We studied the 'crowding effect' in Tribolium confusum infected with Hymenolepis diminuta. Manipulations included age and number of parasites, and diet, sex, age and number of exposures of hosts. Volume per parasite was unaffected until an intensity of at least 5-10 parasites per host, then declined approximately inversely as intensities increased. Parasite size was affected by host sex but not age or reproductive status. Host diet affected parasite size and the impact of crowding. Daily gain in parasite volume peaked partway through the developmental period and preceded the first evidence of a crowding effect. Parasites that established during a second exposure had a transient developmental delay but eventually grew as large or larger than parasites from a single exposure with the same total intensity. Parasites responded to crowding by differential allocation of resources. Cercomer volume decreased even with slight crowding, the capsule surrounding the scolex was not reduced until crowding became more severe, and scolex width was reduced only in the most extreme conditions. The data support the hypothesis that the crowding effect in this system is driven primarily by nutrient, rather than space limitations.