Campsites, forest fires, and entry point distance affect earthworm abundance in the Boundary Waters Canoe Area Wilderness.

Factors controlling the spread of invasive earthworms in Minnesota's Boundary Waters Canoe Area Wilderness are poorly known. Believed to have been introduced by anglers who use them as bait, invasive earthworms can alter the physical and chemical properties of soil and modify forest plant communities. To examine factors influencing earthworm distribution and abundance, we sampled 38 islands across five lakes to assess the effects of campsites, fire and entry point distance on earthworm density, biomass and species richness. We hypothesized that all three parameters would be greater on islands with campsites, lower on burned islands and would decrease with distance from the wilderness entry point. In addition to sampling earthworms, we collected soil cores to examine soil organic matter and recorded ground and vegetation cover. Campsite presence was the single most important factor affecting sampled earthworm communities; density, biomass and species richness were all higher on islands having campsites. Fire was associated with reduced earthworm density, but had no direct effects on earthworm biomass or species richness. Fire influenced earthworm biomass primarily through its negative relationship to groundcover and through an interaction with entry point distance. Entry point distance itself affected earthworm density and biomass. For islands with campsites, earthworm biomass increased with distance from the entry point.

Activity levels and predator detection by amphipods infected with an acanthocephalan parasite, Pomphorhynchus laevis.

The acanthocephalan parasite Pomphorhynchus laevis (Müller, 1776) uses freshwater amphipods as its intermediate host. In order to complete the life cycle, the infected amphipod must be consumed by a fish, where the acanthocephalan will mature and reproduce. Parasite transmission, and therefore fitness, could be enhanced if infected amphipods fail to detect or avoid predatory fish. We compared the activity levels of infected and non-infected amphipods, Echinogammarus stammeri (Karaman, 1931), in both the presence and absence of odours from its natural, definitive host, the fish Leuciscus cephalus (L.). Throughout the experiment, infected amphipods were more active than were non-infected individuals. The non-infected amphipods reduced their activity after the addition of fish odours, but the infected amphipods failed to show a significant decrease. The failure of infected amphipods to reduce activity levels in the presence of fish odour may reflect a parasite strategy to increase its chances of transmission by making its amphipod host more vulnerable to predation by fish.

Redundancy among three herbivorous insects across an experimental current velocity gradient.

We conducted streamside experiments to determine if the ability of herbivorous insects to remove algal periphyton varies with local current velocity. We used two mayfly species (Baetis bicaudatusand Drunella grandis) and one caddisfly species (Glossosoma verdona), which differ from one another in body morphology and mobility. Periphyton was grown for 30 days on ceramic tiles in constant velocity to create similar initial forage conditions for grazers. Tiles were transferred to three velocity regimes characteristic of the natural streambed: slow (3-5 cm s(-1)), medium (15-20 cm s(-1)) and fast (32-41 cm s(-1)). Four grazer treatments (Baetis, Drunella, and Glossosoma alone, and all species combined) were repeated for each velocity treatment to isolate the effect of local current on grazer ability to crop periphyton. Grazers differed in their abilities to remove periphyton across current treatments. Glossosoma removed significantly (P<0.05) more periphyton at fast versus either slow or medium velocities; Baetis showed a similar (but non-significant) trend; and, Drunella always removed about 75% of periphyton, irrespective of current. At fast current, periphyton removal was equivalent among the species. At medium current, Drunella removed significantly more than both Baetis and Glossosoma, whereas at slow current, Drunella removed more than Baetis, which removed more than Glossosoma. Periphyton removal under the combined three-grazer treatment was similar qualitatively to the combined effects of individual grazers. More periphyton tended to be removed as current increased, with the fast versus slow contrast showing marginal significance (P=0.10). Under all current regimes, the quantity of periphyton removed did not differ from the null model expectation of simple additive effects among individual grazers (i.e., no facilitation or inhibition). These experiments show that for some species, herbivory varies with current, which suggests that the herbivore "function" of cropping periphyton may vary with the environmental context of local current. Under some local velocities, however, different herbivore species "function" similarly and are potentially redundant with respect to periphytic removal. In naturally heterogeneous streams characterized by sharp gradients in local current velocity, we expect current-dependent species interactions to be common and at least partially contribute to intra-guild co-existence of species.