Hydrodynamic surveys were conducted in a partially stratified estuary. Longitudinal salt fluxes through a cross-section were measured and the optimal number of lateral and 'vertical points for observation and estimation of fluxes ofpollutants found. The hydrodynamic data were decomposed to investigate the mechanism of instantaneous and tidally averaged mass fluxes.The decomposition method developed showed that instantaneous salt flux arose principally from the combination of velocity fluctuation with the tidal mean salinity. The error analysis showed that the estimate of mass flux was sensitive to the error inherent in the velocity data and the error range was within + 10% of the total flux. Seven. depth and four width points gave acceptable estimates of the water flux, salt flux, and longitudinal dispersion.An important feature of estuarine flow systems is the great spatial and temporal (seasonal, semidiurnal, and diurnal) variation resulting from interaction between tidal oscillation and freshwater flow. Having field observations be representative in an estuary is thus highly dependent on the sampling design (sampling locations, sampling rate, and sampling durations) as well as a rational procedure of data analysis and synthesis (Kjerfve 1979). Cross-sectional variations in velocity and pollutant (dissolved or suspended) concentrations may have a significant effect on longitudinal mass flux. One must decide how many lateral and vertical measuring positions are required to estimate longitudinal mass fluxes through a cross-section within an acceptable error. The standard sampling rate is typically once every hour, but only a few studies have been conducted with sufficient spatial resolution to estimate longitudinal mass fluxes precisely.Velocity and concentration values have been decomposed to investigate the relative importance of transverse and vertical variations (e.g. Hansen 1965; Fischer 1972;Smith 1980;Lewis and Lewis 1983). These AcknowledgmentsWe thank R. E. Lewis for advice during this research and K. R. Dyer for supplying the data for Southampton Water.