Colloidal fouling of reverse osmosis (RO) elements can seriously impair the performance of membranes resulting in loss of productivity. The source of silt in water is varied and often includes bacteria, clay, colloidal silica, and corrosion products. Various pre-treatment chemical can be used to agglomerate the fine colloidal particles which in-tern can be removed easily however if the incorporation of pre-treatment chemicals with the fine colloidal particles fails then these pre-treatment chemicals can also become a source of membrane fouling. In addition, cationic polymers may co-precipitate with the negatively charged anti-scalants resulting in membrane fouling. Several methods have been proposed to predict the colloidal fouling potential of feed water such as the modified fouling index (MFI) and the most commonly used silt density index (SDI). SDI is an important design parameter for an RO membrane water treatment process. In this investigation the effect of SDI index was investigated to predict the fouling tendency of different pore size membranes. The SDI index showed elevated values (~12) with big pore size (2, 3, 5, 10μm) membranes however lower fouling index values (~6) were recorded with membrane pore sizes near to 0.45μm even with different kind and make of membranes.