LAND AT DOMGAY LANE FOUR CROSSES POWYS
POST-EXCAVATION ASSESSMENT AND UPDATED PROJECT DESIGN
For
EDP
on behalf of
DAVID MCCLEAN HOMES
CA REPORT: 06046
JUNE 2007
LAND AT DOMGAY LANE FOUR CROSSES POWYS
POST-EXCAVATION ASSESSMENT AND UPDATED PROJECT DESIGN
CA PROJECT: 9018 CA REPORT: 06046
Author: Tim Havard and Annette Hancocks
Approved: Martin Watts
Signed: …………………………………………………………….
Issue: 01 Date: 14 June 2007
This report is confidential to the client. Cotswold Archaeology accepts no responsibility or liability to any third party to whom this report, or any part of it, is made known. Any such party relies upon this report entirely at their own risk. No part of this report may be reproduced by any means without permission.
© Cotswold Archaeology Building 11, Kemble Enterprise Park, Kemble, Cirencester, Gloucestershire, GL7 6BQ Tel. 01285 771022 Fax. 01285 771033 E-mail: [email protected] Land at Domgay Lane, Four Crosses, Powys: Post-Excavation Assessment and Updated Project Design © Cotswold Archaeology
CONTENTS
SUMMARY...... 3
1. INTRODUCTION ...... 5
Location ...... 5 Archaeological background...... 5
2. AIMS AND OBJECTIVES ...... 7
3. METHODOLOGY...... 7
4. RESULTS (Figs 3 to 5) ...... 8
Stratigraphic Record: factual data...... 14 Stratigraphic record: statement of potential ...... 14 Artefactual record: factual data ...... 15 Artefactual record: statement of potential ...... 16 Biological record: factual data...... 17 Biological record: statement of potential ...... 20
5. STORAGE AND CURATION ...... 22
6. UPDATED AIMS AND OBJECTIVES ...... 22
7. PUBLICATION ...... 24
Synopsis of Proposed Report ...... 24
8. PROJECT TEAM ...... 26
9. TASK LIST ...... 28
10. TIMETABLE ...... 29
11. REFERENCES ...... 29
APPENDIX 1: STRATIGRAPHY BY TIM HAVARD...... 35
APPENDIX 2: WORKED FLINT BY E.R. McSLOY ...... 36
APPENDIX 3: POTTERY BY E.R. McSLOY...... 37
1 Land at Domgay Lane, Four Crosses, Powys: Post-Excavation Assessment and Updated Project Design © Cotswold Archaeology
APPENDIX 4: WORKED STONE BY E.R. McSLOY ...... 40
APPENDIX 5: AMBER BEAD BY E.R. McSLOY ...... 41
APPENDIX 6: METALWORK BY E.R. McSLOY ...... 42
APPENDIX 7: METALLURGICAL RESIDUES BY DR T.P. YOUNG...... 43
APPENDIX 8: HUMAN BONE BY TERESA GILMORE...... 47
APPENDIX 9: ANIMAL BONE BY S. WARMAN...... 50
APPENDIX 10: CHARRED PLANT REMAINS BY ELIZABETH PEARSON ...... 53
APPENDIX 11: CHARCOAL BY ROWENA GALE ...... 62
APPENDIX 12: POLLEN BY DR. KATIE HEAD ...... 69
APPENDIX 13: GEOARCHAEOLOGY BY KEITH WILKINSON...... 73
LIST OF ILLUSTRATIONS
Fig. 1 Site location plan (1:50,000) Fig. 2 The site, with nearby cropmarks and earthworks (1:10,000) Fig. 3 Plan of excavated features (1:2000) Fig. 4 Periods 1 and 2: Neolithic and Bronze Age features (1:1000) Fig. 5 Periods 3 and 4: Iron Age and Roman features (1:1000) Fig. 6 Period 4: Detail plan of postholes containing metalworking debris (1:100) Fig. 7 Periods 5 and 6: Post-medieval and undated features (1:1000) Fig. 8 Sketch plan of the barrow and sample locations Fig. 9 The monolith samples Fig.10 Magnetic properties of the sampled fine-grained layer
2 Land at Domgay Lane, Four Crosses, Powys: Post-Excavation Assessment and Updated Project Design © Cotswold Archaeology
SUMMARY
Site Name: Land at Domgay Lane Location: Four Crosses, Powys NGR: SJ 2707 1878 Type: Excavation and watching brief Date: October 2004 to October 2006 Location of archive: To be deposited with Powysland Museum (Welshpool) Accession Number: A467/2005 Site Code: DFC 04/DFC 05
A programme of archaeological investigation, comprising an excavation and a subsequent watching brief, was undertaken by Cotswold Archaeology (CA) between October 2004 and October 2006 at the request of Waterman CPM Limited and subsequently The Environmental Dimension Partnership (EDP) on behalf of Jennings Homes Limited (now part of David McClean Homes Limited) on land at Domgay Lane, Four Crosses, Powys.
The site had previously been investigated archaeologically in 1983 and 1984 as part of an investigative programme of cropmark complexes around Four Crosses, when three ring ditches forming part of an extensive barrow cemetery of probable Bronze Age date were excavated within the site. In response to a planning consultation for development of the site, a programme of geophysics and archaeological evaluation was undertaken in 2002 and 2003 respectively. The CA excavation was subsequently carried out as a condition of planning consent.
The excavation revealed numerous features dating from the Neolithic to post-medieval period. One pit produced several large potsherds of late Neolithic date. Eleven ring ditches and the southern half of a barrow mound, which were part of the barrow cemetery, were excavated. The barrow mound was shown to have been constructed in two phases, with a small primary mound being replaced by a larger secondary mound. Graves were revealed within several of the ring ditches. Other features included several undated pit alignments and two Romano-British enclosures and field system. Evidence for Roman metalworking was also uncovered, comprising postholes and pits with copper-smelting waste including slag, daub superstructure and a fragment of block tuyère. A small cemetery of probable Roman date was also excavated.
3 Land at Domgay Lane, Four Crosses, Powys: Post-Excavation Assessment and Updated Project Design © Cotswold Archaeology
This document presents a quantification and assessment of the significance of the evidence recovered from the fieldwork. It considers the evidence collectively in its local, regional and national context, and presents an updated project design for a programme of post- excavation analysis to bring the results to appropriate publication.
4 Land at Domgay Lane, Four Crosses, Powys: Post-Excavation Assessment and Updated Project Design © Cotswold Archaeology
1. INTRODUCTION
1.1 Between October 2004 and October 2006 Cotswold Archaeology (CA) carried out an archaeological excavation and watching brief on land at Domgay Lane, Four Crosses, Powys (centred on NGR: SJ 2707 1878; Fig. 1). The work was undertaken at the request of Waterman CPM Limited, and subsequently The Environmental Dimension Partnership (EDP), on behalf of Jennings Homes (now part of David McClean Homes) in accordance with a Design Brief for Archaeological Excavation prepared by Clwyd Powys Archaeological Trust (CPAT) Curatorial Section (2004), archaeological advisors to the Local Planning Authority, and with a subsequent detailed Written Scheme of Investigation (WSI) produced by CA (2004). The fieldwork also followed the Standard and Guidance for Archaeological Field Evaluation, the Standard and Guidance for Archaeological Excavation and the Standard and Guidance for an Archaeological Watching Brief issued by the Institute of Field Archaeologists (1999), and was guided by the Management of Archaeological Projects II (EH 1991).
Location 1.2 The site comprises an area of approximately 2.4ha on the north-eastern side of the village of Four Crosses (Figs 1 and 2). The site is located towards the western extent of a low gravel terrace (c. 60m AOD) by the confluence of the Rivers Severn and Vyrnwy. The flood plain of these rivers extends for at least 2km to the north, east and south of the site, while to the west hills like Bryn Mawr rise to c. 140m AOD. The site is bounded to the east by a caravan park and a dismantled railway; to the south by Domgay Lane; to the west by several houses and the upstanding earthen bank of Offa’s Dyke, and to the north by a tarmac road. It was under pasture prior to excavation and was generally flat although the southern half of an extant barrow lay towards the northern edge of the site on top of a slight natural rise in the ground. The underlying geology of the site is mapped as red marl and sandstone (BGS 1990). These are sealed by glacial and fluvioglacial deposits forming ridges supporting Brown Earths (Marches Archaeology 2003, 6).
Archaeological background 1.3 The site lies within an area of previously identified archaeological significance and immediately adjacent to Offa’s Dyke (Fig. 2). Aerial photographs taken in the 1960s and 1970s show the existence of cropmarks including ring ditches that appear to form an extensive barrow cemetery to the east of Four Crosses, at least five of which lay within
5 Land at Domgay Lane, Four Crosses, Powys: Post-Excavation Assessment and Updated Project Design © Cotswold Archaeology
the site. Other cropmarks include linear features and a complex of pit alignments, two of which approach the site from the north and the south-east.
1.4 Between 1981 and 1985 an investigative programme of these prehistoric remains around Four Crosses was carried out to assess the effects of plough damage. Two areas excavated as part of that programme lay within the boundaries of the site (Fig. 3). Ring ditches excavated at Site 2 (two ring ditches) and Site 3 (a single ring ditch) were dated to the Late Neolithic/Early Bronze Age periods (Warrilow et al. 1986). Three stretches of pit alignment were also investigated but produce no dating evidence (ibid., 55).
1.5 The upstanding earthen bank of Offa’s Dyke runs along the western extent of the site and is scheduled as an Ancient Monument (SAM ref: Mg033).
1.6 In response to pre-planning consultation originating in April 2002, a geophysical survey of the site and a subsequent archaeological evaluation were carried out (ArchaeoPhysica 2002; Marches Archaeology 2003). The geophysical survey enhanced the aerial photographic coverage and suggested the presence of further features not visible on the aerial photographs, including ring ditches, linear features apparently forming field systems, and further pit alignments crossing the site.
1.7 The evaluation confirmed the general accuracy of the geophysical survey and demonstrated that Pit 513, part of the pit alignment 1 was apparently of Late Iron Age date as charcoal recovered from this pit produced a radiocarbon date of 385–115 cal. BC at 95% confidence level (Beta-181726, 2200 +/- 50 BP). However, there is a distinct possibility that the charcoal may be intrusive as animal burrows were dug into the side of this feature. Examination of several of the possible ring ditches identified by geophysics confirmed their existence, but no surviving barrow mound material was observed. Few finds were recovered although the field system and enclosure ditches in the western part of the site were shown to be of likely Roman date. Two ditches were shown to be post-medieval in date.
1.8 After consultation on the full planning application between CPAT Curatorial Section and Waterman CPM, CPAT advised the local planning authority that an archaeological excavation be undertaken as a condition of planning consent ahead of residential development.
6 Land at Domgay Lane, Four Crosses, Powys: Post-Excavation Assessment and Updated Project Design © Cotswold Archaeology
2. AIMS AND OBJECTIVES
2.1 The objectives of the work were to identify any archaeological remains revealed during the course of the development works and to ensure the preservation by record of all archaeological remains revealed. The programme of recording in general terms was to:
• elucidate the form and function of archaeological remains identified during the evaluation and confirm their form and function; • recover artefactual and ecofactual material in order to assist with understanding the form, date and function of site; • confirm the presence or absence of settlement activity associated with the linear boundaries and field systems, and if possible characterise its form; • place the site within its local/regional/national context.
3. METHODOLOGY
3.1 Fieldwork was undertaken in three phases; the southern half of the site was excavated in the autumn of 2004 and the northern half in the summer of 2005. In both phases the topsoil and subsoil were removed by mechanical excavator under archaeological supervision. The topsoil and subsoil were not initially stripped across the barrow mound at the north-western edge of the site; instead two slots were excavated by hand across its length and width to determine the extent of mound material and the extent of the natural rise. Once this had been achieved, bulk deposits in the two quadrants left were reduced archaeologically once their nature had been understood. This barrow was the focus of a subsequent watching brief in October 2006, when deposits were mechanically excavated in shallow spits under constant archaeological supervision down to the top of the ditch fills. Once these were exposed, only one ditch was observed to lie within the area affected by the groundworks; this was hand-excavated.
3.2 Archaeological features exposed by machine stripping were then hand-excavated with sample rates (as specified in the WSI) of 50% minimum for discrete features (pits) and ring ditches, 20% for linear features (ditches, gullies), 5% for post-medieval field boundaries and 100% for any deposits directly relating to funerary/ritual activity
7 Land at Domgay Lane, Four Crosses, Powys: Post-Excavation Assessment and Updated Project Design © Cotswold Archaeology
(e.g. burials, cremations) and domestic/industrial activity (e.g. walls, postholes, hearths, floor surfaces/floor make-up deposits).
3.3 The two areas within the site that were excavated in 1984 were identified following machine stripping and were cleaned by hand to define their extent. Both ring ditches excavated from Site 2 were visible and recorded in plan; only the approximate area of excavation of Site 3 was defined (Fig. 3).
3.4 All features were recorded in accordance with CA Technical Manual 1: Excavation Recording Manual (CA 2005). The whole site was planned at a scale of 1:50 with more detailed plans (1:20) and sections (1:10 and 1:20) where appropriate. A photographic record of the excavation in progress and of all significant archaeological features was compiled on colour slide, monochrome film and digitally.
3.5 All deposits were assessed for their environmental potential and sampled appropriately in accordance with CA Technical Manual 2: The taking of samples for palaeoenvironmental and palaeoeconomic analysis from archaeological sites (CA 2003). All artefacts recovered from the excavation were retained in accordance with CA Technical Manual 3: Treatment of finds immediately after excavation (CA 1995).
4. RESULTS (Figs 3 to 5)
4.1 Six provisional periods of archaeological activity were identified. Features have been assigned to these provisional periods on the basis of initial dating evidence (derived from the ceramic assemblage and other dateable artefacts), functional interpretation, and spatial and stratigraphic relationships with other features.
Period 0: Geological Period 1: Neolithic Period 2: Bronze Age Period 3: Iron Age Period 4: Romano-British Period 5: post-medieval Period 6: modern Undated
8 Land at Domgay Lane, Four Crosses, Powys: Post-Excavation Assessment and Updated Project Design © Cotswold Archaeology
4.2 Brief summaries of the fieldwork results, artefactual evidence and environmental evidence are given for each period below. More detailed stratigraphic, artefactual and environmental information is provided in Appendices 1 to 13.
Period 1: Neolithic (Figs 4 and 6) 4.3 A single pit, 2548, measuring 1.5m in diameter and 0.5m in depth, in the western half of the site produced several large sherds of Late Neolithic Grooved Ware pottery dating to approximately 2700 BC to 2200 BC. It was cut by a later Roman ditch.
Period 2: Bronze Age (Fig. 4) 4.4 A total of eleven ring ditches and the southern half of a surviving barrow mound were excavated within the site. These were all within a 50m-wide corridor on a general north/south alignment. The two re-exposed ring ditches (Ring Ditches 12 and 13) of Site 2 and the excavated area of Site 3 of the previous excavations also lay within this corridor. At present all of the ring ditches are interpreted as being of Bronze Age date by their association with those previously excavated, which were dated to the Bronze Age (Warrilow et al. 1986, 53). Probable grave cuts were distinguished inside many of the ring ditches. Although the majority of them did not contain human bone they are interpreted as graves based on similarity of form to those graves which did contain human bone. All are undated at present.
4.5 At the southern end of the alignment was the largest excavated ring ditch (Ring Ditch 1), approximately 30m in diameter, with the ditch on average 1.35m wide and 0.8m deep with a pronounced notch at the base. Upcast derived from a comparatively narrow ditch enclosing a large area would have created a low mound so this ditch is provisionally interpreted as being for a disc barrow with a small central mound. Several internal features were investigated although none resembled grave cuts and none produced any dating evidence.
4.6 Ring Ditches 2, 3, 4 and 5 lay to the north and north-west of Ring Ditch 1. They were all of similar size, between 7.5m and 8.7m in diameter, with ditches between 0.2m and 0.45m in depth and 0.4m to 1m in width. Possible grave cuts were identified inside Ring Ditches 2 and 4, and a large pit (1.8m in diameter and 0.7m in depth) was recorded within Ring Ditch 5. No human bone was recovered from any of these features although a part of a modern calf skeleton was recovered from the pit inside Ring Ditch 2 is deemed intrusive. No features were identified inside Ring Ditch 3.
9 Land at Domgay Lane, Four Crosses, Powys: Post-Excavation Assessment and Updated Project Design © Cotswold Archaeology
4.7 Ring Ditch 6 measured approximately 10m in diameter, with a ditch 1.5m wide and between 0.4m and 0.8m deep that had been re-cut on its southern side. The ditch had almost totally cut away an earlier grave from which sk1/2 was recovered. Two grave cuts lay interior to the ditch, one of which (grave 2558) contained a partial burial (sk3), both are undated. A buckle of medieval date was recovered from the top of the fill of the ring ditch is likely to be intrusive.
4.8 Ring Ditches 7, 8, 9, 10 and 11 were grouped within 20m to the north and west of Ring Ditch 6. They were all of similar diameter (between 5m and 7m) and the varying widths (0.3m to 0.8m) and depths (0.1m to 0.4m) of their ditches may be attributed to varying degrees of truncation; indeed Ring Ditches 7 and 11 did not survive as continuous features. Possible graves were identified on the interior of all of these ring ditches. A small amount of skeletal material (sk6) was recovered from grave 2639 inside Ring Ditch 10. A copper-alloy finger ring, stylistically of Middle or Late Iron Age/Early Roman date, came from a fill towards the base of grave 2596 inside Ring Ditch 7, and is thus unlikely to be intrusive. The grave may therefore be a later burial within an earlier ring ditch, or the ring ditch itself may also be of Iron Age date. A small pit was situated on the southern side of Ring Ditch 11 where the ditch was discontinuous.
4.9 The southern half of an extant barrow mound lay towards the northern edge of the site, excavation of which revealed two concentric ring ditches. The inner ring ditch, Ring Ditch 15, measured between 1.9m and 2.5m in width and about 1.2m in depth. It was not continuous as two termini lay within the area excavated. The gravel upcast from its construction formed a primary mound around 8m in diameter, which extended to the inner lip of the ditch. An intrusion cut into the primary mound was observed in section, its full extent lying beyond the area of excavation. This was sealed by erosion from the primary mound which, together with silting, also filled the ditch. The final deposit in this sequence was sealed by a thin layer of light grey clay silt which is interpreted as a turf line forming over the primary mound and infilled inner ditch similar to that observed at Site 6 (Warrilow et al. 1986, 65–8). Monolith samples were taken through the possible buried soil underlying the primary mound and the turf line which formed over the primary mound.
4.10 The outer ring ditch, Ring Ditch 14, measured approximately 20m in diameter, the ditch being 1.1m in width and 0.6m in depth, and continuous within the area excavated. The gravel upcast derived from its construction was deposited over the
10 Land at Domgay Lane, Four Crosses, Powys: Post-Excavation Assessment and Updated Project Design © Cotswold Archaeology
infilled inner ditch and primary mound, creating a larger secondary mound. This mound was heightened by the deposition of soil over the gravel, creating a mound of at least 18m in diameter. The outer ditch was infilled by erosion from the secondary mound and ditch silting. This was sealed by a thin deposit of light grey clay silt again interpreted as a turf line. A monolith sample was taken through these deposits. A heavily truncated undated grave 2613 (sk5) lay immediately outside Ring Ditch 14.
4.11 The only other feature to be provisionally assigned to the Bronze Age was isolated grave 1327 towards the southern extent of the site, from a single amber bead, stylistically of likely Early Bronze Age date, recovered from its fill. No stratigraphic relationship could be determined with the immediately adjacent post-medieval ditch due to similarity of fills, and heavy root disturbance in the upper part of the two features.
Period 3: Iron Age (Fig. 5) 4.12 A single cremation (C3) north-west of Ring Ditches 3, 4 and 5 and equivalent to Pit 513 from the evaluation was identified (Marches Archaeology, 2003). Charcoal recovered from this feature during the evaluation produced Late Iron Age radiocarbon date of 385–115 cal. BC at 95% confidence level (Beta-181726, 2200 +/- 50 BP). Although the pit has animal burrows dug into the side of it, so this date must be treated cautiously as the charcoal may well be intrusive.
4.13 Two cremations (C1 and C2) lay towards the north-eastern extent of the site, on the same alignment as C3. A particularly large amount of burnt bone was preserved within C2, where the bone had been deposited in the base of the pit before being backfilled. None of the cremations contained any vessels and both were undated. However, it is likely that given the alignment of cremations 1, 2 and 3, that in all likelihood they are contemporary.
Period 4: Romano-British (Figs 5 and 6) 4.14 A frequently re-cut boundary ditch (Boundary Ditch 1) ran across the western part of the site on a north-west/south-east alignment. Enclosure 1 was located at the northern end of this boundary, with a smaller enclosure, Enclosure 2, to the south of Enclosure 1. Several pieces of Romano-British pottery and a stone spindlewhorl were recovered from the ditches defining Enclosure 1. No internal features to either of these enclosures were defined, although several burnt pits (Pit Group 1) were cut into the western terminus of the southern ditch of Enclosure 2.
11 Land at Domgay Lane, Four Crosses, Powys: Post-Excavation Assessment and Updated Project Design © Cotswold Archaeology
4.15 Field System 1, which was defined by a north/south ditch and several segmented east/west ditches, lay to the east of Enclosures 1 and 2. The field system is dated by a single piece of Romano-British pottery recovered from the main north/south ditch of the field system. This ditch ran parallel to the eastern side of Enclosure 1, cut Ring Ditch 6 and two pits of Pit Alignment 1, and was truncated by later post-medieval ditches.
4.16 Three intercutting ditches (Curvilinear Ditches 1–3), which had not been detected by geophysical survey, were located in the northern corner of the site. The latest of these, Curvilinear Ditch 2, cut Curvilinear Ditches 1 and 3 destroying the relationship between the two earlier ditches. Romano-British pottery was recovered from Curvilinear Ditches 2 and 3. The purpose of these ditches is unclear but they may have been related to the adjacent cemetery (see below).
4.17 To the west of Boundary Ditch 1 was Ditch 8080 (Fig. 6). Whilst there was no relationship between the two ditches, Ditch 8080 appeared to respect the larger boundary ditch. An area of cut features, some of which contained possible copper- smelting residues was located close to Ditch 8080 and to the west of Boundary Ditch 1. The metalworking debris included slag and hammerscale. In addition, fragments of a fired-clay superstructure and tuyères were retrieved. These fragments did not appear to be in situ and as such are unlikely to represent in situ hearths or furnaces. Of particular interest was a fragment of block tuyère recovered from feature 2242. The residues were recovered from a cluster of four features (2219, 2257, 2242 and 2252), a line of six circular features (2272, 2210, 2281, 2297, 2290 and 2313) which respected Boundary Ditch 1 and five other features (2373, 2331, 2324, 2335 and 2344). Additionally, several postholes (2139, 2142, 2324, 2331 and 2335) contained stone-packing although no form of possible structure could be discerned from the distribution of these. Possible copper-smelting residues were also recovered from two postholes (2034 and 2071) located in the north-west corner of the site and two more (1756 and 1796) just inside Ring Ditch 1 (Fig. 5). No dateable material was recovered from pits/postholes containing these residues, but preliminary analysis of the waste material shows it is likely to be of Roman date (see Appendix 7).
4.18 An area of at least eight graves, all aligned north-east/south-west, were located towards the northern end of the site (Fig. 5). Skeletal evidence only survived in grave 2551 (sk4) which also yielded several sherds of Romano-British pottery. A large
12 Land at Domgay Lane, Four Crosses, Powys: Post-Excavation Assessment and Updated Project Design © Cotswold Archaeology
stone-packed posthole (2488) within the area of graves may represent a grave marker.
4.19 Some cremated material (C4) was also recovered from the fill of the eastern ditch of Enclosure 1. This cremation is therefore deemed to be of Roman or later date.
Period 5: post-medieval (Fig. 7) 4.20 Field System 2 was defined by three parallel north-west/south-east ditches and a segmented ditch running north-east/south-west across the southern part of the site. Post-medieval pottery was recovered from one of these ditches and another appears as a boundary on a map of 1790 (Marches Archaeology 2003, 26). To the south of this segmented ditch ran a slightly meandering trackway (Trackway 1) from the ditches of which a small quantity of post-medieval pottery was also recovered.
Period 6: modern 4.21 Modern activity on the site was chiefly represented by several drains which ran across the southern part of the site and a soakaway in the north-west corner. The drains have been omitted from the plan for clarity.
Undated 4.22 Features assigned as undated comprised three pit alignments (nos 1-3) and those identifiable features within Ring Ditches . Pit Alignment 1 ran across the centre of the site on a north/south alignment. The pits were generally ovoid in shape with the long axis along the alignment, between 1.6m and 2.5m in length and an average of 0.33m in depth. No dating evidence was recovered from any of the pits during excavation although a Late Iron Age radiocarbon date was obtained from a charcoal fragment recovered within Pit 513 (now C3) during the earlier evaluation (see archaeological background, above). A gap in the alignment corresponds to the position of Ring Ditches 2, 3 and 4, suggesting that the pits post-date the ring ditches, and two pits of alignment 1 were cut by ditches of Romano-British date. Grave 2648 (sk7) was aligned parallel to the pit alignment, suggesting that it may have been contemporary. The orientation of Pit Alignment 1 suggests that it is a continuation of the cropmark pit alignment that approaches the site from the north (Fig. 2).
4.23 Pit Alignment 3 ran for a length of approximately 16m in a south-east/north-west direction, from pit alignment 1 in the direction of the surviving barrow mound. The pits were sub-rectangular in plan with the long axis along the alignment, and between 2m
13 Land at Domgay Lane, Four Crosses, Powys: Post-Excavation Assessment and Updated Project Design © Cotswold Archaeology
and 3.5m in length, 1m and 1.3m in width and 0.4m to 0.6m in depth. No dating evidence was recovered from any of these pits but their nature and location suggests they are contemporary with Pit Alignment 1. There was no stratigraphic relationship between the pit alignments and adjacent group of Romano-British graves; the pit alignments cut several earlier features but these were undated and of a different form to the graves.
4.24 A double pit alignment (Pit Alignment 2), probably representing a trackway, was situated in the south-east corner of the site. No dating evidence was recovered from it and, unlike Pit Alignments 1 and 3 it is not interpreted as being of contemporary date due to the dissimilar forms of its pits from Pit Alignments 1 and 3. The remaining undated features on site chiefly comprised isolated postholes and tree-throw pits that could not be spatially associated with dated features.
Stratigraphic Record: factual data 4.25 Following the completion of the excavation an ordered, indexed, and internally consistent site archive was compiled. A database of contextual and artefactual evidence and a site matrix was also compiled and cross-referenced to spot-dating. The excavation comprises the following records:
Watching Excavation Total Brief Context sheets 2 1827 2054 Plans (1:50) 2 244 294 Sections (1:10, 1:20) 3 586 589 Sample sheets 0 223 225 Monochrome Films 1 33 38 Colour slide Films 1 26 31 Matrices (A3 sheets) 0 15 15
4.26 The survival and intelligibility of the site stratigraphy was good with archaeological remains having survived as negative features. Despite a relative paucity of dating evidence, the majority of features have been assigned a provisional period.
Stratigraphic record: statement of potential 4.27 A secure stratigraphic sequence is essential to elucidating the form, purpose, date, organisation and development of the various phases of activity represented. This can be achieved through detailed analysis of the sequence and further integration of the artefactual dating evidence from the excavation and watching brief phases of work.
14 Land at Domgay Lane, Four Crosses, Powys: Post-Excavation Assessment and Updated Project Design © Cotswold Archaeology
The refined sequence will then serve as the spatial and temporal framework within which other artefactual and biological evidence can be understood.
4.28 The evaluation record has the potential to better define and date some of the archaeological features. Where appropriate, evidence from the evaluation will be integrated into the overall stratigraphic sequence.
4.29 Features dating to Periods 1 to 4, and some of the undated features, have been identified as being of archaeological significance.
Artefactual record: factual data 4.30 All finds collected during the excavation have been cleaned, marked, quantified and catalogued by context. Non-ceramic artefacts have been examined by a conservator and stabilised where appropriate.
Type Category Sherd/frag. Weight (g) Count Flint Worked 3 - Pottery Neolithic to Bronze Age 30 952 Roman 29 460 post-medieval/modern 102 620 Total 167 2169 Fired/burnt clay Hearth/furnace lining and - 4500 tuyère fragments Amber Bead 1 - Metals Copper alloy 2 - Iron 11 - Metallurgical Ironworking (hammerscale) present - residues Copper waste present - Stone Worked 2 -
Flint 4.31 Worked flint amounted to only three pieces of which one was an unstratified find. All comprise unutilised waste flakes which on the basis of observed technology most likely date to the Late Neolithic to Bronze Age periods.
Pottery 4.32 A small pottery assemblage was recovered dating to between the Late Neolithic and modern periods. Most significant was a group of 28 sherds from a single Grooved ware vessel of Late Neolithic (c. 2700-2200 BC) date from Pit 2548 (Fig. 4). None of the recovered pottery deriving from the ring ditch monuments or associated contexts is Bronze Age in date. Romano-British pottery occurs in a restricted range of fabrics
15 Land at Domgay Lane, Four Crosses, Powys: Post-Excavation Assessment and Updated Project Design © Cotswold Archaeology
which is typical for the region. Post-medieval or later material derived primarily from linear features interpreted as field boundaries.
Metal objects, amber and worked stone 4.33 The quantity and range of non-ceramic artefacts recovered is small. A small amber bead, the sole find from a probable grave fill, 1327, probably dates to the Early or Middle Bronze Age and may be residual in a later feature. Among the metal objects, a copper-alloy finger ring from the fill of grave 2596 (Ring Ditch 7) is noteworthy and appears to be of Mid/Late Iron Age date. Iron objects comprising hobnails and carpentry nails were restricted to Roman and post-medieval/modern dated contexts. The worked stone assemblage consists of a spindlewhorl recovered from Enclosure 1 and a polisher recovered from Field System 1, both of which were associated with pottery of Roman date.
Metallurgical Residues (includes associated fired clay) 4.34 Metallurgical residues from the site include a very small amount of hammerscale, possibly indicative of ironworking, and a larger amount of residues from copper or copper-alloy working. The very low densities of micro-residues relating to ironworking suggest that this was not a significant activity within the excavated area. The far more extensive evidence for copper-working occurs in the form of copper waste and fired clay, including fragments from tuyères and is entirely from contexts provisionally dated to the Roman period. The bulk of the material was recovered from features currently interpreted as postholes. The fragments of hearth/furnace debris recovered suggest the use of pre-formed tuyères, of which one example is represented by three fragments. No evidence for crucibles or moulds was recovered. This absence, coupled with the lack of evidence for tin contamination of the metallurgical ceramics and the presence of single fragment of probable azurite-bearing ore, all suggest that the processing carried out on the site may have been primary copper-smelting. A possible source for the copper ore, Llanymynech Hill (a significant source of copper ore in early times), lies only 3km to the north-west of the site.
Artefactual record: statement of potential
Flint 4.35 Perhaps surprisingly given the nature of the archaeology, worked flint from the site was very scarce and there is no potential for further analysis.
16 Land at Domgay Lane, Four Crosses, Powys: Post-Excavation Assessment and Updated Project Design © Cotswold Archaeology
Pottery 4.36 Of greatest significance within the small pottery assemblage is the Neolithic Grooved ware vessel. This vessel should be drawn and described for publication as it is both rare and distinctive in form and fabric. Roman and later components are few and merit little further attention. A short summary report, supported by basic recording of this material, will fulfil the obligations for archiving purposes and permit integration with the stratigraphic record to assist final site phasing.
Metal objects, amber and worked stone 4.37 The metalwork, worked stone, and amber objects are restricted in size and the potential for further analysis is considered poor. Two items, an amber bead and copper-alloy finger ring, are of intrinsic interest and should be drawn and described in the publication and further work undertaken to determine the date and aid in the chronological interpretation of the site. Additionally the finger ring should be the subject of x-ray fluorescence analysis to determine its composition and its possible source, as it may have been produced locally. Remaining items, including the worked stone items and iron nails should be included in a brief catalogue for purposes of the archive.
Metallurgical Residues (includes associated fired clay) 4.38 The evidence for copper working, most likely representative of primary smelting, is highly significant and quite possibly the first evidence of its kind for the Roman period from Britain. Analysis is recommended to confirm the process evidenced here and the source of raw material. This can be achieved through examination and comparison of the chemical and mineralogical composition of the possible ore fragment and the various residues (spheroids, slag, droplets and ceramics). The precise distribution of analyses and choice of specimens would depend on re- examination of the collection to identify the least corroded samples. Confirmation of the dating of residues through absolute means (radiocarbon dating) is required.
Biological record: factual data 4.39 Where appropriate all ecofacts recovered from the excavation have been cleaned, marked, quantified and catalogued by context. A 10-litre sub-sample of each environmental sample taken was processed for the purposes of assessment. Samples from cremations and inhumations were fully processed. Three monolith samples were taken from the section of the upstanding barrow and ditch.
17 Land at Domgay Lane, Four Crosses, Powys: Post-Excavation Assessment and Updated Project Design © Cotswold Archaeology
Type Category Count Human bone Inhumations/cremations 6/4 Animal Bone Hand collected/ from samples 246/163 Samples Environmental/burial/monolith 163/54/3
Human Bone 4.40 A total of six inhumations (Sk 1-7) and four cremation burials (Cremations 1-4) were
assessed. Skeletons 1 and 2 are the same skeleton, but have been assigned different numbers as the grave was exposed in two different excavated sections. Five of the inhumations came from burials and skeleton 1/2 represented the remains of a burial after truncation by Ring Ditch 6. Three of the cremations were discrete burials
and the fourth consists of pyre debris apparently dumped into (Enclosure 1). Skeleton 1/2 is likely to be earlier than Period 2, Sk4 is Roman (Period 4) and Sk 3, 5, 6 and 7 are undated. Cremations 1-3 are undated and Cremation 4 is of Roman or later date.
Animal bone 4.41 The hand-collected assemblage totalled 246 fragments from 80 bones weighing 1.46kg. Of these only 23 bones could be identified to species. The processed samples yielded a further 163 fragments from 106 bones weighing just 7g. None of these could be identified to species. Animal bone was recovered from contexts dating to Period 4 (Roman), 5 (post-medieval), 6 (modern) and from some undated deposits. This small and poorly preserved assemblage is consistent with the predominantly acidic soil conditions. The only species identified were horse and cattle. A relatively well-preserved calf skeleton from a feature within Ring ditch 2 is an intrusive modern specimen and represents 40% of the bone identifiable to species.
Charred plant remains 4.42 Samples from 138 deposits of Bronze Age to Roman date were selected for assessment. Tables 2-4 detail the range and quantity of material recovered. Food residues in the form of charred spelt, emmer/spelt and hazelnut shell fragments were found in Bronze Age deposits, and are possibly associated with ritual activities relating to the burials. A low level of charred plant remains in Roman contexts was the only evidence of crop processing on the site. Seeds associated with pyre deposits of the same date may indicate that these took place on grassy land, or that grassy material was used as tinder for the pyre.
18 Land at Domgay Lane, Four Crosses, Powys: Post-Excavation Assessment and Updated Project Design © Cotswold Archaeology
Charcoal 4.43 Charcoal was recovered from 181 samples (in flots and residues) but frequently only in very small quantities. Charcoal-rich samples were rare. The samples from dated deposits were all from Period 4 features or activity. The fuel debris included pine (Pinus sp.), oak (Quercus sp.), hazel (Corylus avellana), ash (Fraxinus excelsior) and elder (Sambucus sp.). Charcoal from Period 4 deposits included the remains of possible pyre debris from Enclosure 1 which identified oak (Quercus sp.), hazel (Corylus avellana), alder (Alnus glutinosa) and elder (Sambucus sp.). Undated deposits contained charcoal identified as ash (Fraxinus excelsior), alder (Alnus glutinosa), oak (Quercus sp.) and hazel (Corylus avellana).
Pollen 4.44 Three monolith samples (177, 178, and 179) were taken from the Bronze Age barrow mound (Ring Ditch 14). With the exception of one sample, counts of 250 total land pollen grains (TLP) could not be achieved due to the degraded nature of the pollen grains and their low concentrations. The vegetation observed in the samples above appears to represent a stable community of grassland and herbs associated with grassy environments. There was also a notable presence of ribwort plantain suggestive of clearance and activity around the site. Pollen remains, however, were badly preserved and in low concentrations, partly due to the exceptionally dry nature of the sediments. The remains however, did provide a record of vegetation at the site with little variability while the barrow was in use and extended.
Geoarchaeology 4.45 Three monolith samples were taken from the upstanding barrow and associated ditch in order to investigate the potential presence of palaeosols (buried soils) in Ring Ditch 14. Field and laboratory description suggest that contexts 3005, 3010 and 5023 are palaeosols that were buried by various stages of barrow construction. The magnetic susceptibility results undertaken on the magnetic properties of the monolith samples do not provide proof that the buried red-brown fine-grained layers are palaeosols. Indeed none show evidence of been characteristically mature soils. Therefore on the basis of the magnetic evidence the layers are either immature buried soils or they are sediments in which no pedogenesis has occurred.
19 Land at Domgay Lane, Four Crosses, Powys: Post-Excavation Assessment and Updated Project Design © Cotswold Archaeology
Biological record: statement of potential
Human bone 4.46 The poor preservation and completeness of the inhumations means that there is relatively little that can be gained from further analysis apart from the two individuals displaying some pathological changes, Sk 1/2 and Sk 5. The cremation burials, especially C2, are suitable for further more detailed analysis due to the level of preservation. The weight of bone recovered would suggest that the cremation burial is in situ or close to the pyre site (McKinley 1993) and is a good candidate for radiocarbon dating.
Animal bone 4.47 This small poorly preserved assemblage does not require any further work. A brief summary of the results presented in this assessment should be summarised for inclusion in the overall results on the project. None of the animal bone is sufficiently well preserved to be suitable for radiocarbon dating.
Charred plant remains 4.48 The low level of charred cereal crop remains from Bronze Age deposits is not unexpected from a barrow complex and plant remains associated with Bronze Age deposits are generally rare, and are therefore significant when recorded. Full quantification of plant remains from all samples from Period 2 deposits (Tables 7-9) and those Period 4 deposits associated with pyre material C4 (contexts 2010 and 2011) from the eastern ditch of Enclosure 1 and from context 2298 the posthole/pit fill of 2297 containing free-threshing wheat is recommended in order the understand the significance of this material. Charred plant remains from cremation burial C4 may be suitable for radiocarbon dating.
Charcoal 4.49 For Period 4 the full analysis of samples 103 and 104 (pyre deposit C4 from 2028 part of ditch 8075), and posthole samples 127 (posthole 2242), sample 135 (posthole 2281), sample 140 (pit 2313), sample 147 (posthole 2282) and sample 157 (2374) are recommended in order to determine the type of fuel used within the cremation process and as possible accelarants. It is recommended that no further work should be undertaken on the material from the undated deposits. However, suitable material for radiocarbon dating has been identified from undated samples which include cremations C1 and C2. Additionally charcoal from cremation burial C4 may help to
20 Land at Domgay Lane, Four Crosses, Powys: Post-Excavation Assessment and Updated Project Design © Cotswold Archaeology
refine the date of this deposit which is currently described as Roman or later. The charcoal from the fills of posthole 2242 (2243 and 2245) may be used to obtain a radiocarbon date for the two fragments of tuyères found there.
Radiocarbon dating 4.50 Due to the lack of secure dating evidence obtained from the excavations a programme of radiocarbon dating is recommended. The charred plant remains and charcoal offer the best potential for radiocarbon dating. Many features are currently undated and would benefit from this process, taking care to avoid material that could be either intrusive or residual. Ideally paired samples from primary fills of features should be used. Due to the poor preservation, none of the inhumation material is suitable for radiocarbon dating. The inhumation burial (Grave 2596) contained a copper-alloy finger ring (of Middle Iron Age/Early Roman date) and is associated with an unprocessed soil sample which, if processed, may yield charcoal or charred plant material suitable for radiocarbon dating. The undated cremation burial 1764 (C2) is the only cremation preserved enough to be suitable to send for radiocarbon dating and also has suitable oak and hazel charcoal associated with it. A good dating sequence should therefore be established for this cremation. If the cremated bone and charcoal returned similar dates this would confirm their deposition at the same time. Cremation burial C1 can be dated by use of charcoal from short lived species (see table 11). Cremation burial C4 has a broad stratigraphic date which could be refined if a radiocarbon is obtained using charcoal and charred plant material. The fills of the posthole 2242 which contained the tuyère fragments would also benefit from absolute dates and suitable charcoal has been identified. If paired samples are used in all cases then eight dates would be required. In addition, confirmation of the dating of the metallurgical residues through absolute means is viewed as highly desirable. Several oak charcoal fragments have been deemed suitable for radiocarbon dating from several Period 4 postholes associated with metalworking debris. These features currently remain undated, but in conjunction with the metalworking residues a good chronological framework could be established through a targeted programme of radiocarbon dating.
Pollen 4.51 No further work is recommended due to the poor preservation of the pollen remains.
21 Land at Domgay Lane, Four Crosses, Powys: Post-Excavation Assessment and Updated Project Design © Cotswold Archaeology
Geoarchaeology 4.52 In order to provide further information of the environment in which the barrow was originally constructed, and in particular whether Ring Ditch 14 was constructed on cultivated land, pasture or scrub, consideration should be given to the micromorphological examination of sample 178. It is also recommended that sample 177 be examined using micromorphological techniques in order to further investigate the burning event that immediately preceded the first extension of the barrow. No further work is recommended on the existing magnetic susceptibility sub-samples although it is suggested that provision be made to enable the present geoarchaeological assessment data to be worked into a publication text. This latter should be integrated with the report produced by the micromorphologist.
5. STORAGE AND CURATION
5.1 Artefactual material, including pottery, worked flint, and fired clay is stable and requires no further treatment for long-term storage. Such material is stored by context in plastic bags within acid-free, brass wire-stitched cardboard boxes. The amber bead has been treated to ensure long-term preservation. Metal artefacts have been assessed and stabilised by a specialist conservator and are currently stored in sealed, plastic boxes with humidity controlled, in accordance with the guidelines of the Society for Museum Archaeologists (1993). Suitable arrangements have been made for transfer of the site archives (both written and finds) to Powysland Museum (Welshpool) under accession number A467/2005. Until the final analysis and publication report have been completed the archives is stored at Cotswold Archaeology’s offices at Kemble, along with a security copy of the archive.
6. UPDATED AIMS AND OBJECTIVES
6.1 The original project aims and objectives have been reviewed and updated in the light of the assessment. New aims and objectives have been established and considered in the light of the research agenda for Wales (Briggs 2003).
A. Establish full integration of the cropmark, geophysical, and excavation data.
22 Land at Domgay Lane, Four Crosses, Powys: Post-Excavation Assessment and Updated Project Design © Cotswold Archaeology
6.2 Do the excavated features relate to anomalies identified in the geophysical surveys? Was the interpretation of these features correct in view of the excavation results? Do these features relate to the cropmarks previously identified in the 1960s and 1970s?
B. Establish the dates, chronology and character of early prehistoric activity.
6.3 How does the limited early prehistoric pottery assemblage recovered compare to contemporary material recovered from the vicinity of Four Crosses? Does the Late Neolithic pit and associated pottery assemblage represent significant activity or is it an isolated feature? Have other such features and/or artefactual material been identified in the vicinity?
C. Establish the dates, chronology and character of Bronze Age activity.
6.4 The spatial and chronological relationships between the ring ditches and internal features should be considered with reference to those excavated between 1981 and 1985 and those excavated during this phase of work.
6.5 The spatial and chronological relationship between the ring ditches and the hillfort at Breiddin, and other sites in the locality such as Llanymynech Hill, Collfryn and Trelystan upland cemetery, should be considered as should the artefactual and ecofactual assemblages recovered from them (CPAT 2004). How does this Bronze Age to Iron Age activity relate to the broader landscape? How did the activity in this landscape change through time?
D. Establish the dates, chronology and character of undated activity.
6.6 Attempt to refine the dating and chronology of the undated pit alignment and its relationship with other known pit alignments in the locality, the Bronze barrow cemetery and subsequent Roman field system.
E. Establish the dates, chronology and character of Romano-British activity.
6.7 The spatial and chronological relationship between the field system, area of metalworking debris and cremations should be considered. How does this activity relate to contemporary features and/or finds in the locality?
23 Land at Domgay Lane, Four Crosses, Powys: Post-Excavation Assessment and Updated Project Design © Cotswold Archaeology
6.8 The presence of small-scale copper smelting would appear to be of early Roman date and may well add to the emerging corpus of information about the early development of copper working in the region and beyond.
6.9 The XRF analysis of the copper-alloy ring in conjunction with the detailed analysis of the micro-residues and other debris will clarify the potential source of the ore and may address the role of Llanymynech Hill as a potential source of ore at this time. A considered programme of AMS dating would therefore be highly desirable.
7. PUBLICATION
7.1 The results from this excavation merit publication and are of obvious regional significance, it is proposed that a full report be published in Archaeological Journal.
Synopsis of Proposed Report
Further Ring-ditches at Four Crosses, Llandysilio, Powys, 2004-2006
by, Tim Havard, Neil Holbrook and Tim Darvill
with contributions by Rowena Gale, Teresa Gilmore, Helen Lewis, E.R. McSloy, Elizabeth Pearson, Sylvia Warman, Tim Young and Keith Wilkinson
Abstract Brief summary of main findings of the project 200 words (1/4 page)
Introduction Project background, archaeological background, topography, geology 500 words (1/2 page) Excavation Results (Tim Havard) Chronological discussion of the major phases and features of the site 3000 words (3 pages)
The Finds
24 Land at Domgay Lane, Four Crosses, Powys: Post-Excavation Assessment and Updated Project Design © Cotswold Archaeology
Neolithic Pottery (E.R. McSloy) 750 words Amber Bead (E.R. McSloy) 250 words Finger ring (E.R. McSloy) 750 words Metallurgical residues (Tim Young) 1500 words
Biological evidence Human Bone (Teresa Gilmore) 750 words Charred Plant Remains (Elizabeth Pearson) 1000 words Radiocarbon Dating (Sylvia Warman) 500 words Charcoal (Rowena Gale) 1500 words
Geoarchaeology Geoarchaeology (Dr. Keith Wilkinson) 1000 words Micromorphology (Dr. Helen Lewis) 1000 words 9000 words (9 pages)
Discussion Prehistoric (Timothy Darvill) 1800 words Romano-British (Neil Holbrook) 900 words 2700 words (2.5 pages)
Acknowledgements & Bibliography (Annette Hancocks) 800 words (1.5 pages) TOTAL 16,000 words (16 pages)
Illustrations: Location of site 1 page Site plans with phasing/sections 5 pages Pottery ½ page Amber Bead ½ page Finger Ring ½ page Metallurgical Residues 3 pages Geoarchaeology/Micromorphology 2 pages 12.5 pages
25 Land at Domgay Lane, Four Crosses, Powys: Post-Excavation Assessment and Updated Project Design © Cotswold Archaeology
Tables: Pottery: 1 page Charred Plant Remains 1 page 2 pages
Total Publication Estimate: 30.5 pages
8. PROJECT TEAM
8.1 The post-excavation and publication programme will be under the management of Annette Hancocks MIFA (PX Manager), who will co-ordinate the work of the following personnel:
Mary Alexander MIFA (PX Manager: PXM): Quality-assure stratigraphic analysis
Tim Havard AIFA (Project Officer: PO): Post-excavation phasing, draft report preparation, research and archive.
Neil Holbrook MIFA Romano-British Discussion
E.R. McSloy MIFA (Finds Officer: FO): Specialist report preparation and liaison, post-excavation phasing.
Sylvia Warman MIFA (Environmental Officer: EO): Specialist report preparation and liaison,
Teresa Gilmore (Finds Supervisor: FS): Specialist report preparation
Peter Moore (Senior Illustrator: SI): Production of all site plans, sections and artefact drawings
26 Land at Domgay Lane, Four Crosses, Powys: Post-Excavation Assessment and Updated Project Design © Cotswold Archaeology
8.2 Contributions by the following external consultants will be co-ordinated by the Post- Excavation Manager and Finds and Environmental Officers
Prof. Timothy Darvill (University of Bournemouth) Prehistoric discussion University of Waikato (New Zealand): Radiocarbon dating Dr TimYoung Metallurgical residues Dr Keith Wilkinson Geoarchaeological work Dr Helen Lewis Micromorphology Liz Pearson Charred Plant Remains Rowena Gale Charcoal 8.3 The final publication report will be edited and refereed internally by CA senior project management, and will be externally refereed by Frances Lynch prior to submission to the Archaeological Journal.
27 Land at Domgay Lane, Four Crosses, Powys: Post-Excavation Assessment and Updated Project Design © Cotswold Archaeology
9. TASK LIST
TASK PERSONNEL DURATION/ COST Project Management PXM 6 days Stratigraphic Analysis PO 10 days PXM 3 days FO 1 day Research, comparanda PM 1 day PO 1 day Pottery Analysis and report FO 1.5 days Illustration (1 vessel) SI 1 day Small finds Analysis and report FO 2.5 days Illustration (2 items) Ill 0.5 day XRF Specialist FEE Metallurgical Residues Analysis and report Specialist FEE Human Bone Analysis and Report FS 3 days Charred Plant Remains Processing & sorting of 18 samples FP 4 days Preparation of publication report Specialist FEE Charcoal Analysis and report Specialist FEE Micromorphology Reporting and analysis Specialist FEE Geoarchaeology Reporting and analysis Specialist FEE Radiocarbon dating Analysis (6 AMS dates) Specialist FEE Report preparation EO 1 day Preparation of publication report Abstract and introduction PO 1 day SI 1 day Excavation results PO 5 days SI 4 days Compilation of specialist reports, tables etc. PXM 2 days Discussion, conclusions TD FEE NH 1 day SI 1 day PM 1 day Acknowledgements, bibliography PXM 1 day Submission to referees Quality Assurance HP 2 days Editing PXM 2 days Revision PO 2 days SUBMISSION OF PUBLICATION TEXT Final editing and proofs PXM 3 days TD 1 day NH 1 day PO 3 days Archive Research archive completion PO 0.5 day FO 0.5 day Microfilm external FEE Deposition museum FEE
28 Land at Domgay Lane, Four Crosses, Powys: Post-Excavation Assessment and Updated Project Design © Cotswold Archaeology
Publication Printing 30.5 pages Journal FEE
10. TIMETABLE
10.1 For a publication project, CA would normally aim to have completed a publication draft within twelve months of approval of the updated analysis and publication project design. A detailed programme will be produced on approval of the updated publication project design.
11. REFERENCES
AEA (Association for Environmental Archaeology) 1995 Environmental archaeology and archaeological evaluations. Recommendations concerning the environmental component of archaeological evaluations in England, Working Papers of the Association for Environmental Archaeology, 2 (online). Available from http://www.envarch.net/evaluations.html Accessed 01/07/02
ArchaeoPhysica Ltd. 2002 Magnetic gradient and electrical resistance surveys at Four Crosses, Powys: Geophysical Survey Report
Allen, J.R.L. 1986 ‘Interpretation of some Romano-British smithing slags from Awre, Gloucs’, Historical Metallurgy 20, 97-104
Beck, C. and Shennan, S. 1991 Amber in Prehistoric Britain Oxford: Oxbow Monograph 8
Beijerinck, W. 1947 Zadenatlas der Nederlandsche Flora Wagoningen: Veenman and Zonen
Bennett, K.D. 1994 Annotated catalogue of pollen and pteridophyte spore types of the British Isles Department of Plant Sciences University of Cambridge unpublished report
BGS (British Geological Survey) 1990 Mid Wales and Marches: Solid Geology Sheet 52˚N 04˚W
29 Land at Domgay Lane, Four Crosses, Powys: Post-Excavation Assessment and Updated Project Design © Cotswold Archaeology
Briggs, S. 2003 Towards a Research Agenda for Welsh Archaeology Brit. Archaeol. Rep. Brit. Ser. 343 Oxford: British Archaeological Reports
Brothwell, D.R. 1981 Digging up Bones British Museum (Natural History) Oxford University Press
Buisktra, J.E. and Ubelaker, D.H. (eds) 1994 Standards for data collections from human skeletal remains Arkansas Archaeological Survey Research Series No. 44, Fayetteville Arkansas Archaeological Survey
Bulleid, A. and Gray, H.G. 1917 The Glastonbury Lake Village ii Glastonbury Taunton: Glastonbury Antiquarian Society
Caseldine, A.S. 1990 Environmental archaeology in Wales Lampeter: Saint David's University College
Caseldine, A. 2004 A Research Framework for the Archaeology of Wales: Nationals Seminar Paper, 4th September 2004 – Environmental Archaeology. Available from http://www.cpat.org.uk/research/ns4env.htm, accessed 4th August 2006
Chambers, F.M. 1982a ‘Environmental history of Cefn Gwernffrwd, near Rhandirmwyn, Mid- Wales’, New Phytologist, 92, 607-15
Chambers, F.M. 1982b ‘Two radiocarbon-dated pollen diagrams from high altitude blanket peats in South Wales’, J. Ecol. 70, 445-59
Chambers, F.M. 1983 ‘Three radiocarbon-dated pollen diagrams from upland peats north- west of Merthyr Tydfil, South Wales’, J. Ecol. 71, 475-87
Cleal, R. and MacSween, A. 1999 Grooved Ware in Britain and Ireland: Neolithic Studies Group Seminar Papers 3 Oxford. Oxbow Books
CPAT (Clwyd Powys Archaeological Trust) 2004 Proposed Residential Development On Land Off Domgay Lane, Four Crosses: Design Brief For Archaeological Excavation
30 Land at Domgay Lane, Four Crosses, Powys: Post-Excavation Assessment and Updated Project Design © Cotswold Archaeology
Crampton, C.B. and Webley, D.P. 1964 ‘Preliminary studies of the historic succession of plants and soils on selected archaeological sites in South Wales’, Bull. Board Celtic Stud. 20, 440-9
Crew, P. 1996 ‘Bloom refining and smithing, slags and other residues’, Historical Metallurgy Society Archaeology Datasheet No 6
Doring, P. and Chambers, F.M. 1990 ‘Field survey, excavation and pollen analysis at Mynydd y Drum, Ystragynlais, Powys, 1993 and 1987’, Bull. Board Celtic Stud., 37, 215-46
EH (English Heritage) 2002 Environmental Archaeology: A guide to the theory and practice of methods, from sampling and recovery to post-excavation English Heritage Centre for Archaeology Guidelines 2002/01
Gale, R. and Cutler, D. 2000 Plants in Archaeology Otley/London: Westbury Publishing and Royal Botanic Gardens, Kew
Gale, S.J. and Hoare, P.J. 1991 Quaternary sediments: petrographic methods for the study of unlithified rocks London: Belhaven Press
Head, K. 2006 Pollen remains from Domgay Lane, Four Crosses, Powys. Worcestershire Historic Environment and Archaeology Service internal report, 1449
IFA (Institute of Field Archaeologists) 1999 Standard and guidance for archaeological excavation Institute of Field Archaeologists
Jones, A.P., Tucker, M.E. and Hart, J.K. 1999 ‘Guidelines and recommendations,’ in Jones et al. 1999, 27-76
Jones, A.P., Tucker, M.E. and Hart, J.K. (Eds.) 1999 The description and analysis of Quaternary stratigraphic field sections Quaternary Research Association technical guide 7, London: Quaternary Research Association
31 Land at Domgay Lane, Four Crosses, Powys: Post-Excavation Assessment and Updated Project Design © Cotswold Archaeology
Le Borgne, E. 1960 ‘Influence du feu sur les propriétés magnétiques du sol et sur celles du schiste et du granite’, Annals of Geophysics 16, 159-95
Longworth, I.H. 1971 ‘The Neolithic pottery’, in Wainwright and Longworth 1971, 75-124
MacGregor, A. and Bolick, E. 1993 A summary catalogue of the Anglo-Saxon Collections (Non-Ferrous Metals) in the Ashmolean Museum Brit. Archaeol. Rep. Brit. Ser. 230 Oxford: BAR
Maher, B.A. 1986 ’Characterization of soils by mineral magnetic measurements’, Physics of Earth and Planetary Interiors 42, 76-92
McKinley, J.L. 1993 ‘Bone Fragment size and weights of bone from modern British cremations and their implications for the interpretation of archaeological cremations’, Internat. J Osteoarch. 3, 283-7
Manchester, K. and Roberts, C.A. 1995 Archaeology of Disease Stroud: Sutton Publishing
Marches Archaeology 2003 Land off Domgay Lane, Four Crosses, Powys: Report on an archaeological evaluation Marches Archaeology Series 297
Miket, R. 1985 ‘Ritual enclosures at Whitton Hill, Northumberland’, Proc. Prehist. Soc., 51, 137-48
Milles, A. 1986 ‘Charred plant remains from a beaker pit at Site 2, Four Crosses, Microfiche 2’, in Warrilow et al. 1986, Microfiche 2, 10
Moore, P.D., Webb, J.A. and Collinson, M.E. 1991 Pollen analysis Oxford: Blackwell Scientific Publications
Munsell Color 2000 Munsell soil color charts Munsel Color New York: New Windsor
Musson, C.R., Britnell, W.J., Northover, J.P. and Salter, C.J. 1992 ‘Excavations and metal- working at Llwyn Bryn-dinas Hillfort, Llangedwyn, Clwyd’, Proc. Prehist. Soc. 58, 265-83
32 Land at Domgay Lane, Four Crosses, Powys: Post-Excavation Assessment and Updated Project Design © Cotswold Archaeology
Musson, C.R. and Northover, J.P. 1989 ‘Llanymynech Hillfort, Powys and Shropshire: observations on construction work, 1981’, Montgomeryshire Collect. 77, 15-26
Musson, C.R. 1991 The Breiddin Hillfort: a later Prehistoric Settlement in the Welsh Marches Counc. Brit. Archaeol. Res. Report 76, London: Council for British Archaeology
Northover, J.P. 1991 ‘Further comments on the composition of the Breiddin bronzes’, in Musson 1991, microfiche appendix 13.4
Pearson, E. 2005 Environmental remains from Walton, Transco Newbold Pacey (Warwickshire) to Honeybourne (Worcestershire) pipeline Worcestershire Historic Environment and Archaeology Service internal report, 1300
Sharples, N. (ed.) 2005 A Norse Farmstead in the Outer Hebrides. Excavations at Mound 3, Bornais, South Uist Cardiff Studies in Archaeology Oxford: Oxbow Books
Stace, C. 2001 New Flora of the British Isles Cambridge: Cambridge University Press
Starley, D. 1995 Hammerscale Historical Metallurgy Society, Archaeology Datasheet No 10
Schwartz, J.H. 1995 Skeleton Keys: An Introduction to Human Skeletal Morphology, Development and Analysis. New York: Oxford University Press
Tucker, M.E. 1982 Sedimentary rocks in the field Chichester:Wiley
Tylecote, R.F. 1986 The Prehistory of Metallurgy in the British Isles London: The Institute of Metals
Unglik, H. 1991 ‘Observations on the structures and formation of microscopic smithing residues from the Bixby Blacksmith Shop at Barre Four Corners, Massachusetts, 1824-55’, Hist. Metal. 25, 92-8
Van der Veen, M. 1985. ‘Plant remains’, in Miket 1985, 146-7
33 Land at Domgay Lane, Four Crosses, Powys: Post-Excavation Assessment and Updated Project Design © Cotswold Archaeology
Wainwright, G.J. and Longworth, I.H. 1971 Durrington Walls: Excavations 1966-1968 London, Research Committee of the Society of Antiquaries of London Report 37
Warrilow, W., Owen G. and Britnell, W. 1986 ‘Eight Ring-ditches at Four Crosses. Llandysilio, Powys, 1981-85’, Proc. Prehist. Soc. 52, 53-87
Webster, P.V. 1976 ‘Severn Valley ware: A Preliminary Study’, Trans. Bristol Gloucestershire Archaeol. Soc. 94, 18-46
Whitehead, R. 1996 Buckles 1250-1800 Chelmsford: Greenlight Publishing
Young, T. 2005 ‘Site Activities: slag and related materials’, in Sharples 2005, 174-6
34 Land at Domgay Lane, Four Crosses, Powys: Post-Excavation Assessment and Updated Project Design © Cotswold Archaeology
APPENDIX 1: STRATIGRAPHY BY TIM HAVARD
Quantification 2054 contexts were recorded during the course of the project, of these contexts, 19 were unassigned and 2 were assigned to evaluation trenches. Additionally, 225 contexts recorded during the evaluation. The remaining 1808 contexts were assigned to the following provisional periods:
Natural/geological (Period 0) 57 Neolithic (Period 1) 3 Bronze Age (Period 2) 567 Iron Age (Period 3) 2 Romano-British (Period 4) 456 post-medieval (Period 5) 242 modern (Period 6) 46 Undated 435
Total 1808
Potential for further analysis Further analysis is proposed for Periods 1, 2, 3 and 4. The features assigned to these periods are generally well- preserved and are of archaeological significance. The features dating to Periods 5 and 6 generally relate to modern field boundaries or drains and are of little archaeological interest. Of the undated features, further analysis is proposed on the undated pit alignments (161 contexts) and postholes in the vicinity of areas of metalworking residue (approximately 90 contexts). Further analysis is therefore proposed for 1336 contexts.
Stratigraphic analysis Project Officer 10 days Post-Excavation Manager 3 days Finds Officer 1 day
35 Land at Domgay Lane, Four Crosses, Powys: Post-Excavation Assessment and Updated Project Design © Cotswold Archaeology
APPENDIX 2: WORKED FLINT BY E.R. McSLOY
Worked flint is restricted to three waste flakes, one of which is unstratified. Absence of secondary working inhibits close dating. The irregularity and overall proportions of the three flake removals is perhaps most likely to indicate Late Neolithic to Bronze Age dating.
All material was hand collected from site during the normal course of excavations. Of the stratified pieces, one flake derives Period 2 Ditch 5004 and the second from Pit 2630, part of undated pit alignment 1. Condition is poor with all pieces broken or showing signs of edge damage which is perhaps suggestive of re-deposition. Raw material consists of unpatinated poor quality grey-brown coloured flint (two pieces) and yellowish-brown (honey- coloured) flint. Only the latter retains any cortex and this suggests derivation from a secondary (probably gravel) source.
Statement of potential and requirements for analysis
The worked flint represents a small and restricted assemblage which presents no potential for further analysis. Extant records and reporting are considered adequate for archive purposes and no further work is recommended.
36 Land at Domgay Lane, Four Crosses, Powys: Post-Excavation Assessment and Updated Project Design © Cotswold Archaeology
APPENDIX 3: POTTERY BY E.R. McSLOY
Pottery of all periods, recovered from the excavation, amounted to 167 sherds (2169g), of which 10 sherds were recovered as part of the evaluation undertaken by Marches Archaeology (2003). Condition of the pottery was very variable and is described below relative to each dated group.
Prehistoric A total of 30 sherds (952g) of Earlier Prehistoric date (Neolithic to Bronze Age) were recovered. An unstratified sherd in a fine grog-tempered fabric and a sherd in a fabric with coarse quartz and leached calcareous inclusions were undecorated and attributable on the basis of fabric and firing characteristics. The condition of both sherds is poor, with surfaces heavily abraded.
Of greater significance are 28 sherds (947g) from a single Grooved ware vessel of Late Neolithic date, which were recovered from fill 2493 of Pit 2548. The condition of this vessel, which occurs in a coarse, grog-tempered fabric is excellent. Sufficient of the vessel is present for its decorative scheme to be reconstructed in full: the rim zone, which is separated from the body by a horizontal raised cordon is further sub-divided into panels with incised infill by diagonal cordons. The body zone is divided by vertical cordons into panels of alternating broad and narrower width filled with incised diagonal lines. The decoration and the large, bucket-like proportions of the vessel are distinctive of the Durrington Walls style, which is the most widespread of the styles defined by Longworth (1971).
Grooved ware in England and Wales coincides with the Later Neolithic period from c. 2700 to c. 2200 BC. Utilitarian function suggested by the form of Grooved ware vessels is supported by excavated assemblages, primarily from southern British sites and presence of carbonised and/or lipid residues. A similar interpretation would appear to be likely in this instance, indicated by thick sooting to its outer surfaces. Grooved ware was not recovered from previous excavations from Four Crosses (Warrilow et al. 1986) and is rare generally from Wales. Finds are however recorded in the region of Welshpool, approximately 12km to the south (Cleal and MacSween 1999, 204).
Roman Pottery of Roman date amounted to 25 sherds (460g) from the excavation. An additional 4 sherds of Roman pottery were recovered from the evaluation. Comment on this material by Jeremy Evans is included in this assessment.
The condition of the Roman pottery is generally poor, with considerable surface deterioration in some instances. Severn Valley wares, which make up the majority of the assemblage were particularly adversely affected.
A narrow range of fabrics is represented (Table 1), which is typical for the Roman period in the area. Most material is present as bodysherds, making determination of form impossible in most instances. Material of non- local origin consists of a Dorset Black-Burnished ware (BB1) sherd from ditch E922 and 3 sherds from a handmade vessel in Malvernian rock-tempered ware from fill 2063 ditch 2056 (Curvilinear ditch 3). The Malvernian vessel, probably a bowl or dish, exhibits burnished lattice decoration, in imitation of Black-Burnished ware. The remainder of the assemblage consists of Severn Valley ware which is almost certainly local in origin,
37 Land at Domgay Lane, Four Crosses, Powys: Post-Excavation Assessment and Updated Project Design © Cotswold Archaeology
the fabric similar to Warwickshire Museum fabric O23. Forms are restricted to jars, including wide-mouthed, hooked-rim forms (Webster’s type C; 1976, 25-7) from ditches E922 and fill 2052 of ditch 2051 (Enclosure 1).
The small and restricted nature of the assemblage means it is of limited significance either for dating or understanding site function or ‘status’. Severn Valley and related oxidised wares are long lived, extending in use between the 1st and 4th centuries in the region. The BB1 sherd from Enclosure 1 suggests dating after c. AD 120 and the Malvernian sherds, a date most likely in the 2nd to earlier 3rd century range.
Medieval A single unstratified sherd in a fine sandy redware fabric was recovered during the evaluation and identified by Stephanie Rátkai (2003).
Post-medieval and modern 102 sherds (620g) recovered from the excavation and a further 5 sherds from the evaluation (examined by Stephanie Rátkai) dates to the post-medieval/modern period and most to the 18th and 19th centuries.
Earliest pottery present consists of two sherds of Midlands Purple type wares dating to between the 15th and 17th centuries. Of this material only that from Pit E607 was stratified. The bulk of the assemblage is composed of slipwares and mottled brown glazed earthenwares, almost certainly originating from Staffordshire date to the final years of the 17th and throughout the 18th centuries. Such material is mainly in poor condition and residual, occurring s in most instances alongside material of probable 19th century date. Pottery of 19th century date consists of creamwares/refined whitewares including transfer-decorated ‘china’; black-glazed and unglazed flowerpot type earthenwares and English stonewares.
Statement of potential and requirements for further analysis Small quantities of pottery were recovered. Most significant is an unusually complete and well-preserved Grooved ware vessel of Late Neolithic from fill 2493 of Pit 2548. The small Roman and post-medieval pottery assemblage present very little potential for further analysis and little further work is required.
Perhaps unusually, no material was recovered from burial deposits associated with excavated ring-ditch monuments. Later Prehistoric pottery was entirely absent, despite previously attested radiocarbon determinations and non-ceramic finds of this date. A largely aceramic local Iron Age tradition likely accounts for this absence.
Recommendations The Grooved ware vessel merits illustration as of intrinsic interest and as evidence for possible ‘domestic’ Neolithic activity on the site. This vessel, which is certainly among the most complete and best-preserved examples of its kind form Wales should be drawn. A short report fully describing this vessel and summarising other ceramic material should be prepared for publication.
Report production 1.5 days (FO) Illustration 1 day (SI)
38 Land at Domgay Lane, Four Crosses, Powys: Post-Excavation Assessment and Updated Project Design © Cotswold Archaeology
Table 1: Summary of pottery by Period Period Type Count Weight Comments (Evaluation contexts) Roman 4 - SVW, BB1 med 1 2 Fine sandy redware pmed 5 135 Midlands Purple, unglazed redware Subtotal 10 137 1 Prehist. 28 947 Ra. 15. Grooved ware vessel 2 Roman 2 18 SVW pmed/mod 2 9 modern china, flowerpot 4 Roman 20 429 SVW; gritty oxidised; Malvernian pmed/mod 4 22 Misc. glazed earthenwares (intrus.) 5 pmed/mod 9 46 Black-glazed earthenware; china, flowerpot 6 pmed/mod 82 520 C18 slipwares + iron glazed + white china Roman 3 13 RB (SVW) Unphased Prehist. 1 2 SS 137. 1 x calcar.-temp BA pmed/mod 1 0 SS 137. pmed black-glazed C18+ pmed/mod 1 3 C19+ china Unstratified pmed/mod 1 9 brown glazed pmed: C18+ Prehist. 1 3 grog-temp - Beaker/BA? mod 2 11 C19 china Subtotal 157 2032 Total 167 2169
39 Land at Domgay Lane, Four Crosses, Powys: Post-Excavation Assessment and Updated Project Design © Cotswold Archaeology
APPENDIX 4: WORKED STONE BY E.R. McSLOY
Two items of worked stone, a spindlewhorl (Ra. 3) from Period 4 fill 2052 of ditch 2051 (Enclosure 1) and polisher from Period 5 ditch fill 3034 of ditch 3033 (Field System 2). Neither item is dateable by form although each was associated with Roman pottery and probably belong to this period. The stone used for spindlewhorl Ra. 3 appears to be a micaceous mudstone, which is sufficiently soft for it to be shaped and clearly drilled. The polisher is of harder fine-grained sandstone. The stone for both items could have been obtained locally.
Statement of potential and requirements for analysis
The stone objects require catalogue descriptions for the purposes of the archive to include confirmation of stone type. Beyond this further work is not recommended.
Catalogue descriptions 0.5 day (FO)
40 Land at Domgay Lane, Four Crosses, Powys: Post-Excavation Assessment and Updated Project Design © Cotswold Archaeology
APPENDIX 5: AMBER BEAD BY E.R. McSLOY
A single small flat, disc-like bead of amber was recovered from fill 1328 of Period 2 grave 1327. This was the sole find from the context and retrieved from soil sample 66. Following retrieval this item was examined by a specialist conservator and chemically treated to ensure long-term preservation.
The Bronze Age dating for (possible) grave 1327 was prompted by its alignment with pit alignment 1, although there is some ambiguity in the relationship between this feature and Field System 2. Amber, almost certainly originating in the Baltic, is attested archaeologically in Britain from the Neolithic and subsequent periods (Beck and Shennan 1991). As an isolated find this example is difficult to date, however its small size (9mm in diam.) and form (approximating Beck and Shennan’s Group 1b), make an earlier prehistoric, probably Early Bronze Age date, most likely.
Statement of potential and requirements for analysis
Although potentially re-deposited, this item is of significance as one of very few artefacts from the site of probable earlier prehistoric dating. As such and as a possible grave find, this object should be drawn and fully described for the publication. Further research, particularly with regard to regional occurrence of amber beads of corresponding form, is recommended and may help confirm dating.
Description and research 1 day (FO) Illustration 1 hour (SI)
41 Land at Domgay Lane, Four Crosses, Powys: Post-Excavation Assessment and Updated Project Design © Cotswold Archaeology
APPENDIX 6: METALWORK BY E.R. McSLOY
Eleven items of iron and two of copper-alloy were recovered. Iron items comprise mainly nails of which shank fragment from Period 5 ditch 1480, fill 1479 (Trackway 1) is likely to be Roman and the remainder of post- medieval date. Hobnails of Roman date were recovered from Period 4 ditch 2051 (Enclosure 1) and unphased posthole 2331.
Copper-alloy objects Ra. 5 from fill 2106 Ring Ditch 6 and Ra. 13 from grave 2596 (interior to Ring Ditch 7) derived from contexts attributed to Period 3 (Iron Age). Ra. 13 consists of a finger ring made from a strip of metal with overlapping tapered terminals. Simple coil/spiral rings similar in form to Ra. 13 are most often associated with the Middle or Late Iron Age and Early Roman periods and there exist large groups of this date from Glastonbury Lake Village (Bulleid and Gray 1911, 209-10). The type also occurs in the post-Roman period as indicated by Anglo-Saxon grave finds of 5th to 6th century date from the Thames Valley (MacGregor and Bolick 1993, 169-71).
Ra. 5 consists of a small D-shaped buckle frame of a type seeing common usage in the period c. 1250-1650 (Whitehead 1996, 16-7). The small size of this item, which misses its pin and plate, could lend weight to the notion that this is an intrusive find.
Statement of potential and requirements for analysis
The metalwork assemblage is of limited usefulness, either as dating evidence or as evidence of site function and as such warrants little further attention. A catalogue, listing all items should be prepared for the purposes of the archive. A description, listing parallels and a drawing of the most significant item, finger ring Ra. 13 should be included with the publication.
Ring Ra. 13 is of interest as the only artefact associated with a grave located within any of the excavated ring ditch monuments. Its dating, either to the Iron Age or (less likely) the Post-Roman period is at odds with the provisional dating for all the ring ditch monuments and, if possible, this discrepancy should be addressed at the analysis using radiocarbon dating of associated material. It is further recommended that the composition of this item should be determined through x-ray fluorescence analysis. Similar analysis has been used previously to establish a link between a group of late prehistoric artefacts from central and northern Wales (Musson and Northover 1989; Northover 1991; Musson et al. 1992), and copper deposits characterised by high natural zinc content, from Llanymynech Hill, 3km north-west of the site.
Catalogue and research 1 day (FO) XRF analysis (FEE) Illustration 2 hours (SI)
42 Land at Domgay Lane, Four Crosses, Powys: Post-Excavation Assessment and Updated Project Design © Cotswold Archaeology
APPENDIX 7: METALLURGICAL RESIDUES BY DR T.P. YOUNG
Summary Metallurgical residues from the site include a very small amount of microresidues suggestive of ironworking and a larger amount of residues from copper (or copper-alloy) working.
The evidence for ironworking is very sparse and includes only micro-residues in very low concentrations, recovered from a wide range of contexts from Bronze Age to Roman. At such low densities such material may well be intrusive into the earlier contexts and the small amount recovered (together with the complete lack of associated macro-residues) suggests that iron working was not a significant activity within the excavated area.
The evidence for copper working is entirely from contexts interpreted as Roman. Much of the material was recovered from features interpreted as postholes, although there are strong indications that these in fact may have been hearths or furnaces. The assemblages are consistent in composition, including small droplets of copper or copper-alloy, a small proportion of slags, and clay lining. Air blast to the hearths/furnaces seems to have been through pre-formed tuyères, of rectangular or D-shaped block form. Use of a pre-formed tuyère may have been necessitated by the poor quality of the gravely-fired clay.
The site has not produced any evidence for crucibles or moulds, but one hearth did produce a single fragment of probable azurite-bearing ore. The analyses of archaeometallurgical ceramics previously undertaken for CA by David Dungworth indicated working of a copper-alloy bearing lead, arsenic and zinc, but not tin. This suite of elements has been associated with copper produced from the Llanymynech deposits just to the north of the site.
No Roman copper smelting furnaces are yet known in Britain. This assemblage would add significantly to understanding if the suggested identification were to be confirmed and would provide a discovery of national significance.
Methods All the material submitted has been examined by eye, or under a low-powered binocular microscope. This allows attribution of residues to particular classes, but the interpretation in this assessment is necessarily limited by the scope of the observations.
Results Micro residues from ironworking Micro residues from ironworking include the two classes of hammerscale: spheroidal and flake. Both classes are represented by very small amounts of material. Spheroidal hammerscale was recorded from undated features: primary burial fill 2263 (Grave 2262), primary posthole fill 2336 and third fill (1919) of Pit 1916. In addition, possible flake hammerscale occurred in undated posthole 2335 context 2336. In each case the hammerscale was represented by only a single particle. Indeterminate microscopic slag fragments occurred in the final fill 1420 of Ring Ditch 2; secondary fill 1484 of Ring Ditch 4 and primary fill 1975 of Pit 1974.
Two further undated contexts each associated with the pit alignment 1, yielded hammerscale: primary fill 1404 of Pit 1403 yielded two pieces of flake hammerscale and the primary fill 1723 of Pit 1722 produced a possible microscopic slag fragment.
43 Land at Domgay Lane, Four Crosses, Powys: Post-Excavation Assessment and Updated Project Design © Cotswold Archaeology
Period 4 (Romano-British) contexts yielding hammerscale were rather more abundant and provide slightly more evidence for ironworking at this period. Flake hammerscale was recorded from final fill 2123 of Pit 2121; primary fill 2153 of Pit 2152 and primary fill 2510 of Pit 2509. Spheroidal hammerscale occurred in the primary fill 2437 of Grave 2436. Both classes of hammerscale occurred together in primary contexts 2345/2346, 2371 and 2431 from postholes and the primary fill 2489 of posthole 2488.
In addition, spheroidal hammerscale was recovered from undated pits 1796, 2210, 2281 and 2373 (which also contained flake hammerscale), all of which yielded assemblages otherwise indicative of copper working. It is uncertain whether these spheroids are a residue from the copper working process, or whether they are indicative of iron working.
Possible copper smelting residues Residues from probable copper working are extremely significant. Each assemblage from the site may contain some or all of various classes of residue: slag, metal droplets, fired gravely clay, fired prepared clay (including tuyères and probable daub superstructure) and an iron- and charcoal-rich sintery material. Such assemblages derive from fifteen undated features, variously described: pits 1766, 1796, 2034, 2071, 2219, 2257, 2290 2297, 2313; postholes 2210, 2242, 2252, 2272, 2281 and cut feature 2373.
The rich assemblages of fine-grained residues in these contexts, together with the slabs of friable hearth or furnace lining, suggest that these contexts represent either in situ metalworking, or primary dumps from hearth clearance. One context (2220; primary fill of pit 2219) yielded a very small fragment of what appears to be an azurite-bearing copper ore.
Fired clay associated with possible copper smelting residues The fired clay from these contexts contained four broad categories of material, summarised below, all with variable degrees of vitrification:
• fired porous gravely clay. It is not clear whether this is a construction material, or simply the natural substrate; • prepared lining. This material is of a fine-grained reduced-fired clay, sometimes containing some organic temper and having a smooth surface, which appears to have been applied to the surface of material of type 1; • prepared clay superstructure? This material is of a fine-grained clay and contains holes of 20-25mm diameter at low angle to its face. The hole size is similar to the bore of the tuyère material (see below), but the low angle of the holes to the surface suggests that instead these may be the impressions of wattles supporting a furnace superstructure; • tuyère. Three conjoining fragments (contexts 2243 and 2245) from posthole 2242 indicate a flat-sided tuyère block. It is uncertain whether the block would have originally been rectangular (compare the iron- working tuyère from Carpow; Tylecote 1986), or whether the flat side is a planar base and the whole tuyère had a “D”-shaped section (compare the Saxon tuyères from Rook Hall, Essex).
The slags from these contexts are relatively sparse and occur in small blebby pieces. A few pieces of dense black glassy slags also occurred. Many of the slags may be the result of interaction between copper and the debris in the hearth/furnace. Small droplets of heavily corroded copper (or copper alloy) metal occur widely in these contexts.
44 Land at Domgay Lane, Four Crosses, Powys: Post-Excavation Assessment and Updated Project Design © Cotswold Archaeology
Other materials Small indeterminate slag fragments occurred in a number of the sieved residues.
Most of the sieved samples did not contain slag, but were dominated by magnetic grains of natural rock (79 samples; probably mainly heated rather than originally magnetic) and by large quantities of burnt organic material with a coked texture (32 samples). This material contains particles originally both of wood/charcoal and bone, but for most particles the original material was not determinable from visual inspection. A few samples contained iron-rich magnetic material with a rough flake-like morphology; whether these were derived from oxidation crusts on rocks or from rusted iron is not known.
Period 2 context 1369, the primary fill of Ring Ditch 3 yielded a tiny strand of copper-alloy, probably either a small waste scarf fragment from metalworking, or possibly from damage to an artefact.
Interpretation The locations providing evidence for copper processing are scattered across the south-western part of the site. Features 1766 and 1796 were interior to large Ring Ditch 1. Features 2034 and 2071 lie close to the north-west limits of the site. The main group of 11 features lies just to the west of Period 4 Boundary Ditch 1. The linear arrangement of features 2272, 2210, 2281, 2297, 2290, 2313, parallel with Boundary Ditch 1, is the main evidence for a Roman date. Features 2219, 2242, 2252, 2257, lie 10m north-west of this linear group and posthole 2373 a further 5m distant to the west.
The complete lack of evidence for crucibles or moulds from the site, coupled with the lack of evidence for tin contamination of the metallurgical ceramics previously examined by David Dungworth for CA, and with the possible example of copper ore, all suggest that the processing carried out on the site may have been primary smelting. Llanymynech Hill, 3km north-north-west of the site, was a significant source of copper ore in early times. The ore appears to be characterised by a high natural zinc content, and use of copper from this source has been implicated in a group of prehistoric artefacts from central and northern Wales (Musson and Northover 1989; Northover 1991; Musson et al. 1992). The extensive use of intentional copper-zinc alloys (brass, gunmetal) in the Roman period means that the use of zinc-rich copper is not readily identifiable.
It has been suggested, on the basis of the high iron content of some of the high-zinc copper materials from the area, that the ore smelted at Llanymynech would have been a chalcopyrite ore (Northover in Musson et al. 1992, 275). The smelting of chalcopyrite would be expected to produce significant quantities of iron-rich slag, which have not been observed. The possible azurite ore fragment, suggests that carbonate ores might also have been available, and these could have been smelted with very little slag production.
Statement of potential The possibility of copper smelting being undertaken on the site requires detailed investigation. If confirmed this would be of national significance. Extremely few early copper smelting sites have been identified in Britain and no Iron Age or Roman smelting furnaces are known in Britain at all. The site may therefore have potential to enhance understanding of one of Britain’s copper sources and also to contribute to understanding of furnace technology. On a much smaller scale, the occurrence of slag spheroids in the copper processing residues is unexpected. Although the best-known form of slag spheroid is spheroidal hammerscale produced during iron working (Allen 1986; Starley 1995; Unglik 1991; Crew 1996), other origins for such bodies are known (e.g. production from iron-rich fuel ash, Young 2005), and the origin of these spheroids should be further investigated.
45 Land at Domgay Lane, Four Crosses, Powys: Post-Excavation Assessment and Updated Project Design © Cotswold Archaeology
It is recommended that further detailed analyses be undertaken of a suite of the copper processing residues to provide additional information on their origin. Analysis (chemical and microstructural) should embrace slags, metal droplets, possible ore and slag spheroids, as well as further microscopy and ED-XRF on the ceramics to continue the investigations started by David Dungworth.
The archaeometallurgical investigations should be accompanied by re-appraisal of the field records of the contexts that produced the residues, with a view to re-evaluating whether any or all of those features might be furnaces. Ideally, independent dating of these features would also be highly desirable.
RECOMMENDED ANALYSIS: BREAKDOWN OF TASKS
1. Ore SEM investigation of possible ore fragment in polished block ICP-MS and ICP-OES investigation of possible ore fragment
2. Slag spheroids SEM investigation of slag spheroids in polished block ICP-MS investigation of slag spheroids (if sufficient material available)
3. Metal droplets and other furnace floor materials SEM investigation of metal droplets in polished block SEM investigation of drossy material in polished block
4. Copper slags SEM investigation of copper slag in polished block ICP-MS investigation of copper slag XRF investigation of copper slag
5. Archaeometallurgical ceramics ED-XRF investigation of all archaeometallurgical ceramics not already seen by David Dungworth SEM investigation of furnace lining materials in polished block (2 types) XRF investigation of furnace lining materials ICP-MS analysis of furnace lining materials XRF investigation of tuyère material ICP-MS analysis of tuyère materials
Recommended analysis: estimated costs XRF for major elements (3 slags, 3 ceramics) = 6 ICP-MS for trace elements (3 slags, 3 ceramics, 1 ore, 1 slag spheroids) = 8 Preparation of SEM polished blocks (3 slags, 3 ceramics, 1 ore, = 12 1 slag spheroids, 2 metal droplets, 2 hearth bottom residues) SEM machine time 2 days ED-XRF 1 hour machine time Investigation and reporting 3 days
46 Land at Domgay Lane, Four Crosses, Powys: Post-Excavation Assessment and Updated Project Design © Cotswold Archaeology
APPENDIX 8: HUMAN BONE BY TERESA GILMORE
A total of six inhumations (Sk 1-7) and four cremation burials (Cremations 1-4) were assessed for the potential of further analysis and to establish a Minimum Number of Individuals, ages, sexes and any obvious pathological changes. The observations noted below are provisional and subject to change when further, more detailed, skeletal analysis is undertaken. Five of the inhumation burials had grave cuts and skeleton 1/2 came from the same grave heavily truncated by Ring Ditch 6. Three of the cremations were discrete burials and the fourth consists of pyre debris.
Method Sex was determined on adult remains only using standard criteria (Brothwell 1981; Buisktra and Ubelaker 1994; Schwartz 1995). Adult age was assessed dependent on skeletal parts present. In the sample submitted only dental attrition (Brothwell 1981) could be used. Pathological features were determined by macroscopic inspection, using criteria in Manchester and Roberts (1995) and Schwartz (1995). For the cremations, the entire cremation was weighed and then separated into different sieve fractions, 10mm, 5mm and 2mm. Each section was weighed and the number of identifiable fragments were noted and weighed. The colour and degree of fragmentation was noted as well.
The Inhumations Preservation of human skeletal remains depends on a variety of taphonomic factors including size, shape, robusticity of the bones. The conditions after burial can affect the preservation as well including disturbance of the burial condition, soil characteristics and treatment after excavation. A five-grade system was used to assess preservation: excellent, good, moderate, poor, very poor. Excellent preservation indicates little erosion or fragmentation to the bone surface and integral structure, with all features being easily determined. Very poor preservation indicates complete or practically complete erosion of the bone surface and severe fragmentation. Completeness is determined on a percentage basis depending on the amount of the skeleton present for analysis. Table 2 has a summary of the inhumated human bone recovered. All six inhumations were very poorly preserved and were very incomplete. The poor preservation was mainly due to the acid nature of the soil. Sk 1/2 and Sk 5 were the best preserved and were both had between five and 10% of the skeleton present. The remaining individuals (Sk 3, Sk 4, Sk 6 and Sk 7) were the least well preserved and had around five percent of their skeleton present.
Comments All six individuals were adult in age, but due to the poor preservation, an exact age could not be determined. Sex could potentially be determined on two individuals, Sk 5 and Sk 7. However due to the fragmentation and low level of completeness, the sex assigned is only that of ?female for Sk 5 and ?male for Sk 7. Sk 1 is same as Sk 2 and is cut by Ring Ditch 6 is assigned to Period 2 (Bronze Age) whereas the remaining individuals are undated with the exception of Sk 4 which is of Roman date.
Pathology was noted on two individuals, Sk 1/2 and Sk 5. Sk 1/2 demonstrates severe osteoarthritis in the surviving cervical vertebra, especially cervical vertebrae C2 and C3 which have fused together at the vertebral body and the inferior articular facets. The remaining cervical vertebrae all demonstrate osteophytic lipping of the anterior rims of the vertebral body and porosity of the inferior and superior vertebral body surfaces. Sk 5 has a potential compression fracture in one of the surviving lumbar vertebrae. The surface of the vertebral body is
47 Land at Domgay Lane, Four Crosses, Powys: Post-Excavation Assessment and Updated Project Design © Cotswold Archaeology
aligned towards the front of the body. None of the other vertebral bodies present demonstrate this alignment. Dental pathology was only noted in Sk 7. Calculus was present on the buccal surface of the upper maxillary teeth.
Cremated Bone Overall the preservation of the cremation burials was significantly better than the preservation of the inhumated burials. Table 3 detailed the cremated bone recovered by context. C2 (1764) was the best preserved with over two kilograms of cremated bone surviving from the six contexts. C1 and C3 were the least well-preserved with less than 10 grams of cremated bone surviving. The potential pyre debris, C4 consisted of over 100g of material present.
An adult age could be assigned to all four cremation burials but due to the degree of fragmentation and lack of suitable parts present (i.e. pubic symphysis, auricular surface and dental attrition) an exact age could not be determined. C2 was the only cremation burial sufficiently well-preserved to attempt to diagnose sex and as such, is ?Male due to the nuchal crest and chin exhibiting male traits.
Comments on the Cremation Burials Green staining was present on some of the fragments in C2 suggesting that a copper-alloy object had been present either on the pyre or in the burial afterwards. Cremations 1 and 2 are undated, C3 is Late Iron Age/Early Roman and C4 is Roman or later. The weight of C2 (1764) is within the range expected for a typical adult (McKinley 1993) and would suggest burial in situ or close to the pyre debris. The smaller bones such as phalanges, tooth roots and carpals were present amongst the cremated bone which would suggest that if the burial was not in situ, then the pyre site was raked with the all of the bone being deposited in the same location. The weight of burials C1 and C3 are too small to be considered as a complete cremation burial, the fragments are identifiable as human bone but not identifiable to element.
Further Recommendations Due to the poor preservation, none of the inhumated material is suitable for radiocarbon dating. The cremation Burial C2 is the only one preserved well enough to be suitable to send for radiocarbon dating, which could be advisable due to the lack of secure dating obtained from the site. The poor preservation and completeness of the inhumations means that there is relatively little that can be gained from further analysis apart from the two individuals displaying some pathological changes, Sk 1/2 and Sk 5. Further analysis will allow the pathological changes to be fully recorded and differential diagnoses to be produced. The burials can then be placed within the local, regional and national trends. The cremation burials, especially C2 due to the large amount present, are suitable for further more detailed analysis. This will involve looking at pyre temperature (how efficient the cremation process was), the presence of pyre goods and debris, what methods they used for deposition of the cremation and how the undated cremations may produce suitable material for radiocarbon dating and therefore refine the chronological sequence. Cremated bone from C2 has the best potential for dating.
Time recommended for further analysis Further more detailed analysis on Sk 1/2, Sk 5 and the cremations is recommended. Analysis 2 days (FS) Reporting 1 day (FS)
48
Table 2: Summary of the inhumation material Radiocarbon dating Elements present present Elements analysis? Further Skeleton Number Number Skeleton Completeness Completeness Context Type Preservation Preservation Comments potential potential Context Feature Feature Age Sex
Period 2 Early Bronze Age 1 2106 Ditch fill of cut Same Very poor <5% Adult ? R. clavicle midsection Yes 2105 skeleton 1 molar cut by 1 premolar 2 2326 Skeleton in 8050 Poor 5-10% Adult ? L&R temporal All cervical vertebrae Yes grave 2333 (Ring L&R mandible demonstrate lipping of Ditch 6) Occipital anterior rim & porosity of L Maxilla vertebral body. L. inner ear bones C2 & C3 are fused R. malleus & incus together 7 cervical vertebra (R side only) at body and posterior 1 thoracic vertebra facets. 3 molars Caries present in 1 2 incisors molar 1 premola root. 2 canines 6 roots 1327 Fill of grave No human bone present Amber bead recovered cut from processed sample Period 3 Mid-Late Iron Age 2596 Fiil of grave Ring ditch No human bone present Middle Iron Age/Early cut 7 Roman finger ring recovered Period 4 Roman 4 2567 Skeleton in Grave Very poor <5% Adult ? L. temporal bone No grave 2551 group L. petrous portion
Table 2: continued Radiocarbon dating Elements present present Elements analysis? Further Skeleton Number Number Skeleton Completeness Completeness Context Type Preservation Preservation Comments potential potential Context Feature Feature Age Sex undated 3 2561 Skeleton in Interior of Very poor <5% Adult ? L. prox tibia grave 2558 8050 L. prox femur (Ring R. prox tibia Ditch 6) R. greater trochanter. R. femoral midshaft 3 2564 Grave fill of Very poor <5% Adult ? Long bone fragments Bones belong to Sk 3 No 2558 2 molar crowns 5 2622 Skeleton in Exterior to Poor 5-10% Adult ?F L. sciatic notch ? compression fracture in Yes grave 2613 first phase 3 lumbar vertebrae lumbar vertebrae of barrow L. parietal Occipital L temporal L mandibular condyle Skull frags 6 2641 Skeleton in Interior to Very poor 5-10% Adult ? Acromion process No grave 2639 8056 Ring 2 thoracic vertebrae Ditch 10 Petrous portion 21 tooth roots 7 2649 Grave fill of Adjacent Very poor <5% Adult ? Long bone bones Bones belong to Sk 7 (2648) No 2648 to Iron Maxilla Age pit Cervical vertebral frag alignment 4.5 molar crowns 2.5 premolars 2 molar roots 2664 Skeleton in Very poor 5-10% Adult ?M R. distal femur Calculus buccal side on No grave 2648 R. prox. tibia maxillary teeth R. mandible R. maxilla. R side. atlas R. temporal bone R. Sphenoid 1 molar 1 premolar
Table 3: Summary of the Cremated Human Bone Weight ofidentifiable Cremation Number Number Cremation Elements present present Elements Further Anaylsis Anaylsis Further Weight of 10mm Sample Number Total Weight (g) Weight of 5 mm Weight of 2mm sieve fraction sieve fraction sieve fraction sieve fraction Context type fragments fragments Context Colour Age Sex
Late Iron Age/Early Roman C3 1819 78 Fill of White 1 - - 1 - Not identifiable ? ? N grave 1817 C3 1820 79 Fill of White 2 - 1 1 - Not identifiable ? ? N grave 1817 Roman or later C4 2010 103 Pyre White / 28 - 10 16 2 Alveolar bone Adult ? Y deposit in light grey Tooth root ditch 2028 C4 2011 104 Pyre White / 96 10 52 32 1 Cervical vertebra Adult ? Y deposit in light grey 2028 Undated C1 1733 56 Fill of Grey 4 - - 4 - Not identifiable ? ? N grave 1732 C1 1734 57 Fill of Grey 1 - - 1 - Not identifiable ? ? N grave 1732
Table 3: continued
identifiable fragments identifiable Cremation Number Number Cremation Elements present present Elements Further Anaylsis Anaylsis Further Weight of 10mmWeight of Sample Number Total Weight (g) Weight of 5 mm Weight of 2mm sieve fraction sievefraction sievefraction sieve fraction sieve fraction Context type Weight of Context Colour Age Sex
C2 1764 Fill of White / 818 652 72 16 180 Humerus, fibula, carpals, tarsals, Adult ? Y grave 1763 light grey ribs, vertebra, ulna, radius, phalanges, tooth roots, acetabulum Femur patellas C2 1764 60 Fills of White / 142 - 132 10 6 Radial head, hamate, tooth roots Adult ? Y grave 1763 cream / phalanges light grey
C2 1775 Fill of White / 378 216 90 12 16 Radius, vertebrae, carpal, Adult ? Y grave 1763 cream metacarpals, phalanges, molar root C2 1776 70 Fill of White / 88 - 68 20 4 Tooth roots, phalanges, carpals. Adult ? Y grave 1763 grey
C2 1777 Fill of White / 540 452 38 6 124 Clavicle, fibula, pelvis, glenoid Adult ?M Y grave 1763 cream cavity, humerus, mandible, occipital, femora, parietals, temporals, phalanges, zygomatic bone, tarsels C2 1777 71 Fill of White / 92 - 82 10 6 Phalanges, tooth roots, vertebra Adult ? Y grave 1763 light grey
4001 topsoil White / 2 - 2 - 2 Cranial fragment Adult ? N light grey
Land at Domgay Lane, Four Crosses, Powys: Post-Excavation Assessment and Updated Project Design © Cotswold Archaeology
APPENDIX 9: ANIMAL BONE BY S. WARMAN
Introduction The majority was hand collected during excavation. A large number of environmental samples were taken but very few produced animal bone when processed. The hand collected assemblage totalled 246 fragments from 80 bones weighing 1.46kg. Of these only 23 bones could be identified to species. The processed samples yielded a further 163 fragments from 106 bones weighing just 7g. None of these could be identified to species. This is a small and poorly preserved assemblage consistent with the predominantly acidic soil conditions.
Methods The assessment conforms to the guidance on best practice as described by English Heritage (2002). The animal bone was rapidly scanned and recorded at context level using a Microsoft Access database. Information recorded included; number of bones, number of fragments, weight of bones in grams, number of bones identifiable to species, fragmentation and preservation, numbers of mandibles, epiphyses and whole bones, species and body parts identified, age and state (including modifications such as butchery, burning, gnawing etc). This information is presented by period and generic number is presented in the accompanying Tables (4 and 5).
Results Animal bone was recovered from contexts dating to Periods 4 (Roman), 5 (post-medieval), 6 (modern) and from some undated deposits. The only two species positively identified were cattle and horse. From Period 5 a substantial part of what may be the carcass of a modern calf was recovered from fill 1589 from the interior of Ring Ditch 2, which had survived well considering the aggressive conditions. Horse was identified in 1920 the upper fill of pit 1916 (interior to Ring Ditch 5) and may well be a later addition. The animal bone from Period 4 includes a single cattle tooth from 2481 the primary fill of ditch 2480 part of Enclosure 1. Fill 1018 of the as yet undated ditch 1017 contained a single horse tooth. The fact that much of the identifiable material comprises teeth is reflects the poor preservation of the assemblage in general. Much of the sieved material only survives because it has been burnt.
Discussion The poor condition of the assemblage is reflected in the small proportion of identifiable material. The animal bone from the samples has only survived as it has been burnt and in some cases calcined. The acidic soil conditions are no doubt to blame for this. Many specimens show signs of weathering, and there may have been a period of exposure on the ground surface prior to burial which would further reduce the level of preservation. The assemblage provides proof for the presence horse and cattle during Periods 4 and 6 and horse from an undated deposit. The cattle remains from Period 5 appear to be from the burial of a significant part of a carcass from a single modern calf. The fact that the porous juvenile bone has survived is surprising considering the generally bad preservation in the assemblage thus it may be that this specimen is intrusive and dates to a more recent period.
Recommendations This small poorly preserved assemblage does not require any further work. A brief summary of the results presented in this assessment should be included in any future publication. The animal bone is poorly preserved and is not suitable for radiocarbon dating.
50
Table 4: Hand collected animal bone by period and feature
Period context Description of Feature no of no of weight No of mandibles epiphyses species/part state age context keyword frags bones bones id 4 2235 Primary fill of 2234 DITCH 1 1 0.5 0 UNID(F) BN 4 2481 Primary fill of 2480 DITCH 38 5 86 5 B(H) WE A 4 2204 Final fill of 2203 PIT 10 3 4 0 SSZ(LB) BN 5 2017 Primary fill of 2016 DITCH 4 4 0.25 0 UNID(F) 5 1298 Primary fill of 1299 DITCH 36 3 80 0 CSZ(LB) WE MB SA 5 1589 Burial 2 in 1501 GRAVE 78 20 481 5 1 7 B(H,UL,LL) WE MB J CSZ(V,R) 5 1589 Burial 2 in 1501 GRAVE 18 8 200 4 1 B(H,UL) CSZ(R) WE J 5 1502 Primary fill of 1501 GRAVE 21 21 78 5 2 B(H) CSZ(LB) WE MB SA UNID(F) 5 1503 Burial 1 in 1501 GRAVE 10 3 205 0 3 CSZ(UL,LL) WE MB J 5 1920 Final fill of 1916 PIT 10 2 271 2 1 2 E(H) CSZ(LL) WE MB A, SA 5 1918 Secondary fill of PIT 4 1 2 0 UNID(F) BN 1916 6 4001 Topsoil LAYER 1 1 2 0 SSZ(H) BN 6 5001 Topsoil LAYER 1 1 0.5 0 UNID(F) MB 6 4001 Topsoil LAYER 7 6 18 1 1 B(H) CSZ(LB) MB A Undated 1018 Primary fill of 1017 DITCH 7 1 30 1 E(H) WE, MB A
Key to codes used in tables Species; E = Equus caballus (Horse), B = Bos taurus (cow), CSZ = cow-sized, SSZ = sheep-sized, UNID = unidentified, Parts; H = head, HC = horncore, V = vertebra, R = rib, UL = upper limb, LL = lower limb, MP metapodial, P = phalange, FB = flat bone, LB = long bone, F = fragment. Ageing data; epiphyses = simple count, mandibles = simple count State; WE = weathered, BT = butchery marks, BN = burnt, GN = gnawed, RT = root etching, MB = modern break, PA = pathology. Age; F/N = foetal/neonatal, I = infant, J = juvenile, SA = sub-adult, A = adult, O = old adult.
Table 5: Animal bone recovered from processed samples by period and feature
Period context Description of Feature type Sample no of no of weight No of species/part state context no frags bones bones id 4 2085 Same as 2072 PIT 110 1 1 0.01 0 UNID(F) BN 4 2153 Primary fill of 2152 PIT 115 7 7 0.02 0 UNID(F) BN 4 2291 Primary fill of 2290 PIT 136 1 1 0.01 0 UNID(F) BN 4 2220 Primary fill of 2219 PIT 125 8 8 0.02 0 UNID(F) BN 4 2518 Upper fill of 2515 HEARTH 181 4 4 0.05 0 UNID(F) BN 4 2243 Primary fill of 2242 POSTHOLE 127 4 4 0.01 0 UNID(F) BN 4 2123 Final fill of 2121 PIT 114 5 5 0.2 0 UNID(F) BN 4 2111 Primary fill of 2110 POSTHOLE 111 2 2 0.1 0 UNID(F) BN 4 2204 Final fill of 2203 PIT 121 23 8 2 0 CSZ(H) UNID(F) BN 4 2204 Final fill of 2203 PIT 121 9 1 0.25 0 CSZ(H) 5 2403 Primary fill of 2402 GRAVE 159 1 1 0.01 0 UNID(F) BN 5 1502 Primary fill of 1501 GRAVE 31 6 1 0.25 0 CSZ(H) 5 1918 Secondary fill of PIT 99 9 2 1 0 UNID(F) BN 1916 5 2334 Primary fill of 2333 GRAVE 141 27 5 2 0 UNID(F) MB WE U 2192 Primary fill of 2191 GRAVE 119 6 6 0.25 0 UNID(F) BN U 2611 Upper fill of 2610 PIT 194 1 1 0.05 0 UNID(F) BN U 2118 Primary fill of 2117 POSTHOLE 112 16 16 0.25 0 UNID(F) BN U 2336 Primary fill of 2335 POSTHOLE 142 3 3 0.02 0 UNID(F) BN U 2165 Primary fill of 2164 PIT 117 6 6 0.25 0 UNID(F) BN U 2120 Primary fill of 2119 POSTHOLE 113 1 1 0.01 0 UNID(F) U 2337 Same as 2336 POSTHOLE 154 5 5 0.01 0 UNID(F) BN U 2120 Primary fill of 2119 POSTHOLE 113 18 18 0.25 0 UNID(F) BN
Land at Domgay Lane, Four Crosses, Powys: Post-Excavation Assessment and Updated Project Design © Cotswold Archaeology
APPENDIX 10: CHARRED PLANT REMAINS BY ELIZABETH PEARSON
Samples from 138 deposits of Neolithic to Roman date were selected for assessment. Food residues in the form of charred spelt, emmer/spelt and free-threshing wheat grains, and hazelnut shell fragments were found in Bronze Age barrow deposits, and are possibly associated with ritual activities relating to the burials. Occasional fragments of human tooth or bone were also noted in the Bronze Age burial deposits. A low level of charred plant remains in Romano-British contexts was the only evidence of crop processing on the site. Seeds associated with pyre deposits of the same date may indicate that these took place on grassy land, or that grassy material was used as tinder for the pyre.
Aims The aims of the assessment were to determine the state of preservation, type, and quantity of environmental remains recovered, from the samples and information provided. This information was be used to assess the importance of the environmental remains.
Methods Fieldwork and sampling policy The environmental project conforms to relevant sections of the Standard and guidance for archaeological excavation (IFA 1999); and Environmental Archaeology: a guide to the theory and practice of methods, from sampling and recovery to post-excavation (English Heritage 2002). Samples were taken from deposits considered to be of potential for the recovery of environmental remains. A total of 223 bulk samples were taken during excavation and a total of 138 samples were selected for assessment, focusing on Periods 1 to 4 (Table 6).
Processing and analysis The samples were processed by flotation by Cotswold Archaeology, by either a floatation tank set up on-site or by bucket floatation or ‘washover’ at the CA offices. The flots were collected on a 250µm sieve and the residue retained on either a 1mm or 500µm mesh. This allows for the recovery of items such as small animal bones, molluscs and seeds. The residues were fully sorted by Cotswold Archaeology and any ecofactual remains passed on to the author for examination. The flots were scanned by the author using a low power MEIJI stereo light microscope and plant remains identified using modern reference collections maintained by the Service, and a seed identification manual (Beijerinck 1947). Nomenclature for the plant remains follows the New Flora of the British Isles, 2nd edition (Stace 2001).
Results Environmental remains were generally sparsely distributed across the site. These included charred plant remains and bone fragments (probably human) from graves, while many seed and insect remains are likely to be intrusive. The seeds and insect remains are uncharred and are unlikely to have survived in the soil for long as there was no evidence of waterlogging or anoxic (oxygen reduced) conditions on the site. Moreover, a similar range species from uncharred plant remains are found in all phases, suggesting they are not characteristic of any particular period. The plant remains recovered from the samples are detailed in Tables 6-9, where charred plant remains are present.
53 Land at Domgay Lane, Four Crosses, Powys: Post-Excavation Assessment and Updated Project Design © Cotswold Archaeology
Period 1: Neolithic One context (2493), a secondary fill of a Pit 2548, was sampled from this period, but no remains of significance were identified.
Period 2: Bronze Age Environmental remains were sparse, but extensive in samples of this period. A number of ditch fills from the ring ditches also produced charred plant remains (see Table 7 for details). These were; 1355 the primary fill of 1356 part of Ring Ditch 3, 1367 the primary fill of 1368 also part of Ring Ditch 3, 1415 the primary fill of 1414 part of Ring Ditch 2, 1430 upper fill of 1428 part of Ring Ditch 2, 1476 the primary fill of 1475 part of Ring Ditch 4, 1487 second fill of 1485, 1565 second fill or 1559 part of Ring Ditch 1, 1571 primary fill of 1570 part of Ring Ditch 1, 1593 the third fill of 1596 part of Ring Ditch 1, 1599 the primary fill of 1598 part of Ring Ditch 1 and 1625 the primary fill of 1624 part of ditch 1822.
The material identified to species would be suitable for radiocarbon dating if the deposits can be confirmed as secure and undisturbed. Free-threshing wheat (Triticum sp free threshing) was also identified in the primary fill 1936, of Pit 1935, within the interior of Ring Ditch 4, which compared to emmer and spelt wheat is uncommon in deposits of this period. Emmer and spelt appear to have been the main wheat crops in cultivation at this time. Only occasional charred plant remains were recorded from grave samples, which included unidentified charred cereal and grass grains, and possible vetch (cf Vicia sp) seed.
Period 3: Iron Age No significant remains were identified from these deposits.
Period 4: Roman These samples (mostly associated with a field system) were generally much more charcoal-rich although identifiable remains were only occasionally present (Table 10). The only cereal grain identified was free-threshing wheat (Triticum sp free-threshing) in context 2298 the fill of 2297 a posthole, which is uncommon in deposits of this date. A few remains were identified in pyre deposits (2010 and 2011) within cut 2028 part of Enclosure 1. The moderate number of charred seeds of sheep’s sorrel (Rumex subgen acetosella ) may indicate the burning of the pyre on grassy land on sandy, slightly acidic soil, or that grassy material was used as tinder for the pyre. The charred plant material identified here may provide a means of providing an absolute date for Cremation burial 4, which is currently dated stratigraphically as Roman or later. Similar remains of Iron Age date have been identified by the author in pyre deposits at Walton in Warwickshire (Pearson 2005).
Undated Only occasional charred grains of spelt wheat (Triticum spelta), emmer/spelt wheat (Triticum dicoccum/spelta), unidentified grass grains (Poaceae sp indet) and fragments of hazelnut shell (Corylus avellana) were recorded from ditch and pit fills 1621 (1620); 1917 and 1918 (1916) and 1936 (1935) (Tables 7 and 8). These may be suitable for radiocarbon dating by AMS. Free-threshing wheat (Triticum sp free threshing) was also identified in the primary fill 1936, of Pit 1935, within the interior of Ring Ditch 4, This feature itself is undated which may explain its presence as it is uncommon in Bronze Age deposits. Occasional weed seeds were noted in pit deposits, which were presumably harvested and charred with the crop.
54 Land at Domgay Lane, Four Crosses, Powys: Post-Excavation Assessment and Updated Project Design © Cotswold Archaeology
Discussion The low level of charred cereal crop remains from Bronze Age deposits is not unexpected from a barrow cemetery complex. Sparse remains are consistent with other ceremonial sites in the British Isles, with no abundance of waste being expected, as agricultural processing is more likely to have taken place close to settlement areas. However, any scant remains recorded may represent the remains of food gifts associated with burials (Van der Veen 1985), and therefore need to be considered in the full report. Charred plant remains were similarly sparse in a beaker pit associated with the barrow complex, at Four Crosses, excavated during 1981-85 (Milles 1986). At Domgay Lane, emmer, and to some degree spelt wheat, are likely to have been the main wheat crops in cultivation, evidence of which was present in these samples. Hazelnut fragments, however, are frequently found in deposits of this date and were probably commonly collected and part of the staple food resource. Pollen evidence from Domgay Lane indicates a relatively open and grassland dominated landscape surrounding the barrow (Head 2006), with no specific evidence of cereal cultivation. The recent Research Frameworks for the Archaeology of Wales have shown a generally been a low level of plant remains from Bronze Age sites in Wales, but the role that plant remains may have played in burial and ritual practices both in prehistory and later periods in Wales has received comparatively little attention and requires further investigation (Caseldine 2004).
The low level of charred cereal crop waste associated with a Romano-British field system suggests that processing of cereal crops was only undertaken on a small-scale in this area. Little could be interpreted from these remains about the agricultural economy of the site, although evidence associated with pyre deposits may provide some limited information on the environment and fuel used in the pyre.
Significance Plant remains associated with Bronze Age deposits are rare, and are therefore of significance for the full report. However, environmental remains were generally found in low levels in all samples. Further work is also recommended on plant remains from pyre deposits of Romano-British date, and those containing free-threshing wheat.
Recommendations The following recommendations are made with regard to further work on the samples considered as part of this report: • Full quantification of plant remains from all samples from Period 2 ring ditch deposits (see Tables 7); • Full quantification of Period 4 charred plant remains associated with pyre material (contexts 2010 and 2011); Cremation burial C4 within cut 2028 part of ditch 8075 Enclosure 1 and from context 2298 the posthole fill of 2297 the containing free-threshing wheat (See Table 10); • Integration of the results of the assessment in the full report.
A total of 7 samples are recommended for further work. Charred plant material may be required for radiocarbon dating. Charred seeds identified to species from the primary fill of the deeper features are suitable, as is the hazelnut shell fragment. Material from upper fills should be avoided as should any material that hasn’t been charred as it is likely to be intrusive given the soil conditions. In particular a date for Cremation burial C4 would be helpful as it is currently dated as Roman or later.
55 Land at Domgay Lane, Four Crosses, Powys: Post-Excavation Assessment and Updated Project Design © Cotswold Archaeology
Processing implications To process (float and sort) the additional samples recommended. Processing (float and sort ten litre samples) 4 days (FP) Publication report and analysis 2.5 days (FEE)
Table 6: List of samples selected for assessment
Period Context Sample Residue Flot sieve Feature keyword Description of sieve context 1 2493 212 PIT Secondary fill of 2548 2 1582 035 1.0 mm 0.25 mm Ring ditch 1 Primary fill of 1568 2 1482 028 1.0 mm 0.25 mm Ring ditch 4 Primary fill of 1481 2 1656 047 1.0 mm 0.25 mm Ring ditch 1 Primary fill of 1655 2 1561 037 1.0 mm 0.25 mm Ring ditch 1 Primary fill of 1560 2 1593 040 1.0 mm 0.25 mm Ring ditch 1 Third fill of 1590 2 1508 032 1.0 mm 0.25 mm Ring ditch 4 Secondary fill of 1506 2 1780 067 1.0 mm 0.25 mm Ring ditch 5 Primary fill of 1779 2 1550 034 1.0 mm 0.25 mm Ring ditch1 Primary fill of 1551 2 1358 09 1.0 mm 0.25 mm Ring ditch 3 Primary fill of 1357 2 1571 038 1.0 mm 0.25 mm Ring ditch 1 Primary fill of 1570 2 1604 043 1.0 mm 0.25 mm Ring ditch 1 Secondary fill of 1602 2 1554 033 1.0 mm 0.25 mm ring ditch 1 Secondary fill of 1552 2 1625 048 1.0 mm 0.25 mm Ring ditch 1 Primary fill of 1624 2 1369 013 1.0 mm 0.25 mm Ring ditch 3 Primary fill of 1370 2 1352 006 1.0 mm 0.25 mm Ring ditch 3 Primary fill of 1351 2 1487 030 1.0 mm 0.25 mm Ring ditch 2 Secondary fill of 1485 2 1415 022 1.0 mm 0.25 mm Ring ditch 2 Primary fill of 1414 2 1565 036 1.0 mm 0.25 mm Ring ditch 2 Secondary fill of 1559 2 1599 039 1.0 mm 0.25 mm Ring ditch 1 Primary fill of 1598 2 1787 068 1.0 mm 0.25 mm ring ditch 5 Primary fill of 1786 2 1612 041 1.0 mm 0.25 mm Ring Ditch 1 Primary fill of 1613 2 1484 029 1.0 mm 0.25 mm Ring Ditch 4 Secondary fill of 1483 2 1611 042 1.0 mm 0.25 mm Ring ditch 1 Secondary fill of 1613 2 1360 011 1.0 mm 0.25 mm Ring ditch 3 Primary fill of 1359 2 1420 023 1.0 mm 0.25 mm Ring ditch 2 Final fill of 1418 2 1355 008 1.0 mm 0.25 mm Ring ditch 3 Primary fill of 1356 2 1478 027 1.0 mm 0.25 mm Ring ditch 3 Primary fill of 1477 2 1367 012 1.0 mm 0.25 mm Ring ditch 3 Primary fill of 1368 2 2146 144 0.5 mm 0.25 mm Ring ditch 6 Primary fill of 2144 2 1345 007 1.0 mm 0.25 mm Ring ditch 3 Primary fill of 1346 2 1406 018 1.0 mm 0.25 mm Ring ditch 2 Primary fill of 1405 2 1379 015 1.0 mm 0.25 mm Ring ditch 2 Final fill of 1377 2 1430 024 1.0 mm 0.25 mm Ring ditch 2 Final fill of 1428 2 1373 014 1.0 mm 0.25 mm Ring ditch 3 Primary fill of 1372 2 1469 025 1.0 mm 0.25 mm Ring ditch 4 Primary fill of 1468 2 1412 020 1.0 mm 0.25 mm Ring ditch 2 Primary fill of 1413 2 1376 016 1.0 mm 0.25 mm Ring ditch 2 Final fill of 1374 2 1476 026 1.0 mm 0.25 mm Ring ditch 4 Primary fill of 1475 2 1408 019 1.0 mm 0.25 mm Ring ditch 2 Primary fill of 1407 3 1756 094 0.5 mm 0.25 mm PIT alignment 1 Primary fill of 1755 3 1399 017 1.0 mm 0.25 mm PIT alignment 1 Primary fill of 1398 3 1404 021 1.0 mm 0.25 mm PIT alignment 1 Primary fill of 1403 3 1723 081 0.5 mm 0.25 mm PIT alignment 1 Primary fill of 1722
56 Land at Domgay Lane, Four Crosses, Powys: Post-Excavation Assessment and Updated Project Design © Cotswold Archaeology
Table 6: continued
Period Context Sample Residue Flot sieve Feature keyword Description of sieve context 4 2554 184 1.0 mm 1.0 mm GRAVE SK4 Primary fill of 2551 4 2553 180 1.0 mm 1.0 mm GRAVE SK4 Primary fill of 2551 4 2282 135 0.5 mm 0.25 mm pit Primary fill of 2281 4 2254 130 0.5 mm 0.25 mm POSTHOLE Same as 2252 4 2245 128 0.5 mm 0.25 mm POSTHOLE Same as 2243 4 2212 122 0.5 mm 0.25 mm POSTHOLE Secondary fill of 2210 4 2243 127 0.5 mm 0.25 mm POSTHOLE Primary fill of 2242 4 2011 104 0.5 mm 0.25 mm enclosure 1 Pyre deposit in 2028 4 2220 125 0.5 mm 0.25 mm PIT Primary fill of 2219 4 1798 072 1.0 mm 0.25 mm PIT Final fill of 1796 4 2298 138 0.5 mm 0.25 mm POSTHOLE Primary fill of 2297 4 2213 123 0.5 mm 0.25 mm POSTHOLE Final fill of 2210 4 2314 140 0.5 mm 0.25 mm PIT Primary fill of 2313 4 2375 157 0.5 mm 0.25 mm Pit/posthole Same as 2374 4 1804 076 1.0 mm 0.25 mm Pit alignment 1 Primary fill of 1803 4 2229 126 0.5 mm 0.25 mm Boundary ditch 1 Same as 2220 4 2072 109 0.5 mm 0.25 mm PIT Final fill of 2071 4 2039 105 0.5 mm 0.25 mm PIT Primary fill of 2034 4 2291 136 0.5 mm 0.25 mm PIT Primary fill of 2290 4 2258 131 0.5 mm 0.25 mm PIT Primary fill of 2257 4 2253 129 0.5 mm 0.25 mm POSTHOLE Primary fill of 2252 4 2374 156 0.5 mm 0.25 mm ?KILN Primary fill of 2373 4 2010 103 0.5 mm 0.25 mm Cremation 4 within Pyre deposit in 2028 Enclosure 1 4 2345 143 0.5 mm 0.25 mm POSTHOLE Primary fill of 2344 4 2346 155 0.5 mm 0.25 mm POSTHOLE Same as 2345 4 2489 171 0.5 mm 0.25 mm GRAVE Primary fill of 2488 U 2614 203 1.0 mm 1.0 mm Grave outside ring Primary fill of 2613 ditch 14 U 1919 095 0.5 mm 0.25 mm PIT inside ring ditch 5 Third fill of 1916 U 1917 100 0.5 mm 0.25 mm PIT inside ring ditch 5 Primary fill of 1916 U 1975 091 0.5 mm 0.25 mm Pit inside ring ditch 1 Primary fill of 1974 U 1918 099 0.5 mm 0.25 mm Pit within ring ditch 5 Secondary fill of 1916 U 2641 210 1.0 mm 1.0 mm Grave within ring Primary fill of 2639 ditch 10 U 2657 215 1.0 mm 1.0 mm Grave within ring Primary fill of 2639 ditch 10 U 2564 189 1.0 mm 1.0 mm GRAVE inside ring Primary fill of 2558 ditch 6 U 2559 185 1.0 mm 1.0 mm GRAVE inside ring Tertiary fill of 2558 ditch 6 U 2560 186 1.0 mm 1.0 mm GRAVE inside ring Secondary fill of 2558 ditch 6 U 2564 188 1.0 mm 1.0 mm Grave inside ring Primary fill of 2558 ditch 6 U 2564 190 1.0 mm 1.0 mm GRAVE inside ring Primary fill of 2558 ditch 6 U 2564 192 1.0 mm 1.0 mm GRAVE inside ring Primary fill of 2558 ditch 6 U 1734 057 1.0 mm 0.25 mm Cremation1 same as 1733 U 1764 060 0.5 mm 0.25 mm Cremation 2 Primary fill of 1763 U 1777 071 0.5 mm 0.25 mm Cremation 2 Same as 1764 U 2649 218 1.0 mm 1.0 mm GRAVE containing Primary fill of 2648 SK7 U 2649 216 1.0 mm 1.0 mm GRAVE containing Primary fill of 2648 SK7 U 2649 217 1.0 mm 1.0 mm GRAVE containing Primary fill of 2648 SK7
57 Land at Domgay Lane, Four Crosses, Powys: Post-Excavation Assessment and Updated Project Design © Cotswold Archaeology
Table 6: continued
Period Context Sample Residue Flot sieve Feature keyword Description of sieve context ? 1972 092 0.5 mm 0.25 mm ?GRAVE Primary fill of 1973 ? 1502 031 1.0 mm 0.25 mm GRAVE Primary fill of 1501 ? 1621 044 1.0 mm 0.25 mm PIT Primary fill of 1620 ? 1936 088 0.5 mm 0.25 mm PIT Primary fill of 1935 ? 1911 086 0.5 mm 0.25 mm PIT Primary fill of 1910 ? 1932 090 0.5 mm 0.25 mm ?GRAVE Secondary fill of 1931 ? 2005 101 0.5 mm 0.25 mm ?PIT Final fill of 2003 ? 1588 059 0.5 mm 0.25 mm GRAVE Primary fill of 1587 ? 1908 085 0.5 mm 0.25 mm GRAVE Secondary fill of 1906 ? 1727 058 1.0 mm 0.25 mm PIT Primary fill of 1725 ? 1939 087 0.5 mm 0.25 mm PIT Primary fill of 1938 ? 2403 159 0.5 mm 0.25 mm GRAVE Primary fill of 2402 ? 2621 206 1.0 mm 1.0 mm GRAVE Secondary fill of 2618 ? 2624 207 1.0 mm 1.0 mm GRAVE Primary fill of 2618 ? 2644 213 1.0 mm 1.0 mm GRAVE Upper fill of 2642 ? 2643 214 1.0 mm 1.0 mm GRAVE Primary fill of 2642 ? 2263 133 1.0 mm 0.25 mm GRAVE Primary fill of 2262 ? 2153 115 0.5 mm 0.25 mm pit Primary fill of 2152 ? 2476 168 0.5 mm 0.25 mm GRAVE Primary fill of 2475 ? 2437 166 0.5 mm 0.25 mm GRAVE Primary fill of 2436 ? 2434 164 0.5 mm 0.25 mm GRAVE Primary fill of 2433 ? 2218 124 0.5 mm 0.25 mm POSTHOLE Primary fill of 2217 ? 2299 148 0.5 mm 0.25 mm POSTHOLE Same as 2298 ? 2060 108 0.5 mm 0.25 mm PIT Same as 2039 ? 2432 163 1.0 mm 1.0 mm PIT Same as 2430 2 1818 077 1.0 mm 0.25 mm Cremation 3 Primary fill of 1817 ? 2518 181 0.5 mm 0.25 mm HEARTH Upper fill of 2515 ? 2431 162 0.5 mm 0.25 mm PIT Primary fill of 2430 ? 2111 111 0.5 mm 0.25 mm POSTHOLE Primary fill of 2110 ? 2198 120 0.5 mm 0.25 mm ?GRAVE Same as 2195 ? 2691 220 1.0 mm 1.0 mm GRAVE Primary fill of 2692 ? 2371 152 0.5 mm 0.25 mm POSTHOLE Same as 2325 ? 2123 114 0.5 mm 0.25 mm PIT Final fill of 2121 ? 2438 167 1.0 mm 1.0 mm GRAVE Same as 2437 ? 2435 165 1.0 mm 1.0 mm GRAVE Same as 2434 ? 2477 169 1.0 mm 1.0 mm GRAVE Primary fill of 2475 ? 2428 160 0.5 mm 0.25 mm GRAVE Primary fill of 2427 ? 2048 106 0.5 mm 0.25 mm PIT Same as 2039 ? 1769 063 1.0 mm 0.25 mm PIT Same as 1767 ? 2274 146 0.5 mm 0.25 mm POSTHOLE Same as 2273 ? 2292 149 0.5 mm 0.25 mm PIT Same as 2291 ? 2085 110 0.5 mm 0.25 mm PIT Same as 2072 ? 1768 062 1.0 mm 0.25 mm PIT Same as 1767 ? 2496 173 0.5 mm 0.25 mm PIT Secondary fill of 2495 ? 2057 107 0.5 mm 0.25 mm PIT Same as 2039 ? 2510 183 0.5 mm 0.25 mm PIT Primary fill of 2509 ? 1770 064 1.0 mm 0.25 mm PIT Primary fill of 1766 ? 2315 150 0.5 mm 0.25 mm PIT Same as 2314 ? 2520 182 0.5 mm 0.25 mm PIT Primary fill of 2519 ? 2283 147 0.5 mm 0.25 mm POSTHOLE Same as 2282
58
Table 7: Plant remains from Period 2 Ring ditches
RD 3 RD 3 RD 2 RD 2 RD 4 RD 4 RD 1 RD 1 RD 1 RD 1 RD 1 8014 8014 8016 8016 8017 8017 8022 8022 8022 8022 8022 1356 1368 1414 1428 1475 1485 1559 1570 1596 1598 1624 Latin name Family Common name Habitat 1355 1367 1415 1430 1476 1487 1565 1571 1593 1599 1625 Charred plant remains Triticum dicoccum/spelta Poaceae emmer/spelt wheat F + + grain Cereal sp indet grain Poaceae cereal F + + Poaceae sp indet grain Poaceae grass AF + + Corylus avellana shell Betulaceae hazelnut C + + fragment Melilotus/Medicago sp Fabaceae melilot/medick ABD + unidentified seed unidentified + + + + unidentified catkin? Unidentified + +
Uncharred plant remains Fumaria sp Fumariaceae fumitory ABC + + + Atriplex sp Chenopodiaceae orache AB + + + + Chenopodium/Atriplex sp Chenopodiaceae goosefoot/orache AB + Stellaria media Caryophyllaceae common chickweed AB + Silene sp Carypohyllaceae campion AB ++ Rubus series Glandulosi Rosaceae blackberry/bramble CD + unidentified root fragments unidentified ++
Key to Tables 6-9 Habitat Quantity A= cultivated ground + = 1-10 B= disturbed ground ++ = 11-50 C= woodlands, hedgerows, scrub etc +++ = 51-100 D = grasslands, meadows and heathland ++++ = 101+ E = aquatic/wet habitats F = cultivar
Table 8: Plant remains from pits
1620 1916 1916 1935 Latin name Family Common name Habitat 1621 1917 1918 1936 ? U U ? Charred plant remains Triticum spelta grain Poaceae spelt wheat F + Triticum sp (free-threshing) grain Poaceae free-threshing wheat F + Cereal sp indet grain (fragments) Poaceae cereal F + Poaceae sp indet grain Poaceae grass AF + Corylus avellana shell fragment Betulaceae hazelnut C + + Polygonum/Fallopia sp Polygonaceae knotweed/bindweed ABE + Fabaceae sp indet Fabaceae legume ABCDE + Melilotus/Medicago sp Fabaceae melilot/medick ABD + Asperula/Galium sp Rubiaceae woodruff/cleavers ABC +
Unchared plant remains Atriplex sp Chenopodiaceae orache AB +
Table 9: Plant remains from grave samples
Latin name Family Common name Habitat 2263 2643 2657 ? P4 U Charred plant remains Cereal sp indet grain Poaceae cereal F + cf Cereal sp indet grain Poaceae cereal F + + Poaceae sp indet grain Poaceae grass AF + cf Vicia sp Fabaceae vetch AB +
Table 10: Plant remains from Period 4 samples Contexts in bold are those for which further work is recommended
Latin name Family Common name Habitat 1798 2010 2011 2039 2298 2345 2496 2517 ? Charred plant remains Triticum dicoccum/spelta grain Poaceae emmer/spelt wheat F + Triticum sp (free-threshing) grain Poaceae free-threshing wheat F + cf Triticum sp (free-threshing) grain Poaceae free-threshing wheat F + Poaceae sp indet grain Poaceae grass AF + Poaceae sp indet grain (small) Poaceae grass AF + Rumex subgen acetosella Polygonaceae sheep's sorrel ABD ++ + Rumex sp Polygonaceae dock ABCD + Vicia sativa Fabaceae common vetch AB + cf Vicia sp Fabaceae vetch AB + Melilotus/Medicago sp Fabaceae melilot/medick ABD + unidentified twig/bud fragments unidentified +
Uncharred plant remains Fumaria sp Fumariaceae fumitory ABC + Atriplex sp Chenopodiaceae orache AB + + Stellaria media Caryophyllaceae common chickweed AB + unidentified root fragments unidentified ++
Land at Domgay Lane, Four Crosses, Powys: Post-Excavation Assessment and Updated Project Design © Cotswold Archaeology
APPENDIX 11: CHARCOAL BY ROWENA GALE
A programme of environmental sampling, undertaken during the excavation of the site, resulted in the collection of over 200 samples. These were processed by flotation and sieving. Charcoal was recorded in 181 samples (in flots and residues) but frequently in very small quantities. Charcoal-rich samples were relatively rare.
Although the precise morphology of some features was uncertain, it is probable that many ditch features represent weathered out barrows associated with a Bronze Age cemetery. During the Romano-British period the site appears to have been used extensively for metalworking and waste materials from this activity were dumped or packed into (?pre-existing) pits and postholes (Table 11). A number of graves and pyre deposits were also recorded.
This report presents the assessment of an assemblage of charcoal, mostly Period 4 (Romano-British) or undated contexts. As indicated above, many samples included insufficient material to warrant examination. This report refers to 50 samples that each contained >10 fragments of charcoal considered suitable for identification.
Aims and objectives The assessment was undertaken to gauge the potential of the samples to provide significant data in the following areas: • The economic use of fuel woods (species selection) for funerary purposes (pyres), industrial activities (iron-working) and possibly domestic use; • Environmental data and evidence of woodland management; • The potential of charcoal to provide samples suitable for radiocarbon dating (see Table 11).
Methods The assessment is based on the overall observation of the character of each sample and the identification of three randomly selected fragments from each to give some indication of the range of species present. The selected fragments were prepared using standard methods (Gale and Cutler 2000). Anatomical structures were examined using incident light on a Nikon Labophot-2 compound microscope at magnifications up to x400 and matched to prepared reference slides of modern wood. When possible, the maturity of the wood was assessed (i.e. heartwood/sapwood) and stem diameters recorded. It should be noted that during the charring process wood may be reduced in volume by up to 40%.
Results Details of the samples examined and the taxa identified are presented on Table 11.
Period 2 Bronze Age Six samples were examined from the fills of ring ditches and pits. Although these features are provisionally assigned to the Bronze Age, the inclusion of metalworking debris (slag and magnetic material) in contexts 1355 (the primary fill of cut 1356, Ring Ditch 3), 1625 (the primary fill of cut 1624 (Ring Ditch 1), and 1975 (the primary fill of pit 1974, interior to Ring Ditch 1) suggests that backfilling material in these features may have originated from later activities. The fuel debris included pine (Pinus sp.), oak (Quercus sp.), hazel (Corylus avellana), ash (Fraxinus excelsior) and elder (Sambucus sp.) (Table 11).
62 Land at Domgay Lane, Four Crosses, Powys: Post-Excavation Assessment and Updated Project Design © Cotswold Archaeology
Period 4 Roman period Charcoal was examined from 30 samples, mostly from the fills of postholes and pits, although a pyre deposit was recorded in ditch cut 2028 part of Enclosure 1. Charcoal in samples 103 and 104, associated with the pyre deposit, was sparse but included alder (Alnus glutinosa) and elder (Sambucus sp.). These may be suitable for radiocarbon dating which has the potential to refine the date range for Cremation burial C4 (currently Roman or later on stratigraphic grounds) considerably.
Undated samples The remains of possible pyre debris was collected from undated cremations C1 (sample 56) and C2 (samples 60, 71 and 70). The primary fill (contexts 1764 and 1777) of C2 included large amounts of bone. The charcoal identified included oak (Quercus sp.) and hazel (Corylus avellana) (Table 11). The hazel charcoal may provide a means of obtaining an absolute date for these cremations which are currently undated. Samples 69, 74, 82, 84 and 139 originated from the fills of tree-throw pits (1781, 1792, 1854, 1893 and 2302) which mostly included pine (Pinus sp.) (Table 11). It was not clear however whether this material originated from burning the debris from the fallen trees or from some unrelated activity and subsequently dumped in the resulting pits. Pine was also present in sample 73 (the fill of posthole 1790). Samples 113 and 118 (postholes 2119 and 2184) and sample 170 (pit 2484) contained very little viable charcoal; the taxa named included ash (Fraxinus excelsior), alder (Alnus glutinosa), oak (Quercus sp.) and hazel (Corylus avellana) (Table 11).
The inclusion of charcoal and animal bone and the absence of industrial waste in context 1918 (the secondary fill of pit 1916, interior to Ring Ditch 5) suggests that this deposit is more likely to be domestic in origin. The charcoal consisted almost entirely of narrow roundwood (<10mm in diameter) and included the hawthorn/ Sorbus group (Pomoideae) and willow (Salix sp.) or poplar (Populus sp.).Many of the contexts listed on Table 11 included industrial waste (metal/slag, magnetic material and sometimes fired clay). Associated charcoal probably represents industrial fuel but in many instances this was either too sparse or too degraded for identification. The range of species named included oak (Quercus sp.), hazel (Corylus avellana), alder (Alnus glutinosa), blackthorn (Prunus spinosa), the hawthorn/ Sorbus group (Pomoideae), gorse (Ulex sp.) or broom (Cytisus scoparius), ash (Fraxinus excelsior) and willow (Salix sp.) or poplar (Populus sp.) (Table 11). The structure of oak roundwood in sample 135 (the primary fill of posthole 2281) was consistent with that of coppiced stems.
Recommendations For Period 4 the full analysis of samples 56 (C1) and samples 71 and 70 (the primary and secondary fills of C2) is recommended. For Period 4 full analysis of samples 103 and 104 (pyre deposit C4 from 2028 part of ditch 8075; Enclosure 1), sample 140 (pit 2313), sample 147 (posthole 2282) is recommended. No further work should be undertaken on the material from the undated deposits, unless these can be dated by means of radiocarbon dating.. Suitable material for radiocarbon dating has been identified from these samples, in particular it would be useful to date the undated cremations and the deposits which the Tuyère fragments were recovered from. C4 is dated as Roman or later on stratigraphic grounds thus a radiocarbon date for this feature would tighten the chronology.
Summary of samples recommended for further work The identification of 8 samples: Period 4: 103, 104 Undated: samples 56, 71, 70, 127 and 128 (depending on result of C14 date for these) and 140.
63 Land at Domgay Lane, Four Crosses, Powys: Post-Excavation Assessment and Updated Project Design © Cotswold Archaeology
The identification of 8 samples 1.5 days (FEE) The preparation of a full report 2 days (FEE)
64
Table 11: Charcoal identifications by Period and context Key. h/w: heartwood; r/w: roundwood (diameter <20mm); s/w: sapwood (diameter unknown) Quantity of charcoal with dimensions suitable for identification: X = <10 fragments; XX = 10-20 fragments; XXX = 50+ fragments Radiocarbon dating Charcoal suitable for C14 dating is indicated in bold typeface.
Sample Context Description Quantity Taxa identified Potential Recommendations and comments Period 4, Romano-British 103 2010 Pyre deposit in ditch 2028 XX 2x alder (Alnus glutinosa); Mod Full analysis plus possible C14 1 x elder (Sambucus sp.) date to tighten chronology 104 2011 XX 3 x alder (Alnus glutinosa) Mod Full analysis plus possible c14 date to tighten chronology 105 2039 Primary fill of pit 2034. XX 1 x hazel (Corylus avellana); Low In poor condition and very (Metalworking res.). 1 x oak (Quercus sp.) h/w; friable. No further work. 1 x oak (Quercus sp.) r/w 122 2212 Secondary fill of posthole XXX 3 x alder (Alnus glutinosa) Good No further work 2210. (Metalworking res.). 123 2213 Final fill of posthole 2210. XXX 3 x alder (Alnus glutinosa) Good No further work (Metalworking res.). 125 2220 Primary fill of pit 2219. (Poss. XXX 1 x alder (Alnus glutinosa); Good No further work metalworking res.). 1 x oak (Quercus sp.) r/w ø5mm 126 2229 Fill of pit 2220 XXX 2 x alder (Alnus glutinosa); Good No further work 1 x oak (Quercus sp.) h/w 129 2253 Primary fill of posthole 2252. XX 3 x oak (Quercus sp.) h/w Low Plus cindery material. No further (Metalworking res.). work 134 2273 Primary fill of posthole 2272. X 2 x alder (Alnus glutinosa); Low Plus slag-like material. No (Poss. metalworking res.). 1 x oak (Quercus sp.) h/w further work 136 2291 Primary fill of pit 2290. (Poss. XX 3 x alder (Alnus glutinosa) Mod No further work metalworking res.). 138 2298 Primary fill posthole 2297. XX 2 x oak (Quercus sp.) h/w; Low No further work (Poss. metalworking res.). 1 x oak (Quercus sp.) r/w 140 2314 Primary fill of pit 2313. (Poss. XXX 3 oak (Quercus sp.) h/w Good From largewood. Full analysis metalworking res.). 146 2274 Primary fill of posthole 2273 XX 3 x oak (Quercus sp.) h/w Mod From largewood. No further work 156 2374 Primary fill of 2373 X 3 x oak (Quercus sp.) h/w Low Poor condition; plus cindery material. No further work
Table 11: continued
Sample Context Description Quantity Taxa identified Potential Recommendations and comments Undated samples 8 1355 Primary fill of ditch 1356 XX 3 x oak (Quercus sp.) s/w Mod No further work 11 1360 Primary fill of ditch 1539 X 1 x ash (Fraxinus excelsior); Mod No further work 2 x hazel (Corylus avellana) r/w 44 1621 Primary fill of pit 1620 XX 3 x pine (Pinus sp.) Mod No further work 48 1625 Primary fill of ditch 1624 XX 3 x pine (Pinus sp.) Mod 56 1733 Primary fill of grave 1732 XXX 3 x oak (Quercus sp.) s/w Good Full analysis pending c14 date 57 1734 XXX 3 x oak (Quercus sp.) s/w Good Full analysis pending c14 date 60 1764 Primary fill of grave 1763 XXX 2 x oak (Quercus sp.) s/w; Good Full analysis on sample 71 1 x oak (Quercus sp.) h/w pending c14 date 71 1777 XXX 1 x hazel (Corylus avellana); Good Full analysis pending c14 date 1 x oak (Quercus sp.) s/w; 1 x oak (Quercus sp.) h/w 70 1776 Final fill of grave 1763. XXX 1 x oak (Quercus sp.) s/w; Good 1 x oak (Quercus sp.) h/w; 1 x hazel (Corylus avellana) 72 1798 Final fill of pit 1796. X 1 x blackthorn (Prunus spinosa) Low No further work (Metalworking res.). r/w; 1 x hawthorn/ Sorbus group (Pomoideae); 1 x oak (Quercus sp.) s/w 91 1975 Primary fill of pit 1974 XX 1 x pine (Pinus sp.); Mod 1 x oak (Quercus sp.) s/w; 1 x elder (Sambucus sp.) r/w ø 5mm 99 1918 Secondary fill of pit 1916 XXX 1 x hawthorn/ Sorbus group Good Mostly narrow roundwood. (Pomoideae) r/w; Full analysis 2 x willow (Salix sp.) and/ or poplar (Populus sp.) r/w 69 1782 Primary fill of tree throw X 3 x pine (Pinus sp.) Low No further work 1781 73 1791 Primary fill of posthole XX 3 x pine (Pinus sp.) Mod No further work 1790 74 1793 Primary fill of tree throw XX 2 x oak (Quercus sp.) h/w; Mod No further work 1792 1 x pine (Pinus sp.) 82 1855 Primary fill of tree throw X 3 x pine (Pinus sp.) Low No further work 1854 84 1894 Primary fill of tree throw XXX 3 x pine (Pinus sp.) Good No further work 1893
Table 11: continued
Sample Context Description Quantity Taxa identified Potential Recommendations and comments 106 2048 Fill of pit 2039 XX 2 x oak (Quercus sp.) h/w; Mod No further work 1 x hazel (Corylus avellana) 107 2057 Pit fll X 3 x oak (Quercus sp.) r/w Low No further work 108 2060 Pit fill XX 3 x oak (Quercus sp.) s/w Low No further work
113 2120 Primary fill of posthole 2119 X 1 x ash (Fraxinus excelsior) ; Low No further work 2 x alder (Alnus glutinosa) 118 2185 Primary fill of posthole 2184 X 1 x ash (Fraxinus excelsior); Low No further work 2 x hawthorn/ Sorbus group (Pomoideae) 127 2243 Primary fill of posthole 2242 XX 2 x oak (Quercus sp.); Mod Associated with fragment of tuyère block. also contained Tuyere 1 x alder (Alnus glutinosa) Full analysis pending c14 date fragment 128 2245 Fill of posthole 2242 XXX 2 x oak (Quercus sp.) h/w; Low Oak from largewood. No further work also contained Tuyere 1 x blackthorn (Prunus Radiocarbon date would be useful fragment spinosa)
135 2282 Primary fill of posthole XXX 1 x gorse (Ulex sp.) or broom Good Oak possibly coppiced. Full analysis 2281. (Poss. Metalworking (Cytisus scoparius) r/w; res.) 2 x oak (Quercus sp.) r/w
139 2304 Final fill of tree throw 2302 XX 3 x pine (Pinus sp.) Mod No further work 147 2283 Fill of posthole XXX 2 x oak (Quercus sp.) r/w; Good Full analysis 2282 1 x oak (Quercus sp.) s/w 148 2299 Fill of posthole 2298 X 3 x oak (Quercus sp.) h/w Low No further work 149 2292 Fill of pit 2291 X 3 x oak (Quercus sp.) h/w Low No further work 150 2315 Fill of pit 2314 XX 3 x oak (Quercus sp.) h/w Mod No further work 157 2375 Fill of 2374 XX 2 oak (Quercus sp.) h/w; Mod Full analysis 1 x oak (Quercus sp.) 170 2494 Primary fill of pit 2484 XX 1 x oak (Quercus sp.) s/w; Mod No further work 1 x alder (Alnus glutinosa); 1 x hazel (Corylus avellana) 181 2518 Upper fill of hearth 2515 XX 2 x willow (Salix sp.) or poplar Mod Mostly narrow roundwood. (Populus sp.) r/w No further work 182 2520 Primary fill of pit 2519 XXX 1 x gorse (Ulex sp.) or broom Good No further work (Cytisus scoparius) r/w; 1 x ash (Fraxinus excelsior); 1 x hazel (Corylus avellana)
Table 11: continued
Sample Context Description Quantity Taxa identified Potential Recommendations and comments 183 2510 Primary fill of pit 2509 XXX 2 x oak (Quercus sp.) s/w; Good No further work 1 x oak (Quercus sp.) h/w 187 2517 Lining of hearth 2515 XX 3 x willow (Salix sp.) or poplar Low Narrow roundwood. No further work (Populus sp.) r/w
Land at Domgay Lane, Four Crosses, Powys: Post-Excavation Assessment and Updated Project Design © Cotswold Archaeology
APPENDIX 12: POLLEN BY DR. KATIE HEAD
Summary Samples from three contexts from the site of a Bronze Age barrow were selected for analysis. Vegetation at the site appears to represent a stable community of grassland and herbs associated with grassy environments. There was also a notable presence of ribwort plantain suggestive of clearance and activity around the site. Pollen remains, however, were badly preserved and in low concentrations, partly due to the exceptionally dry nature of the sediments. The remains however, did provide a record of vegetation at the site with little variability while the barrow was in use and extended.
The barrow mound was constructed from Pleistocene gravels, which overlaid three possible palaeosols (Wilkinson, Appendix 13). Six sub-samples were taken from the three palaeosols, two from each monolith, through the Bronze Age barrow. The environmental project conforms to standards and guidance for environmental projects (English Heritage 2002) and (AEA 1995).
Aims The aims of the analysis were to determine the state of preservation, type, and quantity of pollen remains recovered, from the samples and information provided. This information will be used to assess the importance of the pollen remains. More specifically the following aim has been identified to determine whether the preservation is sufficient to provide a detailed reconstruction of vegetation at the site.
Methods Fieldwork and sampling policy Samples were taken from deposits considered to be of high potential for the recovery of pollen remains. Three contexts were selected. A total of six sub-samples were taken from the site, two from each context (see Wilkinson, Appendix 13):
• Context 3005, Monolith 177 (sub-samples at depths of 9-10cm and 16-17cm), possible palaeosol. Located at the top of the secondary fill of the first phase barrow ditch and sealed by gravels of the extended second phase barrow; • Context 3010, Monolith 178 (sub-samples at depths of 1-2cm and 9-10cm), possible palaeosol. Located beneath the first barrow mound; • Context 5023, Monolith 179 (sub-samples at depths of 3-4cm and 10-11cm), possible palaeosol. Located above the primary fill in the second phase barrow ditch and sealed by gravels eroding off the second phase barrow.
Processing and analysis Sediment samples of 1cm3 were measured volumetrically. The samples were soaked for 24 hours and then heated in tetra-Sodium Pyrophosphate for 30mins, sieved through a 120μm mesh, washed onto a 10μm mesh, and the residue collected. 10% Hydrochloric acid was then added in order to remove any calcium carbonate within the samples. Due to the siliceous nature of the sediments, the samples were soaked overnight and then digested using Hydrofluoric Acid in a hot water bath for 30mins. Finally the pollen pellet was stained with safranine, washed in alcohol to dehydrate the sample, and preserved in silicon oil.
69 Land at Domgay Lane, Four Crosses, Powys: Post-Excavation Assessment and Updated Project Design © Cotswold Archaeology
Pollen grains were counted, where possible, to a total of 250 land pollen grains (TLP) for assessment purposes, on a GS binocular polarising microscope at 400x magnification, and identification was aided by using the pollen reference manual by Moore et al. (1991) and reference collection maintained by the Service. Nomenclature for pollen follows Stace (1997) and Bennett (1994), and results are listed in taxonomic order (Table 12).
Results With the exception of one sample, counts of 250 TLP could not be achieved due to the degraded nature of the pollen grains and their low concentrations.
Context 3005, Monolith 177 This context comprised an upper and lower sample. The lower sample located at a depth of between 16cm and 17cm contained exceptionally degraded pollen grains in very low concentrations and so counts of 250 TLP were abandoned. Therefore this lower sample could not be used in the interpretation except to say that this sample, as with the others, was dominated by Poaceae undiff. (grasses).
The upper sample (9-10cm depth) however, did contain better preserved pollen and counts of 200 TLP were achieved. Poaceae undiff. (grasses) made up 63% TLP and so was the dominant taxon. Other herbs included comparable species to context 3010, again dominated by Plantago lanceolata (ribwort plantain), although unlike the previous context, Scabiosa sp (scabious) was only recorded in minimal numbers. Trees and shrubs contributed just 19% to the pollen sum, again. As before, Alnus (alder) and Corylus (hazel) were in slightly greater values than other arboreal taxa, although, Tilia (lime) was notable. As this latter taxon is under represented in the pollen record due to it being insect pollinated, lime would have been significant in the regional vegetation. Polypodium (polypody fern) was in significant numbers, and again, like context 3010, there was a single grain of Calluna vulgaris (heather).
Context 3010, Monolith 178 The lower sample within this context (9-10cm) was dominated by Poaceae undiff. (grasses), making up almost 50% TLP (Table 12). Other herbs were dominated by Plantago lanceolata (ribwort plantain) and Scabiosa sp (scabious), while Ranunculus acris-type (meadow buttercup), Caryophyllaceae (e.g. fat hen), Potentilla-type (cinquefoil), and Bidens-type (bur-marigold) were in very low numbers. Trees and shrubs only made up 16% TLP, dominated by Alnus (alder), followed by Corylus (hazel). Other arboreal taxa included occasional examples of Quercus (oak), Salix (willow), and Tilia (lime). There were high numbers of Polypodium (polypody fern), and just one example of Calluna vulgaris (heather).
The upper sample of this context included similar taxa, with Poaceae undiff. (grasses) still dominant but with a slight increase in tree and shrub taxa, seen in the previous sample. With regard to herbs, there was a significant fall in Plantago lanceolata (ribwort plantain), while other herbs were in comparable numbers to before, with the additions of Filipendula (meadowsweet) and Urtica dioica (common nettle). Polypodium (polypody fern) continued to be represented in high numbers although there was a slight fall in this taxon.
Context 5023, Monolith 179 The final context was again represented by an upper and lower sample, with counts of between 200 and 250 TLP achieved. The lower sample (10-11cm depth) was dominated by Poaceae undiff. (grasses) which made up 64% TLP. Other herbs were as in previous contexts, with Plantago lanceolata (ribwort plantain) dominant but not as significant. Unlike context 3005, however, Scabiosa sp (scabious) was now in values comparable to ribwort
70 Land at Domgay Lane, Four Crosses, Powys: Post-Excavation Assessment and Updated Project Design © Cotswold Archaeology
plantain. Trees and shrubs were represented by species and numbers similar to former contexts, with the exception of Tilia (lime), which was absent. Polypodium (polypody fern) was also in high numbers.
The upper sample (3-4cm depth) of context 179 mainly comprised Poaceae undiff. (grasses), while trees and shrubs only contributed 10% TLP. Other herbs were as in previous samples, although there was a large increase in Scabiosa sp (scabious), overshadowing Plantago lanceolata (ribwort plantain). Polypodium (polypody fern) was still notable but had significantly fallen in number.
All three contexts appear to represent a substantially cleared landscape, dominated by grasses and herbs of grassy places such as scabious. Clearance indicators were also present, particularly ribwort plantain. Trees and shrubs most probably reflect a more regional component of the vegetation, with alder and hazel growing in wetter areas and the valley bottoms, while lime in particular would have been colonising the drier lowland slopes. If all contexts represent palaeosols (Wilkinson, this report) then these samples appear to represent periods of stability in between the two phases of the barrow’s construction. The pollen evidence from all contexts is very similar in species composition possibly supporting the hypothesis of landscape stability. The presence of herbaceous vegetation including clearance indicators, and the large number of polypody ferns supports the idea that populations constantly used the site and between construction phases then polypody fern was able to colonise the banks of the barrow.
Discussion A number of Bronze Age barrows and cairns have been analysed for pollen in both the uplands and lowlands of Wales. Although the evidence in general for this period indicates a primarily pastoral economy, evidence has increased for a mixed farming regime and evidence of cereal cultivation is now being recorded (Caseldine 1990). Pollen sequences indicate that the Welsh landscape had been substantially cleared by the Bronze Age in many areas, and this is supported by the pollen record at Four Crosses.
A nearby upland cairn group which has been excavated and analysed for pollen, is that of Nant Helen on Mynydd y Drum, Ystragynlais in Powys (Doring and Chambers 1990). Analysis of the pollen from soils beneath the cairns indicated an already largely deforested landscape by the Bronze Age, and dominated by heathland or hazel scrub when they were constructed (ibid). The pollen profile indicated a dramatic impact on the woodland during the Bronze Age, an open landscape of heaths, sedges, and grasses developing, followed by a short-lived regeneration of hazel scrub.
At the Bronze Age site of Cefn Gwernffrwd, near Rhandirmwyn in mid-Wales (Chambers 1982a), the mixed oak- hazel woodland was substantially cleared during this time, leading to an open landscape of grasses and ribwort plantain, comparable to the Four Crosses site. Other sites also show the most marked clearances occurring during the Bronze Age, including Coed Taf near Merthyr Tydfil, South Wales (Chambers 1983), which showed a slight regeneration of hazel as scrub during the later Bronze Age.
At Four Crosses there is also a notable presence of polypody fern in a number of the samples. At another cairn site, Nant-maden cairn, at Cwm Cadland, South Wales it was found that this taxon marked the development of podzols during the late Bronze Age/early Iron Age (Crampton and Webley 1964).
The Welsh uplands seems therefore, to have been characterised by the clearance of oak woodland and hazel scrub, a result of the early Bronze Age cairn builders, which was replaced by Calluna heath (Chambers 1982b).
71 Land at Domgay Lane, Four Crosses, Powys: Post-Excavation Assessment and Updated Project Design © Cotswold Archaeology
At sites in the lowlands, following the construction of Bronze Age monuments, the landscape was newly dominated by grassland herb communities rather than heathland, which was the case at the Four Crosses site.
Significance Pollen remains in all the samples were in low concentrations, with grains badly degraded, primarily due to the extremely dry nature of the sediments. This prevented counts of 250 TLP being achieved, with the exception of one sample. All contexts of Bronze Age date seemed to highlight periods of stability in that vegetation did not change a great deal while the barrow was in use. Due to poor preservation and the lack of change in the vegetation, the remains are not of great significance.
Recommendations The following recommendation is made for consideration when designing any further archaeological project for this site: • If additional samples are located where at least some waterlogging has occurred or where anaerobic conditions exist, then these should be evaluated for pollen.
No further work is recommended on the samples currently under investigation due to the poor preservation of the pollen remains.
72
Table 12: Pollen results from selected contexts
Latin name Family Common Habitat 3010/1-2cm 3010/9-10cm 3005/16-17cm 3005/9-10cm 5023/10-11cm 5023/3-4cm Name Pteropsida (monolete) Pteropsida ferns BCDE 4 2 2 2 2 9 indet Polypodium Polypodiaceae polypody CD 26 35 5 12 27 17 Ranunculus acris- Ranunculaceae meadow CD 1 1 1 2 type buttercup Urtica dioica Urticaceae common CD 1 1 nettle, stinging nettle Quercus Fagaceae oak C 2 1 1 2 Alnus glutinosa Betulaceae alder C 16 20 15 11 9 Corylus avellana-type Betulaceae hazel C 20 11 4 13 16 14 Chenopodiaceae sp Chenopodiaceae ABCDE 1 Caryophyllaceae sp Caryophyllaceae 5 9 1 7 7 Tilia cordata Tiliaceae small-leaved C 3 1 6 3 lime Salix Salicaceae willow C 1 1 Calluna vulgaris Ericaceae heather CD 1 Rosaceae sp Rosaceae ABCDE 3 2 5 3 Filipendula Rosaceae meadow CDE 2 1 1 1 sweet Potentilla-type Rosaceae cinquefoil BCD 2 3 1 1 Ilex aquifolium Aquifoliaceae holly C 1 Plantago lanceolata Plantaginaceae ribwort D 12 41 25 15 19 plantain Scabiosa sp Dipsacaceae scabious D 8 14 1 2 16 31 Cichorium intybus- Asteraceae chicory, wild BD 5 3 1 1 type succory Bidens Asteraceae bur-marigold E 1 1 Poaceae undiff. Poaceae grass ABCD 123 106 25 125 129 162
Key: A = cultivated ground; B = disturbed ground; C = woodlands, hedgerows, scrub, etc; D = grasslands, meadows, heathland; E = aquatic/wet habitats; F = cultivar.
Land at Domgay Lane, Four Crosses, Powys: Post-Excavation Assessment and Updated Project Design © Cotswold Archaeology
APPENDIX 13: GEOARCHAEOLOGY BY KEITH WILKINSON
Introduction This assessment reports on both a field visit and a subsequent assessment made on a series of monolith samples that were taken during the visit. The purpose of the visit was to comment on and sample stratigraphy exposed in trenches dug through an upstanding barrow in the northern part of the site. The aims of this assessment report are: • To determine whether the fine-grained units sampled in the field are palaeosols; • To assess the potential of the samples to provide useful palaeoenvironmental information relating to the construction and use of the barrow; • Assuming that the answer to the first aim is in the affirmative, to determine the time span over which the palaeosols might have formed; • To make recommendations for work that could usefully be carried out as part of an analytical programme.
Barrow stratigraphy: fieldwork observations Approximately half of the barrow sampled for geoarchaeological assessment lay within the excavation area (Figure 10). The remaining area of the barrow had been stripped of topsoil, but otherwise had not been archaeologically investigated. The barrow seems to have developed in two distinct phases, the later phase increasing the diameter of the barrow and thereby infilling the first ditch.
The barrow mound was composed of Pleistocene gravels cast up from the surrounding ring ditches (14 and 15). These gravels are of pebble to cobble-sized sub-angular clasts in a medium sand matrix. They are also decalcified. Three possible palaeosols are buried by the gravels at the following locations:
• At the top of the secondary fill of the first phase barrow ditch and sealed by gravels of the extended second phase barrow (3005) Monolith 177; • Beneath the first phase barrow mound (3010) Monolith 178; • Above the primary fill in the second phase barrow ditch and sealed by gravels eroding off the second phase barrow (5023).
In the field the properties of all three ‘palaeosols’ appeared similar, although the third (5023) was less distinct than the others. They comprised silts and clays and were of a reddish brown colour. Soil structures were absent (no horizonation, ped structure etc), although clay translocation had taken place and as a result an illuvial fine grained matrix was found surrounding gravel clasts below the first and second (but not the third) ‘palaeosols’.
A series of monolith samples were taken from contexts (3005 sample 177), (3010 sample 178) and (5023 sample 179) in order to carry out laboratory tests to determine whether these contexts are in fact palaeosols and to provide material for detailed analytical work should this attribution prove correct. Each monolith measured 20x7x3 cm and was taken from cleaned sections through all three ‘palaeosols’ by the author and Dr Sylvia Warman (Cotswold Archaeology Environmental Officer) at the locations marked on Figure 6. In addition four bagged samples were taken for control purposes from undisturbed areas of the modern soil A horizon (two samples) and undisturbed gravels (two samples). Only the monoliths have been studied for this assessment.
73 Land at Domgay Lane, Four Crosses, Powys: Post-Excavation Assessment and Updated Project Design © Cotswold Archaeology
Barrow stratigraphy: laboratory assessment
Method In the laboratory the monolith samples were unwrapped and a fresh surface exposed by removal of c. 0.5mm from the upper sediment/soil surface with a scalpel. The stratigraphy revealed was then photographed and descriptions were made using standard geological terminology (Tucker 1982, Jones et al. 1999, Munsell Color 2000) (Figure 9). Samples for palynological assessment were subsequently taken by removing 1cm3 blocks from the locations shown in Figure 9. The six sub-samples taken were submitted to Dr Katie Head (Historic Environment and Archaeology Service, Worcestershire County Council see Appendix 12 Pollen). Sub-samples for sedimentological assessment were later taken by removing 10mm thick by 15mm wide slices from the left hand side of each monolith. These sub-samples were weighed, placed in a drying cabinet at 40oC for 72 hours and then re-weighed to determine field moisture content (a measurement that approximately correlates with grain-size). The dried sub-samples were then homogenised with a mortar and pestle and passed through a 250μm mesh. Sediment/soil coarser than 250μm was discarded, while finer material was used for high and low frequency magnetic susceptibility measurement. Magnetic susceptibility measurements were made using the methodology of Gale and Hoare (1991, 221-6).
The sample archive comprises the three original sample monoliths – which exist in the state shown in Figure 9 except with 15mm of sediment missing from the left hand side of the sample tins - 50 pots containing <250μm sub-samples from the monoliths, and 4 bulk samples from the modern soil and the Pleistocene gravel. The paper archive comprises pro-forma description sheets for the three monoliths, field notes made by the present author and digital photographs made while undertaking the sampling and of the sample in a laboratory state (Figure 9). Both paper and sample archives are currently stored in the Department of Archaeology, University of Winchester.
Sample description Detailed laboratory description demonstrates that all three silt/clay layers buried by the barrow have similar properties. Each comprises two sub-units, an upper consisting of a 10 YR 5/6 yellowish brown silt/clay with occasional sub-rounded pebbles, and a lower of 10 YR 5/4 yellowish brown silt/clay with moderate quantities of sand and a higher quantity of gravel than the overlying sub-unit (Figure 9). It is also notable that the lower unit has a slightly better developed colloidal structure than the upper. These properties suggest that pedogenesis has occurred in the fine-grained layers. The upper sub-units are probably A horizons, i.e. former near surface layers. Here soil forming processes would have been most active and consequently the unit properties most dissimilar to those of the parent material (Pleistocene gravels in this case). The lower sub-units would be B horizons. These have properties which are more akin to that of the parent material, i.e. in this case a higher gravel content. The slightly redder colour of the putative B horizons may reflect clay translocation down the profile, a process that is associated with most soil-forming regimes. However, although it would appear probable that the fine-grained layers are palaeosols, it is probably the case that none is particularly mature. Clay translocation is a time transgressive process, leading to distinct illuvial (Bt) horizons, often associated with a developed ped structure. In the case of the Four Crosses examples there is neither a visible illuvial horizon (clay concentrations were noted around gravel clasts) nor a ped structure.
Magnetic susceptibility Enough sub-samples were examined from reworked Pleistocene gravels to establish the magnetic properties of the parent material of the putative palaeosols. Low frequency magnetic susceptibility (χlf) in these deposits varies between 25 and 50 –8 m3 kg-1, while the percentage difference between low and high frequency magnetic
74 Land at Domgay Lane, Four Crosses, Powys: Post-Excavation Assessment and Updated Project Design © Cotswold Archaeology
susceptibility (χfd) is between 3 and 10%. The magnetic properties of context 3010 (turf layer above subsoil 3016), which may be the land surface buried by the original barrow structure, has magnetic properties that vary little from that of overlying re-worked parent material (see Figure 10, sample 178), albeit that χlf is slightly enhanced in the B horizon, while χfd is enhanced in the possible A horizon. The latter is characteristic of soil formation, but in a mature soil higher χlf measurements would be expected in the B horizon than the A as a result of translocation of magnetic clays from the overlying A horizon. In other words if context 3010 is a palaeosol, the magnetic evidence would suggest that it is not of any great maturity.
The magnetic properties of context 3005 (turf layer sealing deposits in ditch 3023) are in contrast to those discussed for 3010. Low frequency magnetic readings exceed 150 –8 m3 kg-1 in the lower parts of the possible A horizon and rapidly tail off to c 30 –8 m3 kg-1 in the ‘B horizon’. Percentage frequency difference also averages 8% in the ‘A horizon’, climbs to 10% at the top of the ‘B horizon’, and drops back to around 5-7% lower in the ‘B horizon’. Although pedogenesis leads to enhanced magnetic susceptibility in soils, it is usually in B horizon where readings are at their highest (Maher 1986). This factor combined with the fact that magnetic susceptibility in the putative ‘A horizon’ is five times the level of that in the B horizon, suggests that a cause other than soil formation has to be sought for the enhanced χlf readings. Given the magnitude of the readings the most likely mechanism is heat (Le Borgne 1960). It is therefore likely that some form of burning took place on the surface of context 3005 before it was sealed by the expanded barrow. Gradually diminishing χlf measurements from the peak at 2.5cm depth suggest that the burning occurred at that level, impacted on underlying strata, while mixing (i.e. by earthworms or human disturbance) also transported heated particles upwards. Unfortunately the magnetic data alone cannot determine what form this heating took.
Low frequency magnetic susceptibility of the third possible palaeosol (5023) averaged c 50 –8 m3 kg-1 in the ‘A horizon’ and somewhat less than this in the ‘B horizon’. The magnetic properties are almost identical to those of the parent material (5026) the natural, suggesting that if context 5023 (turf horizon) is a buried soil, it is immature. Indeed, as was also the case in context 3005, χlf measurements in the ‘B horizon’ are lower than those in the ‘A horizon’ suggesting that little clay translocation has taken place. However, there is another similarity to context 3005 in that at the very top of the ‘A horizon’ there is χlf enhancement of c. 90 –8 m3 kg-1, which may again be suggestive of a burning event at the top of this buried fine grained unit.
The magnetic susceptibility results do not provide proof that the buried red-brown fine-grained layers are palaeosols. Indeed none show evidence of enhanced χlf and χfd readings in the B horizon, both of which are characteristic of mature soils. Therefore on the basis of the magnetic evidence the layers are either immature buried soils or they are sediments in which no pedogenesis has occurred.
Assessment Field and laboratory description suggest that contexts 3005, 3010 and 5023 are palaeosols that were buried by various stages of barrow construction. Characteristic soil forming features such as the development of horizons and clay translocation were noted, although features of mature soils, such as distinctive ped structures and a well developed illuvial (Bt) horizon were not present. The evidence from magnetic susceptibility measurement is more equivocal. Only in context 3010 are low frequency magnetic readings higher in the supposed B horizon than the A, while both low frequency magnetic susceptibility and percentage frequency difference are little different to that of the supposed parent material. These data suggest very immature soils or even the absence of soil forming processes. The balance of probabilities therefore suggests that the fine-grained units are immature palaeosols,
75 Land at Domgay Lane, Four Crosses, Powys: Post-Excavation Assessment and Updated Project Design © Cotswold Archaeology
which in the case of contexts 3005 and 5023 have undergone only short-term pedogenesis (decadal time scales or less). The low frequency magnetic susceptibility evidence from 3010 suggests a slightly greater maturity, but certainly not one that could be measured in centuries. The archaeological implications of these results are important:
• Although construction of the original barrow took place on a stable land surface, this stability was a relatively recent phenomenon. It is highly likely that a previous soil of greater maturity had been removed decades or hundreds of years before barrow construction. Indeed the pollen evidence (see Appendix 11) suggests that arable or grassland conditions predominated in the local environment, suggesting that early Holocene forests had been removed long before barrow construction; • Phases of barrow remodelling took place within decades of the original barrow construction (assuming that infilling of the original barrow ditch occurred rapidly). Contexts 3005 and 5023 would display the properties of more mature soils were there to have been a longer delay.
Low frequency magnetic measurements of contexts 3005 and 5023 both suggest that burning took place in the A horizons of these palaeosols. These burning events are likely to be a result of cultural processes and it is therefore important to understand what form the burning took, e.g. an effort to remove vegetation prior to the enlargement of the barrow.
Recommendations In order to provide further information of the environment in which the barrow was originally constructed, and in particular whether the barrow was constructed on cultivated land, pasture or scrub, consideration should be given to the micromorphological examination of sample 178. It is also recommended that sample 177 be examined using micromorphological techniques in order to further investigate the burning event that immediately preceded the first extension of the barrow.
No further work is recommended on the existing magnetic susceptibility sub-samples although it is suggested that provision be made to enable the present geoarchaeological assessment data to be worked into a publication text. This latter should be integrated with the report produced by the micromorphologist.
Summary of geoarchaeological assessment for publication 2 days (FEE) Micromorphology report 2 days (FEE)
76 Land at Domgay Lane, Four Crosses, Powys: Post-Excavation Assessment and Updated Project Design © Cotswold Archaeology
Figure 8: Sketch plan of the barrow and sample locations
77 Land at Domgay Lane, Four Crosses, Powys: Post-Excavation Assessment and Updated Project Design © Cotswold Archaeology
Figure 9: The monolith samples
78 Land at Domgay Lane, Four Crosses, Powys: Post-Excavation Assessment and Updated Project Design © Cotswold Archaeology
Figure 10: Magnetic properties of the sampled fine-grained layers
79 Reproduced from the 2000 Ordnance Survey Landranger map with the permission of Ordnance Survey on behalf of The Controller of Her Majesty's Stationery Office N c Crown copyright Cotswold Archaeological Trust AL50196A
0 5km
COTSWOLD ARCHAEOLOGY
PROJECT TITLE Domgay Lane, Four Crosses, Powys
POWYS FIGURE TITLE Site location plan
SCALE PROJECT NO. FIGURE NO. 1:50,000@A4 9018 1 Reproduced from the 2006 Ordnance Survey Landplan map with the permission 327000 3 2 of Ordnance Survey on behalf of The Controller of Her Majesty's Stationery Office 7 N c Crown copyright Cotswold Archaeological Trust AL50196A 0 0
0
328000 3
2
8
0
0 0
319000319000
318000318000
0 1km
COTSWOLD ARCHAEOLOGY
site Cropmarks PROJECT TITLE (after CPAT 2004, fig. 1) Domgay Lane, Four Crosses, Offa’s Dyke (SAM ref: Mg033) linear features Powys other excavated sites FIGURE TITLE pit alignments The site, with nearby cropmarks ring ditches and earthworks SCALE PROJECT NO. FIGURE NO. 1:10,000@A4 9018 2 N
SITE 2
SITE 3
0 200m
COTSWOLD ARCHAEOLOGY archaeological features PROJECT TITLE Domgay Lane, Four Crosses, evaluation trenches Powys 1984 excavations FIGURE TITLE Plan of excavated features Scheduled Ancient Monument (Offa’s Dyke) SCALE PROJECT NO. FIGURE NO. 1:2000@A4 9018 3 N
Ring Ditch 15 Ring Ditch 14
Ring Ditch 13 Ring Ditch 12 SITE 2
Ring Ditch 9
Ring Ditch Ring Ditch 8 10
Neolithic Ring Ditch pit 2548 11
Ring Ditch 7 Fig. 6 Ring Ditch 6
SITE 3 Ring Ditch Ring Ditch 5 3
Ring Ditch 4
grave Ring Ditch 1327 2
Ring Ditch 1
0 100m
COTSWOLD ARCHAEOLOGY
PROJECT TITLE Neolithic Domgay Lane, Four Crosses, Powys Bronze Age FIGURE TITLE 1984 excavations Period 1 Neolithic and Period 2 Bronze Age features SCALE PROJECT NO. FIGURE NO. 1:1000@A4 9018 4 N
C4
Enclosure Curvilinear Ditch 1 1 Boundary Field Curvilinear Ditch 2 Ditch System 1 1 grave Ring Ditch 2436 Curvilinear Ditch 3 13 Enclosure Ring Ditch 2 12 SITE 2 posthole 2488
Pit sk4 Group grave 1 2551
Field System 1
Ring Ditch 7 Iron Age grave Fig. 6 2596
Field System SITE 3 1
Iron Age C3
0 100m
COTSWOLD ARCHAEOLOGY
PROJECT TITLE Iron Age Domgay Lane, Four Crosses, Powys Roman FIGURE TITLE 1984 excavations Period 3 Iron Age and Period 4 Roman features SCALE PROJECT NO. FIGURE NO. 1:1000@A4 9018 5 N
Field System 2 2373
Boundary Ditch 1
2252 2257 2219 2242
Ditch pit 2548 8080
2331
2272 2324
2335 2210 2142
2281 2139 2344
2297
2290 2313
0 10m
COTSWOLD ARCHAEOLOGY feature containing metalworking debris PROJECT TITLE Neolithic Domgay Lane, Four Crosses, Powys Roman FIGURE TITLE post-medieval Period 4: Detail plan of postholes containing metalworking debris probable geological feature SCALE PROJECT NO. FIGURE NO. 1:100@A4 9018 6 N
posthole posthole 2034 2071
sk5 grave 2613
Pit Alignment 3
Ring Ditch 13 Pit Alignment 1
Ring Ditch 12 SITE 2 Field System 2
pit sk6 grave 2509 2639
Field System 2 (sk1/2) pit 2106 sk3 grave 2258
grave grave pit 2630 2596 2262 Fig. 6 sk7 grave 2648 pit 1722 C1 C2
Field System SITE 3 2
pit 1620 pit Field System 1966 Trackway 1 2
pit 1935
pit 1974
pit 1403 Pit Alignment 2
Pit Alignment 1
0 100m
COTSWOLD ARCHAEOLOGY
PROJECT TITLE post-medieval Domgay Lane, Four Crosses, Powys Undated FIGURE TITLE 1984 excavations Period 5 Post-medieval and Period 6 undated features SCALE PROJECT NO. FIGURE NO. 1:1000@A4 9018 7