DEVON ARCHAEOLOGICAL SOCIETY PROCEEDINGS No. 75 2017 Excavation of a Medieval Farmhouse with an adjacent Droveway, at Beaple’s Moor Cross, Knowstone Kirsten Dinwiddy Proc. Devon Archaeol. Soc. 75 (2017), 135–153 EXCAVATION OF A MEDIEVAL FARMHOUSE WITH AN ADJACENT DROVEWAY, AT BEAPLE’S MOOR CROSS, KNOWSTONE By KIRSTEN EGGING DINWIDDY with contributions by Dana Challinor, Lorraine Mepham and Sarah F. Wyles Archaeological works undertaken in advance of the installation of a wind farm between Batsworthy Cross and Beaple’s Moor Cross, Knowstone, Devon, revealed the remains of a medieval farmhouse, associated field boundaries and a droveway. No earlier settlement was detected, though the droveway is considered prehistoric in origin. The farmhouse was probably constructed in the 13th century, when favourable environmental and economic conditions led to agrarian expansion into the upland fringes. The subsequent deterioration in the weather and the socio-economic changes wrought by famine and plague are likely to have been influential in the site’s abandonment, sometime in the 14th century. An increased demand for agricultural commodities in mid-16th to early 20th centuries brought about the re-establishment of the region’s agricultural usage and improvement of the routeways. Later, as the trend towards urbanisation increased, farming declined and expanses of upland reverted back to moorland. INTRODUCTION Project background Archaeological investigations were undertaken on 190 ha of land to the south of the village of Knowstone, near South Moulton, North Devon (Fig. 1), preceding the installation of a nine-turbine wind farm and associated infrastructure. Preliminary investigations comprising a geophysical survey and an environmental impact assessment (Npower 2006; 2010; Archaeological Surveys 2006) led to an evaluation of each turbine location and associated development areas, as well as an earthwork survey and open area excavation at the proposed location for Turbine 2 (NGR SS 82351 21535) (Wessex Archaeology 2015). Installation works in the field to the west of the excavation were archaeologically monitored in the autumn of 2015. This article focuses predominantly on the Turbine 2 location excavation results. The wind farm covers pasture fields to the south of the A361, on either side of the road running through Batsworthy Cross in the south-west to the Beaple’s Moor Cross towards the north-east, much of which follows the Nutcombe Ridge. Turbine 2 is situated at the north-east end, between Beaple’s Moor Cross, the disused Moortown Quarry and Three Acre Plantation (Fig. 1). The Exmoor National Park lies approximately 8 km to the north, whilst around the site are smaller zones of moorland and woodland. The River Sturcombe flows 1.4 km to the south-east. Underlying superficial geological deposits are recorded as Regolith, in this instance a stony clay, which overlies sandstone and mudstone of the Crackington Formation bedrock (British Geological Survey online viewer). The site is situated on high ground, at approximately 260 m above Ordnance Datum. The ground slopes gently down from north to south, with a well-established turf growing in a 0.14 m thick topsoil. An 0.12 m thick, slightly stony subsoil overlies the natural silty clay and sandstone geology. 135 Fig. 1. Site location plan. 136 Archaeological and historical background Many of the existing roads and trackways, earthworks, cropmarks and geophysical anomalies in the region, are believed to have prehistoric origins, probably associated with the various sites (e.g. burial mounds) that feature in the wider landscape. A small standing stone – The Knowstone (boundary stone of a man named Knutr or ‘Knut’s Stone’; Gover et al. 1932, 340; Mills 1998; Fig. 2) – once marked the junction between three ancient routes. The stone (Devon and Dartmoor Historic Environment Record (DDHER) No. MDV12330) was relocated during the construction of the A361 North Devon Link Road in the 1980s. Much of the road between Batsworthy Cross and Beaple’s Moor Cross follows the route to the south-west, the ditch and bank earthworks of which cross the site (the droveway). According to late 19th-century maps, the north-eastern section (approximately 1 km) had Fig. 2. Site plan. 137 long been decommissioned, traffic instead using the road to the junction at Beaple’s Moor Cross (OS 1st Edition 1889–90). The nearby village of Knowstone (Chenutdestana/Chenueston Cnudstane), named after the above-mentioned standing stone, is noted in Domesday (AD 1086) four times; later documents link the estate, or parts of it, to the medieval families of De Brett, Beuple, Loring, Harington, Bonville and Grey. A number of properties in Knowstone have 11th century origins, whilst in the 13th century the masons working on the church also built the present Mason’s Arms. As with much of the south-west of England, the rural settlement pattern comprised small hamlets and isolated farmsteads, the farming inhabitants typically involved in ‘convertible husbandry’ (i.e. mixed rotational cultivation and pastoral agriculture) into the 16th century, around the time when a patchy retraction from marginal land becomes apparent (Rippon et al. 2006; Turner 2006; Miller and Hatcher 2014, 93; Beresford 1979). Having been declining for years, the trade in wool was much reduced, though other products, such as metal, coal and cloth, contributed to a healthy commercial economy and the beginnings of consumerism (Dyer 2012, 18). The demand for supplies for, and as a result of, the Industrial Revolution, a resurgence of agriculture was seen across the region (Countryside Agency 2005, 2; Rippon et al. 2006, 65). Surrounding farms, including the Moortown Barton farm and cider house, and potentially the Moortown Quarry have their origins in this period. Latterly, a decrease in farming, greater urbanisation and the tourist trade have allowed some moorland revival. EXCAVATION RESULTS The excavation revealed the remains of a small L-shaped farmhouse, comprising two ranges and a yard area (Figs 3 and 4). These were enclosed by a series of ditches and banks which had been modified over time. The surrounding land was divided into a series of fields, the two to the rear being tapered at alternate ends. The earthworks of the droveway, upon which the farmhouse fronted, were probably substantially altered during the life of the settlement, as well as during a later period (Fig. 2 and see Fig. 11). The farmhouse The farmhouse (Fig. 5) was represented by a series of rubble-cored, drystone walls up to 0.65 m wide and 0.50 m high, and predominantly of local dense grey sandstone, with occasional quartz veins (Fig. 6): context numbers 4111−2 and 4114−5 (see Figs 4–7). It is clear that – at least seasonally – rainwater needed to be managed in and around the building. A broad eaves-drip gully (4108 and 4109, up to 1.5 m wide; Fig. 4) extended around the outside of the south-west to north-east aspects, suggesting a thick, low-eaved thatch or turf roof. The various channels and stone-lined drains along and through the walls, and across the floors, are testament to water seepage into the building (e.g., 4341, Fig. 3; 4436 and continuations of 4108–9, Fig. 4). These formed part of the wider arrangement which would have carried the water away from the house and into the ditches defining the settlement plot. The building comprised two rectangular ranges, one to the south-east (the living area) measuring 3.75 m by 12.50 m, and one to the west (3.00 m by at least 3.65 m). The yard to the north-east was approximately 8.10 m wide, and no longer than 12.60 m. Two 1.25 m-wide probable doorways were identified in the northern wall of the living area; one with evidence for a doorpost (Fig. 7c) would have led to the yard, and one which communicated with the western range. A gap in the western wall, at the junction between the two ranges, may also indicate an entrance way. The drainage channels in this location (Figs 3 and 4) suggest that a further internal division may once have existed, perhaps installed in order to shield the hearth (see below) and occupants from draughts. Drainage channels also lined the inside of the walls of the living space and the yard, apparently 138 Fig. 3. Site plan: detail of farmhouse and plot. 139 140 Fig. 4. Site plan: detail of farmhouse and associated features. Fig. 5. View of the farmhouse from the south-west. Fig. 6. External elevation of wall 4112, from the north-east (scales 1 m and 0.20 m). 141 142 Fig. 7. a) West-facing elevation of wall 4112; b) North-west-facing section through wall 4112 and gullies 4108 and 4436; c) South-west-facing section showing end of wall 4113 and probable doorpost 4457. draining towards the south-west. The hearth (4231; Fig. 4), situated in the south-west end of the living area, comprised a 0.68 x 0.49 m sub-rectangular shallow depression lined with charcoal and fire-reddened clay. Charred remains (Wyles and Challinor, below) demonstrate the use of gorse/broom and cereal-processing waste to kindle and fuel the fires, which would have been used to prepare the meals for the household, as well as to provide some heat and light. A few associated stakeholes likely represent the use of cooking spits and pot-hangers. Similar hearths have been recorded in the region, including some with a granite hearthstone, the remains of a chimney hood and associated cooking pits (Beresford 1979, 135). The lack of dating and evidence for function from the adjacent shallow pit (4426) precludes definitive interpretation. The full extent of the western range was not revealed during excavation, though earthworks to the north-west (Fig. 2) suggest that the farmhouse continued for a few more metres, stopping 4 m or so short of the droveway. It appears that boundary ditch 4104 (see below), continued along the south-western side and around to the north-east, across the building’s frontage (Fig. 3). Whilst no internal features were identified to aid the interpretation of the function of the western range, it has been suggested elsewhere that such spaces may have been used as a store, craft space, or private quarters (Clarke 1984, 38). External access to the yard area was probably just beyond the end of the eastern wall, opposite the rock-filled causeway leading into the adjacent field (Fig. 3; see below). The continuation of the eaves-drip gully, a short length of wall and perpendicular, short drainage gullies on the north-eastern side of the complex suggests some form of covered area, or lean-to, at least in that part of the yard. The geophysical survey results (Fig. 1) hinted at a continuation of this wall to the north-west of the entrance, though no earthworks were identified here, and there was no evidence upon excavation. Further drainage features were observed along the eastern (?outdoor) aspect of the western yard wall. Despite careful cleaning, no other features were found within the farmhouse. Medieval pottery, probably dating to the 13th to 14th century, was recovered from several of the farmhouse features (Mepham, below). Associated features Earthworks around the farmhouse include ditches and associated banks, designed to drain and delineate the plot and protect the enclosed area from livestock (Figs 2 and 3). Immediately around the farmhouse, drainage comprised an eaves-drip gully (4108–9; Fig. 7b) with projections at the corners extending to the north-east and south-west, and a series of north-east to south-west channels. One of the earliest included a 30 m-long ditch 4105, along with the drip-gully (Fig. 3). Ditch 4106, 19 m to the south of the farmhouse, was likely to have been of a similar phase. This formed the boundary between two fields, one, tapered to the south-west and bounded to the north-west by the droveway, and the other (tapered to the north-east) delineated to the south by ditch 4107 (Figs 3 and 8). These correlate with the earthwork survey findings, which defined a bank along its northern aspect. Pollen from soils buried below this bank indicate an arable environment with some alder trees, followed by an increase in grassland and pastoral activity (Landon and Scaife 2015). A later phase of activity was represented by enclosure ditch 4104 (Fig. 9), which cut through ditches 4106 and the drip-gully extensions (Fig. 3). The enclosure ditch defined a trapezoidal area approximately 25 x 40 m, the northern boundary being formed by the southern edge of the droveway – which may have been re-established as part of the later alterations (see below). The parallel east and west elements of the enclosure led to an almost square southern boundary, possibly widening the tapered end of the field to the south, or forming an island/bank between ditches 4104 and 4106. In addition to the benefits of drainage, the enclosure earthworks will have afforded any gardens protection from 143 Fig. 8. West-facing section of ditch 4107 (scales 1 m and 0.20 m). Fig. 9. South-facing section of ditch 4104 (scales 1 m and 0.20 m). 144 Fig. 10. Causeway in ditch 4104, from the south-east (scales 2 m, 1 m and 0.20 m). grazing livestock. A 1.5 m long rock-filled constriction in the eastern leg of the enclosure ditch, opposite the yard entrance, probably represented a causeway (Figs 3 and 10), allowing controlled access to the domestic plot. An additional channel, extending from the eaves-drip gully to the rear of the building, appears to skirt around tree-throw hole 4392, and halt at larger, similar feature 4401 (Fig. 3). A stone-lined drain (4341) was seen to extend from the south-west corner of the building. Though not early, it is not clear which stage of the use of the farmhouse these relate to. Feature 4186 was circular, steep-sided and flat-based, and was probably a fire-pit. Basal fills comprised a lens of charcoal-rich silt, followed by a thin deposit of fire-reddened clay and stones. The main fill comprised a pale greyish silty clay and moderate-sized sandstone pieces. Though undated, the nature and contents of this feature suggests it was associated with the occupation of the farmhouse. The network of ditches form part of a larger field system identified to the south and south-west of the farmhouse (DDHER MDV19726). Abraded 13th- to 14th-century pottery, probably derived from manuring of the surrounding fields, was found in some of these ditches, as well as amongst the assemblage from the southern flanking ditch of the droveway. The droveway The remains of the droveway extended across the site, immediately north-east of the farmhouse plot. It continues as hedge banks and field boundaries towards the A361 in the 145 Fig. 11. South-west-facing section through southern droveway ditch 4103. north-east, and the road between Batsworthy Cross and Beaple’s Moor Cross (Figs 1 and 2; DDHER MDV19724; Wessex Archaeology 2015). By the mid-19th century this part of the route was already long-disused and the surrounding fields restored to heathland (Tithe map 1841; OS 1st edition 1889–90). The HER entry describes the droveway as being cut by a small wall or bank close to the farmhouse site, though this appears to have been outside the excavation area. The 5 m-wide course of the droveway, which lacked any sign of a formally laid surface, is defined along each side by an earthen bank (partly of segmented construction). Features indicating adaptations associated with the farmhouse include entrance-ways into the domestic plot to the south, and into the fields to the north. A few residual sherds of 13th- to 14th-century pottery were recovered from one of the accumulated ditch fills. An internal division, in the area where both flanking ditches were revealed, would have constricted the droveway path to around 2 m wide, presumably designed to aid stock control around the settlement, or if of a later date, between the surrounding fields. Evidence for maintenance of the flanking ditches during the post-medieval period includes a few sherds of pottery found in the soils buried below part of the northern bank, and a stone-filled land drain which had been installed along the length of the southern ditch (4103; Fig. 11). Pollen from here suggests open grassland, ferns and patches of hazel and heathland, probably indicating expansion of the moorland – though there was evidence for some cereal cultivation (Landon and Scaife 2015). Later features Ditch 4116 (Fig. 3) post-dated the farmhouse enclosure and associated drainage channels, though it roughly corresponds to the outline of the earlier field to the rear. No dateable artefacts were recovered, but stratigraphic, cartographic and historic evidence suggests it may date to the 18th or 19th century, when the industrial revolution increased the demand for agricultural products sufficiently to warrant the cultivation of marginal land. Traces of later, potentially similarly dated, ditches were identified in the northern part of the site, 135 m to the north-east of the farmhouse, and to the north of the droveway (4202, 4204 and 4243; Fig. 2). A subsequent watching brief revealed a 20 m-long section (0.60 m wide, 0.07 m deep) of a probable stock enclosure evident on aerial photographs approximately 155 m to the west of the farmhouse. No dating evidence was recovered, though it is assumed to be of medieval or later date. 146 FINDS Pottery by Lorraine Mepham The pottery assemblage recovered from the site amounts to 132 sherds (831 g), of which 129 sherds (755 g) are medieval, and the remaining three sherds (76 g) are post-medieval. Only the medieval material will be discussed here. The condition of the assemblage is fair to poor, and sherds have suffered a relatively high level of surface and edge abrasion. Mean sherd weight for the medieval assemblage is 5.9 g. The medieval sherds are all in similar coarse fabrics, tempered with prominent subangular quartz sand and occasional rock inclusions. These fall into the category of North Devon medieval coarsewares, which were certainly made in Barnstaple and possibly in Bideford, and perhaps also at other sources in the area; dating evidence, mainly from Okehampton Castle, indicates a starting point for the industry in the early 13th century, and it continued in use until the 15th century (Markuson 1980, fabric A; Allan and Perry 1982, fabric 1; Allan 1994). Petrological analysis of samples from inland sites, including Okehampton, has confirmed a link with the Barnstaple production centre (e.g. Vince and Brown 1982). The industry had a very limited repertoire in terms of vessel forms, and only jars are represented here, with slightly ‘dished’ rims profiles, some internally bevelled; comparable forms were found at Okehampton in 13th- to 14th-century contexts (Allan and Perry 1982, fig. 43). Several sherds exhibit external sooted residues. In the absence of any more diagnostic forms the pottery cannot be dated more closely, but it seems likely that this small assemblage represents a relatively short-lived occupation rather than any lengthy sequence. One complete jar profile (comprising 26 sherds; Fig. 12), with a narrow, scored band around the girth, was recovered from the southern droveway ditch. A further 49 sherds came from the drainage system associated with building 4110, while minimal quantities were found in other features and deposits in and around the building. Fig. 12. Medieval pottery from the southern droveway ditch 4103. ENVIRONMENTAL EVIDENCE Charred plant remains by Sarah F. Wyles Two bulk samples from hearth 4231 were processed by standard flotation. All identifiable charred plant macrofossils were extracted; identification followed the nomenclature of Stace (1997) and Zohary and Hopf (2000, tables 3 and 5), and with reference to modern reference collections. The results are presented in Table 1. 147 Phase Medieval Hearth 4231 Feature Context 4232 S quad 4232 W quad Sample 5 6 Vol (L) 0.5 0.4 Flot size 250 175 %Roots 5 5 2 2 Cereals Common Name Hordeum vulgare L. sl (grain) barley Hordeum vulgare L. sl (rachis frag) barley - 1 Triticum turgidum/aestivum (grain) free-threshing wheat 35 37 Triticum turgidum/aestivum (rachis frags) free-threshing wheat 10 6 Secale cereale (grain) rye cf. 28 cf. 18 Secale cereale (rachis frag) rye 14 7 Cereal indet. (grains) cereal 60 50 Cereal frag. (est. whole grains) cereal 12 15 Cereal frags (rachis frags) cereal 127 69 Other Species Stellaria sp. L. stitchworts 2 - Rumex sp. L. docks 1 1 Brassica sp. L. brassica 1 - Raphanus raphanistrum L. runch - 1 Calluna vulgaris fruit heather cf. 18 cf. 14 Erica /Calluna type stems heather + ++ Erica /Calluna type pinules heather cf.+ cf. + Vicia L./Lathyrus sp. L. vetch/wild pea 5 - Anthemis cotula L. (seeds) stinking mayweed 2 1 Leucanthemum vulgare Lam. oxeye daisy 2 4 Poaceae culm node grass 24 13 Lolium/Festuca sp. rye-grass/fescue 2 5 Poa/Phleum sp. L. meadow grass/cat’s-tails - 4 Avena sp. L. (grain) oat grain 16 15 Avena sp. L. (spikelet) oat spikelet 1 8 Avena sp. L. (awn) oat awn 2 3 Avena L./Bromus L. sp. oat/brome grass 42 50 + + Monocot. Stem/rootlet frag Table 1. Charred plant remains from medieval hearth 4231. 148 The assemblages were dominated by cereal grains including free-threshing wheat (Triticum turgidum/aestivum type), barley (Hordeum vulgare) and rye (Secale cereale). Fragments of rachis were also found. All are common in Saxon and medieval contexts from southern Britain (Greig 1991). Oat/brome grass (Avena/Bromus sp.) were most common amongst the weed seeds; other species include vetch/wild pea (Vicia/Lathyrus sp.), rye-grass fescue (Lolium/Festuca sp.), meadow grass/cat’s-tails (Poa/Phleum sp.), stinking mayweed (Anthemis cotula) and oxeye daisy (Leucanthemum vulgare). There were also a large number of heather stems (Erica/ Calluna type) and possible pinules, together with heather fruits (Calluna vulgaris), other monocotyledon stem/rootlet fragments and grass culm nodes. Threshing, winnowing and coarse sieving of free-threshing wheat in the field tends to remove the majority of the chaff elements prior to storage as relatively clean grain (Hillman 1981; 1984). The hearth assemblages are consistent with the burning of waste from an early stage crop processing such as coarse sieving. The waste (possibly specifically retained), as well as gorse/broom (Challinor, see below) may have been used as tinder, though a shortage of alternative fuel cannot be discounted. The remains of heather demonstrate the use of heathland resources, whilst an increase in stinking mayweed – as is common in the Saxon and medieval periods – indicates cultivation of heavier clay soils (Greig 1991; Green 1984), associated with the increasing use of mouldboard ploughs in preference to ards (Jones 1981; Stevens with Robinson 2004; Stevens 2009). These assemblages, with a lack of high status/exotic species, are consistent with a rural medieval site, and indicate a feature used solely for domestic purposes. Wood charcoal by Dana Challinor Samples from hearth 4231 were examined to identify which fuelwoods were utilised in the firing, and whether these would shed any light on specific function and to examine, to some extent, the use and availability of local woodland resources. Charcoal >2 mm in transverse section was considered for identification with 30 fragments (of variable size) randomly selected for identification from each sample. This was considered adequate to characterise the fuel used and to determine any significant differences between the quadrants. The charcoal was fractured and sorted into groups based on the anatomical features observed in transverse section. Representative fragments from each group were further examined and identifications were made with reference to Schweingruber (1990), Hather (2000) and modern reference material. Classification and nomenclature follow Stace (1997). Observations on maturity and character of the wood were recorded. The abundant charcoal fragments were generally mid-sized and in quite poor condition, with strong orange sediment infusion and notable blue-green staining; the latter being characteristic of vivianite, a ferrous phosphate linked to the decomposition of organic waste in wet sediments. The preservation is consistent with fluctuating levels of waterlogging. Four taxa were identified (Table 2): Quercus sp. (oak), Betula sp. (birch), Ericaceae (heather family) and Cytisus/Ulex (broom or gorse). Although the members of the Ericaceae family cannot be easily distinguished from the wood anatomy, the presence of fruits (and possible stems) of Calluna vulgaris (heather), makes it likely that the charcoal derived from this taxon. Likewise, Cytisus (broom) and Ulex (gorse) are not readily separated. Almost all of the charcoal fragments exhibited moderate to strong ring curvature, and many of the Cytisus/Ulex had preserved pith and bark. Variable sizes and ages were noted, but most roundwood was <10 years old. Charring causes a significant level of shrinkage (up to 40%), but it was clear that the stems would have originally been of small diameter. 149 Feature Hearth 4231 Context 4232 N quad 4232 E quad 4232 S quad 4232 W quad Sample 3 4 5 6 3r 2r Quercus sp. oak 10r Betula sp. birch 3r Ericaceae heather family Cytisus/Ulex broom/gorse Total 1r 17r 30r 26r 28r 30 30 30 30 KEY: r – roundwood Table 2. Results of the charcoal analysis. The results show that heathland resources were exploited for fuelling the hearth; representing 85% of the whole assemblage. Oak and birch were also used, to a more limited extent (15%), as supplementary fuel. It is impossible to say on the basis of this one feature whether this reflects a shortage of deciduous woodland and/or an increase in heathland habitats since it is possible that the charcoal reflects a deliberate selection of fuel for a specific function. It is recorded that, traditionally, gorse was commonly used to provide a fast, high heat, for example suitable for heating up oven structures, before being swept out for baking (Gale and Cutler 2000, 260). Given the quantities of chaff, rootlets and monocot stems (see Wyles, above), it is likely that a variety of fast-lived fuels were used, perhaps as kindling, or if other resources were limited, as the main source of fuel. This indicates that, at least to some extent, the dominance of gorse (or broom) reflects accessibility to the moorlands of northern Devon and the management of heath resources. DISCUSSION The earliest evidence for activity on the site is probably the droveway, perhaps part of a purported ancient route (DDHER Number MDV12330) which extends at least from Oldways End, the site of the junction of several similar routeways some 5 km to the north-east, to Chulmleigh 15 km in the south-west, where the Rivers Taw, Little Dart and Hollacombe Water converge. Evidence suggests that the 1 km stretch that includes the part that crosses the site was largely abandoned by the late post-medieval period or early 19th century. It would appear that modifications were made when the medieval farmhouse was constructed and in use, as well as some time during the post-medieval period, after the house had long-since been deserted. The droveway would have been important to the occupants of the house, who would have needed to move their grazing stock between areas of rough pasture, usually away from the adjacent arable fields (Astill 1988, 47). The farmhouse was probably a 13th- to 14th-century vernacular building likely to have been inhabited by wealthier farmers. Roofing depended on the availability of materials (e.g. turf, thatch), while walls may have consisted of masonry to full height, or ‘dwarf’ walls, topped with a timber-framed structure. Internal spaces typically included a living room with a hearth, stores, craft or private space, a yard and a garden area. Floors were usually trodden clay or earth, and regularly swept clean, with the debris being deposited outside (Clarke 1984, 38). 150 Enclosure of the plot to include some garden space is not unusual, several similarlydated but larger and longer-lived settlements in the region (e.g. Hutholes and Hound Tor, Dartmoor; Beresford 1979) are similarly defined, and are considered to be largely associated with stock control. The fields around the farmhouse, known as ‘parks’ or ‘closes’, are also characteristic of the region and period (Rippon et al. 2006, 58; Miller and Hatcher 2014, 93). The lack of evidence for lengthy period of use is supported by the pottery assemblage, which implies a fairly short-lived occupation during the 13th to 14th centuries. A combination of the favourable conditions provided by a climatic optimum, a booming economy and a lucrative wool trade, prompted the expansion of mixed farming into the upland fringes (Steane 1985, 174; Rippon et al. 2006, 58; Miller and Hatcher 2014, xii). The consequences of a subsequent deterioration in the weather – evident at the farmhouse and contemporaneous settlement sites in the region as attempts to ameliorate the effects of the wetter conditions, such as drainage channels along the insides of walls, an increase in the use of stone-walling, paved thresholds and yards, and improvements to field drainage systems (Beresford 1979; Steane 1985, 175) – and of the famines and plague of the 14th century are well-recognised and have often been implicated in the desertion of settlements (e.g. Beresford 1979; Platt 1997; Dyer 2002, 228–29, 237; Turner 2006). However, pottery from deserted houses (e.g. Hutholes and Hound Tor) indicates occupation between the mid-13th and mid-15th centuries (English Heritage nd.; Allan 1994, 145). Rippon et al. (2006, 65) found that the pollen record of the mid- and North Devon region implies a general continuation of the agricultural status quo until the 16th century, and only patchy arable retreat from marginal areas during the late 15th century, when documentary evidence attests to a shift from arable to pastoral farming (ibid.; English Heritage 2006, 7). In the case of the Beaple’s Moor Cross farmhouse, the evidence points towards the challenging conditions of the 14th century as being instrumental in its demise, though it is possible that there was some agrarian use of the neighbouring fields. After some reversion to moorland, the increasing demands for agricultural products between the mid-16th and 19th centuries necessitated greater exploitation of the region including some reclamation of the upland fringes and improvements to the connecting thoroughfares, such as those represented by the changes to the droveway (Countryside Agency 2005, 2; Rippon et al. 2006, 65). Increasing urbanisation and a decline in farming have allowed large areas of upland to revert back to moorland. ACKNOWLEDGEMENTS Wessex Archaeology would like to thank RWE Innogy UK Ltd for commissioning and funding the project. The assistance of Michael Williams (RWE Innogy UK Ltd) during the course of the project is gratefully acknowledged. Stephen Reed – Archaeological Officer for Devon County Historic Environment Team – is also thanked for his assistance and advice. The project was managed throughout by Gareth Chaffey, and directed in the field by Susan Clelland. The illustrations are by Elizabeth James. Andrew Powell is thanked for his comments on an earlier draft; Philippa Bradley edited the report. Rachael Seager Smith is thanked for assistance with the stone identification, and Bob Davies for his knowledge of historic buildings. The records and finds are currently held at the Wessex Archaeology Offices in Salisbury, under the project code 104840–1. They will be deposited with the Museum of Barnstaple and North Devon in due course. 151 BIBLIOGRAPHY Unpublished Archaeological Surveys 2006: Geophysical Survey Report, Batsworthy Cross Wind Farm, Geophysical Survey of Proposed Turbine Locations for RWE Npower. Unpublished client report ref. 153. Landon, C. and Scaife, R. 2015: ‘Pollen’. In Wessex Archaeology 2015, 18–21. Wessex Archaeology 2015: Batsworthy Cross, Devon: post-excavation assessment. Unpublished client report ref 104840.04. Tithe Map 1814. Published Allan, J. 1994: ‘Medieval pottery and the dating of deserted settlements on Dartmoor’, Proc. Devon Archaeol. Soc. 52, 141–47. Allan, J.P. and Perry, I. 1982: ‘Pottery and tiles’. In R.A. Higham, J.P. Allan and S.R. Blaylock, ‘Excavations at Okehampton Castle, Devon. Part 2 – the Bailey’, Proc. Devon Archaeol. Soc. 40, 86–101. Astill, G. 1988: ‘Rural settlement: the toft and croft’. In G. Astill and A. Grant (eds), The Countryside of Medieval England (Oxford: Basil Blackwell), 36–62. Barnett, C. 2009: ‘Wood charcoal’. In C. Ellis ‘Archaeology of the West Malling and Leybourne Bypass’. In P. Andrews, K. Egging Dinwiddy, C. Ellis, A. Hutcheson, C. Phillpotts, A. Powell and J. Schuster, Kentish Sites and Sites of Kent: A miscellany of four archaeological excavations. Wessex Archaeology Report 24 (Salisbury: Wessex Archaeology), 48–51. Beresford, G. 1979: ‘Three deserted medieval settlements on Dartmoor: a report on the late E. Marie Minter’s excavations’. Medieval Archaeol. 23, 98–158. British Geological Survey (BGS): Geology of Britain viewer. http://mapapps.bgs.ac.uk/ geologyofbritain/home.html (accessed 9th October 2015). Challinor, D. 2012: ‘Wood Charcoal’. In T. Allen, M. Donnelly, A. Hardy, C. Hayden and K. Powell, A Road through the Past: Archaeological discoveries on the A2 Pepperhill to Cobham road-scheme in Kent. Oxford Archaeology Monograph 16 (Oxford: Oxford Archaeology), 567–68. Clarke, H. 1984: The Archaeology of Medieval England (London: British Museum Publications Ltd). Countryside Agency 2005: Knowstone Parish Plan. Available on line at: http://www.knowstone.org. uk/knowstone-parish-plan (accessed October 2015). Dyer, C. 2002: Making a Living in the Middle Ages: the people of Britain 850–1520 (New Haven: Yale University Press). Dyer, C. 2012: A Country Merchant, 1495–1520. Trading and Farming at the End of the Middle Ages (Oxford: Oxford University Press). English Heritage nd.: History of Hound Tor Deserted Medieval Village. Available on line at: http://www.english-heritage.org.uk/visit/places/hound-tor-deserted-medieval-village/history/ (accessed January 2018) English Heritage 2006: Historic Farmsteads, Preliminary Character Statement: south-west region (Cheltenham: University of Gloucestershire, English heritage and the Countryside Agency). https://content.historicengland.org.uk/images-books/publications/historic-farmsteads-preliminarycharacter-statement-south-west/south-west-part1.pdf/ (accessed January 2018) Gale, R. and Cutler, D. 2000: Plants in Archaeology. Identification manual of vegetative plant materials used in Europe and the southern Mediterranean to c. 1500 (Otley: Westbury Publishing and Royal Botanic Gardens, Kew). Gover, J., Mawer, A. and Stenton, F. (eds) 1932: The Place-Names of Devon, Part Two (Cambridge: Cambridge University Press). Green, F.J. 1984: ‘The archaeological and documentary evidence for plants from the Medieval period in England’. In W. van Zeist and W.A. Casparie (eds), Plants and Ancient Man: Studies in Palaeoethnobotany. Proceedings of the 6th Symposium of the IWGP (Rotterdam: Balkema), 99– 144. Greig J. 1991: ‘The British Isles’. In W. van Zeist, K. Wasylikowa and K-E. Behre (eds), Progress in Old World Palaeoethnobotany (Rotterdam: Balkema), 229–334. Hather, J.G. 2000: The Identification of Northern European Woods; A Guide for Archaeologists and Conservators (London: Archetype Publications). 152 Hillman, G.C. 1981: ‘Reconstructing crop husbandry practices from charred remains of crops’. In R.J. Mercer (ed.), Farming Practice in British Prehistory (Edinburgh: Edinburgh University Press), 123–62. Hillman, G. 1984: ‘Interpretation of archaeological plant remains, the application of ethnographic models from Turkey’. In W. van Zeist and W.A. Casparie (eds), Plants and Ancient Man: Studies in the Palaeoethnobotany. Proceedings of the 6th symposium of the international work group for Palaeobotanists (Rotterdam: Balkema), 1–42. Jones, M.K. 1981: ‘The development of Crop Husbandry’. In M.K. Jones and G. Dimbleby (eds), The Environment of Man, the Iron Age to the Anglo-Saxon Period (Oxford) BAR 87, 95–127. Markuson, K.W., 1980: ‘Excavations on the Green Lane access site, Barnstaple, 1979’, Proc. Devon Archaeol. Soc. 38, 67–90. Miller, E. and Hatcher, J. 2014: Medieval England: Rural Society and Economic Change, 1086–1348 (London: Routledge). Mills, A.D. 1998: Oxford Dictionary of English Placenames (Oxford: Oxford University Press). Npower 2006: Batsworthy Cross Windfarm Environmental Statement, Volume 1, Chapter 7, 322–49. Npower 2010: Batsworthy Cross Windfarm, Statement of Environmental Impact, Chapter 5 Cultural Heritage. OS Map 1:2500, Devonshire XXIII.13. Northern or South Molton Division (1888–90 edn). Available on line at: http://maps.nls.uk/view/105998900 OS Map 1:2500, Devonshire XXIII.1. Northern or South Molton Division (1888–90 edn). Available on line at: http://maps.nls.uk/view/105999617 Platt, C. 1997: King Death: the Black Death and its Aftermath in late-medieval England (London: UCL Press). Rippon, S.J., Fyfe, R.M. and Brown, A.G. 2006: ‘Beyond villages and open fields: the origins and development of a historic landscape characterised by dispersed settlement in South-west England’, Medieval Archaeol. 50, 31–70. Schweingruber, F.H. 1990: Microscopic wood anatomy, 3rd edn (Birmensdorf: Swiss Federal Institute for Forest, Snow and Landscape Research). Stace, C. 1997: New flora of the British Isles, 2nd edn (Cambridge: Cambridge University Press). Steane, J.M. 1985: The Archaeology of Medieval England and Wales. Croom Helm Studies in Archaeology (Beckenham: Croom Helm Ltd). Stevens, C.J. 2009: ‘Charred Plant Remains’. In C. Ellis ‘Archaeology of the West Malling and Leybourne Bypass’. In P. Andrews, K. Egging Dinwiddy, C. Ellis, A. Hutcheson, C. Phillpotts, A. Powell and J. Schuster, Kentish Sites and Sites of Kent: A miscellany of four archaeological excavations. Wessex Archaeology Report 24 (Salisbury: Wessex Archaeology), 41–47. Stevens, C. J. with Robinson, M. 2004: ‘Production and consumption: Plant cultivation’. In G. Hey (ed.), Yarnton: Saxon and Medieval Settlement and Landscape. Thames Valley Landscape Monograph (Oxford: Oxford Archaeology), 81–82. Turner, S. 2006: Medieval Devon and Cornwall: shaping an Ancient Countryside (Landscapes of Britain) (UK: Windgather Press). Vince, A.G. and Brown, D.H., 1982: ‘The petrology of some pottery from Okehampton’. In Allan and Perry 1982, 101–02. Zohary, D. and Hopf, M. 2000: Domestication of plants in the Old World: the origin and spread of cultivated plants in West Asia, Europe, and the Nile Valley. 3rd edn (Oxford: Clarendon Press). This paper is published with financial support from RWE Innogy UK Ltd 153