To test the role of vertebrate predation on benthic community structure, experiments were performed in two third-order streams in northern Québec, Canada. Predators were either enclosed in or excluded from replicate benthic cages (20 × 30 × 10 cm). Ruisseau du Cran Carré (RCC) has a normal complement of fish, while Low Hope Creek (LHC) is fishless. In June 1983, two species of fish were manipulated (one fish per cage, five replicate cages per treatment) in RCC, while the salamander Eurycea bislineata was manipulated in LHC. In August 1983, brook trout fry were manipulated in both streams. It was anticipated that the fauna in LHC, having had no previous contact with fish, would be more susceptible to fish predation than the fauna of RCC. Gut analysis revealed that fish did feed in the cages. However, neither fish predation nor salamander predation had significant effects on prey communities in either stream in June. Community parameters (species richness and diversity, and total numbers of individuals) were not influenced by the presence or absence of fish or salamanders. In August, three-quarters of the taxa were more abundant without trout than with trout. However, few individual taxa were significantly affected by the fish. This effect was similar in both streams. By enclosing fish in a fishless stream, this experiment was designed to enhance the possible effects offish predation on the lotic macrobenthic community. This work strengthens the idea that brook trout are not major structuring agents of the macrobenthic communities of small streams.
Investigating the role of refuges and drift on the resilience of macroinvertebrate communities to drying conditions: An experiment in artificial streams
