The main objective of this study was to investigate the influence of the feed solution chemical composition on the performance and fouling, both biofouling and abiotic fouling, of various reverse osmosis (RO) and nanofiltration (NF) membranes. Fouling experiments were performed as a factorial design with four independent variables: divalent cation concentration, monovalent cation concentration, dissolved organic carbon (DOC) concentration and membrane type denoted by Molecule Weight Cut-Off (MWCO) and roughness, each with three levels to produce a 34 factorial design. The membranes tested were the following: TFC-S (Koch Membrane Systems, San Diego, CA) with MWCO of 200Da and the roughness of 35.64nm; 4040-UHA- ESPA (Hydranautics, San Diego, CA) with MWCO of 100-300Da and the roughness of 78.39nm; and, NTR 7450 (Hydranautics, San Diego, CA) with MWCO of 500Da and the roughness of 27.84nm. These were run at pressures of 280 psi, 120 psi, and 120 psi, respectively. Experiments were conducted using a flat sheet test unit (Osmonics, Minneetonka, MN), which contained a cell with 155cm2 active membrane area. Feed water was pumped into the test unit from feed reservoir, and all the concentrate flux was recycled to the feed tank to increase the fouling potential on the membranes. New membranes were rinsed with deionized (DI) water, then soaked membrane-skin-down in DI water overnight, and rinsed again immediately before installation into the filtration cell unit. Abiotic fouling and biofouling tests on the fouled membrane were also conducted to characterize the fouling on the membrane after each experiment. The membrane specimens were soaked in a basic (0.1M NaOH) solution overnight and then UV254 and turbidity were measured to characterize the abiotic fouling of the membrane. Biofouling of the membrane was measured by detaching the biofilm from the membrane surface and measuring the heterotrophic plate counts (HPC). Includes 9 references, figures.