of parasites in hosts under environmental stress: experiments with anoxia and starvation. Á/ Oikos 108: 156 Á/164.Most environments periodically impose severe stress that may cause high mortality and alter population structure, for example, by removing sick and old individuals. We examined how anoxic conditions and starvation of the host affect virulence of two closely related trematode parasites, Rhipidocotyle campanula and R. fennica . These parasites differ by prevalence of infection and by exploitation rate of individual hosts (freshwater clam, Anodonta piscinalis ). Infection by R. campanula is rare (B/5% prevalence of infection) and destroys on average 90% of the gonad tissue of the individual host. Infection by R. fennica is more common (20 Á/60% prevalence of infection) and leads to on average 30% gonad destruction. In the end, both infections lead to host infertility. We predicted that R. campanula induces higher host mortality than R. fennica under host stress. In two laboratory experiments, we exposed naturallyinfected and uninfected clams to anoxia and to starvation. Anoxia occasionally takes place during winter in eutrophic lakes, while some degree of starvation should occur seasonally. We found that mortality rate of clams was much higher under anoxia than under starvation, and that infection increased mortality rate under both types of host stress. As predicted, R. campanula induced higher host mortality than R. fennica . Host survival was population-specific, suggesting that clams of different origins carried different amount of energy reserves. Severe environmental perturbation may remove R. campanula infected individuals from the host population, but recolonization from the fish host is likely to prevent extinction of the parasite suprapopulation. The observed high host mortality induced by R. campanula may be one ecological explanation for the consistently lower prevalence of infection of R. campanula when compared to R. fennica .

Introduced species are recognized as a major threat to native species. One factor that facilitates their spread is that they are often resistant to natural enemies of their native competitor. Negative effects of the invaders are often documented, but invaders may also convey benefits to the natives if they interfere with the native hostparasite interactions. If invaders act as resistant targets for the native parasites, they may reduce the density of the infectious transmission stages ('dilution effect') and decrease the risk of infection for the natives. We tested this hypothesis by exposing coexisting native and introduced freshwater snails to infectious stages of a native parasite. The native hosts showed a significantly reduced infection rate when exposed together with the resistant invader. A significant amount of the parasite transmission stages was wasted on the resistant invader leading to lower risk of infection for the natives. Our results show that invaders may convey benefits to the native competitors by perturbing the native host-parasite interaction and support the idea that the 'dilution effect' could be important for invaded communities.

Abstract. Genetic compatibility, nonspecific defenses, and environmental effects determine parasite resistance. Host mating system (selfing vs. outcrossing) should be important for parasite resistance because it determines the segregation of alleles at the resistance loci and because inbreeding depression may hamper immune defenses. Individuals of a mixed mating hermaphroditic freshwater snail, Lymnaea ovata, are commonly infected by a digenetic trematode parasite, Echinoparyphium recurvatum. We examined covariation between quantitative resistance to novel parasites and mating system by exposing snail families from four populations that differed by their inbreeding coefficients. We found that resistance was unrelated to inbreeding coefficient of the population, suggesting that the more inbred populations did not carry higher susceptibility load than the less inbred populations. Most of the variation in resistance was expressed among the families within the populations. In the population with the lowest inbreeding coefficient, resistance increased with outcrossing rate of the family, as predicted if selfing had led to inbreeding depression. In the other three populations with higher inbreeding coefficients, resistance was unrelated to outcrossing rate. The results suggest that in populations with higher inbreeding some of the genetic load has been purged, uncoupling the predicted relationship between outcrossing rate and resistance. Snail families also displayed crossing reaction norms for resistance when tested in two environments that presented low and high immune challenge, suggesting that genotype-by-environment interactions are important for parasite resistance. In many plant-pathogen systems, resistance is determined by a genotype-specific interaction between the plant and the pathogen genotypes (reviewed in Thompson 1994;Kover and Caicedo 2001). Recently, genotype-specific interactions have also been found between hosts and their parasites in several animal systems (Lively 1989;Ballabeni and Ward 1993;Ebert 1994;Henter 1995;Ebert et al. 1998;Webster and Woolhouse 1998;Little and Ebert 1999;Webster 2001). Specific interactions of few host and parasite genes are therefore likely to be important in determining which individuals will suffer infections and which will remain healthy.Resistance is also determined by defenses that are independent of the parasite genotype (nonspecific defenses; Frank 2000). These host traits are often quantitative traits governed by several genes (e.g., innate immune system) and evolve through selection on heritable additive genetic variance. Phenotypic value (i.e., effectiveness) of this type of general defense may vary due to both host genotype and environment. For instance, when defense traits are energetically costly, resistance may be related to available resources (e.g., Oppliger et al. 1998;Wiehn and Korpimäki 1998;Ilmonen et al. 2000;Moret and Schmid-Hempel 2000;Rigby and Jokela 2000).Independent of costs, inbreeding may increase susceptibility if inbreeding hampers integri...

In aquatic ecosystems invasive species are among the most important threats to biodiversity worldwide. Understanding the dispersal mechanisms of aquatic invaders is very important for protection and management of vulnerable water bodies. Here we ask how recreational boats that are transported overland could contribute to the dispersal of invasive zebra mussels among lakes in Switzerland. Using a questionnaire sent to registered boat owners, we surveyed properties of transported boats and collected information on self-reported mussel fouling and transport activities of boat owners. We also sampled boat hulls at launching ramps and harbors for biofouling invertebrates. Boats that were kept seasonally or yearround in water were found to have high vector potential with mussel fouling rates of more than 40 %. However, only about 6 % of boats belonging to these groups were transported overland to other water bodies. Considering that approximately 100,000 recreational boats are registered in Switzerland, we estimated that every year around 1400 boats fouled with mussels are transported overland. Such boats pose a high risk of distributing zebra mussels between water bodies. Our results suggest that there is a considerable risk that recreational boats may spread new fouling species to all navigable water bodies within the study area. We speculate that one such species could be the quagga mussel, which has not yet invaded lakes in Switzerland. On a more positive note, our study has identified the group of high-risk boats so that possible control measures would only affect a relatively small number of boat owners.

Harbours are known introduction foci of marine alien species. They act as recipients of new introductions and as sources for regional spread. We report on subtidal surveys of fouling communities from 14 harbours along the coastline of South Africa that were used to identify predictors of high alien species numbers in support of prioritisation of monitoring actions by authorities. The harbours varied in nature from large, international shipping hubs to small, regional fishing harbours and recreational marinas. Fouling assemblages were assessed using visual and scrape sampling to ensure the detection of large, mobile and small inconspicuous species. In total, 29 alien species were recorded, 15 of which were detected outside of their previously known ranges. The number of species recorded per harbour varied from five to. Results revealed that high numbers of alien species were associated with the presence of yachts and low primary productivity. Harbours which had yachts and occurred in areas with mean Chl a minimum levels lower than 0.21 mg.m -3 had the highest number of alien species, while harbours without yachts that were larger than 0.1km 2 supported the fewest alien species. These findings suggest that the presence of yachts can be used to identify harbours with high numbers of alien species, particularly in regions with low productivity. While the applicability of these findings to other regions remains to be tested, this work suggests that harbours that fall within this category could be prioritised for monitoring of marine alien species.

Colonization events like range expansion or biological invasions can be associated with population bottlenecks. Small population size may lead to loss of genetic diversity due to random genetic drift, to loss of heterozygosity due to increased inbreeding and should leave a signature on the genetic polymorphism and genetic structure of populations. The mating system might additionally influence the outcome of such a process. Here, we compare invasive and native populations of the hermaphroditic freshwater snail Lymnaea stagnalis. In the native range we included populations that were ice-free during the last glaciation period and populations that were glaciated and are located at the edge of the species' native distribution range. The microsatellite data show substantial loss of genetic variation in the introduced range and no signs of high propagule pressure or admixture. The expressed polymorphism was so low that mating system analysis was not possible. In the native region, all populations display strong levels of differentiation (global F ST : 0.341) independent of colonization history and exhibit no significant pattern of inbreeding. However, the populations in more recently colonized habitats show diminished genetic diversity. Overall, these results illustrate how dramatic the reduction in genetic diversity can be for hermaphroditic animals and that gene flow in the native range can be surprisingly low despite short distances.

No abstract

Mixed-mating animals self-fertilize a proportion of their offspring. Outcrossing rate may covary with the ecological and historical factors affecting the population. Theory predicts that outcrossing is favored when inbreeding depression is high and when individual heterozygosity is important. Self-fertilization is predicted to be favored when costs of male function, or mate finding are high, for example, when empty patches are colonized by few individuals. In this study, we assessed primary (after hatching) and secondary (after juvenile mortality) outcrossing rates of two mixed-mating snail populations. Our purpose was to assess the variation in mating-system parameters and estimate significance of inbreeding depression for secondary outcrossing rate (the realized outcrossing rate of parents that produce the next generation). Secondary outcrossing rate was higher than the primary outcrossing rate in one of the two populations, suggesting considerable inbreeding depression. In the other study population, secondary outcrossing rates were found to increase when initially low, or decrease when initially high, depending on the family. Moderate outcrossing rates were found to be more stable. Parental inbreeding coefficients were close to zero in both populations. Outcrossing rate was much more variable among families in the population with the lower average outcrossing rate, suggesting that individuals differed considerably in their mating system. Our results add to recent studies suggesting that populations of mixedmating animals may differ in their mating system parameters and expression of inbreeding depression.