1. Grazer and nutrient controls of phytoplankton biomass were tested on two reservoirs of different productivity to assess the potential for zooplankton grazing to affect chlorophyll/phosphorus regression models under Australian conditions. Experiments with zooplankton and nutrients manipulated in enclosures, laboratory feeding trials, and the analysis of in-lake plankton time series were performed.2. Enclosures with water from the more productive Lake Hume (chlorophylla = 3–17.5 μg l–1), revealed significant zooplankton effects on chlorophyll a in 3/6, phosphorus limitation in 4/6 and nitrogen limitation in 1/6 of experiments conducted throughout the year. Enclosures with water from the less productive Lake Dartmouth (chlorophyll a = 0.8–3.5 μg l–1), revealed significant zooplankton effects in 5/6, phosphorus limitation in 5/6 and nitrogen limitation in 2/6 of experiments.3. While Lake Hume enclosure manipulations of the biomass of cladocerans (Daphnia and Diaphanosoma) and large copepods (Boeckella) had negative effects, small copepods (Mesocyclops and Calamoecia) could have positive effects on chlorophyll a.4. In Lake Hume, total phytoplankton biovolume was negatively correlated with cladoceran biomass, positively with copepod biomass and was uncorrelated with total crustacean biomass. In Lake Dartmouth, total phytoplankton biovolume was negatively correlated with cladoceran biomass, copepod biomass and total crustacean biomass.5. In both reservoirs, temporal variation in the biomass of Daphnia carinata alone could explain more than 50% of the observed variance in total phytoplankton biovolume.6. During a period of low phytoplankton biovolume in Lake Hume in spring–summer 1993–94, a conservative estimate of cladoceran community grazing reached a maximum of 0.80 day–1, suggesting that Cladocera made an important contribution to the development of the observed clear-water phase.7. Enclosure experiments predicted significant grazing when the Cladocera/Phytoplankton biomass ratio was greater than 0.1; this threshold was consistently exceeded during clear water phase in Lake Hume.8. Crustacean length had a significant effect on individual grazing rates in bottle experiments, with large Daphnia having highest rates. In both reservoirs, mean crustacean length was negatively correlated with phytoplankton biovolume. The observed upper limit of its variation was nearly twice as high compared to other world lakes.