The trophic efficiency of the planktonic food web in the Phaeocystis-dominated ecosystem of the Belgian coastal waters was inferred from the analysis of the carbon flow network of the planktonic system subdivided into its different trophodynamic groups. A carbon budget was constructed on the basis of process-level field experiments conducted during the spring bloom period of 1998. Biomass and major metabolic activities of auto- and heterotrophic planktonic communities (primary production, bacterial production, nanoproto-, micro- and mesozooplankton feeding activities) were determined in nine field assemblages collected during spring at reference station 330. In 1998, the phytoplankton spring flowering was characterised by a moderate diatom bloom followed by a massive Phaeocystis colony bloom. Phaeocystis colonies, contributing 70% to the net primary production, escaped the linear food chain while the early spring diatom production supplied 74% of the mesozooplankton carbon uptake. The rest of mesozooplankton food requirement was, at the time of the Phaeocystis colony bloom, partially fulfilled by microzooplankton. Only one-third of the microzooplankton production, however, was controlled by mesozooplankton grazing pressure. Ungrazed Phaeocystis colonies were stimulating the establishment of a very active microbial network. On the one hand, the release of free-living cells from ungrazed colonies has been shown to stimulate the growth of microzooplankton, which was controlling 97% of the nanophytoplankton production. On the other hand, the disruption of ungrazed Phaeocystis colonies supplied the water column with large amounts of dissolved organic matter available for planktonic bacteria. The budget calculation suggests that ungrazed colonies contributed up to 60% to the bacterial carbon demand, while alternative sources (exudation, zooplankton egestion and lysis of other organisms) provided some 30% of bacterial carbon requirements. This suggests that the spring carbon demand of planktonic bacteria was satisfied largely by autogenic production. The trophic efficiency was defined as the ratio between mesozooplankton grazing on a given source and food production. In spite of its major contribution to mesozooplankton feeding, the trophic efficiency of the linear food chain, restricted to the grazing on diatoms, represented only 5.6% of the available net primary production. The trophic efficiency of the microbial food chain, the ratio between mesozooplankton grazing on microzooplankton and the resource inflow (the bacterial carbon demand plus the nanophytoplankton production) amounted to only 1.6%. These low trophic efficiencies together with the potential contribution of ungrazed Phaeocystis-derived production to the bacterial carbon demand suggest that during spring 1998 most of the Phaeocystis-derived production in the Belgian coastal area was remineralised in the water column.