Knowledge of the factors that limit parasite numbers offers hope of improved intervention strategies as well as exposing the selective forces that have shaped parasite life–history strategies. We develop a theoretical framework with which to consider the intra–host regulation of malaria parasite density. We analyse a general model that relates timing and magnitude of peak parasite density to initial dose under three different regulatory processes. The dynamics can be regulated either by top–down processes (upgradable immune regulation), bottom–up processes (fixed immune response and red blood cell (RBC) limitation) or a mixture of the two. We define and estimate the following key parameters: (i) the rate of RBC replenishment; (ii) the rate of destruction of uninfected RBCs; and (iii) the maximum parasite growth rate. Comparing predictions of this model with experimental results for rodent malaria in laboratory mice allowed us to reject functional forms of immune upregulation and/or effects of RBC limitation that were inconsistent with the data. Bottom–up regulation alone was insufficient to account for observed patterns without invoking either localized depletion of RBC density or merozoite interference. By contrast, an immune function upregulated in proportion to either merozoite or infected RBC density was consistent with observed dynamics. An immune response directed solely at merozoites required twice the level of activation of one directed at infected RBCs.