||This dissertation addresses the role of mesozooplankton feeding in influencing the trophic interactions in marine planktonic food web. Several related hypotheses were tested in the framework of five individual chapters. Research started with two field investigations studying the response of mesozooplankton biomass and composition to estuarine nutrient loading in Hong Kong coastal waters and the spatial and temporal grazing importance in northern South China Sea (SCS). Trophic cascades induced by mesozooplankton resulted from the release of microzooplankton grazing was demonstrated in laboratory feeding experiments involving simplified triangle food chains and the strengths of trophic cascades were examined for a range of prey concentrations and compared for different predator species with different feeding characteristics. Finally, I went back to field works to test the hypothesis that mesozooplankton carnivory causes a decrease of clearance rate on phytoplankton community. To address this hypothesis, feeding experiments were conducted at two contrasting sites of Hong Kong coastal waters over a two year period to measure mesozooplankton clearance rate and the identity of mesozooplankton grazers in feeding experiments was determined to estimate a net carnivory degree of each assemblage. Field investigation in Hong Kong water indicated that mesozooplankton biomass in the estuarine eutrophic station was not elevated by estuarine nutrient loading compared to the coastal mesotrophic station. Instead, estuarine eutrophic conditions significantly modified the composition of mesozooplankton by increasing the proportion of omnivorous copepods in community and by decreasing some particular groups that are primarily herbivorous such as marine cladocerans, bivalve larvae and gastropod larvae. Field investigation carried out in northern SCS indicated that mesozooplankton in shelf water generally have higher biomass and grazing impacts on phytoplankton than in oceanic water because mesozooplankton in shelf water were more herbivorous. On the other hand, oceanic water was dominated by small-sized phytoplankton that were grazed less efficiently by mesozooplankton. Results from laboratory feeding experiments indicated that because of ambush feeding behavior, the calanoid copepod species Acartia erythraea can induce trophic cascades on marine diatoms and nanoflagellates through predation of heterotrophic dinoflagellates and ciliates, respectively. The magnitude of these trophic cascades was mediated by prey concentrations, but it was primarily determined by the intensity of the predation of top predator on the intermediate trophic levels. The occurrence and magnitudes of trophic cascades were significantly different among three different filtering feeders (two copepod species Parvocalanus crassirostris and Temora turbinata and one cladocean species Penilia avirostris). P. crassirostris significantly preferred ciliates to algae due to size-selectivity and it induced the highest trophic cascading effect among the three species. As a result, the feeding of P. crassirostris caused an increase of algal density. On the contrary, P. avirostris significantly preferred algae over ciliates due to its small upper limits of size range of food particles. Thus this species did not induced trophic cascades, causing a decrease of algal density. The feeding size range of T. turbinate was so broad, feeding the two prey at similar rates that trophic cascades induced by this species balanced with the effect of direct consumption. Results of our feeding experiments conducted in Hong Kong coastal waters supported the hypothesis that mesozooplankton carnivory affect the clearance rate on phytoplankton. The mechanism is that carnivory reduces direct consumption and enhances trophic cascades. Overall, the assemblages of mesozooplankton in Hong Kong coastal waters were omnivorous with varying carnivory degree. Because of size-selectivity of mesozooplankton community, mesozooplankton exert two opposite effects on phytoplankton community: suppressing large-sized phytoplankton (>20 μm) and enhancing small-sized phytoplankton (< 5 μm).