Differential grazing on natural planktonic populations by the mussel Perna canaliculus

Phytoplankton biomass alone cannot predict growth and condition in the mussel Perna canaliculus. This study investigates whether food selection processes, low removal rates of small picophytoplankton-sized particles (0.2 to 2 mu m) or additional heterotrophic food sources such as microzooplankton and bacteria may help explain this observation and therefore aid in predicting growth and condition in P. canaliculus. Three grazing experiments were run at different times of the year, using naturally varying planktonic assemblages. P. canaliculus did not feed efficiently on picophytoplankton or bacterial particles and higher filtration rates were always observed on larger phytoplankton (>2 mu m). Much of the 0.2 to 2 mu m size fraction, whether heterotrophic or autotrophic, was not directly utilised for mussel growth. In terms of phytoplankton biomass, grazing and ingestion was often highest on naked flagellated cells. Results indicated that selection of some phytoplankton for ingestion was based on food quality, with both small and large naked flagellated cells with high carbon content often preferred over other morphotypes. P. canaliculus also supplemented its diet with heterotrophs, heavily grazing a range of microzooplankton. Grazing overall showed selection for naked flagellated or ciliated microzooplankton with high carbon content. The results indicate that rather than directly accessing picophytoplankton and bacteria, P. canaliculus accesses these populations through the food web by grazing upon microzooplankton, mixotrophic algae and, potentially, aggregates. This study adds to our ability to predict mussel condition by identifying key food sources for the growth of P. canaliculus.