Clonal analysis of the murine B-cell repertoire has been used to investigate the possible role of tandem repeat sequence epitopes of Plasmodium falciparum in immune evasion. A limiting dilution culture system was used whereby murine spleen cells were stimulated with the B-cell mitogen lipopolysaccharide (LPS) in the presence of 3T3 fibroblast filler cells. One in three B cells were shown to produce clones secreting immunoglobulin measurable by an ELISA. The frequency of antibody forming cell precursors (AFCp) specific for the 3' repeat epitopes of the ring injected erythrocyte surface antigen (RESA) was estimated in non-primed mice and found to be low. However, an accurate frequency determination was not possible using this method since the detection of the few positive cultures was found to depend on the presence of more than one AFCp or its products. Limiting dilution analysis was used to assess the frequency and repertoire of splenic AFCp at various times after immunization with a synthetic peptide of the RESA 3' repeat epitope (8 x 4-mer), presented in various ways. There was no marked increase in LPS-responsive AFCp specific for this antigen at the level of either IgM or IgG secretion. This was in marked contrast to the antibody response in vivo, where moderate IgG antibody titres, normally indicative of a secondary response, were seen in the serum of the same mice used for AFCp assay. This discrepancy between serum titre and AFCp frequency following immunization was not apparent with a non-malarial antigen, keyhole limpet haemocyanin (KLH). It was concluded that the LPS-stimulated limiting dilution culture system was not registering RESA-specific memory AFCp. These results raise the possibility that the malarial antigens are deficient in memory B-cell generation, or that secondary responses to these determinants may arise from a distinct B-cell progenitor which is non-responsive to LPS in vitro.