Cookies

We use cookies to ensure that we give you the best experience on our website. You can change your cookie settings at any time. Otherwise, we'll assume you're OK to continue.

Durham University

Margins of the East Fen: Historic Landscape Evolution

Appendix 2.2.A

There is broad agreement that between Roman times and the mid-seventeenth century there was salt-making along the coast of Lincolnshire. There is also evidence for Bronze Age and Iron Age salting, with the latter very likely subsumed into the Roman industry. In the course of accumulating the archaeological, documentary, aerial photographic and ground observational evidence for the process, it has emerged that there was no one single method employed to produce salt and this Appendix is an attempt to summarise the technologies used at various periods, with an emphasis on what remains today as raw material for further investigation, with special emphasis on landscape features.

A number of contextual features need examination. The first is to recall that the current outline of the coast is the product of both accumulation and erosion in the post-Roman era. Between Grimsby and Skegness, the picture has been largely one of erosion to the point that some remains of coastal features of the last 2000 years (both natural and man-made) have been submerged. Around the Wash the reverse is generally true, since reclamation has been a constant theme. The ribbon of post-sixteenth century land landward of the present shore is simply the most recent set of land-takings of the kind found in a wetland embayment that might have stretched up to the foot of the Wolds at the end of the Roman era. Reclamation and salt-making are likely to have been affected by changes in sea-level (usually referred to as Relative Sea-Level, RSL) though exactly which way is uncertain. A rising sea-level may have brought more storms and pushed water and silt over installations but it may also have accelerated the deposition of salt-laden silt on salt-marshes. So careful evaluation of local stratigraphies is always necessary, the more so since recent work in places like Romney Marsh and the Low Countries has emphasised that ‘major events’ such as the ‘Late Roman incursion’ of RSL had varying local results and cannot easily be applied to whole regions.1

No formal classification of salt-making methods and their landscape consequences has been formulated and here, at any rate initially, a mixed-criteria approach is taken which uses both location and salt water treatment procedure as a basis for distinguishing types.2 Maybe by the end of the piece, some more logical taxonomy can be proposed but no breath need be held.3

North-East Lincs Coast

Aerial photographs of the extensive mounds of waste in the northern section of the Lincolnshire Marsh between Humberston and North Somercotes has led to extensive survey and excavation of the salt industry. The investigations have been much helped by a map of Fulstow and Marshchapel dated to 1595. The overall conclusion was that seawater was channelled into storage pits at the site; somewhere nearby this was boiled, probably on open lead pans which have not survived. The pans may have been raised on clay supports though the alternative process of filtration through mats on wooden trays is also consistent with the archaeological evidence. The waste mounds start out as irregularly rounded heaps but these are gradually incorporated into field boundaries that have a strong linear shape (east-west in this case); their history is well described on the 1595 map:

The round groundes at the Easte end of Marshchappel are called maures and are first framed by laying together of great quantities of moulde for the making of Salte. When the maures grow greate the Salt makers remove more easte and come nearer to the Sea and then the former maures become in some fewe years good pasture groundes.

Although direct dating evidence is lacking, it seems that the main area of mounds is seawards of a bank (currently occupied by the A1031) and its relationship to settlements suggests a post-DB date and a terminal phase not long after 1600, with the 1571 floods as a great disincentive to continuing a largely uneconomic industry. However, excavations at Marshchapel itself also revealed the water management features of a Saxon saltern, probably of the tenth century and which had been in use for perhaps 200 years: its ditches appear to feed or feed off a palaeochannel that runs through the middle of the site. The excavators have compared these details with a fourteenth century saltern at Parson Drove in Cambridgeshire, so the general method seems to have been long-lasting.4

From Saltfleet to Mablethorpe has been identified as another landscape region but one without much obvious evidence of salt-making; the churches stand on mounds which might have been salt waste but then again, the surrounding landscape may have shrunk. A brief scan on GE suggests that a close examination of aerial imagery might be productive.

Caravan Paradise

Were the sea to invade the coast and Marsh from Mablethorpe to the northern limits of Skegness, it would wash up against the Wolds a strand-line of debris of tsunami-like proportions with the outstanding feature of being mostly composed of caravans, both mobile and supposedly static. The relevance is that these developments cover many areas which in aerial imagery might contribute to the history of salt-marsh in this southerly block of the Lincolnshire Marsh. Archaeologically this is particularly rich in LIA and RB salterns but they are usually covered with 3 - 4m of marine clays and silts and so not visible from the air. They have mostly been discovered in section alongside drainage channels.5 The final stage was boiling in open pans held up on clay supports; the hardened clay remains are usually the most common finds on a saltern site and are known collectively as briquetage. Maps of these discoveries show a concentration in the area between Chapel St Leonards and Skegness, with outliers as far south as Croft and Wainfleet All Saints parishes.6 It has been rare to find the water management features for this kind of saltern in this area, but the notion of boiling a saline concentrate in open pans using wood or peat as fuels has widespread acceptance as a late stage in the process. Roman connections have been stressed in the argument that ‘Salinae’ on Ptolemy’s map of Britain related to an area round Skegness, perhaps even a Roman town now vanished beneath the sea.7 At Hogsthorpe there is a Bronze Age saltern and others may have been covered by later alluvium or eroded away by RSL rise.8 Although the local evidence is sparse, analogous areas further south (in Cambs and at Morton Fen) usually have a channel leading to a pond where preliminary evaporation might take place. In some places it looks as if tidal creeks are modified to supply a settling tank. Here too, it is asserted that the turbaries used to fuel the boiling phase have silted up and that this infill has been left as ridges by the shrinkage of the peat. But in medieval times, ‘extensive turbaries were associated with each village, but, with peat wastage, no trace of them remains on the ground’.9

Greva Along the Wash

The presence of salterns between Gibraltar Point and the Welland between Saxon times and the fourteenth century has been attested by excavation, observation (field patterns in the open and on maps) and documentary evidence as in DB and chartularies. Aerial imagery has been less used though it has perhaps been overlooked rather than absent. The key feature is the upper sector of the intertidal zone, which in Latin documents is called the greva, as part of a salina. The English equivalents are ‘sandacre’ and ‘saltern’.10

In the Low Countries there is a third method in which large quantities of peat are dug out from under salty sites (i.e., where fresh-water peats have been overlain by saline environments) and burned. If there is a sea-level rise then very large areas are drowned, not just the medium-size areas like the Norfolk Broads or small ones like the Deeps of the East Fen. Evidence for this method in the fenland seems to be lacking: the Broads and (probably) the Deeps are turbaries, not salt sources.

Using the sandacre as a basis, each salt-maker is assigned a strip extending from dry land (sometimes in the form of the sea-bank) down the salt-marsh towards open water. The strip (which in Latin documents is usually either sandacris or greva) is marked out to delimit the individual’s domain. The plan is to capture as much salt was possible in the top few inches/cm of the substrate: to that end (a) the substrate was raked or harrowed to increase the surface area on which salt might crystallize out from sea-water and (b) the prepared area was confined to the zone inundated for a few days at high spring tides. The saturated marsh was then scraped up with a scoop-shaped sledge-like implement pulled by oxen or horses, piled up on the dry land and covered against the rain; it was known as ‘mould’ or ‘sleech’. Some versions then include a settling pond in which evaporation is taken further but others take the salt-laden silt and boil it in open-topped lead pans over peat fires. This was done inside a boiling house or salt-cote, a small shed with mud walls and a thatched roof. High quality ‘white salt’ was produced by skilled salters, who had to know when to remove the sodium chloride and leave the other salts to be thrown away. A supply of fresh water to pre-wash the mould reduced the amount of the less valuable salts. The saleable product was hung up to dry in wicker baskets. A depiction of the elements of this process is found as Fig 2.2.A.1.

The second method is encapsulated in Fig 2.2.A.2 This takes Roman techniques and develops them but has the same basis in using a creek that fills at high tide. The creek may be ‘natural’ or obviously dug out. From it a main storage pond is allowed to fill and then is isolated by means of a wooden sluice. From the main pond, salt water is led off into filtration units, which used a series of boxes lined with peat as a filter. Saline water collected at the base was then boiled and (presumably) the peat itself was burned in turn. There may be a number of filtration units, (lining with clay was an alternative to peat), and their spatial relation to the storage pond seems to be variable. All this is likely to take place behind a sea-bank with the breach protected by a sluice, though if there is no sea-bank then the highest ground at the upper limit of the salt-marsh is the most likely spot.

In both cases, a great deal of waste is produced. This may be more or less all silt particles or it may have a proportion of other salts such as magnesium sulphate (Epsom salts); it is assumed that this material accumulated near to the boiling sites and that any heaps coalesced with time and formed lobate areas (called ‘floors’) which became good grazing areas as remaining salts were leached out. The best examples of these spread-out mounds are seen in north-east Lincolnshire around Marshchapel. Their environmental impact would have been to cover any natural raised areas beneath them and thus to provide yet another piece of land above the reach of most tides, though probably not of surges and other major floods. Their soils would have remained saline for some time but that condition precluded few land uses and certainly not sheep-grazing. Little consideration has been given to the impact of scraping up the top layer of the salt-marsh, with the implicit assumption that the sea will renew the ecosystem. Without experimental data, no conclusions are possible but if the removals are from areas only infrequently inundated then the chances of a more long-lasting manipulation of the ecology are greater. The upper levels of salt-marsh are usually densely vegetated and hence (a) the plant material will have to be disposed of as part of the salt-making process, either at the sites of growth or at the salt-cote and (b) the marsh surface is left bare so that it could erode more quickly (including wind-blow) but at the same time be easier to work on the next spring tides. It might be worth examining aerial photographs of salt-making areas to see if any traces of such impacts are still detectable, especially since some work has shown that salt-marsh accretion may depend on vegetation growth rather than sedimentation. The last immediate impact is the provision of fuel in the sense of stores of it on site which may well have been one cause of the black stains seen in soils from coastal areas, though often these are assigned to hearth remains.

With minor variations, the bulk of the interpretation of salt-making in this period focuses on the salina.11 Here, an area of foreshore yields salt-laden silt which is scraped up, stored and then boiled within a small hut. The linear bounds of the foreshore are clearly marked with stones, stakes and furrows. The post-boiling waste then piles up in mounds, which, as with the types in NE Lincolnshire, are converted to pasture and arable. The huts (salt-cotes) become ‘tofts’ with cottages and sometimes farms (‘messuages’) and the whole zone is converted from salt-marsh to dry land as the process moves seawards. A holding’s layout forms an elongated rectangle, as further north, and this may become ‘fossilised’ into the field patterns. The boiling of the salt consumes large amounts of peat and charters may link the gift of a saltern to a turbary. Dark soil marks in the field and on APs can be seen as evidence of burning fuels or of stacks of peat.12 High quality salt production is facilitated by the availability of fresh water. Combining all forms of evidence, it seems as if this form of salt-making was practised all the way from Wainfleet Haven to Bicker Haven and beyond, with the clearest remains between Croft and Wrangle and in Bicker Haven. Examples can come from any time within this long period, though Saxon examples are much less frequent than those documented by medieval monasteries; their contribution to the landscape may however have been substantial if indeed the mounds of material near Quadring correspond to continental terpen. i.e., settlement mounds, which were probably abandoned in the early ninth century as nucleation of settlement became more common.13 (The best example is at QUA 33 (TF 171 332)).

The contribution of the monasteries to salt-derived landscape change is best seen in the evidence from Wainfleet St Mary where the Sailholme area was dominated by Bury St Edmunds, with Stixwold a junior partner and small contributions from Revesby; and from Wrangle where only a scatter of private tenants and a minor holding by Dereham Abbey interrupted the stretches of foreshore under the control of Waltham Holy Cross. In both cases, the salterns occupied parts of the estuaries of havens and also were probably heirs to the salterns recorded in DB, though if there were Wrangle examples they are not recorded in that vill, contrasting with neighbouring Leake. The DB Wainfleet salterns are not all unequivocally in that vill. Both Wainfleet (undifferentiated) and Wrangle have entries of wasta est, and in the case of the latter Guy of Craon had 2 carucates of land that were in that condition of account of the action (fluxum) of the sea; Darby hypothesized that ‘waste’ may have referred to a lapse from a previous form of land use.14 Between Wainfleet and Wrangle lies Friskney, which housed a number of Bardney Abbey salterns, though probably along with other tenants.15 No salting was recorded in DB: Friskney was then under the aegis of thegns and not Normans but no scholar has yet suggested this is relevant.

The obvious landscape effect of the salt-making is the creation of the Tofts as a swale of raised land about 0.5 - 1.5km wide running from Wrangle to Wainfleet town. This is most easily seen in LiDAR imagery but the patterns are visible on conventional APs and on satellite views. Stratigraphies suggest a structureless silt with debris from fires and pottery, mostly of medieval age. It is still uncertain whether the tofts are anchored, as it were, on former sand-banks or whether they are totally human-created structures on top of former salt-marsh. But there is probably a minimum of 2m of salt waste. Thus if there is a 13 x 0.75 x 0.002km deposit of material it must be by volume the largest human-created structure in medieval and early-modern England, albeit an inadvertent effort.16 There are, however, less obvious landscape consequences:

  1. That a large area of land now looking like ‘dry’ farmland and urban area was actually created by salt-making. In the twelfth century, one grant of a saltern referred to ‘the increase of land’.
  2. That sea-banks may have been adjuncts to the medieval salt-making. The footpath along the crest of the Tofts called ‘High Street’ may well be the feature of medieval documents called
    ‘sedykstight’, the sea-dyke path. There is a parallel path, not so afr noticed, nearer the sea but not yet at the A52 line which might have been another sea-bank. Hallam thought that ‘High Street’ was pre-Conquest in date.17
  3. In the nineteenth century the O.S. mapped the plentiful ponds that were found in many fields in the region. They concentrate on Toftland. Some of the grants of Wrangle salterns to Waltham Abbey included a pond (tenta in Ransford’s transcription);18 most have now been filled in but on Sailholme (Wainfleet St Mary) there is a small pond connected with a ditch system across the Toftland.

Close inspection of some aerial imagery suggest that there are saltern remains in e.g., the Low Grounds of Wainfleet St Mary-Friskney and possibly north in Wainfleet All Saints and Thorpe St Peter as well. The land between the Tofts and the Old Fen Bank may therefore have been salt-marsh (field patterns are of the shape described by Hallam as representing former salterns), as may parts of Wrangle-Friskney Low Grounds; a tidal inlet north of Wainfleet Haven centred roughly along today’s line of the Bell Water drain is a more radical idea but not entirely impossible.

Altogether, there is a virtually incontrovertible case for land creation/reclamation by salt-making in this stretch of country; perhaps the main remaining question is, how far ‘inland’ was it found?

Late Medieval-Early Modern Salt-Making Along the Wash

Two sources are important here: the excavations at Wainfleet St Mary in the 1980s by McAvoy with subsequent amplifications along Groose Lane now read into the Lincs HER,19 and the 1606 map of part of the foreshore at Wrangle.20 The 1984 excavations focussed on the arcuate mounds near the A52 in Wainfleet St Mary, now preserved as Ancient Monuments and often referred to erroneously as Anglo-Saxon or immediately Post-Conquest. The outstanding findings for landscape evolution are (a) that although the area was probably used in earlier times, the main salt extraction was at the turn of the 15-16th century and (b) that the idea of a sandacre outside a sea-bank had been abandoned in favour of a channel leading to a lagoon, which fed small filtration units. An extract from the HER account summarises the findings of both investigations:

A total of twenty one filtration units were recorded during the excavations at TF 4978 5773. Twenty of these units were all aligned in east-west rows and arranged so that there are vats or filterbeds lying together. Eighteen of the filtration units lay approximately 10 meters apart. The filtration units consist of rectangular filterbeds joined to a circular vat by a short pipe. The collecting vats were circular with a diameter of approximately one meter. They had vertical sides and a domed roof. The roofs where present gave an overall depth of 1.2 meters. The midden that was originally recorded during trial trenching was identified again. This was found to lie at the edge of a lagoon. Analysis of some of the ten waste mounds identified confirmed that the silt that was used during processing derived from a lagoon, that is now proven to have been here. The lagoon fed directly into the filtration units negating the need for a complicated system of sluices and inlets ... Peat charcoal found on site indicates that this was the most likely fuel used in the boiling process. Coal was also recovered and this may have been used both domestically and for industrial purposes. The saltern dates appear to be 15th - 16th century as the majority of the dateable evidence indicates this. Thirteenth century sherds were also recovered ...

South of these mounds, the O.S. One-Inch first edition shows salt waste ‘hills’ all the way to the Friskney-Wrangle boundary and close inspection of the map shows a row of small tofts with buildings at the landward side of the mound zone. These features may be among the sites documented above for the 14 - 16th centuries and look like the descendants of the houses of the workers. At Wainfleet St Mary an outer continuous footpath runs along the landward side of the waste mounds all the way into Friskney and so if it was a sea-bank like High Street then the 1500-era salt-making sites were on the seaward side of it. The arcuate mounds are found only in northern Wainfleet St Mary and in one locality in Wrangle: either they were flattened out elsewhere or there was something singular about the workings at those places. Ploughed-out arcuate structures are visible on some APs and so they have probably not been overlooked.

In his detailed contribution to the work of the Fenland Survey, T.W. Lane published a copy of the 1606 map.21 The original is in colour (reproduced as 1606 map) and monochrome reproductions of it are generally rather faint, lacking some of the detail. The map of 1606 is perhaps most notable for showing no active salt-making at all. The zones between the Low Grounds and the furthest seaward extents of reclaimed land show a road running NE - SW just inside the left-hand margin. This starts at Wrangle Hall (432 504), debouches into a ‘village square’ centred on 438 512 and then turns through 90 degrees to run into the map’s frame. A continuation northwards looks like a lesser thoroughfare but is labelled ‘Waineflett’. Some lanes run seawards at right angles to this lower road; they are one type of boundary of a series of SE-running holdings of varying widths which end seawards at a dotted line on the map. ‘The draine’ just south of a windmill also has ‘The Clowe’ lower in its course, a clow being a sea-gote draining fresh water at low tide and excluding salt at high tides. (‘Simon goate’ is further to the NNE). Several of the rectilinear holdings are called ‘Toft’ and it seems highly probable that at the landward end at least they are co-terminous with the raised silts (as high as 4.7m OD) which carry that name today. The map carries a linear scale and so its main features can be plotted onto modern O.S. maps. The seaward ends of the toft plots, which was presumably the limit of reclaimed land at the time, are shown on the 1606 map by a dotted line. On the modern map this line is about half-way to the present high tide mark but is not the site of a distinct sea-bank, as are the later and parallel reclamations. The seaward edge of the toft land can be detected in the field as a dark shadow just in front of Marsh Farm (TF 447 506) when viewed from the seaward side, a feature confirmed on the 1:25,000 map of the Soil Survey. There is a quite definite rise in the region of Toft Mill (TF 444 516), though here in fact the outer edge of the toftland is at least 800m from the mill, with a seaward edge now merging with the trace of the ditch. So the seaward edge of the tofts, as seen in the landscape, is the demarcation of the last phase of salt-making, probably before 1600 but perhaps after 1550. There is only one occurrence of the crescentic mounds of the type seen at Wainfleet St Mary; the LiDAR shows them at TF 457 526 and there are no ‘hills’ shown on the 1829 one-inch map. Perhaps salt waste was largely tipped back into the sea to be dispersed by the tide.

Although the landscape evidence at Wrangle is clear about salt having been made, it is obscure about the methods used in later phases of the industry. The medieval workings are of the greva type, attested by the many charters edited by Ransford22 and this may well have persisted here until the demise of the enterprise. The filling-in of Wrangle Haven was probably aided by salt-making as well as silting from land and sea.

Were there Sun-Works?

In the making of sea-salt, regional literature and current practice combine to emphasize one method above all others in Europe south of Brittany and in the Mediterranean. This is the use of the sun to evaporate salt water: the sea-water is drawn across a series of small ponds and then fed into larger spaces. All these are shallow and eventually the salt is raked to the side and piled up; if rainfall is likely then it is covered. As evaporation proceeds, the solids tend to gather in a sub-circular coagulation in the centre of the pond. If water needs to be pumped then windmills are the traditional source of power. In landscape terms the result can be seen on aerial imagery such as GE in places like the Île de Ré on the southern coast of Bretagne, at Trapani in Sicily and Sfax in Tunisia, plus a hundred others. This method, says Kurlansky, was introduced to the Mediterranean by North African Muslims, first to Ibiza in the ninth century, to Cervia (near Ravenna) by 965 and to the Venetians by the eleventh century. The techniques were an advance on the regional Roman system of a single shallow pond, where boiling in pottery vessels was sometimes part of the process.23 In general, therefore, sun-works seem to be confined to lower latitudes than Britain but the coastal exploitation of salt in the Baltic is a reminder that if salt-making was widespread this far north then (a) it is likely to be virtually ubiquitous in more southerly latitudes where solar evaporation becomes ever more possible, (b) that in Britain there should be widespread representation of sea-salt extraction (i) wherever rock salt was not a vigorous competitor, (ii) whenever imported salt was more expensive, (iii) in places and at times that an entrepreneur decided that solar evaporation was worthwhile for at least part of the process, and (iv) if cheap fuels (peat, wood, coal) were available for boiling brine.

The possibility of a sun-works in Lincolnshire was raised by David Robinson in his book on the Lincolnshire Seaside24 using an AP of the shore at Sutton on Sea after the 1953 flood:

Another equally shortlived method of producing salt was tried in the late 16th and early 17th centuries at Sutton, Chapel and Ingoldmells: This was the Sunworks, the remains of which are to be seen when storms have scoured sand from the clay foreshore, although most of the rectangular excavations are the result of digging clay to repair post-mediaeval sea banks. A sluice in the protective sea bank allowed a stream of water to flow up a ditch and into shallow pans, 60 feet wide and about 100 feet long, [18 x 30m] which had been cut into the clay. These were sealed off and the water allowed to evaporate naturally for a time. The sea water was then slowly drained into a smaller pan at a slightly lower level where evaporation again occurred, and then into a still smaller and lower pan. By this time, given good weather, the brine was strong enough to be boiled, and the salt hung in cone-shaped baskets on the wall of the boiling house to drain and dry. But by 1640 the Lincolnshire salt industry was finally dead.

No supporting documentation is given for the dating, though he probably espouses the suggestions made by a local historian in 1975.25 The well-known treatise by Brownrigg suggest that in the UK in the eighteenth century sea-water was led into a cistern and then into boiling-houses, largely fuelled by coal. But he describes at length the French system of three sizes of pond through which salt water led, starting with a deep ‘brine pond’ and ending with a fully saturated brine in a ‘salt pit’. The first reservoir is required to be ten inches (sic) deep and the last only an inch and a half. (The linear dimensions of the pits are not given since readers are referred to plans published in the Transactions of the Royal Society; the plans are missing from the on-line scans but the final beds are 15 x 13ft or 4.5 x 4.25m). The final product is raked out into pyramidal heaps which are thatched over with straw.26 So in France during the seventeenth century the sequential-pool sun-driven system is common and bears a strong resemblance to Robinson’s description.

These details give additional interest to some patterns which are evident on aerial imagery of East Lincs. In a number of places, there seems to be a layout of large quasi-square enclosures with thick boundaries and no obvious entrances coupled with an area of small enclosures with thinner bounds. Sometimes the small features are oblong rather than square. Typical dimensions for the large enclosures might be 25 x 35m or 45 x 45m; for the smaller ones as low as 8 x 8m. One example is found immediately west of Wainfleet railway station and is cut by the railway (Fig 2.2.A.3); another takes up much of the land between Wainfleet Bank and St Mary’s church, where it has been identified in the NMR as a DMV (Fig 2.2.A.4). On one Wainfleet LiDAR, there is a circular low area to the west of the pond zone, with a connecting channel to the Haven. The large ponds are in the deeper area, with the smaller ones in a shallow zone.

It seems feasible therefore at least to consider these features as sun-works. But given a wide distribution in the region and a lack of mention in documentary sources, what else might they be?

  1. They might be ponds but not for salt. Possibilities include:
    • Fish. This area was the source of a rich eel fishery in the thirteenth century and no doubt at other times as well. There is no evidence of storage in ponds in the documentary record.
    • Hemp. Another regional product, which requires soaking in still water to separate out the fibres. It was forbidden in the East Fen and is noisome so that an installation near a town seems unlikely.27 The same is true of flax. Although growing of hemp on certain farms was mandated temp Henry VIII, the size and spread of pond systems seems excessive for the likely acreage involved.
  2. They are not ponds but fields. The main difference here is the boundary trace. The ponds seem to have very thick boundaries, are very regular, and there are no obvious entrances. Compared with known tofts at Sailholme and around the church at Thorpe St Peter their regularity is striking, as is the difference with the system of former tofts and lanes to the south of Wainfleet St Mary church. Further, the enclosure contain no ridge-and-furrow, as do some of the documented tofts, but many have a ‘gathering’ of a different soil colour in the middle like many of the ponds in recent artisanal salt-making viewed from the air.
  3. The small ponds are traces of some form of soil or drainage management. (Clearly the larger features are real solid boundaries, especially when seen in oblique views). Dr Teresa Maybury showed the GE patterns to a farmer near Saltfleetby and he thought that field drainage patterns were the underlying cause of the rectangular layouts.
  4. Examples from inland locations like Thorpe St Peter and Little Steeping raise the question of access to salt water, especially in the early modern period, when presumably there were no remaining tidal creeks so far from the then coast. It is not too difficult to envisage salt creeks getting into Thorpe St Peter from the line of the current Croft-Wainfleet All Saints road, which was probably a sea-bank temp Henry VIII, but Steeping might look as if it had to be supplied from the Lymn, which is very unlikely since after 1560 there were sluices downstream from Wainfleet town. But the whole question of whether there might have been a subterranean saline aquifer and then upwelling ‘hot-spots’ is open. (There are solar - and wind-powered salinas using ground-water in Spain). Also, Wainfleet Haven might have been much wider than the recent roddon implies.

There seems a good case for looking to produce a distribution map of this phenomenon as a next step. But what is odd is that there is no documentary evidence. None of the medieval documents refers to anything that could be interpreted as a pond system and the management of the pre-drainage East Fen was predicated on keeping salt water out of it. Robinson’s date is not one where the documents have been closely examined, except (in general) for those dealing with salt-marsh reclamation. In Wainfleet St Mary, much of the land was acquired in the period 1610-1650 by Edward Barkham and the record of his acquisitions has survived.28 Again, none of these land parcels has any reference to salt except for tofts often with a salt-marsh and a fishery, which seem very likely indeed to have occupied the coastal stretch between the Haven and the boundary with Friskney. A cursory examination of the Commissioners of Sewers records in the LAO has not thrown up any obvious possibilities, though numerous windmills were licensed and these might have had a role in pumping salt water between ponds rather than fresh water out of ‘polders’. Yet the Barkham windmill on Wainfleet Bank very near the pond system was almost certainly a flood-control installation.29

There is also the enigma of whether such systems could produce salt without a final boiling process, In this latitude it seems a bit unlikely except perhaps in a very warm and windy summer. Fuel is unlikely to have been a problem (discounting cost for the moment): the East Fen was very peaty and before the drainage the equivalent of the Norfolk Broads was present in the Deeps, shown for example on Dugdale’s map from the mid-seventeenth century: it is assumed that they are medieval.30 By the seventeenth century coal was being imported into the region.31 So any further investigation of the pond systems needs also to look for the presence of boiling houses, which might have been substantial structures, if the examples described by Brownrigg are typical.

One last consideration is the left-field thought that these set-ups might be Roman in age, given the known importance of this region to the Empire. Since the layout is very similar to those of the Mediterranean, might there have been a transfer of technology? An extensive collection of data on Roman salt-making in the Fenland says relatively little about water management but such evidence as appears seems to focus on tidal channels leading to small storage vats: no extensive pond systems are postulated.32 In the field, the pond systems look rather too fresh for Roman age and in any case Roman briquetage in this area is usually buried beneath the clay of a marine transgression. Whereas some buried landscapes may be coming to light due to wastage of soils under intensive cultivation, this seems unlikely for the pond systems’ presence. So, attractive as the thought of a direct influence from the Mediterranean may be, Kurlansky’s assertion that such systems were post-Roman in age rather denies the usefulness of such an hypothesis.

Overview

In terms of importance in the present landscape, the grevas’ legacy is probably the most pervasive. There are many examples well inland of ridges of material which were probably waste mounds as well as the elongate field pattern which survived unto the 1960s and which was mapped by the OS in the nineteenth century and transferred to the early editions of the 1:25,000 series. At a remote imagery level, the pond systems are widespread (though no map has yet been produced) but their status is more than a little unresolved. Underground, the Roman and earlier contributions are no longer obvious elements of today’s landscapes though where the overlying clays are thin, then they may well become so. If the coming rise in RSL does not repeat the experience of the 4th century AD, of course.