Romney Marsh: the Field-Walking Evidence
Introduction
A programme of fieldwalking was carried out on Romney Marsh between 1989 and 1991. The work was almost exclusively confined to the Romney Marsh Level, which is only one of the marshes which make up the geographical region known as Romney Marsh.
Fieldwalking involves the systematic collection and analysis of artefacts from the surface of cultivated fields. Former patterns ofsettlement and land use can be detected by scatters of pottery and other material and even by variations in soil colour and texture. The purpose was primarily to test the usefulness of fieldwalking as a research tool and only secondly to gain new, specific information about Romney Marsh. Therefore a variety of methods were used in order to produce comparative data and the presentation of results is not as straightforward as it might have been; but those results have nonetheless proved to be very exciting.
Fieldwalking is particularly useful in areas which lack other sources of historical evidence e.g. where there are heavy clay soils which are unresponsive to aerial photography, or where documentary and other archaeological evidence is scant, or where these sources provide conflicting evidence. In Essex and Northamptonshire fieldwalking has consistently revealed evidence of Romano-British settlement in areas of heavy soil once thought to have been first cleared and colonised in the late Saxon and Medieval periods (Rodwell 1978; Foard 1980; Williamson 1984).
Romney Marsh has always been a dynamic environment and consequently the development and succession of human settlement and land use has been complex and difficult to chart. For a long time it has been generally believed that parts of the Marsh were sufficiently elevated and well drained by the first century to allow some Roman settlement to take place, seasonal or otherwise. Roman finds to date have been confined to the areas of"older" decalcified soils as compared to"newer" calcareous soils, according to the definition of the Soil Survey (Green 1968). Documentary evidence in the form of Saxon land charters from the seventh century onwards confirms activity in the region but archaeological evidence of Saxon and Early Medieval occupation has hitherto been sparse. Romney Marsh is well documented in the Middle Ages but again details of the extent and nature of settlement are lacking and little is known about the timing and progress of later depopulation. From the 16th century on, the Marsh became famous for sheep keeping and much information about the history of the area was sealed under a carpet of permanant pasture which lay undisturbed until the 20th century.
Today Romney Marsh is predominantly an arable farming area but much of the land has been under the plough for less than 30 years and the archaeological material in the ploughsoil has only been recently disturbed. The Marsh therefore provides excellent scope for fieldwalking. Many fields have yielded large numbers of pottery sherds, building debris is clearly visible on the fields' surface and, because of the extremely flat landscape, artefacts remain well clustered at the site of deposition. Best results are obtained when fieldwalking takes place on bare land in winter where the soil surface is well weathered. Initially land was chosen for survey on this basis, that is, only those fields which provided ideal conditions for fieldwalking. This severely limited the amount of land available since so many crops are autumn sown. Subsequently work continued in units based on kilometre squares selected from the O.S. 1 :25000 map, adding to the area already walked and ensuring that fieldwalking took place in a variety of conditions and situations, which were always carefully recorded. These included areas of land on both "old" decalcified and "newer" calcareous soils (Fig. 5.1).
Altogether 265ha of land were surveyed in detail. Of this l19ha was surveyed intensively using a grid system and 145ha was line walked. Statistically this is a small sample. Romney Marsh Level is an area of around 10,000ha and the whole Romney Marsh region amounts to nearly 27,000 ha; but it is a sufficient sample to provide a clear indication of the extent and character of past activities in the region.
Where the grid system was used, fields were divided into 25m squares aligned with the national grid. Each square was numbered and walked through twice so the lines walked were12.5m apart within the grid. Finds from each square were bagged separately. This method enabled the density of pottery scatters to be plotted accurately on the field plan thereby locating occupation or industrial sites of different periods and also establishing background manuring levels. It is generally agreed among archaeologists that stray pottery sherds can be deposited by manuring, and examination of their distribution enables areas of past arable land use to be identified (Hayfield 1980; Le Patourel 1980; Wilkinson 1982; Brown 1987).
When fields were sown with a crop they were line walked using a series of lines 12.Sm apart (except in the case of three fields where the lines were 25m apart), so apart from these exceptions coverage remained consistent with the gridded areas but the lines were orientated in the direction of the sown crop. When line-walking, isolated sherds were noted but not collected. Significant concentrations of pottery were marked on the field plan, then rewalked and samples collected and bagged according to line number. The usefulness of line-walking is limited because the distribution of artefacts within the field cannot be accurately plotted, but it did enable the location and dating of settlement sites which would not have been discovered otherwise.
As well as this area of detailed survey where systematic fieldwalking was carried out, a number of chance archaeological finds have been made during the course of other fieldwork in the area. These were added to the record as incidental finds and range from a single medieval jug handle to a substantial site yielding more than 500 sherds of pottery. This meant that altogether there were finds from a total of 19 parishes as shown in Figure 5.2 (including some former parishes; the 19th century parish boundaries were used throughout this study but some of these have subsequently been reorganised and the names lost). Altogether archaeological finds have been recorded at 103 locations in the Romney Marsh area, 95 of which are within the Romney Marsh Level, and almost all were previously unknown.
Decisions about occupation site definitions are inevitably problematical. In this study distinct concentrations of 25 or more pottery sherds in an area of 25m2 were recorded as a site; concentrations of less than 25 sherds per 25m2 but where there was other evidence of occupation (such as building materials or food refuse) were recorded as possible sites. All other isolated finds were recorded simply as findspots. This is essentially a flexible definition because of the differing conditions under which fieldwalking took place, the varying lengths of time each field has been under the plough and quantitive differences in the material culture of the different historical periods.
Using these criteria, within the area of detailed survey there were 46 sites and possible sites and 31 findspots recorded. And altogether there were 82 sites and possible sites, and 71 findspots. Many of these showed continuous occupation through several periods.
It was also possible to record the extent of medieval arable cultivation within the area of detailed survey by plotting the background pottery scatters that were deposited when this land was worked and manured in the Middle Ages. There was evidence of manuring over 181ha of the detailed survey area. That is 68% of the total area (265ha). Where the grid system was used it was possible to break this figure down even further. Differential manuring levels were revealed by counting the density of the background pottery scatter. At this stage it is only possible to speculate about the significance of the different manuring levels, but these figures provide an interestingbasis for discussion and clearly more work needs to be done.
Settlement Dates Indicated by the Pottery
A detailed fabric analysis was not within the scope of this project. Instead pottery was divided into broad groups of wares for dating and recording. The main purpose of cataloguing the material was to suggest dating for settlement in the area.
A gazetteer of all the collected material has been compiled and the distribution of sites and findspots has been chronologically mapped at a scale of 1 :25000 as part of a series of map overlays. This has enabled the information gained from fieldwalking to be studied within the context of its relationship to other significant landscape features in the area, and to the areas of "old" and "new" marshland as defined by the Soil Survey.
Fig. 5.2. Romney Marsh Level showing mid-19th century parishes and their detached portions. Source: Tithe maps of Romney Marsh parishes c. 1840.
A total of 17,378 sherds of pottery were collected altogether, as shown in Table 5.1.
Table 5.1. Quantity and Dates of the Pottery Sherds
Roman
Roman pottery was found at 20 new locations in the Romney Marsh Level (Fig. 5.3). Previously Roman artefacts from Romney Marsh have only been found beneath the present soil surface on the "older" decalcified soil areas, usually as a result of some accidental disturbance. For example, atJesson Farm, St.Mary-in-the-Marsh, habitation on an older surface was proved by pottery debris and charcoal found at depths of between 37cm and 75cm (Green 1968,27). An extensive saltworking site in the former Romney Marsh parish of Snave first came to light in this way as a result of ditch maintenance, found by the author in 1988. Briquetage from this site is on display at Folkestone Museum.But the fieldwalking has subsequently discovered surface Roman finds where there was no sign of deep soil disturbance. Most surprising of all was the fact that fieldwalking discovered Roman pottery on "newer" calcareous soils at five locations in Romney Marsh.
Fig. 5.3. Roman sites and find-spots.
Most Roman finds were from the first and second centuries and the bulk of the material was coarse grog tempered "Belgic" or Romano-British pottery (similar to Patchgrove and East Sussex Ware) but there were also some finer wares including Black Burnished Wares, Samian and pottery of Upchurch, North Kent type. The earliest pottery was found on "old" soil at Burmarsh. This dated from between c.BC 50 and 25 AD which is strictly speaking Late Iron Age, but as only two sherds were recovered they were not recorded as a distinct category. The latest finds from the period, pottery from the third century, was found at four locations - also exclusively on "older" decalcified soils.
There was evidence of extensive saltworking in the Snave area along the margins of the "old" soil area there. At one site over 200 fragments of briquetage were collected. Today this area is 7 km (nearly 5 miles) from the sea.
Saxon (pre 1050)
Late Saxon pottery dating from the 10th and early 11th centuries was found at 10 locations on the Romney Marsh Level, six of these being on "old" decalcified soils and four on the "newer" calcareous soils (see Fig. 5.4). Most of these locations also had early medieval pottery and many of these early sites could be distinguished on the ground by the presence of a spread of dark soil. Exclusively medieval sites tended to be much more compact and could be located by clearly visible concentrations of shingle, tile and other building materials. These features represent different methods of building and different modes of living. The Saxon pottery included both shell and chalk filled fabrics as well as other coarse sandy and gritty wares.
Fig. 5.4. Saxon sites and find-spots.
Dating of the Saxon pottery has been cautious. There is clear evidence of 10th century occupation even on the "new" soil areas in Newchurch parish but some of the pottery might be eighth or ninth century. Recent work in other areas has resulted in coarse shell tempered wares being dated increasingly earlier, and it may well be that some of the pottery currently being recorded in the Early Medieval category here will subsequently prove to have originatedbefore the Conquest. The only previouslyknown Saxon site on Romney Marsh, Sandtun at West Hythe, produced seventh to ninth century material (Woodcock 1988,185).
Early Medieval (1050-1250)
Quantities of Early Medieval pottery were found all over the Marsh - 7114 sherds of pottery were collected which dated from between 1050 and 1250. A total of 64 Early Medieval sites and possible sites were recorded and altogether Early Medieval pottery was found at 83 locations spread indiscriminately across the Marsh (Fig. 5.5).
Fabrics ranged from coarse shell and flint tempered local wares to hard sandy pottery much of which originated from Canterbury. There were also finer French imports including northern French sandy grey pottery and Beauvaisis red painted ware. Scarborough phase 1 fabric dating from between c.1135 and 1225 was also found.
Fig. 5.5. Early Medieval sites andfind-spots.
Medieval (1250 - 1400/1450)
A total of 5688 sherds of Medieval pottery were found from 85 locations on the Marsh (Fig. 5.6). A total of 52 medieval sites and possible sites were recorded. Local sandy fabrics accounted for much of this pottery but large quantities of Pink Wealden ware were found all over the Marsh, and also significant amounts of Rye and other Sussex pottery. Pottery was also continuing to reach the Marsh from a variety of locations outside south-east England presumably through the port of Romney. Non-local pottery at this date included Scarborough, Surrey and London pottery and Saintonge ware from south west France.
Later Medieval (1400/1450- 1500/1550)
There was significantly less pottery collected (only 2151 sherds) that could be specifically dated to the late 15th and early 16th centuries. Later Medieval pottery was found at 52 locations on the Marsh but only 15 distinctly Later Medieval sites and possible sites were recorded (Fig. 5.7). Much of this pottery was Wealden. Examples from Hareplain, an early 16th-century kiln at Biddenden, Kent, and Spilstead, East Sussex were found; also Later Medieval pottery from Tyler Hill, Canterbury was found at two locations. Small amounts of Later Medieval pottery from further afield came from Surrey and London, along with a few early continental imports.
Fig. 5.6. Medieval sites and find-spots.
Post Medieval (1550-c. l 900)
Post Medieval pottery (totalling 1590 sherds) was found at 63 locations on the Marsh. Much of this came from the manuring of small pockets of old arable land although at some earlier sites occupation continued into the 16th and 17th centuries before they were abandoned. These can be confirmed by reference to the manuscript maps which belonged to the Romney Marsh Corporation and which are now held at the Centre for Kentish Studies (Eddison 1988, 186). Some good examples of continental 16th century pottery have been found on the Marsh, including both stonewares and tin-glazed earthenwares. Post medieval pottery from London was also found, and quantities of 18th and 19th century red earthenware Wealden pottery.
The Sources of the Pottery
Where did it all come from? No kilns were found within the surveyarea on the Romney Marsh Level and yet most of the material must have had a local origin. In the Medieval period the nearest known kiln site was at Potters Corner, Ashford (Grove and Warhurst 1952). Unfortunately it was not possible to identify and isolate material from this source at the outset of the project although this kiln, located only 15km (9 miles) to the north, almost certainly must have been supplying Romney Marsh in the late 12th and 13th centuries. Subsequently examples were noted from around 20 Romney Marsh sites although too late to be recorded as a distinct category within this study. The final total is likely to be considerably higher, and it is now probable that this kiln was an important ceramic source for Romney Marsh in the Early Medieval period. It is also very likely that there was pottery production in or around the town of New Romney in the Middle Ages but no conclusive evidence has yet been found.
Fig. 5. 7. Later Medieval sites andfind-spots.
Non-local pottery from Canterbury, London, Rye, Scarborough, Surrey, Sussex and the Weald was clearly identified. Also imports from Belgium, France, Germany and the Netherlands were found, as shown in Table 5.2.
Table 5.2. Sources of non-local Pottery found on Romney Marsh.
because rates of decalcification could depend on a large number of variables.
The location of Roman finds on the "newer" calcareous soils on the Romney Marsh Level is of crucial importance to furthering our knowledge of this period (see Fig. 5.8). This material was dated exclusively to the first and second centuries and there were no substantial concentrations here, only odd sherds. This may be because cultivation has not disturbed the soil deeplyenough to turn up deposits of this date in any quantity yet, or it may be because the original deposits were scant. In other regions evidence of this kind might be interpreted as Roman manuring, indicating arable land adjacent to settlement areas (Williamson 1984). Here, in this location deposition is more likely to have come from other activities, probably seasonal or temporary exploitation of this central area of calcareous marshland, although the possibility of cultivation cannot be completely ruled out. Clearly these finds suggest favourable conditions here in the first and second centuries, both environmentally, and economically and socially; with the existence of a land surface at that date which subsequently became inundated again, perhaps for centuries.
This does not necessarily contradict the accepted interpretation of the Soil Survey but cautions against adopting too rigid an approach. Fieldwalking has helped to bring the picture into a clearer focus, but the picture remains consistent with existing evidence of changing conditions in the later second and third centuries with the area probably adversely affected by raids and rising sea levels. The shore fort at Lympne was abandoned in the mid fourth century and excavation has revealed the existence of a shoreline 2m below the present level of the Marsh. It has been suggested that rising sea levels caused
South-west France Netherlands
4 silting which continued through the fourth and fifth
3 centuries, possibly also temporarily affecting some of the
"old" soil areas especially in the east (Cunliffe 1980,259;
Note: This does not include imports dating from the Roman period.
Analysis and Interpretation
The pottery recovered by fieldwalking has provided a wealth of new information about the history of Romney Marsh; about the evolution of the Marsh landscape and about the people who lived and worked there for the last 2000 years; but how should this evidence be interpreted?
Roman Settlement
Previous research has taken the distinction between the two soil types to indicate the approximate divide between dry land and tidal marsh in the early centuries of the Roman period. Christopher Green warned in the first Romney Marsh monograph (Green 1988,169) that this interpretation might prove to be "an oversimplification",
1988). The only pottery dating from the third century was found at Bonnington, Bilsington, Orgarswick and Snave, all "old" soil and presumably less vulnerable areas, and no artefacts dating later than the mid third century were found. A high level of activity over the whole area in the first and second centuries is emphasised by evidence of a flourishing saltmaking industry around the edges of tidal creeks on the margins of "old" soils west of Newchurch in Snave parish.
Saxon Settlement
Opinions about the Saxon period are harder to reconcile. In 1968 Green concluded that the "new" calcareous land must have been colonised by the 10th century (Green 1968,16,36). Cunliffe (1980,259) accepted this date in the 1980s but matters did not rest there. In 1989 Witney used the Domesday valuations to argue that the draining and settlement of the area around Newchurch must have taken place in the period immediately following the Conquest (Witney 1989,37). However Tatton-Brown found evidence to suggest that a church at Newchurch (not the present one) was probably built in the mid 10th century (Tatton-Brown 1989,255). A church was recorded at Newchurch in the 11th century in the Domesday Monachorum (Douglas 1944,78) and according to Brooks (1988,94) this new church was probably in existence by the date of the establishment of the hundredal system (generally believed to be around the mid-tenth century) since Newchurch gave its name to one of the four hundreds whose territory lay entirely within the Marsh.
Fig. 5.8. Suggested Roman topography.
Aerial photographs of the calcareous marshland reveal a patchwork of tiny irregular fields within a framework of boundaries clearly extending from and aligned to those on the "old" soil areas. Thus the direction of colonisation is clear and Saxon finds discovered by fieldwalking have helped to date this process of inning and colonisation as the tidal area associated with the Limen channel contracted.
Fig. 5.9. Suggested Saxon topography.
Again it is the locations of the finds that are so important (Fig. 5.9). Pottery from the eighth to 10th century was found on "old" soils, on marginal areas between "old" and ''new" soils and also in the centre of the "newer" calcareous land close to Newchurch. This is clear new evidence of settlement of the area at that date, and indicates that the process of inning and reclaiming must have been well under way. If the Limen was still functioning as a tidal channel in the 10th century then it must have been relatively small and well controlled. Furthermore, it appears that a bar of dry land could have formed across the mouth of the lagoon area parallel with the northern shore of the Limen river at an early date. This would have eventually cut off and left behind it to the south an area of marshy poorly drained land. Tithe maps show that this area of calcareous marshland south ofNewchurch, east of Ivychurch and west of St. Mary-in-the-Marsh contained a number of small detached parts of parishes suggesting a probable late date for reclamation here, which hitherto has been difficult to explain (see Fig. 5.2).
Fieldwalking evidence has therefore helped towards understanding the chronology of settlement of the "newer" calcareous soil areas of Romney Marsh by distinguishing and locating differential phases of activity.
The scenario leaves us with many tantalising questions. The presence of salt houses and a fishery at Eastbridge at the time of Domesday suggest a salt water channel still in existence there by the 11th century (Morgan 1983,9, 10). This might have been a small tidal creek, the last trace of the Limen river, or alternatively this industry might have been located along the Dymchurch shoreline where Eastbridge Manor is known to have also held land (Teichman-Derville 1936, 103).
But what of the significance of the name Eastbridge? And the place-name Westbregge located in the west of Newchurch parish close to the border with Ruckinge? Were these crossings over the Limen from north to south? Or east to west crossings linking old land to a relatively new but firm, dry, settled island of land centred around Newchurch?
Medieval Settlement
In the Medieval period a few places showed continuity of occupation over the whole period but it was the Early Medieval period ( 1050-1250) which produced the greatest quantity of pottery and the most sites. This suggests that population in the region had reached its highest levels before 1250 and that thereafter it declined only gradually until c.1450, after which date a sharp decline set in. Probably the population level of the Marsh was more closely linked to the successful functioning of the Port of Romney than to any other single factor (Reeves, forthcoming). Adverse weather in the second half of the 13th century and famine and disease in the 14th century simply compounded existing difficulties. Later Medieval depopulation might not have been as abrupt as the pottery evidence indicates because other factors need to be taken into account, most importantly the possibility that less pottery was deposited anyway at this time because contemporary domestic utensils were being made of other materials. There were a few new and exclusively Later Medieval sites but generally the trend of decline is unmistakeable.
Taken as a whole, occupation sites of the medieval period are spread fairly uniformly across the landscape. The picture is overwhelmingly one of an area of scattered or dispersed settlement with a population far higher than that of today. Calculations based on the area of intensive survey show an average of one site or possible site for every 6ha (or 15 acres) of land.
Not surprisingly this level of population required a great deal of arable land and fieldwalking evidence shows that 68% of the detailed survey area was manured during the medieval period (see Table 5.3).
If the land within the Parish of Orgarswick is excluded then this average rises to 82%. Fieldwalking shows Orgarswick to be exceptional in two ways.This was a tiny parish of only l 62ha (just over 400 acres) much of which belonged to Canterbury Cathedral Priory throughout the Middle Ages.There were no occupation sites on the land known to have belonged to the monks but settlement was notably clustered around the borders of their land. Secondly, only 18% of the known monastic lands were manured compared to 4 1% of the rest of the surveyed land in that parish. Documentary records from Christ Church show that dairying and sheep keeping were important on the Manorof Orgarswick (Smith 1943, 153) and fieldwalking confirms that the monks were practising specialised farming on Romney Marsh in the Medieval period, pursuing a distinct land-management policywhich contrasted with the activities of the surrounding peasant smallholdings. Here fieldwalking evidence balances a potentially distorted historical picture which can emerge when documentary evidence alone (and especially ecclesiastical evidence) is used as the sole basis for inference about the wider medieval population and its activities.
The extraordinary stability of the Romney Marsh landscape is demonstrated by plotting sites from the medieval period onto maps at the I :25000 scale (Figure 5.10 is an example of this). Sites fit comfortably into the present day network of lanes and field boundaries. They are where you would expect them to be - a few metres from an existing field gateway; beside the start or finish of a public footpath; sometimes grouped along the edges of lanes or trackways. Where a site is found in the middle of a field, reference to older manuscript maps or aerial photographs usually show an adjacent but now lost field boundary. The remarkable permanence of the framework of the Romney Marsh landscape - a landscape which feels ancient and in map outline looks ancient - is thus confirmed.
Table 5.3. Medieval manuring over area ofdetailed survey: summary of statistics.
Source: Great and Little Bilsington, Walsfoot and Sheatie Waterings (part of) 1653. Thomas Boycot.
Centre for Kentish Studies S/RM P2/l.
Fig. 5.10. Land in the north-east of Newchurch parish beside Honeywood Lane. This fine 17th century manuscript map shows Mr Chayer's house (arrowed) and lands. No trace of the house is found on later maps. but fieldwork has confirmed a site here dating from the Medieval period. Field gateways still exist by the site, and the current OS 1:25000 map shows a public footpath connecting it with another site to the east. Although the internal field boundaries were re-aligned in the 1970s, the original layout of the land can still be traced and the essential framework of the landscape remains unchanged. Source: OS 1:25000 1982 edition (selective extract).
Post Medieval Settlement
In the Post Medieval period some fine examples of imported pottery were found. Documentary sources give the impression of an unhealthy isolated rural backwater peopled only with a few "lookers" working for absentee landlords at this date (Lambarde 1926, 180-181) (Hasted 1972,469-470). There can be no doubt that the Marsh became increasingly depopulated through the 16th and 17th centuries, although the pottery evidence shows that some successful yeoman farmers must have remained resident there.
Conclusions
This project has demonstrated the potential of fieldwalking for the study of settlement and landscapes over the ages. It has revealed that what is today a rather bleak and empty tract of agricultural land was once peopled by a thriving populous community in the Middle Ages; and how even a thousand years before that, people worked and lived on the Marsh, exploiting and subtly but permanently shaping the land.
Fieldwalking has shown what an extraordinarily valuable resource the Romney Marsh landscape is, not just for local enthusiasts. So much is unique and exceptional about the Marsh but the implications of this "uniqueness" are of national importance.
Above all, this project has provided pointers for future work, highlighting areas of special interest where further study, more detailed field survey and archaeological excavation can most fruitfully take place.
Acknowledgements
This work was undertaken as part of the research for a doctoral thesis funded by the British Academy. It was also supported by a small grant from the Romney Marsh Research Trust.
For help with pottery identification, my thanks are due to Jessica Vale of Kent County Museum Service, David Kelly of Maidstone Museum, Mark Gardiner from the University of London and especially Nigel Macpherson Grant who gave so much of his own time, and other members of Canterbury Archaeological Trust.
Thanks are also due to Tom Williamson, University of East Anglia, and Miss Eleanor Vollans for help and advice; and to Romney Marsh farmers for their co-operation and support during the initial fieldwork.
References
Brooks, N. 1988: Romney Marsh in the Early Middle Ages. In Eddison, J. and Green, C. (editors), Romney Marsh: Evolution, Occupation, Reclamation. OUCA Monograph 24, 90-104.
Brown, A. 1987: Fieldwork for Archaeologists and Local Historians (Batsford).
Cunliffe, B. 1980: Excavations at the Roman fort, Lympne, Britannia, 11, 227-288.
Cunliffe, B. 1988: Romney Marsh in the Roman Period. In Eddison,
J. and Green, C. (editors), Romney Marsh: Evolution, Occupation, Reclamation. OUCA Monograph 24, 83-87.
Douglas, C.C. 1944: The Domesday Monachorum of Christ Church Canterbury (London).
Eddison, J. 1988: Early maps of the Romney Marsh area. In Eddison, J. and Green, C. (editors), Romney Marsh: Evolution, Occupation, Reclamation. OUCA Monograph 24, 186.
Foard, G. 1980: The recovery of archaeological information by systematic fieldwalking: Research in Northamptonshire and Bedfordshire, in Hayfield C. (editor), Fieldwalking as a Method of Archaeological Research, Directorate of Ancient Monuments and Historic Buildings, Occasional Paper No 2, 34-40.
Green, C. 1988: Paleogeography of marine inlets of the Romney Marsh area. In Eddison, J. and Green, C. (editors), Romney Marsh: Evolution, Occupation, Reclamation. OUCA Monograph 24, 167-174.
Green, R.D. 1968: Soils of Romney Marsh, Soil Survey of Great Britain, Bulletin 4 (Harpenden).
Grove, L.R.A. and Warhurst, A. 1952: A thirteenth century kiln site at Ashford, Arch.Cant. 65, 183-187.
Hasted, E. 1972: The History and Topographical Survey of the County of Kent, vol.8.
Hayfield, C. 1980: Wharram Percy Parish Survey, in Hayfield C. (editor), Fieldwalking as a Method of Archaeological Research, Directorate of Ancient Monuments and Historic Buildings, Occasional Paper No 2, 26-33.
Lambarde, W. 1926: A Perambulation of Kent Conteining the Description, Historie and Customes of that Shire.
Le Patourel, Jean H.E. 1980: Fieldwalking at Collingham, West
Yorkshire, in Hayfield C. (editor), Fieldwalking as a Method of Archaeological Research, Directorate of Ancient Monuments and Historic Buildings, Occasional Paper No 2, 13-19.
Morgan, P. 1983: Domesday Book: Kent (Phillimore).
Reeves, A. (forthcoming): The Historical Development of Romney Marsh: Field Survey, Landscape and Documentary Study, unpublished PhD thesis, University of East Anglia.
Rodwell, W. 1978: Relict landscapes in Essex, in Bowen H.C.and Fowler P.J. (editors), Early Land Allotment, BAR, Oxford, 89-98.
Smith, R.A.L. 1943: Canterbury Cathedral Priory, a study in Monastic Administration (Cambridge).
Tatton-Brown, T. 1989: Church building on Romney Marsh in the later Middle Ages, Arch. Cant. 107, 253-266.
Teichman-Derville, M. 1936: The Level and Liberty of Romney Marsh (Ashford).
Wilkinson, T. 1982: The definition of ancient manured zones by means of extensive sherd sampling techniques, Journal of Field Archaeology, 9, 323-333.
Williamson, T.M. 1984: "The Roman countryside: Settlement and Agriculture in N. W. Essex," Britannia, 15, 225-230.
Witney, K.P. 1989: Development of the Kentish marshes in the aftermath of the Norman Conquest, Arch.Cant. 107, 29-50.
Woodcock, A. 1988: Gazetteer of Prehistoric, Roman and Saxon sitesinRomneyMarshandthesurroundingarea.InEddison,J. and Green, C. (editors), RomneyMarsh:Evolution,Occupation, Reclamation.OUCAMonograph 24,177-185.
