Two billabongs on the floodplain of the Murrumbidgee River, Australia, were partitioned in half with impermeable plastic barriers for four months from summer to winter 1995. The densities of carp were manipulated to establish high- and low-carp biomass treatments in each billabong. Final standing stocks of carp in the high- and low-carp treatments of each billabong were 1181 and 101 kg ha-1 , and 669 and 348 kg ha-1 , respectively. Turbidity, phytoplankton biomass and total and dissolved phosphorus concentrations were similar in both ends of each billabong before the establishment of the plastic barriers. Throughout the experimental period, turbidity and phytoplankton biomass were significantly higher in the high-carp treatment in both billabongs. However, the two billabongs had different temporal patterns for both turbidity and phytoplankton biomass. Concentrations of total phosphorus were usually greater in the high-carp end of each billabong but there was no consistent pattern for dissolved phosphorus. Carp had a significant impact on turbidity and intensity of algal bloom but this varied with carp biomass and billabong sediment type, and factors other than carp usually contributed to most of the variation in measured water quality.