We estimated the grazing impact of the heterotrophic flagellate Collodictyon triciliatum on the harmful, bloom-forming cyanobacterium Microcystis aeruginosa in an experimental pond during a Microcystis bloom from summer to winter in 2010. For these experiments, we calculated the grazing rates from the digestion rate of C. triciliatum and its food vacuole contents. During the study period, M. aeruginosa exhibited one bloom event with a maximum density of 1.1 x 10(5) cells ml(-1). The cell density of C. triciliatum fluctuated from below the detection limit to 291 cells ml(-1). The number of M. aeruginosa cells ingested by C. triciliatum food vacuoles ranged between 0.4 and 10.8 cells flagellate(-1), and the digestion rate of C. triciliatum at 25 A degrees C was 0.73 % cell contents min(-1). The grazing rate of C. triciliatum on the M. aeruginosa prey was 0.2-6.9 cells flagellate(-1) h(-1), and its grazing impact was 0.0-25.3 % standing stock day(-1). The functional response of C. triciliatum to the M. aeruginosa prey followed the Michaelis-Menten model of significance (r (2) = 0.873, p < 0.001) in our experimental systems, in which the prey concentration varied from 1.0 x 10(4) to 2.1 x 10(6) cells ml(-1). The maximum grazing rate was 6.2 prey cells grazer(-1) h(-1), and the half-saturation constant was 1.2 x 10(5) cells ml(-1). We present evidence that C. triciliatum grazing explained the remarkable decrease in M. aeruginosa cell density in the pond. The present study is the first demonstration of the high potential of protistan grazing on M. aeruginosa to reduce cyanobacterial blooms.