As there is an increasing need to minimize the potential of nitrate contamination of groundwater from the effluents released from industries, waste treatment plants and dairies, we need to develop novel remediation techniques. One such remediation technique is to grow nitrogen (N) scavenging crops, commonly referred to as “bio-filters”. Our goal was to evaluate optimal harvest time for Elephant grass (Pennisetum sp.) and Sudan grass (Sorghum bicolor) which were grown in 5 gallon pots in a greenhouse and irrigated with secondary treated municipal waste water (MW) and dairy effluent (DE). The experiments were set up as completely randomized designs (CRD) with three rates (0, 50 and 100 percent) of effluent, and replicated four times with three harvest times (8, 10 and 12 weeks). Application rates had no significant effect on grass yields. The average biomass for the grasses harvested at 8 and 10 weeks were generally higher for plants irrigated with the DE than those irrigated with the MW.  By the 12^th week, similar yields were obtained for each grass regardless of the water source. Generally, the highest crude protein (CP) content, relative feed value (RFV) and total digestible nutrients (TDN) were detected in grasses harvested at eight weeks. The exception was the EG treated with MW, which had its greatest values at 12 weeks. Grasses irrigated with DE exhibited their greatest nitrate content earlier (at 8 weeks) than those receiving MW. More importantly, the grasses receiving MW accumulated as much as five times more nitrate than the grasses treated with DE. These findings concur with those from our previous studies in which Elephant grass has been identified as a highly nutritious forage crop with the ability to readily take up N from soils subjected to high rates of N.