Fluoride concentrations in eastern UK rivers (the Humber, Tweed, Wear, Great Ouse and Thames) are described based on information collected within the Land-Ocean Interaction Study (LOIS) and by the Environment Agency (EA) of England and Wales. The results show varied fluoride concentrations across the region, with a range from <0.01 to >10 mg l(-1); and mean, median and range in mean concentrations of 0.30, 0.21 and 0.05-3.38 mg l(-1) (excluding one outlier point), respectively. Within the main rivers and tributaries, the mean fluoride concentration varied from approximately 0.5 to over 2 mg l(-1) and the highest values occurred within the Don basin (Don, Dearne and Rother) and parts of the Trent basin (upper Tame and mid-upper Derbyshire Derwent) in highly industrialised and urbanised areas (Sheffield and Rotherham in the Don basin; Birmingham and Derby on the Trent). For localised inputs to the rivers, fluoride concentrations were slightly higher, and considerably higher in one outlier case. Correspondingly, the other rivers examined typically had mean fluoride concentrations between approximately 0.2 and 0.5 mg l(-1), but fluoride concentrations were lower in the headwater areas. As there is much less information on fluoride levels in upland areas, extensive data collected as part of an acid waters survey are used to show that fluoride concentrations are generally less than 0.1 mg l(-1) for the upland UK. The data are summarised in terms of both fluoride concentrations and flux, and the values are cross-referenced to other determinands collected within LOIS. The high positive correlation with boron and negative correlation with flow show the importance of point source (sewage) inputs of fluoride, while strong positive correlations between fluoride and barium indicate the relative importance of vein mineralisation in the bedrock in supplying fluoride to the waters of the Yorkshire Ouse and its tributaries. There seems to be some process that limits the fluoride concentrations within the more contaminated rivers, and this is indicated by a negative upper bound between fluoride and calcium. The waters are undersaturated with respect to fluorite and oversaturated with respect to calcium fluoro-phosphates. This upper bound may reflect either physical controls, such as the availability and size of point and diffuse sources for fluoride coupled to mixing of these sources with rain and soil runoff of low concentration, or solubility controls for a pure or mixed-phase mineral that cannot be specified here.