Linear models were developed for predicting mean Secchi disk depth readings as a measure of water clarity for the summer months in Lake Mendota, Wisconsin. The 20-yr (1976-1995) data set also included external phosphorus (P) loadings and in-lake April P concentrations as indices of lake nutrient status, and monthly (28 d) water column stabilities and Daphnia biomasses as indices of lake mixing and algal grazing potentials, respectively. June Secchi depths were mostly controlled by food web dynamics, which dictated whether the larger-bodied Daphnia pulicaria or the smaller-bodied D. galeata mendotae dominated during the spring clear-water phase. May Daphnia biomasses were significantly greater in D. pulicaria-dominated years than in D. galeata-dominated years; D. pulicaria-year biomasses were also greater during the summer months. The model for the midsummer (July-August) months indicated that Secchi depths were inversely related to April P concentrations and positively related to midsummer Daphnia biomasses and lake stabilities. Scenarios for midsummer Secchi depths were tested using the observed minimum and maximum values for each predictor variate. While holding two variates constant, April P, Daphnia biomass, and lake stability each resulted in relatively similar Secchi ranges (0.81, 0.81, and 1.17 m, respectively). Our results suggest that summer water clarity in eutrophic Lake Mendota is controlled by interacting ecosystem processes linked to land use activities, lake food web dynamics, and climate.