South Wales Gas Pipeline Project Site 513 Land North-West of Steynton Milford Haven Pembrokeshire
Archaeological Excavation
for Rhead Group on behalf of National Grid
CA Project: 9150 CA Report: 13261 Event DAT108846 March 2014
South Wales Gas Pipeline Project Site 513
Archaeological Excavation
CA Project: 9150 CA Report: 13261 Event DAT102846
Jonathan Hart, Senior Publications Officer prepared by Alistair Barber, Senior Project Officer Christopher Leonard, Project Officer
date March 2014
checked by Karen E Walker, Post-Excavation Manager
date September 2015
approved by Martin Watts, Project Director, Head of Publications
signed
date
issue 01
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
Cirencester Milton Keynes Andover Building 11 Unit 4 Office 49 Kemble Enterprise Park Cromwell Business Centre Basepoint Business Centre Kemble, Cirencester Howard Way, Newport Pagnell Caxton Close, Andover Gloucestershire, GL7 6BQ MK16 9QS Hampshire, SP10 3FG t. 01285 771022 t. 01908 218320 t. 01264 326549 f. 01285 771033 e. [email protected] © Cotswold Archaeology South Wales Gas Pipeline Project Site 513: Archaeological Excavation
CONTENTS
SUMMARY ...... 5
1. INTRODUCTION ...... 6
The site ...... 6 Archaeological background ...... 7 Archaeological objectives ...... 7 Methodology...... 7
2. RESULTS (FIGS 2–17) ...... 8
3. PROJECT TEAM ...... 18
4. REFERENCES ...... 19
APPENDIX A: CONTEXT DESCRIPTIONS ...... 23
APPENDIX B: GRAVE CATALOGUE ...... 28
APPENDIX C: THE FINDS...... 33
APPENDIX D: PALAEOENVIRONMENTAL EVIDENCE BY JAMES RACKHAM ...... 61
APPENDIX E: THE RADIOCARBON DATING BY SEREN GRIFFITHS ...... 75
3 © Cotswold Archaeology South Wales Gas Pipeline Project Site 513: Archaeological Excavation
LIST OF ILLUSTRATIONS Fig. 1 Site location plan (1:25,000) Fig. 2 Plan of archaeological features (1:250) Fig. 3 Section of pit 268041 (1:20) Fig. 4 Plan of archaeological features (1:75 and 1:30) Fig. 5 Photograph: ring ditch, looking south Fig. 6 Photograph: section through ring ditch, looking north-west Fig. 7 Ring ditch sections (1:20) Fig. 8 Cremation sections (1:20) Fig. 9 Photograph: Burials 1 and 2, looking west Fig. 10 Photograph: Burials 3 and 4, looking south-west Fig. 11 Photograph: Burials 5-10, looking south-east Fig. 12 Photograph: Burial 11, looking south Fig. 13 Photograph: Burials 12 and 13, looking south-west Fig. 14 Photograph: Burials 14 and 15, looking south Fig. 15 Photograph: Burial 16, looking west Fig. 16 Photograph: Burial 20 looking west Fig. 17 Pit sections (1:20) Fig. 18 A Bayesian model for the results from Site 513 (Appendix E) Fig. 19 Key parameters for Site 513 shown in the model calculated in Fig. 18 (Appendix E)
GLOSSARY CA – Cotswold Archaeology CAP – Cambrian Archaeological Projects CPAT – Clwyd Powys Archaeological Trust DAT – Dyfed Archaeological Trust GGAT - Glamorgan Gwent Archaeological Trust FTP – Felindre to Brecon gas pipeline HER – Historic Environment Record MHA – Milford Haven to Aberdulais gas pipeline NAL – Network Archaeology Ltd NLMJV – Nacap Land & Marine Joint Venture UPD – Updated Project Design
4 © Cotswold Archaeology South Wales Gas Pipeline Project Site 513: Archaeological Excavation
SUMMARY
Project Name: South Wales Gas Pipeline Project Location: Site 513, Land North-West of Steynton, Milford Haven, Pembrokeshire NGR: SM 9110 0830 Type: Excavation Date: 24 July–25 August 2006 Location of Archive: To be deposited with RCAHMW (original paper archive) and Scolton Manor Museum (material archive and digital copy of paper archive; accession number 2008.1) Site Code: MHA06
An archaeological excavation was undertaken by Cotswold Archaeology during groundworks associated with construction of gas pipelines (part of the South Wales high pressure gas pipeline scheme) between Milford Haven and Aberdulais, and Felindre and Brecon, which were conducted between 2005 and 2007.
A small number of Mesolithic or Early Neolithic flints were found as residual items. The earliest cut features were two pits containing Late Neolithic Grooved Ware pottery, together with hazelnut shells, worked flints, flint microdebitage and burnt bone. These probably indicate the location of a Late Neolithic settlement and were perhaps associated with a group of undated pits within the same part of the site. Bayesian analysis of data from these pits suggests that this occupation may have been episodic over a period of 110–140 years.
Most of the dated features were associated with a previously unrecorded ring ditch. This enclosed a group of cremation burials, mostly within Early Bronze Age Collared Urns and Food Vessels. The earliest dated use of the ring ditch is 2020–1880 cal. BC and there appears to have been no continuity between the Late Neolithic settlement and the site’s re- use as the setting of a funerary monument during the Early Bronze Age.
5 © Cotswold Archaeology South Wales Gas Pipeline Project Site 513: Archaeological Excavation
1. INTRODUCTION
1.1 NACAP Land and Marine Joint Venture (NLMJV), on behalf of National Grid, commissioned RSK Environment (part of the RSK Group) to manage the archaeological works (non-invasive surveys, desk based assessment, evaluation, watching brief, and open area excavation) on a 216km-long section of pipeline from Milford Haven (Pembrokeshire) to Brecon (in Powys). The high pressure gas pipeline (part of the 316km-long pipeline route from Milford Haven to Tirley in Gloucestershire) was required to reinforce the gas transmission network. The archaeological work performed in advance of this pipeline was undertaken in a number of sections by a number of archaeological companies. The westernmost section of 122km, from Milford Haven to Aberdulais, was investigated by CA (then Cotswold Archaeological Trust) during 2005–2007 with some additional excavation work carried out by CAP. The 89km section from Felindre to Brecon was investigated by CA during 2006–2007 and CAP during 2007. Assessment reports on the works were completed in January 2012 (NLM 2012a, 2012b) and the current reporting stage was commissioned in February 2013.
1.2 In July and August 2006 CA carried out an archaeological excavation at Site 513, Land North-West of Steynton, Milford Haven, Pembrokeshire (centred on NGR: SM 9110 0830; Fig. 1). The objective of the excavation was to record all archaeological remains exposed during the pipeline construction.
1.3 The excavation was carried out in accordance with professional codes, standards and guidance documents (EH 1991; IfA 1999a, 1999b, 2001a, 2001b, 2001c and IfA Wales 2008). The methodologies were laid out in an Archaeological Management Plan (RSK 2006) and a subsequent site-specific Written Scheme of Investigation (CA 2006).
The site 1.4 The site lies within a pasture field on a plateau just below the crest of an unpronounced hill overlooking a minor tributary of Hubberston Pill (Fig. 1). It occupies a moderate north-facing slope and lies at 38m AOD. The surrounding landscape is gently undulating and cut by numerous water courses within steep- sided valleys.
6 © Cotswold Archaeology South Wales Gas Pipeline Project Site 513: Archaeological Excavation
1.5 The underlying solid geology comprises Milford Haven Group Argillaceous Rocks and Interbedded Sandstone of the Devonian and Silurian periods (BGS 2013).
Archaeological background 1.6 No archaeological remains were identified within the site during the preliminary Archaeology and Heritage Survey (CA 2005). An eroded round barrow is located 1.1km south-east of the site (PRN 3182) whilst the HER records the locations of burnt mounds 3km north-west and 1km south-east of the site (PRNs 276065 and 3183). During the pipeline construction, a further burnt mound was found 300m west of the site at Site 512 (CA 2014a; Fig. 1). Other recording along the pipeline route (Fig. 1) in the vicinity of Site 513 revealed an undated but probably prehistoric penannular ditch, postholes and a ditch at Site 519, the remains of either a settlement or a barrow, and a barrow ring ditch dated to the Early Bronze Age at Site 505. Prehistoric pits, probably the remains of Early and Middle Neolithic settlement activity were found at Site 286.
1.7 The geophysical survey (BCC 2005) undertaken in advance of the pipeline construction revealed no potential archaeological features within the site, and consequently no archaeological evaluation was undertaken.
Archaeological objectives 1.8 The objectives of the archaeological works were:- • to monitor groundworks, and to identify, investigate and record all significant buried archaeological deposits revealed on the site during the course of the development groundworks; and • at the conclusion of the project, to produce an integrated archive for the project work and a report setting out the results of the project and the archaeological conclusions that can be drawn from the recorded data.
Methodology 1.9 The fieldwork initially followed the methodology set out within the watching brief Written Scheme of Investigation (WSI; NLM 2006). An archaeologist was present during the removal by machine of topsoil and subsoil within the pipeline easement to the natural substrate. Cremations were first discerned when benching work produced a cleaner finish to the site strip and subsequent work was undertaken in accordance with a site-specific WSI (CA 2006).
7 © Cotswold Archaeology South Wales Gas Pipeline Project Site 513: Archaeological Excavation
1.10 All features were recorded in accordance with CA Technical Manual 1: Fieldwork Recording Manual. A Home Office licence, No. 06-0143, was obtained for the removal of the cremation burials. The ring ditch was 50% excavated by hand and the cremations were 100% excavated. All cremation urns were lifted with their contents intact and underwent micro-excavation in spits and consolidation at Cardiff University’s Conservation Department.
1.11 The post-excavation work was undertaken following the production of the UPD (GA 2012) and included re-examination of the original site records. Finds, environmental and radiocarbon-dating evidence was taken from the assessment report (NLM 2012a) except where the UPD recommended further work, in which case the updated reports were used. The archaeological background to the site was assessed using the following resources:- • the Archaeology and Heritage Survey which was undertaken in advance of the pipeline construction and which examined a 1km-wide corridor centred on the pipeline centre line, including the then existing HER record (CA 2005); • Dyfed Archaeological Trust HER data (received May 2014); and • other online resources, such as Google Earth and Ordnance Survey maps available at http://www.old-maps.co.uk/index.html. All monuments thus identified that were relevant to the site were taken into account when considering the results of the fieldwork.
1.12 The archive and artefacts from the excavation are currently held by CA at their offices in Kemble. Subject to the agreement of the legal landowner the artefacts will be deposited with Scolton Manor Museum under accession number 2008.1, along with a digital copy of the paper archive. The original paper archive will be deposited with the RCAHMW.
2. RESULTS (FIGS 2–17)
2.1 This section provides an overview of the excavation results; detailed summaries of the recorded contexts, burials, finds, environmental samples (palaeoenvironmental evidence) and radiocarbon dates are to be found in Appendices A–E. Full, original versions of the specialist reports are contained within the archive. It should be noted that where human remains from multiple burials within the same urn were
8 © Cotswold Archaeology South Wales Gas Pipeline Project Site 513: Archaeological Excavation
radiocarbon dated, it is not possible to determine which burial the dated remains came from. Radiocarbon dates quoted below are calibrated dates at 95% probability, except those quoted in italics which are posterior density estimates derived from Bayesian analysis. 2.2 The natural geological substrate, comprising pink-red sandstone overlain by orange- brown sandy clay and gravelly clay, was cut by archaeological features (Fig. 2).
Mesolithic/Early Neolithic 2.3 Small numbers of Mesolithic or Early Neolithic flints were recovered as residual items within later features. No cut features of these periods were identified. Given the truncation to the site, it is possible that further remains of this date have been lost.
Late Neolithic c.2890–2450 cal. BC (Figs 2 and 3) 2.4 Two pits contained Late Neolithic pottery. These were found side by side near the western baulk. In addition, 11 fragments of Late Neolithic Grooved Ware were recovered as residual finds from the topsoil. Pit 268019 was circular in plan, 0.2m wide and 0.2m deep with a bowl-shaped profile. It contained clay silt fill 268020 from which 13 Late Neolithic Grooved Ware pottery sherds were recovered, together with 19 sherds broadly dateable as prehistoric. Charcoal from the fill produced Late Neolithic radiocarbon dates of 2890–2660 and 2570–2450 cal. BC (SUERC-54661 and -54662), ranges compatible with the pottery.
2.5 Adjacent pit 268041 was a larger, steeper-sided cut, with a slightly rounded base (Fig. 3, section AA). It was 0.8m wide and 0.4m deep and contained a sequence of fills. The lowest fill, silty clay 268051, was a thin deposit possibly formed as a result of weathering of the pit edges and contained a sherd of Late Neolithic Grooved Ware. It was overlain by clay silt 268050 which contained a further 23 Grooved Ware sherds as well as burnt and unburnt flint microdebitage and flakes. The final fill, clay silt 268049, was devoid of finds and may represent natural infilling of a remnant earthwork. Charcoal from the lower fills returned two statistically consistent radiocarbon dates of 2880–2570 cal. BC (SUERC-54659 and -54660).
2.6 Samples from both pits produced large assemblages of hazelnut shell fragments in addition to which, small quantities of heavily-fragmented burnt bone (not identified to species) came from pit 268019.
9 © Cotswold Archaeology South Wales Gas Pipeline Project Site 513: Archaeological Excavation
Early Bronze Age c. 2020–1700 cal. BC (Figs 2 and 4–16) 2.7 A ring ditch was found 50m east of the Late Neolithic pits (Fig. 2). The northern two thirds of the ring ditch and its interior had been truncated during benching works. The ring ditch comprised a complete ditch circuit, 11.5m in diameter internally (Figs 4 and 5) and the ditch itself was a u-profiled to flat-based cut which, where not recently truncated, was up to 1.8m wide and 0.5m deep (Figs 6 and 7, sections BB– FF). Throughout most of its length the ditch contained a primary red-brown silty clay fill overlain by a grey silty clay fill, both of which probably resulted from natural infilling. No tip lines were present but these deposits may represent infilling from an internal bank since the layout of the cremations within the ring ditch interior suggests that a bank may have been present, covering a width of c. 2m along the inner edge of the ditch. Samples from the primary fills yielded a small quantity of highly fragmented cremated human bone and a piece of fresh flint microdebitage. Other artefactual material from the ring ditch fills consisted only of a Mesolithic/Early Neolithic flake/blade core as well as flint flakes and crumbs of pottery broadly dateable as prehistoric. One sample also produced a large assemblage of wood charcoal suggestive of a localised deposition of fire debris.
2.8 The untruncated (southern) third of the ring ditch interior contained at least 20 and perhaps as many as 23 burials (Burials 1–23) as well as a pit/posthole. The majority of the cremations were found within urns, mainly Collared Urns, although two Food Vessels were also used and a further burial may have been within an organic container of which no physical trace survived. The minimum number of individuals represented is 20, and there was a clear bias towards sub-adults. Sex could not be determined for any of the burials. The burials are described below and full details are contained within the grave catalogue (Appendix B) and other appendices.
2.9 The oldest dated burials were Burials 1 and 2 which were the remains of an adolescent and juvenile respectively, interred together within a Tripartite Vase Food Vessel (Fig. 8, section GG; Fig. 9). The vessel had been placed inverted within pit 268177 and further human bone was found within the pit backfill, although this was in small quantities and perhaps derived from one of the burials. Food Vessels are dateable to 2200–1500 cal. BC (Appendix C) and this dating is compatible with an Early Bronze Age radiocarbon determination on a human bone fragment from the burials (2140–1940 cal. BC; SUERC-54663), which Bayesian analysis has refined to 2020–1880 cal. BC (95% probable; Appendix E).
10 © Cotswold Archaeology South Wales Gas Pipeline Project Site 513: Archaeological Excavation
2.10 Burials 3 and 4 were associated with a comparable radiocarbon date. These comprised the remains of an adult and a sub-adult respectively and had been buried together within a decorated Bipartite Collared Urn (Fig. 10). This had been placed within pit 268126 and at some point in antiquity (perhaps when the pit was backfilled) had fallen on its side and been partially crushed. Cremated bone from one of the burials was radiocarbon dated to 2120–1890 cal. BC (SUERC-54679). This date range is statistically inconsistent with younger date ranges obtained from charred residue on the urn (1890–1690 cal. BC; SUERC-54681) and from charcoal within the pit backfill (1900–1690 cal. BC; SUERC-54680), although these younger dates are statistically consistent with one another.
2.11 Statistically inconsistent radiocarbon dates were also associated with Burials 5–10. These six burials were found within a single pit, 268157 (Fig. 8, section HH; Fig. 11). On the base of this pit, an upright Bipartite Vase Food Vessel contained the remains of an adult, an adolescent and a juvenile (Burials 5–7). Above this vessel had been placed a larger and inverted Tripartite Collared Urn containing the remains of an adolescent and two young juveniles (Burials 8–10). Charred residue on the inverted urn was radiocarbon dated to 1890–1690 cal. BC (SUERC-54673) but a statistically inconsistent and older date came from a fragment of human bone within the urn (2040–1880 cal. BC; SUERC-54672).
2.12 Radiocarbon dating extending into the third millennium BC also came from Burial 11. Here burnt bone identified as probably human had been placed within an upright decorated Tripartite Collared Urn which had been buried within pit 268172 (Fig. 8, section II; Fig. 12). Charcoal from the pit backfill was radiocarbon dated to 2030– 1880 and 1960–1770 cal. BC (SUERC-54678 and -54674).
2.13 Burials 12 and 13 were associated with the next oldest radiocarbon dates. Burial 12 was of a juvenile, whilst Burial 13 was an infant / young juvenile. Both had been deposited within an undecorated Tripartite Collared Urn which was placed inverted within pit 268113 (Fig. 8, section JJ; Fig. 13). Statistically consistent radiocarbon dates were obtained from charred residue on the urn and from a bone fragment (1870–1630 and 1890–1690 cal. BC respectively; SUERC-54668 and -54669). However, an older and statistically inconsistent date came from charcoal within the pit backfill (1960–1770 cal. BC; SUERC-54664).
11 © Cotswold Archaeology South Wales Gas Pipeline Project Site 513: Archaeological Excavation
2.14 Burials 14 and 15 were also associated with radiocarbon dates extending into the 20th century BC. These were the remains of an adult and an infant respectively and had been placed within a Tripartite Collared Urn which had then been buried within pit 268147 (Fig. 8, section KK; Fig. 14). The backfill of this pit contained flints, including fresh microdebitage found in sufficient quantities not to have been residual and therefore likely to have been deposited during the burial rite. A fragment of the human bone was radiocarbon dated to 1940–1750 cal. BC (SUERC-54671) whilst charcoal from the pit fill was dated to 1900–1690 cal. BC (SUERC-54670).
2.15 The remaining burials were not associated with radiocarbon dates. Of these, Burial 16 comprised the remains of an adolescent. Although not found within an urn, these remains were tightly packed within the soil (Fig. 8, section LL; Fig. 15), suggesting that they had been buried within an organic container, of which no physical trace survived, before being buried within pit 268170.
2.16 Burials 17 and 18 were of an adult and a non-adult respectively and had been placed within a decorated Tripartite Collared Urn which was then buried upright within pit 268009. A Mesolithic/Early Neolithic blade core was found within the pit backfill. It is not certain that this latter item was a grave good but it may have been so since a similar item was found alongside Burial 19 (Fig. 8, section MM), the remains of an infant within an undecorated Tripartite Collared Urn which also contained a Mesolithic/Early Neolithic blade core. This urn had been buried upright within pit 268004.
2.17 Burial 20 was of a juvenile placed within an undecorated Tripartite Collared Urn which was buried upright within pit 268142 (Fig. 8, section NN). Burial 21 had been truncated during the benching work and contained no surviving human remains but had probably been a cremation burial since fragments of a decorated Tripartite Collared Urn survived within the remaining pit (268110; Fig. 8, section OO). Burials 22 and 23 were shallow pits containing neither pottery nor human remains but were perhaps the truncated remains of further cremation burials.
2.18 Pit/posthole 268160 (Fig. 8, section PP) was 0.2m wide and 0.15m deep with steep sides and a concave base. Its fill contained a burnt flint flake as well as a large number of charred hazelnut shell fragments.
12 © Cotswold Archaeology South Wales Gas Pipeline Project Site 513: Archaeological Excavation
Medieval and later (Fig. 2) 2.19 Two parallel ditches, 268037 and 268039, terminated within the north-western corner of the site. No relationship between the two was discernible, but modern metalwork (not retained) and a lump of iron ore were found within ditch 268037 (the latter find is discussed in full in Appendix B). Together, these were probably medieval or later field boundaries, perhaps infilled when the railway to the immediate north was constructed, in which case the ore and metalwork may have been imported to site.
Undated (Figs 2 and 17) 2.20 Undated pits were found. Most were in the western half of the site, although two were found east of the ring ditch. These were mostly shallow cuts (Fig. 17) with pale gravelly or silty fills which were notably different to those of the dated Late Neolithic pits. An exception was pit 268129 (Fig. 17, section QQ), a deeper cut with a charcoal-rich upper fill. Two linear features, 268021 and 268036, were also present.
Discussion Late Mesolithic/Early Neolithic 2.21 The small quantities of residual Late Mesolithic/Early Neolithic flints indicate activity of this date in the locality. The small number of these items suggests that they were perhaps discarded during temporary (overnight) occupation of the site, or in transit.
Late Neolithic c.2890–2450 cal. BC 2.22 The two Late Neolithic pits with Grooved Ware are a significant discovery. Radiocarbon determinations on other Welsh sites have suggested that Grooved Ware was in use in Wales during the period c.3000–2300 cal. BC (Appendix B) and the dated Neolithic activity on the current site falls within a range compatible with this dating (c.2890–2450 cal. BC). Pits of this date are uncommon in Wales but further discoveries of Neolithic or possibly Neolithic pits along the pipeline route suggests that they may be more widespread than previously thought and the unexpected discovery of such sites is one of the principal archaeological strengths of linear schemes such as the pipeline.
2.23 Although uncommon, small Neolithic pits are recorded across much of Britain. In the absence of evidence for storage or processing, they are often interpreted as having been specially excavated for the deposition of closure deposits, where specially selected, but often mundane, items representing a range of domestic activities were
13 © Cotswold Archaeology South Wales Gas Pipeline Project Site 513: Archaeological Excavation
placed within the small cuts to commemorate the end of a period of occupation (for example see Carver 2012, 111 and Thomas 2012, 2). This should be placed within the context of the Neolithic environment, which was largely wooded, with periodic occupation having occurred within small man-made clearings. Wide-ranging Neolithic radiocarbon dates from a number of such sites suggests that many of these clearings were re-occupied multiple times and that may be the case on the current site, which Bayesian analysis of the dating evidence indicates may have been occupied episodically over a duration of 110–410 years (95% probable; Appendix E).
2.24 The finds within the Neolithic pits at Site 513 are compatible with domestic debris which could have been collected for burial to commemorate each period of occupation. It may be the case that the Early Neolithic flints within the Late Neolithic pits (which included the only tool found on the site) were curated items, rather than residual pieces included incidentally within the fill. The curation and re-use of earlier tools during the Neolithic period is not without parallel: along the pipeline at Site 47.00 a crude piercer had been manufactured in the Neolithic period using a much older flake (CA 2014c). This re-use may in part reflect the lack of a local flint resource, other than beach pebbles, but could equally have a more esoteric significance, particularly when seen in relation to special deposition.
2.25 It is possible that the undated pits, although morphologically different from the dated Late Neolithic pits, were also Neolithic. If so, seen overall, many of the pits in the western part of the site seem to follow an alignment, albeit a somewhat sinuous one. This might relate to a boundary but other explanations are possible: at Kilverstone, Norfolk, spaces between Neolithic pit clusters were thought to indicate the locations of former buildings (Garrow et al. 2005), no physical remains of which survived. However, the full distribution of the pits on the current site was not exposed (and some may have been lost beneath the railway), so it is not possible to make such interpretations here based on the current evidence.
Early Bronze Age c. 2020–1700 BC 2.26 The discovery and excavation of a Bronze Age ring ditch with multiple cremations is a further significant result. The large assemblage of Collared Urns and Food Vessels used to contain the cremations is again rare in Wales. Statements regarding the number of burials and the construction and abandonment dates of the monument
14 © Cotswold Archaeology South Wales Gas Pipeline Project Site 513: Archaeological Excavation
are restricted by the loss of much data during the benching works. Estimates have been made for the start and end dates of the ring ditch, but these are based on the surviving evidence and it is possible that evidence of earlier and later activity has been lost. It should also be noted that the date of construction of the ditch is not known; although it is assumed here to have been a Bronze Age feature and to be very broadly contemporary with the radiocarbon-dated burials, it is not possible to be certain that it did not significantly pre-date the earliest dated burials.
2.27 Although the construction date of the ring ditch is not known, the earliest dated use of it is 2020–1880 cal. BC (95% probable). Taken at face value, this reveals a gap of up four centuries between the end of Late Neolithic occupation and the first dated use of the ring ditch. The ring ditch was therefore probably constructed within an area that had regenerated with woodland, although it is possible that traces of the earlier occupation were still evident, such as stray finds or differential vegetation growth over infilled pits or above middens.
2.28 The form of the monument represented by the ring ditch is not readily apparent. No physical evidence of a bank or mound was recorded prior to the works and there was no evidence for one in the form of tip lines within the ring ditch fills. However, all of the surviving cremations were found at least 2m from the inner edge of the ring ditch, and it is possible either that a c. 2m-wide inner bank was present and that the cremations were placed within the area thus enclosed, or that a mound extended up to 2m towards the inner edge of the ditch and that cremation burials were dug into this. At the St Clears to Red Roses cremation cemetery, some of the burials were sealed by thin pebbly layers and these may have been the remains of round barrows (CA 2014b). On the current site, the possibility (see below) that several of the burials may have been revisited over time raises the question of whether they might have been physically marked by upstanding features such as cairns of which no physical traces have survived.
2.29 Because of the extent of the truncation, it is not known how many burials were contained within the monument. Certainly, unstratified finds suggest that more burials were originally present than were recorded in situ. The density of the burials within the untruncated part of the ring ditch interior varies and extrapolating from this evidence (which may be unrepresentative of the whole) is therefore problematic but as a broad guide, about a third of the ring ditch interior was untruncated and this contained between 20 and 23 burials, so the monument may perhaps have
15 © Cotswold Archaeology South Wales Gas Pipeline Project Site 513: Archaeological Excavation
contained the remains of some 60 individuals. Alternatively, if burials tended to cluster towards the central area of the interior, perhaps some 40 individuals may have been buried there. It is notable that a disproportionate number of the burials were sub-adults (15 out of 19 burials where age could be assessed were sub adult). Again, this might not be representative of the original total age make-up of the burials but such a bias has been noted at other barrows and has been taken to suggest that they were constructed and used by family groups rather than being monuments used for the burial of social elites (Healy and Harding 2007, 56).
2.30 The presence of multiple burials within single pits and within single vessels is interesting. It was often the case that individuals within the same vessel were of different ages at death and, in some instances the radiocarbon dating evidence suggests that these individuals died at different times. Cremated bone within Burials 3 and 4 was significantly earlier than residue on the urn and charcoal within the pit backfill. Again, human bone from Burials 5–10 was significantly earlier than charred residue from the uppermost urn. Charcoal from the backfill of the pit containing Burials 13 and 4 was also later than a fragment of the bone and residue on the urn containing the burials. Although in the latter instance it is possible that the later charcoal fragment was simply residual, overall these results suggest some chronological and social complexity to the burial rites. It is possible either that some older cremations were curated elsewhere and later interred with more recent cremations, or that older cremations were exhumed so that newer cremations could be added. A third possibility is that older burials were kept in an un-cremated state and then cremated at the same time as an individual who had died much later. Since all of the burials yielding inconsistent radiocarbon dates contained multiple burials, often of widely differing ages at death, this raises the possibility that these were family graves, where the vessels were either exhumed and added to, or where burial was delayed until the deaths of surviving family members. In light of this, it is worth noting the minimum lifespan of the monument, estimated using Bayesian analysis to have been 1–300 years (95% probable) or 130–270 years (68% probable; Appendix E), so perhaps some c. 6–12 generations.
2.31 These details provide insights into the past burial rites. Further glimpses are provided by the presence of finds associated with the burials. Curated flint cores seem to have been selected to accompany some burials; were such items family heirlooms, accidental discoveries or items specially sought after to accompany such events? The flint microdebitage within one of the cremation pits suggests either than
16 © Cotswold Archaeology South Wales Gas Pipeline Project Site 513: Archaeological Excavation
tools were prepared during the burial rite (no tools were found within this burial, so they don’t seem to have been created as grave goods and might instead have been used for food preparation and consumption during the burial rite), or that debitage created elsewhere was carefully gathered and offered as a grave good, perhaps as a reminder of domestic life. The localised charcoal dump within the ring ditch did not include burnt bone and was therefore unlikely to be pyre debris so instead might relate to cooking and eating during the burial, or be hearth waste collected from a domestic context for burial in a mortuary setting. The presence of burnt hazelnut shells within eight of the twelve cremations may provide further evidence of consumption associated with the burial ceremony. The burnt flints and charred hazelnut shells found in the small pit/posthole within the monument provide further evidence of on-site burning or burial of domestic debris within a mortuary setting. The pyre fuel used was largely oak trunkwood, although hazel was used as kindling and was the main fuel used for one of the cremations. It is worth noting that the gathering of this fuel may have formed part of the wider funerary ceremony.
2.32 The last dated activity at the ring ditch occurred in 1870–1700 cal. BC (95% probable). This suggests that it went out of use before the burnt mound activities found at Site 512, 300m to the east, began: Bayesian analysis of the radiocarbon dates from the burnt mound and associated trough suggest that this activity began during 1510–1420 cal. BC (95% probable; CA 2014a). However, an older charcoal fragment found at the mound site as a residual item was radiocarbon dated to 1740– 1560 cal. BC and it is therefore possible that the ring ditch was at least known about, if not actually in use, when the mound activities began. If this was the case, it is difficult to assess the significance of these possibly overlapping dates: if the landscape was still wooded, the ring ditch would not have been visible from the mound but if the area was largely open, the two sites would have been intervisible.
2.33 It is also difficult to assess how the ring ditch would have sat within the landscape during its use. Although it occupies a plateau on a hill, any intervisibility with the surrounding landscape would have been dependent on the extent of the tree cover as the hill is somewhat unpronounced. It is also possible that the monument was sited so as not to have been widely visible.
17 © Cotswold Archaeology South Wales Gas Pipeline Project Site 513: Archaeological Excavation
3. PROJECT TEAM
Fieldwork was undertaken by Kelly Saunders, assisted by Jerry Austin, Justine Biddle, Ben Brogan, Steven Cox, Greg Crees, Jess Davidson, Sarah Dronfield, Jenny Giddins, Dan Heale, Matt Nichol, Tim Power, Anthony Pritchard, Victoria Rees, Sam Thorogood and Mike Turner. This report was written by Alistair Barber, Christopher Leonard and Jonathan Hart with illustrations prepared by Daniel Bashford. The archive has been compiled by Jonathan Hart and prepared for deposition by Hazel O’Neill. The fieldwork was managed for CA by Clifford Bateman and the post-excavation was managed for CA by Karen Walker.
18 © Cotswold Archaeology South Wales Gas Pipeline Project Site 513: Archaeological Excavation
4. REFERENCES
Anderson-Whymark, H. and Thomas, J. 2012 Regional Perspectives on Neolithic Pit Deposition: Beyond the Mundane, Neolithic Studies Group Seminar Papers 12. Oxford, Oxbow Books, 216–225 BCC (Bartlett-Clark Consultancy) 2005 Milford Haven to Aberdulais Proposed Gas Pipeline: Archaeogeophysical Survey 2005 Benson, D. G., Evans, J. G. and Williams, G. H. 1990 ‘Excavations at Stackpole Warren, Dyfed’, Proc. Prehist. Soc. 56, 179–246 BGS (British Geological Survey) 2013 Geology of Britain Viewer. Online resource at http://mapapps.bgs.ac.uk/geologyofbritain/home.html accessed 4 November 2013 Brindley, A. 1999 ‘Irish Grooved Ware’, in Cleal, R. and MacSween, A. (eds) Grooved Ware in Britain and Ireland, 23–35. Oxford, Oxbow Books Britnell, W. J. 1982 ‘The Excavation of two Round Barrows at Trelystan, Powys’, Proc. Prehist. Soc. 48, 133–202 Burgess, C. B. 1986 ‘‘Urnes of no Small Variety’: Collared Urns reviewed’, Proc. Prehist. Soc. 52, 339–351 Burrow, S. and Williams, S., 2008 The Wales and Borders Radiocarbon Database. Amgueddfa Cymru. National Museum Wales (www.museumwales.ac.uk/en/radiocarbon) Caffell, A. and Holst, M. 2009 ‘Assessment Report for Cremated Bone’ in NLM 2012b Carver, G. 2012 ‘Pits and Place-making: Neolithic Habitation and Deposition Practices in East Yorkshire c. 4000–2500 BC’, in Proc. Prehist. Soc. 78, 111–134 CA (Cotswold Archaeology) 2005 Milford Haven to Aberdulais Gas Pipeline: Archaeology and Heritage Survey. CA typescript report 04147 CA (Cotswold Archaeology) 2006 Possible barrow site on land north-west of Steynton (RDX 182), Milford Haven, Pembrokeshire. Milford Haven To Aberdulais Gas Pipeline. Written Scheme of Investigation for Archaeological Excavation CA (Cotswold Archaeology) 2013 South Wales Pipeline Project Site 257, Land East of Rose Hill, Rosemarket, Pembrokeshire: Archaeological Watching Brief. CA typescript report 13208 CA (Cotswold Archaeology) 2014a South Wales Pipeline Project Site 512, Land North of Steynton, Milford Haven, Carmarthenshire: Archaeological Excavation. CA typescript report 13355
19 © Cotswold Archaeology South Wales Gas Pipeline Project Site 513: Archaeological Excavation
CA (Cotswold Archaeology) 2014b A477 St Clears to Red Roses Road Improvement, Carmarthenshire: Post-Excavation Assessment and Updated Project Design. CA typescript report 14014 CA (Cotswold Archaeology) 2014c South Wales Pipeline Project. Site 47.00, Land North- West of Cwm-Camlais-Isaf, Trallong, Penpont and Llanfihangel, Powys: Archaeological Excavation. CA typescript report 13324 Cleal, R. 1988 ‘The occurrence of drilled holes in Later Neolithic pottery’, Ox. J. Arch. 7, 139–45 Cox, M. 2000 ‘Ageing adults from the skeleton’, in M. Cox and S. Mays (eds), Human Osteology in Archaeology and Forensic Science, 61-82 Crane, P. and Murphy, K. 2010 ‘Early medieval settlement, iron smelting and crop processing at South Hook, Herbranston, Pembrokeshire, 2004–05’, Archaeologia Cambrensis 159, 117–196 EH (English Heritage) 1991 The Management of Archaeological Projects 2 GA (Groundwork Archaeology) 2012 Milford Haven to Aberdulais and Felindre to Brecon High Pressure Gas Pipelines: Updated Project Design Garrow, D. Beadsmoore, E. and Knight, M. 2005 ‘Pit Clusters and the Temporality of Occupation: an Earlier Neolithic Site at Kilverstone, Thetford, Norfolk’. Proc. Prehist. Soc. 71, 139–158 Garwood, P. 1999 ‘Grooved Ware in Southern Britain: Chronology and Interpretation’, In Cleal, R. and MacSween, A. (eds) Grooved Ware in Britain and Ireland, 145–176 Gibson, A. M. 1978 Bronze Age pottery in the North-east of England. BAR Brit. Series 56. Gibson, A. M. 1999 The Walton Basin Project: Excavation and Survey in a Prehistoric Landscape 1993-1997. Research Report 118. York, Council for British Archaeology and Cadw Welsh Historic Monuments Gibson, A. 2013 Milford Haven to Aberdulais and Felindre to Brecon Gas Pipeline: The Neolithic and Bronze Age Pottery from the Archaeological Investigations, Gibson typescript report no. 121 Giorgi, S., and Martin, G. 2009 ‘Assessment Report for the Archaeobotanical Remains’, in NLM 2012a Hall, R. 2008 ‘Assessment Report for Fired Clay and Daub’, in NLM 2012a Healy, F. and Harding, J. 2007 ‘A thousand and one things to do with a round barrow’, in Last, J. (ed.) Beyond the Grave: new perspectives on round barrows, 53–71 IfA (Institute for Archaeologists) 1999a Guidelines for Finds Work. IfA, Birmingham IfA (Institute for Archaeologists) 1999b Standard and Guidance for Finds and Ecofact Studies and Curation. IfA, Reading
20 © Cotswold Archaeology South Wales Gas Pipeline Project Site 513: Archaeological Excavation
IfA (Institute for Archaeologists) 2001a Standard and Guidance for the Collection, Documentation, Conservation and Research of Archaeological Materials. IfA (Institute for Archaeologists) 2001b Standard and Guidance for an Archaeological Watching Brief IfA (Institute for Archaeologists) 2001c Standard and Guidance for Archaeological Excavation IfA Wales (Institute for Archaeologists of Wales/Cymru) 2008 Introducing a Research Framework for the Archaeology of Wales, online resource at http://www.archaeoleg.org.uk/intro.html accessed December 2008 Law, R. 2008. The Development and Perpetuation of a Ceramic Tradition: The Significance of Collared Urns In Early Bronze Age Social Life. Unpublished PhD Thesis Lewis, M. E. 2000 ‘Non-adult palaeopathology: current status and future potential’, in M. Cox and S. Mays (eds) Human Osteology in Archaeology and Forensic Science Longworth, I. H. 1984 Collared Urns of the Bronze Age in Great Britain and Ireland. Cambridge, Cambridge University Press Lynch, F. M. 1991 Prehistoric Anglesey (2nd edn). Llangefni, Anglesey Antiquarian Society Lynch, F. M. 1993 Excavations in the Brenig Valley. A Mesolithic and Bronze Age Landscape in North Wales. Cambrian Archaeological Monographs No.5 Mays, S. and Cox, M. 2000 ‘Sex determination in skeletal remains’, in M. Cox and S. Mays (eds) Human Osteology in Archaeology and Forensic Science, 117-130 McKinley, J. I. 1994 ‘Bone fragment size in British cremation burials and its implications for pyre technology and ritual’, J. Arch. Science 21, 339-342 Millson, D., Waddington, C. and Marshall, P. 2011 ‘Towards a Sequence for Neolithic Ceramics in the Milfield Basin, and Northumberland’, Archaeologia Aeliana 5th ser. 40, 1–40 Needham, S. 2005 Transforming Beaker Culture in North-west Europe: Processes of Fusion and Fission. Proc. Prehist. Soc. 71, 171–217 NLM (Nacap Land and Marine) 2006 Milford Haven to Aberdulais Natural Gas Pipeline: Scheme of investigation for a programme of archaeological works NLM (Nacap Land and Marine) 2012a Milford Haven to Aberdulais High Pressure Gas Pipeline: Archaeology Assessment of Potential for Analysis NLM (Nacap Land and Marine) 2012b Felindre to Brecon High Pressure Gas Pipeline: Archaeology Assessment of Potential for Analysis Ortner, D. J. 2003 Identification of Pathological Conditions in Human Skeletal Remains Pannett, A. 2014 Milford Haven to Brecon LNG Pipeline: Lithic Analysis Report Rackham, J., Challinor, D., Langdon, C. and Scaife, R. in prep. Palaeoenvironmental studies along the Milford Haven to Aberdulais and Felindre to Tirley Natural Gas Pipeline
21 © Cotswold Archaeology South Wales Gas Pipeline Project Site 513: Archaeological Excavation
Read, D., Cooper, D. C. and McArthur, J. M. 1987 ‘The composition and distribution of nodular monazite in the Lower Palaeozoic rocks of Great Britain’, Mineralogical Magazine 51, 271–280 Roberts, C. A. and Cox, M. 2003 Health and Disease in Britain Roberts, C. A. and Manchester, K. 2005 The Archaeology of Disease RSK (RSKENSR) 2006 Milford Haven to Aberdulais Natural Gas Pipeline: Archaeological Management Plan. Nacap Land and Marine Final, RSKENSR Environmental Ltd Savory, H. N. 1957 ‘A Corpus of Welsh Bronze Age Pottery. Part II: Food vessels and Enlarged Food Vessels. Middle Bronze Age (c.1500–1200 BC)’, Bulletin of the Board of Celtic Studies 17, 196–233 Savory, H. N. 1980 Guide Catalogue to the Bronze Age Collections. Cardiff, NMW Scheuer, L. and Black, S. 2000a ‘Development and ageing of the juvenile skeleton’, in M. Cox and S. Mays (eds), Human Osteology in Archaeology and Forensic Science Scheuer, L. and Black, S. 2000b Developmental Juvenile Osteology Stuart-Macadam, P. 1992 ‘Anemia in past populations’, in P. Stuart-Macadam and S. Kent (eds) Diet, Demography and Disease: Changing Perspectives of Anemia, 151-170 Thomas, J. 2012 ‘Introduction: beyond the mundane?’, in Anderson-Whymark and Thomas 2012, 1–12 Thompson, G.B. 1999 ‘The analysis of wood charcoals from selected pits and funerary contexts’, in A Barclay and C Halpin, Excavations at Barrow Hills, Radley, Oxfordshire, volume 1: the Neolithic and Bronze Age monument complex. Thames Valley Landscapes, 11, 247-253, Oxford, Oxford Archaeological Unit Wainwright, G. J. and Longworth, I. H. 1971 ‘Durrington Walls: Excavations 1966-1968’, Soc. Antiq. Res. Rep. 29 Walker, P. L., Bathurst, P. R., Richman, R., Gjerdrum, T. and Andrushko, V. A. 2009 ‘The causes of porotic hyperostosis and cribra orbitalia: a reappraisal of the iron- deficiency-anemia hypothesis’ American J. Physical Anthropology 39, 109–125 Ward, G. K. and, Wilson, S.R. 1978 ‘Procedures for Comparing and Combining Radiocarbon Age Determinations: a critique’, in Archaeometry 20, 19–31 Young, T.P. 2000. ‘Chapter 10. The Paviland Ochres: characterisation and sourcing’, in Aldhouse-Green, S. (ed) Paviland Cave and the 'Red Lady': a definitive report. Western Academic and Specialist Press Limited, 205- 225 Young, T.P. 2010 Analysis of archaeometallurgical residues from Brownslade, Pembrokeshire [NPRN 94225]. GeoArch typescript report no. 2010/07 Young, T. 2014 Archaeometallurgical residues from the Milford Haven–Brecon Pipeline. GeoArch typescript report no. 2013/30
22 © Cotswold Archaeology South Wales Gas Pipeline Project Site 513: Archaeological Excavation
APPENDIX A: CONTEXT DESCRIPTIONS
Context Fill of Interpretation Description L W Depth Spot date No. (m) (m) (m) 268000 Topsoil Mid Brown-grey sand-silt 0.1 268001 Subsoil Mid grey-brown sand-silt 0.25 268002 Natural Mid pink-red and yellow sandstone 268003 Natural Mid orange-brown sands and gravel-clay 268004 Cremation Sub-circular with steeply- 0.35 0.3 pit sloping sides and concave base 268005 268004 Black-brown silt-clay with 0.2 0.15 charcoal and burnt bone 268006 268004 Cremation pit Mid orange-brown silt-clay 0.35 0.2 EBA fill 268007 268004 Urn Cremation urn EBA 268008 268007 Urn fill Mid brown silt-clay with burnt 0.15 0.15 bone 268009 Cremation Sub-circular with steeply- pit sloping sides and concave base 268010 268009 Cremation pit Mid orange-brown silt-clay EBA fill 268011 268009 Urn Cremation urn EBA 268012 268011 Urn fill Mid brown silt-clay with burnt bone 268013- natural features 268015 268016 ?gully Curving feature with gently- 0.55 0.15 sloping sides and concave base 268017 268016 ?gully fill Mid brown grit 0.55 0.15 268018 268016 ?gully fill Mid orange-brown clay-silt 0.5 0.05 268019 Pit Circular with steeply-sloping > 0.25 sides and concave base 0.25 268020 268019 Pit fill Mid pink-brown clay-silt > 0.25 2570–2450 cal BC 0.25 2890–2660 cal BC 268021- natural features 268026 268027 268028 Pit fill Mid red-brown silt-sand 1.05 0.9 0.2 268028 Pit Oval with moderately-sloping 1.05 0.9 0.2 sides and concave base 268029 context not used 268030 context not used 268031 natural feature 268032 natural feature 268033 Ring-ditch Curving ditch with moderately- 0.95 0.3 sloping sides and concave base 268034 268033 Ring-ditch fill Mid grey-brown stony silt-clay 0.95 0.3 268035 natural feature 268036 natural feature 268037 Ditch Straight with steeply-sloping >1 0.75 0.25 sides and flat base 268038 268037 Ditch fill Red-brown silt-clay >1 0.75 0.25 268039 Ditch Straight with steeply-sloping >1 1.1 0.3 sides and irregular base 268040 268039 Ditch fill Red-brown silt-clay >1 1.1 0.3
23 © Cotswold Archaeology South Wales Gas Pipeline Project Site 513: Archaeological Excavation
268041 Pit Circular with steeply-sloping 0.85 0.7 0.4 sides and concave base 268042 context not used 268043 context not used 268044 Ring-ditch Curving with moderately- 1.25 0.5 sloping sides and concave base 268045 268044 Ring-ditch fill Mid red-brown silt-clay 1.25 0.4 268046 Pit Sub-oval with irregular sides > 1.4 0.25 and concave base 0.75 268047 268046 Pit fill Light brown-yellow stony silt- > 1.4 0.25 clay 0.75 268048 268044 Ring-ditch fill Mid red grey-brown stony silt- 0.5 0.2 clay 268049 268041 Pit fill Mid pink-brown clay-silt 0.8 0.7 0.15 268050 268041 Pit fill Mid brown stony clay-silt and 0.75 0.2 2880–2750 cal BC charcoal 268051 268041 Pit fill Mid pink-orange-brown stony 0.7 0.1 2880–2570 cal BC silt-clay 268052 context not used 268053 Ring-ditch Curving with moderately- 1.1 0.3 sloping sides and concave base 268054 268053 Ring-ditch fill Mid red-brown stony silt-clay 1.1 0.25 268055 context not used 268056 Ring-ditch Curving with moderately- 1.25 0.2 sloping sides and concave base 268057 268056 Ring-ditch fill Pink-brown stony silt-clay 1.25 0.2 5 268058 268059 Ring-ditch fill Mid pink-brown silt-clay 0.95 0.35 268059 Ring-ditch Curving with steeply-sloping 1 0.4 sides and uneven base 268060 Ring-ditch Curving with steeply-sloping 1.1 0.25 sides and uneven base 268061 268060 Ring-ditch fill Mid pink-brown stony silt-clay 1.1 0.25 268062 Ring-ditch Curving with gently-sloping 1.05 0.2 sides and concave base 268063 268062 Ring-ditch fill Mid pink-brown silt-clay 1.05 0.2 268064 Ring-ditch Curving with moderately- 0.85 0.25 sloping sides and concave base 268065 268064 Ring-ditch fill Light pink-brown silt-clay 0.55 0.1 268066 268064 Ring-ditch fill Light grey-brown silt-clay 0.85 0.15 268067 268059 Ring-ditch fill Dark pink-brown stony silt-clay 0.45 0.1 268068 268070 Ring-ditch fill Light ink-brown clay-silt 0.55 0.2 268069 268070 Ring-ditch fill Mid red-brown sand-silt 1.8 0.25 268070 Ring-ditch Curving with steeply-sloping 1.8 0.5 sides and flat base 268071 268070 Ring-ditch fill Light grey-brown stony sand- 1.3 0.2 silt 268072 Ring-ditch Curving with moderately- 1.2 0.5 sloping sides and flat base 268073 268072 Ring-ditch fill Mid orange-grey-brown stony 0.8 0.25 BA clay-silt 268074 268092 Ring-ditch fill Mid grey-pink stony silt-clay 0.8 0.2 268075 Ring-ditch Curving with moderately- 1.2 0.45 sloping sides and flat base 268076 268075 Ring-ditch fill Light brown stony-silt 1.2 0.2
24 © Cotswold Archaeology South Wales Gas Pipeline Project Site 513: Archaeological Excavation
268077 268075 Ring-ditch fill Mid orange-brown stony clay- 0.9 0.2 silt 268078 268075 Ring-ditch fill Mid purple-brown stony clay- 0.25 0.05 silt 268079 Ring-ditch Curving with moderately- 1.05 0.4 sloping sides and uneven base 268080 268079 Ring-ditch fill Light grey-brown stony silt- 1.05 0.2 clay 268081 268079 Ring-ditch fill Light grey stony silt-clay 0.75 0.15 268082 268072 Ring-ditch fill Mid grey-brown stony sand-silt 1.2 0.2 268083 268072 Ring-ditch fill Dark orange-brown silt-clay 0.7 0.15 268084 Ring-ditch Rounded terminal with 0.5 0.15 moderately-sloping sides and concave base 268085 268084 Ring-ditch fill Mid orange-brown sand-silt 0.4 0.15 268086 Ring-ditch Curving with moderately- 0.8 0.15 sloping sides and flat base 268087 268086 Ring-ditch fill Mid orange-brown stony-silt 0.8 0.15 268088 Ring-ditch Curving with moderately- 0.85 0.3 sloping sides and uneven base 268089 268088 Ring-ditch fill Pink-brown silt-clay 0.8 0.1 268090 268088 Ring-ditch fill Light brown-pink stony silt- 0.5 0.2 clay 268091 268053 Ring-ditch fill Mid grey-brown silt-clay 1.1 0.1 268092 Ring-ditch Curving with steeply-sloping 1 0.35 sides and uneven base 268093 Ring-ditch Curving with steeply-sloping 0.9 0.25 sides and flat base 268094 268093 Ring-ditch fill Mid pink-brown silt 0.75 0.15 268095 268092 Ring-ditch fill Dark pink-brown stony silt-clay 1 0.15 268096 268093 Ring-ditch fill Grey pink-brown stony silt 0.85 0.15 268097 Ring-ditch Curving with steeply-sloping 1.15 0.4 sides and uneven base 268098 268097 Ring-ditch fill Mid pink-brown stony silt-clay 0.9 0.3 268099 268097 Ring-ditch fill Light pink-brown stony clay 0.9 0.3 268100 Pit Oval with steeply-sloping 0.55 0.8 0.2 sides and concave base 268101 268100 Pit fill Dark red-brown gravel-silt 0.55 0.8 0.2 268102 Pit Oval with gently-sloping sides 0.6 0.45 0.05 and uneven base 268103 268102 Pit fill Light pink-brown silt-clay 0.6 0.4 0.05 268104 Pit or tree- Kidney-shaped with steeply- 0.7 0.2 0.1 throw pit sloping sides and irregular base 268105 26910 Pit or tree- Mid brown gravel sand-silt 0.7 0.2 0.1 throw pit fill 268106 Pit Oval with gently-sloping sides 0.6 0.75 0.1 and uneven base 268107 268106 Pit fill Mid red-brown gravel-sand-silt 0.6 0.75 0.1 268108 Pit Sub-circular with irregular 0.8 0.55 0.15 sides and base 268109 268108 Pit fill Mid yellow-brown gravel-sand- 0.8 0.55 0.15 silt 268110 Cremation pit Circular with steeply-sloping 0.5 0.2 sides and uneven base 268111 268110 Urn and fill Mid pink-brown silt-clay 0.25 0.2 EBA 268112 268110 Cremation pit Light pink-brown silt-clay and 0.25 0.2 fill charcoal
25 © Cotswold Archaeology South Wales Gas Pipeline Project Site 513: Archaeological Excavation
268113 Cremation pit Sub-circular with near vertical 0.3 0.2 sides and concave base 268114 268113 Cremation pit Mid brown-grey sand silt-clay 0.3 0.2 1960–1770 cal BC fill 268115 268113 Urn Cremation urn 0.15 0.2 1870–1630 cal BC 1890–1690 cal BC 268116 Pit Oval with moderately-sloping 0.9 0.15 sides and concave base 268117 268116 Pit fill Dark red-brown sand-silt 0.9 0.15 268118 Ditch Straight with moderately- 0.85 0.25 sloping sides and concave base 268119 268118 Ditch fill Light orange-brown silt-clay 0.6 0.1 268120 Ditch Straight with gently-sloping 0.8 0.3 sides and uneven base 268121 268120 Ditch fill Orange-brown stony silt-clay 0.8 0.3 268122 268120 Ditch fill Light brown silt-clay 0.8 0.1 268123 268118 Ditch fill Light orange-brown silt-clay 0.75 0.15 268124 268125 Pit fill Light brown stony-silt 0.7 0.1 268125 Pit Oval with moderately-sloping 0.7 0.1 sides and flat base 268126 Cremation pit Oval with steeply-sloping 0.5 0.35 sides and concave base 268127 268126 Cremation pit Dark grey to black-brown clay- 0.5 0.07 1890–1690 cal BC fill silt with burnt stone and urn 1900–1690 cal BC frags 2120–1890 cal BC 268128 268126 Cremation pit Mid pink-brown stony silt-clay 0.45 0.65 0.25 fill 268129 Pit Sub-circular with moderately- 2 1.45 0.5 sloping sides and flat base 268130 Pit Sub-square with steeply- 0.75 0.75 0.35 sloping sides and flat base 268131 Layer Mid brown-grey stony silt-clay 0.1 268132 Cremation pit Irregular with irregular sides 0.45 0.2 0.2 and base 268133 268132 Cremation pit Mid orange-brown sand-silt 0.45 0.2 0.2 fill with frequent charcoal 268134 268129 Pit fill Mid red-brown sand-silt and 1.5 0.8 0.15 charcoal 268135 268129 Pit fill Mid grey-brown stony clay-silt 2 0.8 0.35 268136 268129 Pit fill Light red-brown stony clay-silt 1.3 0.8 0.1 268137 268130 Pit fill Mid orange-brown silt 0.75 0.75 0.35 268138 Pit Sub-circular with steeply- 0.9 0.3 sloping sides and uneven base 268139 268138 Pit fill Mid brown-orange stony silt- 0.9 0.3 clay 268140 Cremation pit Circular with steely-sloping 0.45 0.25 0.05 sides and flat base 268141 268140 Cremation pit Dark orange-brown sand-silt 0.45 0.25 0.05 fill with burnt bone 268142 Cremation pit Circular with moderately- > > sloping sides and concave 0.55 0.25 base 268143 268142 Cremation pit Light orange-brown silt-clay > > EBA fill 0.55 0.25 268144 268142 Urn Cremation urn 0.3 0.3 0.2 EBA 268145 268147 Pit fill Mid brown-grey silt-clay and 0.3 0.25 1900–1690 cal BC charcoal 268146 268147 Urn Cremation urn 0.3 0.25 1940–1750 cal BC
26 © Cotswold Archaeology South Wales Gas Pipeline Project Site 513: Archaeological Excavation
268147 Cremation pit Sub-circular with steeply- 0.35 0.25 sloping sides and concave base 268148 Ditch = 268039 2.2 0.45 268149 Ditch = 268037 1 0.25 268150 268148 Ditch fill Mid grey-brown silt-clay 1.6 0.45 268151 268148 Ditch fill Light yellow-brown sand-silt 0.4 1 268152 268148 Ditch fill Mid grey-brown stony clay-silt 1.8 0.35 268153 268148 Ditch fill Mid yellow-brown clay-silt 0.45 0.1 268154 context not used. Finds from this context are from 268001. 268155 context not used 268156 context not used 268157 Cremation pit Circular with steely-sloping 0.35 0.4 sides and flat base 268158 268157 Cremation pit Dark grey-brown stony silt- 0.35 0.4 EBA fill clay 268159 268157 Urns 2 x cremation urns 0.2 0.4 1890–1690 cal BC 2040–1880 cal BC 268160 Pit/posthole Circular with steeply-sloping 0.2 0.2 0.15 sides and concave base 268161 268160 Pit/posthole fill Mid grey-brown silt 0.2 0.2 0.15 268162- context not used 169 268170 Cremation pit Circular with near vertical 0.4 0.15 sides and concave base 268171 268170 Cremation pit Mid yellow-brown silt-clay and 0.4 0.15 fill burnt bone 268172 Cremation pit Sub-circular with steeply- 0.25 0.2 sloping sides and concave base 268173 268172 Urn Cremation urn 0.2 0.2 EBA 268174 268172 Cremation pit Mid orange-brown sand silt- 0.25 0.2 1960–1770 cal BC fill clay with charcoal 2030–1880 cal BC 268175 268177 Cremation pit Dark brown sand silt-clay and 035 0.3 EBA fill charcoal 268176 268177 Urn Cremation urn 0.2 0.25 2140–1940 cal BC 268177 Cremation pit Circular with steeply-sloping 0.35 0.3 sides and concave base
27 © Cotswold Archaeology South Wales Gas Pipeline Project Site 513: Archaeological Excavation
APPENDIX B: GRAVE CATALOGUE
Pit 268177 Burial 1 Age: adolescent 13-17 years Urn: 268176 Burial rite: double burial with Burial 2 within inverted Tripartite Vase Food Vessel Grave goods: human bone within cremation pit fill Dating evidence:- food vessel dateable to 2200-1500 cal. BC from cremated human bone (burial number not known): 2140-1940 cal. BC (SUERC-54663)
Burial 2 Age: juvenile 4-6 years Urn: 268176 Burial rite: double burial with Burial 1 within inverted Tripartite Vase Food Vessel Grave goods: human bone within cremation pit fill Dating evidence:- from cremated human bone (burial number not known): 2140-1940 cal. BC (SUERC-54663); 2020–1880 cal. BC (95% probable)
Pit 268126 Burial 3 Age: adult >18 years Urn: 268127 Burial rite: double burial with Burial 4 within in decorated Bipartite Collared Urn broken in antiquity and compressed on its side Grave goods: none Dating evidence:- from charcoal in cremation pit backfill: 1900-1690 cal. BC (SUERC-54680) from residue on urn: 1890-1690 cal. BC (SUERC-54681) from cremated bone (burial number not known): 2120-1890 cal. BC (SUERC-54679)
Burial 4 Age: sub-adult <18 years Urn: 268127 Burial rite: double burial with Burial 3 within in decorated Bipartite Collared Urn broken in antiquity and compressed on its side Grave goods: none Dating evidence:- from charcoal in cremation pit backfill: 1900-1690 cal. BC (SUERC-54680) from residue on urn: 1890-1690 cal. BC (SUERC-54681) from cremated bone (burial number not known): 2120-1890 cal. BC (SUERC-54679)
28 © Cotswold Archaeology South Wales Gas Pipeline Project Site 513: Archaeological Excavation
Pit 268157 Burial 5 Age: adult >36 years Urn: 268159a Burial rite: multiple burial with Burials 6 and 7 within upright Bipartite Vase Food Vessel along; inverted larger Tripartite Collared Urn containing multiple Burials 8, 9 and 10 placed above this Grave goods: unidentified burnt bone in pit fill Dating evidence:- food vessel dateable to 2200-1500 cal. BC from cremated bone (burial number not known): 2040-1880 cal. BC (SUERC-54672)
Burial 6 Age: adolescent <14 years Urn: 268159a Burial rite: multiple burial with Burials 6 and 7 within upright Bipartite Vase Food Vessel along; inverted larger Tripartite Collared Urn containing multiple Burials 8, 9 and 10 placed above this Grave goods: unidentified burnt bone in pit fill Dating evidence:- food vessel dateable to 2200-1500 cal. BC from cremated bone (burial number not known): 2040-1880 cal. BC (SUERC-54672)
Burial 7 Age: juvenile 5.1-6.2 years Urn: 268159a Burial rite: multiple burial with Burials 6 and 7 within upright Bipartite Vase Food Vessel along; inverted larger Tripartite Collared Urn containing multiple Burials 8, 9 and 10 placed above this Grave goods: unidentified burnt bone in pit fill Dating evidence:- food vessel dateable to 2200-1500 cal. BC from cremated bone (burial number not known): 2040-1880 cal. BC (SUERC-54672)
Burial 8 Age: adolescent <14 years Urn: 268159b Burial rite: multiple burial with Burials 9 and 10 within inverted upper Tripartite Collared Urn along; placed above smaller upright upright Bipartite Vase Food Vessel containing multiple Burials 6, 7 and 8 Grave goods: unidentified burnt bone in pit fill Dating evidence:- from residue on urn: 1890-1690 cal. BC (SUERC-54673) from cremated bone (burial number not known): 2040-1880 cal. BC (SUERC-54672)
29 © Cotswold Archaeology South Wales Gas Pipeline Project Site 513: Archaeological Excavation
Burial 9 Age: young juvenile 1.8-2 years Urn: 268159b Burial rite: multiple burial with Burials 9 and 10 within inverted upper Tripartite Collared Urn along; placed above smaller upright upright Bipartite Vase Food Vessel containing multiple Burials 6, 7 and 8 Grave goods: unidentified burnt bone in pit fill Dating evidence:- from residue on urn: 1890-1690 cal. BC (SUERC-54673) from cremated bone (burial number not known): 2040-1880 cal. BC (SUERC-54672)
Burial 10 Age: young juvenile 1.8-2 years Urn: 268159b Burial rite: multiple burial with Burials 9 and 10 within inverted upper Tripartite Collared Urn along; placed above smaller upright upright Bipartite Vase Food Vessel containing multiple Burials 6, 7 and 8 Grave goods: unidentified burnt bone in pit fill Dating evidence:- from residue on urn: 1890-1690 cal. BC (SUERC-54673) from cremated bone (burial number not known): 2040-1880 cal. BC (SUERC-54672)
Pit 268172 Burial 11 Age: not known (probably human) Urn: 268173 Burial rite: within upright decorated Tripartite Collared Urn Grave goods: none Dating evidence:- from charcoal in cremation pit backfill: 1960-1770 cal. BC (SUERC-54674) from charcoal in cremation pit backfill: 2030-1880 cal. BC (SUERC-54678)
Pit 268113 Burial 12 Age: juvenile 4 years Urn: 268115 Burial rite: double burial with Burial 13 within inverted undecorated Tripartite Collared Urn Grave goods: none Dating evidence:- from charcoal in cremation pit backfill: 1960-1770 cal. BC (SUERC-54664) from residue on urn: 1870-1630 cal. BC (SUERC-54668) from cremated bone (burial number not known): 1890-1690 cal. BC (SUERC-54669)
30 © Cotswold Archaeology South Wales Gas Pipeline Project Site 513: Archaeological Excavation
Burial 13 Age: infant/young juvenile 0.8-2.9 years Urn: 268115 Burial rite: double burial with Burial 12 within inverted undecorated Tripartite Collared Urn Grave goods: none Dating evidence:- from charcoal in cremation pit backfill: 1960-1770 cal. BC (SUERC-54664) from residue on urn: 1870-1630 cal. BC (SUERC-54668) from cremated bone (burial number not known): 1890-1690 cal. BC (SUERC-54669)
Pit 268147 Burial 14 Age: adult >30 years Urn: 268146 Burial rite: double burial with Burial 15 within inverted Tripartite Collared Urn Grave goods: 20 pieces of flint microdebitage, plus burnt flint flakes and chunks from pit backfill Dating evidence:- from charcoal in cremation pit backfill: 1900-1690 cal. BC (SUERC-54670) from cremated bone (burial number not known): 1940-1750 cal. BC (SUERC-54671)
Burial 15 Age: infant 0.3-0.5 years Urn: 268147 Burial rite: double burial with Burial 14 within inverted Tripartite Collared Urn Grave goods: 20 pieces of flint microdebitage, plus burnt flint flakes and chunks from pit backfill Dating evidence:- from charcoal in cremation pit backfill: 1900-1690 cal. BC (SUERC-54670) from cremated bone (burial number not known): 1940-1750 cal. BC (SUERC-54671)
Pit 268170 Burial 16 Age: adolescent <14 years Urn: none Burial rite: buried within organic container? Grave goods: none Dating evidence: none
Pit 268009 Burial 17 Age: adult >18 years Urn: 268011 Burial rite: double burial with Burial 18 within upright decorated Tripartite Collared Urn Grave goods: Mesolithic/Early Neolithic blade core in pit backfill Dating evidence: urn
31 © Cotswold Archaeology South Wales Gas Pipeline Project Site 513: Archaeological Excavation
Burial 18 Age: non-adult (<14 years) Urn: 268011 Burial rite: double burial with Burial 17 within upright decorated Tripartite Collared Urn Grave goods: Mesolithic/Early Neolithic blade core in pit backfill Dating evidence: urn
Pit 268004 Burial 19 Age: infant 0.4-0.8 years Urn: 268007 Burial rite: within upright undecorated Tripartite Collared Urn Grave goods: Mesolithic/Early Neolithic blade core within urn Dating evidence: urn
Pit 268142 Burial 20 Age: juvenile 1.5-4 years Urn: 268144 Burial rite: within upright undecorated Tripartite Collared Urn Grave goods: unidentified burnt bone in pit fill Dating evidence: urn
Pit 268110 Burial 21 Age: no human remains found but pit was truncated during benching Urn: decorated Tripartite Collared Urn: fragments only Burial rite: not known Grave goods: none Dating evidence: urn
Pit 268140 Burial 22 Age: no human remains found Urn: none Burial rite: not known Grave goods: none Dating evidence: none
Pit 268132 Burial 23 Age: no human remains found Urn: none Burial rite: not known Grave goods: none; Dating evidence: none
32 © Cotswold Archaeology South Wales Gas Pipeline Project Site 513: Archaeological Excavation
APPENDIX C: THE FINDS
Prehistoric pottery by Alex Gibson
Weight (kg) No of Contexts Periods Represented 9.2 plus complete 16 + Unstratified Late Neolithic, Bronze Age vessels
Grooved Ware
P1 – 268050 RD: c.380mm (though the sherd may have suffered post-depositional flattening), 55g. Rim sherd and body sherd in a soft, grog-filled fabric with grey surfaces and fabric. The body sherd is buff. The inner surface in particular is severely pitted, but the outer surface also has large, roughly rectangular pits towards the top of the rim sherd. The fabric averages 7mm thick. The rim is sharply rounded with an internal bevel. The vessel has been open in profile. The rim bevel is decorated with a raised cordon accentuated by an upper and lower incised line. The cordon is impressed with regular fingernail impressions forming the ‘rungs’ of a ladder motif. Externally, the decoration comprises three irregular horizontal incised lines below the rim. Below this are 3 further diagonal incised lines and, after an undecorated zone, a 4th parallel incised line. Two well-spaced irregular lines are visible on the body sherd.
P2 – 268050, 268051 167g. Twenty three body sherds in a grog-filled fabric.
They have grey surfaces and a black fabric core. The inner surface is particularly pitted, similar to GW1 above. The fabric averages 9mm thick. The decoration comprises bold incised lines. Multiple parallel horizontal and diagonal lines divide the outer surface into panels. On the larger sherd, the panel is subdivided into two triangular area, 1 is subdivided with incised filled chevrons, the other half is undecorated. A similar motif may be present on another of the body sherds. A number of bold parallel incised lines, some clearly arranged in a chevron motif,
33 © Cotswold Archaeology South Wales Gas Pipeline Project Site 513: Archaeological Excavation
the inner surface and boldness and broadness of the incision suggests that these sherds are from the same vessel. Sherd 268051 has been burnt.
P3 – 268020 BD: c.160mm, 166g. Thirteen body sherds in a hard, well-fired and well-finished grog-filled fabric.
The outer surface is dark grey to light brown. And the inner surface is generally dark grey. The fabric is black and averages 10mm thick. The decoration is comprised fine, incised lines arranged somewhat haphazardly into chevron, triangular and lozenge-shaped panels. The lozenge and triangular panels are filled with oval impressions. One sherd has a large perforation, 10mm in diameter at the outer surface, 6mm at the inner surface. This has been drilled after the vessel had been fired and may well be a repair hole. One sherd from the base angle is also decorated with small randomly-spaced circular impressions.
P4 – 268020 Nineteen small undecorated sherds (44g) that may belong to any of the above vessels.
Discussion Whilst the identification of these sherds as Grooved Ware is not in doubt, their fragmentary nature makes it difficult to attribute them to any particular style. The use of incised, grooved or scored oblique lines on P1-P3 suggests a tub-shaped vessel profile and if that is the case they probably belong to Longworth’s Clacton or Durrington Walls style (Wainwright & Longworth 1971). The former is suggested by the openness of the broadly triangular panels on P1 and P2). The panels filled with oblique lines on P2, however, may be better paralleled in the Durrington Walls style though none of the sherds have the plastic decoration associated with this style. The dot-filled panels of P3 are more common in the Clacton style. Diagnostic rim forms are also absent from the
34 © Cotswold Archaeology South Wales Gas Pipeline Project Site 513: Archaeological Excavation
assemblage though the Durrington Walls style lends itself to less complicated rim forms than those found in the Clacton style.
Grooved Ware is rare in Wales with the two largest assemblages being from Trelystan, Powys (Britnell 1982) and Upper Ninepence, Powys (Gibson 1999). At the former site both incised and scored oblque lines are common as are chevron motifs and filled panels. At Upper Ninepence, diagonal lines and chevron panels are also represented as is the combination of incision and circular stabs (Gibson 1999, fig 56, P63) and panels filled with close-set incised diagonal lines similar to P2. Closer to home, a small Grooved Ware presence has been reported from Stackpole Warren, Dyfed (Benson et al. 1990) one sherd decorated with jab marks close to a groove and another with grooved rim and raised cordon both of which can be paralleld at Upper Ninepence if not in the present assemblage.
The rarity of Grooved Ware in Wales is difficult to explain given that it is spread widely over Britain and has been increasingly recognised in Ireland. Indeed, some thin-walled vessels from Upper Ninepence compare closely with some vessels from Newgrange (Brindley 1999) where the sweeping curved triangular panels of P9 can also be matched.
The perforation in P3 has been drilled after the vessel had been fired and not through the unfired clay. Drilling a hole in an already fired and brittle medium such as ceramic must have been an afterthought as it would have been far easier to have perforated the vessel whilst the clay was still plastic had the perforation been part of the original design. Consequently they have been interpreted as repair holes (Cleal 1988). It is envisaged that originally there would have been a pair of holes one either side of a crack in the vessel wall and that they would have been used to thread through a binding material to hold the crack together. It is assumed that the drilled hole in P3 is a similar device but that its partner has been lost.
Given the paucity of Grooved Ware sites in Wales, not surprisingly, evidence for the dating of the tradition within the Principality is also poor. The Grooved Ware contexts from Upper Ninepence spanned the first half of the 3rd Millennium (c.2900-2500 BC) (Gibson 1999, 44). A similar date range was established by Garwood (1999) for the Clacton and Durrington Walls styles in southern England though perhaps extending as late as Needham’s Fission Horizon c.2200 BC (Needham 2005). In the North of England, though again based on few dates and not always from totally reliable samples, Marshall would place Grooved Ware between 3100 BC and 2245 BC (Millson et al 2011). The present assemblage must also date to this Late Neolithic period.
35 © Cotswold Archaeology South Wales Gas Pipeline Project Site 513: Archaeological Excavation
Food Vessel P5 – 268159 (Lower) Bipartite Vase Food Vessel – H: 195mm, RD: 170mm, BD: 110mm, T: 10mm.
Hard well fired fabric, with a brown external surface and a grey-brown internal surface. The fabric is dark grey throughout. The fabric contains angular crushed stone inclusions up to 7mm across. Many of these break the outer surface giving it a rough, gritty texture. The fabric is also pitted, especially visible on the outer surface. Traces of coil breaks (join voids) van be seen in the breaks especially just below the rim and at the shoulder. The rim itself is well-formed and slightly flattened with finishing/smoothing facets visible. The rim has neither external moulding nor internal bevel but rather is vertical internally for a depth of some 40mm and this area is decorated with two rows of well-spaced, circular reed impressions. The external decoration comprises a row of overlapping horizontal/slightly oblique fingernail impressions (pseudo-cord) immediately blow the rim. These impressions are carelessly executed. The neck has two encircling lines of large reed impressions a single row of which also decorates the shoulder. The top and bottom of the shoulder are defined by two encircling lines of pseudo-cord impressions identical to that below the rim. The belly and base are undecorated. The internal base appears worn and abraded perhaps suggesting long term use. The vessel is incomplete with some 50% of the rim missing.
36 © Cotswold Archaeology South Wales Gas Pipeline Project Site 513: Archaeological Excavation
Discussion Fingernail impressions, especially arranged in a pseudo-cord line, are rare on Food Vessels though multiple horizontal rows of fingernail impressions have been found on a Food Vessel fragment from Pentraeth, Anglesey (Savory 1957). Pseudo-cord can, of course, resemble incision and the combination of circular impressions and incision can be seen on a bipartite vase Food Vessel from Whitford, Clwyd, (Savory 1980, Cat No.383) and a globular ridged bowl from Narberth South, Dyfed (ibid Cat No.384.1). On both these vessels the incision is arranged in herringbone motif rather than horizontal lines. A fragmentary ridged vessel combining circular impressions and horizontal incised lines was found with a Collared Urn at Bedd Branwen in Anglesey (Lynch 1991, fig 45). Though the pot group had been badly damaged, enough survived of the contents to identify the peculiar north Welsh rite of infant ear bones burial. There are 43 Food Vessel associated dates from Wales (Burrow & Williams 2008) dating the tradition to 2200-1500 cal BC.
P6 – 268176 Tripartite Vase Food Vessel – H: 187, RD: 173mm, BD: 85mm, T: 7mm.
37 © Cotswold Archaeology South Wales Gas Pipeline Project Site 513: Archaeological Excavation
Hard well-fired fabric, with a brown external surface and a brown to grey-brown internal surface. The fabric is brown externally and black internally, the colour changing mid-way through the thickness of the wall. The fabric contains abundant large angular crushed stone inclusions up to 5mm across. Both surfaces are smooth and well- finished with horizontal smoothing lines on the outer surface of the neck. The slightly ribbed nature of the internal profile attests to coil construction. The rim is well-formed and rounded. The rim interior is slightly concave and reaches 33mm deep. The external rim moulding is almost vertical and narrow, measuring only 25mm deep. The neck is well defined and sharply distinguished from the rim. It measures up to 50mm deep ending at a sharp carination. The belly is slightly concave. The rim bevel is decorated with 4 encircling lines of well-defined twisted cord impressions. The overlaps between the end and start of a line can be seen in some places. The same technique decorates the rim moulding. The majority of the neck is undecorated though three rows of short diagonal twisted cord impressions arranged in herringbone motif are located just above the shoulder. The belly of the pot is undecorated. The pot is virtually complete and well-preserved though it has suffered some distortion due to taphonomic processes. There is a dark carbonaceous band on the internal surface some 45mm wide and 55mm below the base of the bevel.
Discussion Multiple twisted cord line decoration seems more common on Collared Urns in Wales than on Food Vessels though its occurrence on some Food Vessel Urns clearly points at its Food Vessel association. The vertical internal rim profile of this pot is also a formal trait that is found on both Collared Urns and Food Vessel Urns and is unusual for smaller Food Vessels that usually have an internally bevelled rim. The cultural and stylistic overlap between Food vessels, Food Vessel Urn and Collared Urn, however, have been well documented elsewhere (Law 2008).
The lack of exact parallels should not surprise us given that few prehistoric pots seem to have exact matches. Rather they seem to have an individuality, albeit set within a rigorous formal and decorative tradition. Perhaps the closest parallel for the present Food Vessel is a tripartite vase from Llanddyfnan, Anglesey (Savory 1980, Cat. No. 391). This has multiple cord lines on the rather steep internal bevel and the concave neck. The outer rim moulding and the shoulder are, however, decorated with oblique twisted cord maggots recalling the herringbone motif accentuating the shoulder of the present vessel. A Collared Urn from Rhiw-with-Llanfaelrhys, Gwynedd, found with a bronze awl, burnt dagger pommel and human bones in a cist (Savory 1980, Cat. No. 341.1) has a steep internal rim profile and both this and the collar are decorated with horizontal lines of twisted cord impressions. Vertical impressions in the same technique decorate the neck and the shoulder is accentuated by oval impressions. This again demonstrates the close relationship between these ceramic types. There are 43 Food Vessel associated dates from Wales (Burrow & Williams 2008) dating the tradition to 2200– 1500 cal BC.
38 © Cotswold Archaeology South Wales Gas Pipeline Project Site 513: Archaeological Excavation
Collared Urn P7 – 268159 (Upper) Tripartite Collared Urn – H: 215mm, RD: 225mm, BD: 100mm, T: 10mm.
Hard well fired fabric, with a brown external surface and a brown to grey-brown internal surface. The fabric is dark grey throughout. The fabric contains angular crushed stone inclusions up to 7mm across. Many of these break the outer surface, particularly the lower part of the body, giving it a rough, gritty texture. The fabric is also slightly pitted, especially visible on the outer surface. The rim is simple and rounded with a sloping slightly concave interior 40mm deep. This is smooth and well-finished. The collar is shallow, only some 25mm deep and the neck measures 50mm from the base of the collar to the shoulder. The decoration comprises four encircling lines of well-defined twisted cord impressions on the internal rim moulding. The collar is decorated with short oblique lengths of the same technique sloping from top right to bottom left. The neck is undecorated save for a similar row of short diagonal impressions, mirroring those on the collar, just above the shoulder. The belly of the
39 © Cotswold Archaeology South Wales Gas Pipeline Project Site 513: Archaeological Excavation
vessel is undecorated. Internally there are possible traces of carbonaceous residues occupying a zone some 60mm deep immediately below the shoulder. The vessel is more or less complete with slight and ancient damage to the rim. It has warped slightly due to taphonomic processes and the base has become detached, possibly, judging from the cleanness of the break, during transit. This has been repaired using HMG acetone-soluble adhesive.
Discussion The vessel belongs to Burgess’ (1986) early form though this stylistic dating is as yet unsupported by an adequate absolute chronology. An urn with similar profile, though proportionately slightly taller, comes from Rhiw- with-Llanfaelrhys, Gwynedd (Savory 1980, Cat No. 341.1). The rim profile of this vessel is very similar to the urn under discussion and is similarly decorated externally. The collar decoration on the Gwynedd urn, however, comprises horizontal rather than diagonal lines. The simple diagonal motif on the collar can be compared with that on the collar of an urn from Sutton ‘268’, Glamorgan (Savory 1980, Cat. No.371.18) though this urn is otherwise undecorated. Such simple decoration on the collar is actually quite rare nationally though geographically widespread. Longworth (1984, plates 144-146) illustrates eight vessels with similar simple, unbordered, diagonal twisted cord lines. The range from Prestonkirk, Lothian to a vessel probably from Gloucestershire and another, smaller urn, from Therfield, Hertfordshire. The internal rim forms of all these pots, however, differ from P7 in that they tend towards internally bevelled rather than moulded rims. A vessel from Applegarth and Sibbaldbie, Dumfriess and Galloway does have an internal rim moulding decorated, like P9, with multiple horizontal lines and on the collar with simple diagonal lines, however the collar diagonals are bordered above and below with single horizontal lines.
40 © Cotswold Archaeology South Wales Gas Pipeline Project Site 513: Archaeological Excavation
P8 – 268173 Tripartite Collared Urn – H: 180mm, RD: 185mm, BD: 85mm, T: 10mm.
Hard but friable fabric, with a brown external surface and a grey-brown to dark grey internal surface. The fabric is brown throughout. The fabric contains angular crushed stone inclusions up to 7mm across. These do not break either surface to the degree noted in the previous vessels. Instead the surfaces are smooth and well-finished. The fabric is also slightly pitted, especially visible on the outer surface. The rim is thin, simple and rounded with a deep, slightly concave interior some 30mm deep. The collar is narrow, only 25mm and the neck measures 50mm from the base of the collar to the shoulder. The decoration comprises a bold zone of incised herring bone on the rim interior, executed with a sharp point. The collar is very abraded but is similarly decorated. The neck is also decorated with incised herring bone. The uppermost element generally slopes from top left to bottom right but this direction changes at one point. This change is not, however, reflected in the second element at this point. The first and second elements and the second and third elements are each separated by a single encircling incised line forming interrupted herringbone motif. The third and fourth (final) elements are somewhat ‘squashed into the remaining lower part of the neck above the shoulder to the degree that they often overlap and affectively form a
41 © Cotswold Archaeology South Wales Gas Pipeline Project Site 513: Archaeological Excavation
zone of cross-hatching. There is no decoration below the shoulder. As mentioned, the rim is very abraded, almost certainly as the result of taphonomic processes, and parts have become detached. Judging by the amount of loose material in the box, this has also suffered damage in transit.
Discussion The vessel belongs to Burgess’ (1986) early form though this stylistic dating is as yet unsupported by an adequate absolute chronology. Herringbone and interrupted herringbone motifs are common on Collared Urns in a variety of decorative techniques and especially in twisted and whipped cord. The urn finds close parallels with other Collared Urns in Wales with incised herringbone on the collar and neck from Sheeplays, Glamorgan (Savory 1980, Cat No. 440.1), and Kerry, Powys (ibid 428). Internal and external incised herringbone can be matched at Trefeglwys, Powys (ibid 426) and Coity, Glamorgan (ibid 433). The narrow collar of the present vessel and the ‘squashed’ herringbone in the neck can be found on a vessel from Lower Lledrod, Dyfed (ibid 430) though on this vessel the internal herringbone is in twisted cord technique rather than incised. The interrupted herringbone on the present vessel can be paralleled on the collar of an urn from Llanbeblig, Dyfed (ibid 412) though in this case the incised horizontal interrupting lines divide whipped cord diagonals.
A vessel with finely executed interrupted incised herringbone on both the collar and neck is attributed to North Yorkshire by Longworth (1984, pl 72a) but this is altogether finer than the decoration encountered in P8. A vessel from N Ireland (ibid pl 72c) matches the bold internal herring bone of P8 though P8 only has a single row and the Irish vessel has two. Two tripartite urns from Llanbabo, Gwynedd combine bold incised herring bone both internally and externally (Longworth 1984, Pl 22 b&c) and these vessels also have stop-grooved shoulders: they are so similar that they could be by the same potter. In the case of the Llanbabo urns, however, the herringbone is not interrupted. A tripartite Collared urn from Clynnog, Gwynedd, however, has a similar rim form to P8 but carrying two zones of incised herringbone and on the collar and neck, the same motif is interrupted (Longworth 1984, pl.9c). Once again the variables in these comparanda display the resilience of the potters whilst still working within the confines of the tradition.
P9 – 268127 Bipartite Collared Urn – RD: c.230mm, BD: 120mm, T: 14mm. Hard well-fired fabric, with a brown external surface and a grey-brown to dark grey internal surface. The fabric is brown throughout. The fabric contains abundant large angular crushed stone inclusions up to 10mm across. These break the abraded outer surface. The internal surface is smooth and well-finished. The fabric is also slightly pitted, especially visible on the outer surface. The rim is rounded and well-formed with slight, internal rounding, suggestive of a bevel but only some 10mm deep. The collar is well-defined and some 80mm deep. This gives to a rounded body without any hollow neck. The base of the collar has been applied to a vessel with an otherwise rounded, slightly closed, profile. The scars of the join voids in this area are clearly visible. The vessel is fragmentary with only the upper portion reconstructable. It is completely undecorated. Two rounded pits, each 5mm in diameter, on the inside of one rim fragment appear to be deliberate and to have been made while the clay was still wet. Their purpose is unknown.
Of the 47 sherds (and 2 bags of small crumb) boxed separately are 3 base sherds representing just under half of the base and allowing a basal diameter of 120mm to be estimated. The other sherds are undecorated and have abraded surfaces. Traces of carbonaceous residues are visible on the inner surface some 120mm below the rim. Also some small crumbs from this context, bagged separately but may be from the same vessel.
42 © Cotswold Archaeology South Wales Gas Pipeline Project Site 513: Archaeological Excavation
Discussion Bipartite Collared Urns with deep collars, smooth internal profile and applied collar bases are stylistically late in Burgess’s (1986) scheme. Bipartite Collared Urns are rare in Wales. Savory illustrates an example from Ysceifiog, Clwyd (1980, Cat. No. 423.2) but this has a twisted cord-decorated bevel and collar, though it has to be said that the decorative scheme is less than ambitious. There are, however a number of undecorated Collared Urns from Wales though these are largely tripartite examples such as the pair of undecorated Collared urns from Letterston, Dyfed (ibid Cat. Nos 316.3, 316.4). A bipartite form (ibid Cat No. 438.2) was associated with a secondary cremation below a small cairn at Llandyssul, Dyfed.
Longworth illustrates around two dozen undecorated bipartite vessels of which over half are miniature vessels from as far afield as Cleveland and Berkshire (Longworth 1984, pl 160-162). Of the larger vessels, the Llandyssul urn mentioned above is illustrated as are vessels from Brighton, East Sussex, Great Wilbraham, Cambridgeshire and Goodmanham, Humberside. Two associated undecorated bipartite urns from Fylingdales, N. Yorks, are associated with ovoid tubular jet beads (ibid Pl 161, i & j). A rather asymmetrical vessel with pinched-out, almost cordoned, collar base from Farmington, Gloucestershire, was associated with a perforated piece of bronze (ibid, pl 160 f).
43 © Cotswold Archaeology South Wales Gas Pipeline Project Site 513: Archaeological Excavation
P10 – 268115 Tripartite Collared Urn – H: 210mm, RD: 162mm, BD: 10mm, T: 8mm.
Hard well-fired fabric, with a brown external surface and a grey-brown internal surface. The fabric is brown externally and black internally, the colour changing mid-way through the thickness of the wall. The fabric contains abundant large angular crushed stone inclusions up to 7mm across. These break the abraded outer surface. The internal surface is smooth and well-finished. The rim is simple, rounded and inturned. The collar is convex externally and only 35mm deep. The neck is very poorly defined with a sinuous rather than acutely tripartite profile. The neck is approximately 45mm deep and the belly convexly rounded. The vessel is undecorated. The vessel is virtually complete with some rim damage and a recent (fresh) break in the mid to lower belly. Internally there is a zone of preserved carbonaceous residues, some 45mm deep and some 60mm below the rim.
44 © Cotswold Archaeology South Wales Gas Pipeline Project Site 513: Archaeological Excavation
Discussion This would also appear to be late in Burges’s scheme (Burgess 1986) given its narrow base and smooth internal profile though the shallow collar is an early trait. Like P9, the Urn is undecorated and, as noted above, there are a number of undecorated Collared Urns from Wales largely of tripartite form such as the pair of undecorated Collared urns from Letterston, Dyfed (Savory 1980, Cat. Nos 316.3, 316.4). This said, the shoulder of P10 is very slack and rounded not dissimilar to an undecorated vessel from Welsh St Donats, Glamorgan, (ibid Cat No. 342.1) found with cremated human remains, a bone ring-headed pin and flint knife. Like P10, this also has a narrow collar though it is not as curved as the present vessel. This is also the case with a small tripartite undecorated urn from Bedd Branwen, Anglesey (Lynch 1991) and from cremation 7 at Brenig 45 (Lynch 1993). The date of 3620±100 BP (HAR-1027) for the Brenig urn must be regarded as of poor integrity given that it is from charred soil.
Undecorated tripartite vessels are more common than bipartite ones. Longworth (1984, pl 42 e) illustrates a vessel with rounded, inturned collar and rounded shoulder, very similar to p10 from Christchurch in Dorset, another two from Bournemouth, Dorset (pl 75 a & g) and another from Ogbourne St Andrew 7 in Wilts (ibid pl 43 c). Further north, a vessel from Scunthorpe, Humberside (ibid pl 43 b) and another from Kettering, Northamptonshire (pl 75 f) have similar shapes and sizes but slightly more angled shoulders.
45 © Cotswold Archaeology South Wales Gas Pipeline Project Site 513: Archaeological Excavation
P11 – 268146 Tripartite Collared Urn – RD: 185mm, BD: 10mm, T: 10mm.
Hard well-fired fabric, with a brown external surface and a grey-brown internal surface. The fabric is brown externally and black internally, the colour changing mid-way through the thickness of the wall. The fabric contains abundant large angular crushed stone inclusions up to 5mm across. Both surfaces are smooth and well-finished with vertical smoothing lines on the outer surface. The rim is well-formed and slightly flattened. The interior of the collar is concave and reaches 50mm deep. The collar is almost vertical and measures 34mm deep. The neck is well defined and sharply distinguished from the rim. It measures up to 70mm deep ending at a sharp carination. The belly is only slightly concave. The vessel is undecorated. The upper portion of the pot is virtually complete and well-preserved. The base is missing. An irregular scored line around the base fracture may be excavation
46 © Cotswold Archaeology South Wales Gas Pipeline Project Site 513: Archaeological Excavation
damage. There is a dark carbonaceous band on the internal surface some 45mm wide and 55mm below the base of the bevel.
Discussion The well-defined collar and neck and the concave nature of the collar interior would date this early in Burgess’s (1986) scheme. Like P9 & 10, the urn is undecorated but its sharp tripartite profile can be compared to the vessels from Letterston mentioned above and a rather more squat vessel from Llanboidy, Dyfed (Savory 1980, Cat. No.417.2). The upright form of the pot and well-defined shoulder resembles the Scunthorpe vessel mentioned above (Longworth 1984, Pl 43 b) and to a lesser extent, that from Christchurch (ibid 42e). A Collared Urn from Macclesfield, Cheshire (ibid pl 45b) also shares the comparatively narrow well-defined collar, deep neck and well-defined shoulder of P11. A further vessel from Cocksburnpath, Borders has a slightly concave collar, but nevertheless is similar in form, with deep neck and straight-sided body (ibid, pl 75b). The rim diameter of this vessel is slightly greater than P11.
P12 – 268007– soil block Tripartite Miniature Collared Urn – H: 122, RD: 165mm, BD: mm, T: 9mm.
47 © Cotswold Archaeology South Wales Gas Pipeline Project Site 513: Archaeological Excavation
Hard well-fired fabric, with light brown surfaces. The fabric is brown externally and black internally, the colour changing mid-way through the thickness of the wall. The fabric contains abundant large angular crushed stone inclusions up to 5mm across some of which occasionally break the outer surface. Both surfaces are comparatively rough and poorly finished though this may be due to taphonomic processes. Coil breaks are visible especially at the shoulder. The rim is well-formed and rounded. The interior is very slightly concave, almost vertical and reaches 33mm deep. The collar is shallow and not well-defined. It slopes slightly outwards from the rim and measures only 23mm deep. The neck is shallow but well defined and measures up to 25mm deep. The belly is slightly concave. The vessel is undecorated. The pot is well-preserved though some 50% of the base and lower body are missing. Some fresh and smooth breaks in this area indicate that at least some of this damage is due to excavation.
Discussion This small undecorated tripartite urn measures only 122mm high yet has a perfect Collared Urn form. Though it has been recognised for some time that there are miniature Collared Urns (Longworth 1984) scatter diagrams of vessel height against rim diameter show a continuous series from urns some 75mm to over 500mm high. This is different to the situation with Food Vessels where there appears to be a break at 200mm between Food Vessels proper and the Enlarged or Food Vessel Urns at least in northern England (Gibson 1978). Savory also illustrates small Collared urns from Wales such as the tripartite Collared urn from stone circle 278 on Penmaenmawr (1980, Cat No. 493) at only 143mm high and a vessel of almost Food Vessel form from Holt, Clwyd at 146mm tall (ibid. 416.2). The small undecorated urn from cremation 7 at Brenig 45 (Lynch 1993) has already been mentioned (P10 above) and the date of 3620±100 BP (HAR-1027) warned against given that it is from charred soil. At 140mm high, vessel K from Bedd Branwen shares the same narrow collar and shallow neck of P12 but in this vessel the body is slightly taller and the whole vessel has a less squat appearance (Lynch 1991, fig 43). At only 100mm tall, the small undecorated tripartite Collared Urn from Cae Mickney is even smaller than P12 and, when taken with the other larger undecorated urns above, serves to demonstrate the size-range even within the Principality (Lynch 1991, fig 55). These diminutive vessels may well have served the function of small accessory cups but as mentioned above, there seems to be no logical cut-off point between miniature and large vessels. Indeed Longworth (1984, pl 246) illustrates both decorated and undecorated Collared Urns under 100mm tall.
48 © Cotswold Archaeology South Wales Gas Pipeline Project Site 513: Archaeological Excavation
P13 – 268009 – soil block Tripartite Collared Urn – H: 225, RD: 105mm, BD: 105mm, T: 9mm.
Hard well-fired fabric, with light brown surfaces. The fabric is brown externally and black internally, the colour changing mid-way through the thickness of the wall. The fabric contains abundant large angular crushed stone inclusions up to 10mm across some of which occasionally break the outer surface where it is abraded but which break the inner surface with greater frequency, especially towards the base. The outer surface is comparatively smooth, especially the collar and neck and vertical smoothing marks may be seen. Traces of join voids from coil
49 © Cotswold Archaeology South Wales Gas Pipeline Project Site 513: Archaeological Excavation
construction are visible in the breaks. The rim is rounded and well-formed and the collar is slightly concave internally reaching some 55mm deep. Externally the collar is generally straight (though very slightly concave in places) and measures 55mm deep. The neck is well defined measuring 50mm from the base of the collar to the shoulder. The lower body of the urn is slightly concave. Decoration comprises a line of very light horizontal, evenly spaced, impressions at the top and base of the collar. A similar row of impressions accentuates the shoulder. The urn is approximately two thirds complete. There is a possible circular seed impression on the collar.
Discussion This early Collared Urn (Burgess 1986) at first appears undecorated but then faint elongated oval impressions are visible at the top and bottom of the collar and on the shoulder. This urn appears to be unique in Wales and indeed in Britain and Ireland generally (Longworth 1984). Tripartite vessels from Holt, and Rhiw-with-Llanfaelrhys have deeper and more well-defined oval impressions on the shoulder but are otherwise more elaborately decorated (Savory 1980, Cat Nos 416.1; 341.1). These stop-grooved shoulders are more common in early Collared Urns (Burgess 1986; Longworth 1984, pls 22-24) but these impressions are always considerably deeper than those on P13 and are combined with other decorative techniques and motifs. The appearance of such impressions on the collar of P13 as well as the shoulder, may suggest that the resemblance to shoulder stop- grooves is more apparent than real.
50 © Cotswold Archaeology South Wales Gas Pipeline Project Site 513: Archaeological Excavation
P14 – 268144 Tripartite Collared Urn – BD: 90mm, T: 10mm.
Hard well-fired fabric, with light brown external surface and a dark grey internal surface. The fabric is brown externally and black internally, the colour changing mid-way through the thickness of the wall. The fabric contains abundant large angular crushed stone inclusions up to 10mm across some of which occasionally break the outer surface where it is abraded near the base. The outer surface is comparatively smooth and vertical smoothing marks may be seen. Traces of join voids from coil construction are visible in the breaks. The rim and most of the collar and neck are absent. Traces of a concave neck and shoulder are visible but only the lower portion of the pot survives. The sherds have splayed out to give a diameter much greater than it would originally have had as a result of taphonomic distortion.
Eighteen sherds (plus small crumb) bagged separately may come from the collar of the same vessel though no rim sherds are present. These are decorated with incised crosshatching. Decoration is generally absent from the main body of the vessel but two converging oblique lines on the one neck sherd that survives suggest that the neck had originally carried some kind of incised chevron- or crosshatch based motif similar to that on the collar.
Discussion Little can be said of this vessel given its poor survival however from the reconstructed portion and the collared sherds present, it seems to have been tripartite and decorated both on the collar and in the neck with incised
51 © Cotswold Archaeology South Wales Gas Pipeline Project Site 513: Archaeological Excavation
cross-hatching. This can be matched on a tripartite vessel from Letterston (Savory 1980 Cat No. 316.1) which also has nicks on the top of the rim. Incised lattice in the neck is a common motif nationally (Longworth 1984) and is more common in Longworth’s Primary Series urns but not restricted to them.
P15 – 268001 Tripartite Collared Urn – RD: 115mm, T: 7mm
Hard well-fired fabric, with light brown external surface and a slightly more grey internal surface. The fabric is brown externally and black internally, the colour changing mid-way through the thickness of the wall. The fabric contains abundant large angular crushed stone inclusions up to 5mm across some of which occasionally break the outer surface where it is abraded. The rim is simple and rounded and the collar is slightly concave internally reaching a depth of 35mm below the rim. The collar is quite narrow (25mm deep). The neck is marked by a slight concavity, some 40mm deep giving to a rounded shoulder. The base sherd is too abraded to allow an estimation of diameter. The decoration comprises three encircling lines of abraded twisted cord impressions on the internal rim moulding. The collar exterior is very abraded but traces of twisted cord herring bone motif survive on 1 sherd. Two diagonal twisted cord lines sloping from top left to bottom right survive in the neck of the same sherd suggesting the same or possibly even a lattice or chevron-based motif. The vessel is fragmentary comprising 8 rim sherds, a shoulder sherd and a possible base sherd. Six of the rim sherds conjoin. In addition there are 26 (70g) small body fragments.
Discussion This vessel is early in Burgess’s scheme with its narrow collar and concave decorated internal rim moulding (Burgess 1986). The internal decoration is similar to P7 above and has already been discussed. Normally twisted cord herringbone motif is bordered by horizontal lines but this does not seem to be the case in P15 and this can be matched in primary Series vessels from Stanton Moor, and Castleton, Derbyshire, as well as a vessel, possibly from Northamptonshire (Longworth 1984, pl 81). The Stanton Moor urn also has twisted cord lattice in
52 © Cotswold Archaeology South Wales Gas Pipeline Project Site 513: Archaeological Excavation
the neck, as may be a possibility for P15, but lacks the internal cord lines. An urn from Evercreech, Somerset, has the internal cording and external un-bordered herringbone motif but in this case the herring bone is interrupted (Longworth 1984, Pl 25a). Vessels from Nantglyn, Clwyd and Northamptonshire combine the encircling internal lines with unbordered herringbone on the collar, but in this case the external decoration is whipped rather than twisted cord (ibid pl 5a & d).
P16 – 268111 Tripartite Collared Urn – RD: 160mm, T: 8mm.
Hard well-fired fabric, with a brown external surface and a dark grey internal surface. The fabric is brown externally and black internally, the colour changing mid-way through the thickness of the wall. The fabric contains abundant large angular crushed stone inclusions up to 7mm across some of which occasionally break but lie flush with the inner surface. Join voids representing coil construction are visible in a number of sherds. The rim is simple and rounded and the collar is slightly concave internally reaching a depth of 40mm below the rim. The collar is quite narrow (30mm deep). The neck is marked by a slight concavity, some 35mm deep giving to a rounded shoulder. The decoration comprises twisted cord impressions. The collar is decorated with three lines of oblique impressions arranged in a haphazard and overlapping herringbone motif. A single zone of rather more carefully executed twisted cord herring bone fills the neck bordered at the top and bottom and interrupted by single encircling lines. The urn is fragmentary and has been poorly reconstructed. None of the 27 body sherds (155g) seem to be from lower than the shoulder.
Discussion This urn is also early in Burgess’s (1986) scheme and equates with Longworth’s (1984) Primary Series. As with P14, twisted cord herringbone motif has already been noted elsewhere in Wales but in this case the motif is bordered and interrupted in the neck. This motif is found on the collar of an urn from Llanbeblig (Savory 1980,
53 © Cotswold Archaeology South Wales Gas Pipeline Project Site 513: Archaeological Excavation
Cat No. 412) and in a rather more elaborate way, where multiple horizontal lines are used at Talbenny (ibid Cat No. 372.3). This urn would seem to be close in form to P16.
P17 – 182.4 268113 Collared Urn? – T: 8mm. Five sherds (20g) in a hard well-fired fabric, with light brown outer surface and light grey to black inner surface. The fabric is brown externally and black internally, the colour changing mid-way through the thickness of the wall. The fabric contains abundant large angular crushed stone inclusions up to 6mm across. The surfaces are smooth. One concave neck sherd is decorated with lightly incised lattice. This is similar to the surviving decorated neck sherd on P14 above and may be from the same vessel. The fabric is very similar to P18 below though P18 is generally considerably thicker.
Discussion See P14 above.
P18 – 182.4 Unstratified 01 Tripartite Collared Urn – RD: ?240mm, BD: 140mm, T: 13mm. Nine rim sherds and twenty seven collar sherds from a large Collared Urn. The fabric is hard and well-fired with generally smooth surfaces. The outer surface is brown with some black patches at the rim. The inner surface is black and the fabric changes from brown to black midway through the thickness of the wall. The fabric contains well-crushed stone inclusions most of which do not break the vessel surfaces. The rim is slightly flattened on top, smooth and well-finished without an internal bevel. The collar is slightly concave internally measuring some 105mm deep. Externally the collar has been approximately 95m deep and decorated with well-defined twisted cord impressions. These comprise two encircling lines at just below the rim and one or two encircling lines at the base of the collar. The space between is decorated with fairly carelessly executed opposed filled triangles in the same technique. The inside of the rim and the neck are undecorated. The neck is approximately 60mm deep from the base of the collar to slightly rounded shoulder. Base sherds are fragmentary. The collar, neck and larger boy sherds weigh 7.7kg. In addition there are some 27 (516g) of small undecorated body sherds and a quantity of small crumbs almost certainly from the same vessel.
Discussion The concave internal on this vessel place it early in Burgess’s scheme (1986) and in Longworth’s Primary Series (1984) however the bold decoration the deep collar and the lack of decoration below the collar are indicative of Burgess’s late urns and Longworth’s Secondary series and it must be with these later vessels that P18 finds more affinity. It has obviously been a large, heavy vessel, and the bordered opposed filled triangle motif on the collar can be matched on similarly large vessels from Tredunnock, Kilpaison and, on a smaller vessel, Llanfihangel-y-Creuddyn (Savory 1980, Cat. Nos 339.1, 419.2, 419.3 & 427). The Tredunnock vessel contained a cremation and bone ring-headed pin, and Kilpaison 419.2 contained the cremated remains of an 11-12 year old boy as well as a second, smaller tripartite Collared Urn itself decorated with bordered twisted cord herring bone motif on the collar. The Llanfihangel-y-Creuddyn vessel was reputedly associated with a bronze pin. Large vessels from Whitton, Norfolk, and from Rutland (Longworth 1984, pl 209c & 211 b) match P18 but these have decorated rim bevels. Nevertheless the heavy and slightly erratic nature of the twisted cord opposed filled chevrons is very similar, including the bordering lines, and both vessels are undecorated below the shoulder. The Whitton vessel was associated with a copper alloy awl.
54 © Cotswold Archaeology South Wales Gas Pipeline Project Site 513: Archaeological Excavation
P19 – 182.4 Unstratified 02 Tripartite Collared Urn – T: 9mm. Thirteen sherds (344g) in a hard well-fired fabric, with light brown outer surface and light grey to black inner surface. The fabric is brown externally and black internally, the colour changing mid- way through the thickness of the wall. The fabric contains abundant large angular crushed stone inclusions up to 10mm across some of which occasionally break the outer surface where it is abraded but which break the inner surface with greater frequency. The outer surface, where not abraded, is comparatively smooth. One sherd exhibits the rounded shoulder and lower neck of a tripartite urn. The fabric is very similar to P13 above and these sherds may well belong to the missing portion of this vessel. Six sherds have internal carbonaceous encrustations.
Discussion See P13 above.
P20 – 182.4 Unstratified 03 Collared Urn? – T: 8mm. Two sherds (23g) in a hard well-fired fabric, with light brown outer surface and light grey to black inner surface. The fabric is brown externally and black internally, the colour changing mid-way through the thickness of the wall. The fabric contains abundant large angular crushed stone inclusions up to 6mm across some of which occasionally break the inner surface. The outer surface is smooth. One sherd exhibits converging twisted cord lines suggesting a filled triangular motif. The fabric is very similar to P18 above and these sherds may well belong to this vessel though the fabric of P18 is generally considerably thicker.
Discussion See P18 above
Miscellaneous Ceramics 268145: Crumbs (2g) of abraded material. Probably Bronze Age 268158: Twelve sherds plus crumbs (27g) of abraded material. Probably Bronze Age 268073: Undiagnostic crumbs. Other Materials 268050: Single sherd (11g) of daub or fired clay.
Overall Discussion The rich ceramic assemblage from this site appears to represent two distinct phases of activity represented firstly by Grooved Ware and secondly by Food Vessels and Collared Urns. There is no evidence for Beaker activity that might link the earlier and later phases and thus suggest continuity. The dating of Grooved Ware has been discussed above (p 63-64) and a date range of c.3000–2300 cal BC in rounded terms might be suggested. Of the 79 radiocarbon dates associated with Food Vessels and Collared Urns cited by Burrow and Williams (2008) the few early outliers are charcoal derived dates. Once again the period of currency would appear to be between c. 2200–1500 cal BC. It is therefore very unlikely that the Grooved Ware was in use on this site at the same time as the Food Vessels and urns and so must represent a discreet phase of pre-barrow activity.
55 © Cotswold Archaeology South Wales Gas Pipeline Project Site 513: Archaeological Excavation
Lithics (Pannett 2014) Context Description No. Lithics 268001 Subsoil 4 268008 Fill of cremation urn 268007. 1 268010 Cremation pit fill 1 268020 Single fill of pit 268019. 1 268050 Cremation pit fill 64 268054 Fill of ring-ditch 1 268058 Fill of ring-ditch 1 268073 Fill of ring-ditch 1 268095 Fill of ring-ditch 1 268145 Cremation pit fill 24 268158 Cremation pit fill. 1 Unstratified 6 Total 106
Context 268001 Primary Technology Four complete flakes struck on fresh flint varying from light brown to dark grey were recovered from the subsoil. The flakes are on average 22mm long, 16mm broad and 6.5mm thick. Two flakes retain a cortical platform, two have damaged and indeterminate platforms, while terminations are feathered, hinged and plunging. One flake is a core trimming flake, having been struck to remove hinge fracture scars from the core face. None of the flakes are diagnostic.
Context 268008 Primary Technology A single complete core trimming flake struck on fresh flint was recovered from the fill of cremation urn 268007. The flake is 23mm long, 21mm broad and 6mm thick and has been manufactured on dark grey flint. The platform has been damaged and was indeterminate, while the termination was plunging. The flake has been struck from a single platform blade core, to remove hinge fracture scars from the face of the core. It is diagnostically Late Mesolithic/Early Neolithic.
Context 268010 Primary Technology A single proximal flake fragment was recovered from the fill of cremation pit 268009. The flake fragment is 32mm in diameter and has been struck from a single platform blade core. It retains a planar platform with preparation and a diffuse bulb of percussion, indicting the use of a soft hammer technique. The flake fragment is diagnostic of Late Mesolithic/Early Neolithic knapping.
Context 268020 Primary Technology A single complete flake struck on fresh flint was recovered from the single fill of pit 268019. The flake is 10mm long, 7mm broad and 1mm thick and has been manufactured on dark grey flint. It retains a planar platform and a feathered termination and has a pronounced bulb of percussion indicating the use of a hard hammer technique. The flake is undiagnostic.
56 © Cotswold Archaeology South Wales Gas Pipeline Project Site 513: Archaeological Excavation
Context 268050 Primary Technology An assemblage of 64 struck lithics was recovered from the second fill of pit 268041. The assemblage comprises 16 pieces of burnt flint and 48 pieces of fresh flint, with the colour of the fresh flint varying from white to dark grey. The assemblage is dominated by microdebitage (<5mm in diameter), with 53 pieces recovered, 12 of which were burnt. In addition, four flakes, three flake fragments, two blades and two burnt chunks were also recovered. The complete flakes are on average 22mm long, 10mm broad and 3mm thick while the blades are on average 12mm long, 5mm broad and 1.5mm thick. Three of the complete flakes retain a planar platform, with one retaining a cortical platform, while three retain a feathered termination and one a hinged termination. One blade retains a planar platform, the other damaged, while both retain feathered terminations. One flake has been removed from a single platform flake core while one flake and both blades have been removed from single platform blade cores and are diagnostic of Early Neolithic knapping.
Secondary Technology One flake, struck from the distal end of a single platform flake core has been retouched. Abrupt retouch has been applied to form denticulations along the left hand side dorsal edge. The denticulate is diagnostically Neolithic and would have been used as a cutting tool.
Context 268054 Primary Technology A single complete flake manufactured on fresh flint was recovered from lower fill 268054 of the ring-ditch. The flake is 25mm long, 21mm broad and 3mm thick and was manufactured on mottled grey flint. It retains a planar platform and a stepped termination and has a diffuse bulb of percussion indicting the use of a soft hammer. The flake is undiagnostic.
Context 268058 Primary Technology A single platform flake core manufactured on mottled grey flint was recovered from upper fill 268058 of the ring- ditch. The core is 25mm long, 35mm broad and 29mm thick and had been worked to the end of its useable life, with a large hinge fracture scar on the core face and the angle of the platform edge too steep to successfully remove further flakes. The core is rolled and abraded, suggesting former inclusion in a topsoil deposit in antiquity. The core is diagnostically Neolithic.
Context 268073 Primary Technology A single flake struck from a beach pebble was recovered from the second fill 268073 of the ring-ditch. The flake is 23mm long, 16mm broad and 6mm thick and was manufactured on mottled grey fresh flint. The flake retains a cortical platform and a plunging termination and has been struck to remove the outer surface of a pebble. The flake is undiagnostic.
Context 268095 Primary Technology A single complete flake was recovered from the lower fill 268095 of the ring-ditch. The flake is 4mm long, 8mm broad and 1mm thick and had been manufactured on fresh mottled grey flint. The flake retains a planar platform and a feathered termination. It is undiagnostic.
57 © Cotswold Archaeology South Wales Gas Pipeline Project Site 513: Archaeological Excavation
Context 268145 Primary Technology An assemblage of 24 piece of burnt flint were recovered from the fill of cremation pit 268147. The assemblage is dominated by microdebitage (<5mm in diameter), with 20 pieces recorded. All of the microdebitage pieces are the result of shattering through burning rather than knapping debris. In addition one complete flake, one flake fragment and two burnt chunks were also recovered. The complete flake is 39mm long, 26mm broad and 12mm thick and retains a cortical platform and a plunging termination. It has been struck from a beach pebble and subsequently burnt. None of the pieces were diagnostic.
Context 268158 Primary Technology A single blade fragment was recovered from the fill of cremation pit 268157. It is 7mm in diameter, was manufactured on dark grey fresh flint and had been struck from a single platform flake core. The blade fragment is diagnostically Late Mesolithic/Early Neolithic.
Unstratified Primary Technology Six struck lithics were recovered as unstratified finds. These comprised four complete flakes, a core and a burnt chunk. The complete flakes are on average 19mm long, 13mm broad and 6mm thick and had been manufactured on fresh flint of varying colours. Three retain a planar platform, while 2 retain a plunging termination, 1 a hinged termination and the other an indeterminate termination. One flake was struck from the distal end of a core and is diagnostically Neolithic, the remaining flakes are undiagnostic. The core comprised a rough single platform, bipolar, flake core that had been recently broken into two pieces. It retained a small area of chalky cortex and is diagnostically Early Neolithic.
Discussion The dateable lithics are Late Mesolithic or Early Neolithic and most occur within later features. Neolithic pit 268041 produced 53 pieces of microdebitage, the small chips and flakes resulting from knapping. Microdebitage is not generally found in such quantities unless it has been deliberately deposited as, due to its small size, it is often lost. The presence of this quantity of microdebitage suggests that knapping was occurring in close proximity to the pit and that the knapping waste was collected up and deliberately deposited with the pit fill. Two blades and a flake struck from a single platform blade core were also recovered from the pit fill. These are diagnostically Early Neolithic, earlier than the date of the Grooved Ware pottery also found in the pit fill. The burnt flint from cremation pit 268147 was largely material that had shattered as a result of burning and this is not indicative of deliberate curation.
The lithics found are characteristic of a beach pebble resource, with pebbles probably collected from the nearby coast (approximately 3km to the south) and brought inland to be worked. The lack of tools is interesting, with only a single denticulate recovered. This may, however, be a result of the methodology used during the stripping of the topsoil and subsoil, with the lithics recovered during the excavation of features only representing a small proportion of the original assemblage. It is extremely difficult to identify lithics in the topsoil and subsoil while it is being stripped off by mechanical excavator and it is likely that the majority of the lithic material on the site was removed during this process.
58 © Cotswold Archaeology South Wales Gas Pipeline Project Site 513: Archaeological Excavation
Fired/burnt clay (Hall 2008) A small assemblage of featureless, heavily abraded, fired/burnt clay was recovered. Basal fill 268051 of pit 268041 produced eight fragments, fill 268020 of pit 268019 yielded eight fragments, basal fill 268127 of cremation pit 268126 contained 28 fragments and backfill 268174 of cremation pit 268172 eight fragments. No forms were identifiable.
Metallurgical remains (Young 2014) Introduction Magnetic residues were recovered from 18 contexts. Fill 268038 produced a single block of iron ore, initially misidentified as slag. Finds of iron ore are rare in this area. Growing evidence for early iron smelting in South West Wales suggests that such ores were used as a resource and the analysis of this example aimed at assisting with improving the understanding of early iron smelting. It is worth noting that the site lies 4km from the early medieval iron smelting site at South Hook (Crane and Murphy 2010) and 4km from pipeline site Site 257 (CA 2013), which also had metallurgical remains.
Results The specimen sampled was the largest fragment of three recently broken parts of a single block recovered from the fill of a post-medieval ditch. The age of formation of the ore (apart from its relationship to the ditch fill) is unknown. The specimen (MHB5) showed the development of botryoidal hydrated iron oxide (probably goethite) within a sparse, bioturbated host sediment. The iron oxide comprises approximately 80% of the sample by weight, the host sediment 20%. The ore includes examples of moulds of organic matter, including probable roots.
The ore has a moderately elevated phosphorus content (1.39% expressed as wt% P2O5), but is low in manganese (0.44% expressed as wt% MnO). The silica:alumina ratio (by weight) is low at 4.3. Many trace elements are in elevated concentrations, including Zr at 128ppm, the REE (total REE is 305ppm) and U (24.2ppm). The uranium:thorium ratio is 9.96. The upper crust normalised REE profile is broadly humped.
The sample was of a brownish, porous iron ore. In detail, there are small areas of brown colour, surrounded by a darker, almost black botryoidal material that binds the ore together. In section, the brown areas can be seen to be patches of sediment, bound by the iron mineral. The sediment includes fine-grained materials, probably soil clays, not identifiable by SEM, together with larger grains of quartz, potassium feldspar, micas and clay minerals (illite and chlorite). The patches of sediment generally occur in small areas (400-1000µm across) with rounded outlines, giving a somewhat pelletal texture (though it is not known if the texture is truly pelletal in three dimensions. The texture is likely to be mainly biogenic. Some of the rounded areas show a relationship to bioclasts preserved mainly as moulds. These moulds appear to contain some authigenic minerals as wisps and strands. These materials include silica. One of the bioclasts probably represents the outer parts of an unidentified rootlet with a strongly lobate outline. Another example appears to represent a piece with overlapping, almost imbricate, scale-like components and is perhaps from a moss. The dark material is a botryoidal hydrated iron oxide, probably, but not certainly, goethite. The iron oxides show some penetration of the clastic grains, but there is only limited indication of embayment and corrosion of the detrital materials. Several grains of monazite-Ce were observed within the patches of host sediment. This slightly unusual mineral has been recorded in high concentrations from panned stream samples from settings on late Ordovician rocks of the area south-west of Newcastle Emlyn (Read et al. 1987), and indirect, glacial or fluvial, origin from that area, or an adjacent area with similar later Ordovician bedrock geology is plausible.
59 © Cotswold Archaeology South Wales Gas Pipeline Project Site 513: Archaeological Excavation
Interpretation Although this specimen does not precisely match the chemical properties of the sources interpreted for any of the local iron production sites, it shows some relationship with them. The low silica:alumina ratio, the elevated phosphorus, the elevated zirconium and the uranium:thorium ratio are all features potentially linking the sample to the South Hook Group 2/Site 257 analyses. However, the sample possesses an entirely different REE profile and the manganese content is low. The REE profile can be compared with other examples of samples with humped profiles from South Wales, including a bog ore specimen from South Hook (SHL19, Young 2010; 22 wt% MnO, 51 wt%FeO) and a ‘wad’ specimen associated with the ‘Red Lady of Paviland’ and apparently a pigment (PAV2, Young 2000; 27.7% MnO 3.81%FeO). These samples share a general REE profile, but differ in the size and sense of the cerium anomaly, as well as in the Fe:Mn ratio.
Conclusion This sample not only adds to the discussion surrounding the origin of materials exploited during early iron smelting in South West Wales, but may also, indirectly, help to shed light on the origins of the pigments employed in the Palaeolithic rituals at Paviland.
60 © Cotswold Archaeology South Wales Gas Pipeline Project Site 513: Archaeological Excavation
APPENDIX D: PALAEOENVIRONMENTAL EVIDENCE BY JAMES RACKHAM
Cremated bone (by Anna Fotaki and Malin Holst) A significant amount of cremated bone was retrieved, particularly from the urns (Table 1). Where cremations were urned, they are referred to by their urn number, where remains were un-urned, they are referred to by the cut number of their pit. The bone preservation was very good to very poor, with the majority of bone in a good condition (Table 1).
Table 1 Summary of cremated bone assemblages Context Feature Urned Bone Colour Preservation Weight (g) % of Expected Quantity of Bone Urn 268007 268004 Yes Brown/White Poor 66.2 4.1 (6015-09) Urn 2687011 268009 Yes Brown/White Good 1,019.8 MNI-2 (6015-10) Fill 268020 268019 - - Very Poor 0.8 - Fill 268095 268092 - - Very Poor 0.3 - Urn 268115 268113 Yes Brown/White Good 529.4 MNI-2 Fill 268127 268126 No Grey Good 977.5 MNI-2 Fill 268143 268142 - - Very Poor 0.2 - Urn 268144 268142 Yes Brown/White Good 418.2 0.26 Urn 268146 268147 Yes Brown/White Good 465.4 MNI-2 Fill 268158 268157 - - Very Poor 0.8 - Lower urn 268157 Yes Brown/White/Grey Good 1,220.7 MNI-3 268159 Upper urn 268157 Yes Reddish/Brown Very Good 1,212.9 MNI-3 268159 Fill 268171 268170 No Off White Good 528.5 33 Urn 268173 268172 Yes White/Brown Poor 1.4 0.08 Fill 268175 268177 No Off white/Light Brown Poor 5.4 0.3 Urn 268176 268177 Yes Light red/brown Very Good 1,243.0 MNI-2 u/s - - - Very Poor 1.0 -
The quantity of cremated bone recovered from the Site 513 burials varied in weight from 1.4g to 1,267.6g (see Tables 1 and 2), with an overall mean weight of 645.4g. The majority of the bones were well burnt, causing the complete loss of the organic portion of the bone and producing a white or off white colour (see Table 1). Some bone surface detail had been lost with very limited long bone warping. Bone cracking was also evident. Almost all of the assemblages contained bone fragments that were 10mm in size or larger (Table 2). Fragmentation varied considerably between burials; upper urn 268159 contained the largest quantity of fragments larger than 10mm, with the burials in urns 268115 and 268173 following close behind. Most pits contained cremated bone that was largely within the 10mm or 5mm size range (Table 2).
61 © Cotswold Archaeology South Wales Gas Pipeline Project Site 513: Archaeological Excavation
Table 2 Summary of cremated bone fragment size Context Feature 10mm 10mm 5mm 5mm 2mm 2mm <2mm <2mm Weight (g) (%) (g) (%) (g) (%) (g) (%) (g) Urn 268007 268004 9.4 14.1 21.3 32.3 28.9 43.6 6.6 10 66.2 (6015-09) Urn 2687011 268009 347.2 34 305.1 30 345.2 33.8 22.9 2.24 1019.8 (6015-10) Urn 268115 268113 325.4 61.5 113.2 21.4 77.5 14.6 13.3 2.5 529.4 Fill 268127 268126 500.1 51.2 340 3.5 135.1 13.8 2.3 0.2 977.5 Fill 268143 268142 Urn 268144 268142 114.3 27.3 150.6 36 146.6 35 6.7 1.6 418.2 Urn 268146 268147 172.1 36.8 171.9 36.7 119.1 25.5 4.4 0.9 467.5 Lower urn 268157 726.6 59.5 317.3 26 148 12.1 28.8 2.4 1220.7 268159 Upper urn 268157 898.8 70.8 245.4 19.3 93.3 7.3 30.7 2.4 1267.6 268159 Fill 268171 268170 133.3 25.2 233.7 44.2 154.1 29.2 7.4 1.4 528.5 Urn 268173 268172 - - - - 1.4 100 - - 1.4 Fill 268175 268177 0.6 11.1 1.7 31.5 2.8 51.8 0.3 5.5 5.4 Urn 268176 268177 778.4 62.6 249 20 93.5 7.5 29.6 2.4 1243
There were five smaller bone assemblages from the site which were minute in quantity and due to their poor preservation and small quantity it was not possible to determine species (Table 3). In the case of all burials it is unclear whether post-depositional or post-burning disturbance of the bone caused the fragmentation.
Table 3 Summary of smaller cremations found on site Context Feature Max fragment Total fragment Weight length (mm) no. (g) Fill 268020 268019 10.1 2 0.8 Fill 268095 268092 16.3 1 0.3 Fill 268143 268142 11.2 1 0.2 Fill 268158 268157 11.9 - 0.8 u/s - 14.4 - 1.0
It was possible to identify between 21.7% and 79.1% of the skeletal elements in the cremations (Table 4), with an average of 59% of bone being identifiable. All of the burials in Table 4 could be positively identified as human. The majority of identifiable bone from the burials was either skull elements or long bones. Since the cranial vault is very distinctive and easily recognisable, even when severely fragmented, it often forms a large proportion of identified bone fragments in cremated remains (McKinley 1994). There was a substantial quantity of tooth fragments that allowed for more accurate aging and there were also recognisable fragments of other skull elements, as well as pelvis, vertebrae, lower limb bones and upper limb bones, although the latter in less abundance. No sexually dimorphic skeletal elements survived, so sexing was not possible in any of the assemblages.
62 © Cotswold Archaeology South Wales Gas Pipeline Project Site 513: Archaeological Excavation
Table 4 Summary of identifiable elements in the cremation burials Burial Skull Skull Axial Axial UL UL LL LL UIL UIL Tot ID Tot Tot Tot (g) (%) (g) (%) (g) (%) (g) (%) (g) (%) (g) ID UID UID (%) (g) (%) Urn 6.7 10.1 5.2 7.8 - - - - 2.5 3.7 14.4 21.7 51.8 78.2 268007 (6015-09) Urn 214.1 21 31.1 3 29.3 2.9 31.8 3.1 239.4 23.5 545.7 53.5 474.1 46.5 2687011 (6015-10) Urn 180.5 34 71.7 13.5 29.3 5.5 43.4 8.2 80.8 15.3 405.7 76.6 123.7 23.4 268115 Pit 218.6 22.4 25 2.5 43.3 4.4 58.4 6 408.5 42 753.8 77.1 223.7 22.9 268126 Urn 53 12.7 20.4 4.9 9.2 2.1 20.6 4.9 81.8 19.6 185 44.2 233.2 55.8 268144 Urn 165.1 35.3 24.1 5.1 29.1 6.2 10.2 2.2 109.3 23.4 337.8 72.3 129.7 27.7 268146 Lower urn 232.7 19 216.1 17.7 58.4 4.8 132.3 10.8 228.2 18.7 867.7 71 353 29 268159 Upper urn 275 21.7 286 22.6 62.6 4.9 136.8 10.8 242.8 19.1 1003.2 79.1 264.4 21 268159 Fill 105 20 10.3 1.9 3.8 0.7 13.1 2.5 156.1 29.5 288.3 54.5 248.1 47 268170 Urn 0.5 35.7 ------0.1 7.1 0.6 42.8 0.8 57 268173 Pit 1.6 29.6 0.2 3.7 - - - - 1.2 22.2 3 55.4 2.3 42.6 268177 Urn 205.5 16.4 226.3 18.2 126.8 10.2 113.7 9.1 194.4 15.6 865.7 69.6 377.3 30.4 268176
MNI The minimum number of individuals can be achieved by identifying duplicate elements. There were a noticeable number of multiple burials (Table 5). Based on duplicated bone elements or skeletal elements of individuals of different ages, a total of 21 individuals were identified. The triple burials were found in upper and lower urns 286159. However, it is possible that the remains of these two urns were intermingled, since the upright urn onto which a larger urn had been inverted was connected at the vessel openings. The age of the individuals found in these vessels, particularly the adolescent and young juveniles, are quite close in age. Upon further detailed examination, it was found that both upper and lower urn contained a pubic ramus each of opposite sides, that, based on their size and morphology could belong to the same individual. No other pairs of bone elements were confidently identified, but that does not exclude the possibility that these remains were co-mingled. The ramus of the pubic symphysis under examination could belong to an older juvenile or adolescent.
63 © Cotswold Archaeology South Wales Gas Pipeline Project Site 513: Archaeological Excavation
Table 5 Summary of osteological results Context Feature Preservation MNI Species Age Weight Notes (g) Urn 268004 Poor 1 Human Non-adult (0.4-0.8 66.2 - 268007 years) (6015-09) Urn 268009 Good 2 Human One adult and one 1019.8 - 2687011 non-adult (under 14 (6015-10) years) Fill 268019 Very poor - Unknown - 0.8 - 268020 Fill 268092 Very poor - Unknown - 0.3 - 268095 Urn 268113 Good 2 Human Two non-adults (one 529.4 Cribra orbitalia on 268115 close to 4 years old, one orbit and one between 0.8- 2.9 years) Fill 268126 Good 2 Human One adult and one 977.5 Facet of vertebra 268127 non-adult with DJD Fill 268142 Very poor - Unknown - 0.2 - 268143 Urn 268142 Good 1 Human Young Juvenile (1.5-4 418.2 - 268144 years) Urn 268147 Good 2 Human One adult (30 years 465.4 Facet of vertebra 268146 plus), one infant (0.3- with DJD 0.5 years) Fill 268157 Very poor - Unknown - 0.8 - 268158 Lower urn 268157 Good 3 Human One adult (36 years 1220.7 Osteochondritis 268159 plus), one adolescent dissecans on right (<14), one juvenile adolescent talus, (5.1-6.2 years) adult hand phalanx with osteophyte formation, possible intermixing with remains from Upper Urn Upper urn 268157 Very Good 3 Human One adolescent (over 1212.9 Non-metric 268159 14 years), one young emarginate juvenile, one young patella, possible juvenile (1.8-2 years) intermixing with remains from Lower Urn Fill 268170 Good 1 Human Adolescent <14 years 528.5 - 268171 Urn 268172 Poor 1 Likely - 1.4 - 268173 human Fill 268177 Poor 1 Human - 5.4 - 268175 Urn 268177 Very good 2 Human One adolescent and 1243 Non-metric septal 268176 one Juvenile (4-6 Aperture on two years) humeri fragments u/s - Very poor - Unknown - 1 -
Age Age is split into a number of categories, from foetus (up to 40 weeks in utero), neonate (around the time of birth), infant (newborn to one year), juvenile (1-12 years), adolescent (13-17 years), young adult (ya; 18-25 years), young middle adult (yma; 26-35 years), old middle adult (oma; 36-45 years), mature adult (ma; 46+) to adult (an individual whose age could not be determined more accurately than that they were eighteen or over). Age was
64 © Cotswold Archaeology South Wales Gas Pipeline Project Site 513: Archaeological Excavation
determined using standard ageing techniques, as specified in Scheuer and Black (2000a; 2000b) and Cox (2000).
There was a conspicuous presence of non-adults in this cemetery, the youngest being that in urn 268146, who was aged between three and six months based on the presence of an undeveloped, and unerupted deciduous molar, an unfused hand phalanx and the thin skull fragments. Urn 268176 contained a double burial of two non- adults, one adolescent based on long bone fusion stages, and one juvenile aged between four and six years old, based on comparison with a reference of a mandibular condyle neck. Pit 268170 contained the remains of an adolescent under the age of fourteen based on presence of an unfused proximal humerus fragment.
In pit 268157 upper urn 268159 contained three non-adults. One of these was an adolescent over the age of fourteen based on the presence of a right pelvic ischial tuberosity and partial notch, which appeared to have just fused. There were also undeveloped deciduous molars and canines suggesting the presence of a younger juvenile aged between one and a half and two years. Finally, there was evidence of a slightly older juvenile based on long bones and tooth fragments that did not match the age profile of the other two. Lower urn 268159 also contained one young juvenile who was aged between five and six years of age, based on the presence of incomplete molar crowns. There was also evidence of an adolescent or older juvenile under the age of fourteen based on unfused iliac blades and size of unfused long bones. There is a possibility, however, that material from the upper and lower urns were intermixed, as was previously mentioned. In total, the two urns together contained at least five individuals. Urn 268144 contained the remains of a young juvenile aged between one and a half and four years, apparently interred in a single burial. The ageing was achieved primarily with comparison of the mandibular condyle and surviving thoracic vertebra fragments with a reference collection, taking into account the approximately 15% shrinkage that occurs during cremation of bones. Pit 268126 contained a non-adult of unknown age based on the presence of two left petrous temporal bones, one belonging to the adult; the other was too small. Urn 268115 contained the double burial of two non-adults, one aged close to four years based on the unfused dens of the axis and dental remains and a non-adult aged between six months and two and a half years based on an undeveloped deciduous mandibular molar crown.
Urn 268007 contained a single burial of a non-adult aged three to nine months years based on developing deciduous tooth fragments, whilst urn 268011 contained a non-adult under the age of fourteen based on the non- fusion of long bone elements. The adults were identified by bone size and fusion, but more specific ages were achieved when certain pathological conditions, such as degenerative joint disease or osteophyte formation which tend to occur after the age of 36 were identified.
Non-Metric Traits Urn 268176 contained evidence of a septal aperture on two surviving humeri fragments, which likely belonged to the same individual. Also, upper urn 268159 contained evidence of the emarginated patella non-metric trait.
Pathology A number of pathological manifestations were observed. More specifically, Urn 268146 contained a vertebral facet with evidence of moderate degenerative joint disease (DJD), as did burial 268126. Osteophyte formation on a phalanx was found in the adult found in lower urn 268159. Urn 268115 contained a non-adult eye orbit with evidence of cribra orbitalia. Cribra orbitalia is a term used to describe fine pitting in the orbital roof which develops during childhood and often recedes during adolescence or early adulthood. Until recently, iron deficiency anaemia was the accepted cause of these lesions (Stuart-Macadam 1992), but a strong case has been made by
65 © Cotswold Archaeology South Wales Gas Pipeline Project Site 513: Archaeological Excavation
Walker et al. (2009) for different types of anaemia as the causative factor. These include megaloblastic anaemia in the New World, suggesting a diet deficient in Vitamin B12 (i.e. plant-based and lacking in animal products) and/or folic acid. Such dietary deficiency could have been exacerbated through poor sanitation leading to infection and infestation with gut parasites (ibid). In malarious areas of the Old World, haemolytic anaemia (e.g. sickle cell anaemia and thalassemia) may be important in the development of cribra orbitalia (ibid). However, for areas such as northern Europe they have proposed that cribra orbitalia may be more likely related to conditions such as scurvy (Vitamin C deficiency) or chronic infections (ibid). Cribra orbitalia is often used as an indicator of general stress (Lewis 2000, Roberts and Manchester 2005) and is often found associated with agricultural economies (Roberts and Cox 2003). The presence of this stress factor on this individual may suggest a stressful upbringing or an adaptation to surviving earlier stressful incidences.
Finally, lower urn 268159 contained a right adolescent talus with osteochondritis dissecans. It is caused by fragmentation of the cartilage and the sometimes the underlying subchondral bone and is caused by trauma. It usually appears during adolescence and is more prevalent in males rather than females (Ortner 2003). Dentition A number of tooth fragments were retrieved from the majority of burials, with the noticeable presence of non-adult deciduous dentition.
Summary and Funerary Ritual Multiple cremations are not uncommon in the Bronze Age, and often non-adults were cremated and interred with adults, the reasons being either cultural or practical, or a combination of both. Often non-adults are overlooked in the archaeological record, and in the case of cremation burials this can sometimes be because of the fragility and size of their bones. Non-adults are very well represented in this Early Bronze Age cemetery and the demographic profile of the non-adults suggests a wide age range at death. Of particular interest were the urns (both numbered 268159) in pit 268157, which contained one urn upturned over a lower urn. This is quite a rare deposition of Bronze Age cremation urns and included either five or six individuals co-mingled or three individuals in each urn (a total of six). The fragmented nature of the bones made it hard to be certain of this, but the presence of a pair of pubic ramuses in each urn suggests co-mingling of remains.
Early Bronze Age cremation cemeteries, such as this one, with this many Collared Urns and food vessels used to contain human cremated remains are rare in the South West of Wales. In particular, the burial containing an upturned Urn over a lower one is unusual. Notably, the upper vessel is a rare form nationally, making this an important burial that warrants further research and it is recommended that this burial undergoes AMS dating.
Animal or Human Bone A small assemblage (172 pieces) of very fragmented, largely indeterminate burnt bone (5g) was recovered from sample 268012 (cremation pit fill 268145). A few fragments suggest the thin platy bone of the human cranium, possibly a juvenile. This context is a cremation ‘backfill’ in the pit with urn 268147 and may well be associated with the infant discussed above in this urn.
Environmental soil samples Twenty one environmental samples were taken at this site. The samples were taken from distinct areas of the site, a scatter of probable late Neolithic pits to the west of the field and number of cremation pits, pits and a posthole within a ring ditch, and the ring ditch itself, some 30m to the east. Although no evidence was found for a mound the ring ditch possibly enclosed a barrow. An isolated pit 20m to the south-east of the ring ditch, 268102
66 © Cotswold Archaeology South Wales Gas Pipeline Project Site 513: Archaeological Excavation
was also sampled. Two pits within the western pit group have been dated by radiocarbon to the Late Neolithic (Table 6) and have produced Late Neolithic grooved ware (although one sampled pit, 268129, is undated), while the cremations associated with the ring ditch have been radiocarbon dated to a 100-300 year span in the early Bronze Age (Appendix E), supporting the dating suggested by the collared urns and food vessels associated with the cremations. Both the ceramic and radiocarbon evidence indicate that these two groups represent chronologically distinct episodes on the site. The isolated pit, 268102, to the south east of the ring ditch is undated.
Table 6. Bulk environmental samples collected and processed from Site 513. feature Volume Date (C14 dates at 95% sample context Context type processed probability) 268001 268020 268019 pit fill 24 2890–2660 cal BC (hazelnut) 2570–2450 cal BC (hazel char) 268002 268050 268041 pit fill 40 2880–2570 cal BC (hazel char) 268003 268051 268041 pit fill 8 2880–2570 cal BC (hazel rwd) 268004 268103 268102 pit fill 20 Undated/? Neo/LNeo? 268005 268111 268110 Urn and cremation fill 5 EBA 268006 268114 268113 exterior of cremation urn 18 1960–1770 cal BC (hazel char) 268007 268133 268132 Cremation pit fill 8 EBA 268008 268128 268126 cremation pit fill 44 EBA 268009 268141 268140 cremation pit fill 6 EBA 268010 268134 268129 pit fill 30 undated 268011 268143 268142 cremation pit fill 2 EBA 268012 268145 268147 cremation backfill 16 1900–1690 cal BC (hazel char) 268013 268127 268126 cremation pit fill 40 2120–1890 cal BC (human bone) 1900–1690 cal BC (hazel char) 268014 268095/268098 268092 Ring ditch fill 30 EBA 268015 268083/268069 268072 Ring ditch fill 32 EBA 268016 268161 268160 Posthole/pit fill 4 EBA 268017 268158 268157 cremation backfill 30 2040–1880 cal BC (human bone) 1890–1690 cal BC (pot residue) 268018 268171 268170 cremation pit fill 20 EBA 268019 268174 268172 surrounding cremation urn 6 1960–1770 cal BC (holly char) 2030–1880 cal BC (hazel char) 268020 268175 268177 cremation backfill 18 EBA 268021 268176 268177 cremation urn fill 3 2140–1940 cal BC (human bone)
The samples were processed in the manner described in the assessment report (Giorgi and Martin 2008) with the additional refloating of the dried <2mm sample residues from the samples processed by Cotswold Archaeology, whose flot volume is indicated in Table 6. This second flot was then sorted for charred macrofossils and the residue re-dried and checked with a magnet to recover any further magnetic material. For the Late Neolithic samples where hazel nutshell fragments were abundant a weighed or volumed fraction of the 2nd flot was sorted (between 15 and 25%) and an estimate of the total number of nutshell fragments estimated from the count in the sorted fraction. During the assessment only a proportion (25 or 50%) of the 1st flot was assessed (Giorgi and Martin 2008) the remainder of the 1st flots were sorted for this post-excavation study except for two samples (268017 and 018) where only the >2mm fraction of the un-assessed 1st flot was sorted. The relative absence of identifiable material in the assessed 25% (including the <2mm) was taken as representative.
The results of the sample processing and analyses are presented below in Table 7-10.
67 © Cotswold Archaeology South Wales Gas Pipeline Project Site 513: Archaeological Excavation
Table 7. Data from the environmental bulk samples, Site 513.
1st 2nd burnt volume cremated fired worked burnt magnetic sample context flot flot mammal coal pot CBM slag sample comments proc. l. human Bone clay flint flint material ml* ml $ bone g. Late Neo and undated pits 268001 268020 24 125 nd 0.8 E B A 268002 268050 40 300 40 D D A residues rich in charcoal & cpr 268003 268051 8 100 16 E E C 268010 268134 30 30 18 A+3.2 268004 268103 20 100 23 E A Ring ditch and cremations 268005 268111 5 60 5 E C 268006 268114 18 200 4 4 268007 268133 8 300 54 A+0.8 HNSx1 (2nd flot) 268008 268128 44 300 15 D A+9.6 268009 268141 6 8 25 3.1 A 268011 268143 2 22 11 0.2 E E C HNSx1 (2nd flot) 268012 268145 16 165 3 5 D E D E B 268013 268127 40 900 300 A A+103 268014 268095/268098 30 300 9 0.3 E C 268015 268083/268069 32 20 2 E A 268016 268161 4 100 12 B 268017 268158 30 500 17 0.8 D E A+0.2 268018 268171 20 100 27 A E A 268019 268174 6 95 12 E D A 268020 268175 18 800 20 D D 268021 268176 3 25 15 A D residues rich in charcoal & burnt bone * >1mm; abundance rating – E= 1-10 items; D=11-50, C=51=100, B=101-200, A=>200; nd – no data; HNS – hazel nutshell $ 2nd flot taken from the <2mm residue
68 © Cotswold Archaeology South Wales Gas Pipeline Project Site 513: Archaeological Excavation
The samples fall into two groups, a Late Neolithic group of pits, with two undated pits, one of which may well be Neolithic, and the samples from the ring ditch and internal features.
The samples from the Neolithic pits produced a little pottery, fired clay, burnt flint and a magnetic component (Table 7). An entry of ‘large mammal bone’ from sample 268001 was noted on the processing record, but this is represented by just 0.8g of unidentifiable burnt bone (see Fotaki and Holst, above). The samples are rich in charcoal and contain abundant charred hazel nutshell (Table 8), with rare barley grains and a fruit stone fragment. The frequency of charred nutshell in the second flot of samples 268002 and 268003 (Table 8) suggests that some nutshell will have been thrown away with the >2mm residue when this was discarded after processing, even though the residues were hand sorted by eye.
Table 8 Charred plant macrofossils from the Late Neolithic and undated pits.
Period LNeo LNeo LNeo LNeo LNeo un- un- LNeo LNeo? dated dated Sample 001 002 002 003 003 010 010 004 004 context 020 050 050 051 051 134 134 103 103 nd nd nd nd Flot 1st flot 1st flot 2 flot 1st flot 2 flot 1st flot 2 flot 1ST flot 2 flot Residue fraction floated <2mm <2mm <2mm <2mm Flot volume ml 125 300 40 100 16 30 18 100 23
Cereals Hordeum sp. Barley grain 1 cf Hordeum cf barley grain 1 Cereal indet Cereal grain fragment 1 1 1
Others Corylus avellana Hazel nutshell – count/wt 14/+ 46/1 6131e/10 7/+ 1258e/2 2/+ 538e/1.3
Nutshell count/wt from e Corylus avellana 467/8 2049 /48 216/3 31/0.5 res. Prunus sp. Fruit stone fragment 1 e estimated on the basis of a counted weighed proportion (between 15 and 25%); + not weighed but <1g or present but not counted;
Although technically undated the assemblage of charred nutshell in isolated pit 268102, with a probable barley grain, is sufficiently similar to the dated pits that a similar age can be suggested. Pit 268129 which lies 20m east of the dated late Neolithic pits produced just two fragments of nutshell and an unidentifiable cereal grain, and with no other finds (Table 7), other than a magnetic fraction suggesting some mineral hearth debris, this feature could be of any age. It seems probable that the late Neolithic pits and pit 268102 are domestic in character, with a reasonably high charcoal content, pottery, a little fired clay, burnt flint and bone, and abundant nutshell with grain and a fruit stone, this is characteristic of occupation deposits and contrasts with the much poorer assemblages from the ring ditch features.
The samples from the ring ditch site have been dated or assigned by association to the Early Bronze Age. A number of them quite clearly represent cremation burials (see Fotaki and Holst, above), either urned or un-urned, whilst others have no associated cremated human remains and may not be burials. Sample 268009 from pit 268140 is recorded as producing over 50 fragments of unidentified burnt bone (Table 7) weighing 3.1g (Caffel and Holst 2009), possibly a cremation; possible cremation burial 268132 produced no burnt bone from an eight
69 © Cotswold Archaeology South Wales Gas Pipeline Project Site 513: Archaeological Excavation
litre sample; and five litres of sample from urn and fill 268110 produced no burnt bone, although the ceramics suggest a cremation burial. Two cremation pits were not sampled, 268004 and 268009. Two samples were taken from the ring ditch, 268014 and 268015, and posthole 268160 was sampled. A little burnt bone and burnt flint were recovered from ditch deposit 268095/098, a little ‘slag’ from ditch deposit 268083/069 and nothing but a small magnetic component from posthole 268160 (Table 7).
The environmental assemblages from all the samples associated with the ring ditch are dominated by charcoal with occasional fragments of charred hazel nutshell (Table 9). Charcoal content varied across the burial pits with very little in pit 268140 (5ml per litre) and the highest concentration in the backfill in cremation 268177 at 45ml per litre. Charred hazel nutshell occurs consistently through the samples, occurring in eight of the eleven sampled cremation pits (Table 7 and 9), and the ring ditch and posthole. In comparison with the samples from the late Neolithic pits nutshell fragments were very few in number and were generally recovered during the residue sort and from the second flot. Nevertheless their consistency suggests consumption of the kernel and discard of the nutshell onto some of funeral pyres or offerings since it would seem unlikely that branches bearing nuts were used as fuel in the pyres and there is no evidence for late Neolithic occupation predating the ring ditch from which residual nutshell could have derived. Very rare charred cereal grains are present although one of the three grains derives from the ring ditch. Whilst this evidence would support an inference that eating took place at the monument, presumably in association with the cremation events, or offerings were made it is not perhaps sufficient to imply ‘feasting’.
Two samples, from pits 268170 and 268172, produced no hazel nutshell but a few weed seeds and herbaceous tubers, the latter probably Ranunculus ficaria, the lesser celandine. With seeds of daisy family, cleavers, plantain and possible dock it is difficult to see these two samples as anything other than fortuitous, perhaps some local herbaceous vegetation on the fire.
With several of the samples deriving from cremation burials charcoal assemblages were studied from eight of the cremations and two of the Late Neolithic pits.
70 © Cotswold Archaeology South Wales Gas Pipeline Project Site 513: Archaeological Excavation
Table 9. Charred plant macrofossils from the ring ditch and features within it
Cremations burial pits Ring ditch posthole Period EBA EBA EBA EBA EBA EBA EBA EBA EBA EBA EBA EBA EBA EBA EBA EBA EBA EBA Sample 006 006 008 008 012 012 017 017 018 019 020 020 014 014 015 015 016 016 context 114 114 128 128 145 145 158 158 171 174 175 175 098 098 083 083 161 161 Flot 1st flot 2nd flot 1st flot 2nd flot 1st flot 2nd flot 1st flot 2nd flot 1st flot 1st flot 1st flot 2nd flot 1st flot 2nd flot 1st flot 2nd flot 1st flot 2nd flot Residue fraction floated <2mm <2mm <2mm <2mm <2mm <2mm <2mm <2mm Flot volume 200 4 300 15 165 3 500 17 100 95 800 20 300 9 20 2 100 12 Hordeum sp. Barley grain cf Hordeum cf barley grain Hordeum/Triticum Indet grain, wheat or 1 sp. barley Cereal indet Cereal grain fragment 1 1
Corylus avellana Hazel nutshell – count/wt 1/+ 9/+ 1/+ 30/+ 1/+ 8/+ 3/+* 13/+ 1/+* 6/+ 2/+ 19/+ 1/+ Corylus avellana Nutshell count/wt from 23/+ 29/+ 17/+ 3/+ 4/+ res. Poaceae Indet grass seed, large 1 Poaceae Small indet grass seed 1 Rubus sp. Bramble/raspberry 1 cf Asteraceae Daisy family 1 cf Galium sp. cleaver 9 Plantago sp. plantain 1 Indet Small indet trigonous seed 1 cf Rumex sp. Dock family 1 cf Ranunculus tubers 12 10 ficaria Unidentified seeds 11 1 Buds 6 2 1 + 4 + not weighed but <0.5g or present but not counted; * for 75% of the 1st flot only the > 2mm fraction was sorted.
71 © Cotswold Archaeology South Wales Gas Pipeline Project Site 513: Archaeological Excavation
Charcoal analysis (Dana Challinor) Ten samples were examined for charcoal analysis, following standard procedures, with 30 fragments identified from each sample, excepting two particularly poor assemblages (from cremation pit 268140 and urn 268142) from which only 20 were studied. A relatively restricted range of four taxa were positively identified: Quercus sp. (oak), Alnus glutinosa (alder), Corylus avellana (hazel) and Ilex aquifolium (holly) (Table 10). A total of 280 fragments were identified. Charcoal was abundant in most of the samples, but most of the fragments were mid- small sized, and a heavy covering of sediment obscured anatomical details. Oak heartwood was present in most samples, with occasional sapwood recorded, although this is (to some extent) a reflection of the condition of the charcoal and it is likely to be under-represented. Roundwood fragments were infrequent and incomplete, with the exception of the non-cremation pit samples which produced greater evidence for strong ring curvature.
The Late Neolithic pits The function of these pits is unclear, but may have been used for the deposition of waste remains of domestic cooking. The assemblages contained notable quantities of roundwood, as well as substantial fragments of detached bark. Pit 268019 produced both hazel and oak charcoal, while Pit 268041 was more diverse, with additional taxa; alder and holly. These taxa could all have been readily gathered from the local oak-hazel woodland, with alder preferring a riverside habitat. These samples contrast with the Early Bronze Age cremation assemblages, supporting the interpretation that the latter were the product of deliberate selection.
The Early Bronze Age cremation samples The cremation assemblages were overwhelmingly dominated by oak, mostly trunkwood including mature wood, with the exception of pit 268147, which produced mostly hazel. Generally, the rare fragments of hazel were probably from kindling or brushwood infilling on the pyre. Pit 268147 contained an inverted urn, but so did pit 268177 which was oak dominated, so this does not adequately explain the dominance of hazel in the former. The lack of human bone survival prohibits examination of possible links relating to gender or age. Oak is commonly used in Bronze Age cremations as it provides the high calorific heat necessary for efficient cremation, as well as being suitable for bier support/pyre structure. The dominance of a single taxon in these assemblages is also a pattern commonly seen in Early Bronze Age cremations, with the possibility that it may have performed a ritual as well as functional role in the burial rites (Thompson 1999).
Discussion The site lies on the north facing slopes of a low hill at 38m OD overlooking a small stream valley to the north. For a monument to be sited here in the Early Bronze Age the gentler slopes of the hill are likely to have been cleared of woodland, but this may not have been the case in the Late Neolithic at the time the pits were dug. At this earlier period the Late Neolithic activity may have lain within clearings in a much broader expanse of oak and hazel woodland. The fact that at this early period the settlement exploited alder (Table 10) for its fires suggests that they were accessing as far as the streamside for fuel, where the nearest alder is likely to have grown, a 100m north of the site, although this could indicate that the landscape between the site and the stream had been cleared of woodland. It is possible that the few acres in the stream valley had been cleared making access to the water and the adjacent pasture for stock and the settlement. At sites further east along the pipeline this lowland landscape appears to be dominated by woodland (Rackham et al in prep) in the Neolithic. The Neolithic features have produced a characteristic charred plant assemblage dominated by fragmented hazel nutshell, with just a few grains of barley, and a fragment of fruit stone.
72 © Cotswold Archaeology South Wales Gas Pipeline Project Site 513: Archaeological Excavation
Table 10. Charcoal from Late Neolithic pits and Early Bronze Age cremation pits from Site 513
Feature type Early Bronze Age cremation pits urn Late Neolithic pits Feature number 268113 268126 268132 268140 268147 268172 268177 268142 268019 268041 Context number 268114 268127 268133 268141 268145 268174 268175 268144 268020 268050 Sample number 268006 268013 268007 268009 268012 268019 268020 268011 268001 268002 Quercus sp. oak 27 (hsr) 29 (h) 30 (h) 20 6 (r) 21 (h) 30 (sh) 20 (h) 6r 5 (hr) Alnus glutinosa Gaertn. alder 5 (r) Corylus avellana L. hazel 3r 1 19 (r) 9 14r 8 (r) Alnus/Corylus alder/hazel 5 6 Ilex aquifolium L. holly 1r Indeterminate bark 10 5 Total 30 30 30 20 30 30 30 20 30 30 h=heartwood;s=sapwood; r=roundwood; (brackets denotes presence in some frags only)
73 © Cotswold Archaeology South Wales Gas Pipeline Project Site 513: Archaeological Excavation
Despite the very large number of fragments recorded from the samples, many of these were only recovered from the <2mm fractions, and the total weight of surviving and recovered shell comprises less than 75g, although it must represent only a fraction of what was thrown on the fire and was burnt rather than carbonised. Since the grain, fruit stone and nutshell all have different pathways to the fire their proportions in the samples are no reflection of their importance at the site, just the fact that they were exploited. The implications are that cereals were being cultivated locally, and other foods harvested from the wild. The few fragments of burnt bone are presumably animal rather than human but yield no useful information. The isolated undated pit, 268102, just over 80m east of the dated Late Neolithic pits has a very similar assemblage and may be broadly contemporary, or perhaps reflect an earlier episode of Neolithic activity such as a campsite within the woodland, not dissimilar to other nut-rich deposits in isolated features found elsewhere along the pipeline.
By the Early Bronze Age, some 500 years after the Late Neolithic activity (Appendix E), it is probable on the basis of the siting of the monument, that the hillside was cleared of woodland so that it could be seen from distance and from across the stream valley, and quite possibly from the hill to the south, although there is no way of proving this. It also sits above and overlooking a shallow dry valley that runs broadly north south up the hillside for a couple of hundred metres immediately to the east. The site is interesting for the number of juvenile and adolescent burials (Table 5).
As is to be expected with cremations the fuel used for the pyres was predominantly oak, probably largely trunkwood. The exception was inverted Collared Urn cremation burial 268147 with the remains of an adult of 30+ years and an infant of just a few months. In this sample hazel dominated. Despite the association of the two individuals in the same cremation burial Griffiths (Appendix E, below) has suggested the possibility, on the basis of inconsistent radiocarbon results on one feature from the site, that human material could have been curated and re-buried with later cremations, or residual material picked up and incorporated with a new burial. The hazel might be associated with the infant burial, perhaps even a basket or hurdle it lay in or on when placed on the pyre. This is pure speculation but perhaps an avenue for future research, and the radiocarbon dating of each individual in such double or triple burials (see Table 5) might be warranted. There is a positive association of small quantities of burnt hazel nutshell with the burials. Eight out of twelve of the cremation samples produced small fragments of nutshell. Their occurrence seems more likely to reflect consumption of hazelnut kernels at the ‘event’ or food offerings thrown on the fire than casual inclusion of residual material (for which there is no evidence) or the use of nut carrying boughs on the pyre. The very occasional finds of charred cereal grain are also difficult to account for without some purposive action. These few remains, and the bramble/raspberry, may indicate offerings placed on the pyre or waste from food consumption at the cremation event that was thrown on the fire. Most of the few other plant species identified are typical of grassland and may merely reflect the setting of the site or material used as kindling.
74 © Cotswold Archaeology South Wales Gas Pipeline Project Site 513: Archaeological Excavation
APPENDIX E: THE RADIOCARBON DATING BY SEREN GRIFFITHS
For the analysis, radiocarbon measurements were produced on short-life, single entity charred plant remains. Samples with the ‘SUERC-‘ laboratory code were pretreated using an acid-base-acid process. Samples were combusted and graphitized and then dated by Accelerator Mass Spectrometry (AMS). The results are conventional radiocarbon ages, quoted according to the international standard set at the Trondheim Convention. The results have been calibrated using IntCal13, and OxCal v4.2. The date ranges have been calculated using the maximum intercept method, and have the endpoints rounded outward to 10 years.
Seventeen radiocarbon dates were produced. Results from Neolithic pit 268041 (SUERC-54659 and -54660) are statistically consistent (T’=0.2; T’5%=3.8; df=1; Ward and Wilson 1978), and could be of the same actual age. Results from Neolithic pit 268019 (SUERC-54661 and -54662) are statistically significantly different (T’=0.2; T’5%=28.5; df=1; Ward and Wilson 1978). The first dated event associated with the use of Grooved Ware on this site is 2890–2670 cal BC (95% probable; or 2870–2700 cal BC 68% probable; First Grooved Ware 513; Fig. 18). An estimate for the last dated event associated with the Grooved Ware on this site is 2580–2450 cal BC (95% probable; or 2570–2530 cal BC 50% probable or 2490–2470 cal BC 18% probable; Last Grooved Ware 513; Fig. 18). Activity represented by the radiocarbon dates from these two features could have occurred over 110–410 years (95% probable; or 170–330 years 68% probable; DurationGroovedWare513). The small sample size from this area could mean that these data obscure more punctuated phases of activity. The inconsistency of the results from pit 268019 might indicate that the older result represents residual material.
From the Bronze Age cremations, a single radiocarbon date from pit 268177 containing a Food Vessel estimates presence on the site in 2020–1880 cal BC (95% probable), most probably in 1990–1890 cal BC (68% probable; SUERC-54663; Fig. 18).
Twelve results were produced on features which produced Collared Urns. An estimate for the start of activity associated with the use of Collared Urns on the site, places it in 2020–1890 cal BC (95% probable) or 2010–1920 cal BC (68% probable; First Collared Urn 513; Fig. 18). The end of activity associated with the use of Collared Urns on the site is estimated as occurring in 1870–1700 cal BC (95% probable; or 1860–1730 cal BC 68% probable; Last Collared Urn 513; Fig. 18).
The duration of activity to have been represented at the ring ditch is estimated as 1–300 years (95% probable) or 130–270 years (68% probable; DurationRingDitchActivity).
Three features from which multiple samples were radiocarbon dated are statistically inconsistent. From pit 268113, the charcoal sample (SUERC-54664) is older than the two results from this feature on charred residue from a tripartite urn (SUERC-54668) and cremated human bone (SUERC-54669), which are from the same context in the pit and statistically consistent (T’=1.7; T’5%=3.8; Ward and Wilson 1978). This charcoal sample could therefore be residual, or the materials infilling the feature in some way represent deposits of different ages.
Cremated human remains from pits 268157 and 268126 are older than the results from other sampled materials in the same contexts. In the case of pit 268126, results on charred residue on a bipartite collared urn (SUERC- 54681) and hazel charcoal (SUERC-54680) are statistically consistent (T’=0.3; T’5%=3.8; df=1; Ward and Wilson 1978). In these cases, and the case of 268157 where there are only two results, it is possible that slightly older
75 © Cotswold Archaeology South Wales Gas Pipeline Project Site 513: Archaeological Excavation
human remains were incorporated into the features. There are no reasons to doubt the accuracy of the measurements on cremated bone (Naysmith pers. comm. 2014), therefore there remains the potential that curated, or otherwise stored, human remains were incorporated into some of these features, either for special purposes or through the reworking of deposits on the site (though the sample size of dated contexts with measurements on duplicate materials are few).
Figure 19 shows key parameters from the site 513 model. There is a considerable interval between the Grooved Ware presence (Last Grooved Ware 513; Fig. 19) to the west of the ring ditch and the later activity associated with the use of the ring ditch (Start 513; Fig. 19), in the order of 410–730 years (94% probable; or 480–610 years 68% probable).
It is slightly more probable (64% probable) that activity associated with Collared Urn use within the ring ditch (First Collared Urn 513; Fig. 19) occurred before the activity associated with the deposition of the Food Vessel in pit 268177 (SUERC-54663). However, the radiocarbon sample of use of Food Vessel from the site is very limited, and may not reflect the full currency of this pottery type on the site. The single dated example of Food Vessel deposition was in use between the first and last deposition of Collared Urns in features within the ring ditch.
Context Feature Sampled material Laboratory ref Measured δ13C Calibrated date age (95%) 268020 268019 Corylus sp. nutshell SUERC-54661 4185 +/-29 -25.6 2890–2660 cal BC (GU34668) 268020 268019 Corylus sp. charcoal SUERC-54662 3966 +/-29 -29.4 2570–2450 cal BC (GU34669) 268050 268041 Corylus sp. charcoal SUERC-54659 4138 +/-29 -27.6 2880–2570 cal BC (GU34666) 268051 268041 Corylus sp. charcoal round SUERC-54660 4120 +/-29 -27.7 2880–2570 cal BC wood (GU34667) 268114 268113 Corylus sp. charcoal SUERC-54664 3542 +/-29 -26.2 1960–1770 cal BC (GU34671) 268115 268113 Charred residue on SUERC-54668 3416 +/-29 -27.5 1870–1630 cal BC tripartite collared urn (GU34672) SF268115 268115 268113 Human cremated bone SUERC-54669 3470 +/-29 -27.3 1890–1690 cal BC distal tibia (layer 6) (GU34673) 268127 268126 Human cremated humerus SUERC-54679 3619 +/-29 -23.7 2120–1890 cal BC (GU34680) 268127 268126 Corylus sp. charcoal SUERC-54680 3484 +/-29 -27.2 1900–1690 cal BC (GU34681) 268127 268126 Charred residue on SUERC-54681 3462 +/-29 28.1 1890–1690 cal BC bipartite collared urn. (GU34682) 268145 268147 Corylus sp. charcoal SUERC-54670 3484 +/-29 -25.7 1900–1690 cal BC (GU34674) 268146 268147 Human cremated occipital SUERC-54671 3522 +/-29 -23.7 1940–1750 cal BC and maxilla (layer 2) (GU34675) 268159 268157 Human cremated upper SUERC-54672 3605 +/-29 -23.6 2040–1880 cal BC limb (GU34676) 268159 268157 Charred residue on SUERC-54673 3478 +/-29 -27.7 1890–1690 cal BC collared urn SF268159 (GU34677) 268174 268172 Ilex sp. charcoal SUERC-54674 3549 +/-29 -24.2 1960–1770 cal BC (GU34678) 268174 268172 Corylus sp. charcoal SUERC-54678 3588 +/-29 -29.3 2030–1880 cal BC (GU34679) 268176 268177 Human cremated lower SUERC-54663 3651 +/-29 -23.4 2140–1940 cal BC limb layer (GU34670) Dating undertaken by Scottish Universities Environmental Research Centre
76 © Cotswold Archaeology South Wales Gas Pipeline Project Site 513: Archaeological Excavation
Fig. 18 A Bayesian model for the results from site 513. Each distribution represents the relative probability that an event occurred at a particular time. For each distribution two ranges have been plotted. The range in outline represents the calibrated radiocarbon result, the solid distribution represents the posterior density estimate — or Bayesian statistical model output. The large square brackets and CQL2 OxCal keywords define the model exactly
77 © Cotswold Archaeology South Wales Gas Pipeline Project Site 513: Archaeological Excavation
Fig. 19 Key parameters for site 513 shown in the model calculated in Fig. 18
78 A NNyferyfer
AAC Cothi
PEMBROKESHIREPEMBRP MBROROKESESHIRSSHIREH RE CARMARTHENSHIRECACCARMARARRRMARMARA THENH NSHISH RE n Lly AfonA LlynfiLly dau Brecon w Tywi
A CyninCyni on frfron Ty stern Cleddau AfronAAf Tywi ddau BRECONC BEACONSBBEAEAACONSACACONS NATIONALNNATNATIONATIONAATIONAT O ALA PARKPARPARK WessternWes Eastern Cleddau Cleddau A Taf Taf CywynCy A Cywyn Cyw FawrFawFaawr endraet Milford Gwendraeth Fach draeth A GwendraethGwendr Fach Haven g ugu Site 513 R Loughor
Aberdulais Rhy
AAR Rhondda Fach ondda F Felindre A Rhondda Fawr
awr 0 25km SSWSWANSSWANSEAWWAANANSEEAA
Site 513 Site 512
Site 519
Fig 2 Inset
Site 505
Site 286
N Cirencester 01285 771022 Cotswold Milton Keynes 01908 218320 Milford Haven to Aberdulais pipeline Andover 01264 326549 Archaeology w www.cotswoldarchaeology.co.uk Felindre to Brecon pipeline e [email protected] PROJECT TITLE 0-75m contour South Wales Pipeline. Site 513, Land North-West 75m contour of Steynton, Milford Haven, Pembrokeshire 200m contour FIGURE TITLE Site location plan 400m contour 600m contour 0 1km FIGURE NO. Reproduced from the 2005 Ordnance Survey Explorer map with PROJECT NO. 9150 DATE 07-02-2014 the permission of Ordnance Survey on behalf of The Controller DRAWN BY DJB REVISION 00 of Her Majesty's Stationery Office c Crown copyright Cotswold Archaeology Ltd 100002109 APPROVED BY LM SCALE@A4 1:25,000 1 N 2
FIGURE NO. 1:5000 01908 218320 cotswoldarchaeology.co.uk @ 0 100m www.cotswoldarchaeology.co.uk Site 513 Cirencester 01285 771022 Milton Keynes 01264 326549 Andover w e enquiries 07-02-2014 00 1:250 911 pipeline centreline Late Neolithic Early Age Bronze post-medieval undated DATE REVISION SCALE@A3 pit 268102 9150 DJB LM Cotswold Archaeology c Reproduced from the digital Ordnance Survey Explorer map with the permissionReproduced from the digital Ordnance Survey The Controller of Her Majesty's Stationery Office on behalf of of Ordnance Survey Archaeology Ltd 100002109 Cotswold copyright Crown PROJECT TITLE Pipeline. Site 513, Land North-West South Wales of Steynton, Milford Haven, Pembrokeshire FIGURE TITLE Plan of archaeological features PROJECT NO. BY DRAWN APPROVED BY SM 083 N ring ditch 10m Fig 4 0
extent of truncation by benching 91100 08300 SM 268036 R Q R 268021 268129 268028 Q 268130 268125 X X
268116
W W railway embankment railway 268019 T 268039 A 268104 2890-2660 cal. BC 2890-2660 cal. BC 2570-2450 cal. A U T V V S 268037 268106 S 268041 U 268108 2880-2570 cal. BC 2880-2570 cal. 268100 Section AA
SW NE 39.5m AOD
268049
268050 pit 268041
268051
01m
N Cirencester 01285 771022 Cotswold Milton Keynes 01908 218320 Archaeology w www.cotswoldarchaeology.co.uk e [email protected]
PROJECT TITLE South Wales Pipeline. Site 513, Land North-West of Steynton, Milford Haven, Pembrokeshire
FIGURE TITLE 083 Section of pit 268041 Site 513
FIGURE NO. Reproduced from the digital Ordnance Survey Explorer map with PROJECT NO. 9150 DATE 07-02-2014 the permission of Ordnance Survey on behalf of The Controller DRAWN BY DJB REVISION 00 of Her Majesty's Stationery Office c Crown copyright 01:5000 100m SM 911 Cotswold Archaeology Ltd 100002109 APPROVED BY LM SCALE@A4 1:20 3 K N K 17 18 M O
14 15 19 1940-1750 cal. BC 21 1900-1690 cal. BC J M 12 13 H O 1960-1770 cal. BC 1890-1690 cal. BC L 3 4 N 1870-1630 cal. BC 2120-1890 cal. BC J I I 1900-1690 cal. BC 20 1890-1690 cal. BC 22 16 G 11 2030-1880 cal. BC 5 6 7 H 1960-1770 cal. BC L 8 9 10 P pit/posthole 2040-1880 cal. BC 268160 N 1890-1690 cal. BC 1 2 2140-1940 cal. BC G P F
F 22 268092 possible 0 1:30 1.5m cremation burial
N
083 Site 513
SM 911 01:5000 100m B B
inset 268044 14 & 15 pipeline centreline 19 1 adult 17 & 18 21 1 sub-adult 12 & 13 3 & 4 11 1 age not known 16 22 E 5, 6, 7, 1 & 2 268033 buried within urn 8, 9, 10 pit/posthole 20 E C 268160 buried within organic container
possible cremation burial 23 268138 Reproduced from the digital Ordnance Survey Explorer map with the permission of Ordnance Survey on behalf of The Controller of Her Majesty's Stationery Office c Crown copyright Cotswold Archaeology Ltd 100002109 C
Cirencester 01285 771022 268070 Milton Keynes 01908 218320 Cotswold Andover 01264 326549 268062 Archaeology w www.cotswoldarchaeology.co.uk e [email protected] D PROJECT TITLE SM South Wales Pipeline. Site 513, Land North-West 91123 of Steynton, Milford Haven, Pembrokeshire FIGURE TITLE 08323 Plan of archaeological features D
05m PROJECT NO. 9150 DATE 07-02-2014 FIGURE NO. DRAWN BY DJB REVISION 00 APPROVED BY JB SCALE@A3 1:75 1:30 4 5
Ring ditch, looking south (scales 1m)
6
Section through ring ditch, looking north-west (scale 1m)
N Cirencester 01285 771022 Cotswold Milton Keynes 01908 218320 Archaeology w www.cotswoldarchaeology.co.uk e [email protected]
pipeline centreline PROJECT TITLE Fig. 5 South Wales Pipeline. Site 513, Land North-West Fig. 5 photograph locator of Steynton, Millford Haven, Pembrokeshire FIGURE TITLE Fig. 6 083 Photographs Site 513
FIGURE NO. Reproduced from the digital Ordnance Survey Explorer map with PROJECT NO. 9150 DATE 10-02-2014 the permission of Ordnance Survey on behalf of The Controller DRAWN BY DJB REVISION 00 of Her Majesty's Stationery Office c Crown copyright 01:5000 100m SM 911 Cotswold Archaeology Ltd 100002109 APPROVED BY LM SCALE@A4 N/A 5 & 6 Section BB
E W 38.5m 268001 AOD
268045 ditch 268044
268048
Section CC
SW NE 38.75m AOD
268071
268068
ditch 268063 268070
Section DD Section EE
SSE NNW ENE WSW 46.5m 36.25m AOD 268063 AOD ditch 268062
268034
ditch 268033
Section FF
NW SE
37.75m 268074 AOD ditch 268095 268092
01m
N Cirencester 01285 771022 Cotswold Milton Keynes 01908 218320 Archaeology w www.cotswoldarchaeology.co.uk e [email protected]
PROJECT TITLE pipeline centreline South Wales Pipeline. Site 513, Land North-West of Steynton, Milford Haven, Pembrokeshire
FIGURE TITLE 083 Ring ditch sections Site 513
FIGURE NO. Reproduced from the digital Ordnance Survey Explorer map with PROJECT NO. 9150 DATE 10-02-2014 the permission of Ordnance Survey on behalf of The Controller DRAWN BY DJB REVISION 00 of Her Majesty's Stationery Office c Crown copyright 01:5000 100m SM 911 Cotswold Archaeology Ltd 100002109 APPROVED BY LM SCALE@A4 1:20 7 Section GG: burials 1 and 2 Section HH: burials 5-10 Section II: burial 11
268175 SNNSEW
38.25m 38.5m 268158 38.5m AOD 268176 AOD AOD 268159 268003 268177 268003 268003 268173 268002 268002 268002 268174 268172 268157
Section JJ: burials 12 and 13 Section KK: burials 14 and 15 Section LL: burial 16
SN ENE WSW SN 38.5m 268114 38.5m 268145 38.5m AOD AOD AOD
268003 268115 268146 268113 268111 268147 268003 268003
268170 268002
Section MM: burial 19 Section NN: burial 20 Section OO: burial 21
SNSNSN 38.5m 38.5m 38.5m AOD 268006 AOD AOD
268007 268144 268111 268003 268003 268143 268142 268112 268110
268005 268004
Section PP
SN268161 38.5m AOD
pit/posthole 268160 01m
N Cirencester 01285 771022 Cotswold Milton Keynes 01908 218320 Archaeology w www.cotswoldarchaeology.co.uk e [email protected]
pipeline centreline PROJECT TITLE South Wales Pipeline. Site 513, Land North-West cremation urn of Steynton, Milford Haven, Pembrokeshire
cremated bone FIGURE TITLE 083 Cremation sections Site 513
FIGURE NO. Reproduced from the digital Ordnance Survey Explorer map with PROJECT NO. 9150 DATE 13-02-2014 the permission of Ordnance Survey on behalf of The Controller DRAWN BY DJB REVISION 00 of Her Majesty's Stationery Office c Crown copyright 01:5000 100m SM 911 Cotswold Archaeology Ltd 100002109 APPROVED BY LM SCALE@A4 1:20 8 9 10
9 Burials 1 and 2, looking looking west (scale 0.2m) 10 Burials 3 and 4, looking south-west (scale 0.3m) 11 Burials 5-10, looking south-east (scale 0.2m) 12 Burial 11, looking south (scale 0.2m)
11
12
N
Fig. 11 Fig. 12 Fig. 10
Fig. 9
05m1:300
Cirencester 01285 771022 Milton Keynes 01908 218320 Cotswold Andover 01264 347630 Archaeology w www.cotswoldarchaeology.co.uk e [email protected]
PROJECT TITLE South Wales Pipeline. Site 513, Land North-West of Steynton, Milford Haven, Pembrokeshire
FIGURE TITLE Photographs
PROJECT NO. 9150 DATE 13-02-2014 FIGURE NO. DRAWN BY DJB REVISION 00 APPROVED BY LM SCALE@A3 N/A 9 - 12 13 14
13 Burials 12 and 13, looking south-west (scale 0.2m) 14 Burials 14 and 15, looking south (scale 0.2m) 15 Burial 16, looking west (scale 0.1m) 16 Burial 20, looking west (scale 0.3m))
15 16
N
Fig. 14 Fig. 13
Fig. 16
Fig. 15
05m1:300
Cirencester 01285 771022 Milton Keynes 01908 218320 Cotswold Andover 01264 347630 Archaeology w www.cotswoldarchaeology.co.uk e [email protected]
PROJECT TITLE South Wales Pipeline. Site 513, Land North-West of Steynton, Milford Haven, Pembrokeshire
FIGURE TITLE Photographs
PROJECT NO. 9150 DATE 13-02-2014 FIGURE NO. DRAWN BY DJB REVISION 00 APPROVED BY LM SCALE@A3 N/A 13 - 16 Section QQ
E W 39.75m AOD
268134 268137
268135 268129 268130
268136
Section RR Section SS
SE NW NS 39.25m AOD 268101 268027 268100 268028
Section TT Section UU Section VV
EWEWNS 39.50m 39.0m 39.25m AOD AOD AOD 268105 268107 268106 268109 268104 268108
Section WW Section XX
NSSN
39.5m 39.5m 268124 AOD AOD 268117 268002 268116 268125
01m
N Cirencester 01285 771022 Cotswold Milton Keynes 01908 218320 Archaeology w www.cotswoldarchaeology.co.uk e [email protected]
PROJECT TITLE pipeline centreline South Wales Pipeline. Site 513, Land North-West of Steynton, Milford Haven, Pembrokeshire
FIGURE TITLE 083 Pit sections Site 513
FIGURE NO. Reproduced from the digital Ordnance Survey Explorer map with PROJECT NO. 9150 DATE 10-02-2014 the permission of Ordnance Survey on behalf of The Controller DRAWN BY DJB REVISION 00 of Her Majesty's Stationery Office c Crown copyright 01:5000 100m SM 911 Cotswold Archaeology Ltd 100002109 APPROVED BY LM SCALE@A4 1:20 17