Flow-parallel linear bedforms, consisting of ridges and runnels have been reported widely in soft and consolidated cohesive sediments within intertidal and subtidal coastal waters. Less frequent are reports of these systems incised into soft bedrock. Despite their common occurrence there are few data concerning the hydrodynamic controls on the initiation and maintenance of ridge–runnel systems. Rather, several hypotheses have been advanced to account for them. In tidal environments a theory based upon the evolution of ‘streamwise vorticity’ transverse to the main flow direction is favoured, and has been supported recently by detailed hydrodynamic data obtained from above a ridge–runnel system on an intertidal zone of the Severn Estuary, UK. In the present paper, a detailed description of the sedimentology of the same ridge–runnel system is provided and interpreted within the framework of the ‘streamwise vorticity’ hypothesis. This objective is accomplished by measurements of the morphology and sampling the bulk characteristics of the sedimentary ridges and runnels but also with reference to ridges and runnels incised into soft bedrock beneath the sediment body. Grain-size composition, sediment compaction, radionuclide distributions and elemental distributions and X-radiography are used to describe the sediment structures within ridges and their inferred histories of development. These descriptions are interpreted with respect to the known hydrodynamic climate and the deposition and erosion behaviour of the sediments in laboratory experiments using an annular flume. Runnels contain liquid mud at low tide but contain very little fine gravel and shell debris. In contrast, fine suspended sediment is present at high concentrations in the flow. Thus fluid evorsion, aided by a dense suspension of fine sediment that impacts the mud bed, must be the primary tool to initiate and maintain the runnels, rather than scour by a coarse bedload. The ridges are composed primarily of fine silt or sand laminae some of which are approximately horizontal but most of which are slightly domed upwards such that they drape the ridge topography. This relationship between lamination and form demonstrates that the ridges are distinct accretion units and not residual masses of un-eroded planar-laminated mud incised by runnels. Laminae thicken, or thin and pinch-out, or are truncated, laterally at the sides of the runnels. Truncation occurs probably by damped spanwise turbulence structures which were recorded at this location. A notable characteristic is that the mud in the upper portion of the ridges is of low density and low compaction whilst the sediment in the lower part is more dense and well-compacted; often the two units are distinct, being separated by major reactivation surfaces; the latter indicative of occasional stripping of the surface sediments from above a residual ‘core’ at the base. The presence of coarse-grained laminae and lenses and minor reactivation surfaces throughout the sediment body indicates that high energy conditions, possibly other than the Neap–Spring cycle, are also important in developing the ridges. In this respect, consideration of the erosion potential of deposited sediment, using field and flume data, shows that the relatively low density, low-compaction mud constituting the crests of the ridges is not readily eroded by the present-day Neap–Spring tidal flow regime. Further, for the same flow regime, when the shear stresses on the bed fall below critical for suspension, potential deposition rates are very low, which supports the interpretation of the coarse-grained layers and reactivation surfaces as indicators of the significant role played by additional high-stress events, such as storm surges, in ridge-erosion and rebuilding.