Mathematical models of malaria transmission have been used to inform the design of malaria control programs since the mid 20th century, and many of these models have provided useful insights into the complexity of the disease. Among developing countries, however and particularly in sub-Saharan Africa, malaria remains a major cause of morbidity and mortality. One of the main difficulties in controlling the most virulent human malaria parasite, Plasmodium falciparum, is its genetic diversity, which confounds attempts to design an effective vaccine. The population structure of P. falciparum remains poorly understood but plays a key role in determining epidemiological patterns of disease and the development of immunity. We discuss the seminal model of malaria transmission developed by Ross and MacDonald, and the modifications that have been made since to include more realism. We show that age profiles of disease and serological data support a theoretical model in which the parasite population is diverse and structured into several antigenic types and highlight the implications of this structure for controlling malaria. Lastly, we discuss the current sequence data on parasite antigen genes that are important for the aquisition of immunity, and the results of a new analysis of P. falciparum population structure at the genomic level.