It has been estimated that 6000 tons per year of mercury are released into the environment, 4000 tons from anthropogenic sources. The adverse effects of mercury on both human and animal health have led to regulators enforcing increasingly stringent discharge limits. Limits as low as 10 μg/L are typical in many jurisdictions, and even lower in highly sensitive environments. A proven technique for removing mercury from wastewaters is carbon adsorption; however this technology becomes less economical for high flow streams due to the capital costs associated with constructing adsorption units. These costs would be significantly reduced if the mercury was concentrated into a lower volume stream, reducing the size of the required facility. This would be enhanced by improved adsorption kinetics associated with the higher mercury concentration. This investigation assessed the viability of using cross-flow reverse osmosis filtration (RO) to concentrate the mercury. A series of flat-sheet membrane tests were conducted at pH 2, 7 and 10 using a feed solution with a concentration of 30 mg/L Hg. The data produced was fed into a model for predicting filtration plant design. For each pH tested, it is predicted that RO will concentrate 99% of the mercury into a low volume stream using a 4-pass configuration. This would produce a permeate stream that is potentially suitable for industrial re-use. The model also predicted that the feed stream at each pH could be treated by an 8-pass RO system to produce a permeate water achieving regulatory discharge limits of < 10 μg/L.