Field survey of Romney Marsh proper - interim report
As part of a project designed to evaluate the land as a primary source of historical evidence, a programme of field walking was begun in the Romney Marsh Level in 1989. A grant from the Romney Marsh Research Trust enabled this work to be continued and extended. During the winter and early spring of 1990 sixteen fields totalling 328 acres were intensively surveyed around Newchurch village, either side of Gammons Lane (Newchurch and Burmarsh parishes) and at Orgarswick i.e. land on both younger and older soils. Fieldwalking had previously been carried out only on weathered fallow field surfaces but in 1990 the majority of fields surveyed were sown with cereals. Further fieldwalking took place in the Dymchurch area in the autumn of 1990. Results from fieldwalking during these various states of cultivation and at different seasons are being compared to see how this affects the recovery rate of the surface artefacts.
Of the 328 acres fieldwalked in 1990, just over 100 acres was walked using a 25m grid system. Each 25m square was walked through twice (first south to north, then north to south) along lines 12. Sm apart. Finds for each square were bagged separately. This method enables density of pottery scatters to be plotted accurately on the field plan thus locating occupation and/or industrial sites of different periods. series of Where crops were growing the fields were line-walked using a lines 12.Sm apart following the direction of the sown crop. Significant concentrations of pottery were marked on the field plan, then re-walked and samples collected and bagged according to line number. It was necessary to adopt this method in order to achieve representative coverage since so much of the Marsh is under crops almost all year round with the current trend of ever-earlier autumn sowing. shell, bones, The presence of fire cracked flints, oyster teeth, daub and other building materials were noted but these were not collected, in order to keep the amount of material to be processed down to manageable proportions.
Washing and identifying finds is not yet complete but the results of earlier work seem to be reaffirmed. Significant concentrations of medieval pottery were found in most fields, suggesting approximately one site per twenty acres. These occupation sites were often adjacent to existing roads, footpaths and gateways. In a many places evidence of medieval manuring could be detected by the presence of thin scatter of sherds over the whole field surface. Thus a pattern of dispersed medieval settlement with a population greater than that of today is envisaged and widespread arable cultivation at this early date is indicated. Several sites show continuity of use into the sixteenth century and beyond, but there is a comparative dearth of eighteenth and nineteenth finds in this area.
Most of the pottery recovered consists of local medieval wares including a high proportion of Pink East l,ea l den Wares, Pink-grey ware and other local sandy fabrics of varying degrees of coarseness. pottery has been found. Some Rye and White Painted Sussex Sites containing predominately early medieval pottery (11th to mid 13th c. ) could be distinguished on the ground by the presence of dark soil and a more dispersed character whereas late-and post-medieval sites tended to be more compact and could be located by clearly visible concentrations of shingle, tile and sometimes rock fragments. Early sites were characterised by coarse fabrics tempered with flint, shell and chalk.
Although Roman salt-working sites have been discovered by field-walking in previous years, only two isolated sherds of fine black Roman pottery were found in 1990, at Orgarswick. The farmer had previously found a bronze Roman annular ring broach (identified by Kent County Museum Service) at this location, which is on older (decalcified) soil.
When all the material has been processed, information can be plotted onto maps of the whole region showing chronology and density of sites over the wider area and their relationships with each other and to other significant landscape features. All finds will eventually be deposited with the Kent County Museum Service and recorded in the National Sites and Monuments Record.
Anne Reeves
University of East Anglia