The presence or absence of parasites within host populations is the result of a complex of factors, both biotic and abiotic. This study uses a non-parametric classification tree approach to evaluate the relative importance of key abiotic and biotic drivers controlling the presence/absence of parasites with complex life cycles in a sentinel, the common killifish Fundulus heteroclitus. Parasite communities were classified from 480 individuals representing 15 fish from 4 distinct marsh sites in each of 4 consecutive seasons between 2006 and 2007. Abiotic parameters were recorded at continuous water monitoring stations located at each of the 4 sites. Classification trees identified the presence of benthic invertebrate species (Gammarus sp. and Littorina sp.) as the most important variables in determining parasite presence: secondary splitters were dominated by abiotic variables including conductance, pH and temperature. Seventy percent of hosts were successfully classified into the correct category (infected/uninfected) based on only these criteria. The presence of competent definitive hosts was not considered to be an important explanatory variable. These data suggest that the most important determinant of the presence of these parasite populations in the common killifish is the availability of diverse communities of benthic invertebrates.

Although the normal intermediate hosts of Hymenolepis microstoma are insects, oncospheres invade the intestinal tissue of the definitive hosts, mice, within 30 min after oral egg inoculation.

Infection parameters of Posthodiplostomum cuticola, a digenean parasite with a complex life-cycle, were investigated in fish (the second intermediate host) from 6 floodplain water bodies over 2 years. A broad range of factors related to abiotic characteristics of localities, density of the first intermediate (planorbid snails) and definitive (wading birds) hosts and fish community structure were tested for their effects on P. cuticola infection in juvenile and adult fish. Characters of the littoral zone and flood duration were found to be important factors for the presence of the first intermediate and definitive hosts. Visitation time of definitive bird hosts was also related to adult fish host density. Localities with P. cuticola infected fish were visited by a higher number of bird species. Infection of P. cuticola in fish and similarities in infection among fish host assemblages were correlated with fish host density and fish species composition. Parasite infection in both adult and juvenile fishes was associated with the slope of the bank and the bottom type, in particular in juvenile fish assemblages with snail host density. We conclude that habitat characteristics, snail host density and fish community structure contribute significantly to P. cuticola infection in fish hosts.