The cencept of mean cell residence time, MCRT, as a method of controlling activated sludge process was evaluated by treating artificial sewage through a bench scale activated sludge process unit. Two types of sewages, low organic content sewage, COD = 470 mg/l, and high organic content sewage, COD = 936 mg/l, were used. Six identical units were operated simultaneously, but controlled under six different MCRT values, 3,4,6,8,10 and 12 days. When steady state were achieved analysis and calculations were made and following parameters were obtained; k = 0.0167 $\mbox{day}^{-1}$, Y = 0.54 mg MLSS/mg COD, Kd = 0.33 mg MLSS/day/mg MLSS, a`` = 0.93 mg $O_2$/mg COD, b`` = 0.8 mg $O_2$/day/mg MLSS. During the operation MCRT affected the settling characteristics of activated sludge, i. e, a filamentous bulking was observed at low MCRT such as 3 or 4 days, and a pin point bulking was occured at MCRT of 12 days. It was also observed that throughout the all MCRT the proportion offilamentous microorganisms was higher in the high organic content sewage. The tolerance of the unit against the shock loading was increased as the MCRT increased. Considering the effluent soluble COD, settling characteristics of activated sludge and response to shock loading the MCRT values of 8 and 10 days were found to be suitable for treating the sewages tested.