Abstract
To examine the hypothesis that invasion by zebra mussels ( Dreissena polymorpha ) promotes phytoplankton dominance by the noxious cyanobacterium Microcystis aeruginosa , 61 Michigan lakes of varying nutrient levels that contain or lack zebra mussels were surveyed during late summer. After accounting for variation in total phosphorus (TP) concentrations, lakes with Dreissena had lower total phytoplankton biomass, as measured by chlorophyll a and algal cell biovolume. Phytoplankton biomass increased with TP in both sets of lakes, although the elevations of the relationship differed. The percentage of the total phytoplankton comprised by cyanobacteria increased with TP in lakes without Dreissena (R 2 = 0.21, P = 0.025) but not in lakes with Dreissena (P = 0.79). Surprisingly, there was a positive influence of Dreissena invasion on Microcystis dominance in lakes with TP ≪ 25 µg L −1 (P 5 0.0018) but not in lakes with TP ≫ 25 µg L −1 (P = 0.86). The finding that Microcystis , a relatively grazing‐resistant component of the phytoplankton, was favored by Dreissena in low‐ but not in high‐nutrient lakes is somewhat counterintuitive, but predator‐prey models make this prediction in certain cases when the cost for the prey of being consumption resistant is a low maximum population growth rate. This Dreissena ‐cyanobacteria interaction contradicts well‐established patterns of increasing cyanobacteria with nutrient enrichment in north‐temperate lakes and suggests that the monitoring and abatement of nutrient inputs to lakes may not be sufficient to predict and control cyanobacterial dominance of Dreissena‐invaded lakes.