An ideal vaccine should elicit a long lasting immune response against the natural parasite, both at the T- and B-cell level. The immune response should occur in all individuals and be directed against determinants that do not vary in the natural parasite population. A major problem in designing synthetic peptide vaccines is that T cells generally recognize peptide antigens only in association with one or a few of the many variants of major histocompatibility complex (MHC) antigens. During the characterization of epitopes of the malaria parasite Plasmodium falciparum that are recognized by human T cells, we analysed a sequence of the circumsporozoite protein, and found that synthetic peptides corresponding to this sequence are recognized by T cells in association with many different MHC class II molecules, both in mouse and in man. This region of the circumsporozoite protein is invariant in different parasite isolates. Peptides derived from this region should be capable of inducing T-cell responses in individuals of most HLA-DR types, and may represent good candidates for inclusion in an effective anti-malaria peptide vaccine.