The North Texas Municipal District (NTMWD) experiences seasonal taste and odor (T&O) problems caused by the algal metabolites, geosmin and MIB. As part of a comprehensive T&O study, bench- and full-scale tests were performed to evaluate the effectiveness of powdered activated carbon (PAC) in removing geosmin and MIB from the NTMWD source water, Lake Lavon. The results of the bench- and full-scale tests are reported in this paper. NTMWD operates three conventional surface water treatment plants with a combined capacity of over 600-mgd. Each plant has its own raw water intake and pumping station on Lake Lavon. Ferric sulfate is used to provide enhanced coagulation at each plant. Jar testing was performed at Texas A&M University with four PAC types. The sample water from Lake Lavon was spiked with a Carbon 14 radiolabeled MIB (14C-MIB), and MIB concentrations in the spiked and treated water were measured with a scintillation counter. The bench-scale testing objectives were: to develop dose-response curves for MIB removal specific to the NTMWD source water and treatment plant conditions; to compare the treatment efficiencies for PAC addition at the raw water intake and rapid mix basin; and, to evaluate the cost efficiency of different PAC types. Several researchers have demonstrated that geosmin is more readily removed by PAC adsorption than MIB 2, 3, 4. Consequently, MIB was selected for testing to provide a conservative representation of treatment requirements for geosmin removal. The jar testing protocol was designed to simulate contact time and chemical additions at the fullscale plants. Following jar testing, full-scale tests were conducted at one of the NTMWD water treatment plants. The objectives of the full-scale tests were: to verify bench-scale test results and calibrate the bench-scale dose-response curves to full-scale conditions; and, to test the effects of sufficient PAC doses on filter headloss. PAC addition was tested at the rapid mix basin and with 30 minutes of PCT to simulate addition at the raw water intake. During full-scale testing, filter headloss was measured every two hours to evaluate the impact of the PAC doses on filter run times. Includes 5 references, tables, figures.