(1) A winter survey of seven species of wading birds (Charadrii) at forty intertidal sites in six estuaries in south-west England was made to identify the variables that determined the variation in bird densities between the sites and to develop a method for predicting bird densities should a tidal power barrage be built on the Severn estuary. (2) The densities of the two smallest waders, the dunlin (Calidris alpina) and redshank (Tringa totanus), were comparable on the twelve sites in the Severn to those in the other estuaries. In contrast, the densities of the larger species, the grey plover (Pluvialis squatarola), bar-tailed and black-tailed godwits (Limosa lapponica and L. limosa), curlew (Numenius arquata) and oystercatcher (Haematopus ostragleus), were comparatively low on sites in the Severn. (3) Within most estuaries, bird densities correlated with the densities of one to three widely taken prey species. When data from all estuaries were combined, bird densities correlated with the densities of one or two prey species, the polychaete worm Nereis diversicolor providing the best correlation in five cases. The densities of the larger birds were correlated with the densities of their larger-sized prey. Allowing for the effects of prey density, sediment parameters correlated additionally with the density of bar-tailed godwit and redshank and exposure-time correlated with the density of oystercatcher. (4) When the effects of these variables were taken into account, bird densities on sites on the Severn were only significantly different from those on other estuaries for one or two species. Therefore, the low densities of larger wader species on the Severn can be explained by the low densities there of the larger-sized polychaete worms and bivalve molluscs upon which they feed: there was no reason to invoke a special unidentified factor in the Severn to account for the low densities of these species. It was concluded that other estuaries provide an analogue with which to predict post-barrage bird densities on the Severn from predicted densities of their prey. (5) By holding back the ebbing tide, a barrage would substantially reduce the area of intertidal flats available at low water for the birds to feed. On the other hand, the productivity of the estuary, and thus the invertebrate density, could increase in the generally more benign post-barrage environmental conditions. Wader densities would have to increase approximately twofold to allow the same overall numbers of birds to remain post-barrage as occur on the Severn at present. Provisional estimates are given of the increases in prey density required to allow bird densities to increase by this amount. With the -exception of the prey of dunlin, the required values fall well within the ranges of densities found in other estuaries, and so could in principle be attained in the post-barrage Severn. (6) An attempt was made to derive equations with which to predict post-barrage densities of invertebrates from easily measured, static environmental variables whose post-barrage values could themselves be predicted. Although the densities of six of the ten most important prey categories correlated with static environmental variables, such as the particle-size and organic content of the sediment, the fact that a site was in the Severn had a significant additional effect on invertebrate density in seven cases. This suggests that there is a special feature of the Severn, probably one associated with its highly dynamic nature, whose effect was not captured by measuring static variables. This factor must be identified, and the effect of a barrage upon it evaluated, if the post-barrage densities of invertebrates, and thus of the birds, are to be successfully predicted.