Experiments were designed to look at the relationship between infective dose and disease severity using 2 clones of Plasmodium chabaudi that differ in virulence. We asked whether there were dose-severity relationships, whether clone differences in virulence were maintained over a range of doses, and whether disease severity could be accounted for by parasite dynamics. Groups of mice were infected with parasite doses differing by an order of magnitude, ranging from 100 to 1 x 10(8) parasites. Infective dose affected the probability of death, but only with the more virulent clone. Dose also affected morbidity. For both clones, higher doses induced greater anaemia. Larger doses caused greater weight loss, but only for infections with the more virulent clone. Here, for a given dose, mice lost a fixed amount of weight, irrespective of their initial weight. Larger doses induced earlier mortality and morbidity than did lower dose treatments. Finally, dose affected parasite dynamics, with earlier and higher peak parasite densities in larger dose infections. All these effects were small relative to clone differences in disease severity, which were apparent across the range of doses. Dose effects were manifested through the timing and/or magnitude of peak parasite densities, broadly supporting the idea that dose affects disease severity by altering the time the host has to control parasite densities and ameliorate the effects of parasites. We discuss the possible efficacy of intervention strategies aimed at reducing human disease severity by reducing infective parasite dose.