The suggestion that there may be a limit to the number of niches available to helminth species in the intestine of Anguilla anguilla was investigated by examining the frequency distributions of the number of helminth species per eel and the relationships between maximum and mean infracommunity richness and component community richness in 1 locality over 17 years and in 64 localities throughout Ireland and England. The maximum number of species per eel did not exceed 4 in the 1 locality, or 3 in the 64 localities. In both the single and the several localities, the relationship between maximum and mean infracommunity richness and component community richness was curvilinear and best described by a power or polynomial function. This was interpreted to mean that infracommunity richness became increasingly independent of component community richness, and that infracommunities were saturated at values well below the higher level of helminth richness immediately available for colonization i.e. component community richness. It is argued that these findings cannot be explained by supply-side ecology, pool exhaustion or transmission rates, but only by infracommunity processes acting to impose a fixed limit to the number of species in an infracommunity. Most infracommunities are species poor, and the limiting factors will only operate as species richness rises to determine a maximum. Acceptance of a limit to the number of niches available also resolves the apparent inconsistency between the occurrence and importance of interspecific competition and the nature of isolationist communities.