We have studied the impact of the bioturbating macrofauna, in particular the lugworm Arenicola marina and the bivalve Cerastoderma edule, on abundances and distribution patterns of total bacteria and of bacteria of selected functional groups in sandy intertidal sediments. The selected groups comprised the colourless sulphur-oxidising bacteria and the anoxygenic phototrophic bacteria, which are expected to occupy small zones at the oxygen-sulphide interface in stable (non-bioturbated) sediments. The presence of a wooden wreck buried in the sediment at 10 cm depth within a large area of intertidal sand flat colonised by lugworms provided a unique opportunity to confront field observations with laboratory simulations. The site with the wooden wreck, which was used as control site, was devoid of both A. marina and C. edule, while the composition of the rest of the zoobenthic community was rather similar to that of the surrounding area. In the field, the density of total bacteria was approximately one order of magnitude higher in the control site than in the natural (bioturbated) site. This can be explained by the higher contents of silt and clay particles (higher surface-area/volume ratio) and higher total organic-carbon contents found at the control site. It appears that the presence of macrofauna affects sedimentation processes, which indirectly influence bacterial dynamics. Samples from the control site have been incubated in the laboratory with A. marina and C. edule added (bioturbated core), while an unamended core served as a control. The laboratory experiments contrasted with the field observations, because it was found that total bacteria were actually higher in the deeper layers of the bioturbated core. Moreover, the populations were more homogeneous (less stratified) and colourless sulphur bacteria were on average less numerous in the bioturbated core. In general, laboratory incubations resulted in a decrease of total bacteria with a concomitant increase of colourless and phototrophic sulphur-oxidising bacteria and thus in modifications of the bacterial community structure. Hence, our results demonstrate that care must be taken in extrapolating results from laboratory experiments (e.g. mesocosm research) to field situations.