Heterotrophic nanoflagellates comprise an important consumer group within pelagic food webs and play a vital role in cycling carbon and nitrogen in the world's oceans. Most primary production passes either directly or indirectly through heterotrophic nanoflagellates in the open ocean. Oceanic nanoflagellates generally have a simple, smooth form and capture their picoplankton-size prey by chance contact with their cell surface as they swim randomly through the fluid medium. In theory, this feeding mechanism does not involve the structural complexity, inter-taxonomic variability, or obvious behavioral complications of higher organisms. Consequently, it may be possible to understand a great deal about this ecologically important trophic interaction by viewing it from a purely physical perspective.

Prey capture by marine nanoflagellates is formulated in terms of physical forces arising between nanoflagellates and picoplankton prey as they approach one another. A numerical model developed to study colloidal interactions is adapted to make predictions of nanoflagellate grazing rates. Predictions are formulated as a function of prey size, nanoflagellate swimming speed and the resultant of a balance between hydrodynamic, London-van der Waals, electrostatic and solvation forces.

In short, the relative strengths of attractive or repulsive molecular and hydrodynamic forces determine nanoflagellate grazing rates on picoplankton prey of a particular size. One of the more significant model predictions is that nanoflagellate grazing rates should increase approximately linearly with prey diameter -- a relationship that conflicts with previously published quadratic and cubic dependencies. Experiments aimed at resolving this discrepancy provides support for a linear size dependency.

The model ignores post-contact rejection of prey. If selective prey rejection occurs to a significant extent, a random contact model of the kind described here would be unreliable for studying natural systems that contain variable prey types. To test for selection based on factors other than size, ingestion rates are compared for a chrysomonad population feeding on different species of live or heat-killed bacteria, Synechococcus sp., and latex microspheres. Results indicate that post-contact prey rejection is not significant for this organism. Results from this work also suggested that highly motile bacteria may experience significantly greater predation losses due to an increase in the encounter rate with nanoflagellates.

All grazing rate predictions are low by a factor of 3 or 4 when compared with well-accepted literature values. Varying modeled London-van der Waals and electrostatic double layer forces within reasonable natural limits caused predicted grazing rates to change by less than a factor of two. This led to a reexamination of the importance of the hydrophobic interaction force -- neglected in the current model. Experiments designed to directly measure the effect of the hydrophobic force on grazing rates show that a modest increase in prey-surface hydrophobicity, as revealed by hydrophobic interaction chromatography, can triple the rate at which prey are ingested by nanoflagellates. This result may explain the discrepancy between model predictions and observations of the absolute magnitude of nanoflagellate grazing rates.