Cadmium, chromium, copper, nickel, lead and zinc concentrations in the bivalve mollusc Macoma balthica and the polychaete annelids Hediste diversicolor and Arenicola marina were measured, during winter and summer, at sites throughout the Severn Estuary and Bristol Channel. The mean concentration of each metal in A. marina was greater in the lower Severn Estuary than in the far less contaminated outer Bristol Channel and the concentration of a given metal, e.g. Cr, in a species occasionally peaked at a site, reflecting local metal contamination. The concentrations of each metal in each of these biomonitor species almost invariably differed significantly among sites and often seasons and there were sometimes interactions between site and season. This indicates that the various factors that determine the concentration of a metal in a species operate in a complex manner and that their individual effects can vary among sites and/or seasons. The rank order of each metal concentration in each species at a site within the estuary frequently did not match the sequence for the concentration of that metal measured in the sediment at that site at the same time. This lack of correspondence is likely to be due, at least in part, to one or both of the following: (1) variations in the bioavailability of certain metals among sites due to differences in such features as the metal-binding properties of the sediments; (2) the effects of the constant transport and redistribution of the sediments and thus also of their associated trace metals by the very strong tidal action that characterizes the Severn Estuary. This would mean that single time measurements do not accurately reflect the overall trace metal environment to which the biomonitor organism had been exposed in the weeks/months prior to sampling. Marked differences in the concentrations of certain metals, e.g. Cu and Zn, in co-occurring biomarker species could frequently be related to differences between the ability of these species to regulate certain metals. Non-metric multi-dimensional scaling ordination and associated tests emphasize that the relationships between the concentrations of the various metals differed markedly among species and between sites and seasons in individual species and elucidated which metals contributed most to those differences. If the proposed scheme for harnessing tidal power in the Severn Estuary proceeds, the data in this paper provide a baseline for assessing the impact of such major changes on the bioavailability of trace metals in this estuary. This information will also be invaluable for predicting the changes likely to occur in other estuaries that become subjected to major structural changes.