An experimental study of microflagellate bacterivory: further evidence for the importance and complexity of microplanktonic interactions

By simulating an upwelling event in a laboratory microcosm, it was possible to promote the development of a natural and diverse planktonic community. An initial bacterioplankton community which developed in response to phytoplankton growth was dominated by small coccoid forms (0,14–0,2 μm³) of the genera Vibrio and Pseudomonas. This group was heavily exploited by the heterotrophic microflagellate Pseudobodo sp. (30 μm³). Later, the bacterioplankton community was dominated by large rods (0,7 μm³) which the flagellates seemed unable to exploit. A Lotka-Volterra predator-prey model fitted to the observed data indicated that the flagellates consumed 2,4 times their carbon body mass per day or 19 bacteria·flagellate^?1·h^?1 when prey were not limiting. Clearance rates were inversely proportional to prey density and ingestion rate, ranging from 2 × 10^?3 to 20 × 10^?3) μ?·flagellate^?1·h^?1. At typical field densities of bacteria and heterotrophic flagellates in the southern Benguela region, between 5 and 30 percent of the water column could be cleared per day. Specific growth rates of the flagellates were positively related to prey density, the maximal rate being 0,84 · d^?1. Their initially faster growth rates allowed bacteria to increase in numbers despite predation. The growth yield of the flagellates (34–36 per cent) was also positively related to food density. Such low values suggest inefficient transfer of carbon to higher trophic orders but considerable nitrogen regeneration. Nitrogen excretion rates were approximately 6–7 μg N·mg dry weight^?1·h^?1, comparable to other flagellates but faster than ciliates. These rates are comparable with in situ measurements of NH⁺₄-N excretion in pycnoclinal regions based on ¹⁵N isotope studies but are only about 20 per cent of measured rates in surface waters. This is interpreted to mean that, in pycnoclinal regions where the relative input of "new" nitrogen is high, there are few regenerative steps and the model describes them satisfactorily. In surface waters, observed rates of excretion can only be accounted for by many regenerative steps in a highly complex food chain in which the cumulative total of nitrogen excretion at each step amounts to that based on ¹⁵N labelling studies.