythrocytes infected with the trophozoite stage of P. falciparum were particularly susceptible to lysis by amphotericin B (Fungizone) and, to a lesser extent, nystatin, as determined by ELISA and various microscopy assays. Liposomal amphotericin B (AmBisome) displayed a similar specificity for parasitised erythrocytes, but complete lysis required a longer incubation period. In contrast, filipin and natamycin did not distinguish between normal and parasite-infected erythrocytes, but lysed both at similar concentrations. In addition, when added to ring-stage cultures, the amphotericin B preparations and nystatin produced a marked disruption in parasite morphology in less than 2 h without an accompanying permeabilisation of the infected host cell, suggesting a second plasmodicidal mode of action. The results imply that selected polyene macrolide antibiotics or their derivatives could find application in the treatment of severe malaria caused by of P. falciparum.