Interaction between chemical and tactile cues in mayfly detection of stoneflies

• 1 Sediments and eleven species of rooted submerged aquatic macrophytes were sampled from twenty-five lakes in southern Quebec to examine species patterns in metal content.
• 2 Linear regression models relating the content of metals in the macrophytes to their sediments demonstrated that the isoetoid (Eleocharis acicitlaris, Eriacaulon septangulare, Isoetes sp. and Sagittaria graminae) and the non-isoetoid species had different plant-sediment relationships. Canopy-forming (Myriophyllum spicatum, Potamogeton amplifalius, P. crispus and P. perfoliatus) and understory species (Elodea canadensis, P. robbinsii and Vallisneria americana) had similar linear regression model slopes, but the understory species had significantly higher intercepts, indicating higher plant metal contents for a given sediment metal concentration.
• 3 We grouped species by their biomass density (biomass (kg)/height (m)), a measure of the packing of the plant canopy, and averaged values for each metal, thereby producing one model per group. Comparison of the models by analysis of covariance revealed no significant differences in metal content. However, the canopy-forming species that occurred in monospecific and mixed species stands (M. spicatum and P. amplifolius) had significantly higher metal content for a given sediment metal content when an understory was present.
• 4 These results show that sediment metal content explains the largest proportion of variation in plant metal content across all sites and species. Deviations from what would be predicted by the most general model are not related to large-scale differences in sediment geochemistry, but to the physical structure of the weedbed within geochemically similar sediments.