The effects of intensive hydraulic cockle dredging during one tidal cycle on the sediment and infaunal community compostion of two areas of sediment flat were studied during the period June 1989 - January 1990. The 2 areas selected for the study were on the mid-shore of Lavan Sands, N Wales and on the mid-shore of Blackshaw Flats, Solway Firth. The infaunal community of the Lavan Sands (LS) sites were dominated by the cockle C. edule, the baltic tellin shell M. balthica, the lugworm A. marina and the amphipod C. arenarium, with a variety of small polychaetes and other infaunal species. The infaunal community of Blackshaw Flats (BF) sites was less diverse, and dominated by cockles, baltic tellin shells and the small snail H. ulvae, with a variety of other small infaunal species. 6 sites were established in each area, 3 to act as controls for the 3 experimental sites which were dredged during one tidal cycle during June (LS) or July (BF). The sites were sampled for infaunal and sedimentological analyses on the day before dredging and then one day, one week, 3 months and 6 months after dredging. The data were analysed with the aid of a variety of statistical and graphical techniques. Prior to dredging, there was only a smally variability between the sediment characteristics of the 6 sites, at both LS and BF. However, the abundances of many infaunal species varied significantly between sites and between the control and experimental areas, particularly at LS. This is considered to be due extreme patchiness in the distribution patterns of the infauna, rather than to differences in the environmental characteristics of the sites. Comparison of the experimental sites the day before and the day after dredging found very few differences in sediment characteristics, although there was a small but significant increase in the mean grain size of the sediment at BF. There was no significant difference in the precentage mud content. Immediate effects on the infauna at LS were considerable, with significant reductions in the abundances of most species. The total number of recorded taxa dropped from 27 the day before to 20 the day after. The mean abundance of all individuals was reduced from 7,000 to 3,300 per squ. m, while the precentage reduction in abundance of most individual species was around 50%. However, at BF the effects were less dramatic, with significant reductions of only one species H. ulvae, reduced from a mean abundance of 3.313 to 1.293 per squ. m. Analysis of seasonal trends at all sites over the 6 months after dredging found that the sediment characteristics showed very little variation at either LS or BF. The percentage organic matter of the LS sites did show a small but significant reduction from summer to winter, at both the control and experimental sites. Seasonal variations in the fauna, however, were significant for most species, and at LS there was a peak in the abundance of most species in the October samples. At BF the individual species showed very different trends from each other. Comparison of the control and experimental sites over the 6 month 'recovery period' found that the fauna appeared to have recovered from the efects of the dredging within 3 months, and few effects were visible 1 week after dredging. Although there were significant differences in some faunal parameters at all sampling times, these differences were probably due to natural variation.