In those individuals who live in endemic areas, immunity to malaria is slow to develop and stage-specific. The nature and antigenic specificity of this response, which may involve components of both cell-mediated and humoral immunity, is not well understood. Rodent models provide useful systems to explore the spectrum of host responses that may contribute to resolution of erythrocytic-stage infection or possibly to pathogenesis. Moreover, these models allow identification of plasmodial molecules that can induce different types of host responses. Two different mouse model systems, Plasmodium yoelii yoelii and P. chabaudi adami are presented. These have been selected because resolution of infection by P. yoelii yoelii has been shown to require B cell-dependent mechanisms, while control of acute P. chabaudi adami infection can be achieved by T cell-dependent mechanisms. A monoclonal antibody that provides passive protection to P. yoelii challenge infection has been shown to recognize the cysteine-rich, carboxyl-terminal region of the merozoite surface protein-1. This region, obtained in an appropriate configuration from recombinant Escherichia coli, can induce significant protective immune responses in naive mice. In contrast, cell-mediated immune mechanisms make a major contribution to resolution of asexual-stage P. chabaudi adami infection. An empirical approach using continuous flow electrophoresis has identified several low molecular weight plasmodial proteins that can induce partial protective responses in susceptible hosts. These observations are briefly discussed with respect to human malaria.