Studies of sunfishLepomisspp. have demonstrated that predation regulates size-specific habitat use and competitive interactions both within and among species. However, all of these studies were conducted in clear-water systems with dense littoral vegetation, resulting in well-defined littoral and open-water habitats. In contrast, little is known about predator–prey interactions among fishes that inhabit turbid lakes and reservoirs. In these systems, turbidity reduces detection (reaction) distance for both predator and prey, and can eliminate submerged vegetation that serves as a refuge. In laboratory experiments, we quantified reaction distance of juvenile bluegillsLepomis macrochirusto a predator, largemouth bassMicropterus salmoides, and determined shifts by bluegills between nearshore and offshore habitats across a turbidity gradient. Bluegill reaction distance declined as a negative power function of turbidity from less than 2 m in clear water to 23 cm at 10 nephelometric turbidity units (NTU) and 9 cm at 50 NTU. However, bluegill reaction distance was always greater than largemouth bass reaction distance. In tilted laboratory pools with a depth gradient of 0–50 cm and a largemouth bass predator, bluegills used deepwater habitat (>20 cm deep) less than 20% of the time in clear water and more than 80% of the time at all turbidities greater than 10 NTU. The presence of an obligate open-water prey for largemouth bass, gizzard shadDorosoma cepedianum, did not influence bluegill habitat use. The apparent reduction in open-water predation risk with a relatively minor increase in turbidity suggests that size-specific habitat use by bluegills in turbid systems may not be as simply defined as in clear-water lakes.