Parasites primarily affect food web structure through changes to diversity and complexity. However, compared to free-living species, their life-history traits lead to more complex feeding niches and altered motifs.

While the recent inclusion of parasites into food-web studies has highlighted the role of parasites as consumers, there is accumulating evidence that parasites can also serve as prey for predators. Here we investigated empirical patterns of predation on parasites and their relationships with parasite transmission in eight topological food webs representing marine and freshwater ecosystems. Within each food web, we examined links in the typical predator -prey sub web as well as the predator -parasite sub web, i.e. the quadrant of the food web indicating which predators eat parasites. Most predatorparasite links represented ' concomitant predation ' (consumption and death of a parasite along with the prey/host; 58 -72%), followed by ' trophic transmission ' (predator feeds on infected prey and becomes infected; 8 -32%) and predation on free-living parasite life-cycle stages (4 -30%). Parasite life-cycle stages had, on average, between 4.2 and 14.2 predators. Among the food webs, as predator richness increased, the number of links exploited by trophically transmitted parasites increased at about the same rate as did the number of links where these stages serve as prey. On the whole, our analyses suggest that predation on parasites has important consequences for both predators and parasites, and food web structure. Because our analysis is solely based on topological webs, determining the strength of these interactions is a promising avenue for future research.

The parasite communities of goby species (Teleostei, Percomorphi) from the south-western Baltic Sea were investigated from 1997 to 2000 in three different seasons. In total, 30 parasite species were found in the guild of four goby spp. from Dahmeshöved (Lübeck Bight). The component community of Pomatoschistus minutus comprised 22, Pomatoschistus pictus 20, Gobiusculus flavescens 18 and Gobius niger 14 parasite species, whereas Pomatoschistus microps from the Salzhaff (Mecklenburg Bight) harboured 24 species. The digenean Podocotyle atomon (ingested with prey) and Cryptocotyle concavum (active penetration) were the most common parasites. Cryptocotyle lingua and the nematode Hysterothylacium sp. in Gobius niger as well as the specialists Aphalloides timmi and Apatemon gracilis (Digenea) in Pomatoschistus microps were also very abundant. There were large changes in the parasite communities within the years as well as between the seasons of a year; only Gobius niger presented rather homogeneous communities. The ratio of core parasite species in the hosts was at most 28% ( Gobiusculus flavescens) and at least 9% ( Gobius niger). The core species can attain their maximum values at different seasons, which is not only influenced by the parasite but also by the host species. It is concluded that the composition of parasite communities was predominantly determined by the ways of life of the host as well as of the parasite species. Another important factor is the population density of intermediate hosts.

The parasites of ten fish species, including four Gobiidae, three Gasterosteidae, two Syngnathidae, and one Zoarcidae from the Salzhaff region, Northwest Mecklenburg, Baltic Sea, were investigated in 1995 and 1996. As many as 36 parasite species, represented by diverse groups of helminths and protozoans as well as annelids and copepods, are found during 4 seasons in these hosts. By far the most abundant group is represented by digeneans, comprising 15 species, followed by 7 cestodans and 6 nematodes. With regard to component communities, 8 host-parasite combinations are core and secondary species with more than 40% prevalence in which generalists such as the digeneans Podocotyle atomon and Cryptocotyle concavum (in 3 hosts). C. lingua, Diplostomum spathaceum, and Acanthostomum balthicum, and the nematode Hysterothylacium are involved. Also, specialists such as Aphalloides timmi in Pomatoschistus microps as well as Magnibursatus caudofilamentosa and Thersitina gasterostei in Gasterosteus aculeatus attain high levels of prevalence. A comparison of different investigations reveals greater prevalence of allogenic and autogenic parasite species with 3 host cycles in the Rerik-Riff (free Baltic) and higher levels of prevalence of autogenic parasite species with 1 or 2 host cycles in the entire Salzhaff. The component communities of gobies from Dahmeshöved, Lübeck Bight, attain generally lower degrees of prevalence than those of the Salzhaff region. The infracommunities consist mostly of 1-3 parasite species per host specimen; this value is surpassed on occasion in P. microps (maximum 7 species) and in G. aculeatus (maximum 9 species, which may compete for 5 microhabitats in a host specimen). In this context the theory of empty niches propagated by some parasitologists is critically discussed and substituted by the assumption of variable niche widths. The seasonality of the more abundant parasites is either unclear--as in the case of C. concavum--or evident--as in the case of P. atomon, which prevail in early spring and summer, or A. timmi, which dominate in late summer, as do M. caudofilamentosa, which is absent in spring. The main causes of the infestation of fish hosts may be their ages and the availability of parasites due to the presence of intermediate hosts.