Grasses were grown in hydroponic culture tanks to evaluate their nutrient removal capabilities when supplied with secondary-treated sewage effluent as the sole source of plant nutrients. Statistical methods were employed to determine the effects of the grasses, flow rates, and seasons on nutrient removal. Two control tanks with gravel bed and no grass were maintained throughout the study, two were planted with tall (Ky 31) fescue, and two were planted with perennial ryegrass. Two flow rates were maintained in each pair of tanks, approximately one-day and two-day detention times. All six tanks were effective in reducing the oxygen-demanding organic content of the effluent. Total nitrogen content was reduced appreciably by the control tanks, but the grass tanks were significantly better at nitrogen removal. Total phosphorus concentrations were reduced only slightly by passage through the tanks. The fast flow rate tanks produced the greater grass yields, while the slow flow rate tanks were more effective in nutrient removal from the sewage effluent. From grass yield and analyses data, the amount of plant nutrient material removed by the grasses was small compared to the total quantity supplied. (Author)