Rapid development of seawater desalination applications using membrane technology in recent years has reignited the issue of boron in drinking water. This is reflected by an increasing number of scientific investigations on the removal of boron by reverse osmosis (RO) and to a lesser extent by nanofiltration (NF) membranes over the last decade. This review provides a summary of the current knowledge relevant to the removal of boron by NF/RO membranes in seawater desalination applications. In seawater, boron exists almost exclusively in the form of boric acid. The speciation of boric acid and other physicochemical properties are delineated and systematically related to the rejection of boron by NF/RO membranes. Studies investigating the rejection of boron by NF/RO membranes reported to date are summarised and reviewed to comprehensively elucidate key factors governing the transport of boron in a typical NF/RO filtration process. The speciation of boric acid as a function of the feed solution pH appears to be the single most important factor governing the rejection of boron by NF/RO membranes. In addition, results reported in the literature reveal an intricate relationship between the speciation of boric acid and several other operating parameters. Some of these operating parameters such as feed solution temperature and ionic strength can directly influence the rejection of boron and at the same time alter the intrinsic dissociation constant of boric acid, hence, indirectly affecting boron rejection by NF/RO membranes. This review also delineates key mathematical modelling approaches, capable of describing the transport of boron in NF/RO filtration processes. The studies included in this review highlight the potential for further process optimisation to enhance the removal of boron in seawater desalination applications. However, the discussions provided also emphasize the need for more fundamental research to improve the scientific understanding of boron rejection by NF/RO membranes. In particular, there is a pressing need to evaluate the impact of membrane fouling and chemical cleaning on the long-term performance of NF/RO systems with respect to boron removal.