Redshank selected low tide feeding sites on the Ythan Estuary in a manner consistent with the "Ideal Free" model (Fretwell and Lucas 1970). Predictions derived from the "Despotic" (Fretwell 1972) model were contravened. Redshank feeding rates were independent of the density in the mud of their major prey species, Corophium volutator, the size of prey available however, was crucial in determining food intake. High densities of Corophium too small to be taken had an apparent confusing effect, resulting in increased search costs. The shift in the Redshank distribution towards the mouth of the estuary in late autumn, previously associated with decreasing daylength (Joffe 1978), was consistent with a change in the cue, from energy intake per unit time to energy intake per unit feeding cost, used by Redshank to select their lowtide feeding sites. Increase in the proportion of Redshank feeding at low tide, the switch from roosting to feeding in agricultural land at high tide, increase in the degree of fidelity to low tide sites and decrease in the amount of movement about the estuary at low tide all indicate an increase in the pressure on Redshank to feed efficiently. Although no significant turnover in the winter population was detected, a PASCAL simulation revealed that turnover of 3 0% of the population size, or higher, would have gone undetected. Significant turnover was noted during the autumn period. Lower quality Redshank, e.g. juveniles and light birds, were predominant among the autumn emigrants. Severe weather conditions in 1981/82 resulted in a 22% reduction in the winter population, almost entirely through increased mortality. Juveniles and, despite the "Ideal Free" distribution, Redshank on the lower section of the estuary, the least preferred in autumn, were apparently more vulnerable to severe weather conditions.