south wales gas pipeline project land at conkland hill wiston...south wales gas pipeline project...

South Wales Gas Pipeline Project Site 508

Land at Conkland Hill Wiston

Pembrokeshire

Archaeological Excavation

May 2014

for

on behalf of Rhead Group

National Grid CA Project: 9150

CA Report: 13251 Event: DAT108844

South Wales Gas Pipeline Project Site 508

Archaeological Excavation

CA Project: 9150 CA Report: 13251 Event: DAT102846

prepared by Jonathan Hart, Senior Publications Officer

date 8 May 2014

checked by Karen E Walker, Post-Excavation Manager

date 17 February 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]


1

South Wales Gas Pipeline Project Site 508: Archaeological Excavation

CONTENTS

SUMMARY ..................................................................................................................... 2

1. INTRODUCTION ................................................................................................ 3

2. RESULTS (FIGS 2–7) ........................................................................................ 6

3. PROJECT TEAM ............................................................................................... 12

4. REFERENCES ................................................................................................... 13

APPENDIX A: CONTEXT DESCRIPTIONS .................................................................... 15

APPENDIX B: THE FINDS .............................................................................................. 22

APPENDIX C: THE PALAEOENVIRONMENTAL EVIDENCE BY JAMES RACKHAM .... 26

APPENDIX D: THE RADIOCARBON-DATING EVIDENCE BY SEREN GRIFFITHS ...... 58

LIST OF ILLUSTRATIONS Fig. 1 Site location plan (1:25,000)

Fig. 2 Plan of archaeological features (1:200 and 1:100)

Fig. 3 Plan showing archaeological features and geophysical survey results (1:200)

Fig. 4 Section through Ditch A (1:20) Fig. 5 Photograph: section through Ditch A, looking north-west

Fig. 6 Section and photograph of Ditch B (1:20)

Fig. 7 Photograph: central area of the site, looking east

Fig. 8 Photograph: early medieval hollow 508052, looking east

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


2


SUMMARY

Project Name: South Wales Gas Pipeline Project Location: Site 508, Land at Conkland Hill, Wiston, Pembrokeshire NGR: SN 0188 1766 Type: Excavation Date: 7 July–1 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 Cambrian Archaeological Projects 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.

Ditches and pits were identified on a slight plateau on the southern edge of Conkland Hill.

Most of the features were undated by finds and stratigraphic relationships were not securely

established. However, radiocarbon dates within the Early Bronze Age, Late Iron Age/Early

Roman and early medieval periods were obtained from charred fuelwood and cereals.

Although difficult to both phase and interpret, the site seems to have included limited Early

Bronze Age activity with more extensive Late Iron Age/Early Roman remains including

substantial ditches, potentially part of a previously unrecognised multivallate hilltop

enclosure. Within the enclosed area, an oven was recorded perhaps surrounded by a

windbreak or workshop.

The site was re-used in the early medieval period when a feature that may have been a

sunken-floored building was constructed between the earlier ditches (which had survived as

earthworks). This possible building was found in association with evidence for metal

processing and may be evidence for a rarely seen form of settlement comparable to that

found at South Hook, also in Pembrokeshire.


3


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 section of 89km, 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 CAP carried out an archaeological excavation at Site 508,

Land at Conkland Hill, Wiston, Pembrokeshire (centred on NGR: SN 0188 1766; 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 and IfA Wales

2008). The methodologies were laid out in an Archaeological Management Plan

(RSK 2006) and associated Written Statements of Investigation (WSIs) and Method

Statements.

The site 1.4 The site is located within a field on the south-western face of Conkland Hill (Fig. 1).

The hilltop reaches a height of 115m AOD and is the location of Wiston village. The

site itself lies at 95m AOD on land that falls away sharply to the south and west, with

the Fenton Brook flowing through a valley at the southern foot of the hill.


4


1.5 The underlying solid geology of the area is mapped as the Slade and Redhill

Formation (Mudstone) of the Ordovician Period; no superficial deposits are recorded

(BGS 2013).

Archaeological background

1.6 No archaeological remains were identified within the site during the preliminary

Archaeology and Heritage Survey (CA 2005). Mesolithic, Neolithic and Bronze Age

flints (PRNs 12253 and 7404) have been found 700m south of the site, in the vicinity

of Woodbarn Rath defended enclosure, an Iron Age hillfort (CA 2005, ref. ID S2).

Further hilltop sites lie in the area, including Merryborough Camp Hillfort from which

Roman and post-medieval finds have been recovered, although the camp itself is

not securely dated (CA 2005 ref. IDs 122–124; PRNs 3554, 3555 and 7414).

1.7 The site is overlooked by the remains of Wiston Castle, a motte and bailey castle

with 12th-century origins, which lies 500m north of the site and is the focus of the

village of Wiston (CA 2005, ref. ID 129; PRN 8510). The castle may be on the site of

an Iron Age enclosure (Ken Murphy pers. comm. to Jonathan Hart).

1.8 Recording during the pipeline construction identified numerous sites in the locality

(Fig. 1). These included Early Neolithic pits at Sites 245/253, 250m east of the site

(CA 2014a); a burnt mound and trough associated with a Bronze Age radiocarbon

date at Site 510, 650m east of the site (CA 2013a); undated pit with unidentifiable

burnt bone at Site 246, 520m west of the site (CA 2013b); undated ditches,

postholes and a pit at Site 247, 250m west of the site and remains possibly

associated with a prehistoric hilltop enclosure identified as a cropmark adjacent to

Site 504, 2.3km south-west of the site (CA 2013c). Site 504 also included evidence

for medieval and later agricultural practices.

1.9 Archaeological features were first identified within the site during a geophysical

survey conducted in advance of the pipeline construction works. This recorded

probable ditches (BCC 2005; Fig. 3) and the site was therefore selected for

excavation.


5


Archaeological objectives 1.10 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;

• 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 1.11 The fieldwork followed the methodology set out within the WSI (NLM 2006). An

archaeologist was present during intrusive groundworks comprising stripping of the

pipeline easement to the natural substrate (Fig. 1).

1.12 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 reports (NLM

2012a) except where the UPD recommended further/new 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.

1.13 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.

http://www.old-maps.co.uk/index.html


6


2. RESULTS (FIGS 2–7)

2.1 This section provides an overview of the excavation results; detailed summaries of

the recorded contexts, finds, environmental samples (palaeoenvironmental

evidence) and radiocarbon dates are to be found in Appendices A, B, C and D. Full,

original versions of the specialist reports are available within the archive.

2.2 The natural geological mudstone substrate was cut by ditches, pits, postholes and a

hollow. Very few dateable finds were recovered from the site; this limited evidence

was supplemented by radiocarbon dates for five features but the majority of features

lacked datable artefacts and only a small number of stratigraphic relationships were

established. As a result of this, the majority of features must be considered undated

or poorly dated. Nevertheless, most features have been assigned to periods based

on dating evidence, on spatial relationships with dated features or on the basis of

characteristic cereal assemblages but this suggested phasing should be regarded

as provisional.

Early Bronze Age

2.3 A single charred oat grain from curvilinear ditch 508066 returned a radiocarbon date

of 1890–1680 cal. BC (Beta-249350), a date within the Early Bronze Age. This ditch,

located within the central part of the site, was undated by finds but has provisionally

been assigned to the Late Iron Age/Early Roman period (see below), in which case

the Bronze Age oat would have been residual. Pit/ditch 508246, also found towards

the centre of the site, was undated but was truncated by ditch 508066 and may have

been a Bronze Age feature, and possibly the source of the Bronze Age oat.

Late Iron Age/Early Roman

2.4 A small assemblage of pottery and one radiocarbon date indicated the presence of

Late Iron Age/Early Roman activity. Other features have been assigned to this

period based on their layout in plan. The easternmost of these, Ditch A, was

curvilinear and aligned broadly north/south, corresponding to an anomaly identified

during the geophysical survey (Fig. 3). It was up to 4.85m wide and 2.15m deep with

a steep-sided profile and a narrow, flat base, but became shallower (1.2m deep) to

the south with a wider-based, more open profile (Fig. 4, section AA and Fig. 5). It

had filled naturally and contained few finds. One of the lower secondary fills

(508153) included seven joining sherds of pottery dating to the 1st to 2nd centuries


7


AD, whilst a comparable fill (508130) within another sondage produced four 1st-

century AD sherds.

2.5 Ditch B was another large curvilinear ditch. It was undated by finds but was broadly

parallel to, and 11m to the west of, Ditch A and may therefore have been

contemporary with it. Ditch B was 2.2m wide and 1.4m deep with steep sides and a

narrow flat base (Fig. 6). It contained a series of natural infills and finds were absent.

2.6 Between Ditches A and B feature 508160 was found. It comprised a short linear cut

with bowl-shaped terminals, each of which was lined with stones, and was probably

a hearth or drying oven, although no scorching to the stones or substrate was

recorded. A radiocarbon date of 170 cal. BC–30 cal. AD (Beta-249352) was returned

from charred fuelwood from this hearth/oven and part of a corroded iron object was

recovered from its backfill, but there was no evidence as to what processes had

occurred within it. It was partially enclosed by two concentric curvilinear ditches

(508066 and 508148; Figs 2 and 6). These were up to 0.4m wide and 0.35m deep

and contained up to two stony fills. Ditch 508148 had been re-cut on a slightly

different alignment by ditch 508105. As well as small quantities of charcoal, the

ditches yielded a shale spindle whorl (Fig. 10) and a deliberately notched pebble,

possibly a tally, both of which were not dateable. A number of undated postholes

may have been associated with these ditches but this could not be established for

certain since early medieval postholes/small pits were also present in the vicinity.

However, it does seem likely that the curvilinear ditches and some of the postholes

together represent the remains of a workshop partially surrounding the hearth/oven.

However, it should be noted that the Bronze Age oat discussed above came from

one of these ditches, and the possibility that they were prehistoric cannot be

excluded.

2.7 Pits 508077 and 508114 may have belonged to this period on the basis of cereal

assemblages within them (see Appendix C), although the assemblage within the

latter pit was very small. Pit 508077 was clay-lined and perhaps associated with

oven 508160 to the north, although this is not certain.

2.8 Pit 508230 was found to the west of Ditch B and was a broad, shallow, bowl-shaped

cut up to 3.4m wide and 0.35m deep. It was filled with dark silty clay 508255 which

contained frequent charcoal flecks. A rotary quernstone fragment (Fig. 9) lay flat

within the pit along with other flat-laid stone, a sherd of Late Iron Age or Early


8


Roman pottery, some burnt clay and the remains of a horse skull. The quernstone

comprised the majority of an upper rotary quernstone and is typologically Roman

(Ed McSloy, CA Principal Finds Officer, pers. comm. to Jonathan Hart). The function

of this pit is not readily apparent, although it may have been a construction cut rather

than a pit since the flat laid stones perhaps formed a work surface. However, the

presence of what was originally probably an entire horse skull (see Appendix C)

within the pit may suggest that the quern fragment, skull, flat-laid stones and pottery

were parts of a structured deposit.

Early medieval

2.9 Early medieval activity was first identified upon the return of three radiocarbon

determinations. No dateable finds of this period were recovered. One radiocarbon

date was obtained from the fill of hollow 508052. This hollow was a very regular sub-

rectangular cut with steep sides and a flat base (Figs 2 and 8). It was 0.45m deep,

5m long and at least 5m wide, although its western extent was not established. A

linear ditch/gully (not illustrated) which ran through the centre of the hollow’s base

along the long (north-east/south-west) axis of the hollow was thought by the

excavators to have been part of the hollow, although this was not established

stratigraphically.

2.10 No surface or trample deposits were identified within the hollow, and all of its fills

seemed to post-date its use. These fills were a mixture of silty clays, possibly

colluvial in origin, and several dumps of burnt material, samples from which yielded

burnt stones, featureless fired clay fragments, notable quantities of charred cereals

and pieces of slag, including a fragment from a smelting furnace base. Further slag

was recovered from posthole 508174, which included stone post-packing and which

was found within the hollow itself. A second stone-packed posthole, 508209, located

centrally within the hollow, contained small specks of iron and burnt material. Other

postholes within the hollow may also have been associated with it, although all were

undated and it is not clear how they related stratigraphically to the hollow.

Unfortunately, the metallurgical residue assemblage from the site, including the slag,

was lost prior to analysis and it was not possible to assess the nature of this

material, other than to note its presence. One of the burnt fills within the hollow

included a charred barley grain radiocarbon dated to 810–1010 cal. AD (Beta-

508045).


9


2.11 To the south of the hollow, stone-packed posthole 508193 was found. Its backfill

included charred oat grain which gave radiocarbon dates of 640–720 cal. AD and

740–770 cal. AD (Beta-249353). It was intercutting with posthole 508206, but no

relationship between these features was established. Other postholes and pits were

found truncating the Late Iron Age/Early Roman small curvilinear ditches (508178,

508180, 508238, 508240, 508081, 508191 and 508151) and Ditch B (508079,

508032, 508042 and 508166) and were probably Early Medieval on this basis. Of

these, pit 508081 was possibly clay-lined.

2.12 To the southeast of the hollow, pit 508075 truncated Ditch A. This pit had a broad,

rounded but fairly irregular profile and was 3.2m wide and 0.7m deep. It contained a

lower charcoal-rich fill (508046) which included burnt stones and was overlain by fill

508011/508012, a grey-brown silty clay deposit with frequent charcoal flecks.

Overall, this pit yielded a very large assemblage of charred cereals, characteristic of

early medieval crop processing debris, including a charred oat grain radiocarbon

dated to 690–900 cal. AD (Beta-249348). Further charred cereals likely to be from

early medieval activities came from the upper fills of Ditch A, along with a small

assemblage of unidentifiable iron fragments and it is likely that this ditch persisted as

an earthwork hollow into the early medieval period during which time it was used for

the disposal of waste.

2.13 East of Ditch A, pit 508062 contained a cereal assemblage suggestive of early

medieval dating. On site, this pit was interpreted as a corndrier, although structural

remains were absent and this interpretation is uncertain.

Later medieval/post-medieval/modern

2.14 North/south aligned Ditch C was found towards the western end of the site. Its

alignment and location correspond to those of a former field boundary which is

depicted on Ordnance Survey maps between 1889 and 2005, but which had been

removed by the time of the fieldwork.

2.15 Ditches D, E and F, at the eastern end of the site were undated by finds. They do

not conform to the current field boundary alignment, and are not depicted on the

1889 1st Edition Ordnance Survey (OS) map of the area. Interestingly though, the

alignment of Ditch F suggests that it may have been a continuation of a kink in the

southern boundary of the field within which the site is located. This kink is shown on


10


the 1st edition OS map and it is possible that this is the remnant of a medieval field

system.

Undated

2.16 A number of pits and postholes remained undated. These are most likely to have

been associated with either the Late Iron Age/Early Roman or early medieval

activity, but the possibility that other periods are represented by these cannot be

ruled out.

Discussion 2.17 Given the paucity of the dating evidence and securely established stratigraphic

relationships, interpreting the features found at Site 508 is problematic. It is clear

that at least two phases of activity pre-dating the later medieval and post-medieval

field systems are represented, one dating to the Late Iron Age/Early Roman period,

the other to the early medieval period. It is also possible that Early Bronze Age

activity had occurred. It is interesting to note that both the possible Early Bronze Age

activity and that of the early medieval period were only identified once samples had

returned radiocarbon dates. These periods are poorly represented in the Welsh

archaeological record would not otherwise have been identified, and this has wider

implications for the identification of these periods elsewhere in Wales

2.18 A phased interpretation of the features is presented below, but must be read with the

caveat that this is based on limited evidence and that the interpretations and

phasing are therefore open to re-interpretation.

Late Iron Age/Early Roman

2.19 The Late Iron Age/Early Roman pottery from Ditch A was found in small quantities

but came from well down in the ditch, from two separate sondages and represents

parts of at least four separate vessels. Even so, the dating of this ditch cannot be

regarded as fully secure and it is possible that it had earlier origins. Ditch B was

almost certainly broadly contemporaneous with Ditch A, given that the two features

appear to have been parallel. There were no clear indications within the fills to

suggest the locations of former banks, but if the hearth/oven (and the two associated

smaller shallow curvilinear features) were in use at the same time as the main

ditches were open, this would imply that any banks lay on the western edges of the

ditches which accords with the absence of prehistoric-dated features immediately

west of Ditches A and B.


11


2.20 The proximity to the ditch edge of the putative workshop containing the hearth/oven

seems strange, but it is possible that an otherwise unused space between the

ditches was actually ideal for heating activities which could be usefully kept away

from any other structures in the vicinity.

2.21 The site is located on a slight plateau on the edge of Conkland Hill, raising the

possibility that Ditches A and B were part of a multivallate hilltop enclosure but this

should be a matter for future research. Linear features were recorded to the west of

the site during the geophysical survey, although none were found during the pipeline

construction. It is also worth noting that a Roman fort was discovered just north of

Wiston after the construction of the pipeline and that perhaps the pottery from this

phase of activity is more likely to be Early Roman than Iron Age (Ken Murphy pers.

comm. to Jonathan Hart), dating which accords with the stylistically Roman

quernstone fragment.

Early medieval

2.22 The three early medieval radiocarbon dates have a broad overall range between the

mid 7th to early 11th centuries AD. It is not clear whether these dates represent

prolonged or intermittent use of the site but finding remains of this date is a

significant contribution to the archaeology of this period in Wales.

2.23 The presence of a large, flat hollow found in association with postholes is paralleled

by sunken-floored buildings discovered at South Hook, Herbranston (also in

Pembrokeshire). The structures at South Hook consisted of a number of hollows

associated with postholes, pits and furnaces (Crane and Murphy 2010, 117). The

best preserved example, ‘Complex 1’, comprised a large, well-defined sub-

rectangular hollow 10m x 4m in extent and 0.3m to 1m deep (ibid., 123). It included

evidence for sub-division into working areas, including parts used to house iron

smelting furnaces, and also contained postholes, presumed to have supported a

wooden superstructure. Other buildings were associated with domestic activity and

crop processing. A similar interpretation is possible for the example on the current

site and although the only one fragment from a smelting furnace base was found,

two of the environmental samples from the site produced quite low densities

hammerscale flakes, indicative of iron smithing being undertaken at the site,

probably in the early medieval period (Rackam, Appendix C) and the presence of

charred cereal remains and iron processing residues bears striking resemblance to

the finds from South Hook.


12


2.24 Radiocarbon dating suggested that the South Hook buildings were in use between

the late 8th to mid 12th centuries AD (Murphy et al. 2010, 185), and this too is

compatible with the early medieval radiocarbon dates from Site 508. Murphy et al.

have pointed out that this building style may have Scandinavian, Saxon and Irish

influences but that this may mask as yet unrecognised Welsh traditions and the

cultural affinities of this type of site remain to be determined (ibid. 189–190).

2.25 Based on comparison with the early medieval remains at South Hook it may be that

further early medieval buildings remain to be discovered close to Site 508. The

relationship between this early medieval and the medieval village at Wiston is not

known, but could be a subject for future research and the discovery of this possible

early medieval site is a significant finding.

3. PROJECT TEAM

Fieldwork was undertaken by Cambrian Archaeological Projects. This report was

written by 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 CAP by Kevin Blockley and the post-excavation was

managed for CA by Karen Walker.


13


4. REFERENCES

BCC (Bartlett-Clark Consultancy) 2005 Milford Haven to Aberdulais Proposed Gas Pipeline:

Archaeogeophysical Survey 2005

BGS (British Geological Survey) 2013 Geology of Britain Viewer. Online resource at

http://mapapps.bgs.ac.uk/geologyofbritain/home.html accessed 4 December 2013

Carruthers, W. 2008 ‘Assessment Report for Charred Plant Remains’, in NLM 2012a

CA (Cotswold Archaeology) 2005 Milford Haven to Aberdulais Gas Pipeline: Archaeology

and Heritage Survey. CA typescript report 04147 CA (Cotswold Archaeology) 2013 South Wales Pipeline Project. Sites 504, 249 and 250.1

Land at Merryborough Farm, Wiston, Pembrokeshire: Archaeological Excavation

and Watching Brief. CA typescript report 13179 Crane, P. and Murphy, K. 2010 ‘Early medieval settlement, iron smelting and crop

processing at South Hook, Herbranston, Pembrokeshire, 2004–05’, in Archaeologia

Cambrensis 159, 117–195 Courtney, P. and McSloy, E. 2013 Post-Roman Pottery from the Milford Haven to Aberdulais

and Felindre to Brecon Gas Pipeline, Cotswold Archaeology typescript report

Edwards, N., Lane, A. and Redknap, M. 2011 ‘Early Medieval Wales: an Updated

Framework for Archaeological research’, in A Research Framework for the

Archaeology of Wales. V 02 Online resource at

http://www.archaeoleg.org.uk/pdf/earlymed2011.pdf accessed 4 December 2013

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

Griffiths, S. 2012 ‘Assessment of Extant Radiocarbon Dates, Sites with Potential for

Radiocarbon Dating, and Initial Advice on Radiocarbon Dates’ in GA 2012

Hall, R. 2008 ‘Assessment Report for Burnt Clay’, in NLM 2012a

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

IfA (Institute for Archaeologists) 2001a Standard and Guidance for the Collection,

Documentation, Conservation and Research of Archaeological Materials. IfA,

Reading

IfA (Institute for Archaeologists) 2001b Standard and Guidance for Archaeological

Excavation

http://mapapps.bgs.ac.uk/geologyofbritain/home.html%20accessed%204%20December%20201http://www.archaeoleg.org.uk/pdf/earlymed2011.pdf


14


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

Leahy, K. 2008 ‘Assessment Report for Metal Objects’, in NLM 2012a

Murphy, K., Crane, P. and Young, T. 2010 ‘Discussion’, in Crane and Murphy 2010, 185–

190

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

RSK (RSKENSR) 2006 Milford Haven to Aberdulais Natural Gas Pipeline: Archaeological

Management Plan. Nacap Land and Marine Final, RSKENSR Environmental Ltd

Schmidl, A., Jaques, D. and Carrott, J. 2009 ‘Assessment Report for Charcoal’, in NLM

2012a

Timby, J. R. 2013 South wales Pipeline Project: Roman pottery publication summary.

Unpublished typescript report for National Grid

Wood, J. 2008 ‘Assessment Report for Faunal Remains’, in NLM 2012a

http://www.archaeoleg.org.uk/intro.html%20accessed%20December%202008


15


APPENDIX A: CONTEXT DESCRIPTIONS

Context No.

Fill of Interpretation Description L (m)

W (m)

D (m)

Spot date

508001 508002 Natural feature 508002 Natural feature 508003 Ditch C N/S aligned with u-shaped profile 1.6 0.55 508004 508003 Ditch C fill Upper fill: dark brown 1.6 0.35 508005 508003 Ditch C fill Lower fill: light brown 1.6 0.25 508006- 508008

Natural features

508009 Ditch A 45° sides, flat base, broad profile 4.7 1.2 508010 508009 Ditch A fill Upper fill of ditch: mid brown clay silt

with stones 0.65 0.3

508011 508075 Pit fill Grey-brown silty clay, frequent charcoal

3.2 0.7 690-900 Cal AD

508012 = 508011 508013 Posthole Vertical sides, flat base 0.35 0.4 508014 508013 Posthole fill Dark brown with disturbed packing

stones around edges 0.35 0.4

508015 Posthole Vertical sides, flat base 0.5 0.45 508016 508015 Posthole fill Dark brown with disturbed packing


508017 Posthole Vertical sides, flat base 0.25 0.4 508018 508017 Posthole fill Dark brown with disturbed packing


508019 Posthole Vertical sides, flat base 0.2 0.2 508020 508019 Posthole fill Upper fill: brown with small pebbles 0.2 0.1 508021 508019 Posthole fill Lower fill: brown with occasional

stones and charcoal 0.2 0.1

508022 Pit/posthole Vertical sides, flat base 0.4 0.3 508023 508022 Pit/posthole fill Upper fill: brown silty clay with

frequent stones 0.4 0.25

508024 508022 Pit/posthole fill Lower fill: light brown silty clay with frequent charcoal

0.4 0.05

508025 Posthole Vertical sides, flat base with post packing stones

0.2 0.3

508026 508025 Posthole fill Upper fill: brown silty clay with stones 0.2 0.15 508027 508025 Posthole fill Lower fill: brown silty clay 0.2 0.15 508028 508025 Posthole fill Post-packing: small pebbles packed

against cut edge 0.2 0.05

508029 Posthole Vertical sides, flat base with post packing

0.25 0.35

508030 508029 Posthole fill Upper fill: brown silty clay with stones 0.25 0.2 508031 Ditch Ditch D: 45° sides, flat base with

broad profile 1.95 0.6

508032 Pit Vertical sides, flat base 0.55 0.2 508033 508032

Pit fill Grey-brown silty clay with stones and

charcoal 0.55 0.2

508034 508029 Posthole fill Lower fill: brown silty clay with stones 0.25 0.15 508035 508031 Ditch fill Upper fill: light brown silty clay 1.95 0.15 508036 508031 Ditch fill 3rd fill: orange-brown silty clay and

frequent large stones 1.15 0.25

508037 508031 Ditch fill 2nd fill: dark brown silt 1.95 0.2 508038 508031 Ditch fill Primary fill: yellow-brown sandy silt 0.45 0.1 508039 Posthole Vertical sides, flat base 0.4 0.2


16


508040 508039 Posthole fill Upper fill: brown silty clay, occasional

pebbles and charcoal 0.4 0.1

508041 508039 Posthole fill Lower fill: brown silty clay, occasional

pebbles 0.4 0.1

508042 Pit Irregular in plan with bowl-shaped profile

0.45 0.15

508043 508042 Pit fill Red-brown clay silt 0.45 0.15 508044 = 508052 508045 508052 Fill of hollow Fill of possible working hollow: burnt

stone, burnt clay and frequent charcoal within pinkish silt

0.05 810-1010 Cal AD

508046 508075 Pit fill Dark charcoal-rich fill with scorched (red) stones

0.3 0.1

508047 Natural feature (not illustrated)

Circular in plan with gently sloping edges and slightly rounded base

1.8 0.1

508048 508047 Fill of natural feature

Upper fill: charcoal and burnt stones. Anthropogenic material collected into a natural feature

1.8 0.05

508049 508047 Fill of natural feature

Lower fill: charcoal and burnt stones. Anthropogenic material collected into a natural feature

1.8 0.05

508050 Pit Irregular in plan with rounded profile 0.55 0.25 508051 508050 Pit fill Mid brown silty clay 0.55 0.25 508052 Hollow Sub-rectangular hollow (west edge not

defined). Steep sides, flat base, with central linear hollow running through it

5.0 5.0 0.45

508053 Ditch Ditch E: broad, shallow profile 0.95 0.15 508054 508053 Ditch fill Orange-brown silt 0.95 0.15 508055 Layer Layer of burnt stone and charcoal 0.05 508056 508052 Fill Dark brown silty clay with stones and

slag pieces 0.2

508057- 508059

Contexts not used

508060 Pit Circular in plan with bowl-shaped profile

0.8 0.1

508061 508059 Pit fill Yellow-brown sandy silt with charcoal

and frequent burnt stone 0.8 0.1

508062 Pit Sub-circular in plan with bowl-shaped profile

1.0 0.2

508063 508059 Pit fill Upper fill: brown clay with frequent

charcoal and burnt stone 1.0 0.15

508064 Posthole Vertical sides, flat base 0.2 0.25 508065 508064

Posthole fill Brown silty clay, occasional charcoal 0.2 0.25

508066 Ditch Inner curvilinear ditch, with u-shaped profile

0.5 0.15

508067 508066 Ditch fill Light brown clay silt with frequent to occasional charcoal flecks

0.5 0.15 1890- 1680 Cal BC

508068 Part of 508066 508069 = 508067 508070 Ditch B Broad curvilinear ditch with steep

sides and narrow flat base. 2.2 1.4

508071 508070 Ditch B fill A middle (secondary) fill: brown silt with occasional stones

1.2 0.55

508072 508070 Ditch B fill A possible primary fill: small stones from cut edges

0.7 0.2

508073 508009 Ditch A fill Primary fill: orange-brown stony sand 3.35 0.95 508074 = 508160 1.2 0.8 0.6 508075 pit Pit truncating ditch 508009 3.2 0.7


17


508076 508052 Fill of hollow Mid brown silty clay with frequent stones

0.15

508077 Pit Sub-rectangular in plan with flat base 0.6 0.15 508078 508077

Pit fill Brown-orange silty clay with frequent

charcoal 0.6 0.15

508079 Posthole Vertical sides, rounded base and stone packing

0.4 0.25

508080 508079 Posthole fill Dark brown-grey silty clay with

packing stones around cut edges 0.4 0.25

508081 Pit Circular in plan with flat base and clay lining

0.95 0.25

508082 508077 Pit fill Dark yellow-brown silty clay with

occasional charcoal and burnt stones 0.95 0.25

508083 508077 Pit lining Yellow clay around cut edges: lining? 0.95 0.25

508084 Ditch (not located on plan)

E/W aligned with broad, shallow profile

1.95 0.3

508085 508084 Ditch fill Upper fill: dark brown silty fill with frequent stones

1.95 0.25

508086 508084 Ditch fill Lower fill: dark brown silty fill with occasional stones

1.95 0.25

508087 Part of 508085 508088 = 508086 508089 Layer Burnt stones and charcoal 0.25 0.05 508090 = 508089 508091 Ditch B Part of 508070 4.5 1.45 508092 508091 Ditch B fill Upper fill: dark grey clay 508093 = 508055 508094 Ditch B Part of 508070 1.7 0.65 508095 508094 Ditch B fill Upper fill: brown silt with stones 1.7 0.45 508096 508094 Ditch B fill Lower fill: brown silt with stones 1.4 0.2 508097 508070 Ditch B fill Upper fill: brown silty fill 2.2 0.75 508098 508070 Ditch B fill A lower fill: red-brown clay 1.0 0.25 508099 508070 Ditch B fill A lower fill: large stones (collapsed

bank?) 0.5 0.2

508100 508070 Ditch B fill Primary fill: pale brown silty clay 0.75 0.1 508101 508070 Ditch B fill A lower fill: charcoal 0.4 0.05 508102 508070 Ditch B fill A lower fill: red clay 1.1 0.1 508103 Ditch Ditch F 508104 Ditch fill Mid brown sandy silt 508105 Ditch re-cut Outer curvilinear ditch (re-cut of

508148); u-shaped profile 0.85 0.25

508106 508105 Ditch re-cut fill Mid yellow-brown sandy silt with occasional stones and charcoal

0.85 0.25

508107 Ditch A Part of curvilinear ditch, 45° sides, narrow, flat base

4.85 2.15

508108 508107 Ditch A fill Upper fill: yellow-brown clay silt with a charcoal-rich lens

4.85 0.55

508109 508062 Pit fill Lower fill: brown with charcoal flecks 0.7 0.1 508110- 508113

Natural features

508114 Pit Circular in plan with bowl-shaped profile

0.65 0.15

508115 508114 Pit fill Mid brown silty clay with frequent

charcoal flecks 0.65 0.15

508116 Context not used 508117 508091 Ditch B fill Brown-grey silty clay 3.3 0.3 508118 508127 Ditch B fill Brown clay with stones 1.2 0.6


18


508119 508091 Ditch B fill Brown-orange clay 1.2 0.35 508120 508091 Ditch B fill Light brown clay with stones 1.45 0.15 508121 508091 Ditch B fill Brown silt with pebbles 1.1 0.25 508122 508091 Ditch B fill Orange-brown sandy clay with stones 1.7 0.15 508123 508091 Ditch B fill Brown clay with stones 1.1 0.2 508124 508091 Ditch B fill Orange sandy clay with stones 1.05 0.1 508125 508091 Ditch B fill Brown-orange clay with stones 0.9 0.25 508126 508091 Ditch B fill Brown-orange clay with stones 1.5 0.45 508127 Ditch B Part of 508091 1.15 0.55 508128 Ditch B Part of 508091 2.55 0.3 508129 Ditch A Part of curvilinear ditch, 45° sides,

narrow, flat base 4.6 1.8

508130 508129 Ditch A fill 2nd fill: light orange-brown sandy silt with pebbles

1.5 0.45 C1 AD

508131 Oven Part of 508160: SW bowl terminal 0.55 0.35 508132 508131

Oven fill Dark brown clay silt with lining stones

around cut edges 1.5 0.55 0.35

508133 Context not used 508134 508107 Ditch A fill Light brown silty clay 4.0 0.5 508135 508136

Posthole fill Mid brown silty clay with packing

stones 0.35 0.25


0.35 0.25

508137 508129 Ditch A fill Upper fill: grey-brown clay silt 1.1 0.2 508138 508129 Ditch A fill A middle fill: yellow-brown silt with

stones 4.6 0.55

508139 508129 Ditch A fill A middle fill: dark brown silt 1.95 0.5 508140 508129 Ditch A fill A middle fill: yellow-brown sandy

pebble fill 1.0 0.15

508141 508129 Ditch A fill A middle fill: yellow-brown sandy pebble fill

1.1 0.2

508142 508129 Ditch A fill A middle fill: yellow-brown silt 2.2 0.3 508143 508129 Ditch A fill Sandy silt with stones 1.45 0.25 508144 508129 Ditch A fill 3rd fill: lens of pebbles slumping in

from east side (bank?) 1.45 0.2

508145 508129 Ditch A fill Lower fill: light orange-brown sandy silt with frequent angular stones

2.3 0.3

508146 = 508052 508147 508052 Fill of hollow Dark brown with charcoal flecks 4.0 2.0 0.1 508148 Ditch Outer curvilinear ditch: steep u-

shaped profile 0.4 0.35

508149 508148 Ditch fill Upper fill: dark yellow brown silty clay with stones and occasional charcoal

0.25 0.15

508150 508148 Ditch fill Lower fill: dark yellow brown silt with stones

0.35 0.2


0.4 0.1

508152 508151 Posthole fill Dark brown clay silt with packing

stones 0.4 0.1

508153 508107 Ditch A fill 2nd fill: yellow-brown silt with stones 1.5 0.35 C1-2 AD 508154 Ditch

(not located on plan)

N/S aligned with irregular sides and flat base

4.2 0.65

508155 508154 Ditch fill Red-brown clay with stones 4.2 0.65 508156 508146 Fill of hollow Light brown silt with charcoal 5.5 1.5 0.1 508157 508146 Fill of hollow Spread of charcoal, burnt clay and

burnt stone 2.8 1.2 0.25

508158 = 508131


19


508159 508158 Oven fill Upper fill: orange-brown silty clay with occasional charcoal

0.55 0.15

508160 Oven Linear flue with bowl-shaped pit terminals, terminals stone-lined

508161 508160 Oven fill Light brown sandy silt with charcoal 0.5 0.3 170 Cal BC-30 Cal AD

508162 Part of 508160 Vertical sides, rounded base and stone packing

0.6 0.25

508163 508162 Mid brown sandy clay with packing

stones and occasional charcoal 0.6 0.25


0.5 0.25

508165 508164 Posthole fill Mid brown with packing stones 0.5 0.25 508166 Pit Irregular sub-circular in plan with

bowl-shaped profile 0.8 0.2

508167 508166 Pit fill Mid brown silt with stones 0.8 0.2 508168 Ditch E/W aligned with irregular, steep sides

and flat base 1.85 0.3

508169 508168 Ditch fill Dark brown with stones 1.85 0.3 508170 Part of 508154 (re-cut of 508186) 508171 Ditch N/S aligned with irregular, steep sides

and flat base 2.3 0.3

508172 508171 Ditch fill Brown-orange clay 2.3 0.3 508173 Layer Mid brown silty clay layer found

across large areas of site


0.3 0.15

508175 508174 Posthole fill Orange-brown clay with packing

stones 0.3 0.15

508176 Posthole Vertical sides, flat base and stone packing

0.55 0.3

508177 508176 Posthole fill Mid brown clay silt with packing

stones around cut edges 0.55 0.3

508178 Posthole Vertical sides, flat base and stone packing

0.2 01

508179 508178 Posthole fill Grey clay silt 0.2 0.1 508180 Posthole Vertical sides, rounded base 0.3 015 508181 508180

Posthole fill Mid brown silt with occasional small

stones 0.3 0.15

508182 508170 Ditch fill Upper fill of ditch: brown-grey clay with occasional charcoal and frequent stones

2.8 0.75

508183 508170 Ditch fill Lower fill of ditch: red-brown silt with occasional charcoal and frequent stones

508184 508186 Ditch fill Upper fill: dark brown silty clay with stones

1.5 0.2

508185 508186 Ditch fill Lower fill: red-brown silty clay with stones

1.5 0.2

508186 Ditch Terminal of ditch, steep sides and flat base. Re-cut by 508170

1.5 0.4

508187 Ditch N/S aligned with u-shaped profile 1.6 0.25 508188 508187 Ditch fill Dark clay with charcoal 1.6 0.25 508189 Posthole Vertical sides, rounded base 0.25 0.15 508190 508189 Posthole fill Dark brown clay silt 0.25 0.15 508191 Posthole Vertical sides, rounded base 0.05 0.05 508192 508191 Posthole fill Dark yellow-brown silt with occasional

charcoal 0.05 0.05

508193 Posthole Vertical sides, flat base and packing stones

0.4 0.4


20


508194 508193 Posthole fill Dark grey-brown silt with occasional charcoal and with packing stones

0.4 0.4 640-720 cal AD & 740-770 cal AD

508195 Oven Part of 508160 0.25 0.05 508196 508195 Oven fill Light brown silty clay 0.25 0.05 508197 Pit? Oval in plan with irregular profile;

possibly a natural feature 0.6 0.3 0.1

508198 508197 Pit? fill Mid brown silty clay, rare charcoal 0.6 0.3 0.1 508199 508107 Ditch A fill Dark brown silt 2.55 0.45 508200 508107 Ditch A fill Small stones in sandy matrix 1.0 0.2 508201 508107 Ditch A fill Small stones in sandy matrix 1.85 0.25 508202 508107 Ditch A fill Small stones in dark brown silt 0.7 0.2 508203 508107 Ditch A fill Small stones in sandy matrix 0.8 0.15 508204 508107 Ditch A fill Lower fill: pebbles in sandy matrix 1.5 0.65 508205 508158 Oven fill Lower fill: dark brown-black sandy silt

with charcoal lumps and unspecified burnt material

0.45 0.25

508206 Posthole Vertical sides, flat base, packing stones

0.25 0.15

508207 508206 Posthole fill Light yellow-brown silty clay with packing stones

0.25 0.15

508208 508052 Fill of hollow mid brown silt with charcoal lumps 0.4 0.2 0.1 508209 Posthole Vertical sides, flat base, packing

stones 0.4 0.2

508210 508209 Posthole fill Upper fill: brown clay with occasional charcoal

0.4 0.2

508211 508209 Posthole fill Post pipe: orange-red clay with occasional charcoal

0.1 0.1

508212 508209 Posthole fill Lower fill: small stones and frequent charcoal

0.2 0.1

508213 Pit? Circular in plan but slightly irregular, step sides, rounded base. Possibly a natural feature

0.9 0.25

508214 Part of 508209 (possibly a ‘cut’ made during post-removal)

508215 508214 Stone-rich fill: possibly disturbed packing material

508216 508213 Pit? fill Dark orange-brown sandy silt with occasional charcoal

0.9 0.25

508217 Context not used 508218 Context not used 508219 sondage 508220 Context not used 508221 Layer Colluvium overlying 508056; extent

not recorded 0.05

508222 = 508056 508223 Posthole Vertical sides, flat base, packing

stones 0.25 0.2

508224 508223 Posthole fill Grey with charcoal 0.25 0.2 508225- 508227

Context not used


0.35 0.2

508229 508228 Posthole fill Silt with small stones and occasional charcoal

0.35 0.2

508230 Pit Circular in plan with shallow bowl-shaped profile; contained quernstone

3.4 0.35

508231 Possible pit/posthole

Only partially exposed 0.15


21


508232 508231 Possible pit/posthole fill

Brown silt with burnt clay and charcoal

0.15


Only partially exposed


Brown silt

508235 Layer Burnt clay and charcoal, only partially exposed; extent not known

0.15


Only partially exposed 0.1


Silt with possible packing stones around cut edge

0.1


0.6 0.4

508239 508238 Posthole fill Light brown-grey silty clay with charcoal flecks with post-packing around cut edges

0.6 0.4

508240 Posthole Vertical sides, pointed base (stakehole?)

0.05 0.1

508241 508240 Posthole fill Mid brown silty clay 0.05 0.1 508242 Posthole Vertical sides, flat base 0.25 0.2 508243 508242 Posthole fill Dark brown sandy clay 0.25 0.2 508244 Posthole Vertical sides, flat base, packing

stones 0.45

0.4

508245 508244 Posthole fill Dark brown sandy clay with packing stones around cut edges

0.45

0.4

508246 Oval? pit Truncated at both ends by curvilinear ditches

>1 0.45 0.05

508247 508246 Fill Flat stones within compact sandy clay matrix

>1 0.45 0.05

508248 Posthole Vertical sides, flat base 0.35 0.35 508249 508248 Posthole fill Dark brown silt 0.35

0.35

508250 Pit/fill Circular in plan with steep sides and flat base; fill not recorded

508251 Pit Circular in plan with irregular broadly bowl-shaped profile

0.85 0.2

508252 508251 Pit fill Dark brown sandy silt with stones and charcoal

0.85 0.2

508253 = 508255 508254 = 508230 508255 508230 Pit fill Dark grey-brown to black silty clay

with charcoal flecks; flat stones, including a quernstone fragment, laid flat within fill

LIA/ERB

508256 Context not used 508257 508219 sondage fill Light brown sandy silt with stones 2.0 0.15 508258 508219 sondage fill Dark orange-brown silty clay 0.3 0.3 508259 508219 sondage fill Dark red clay with occasional charcoal 0.6 0.4 508260 508219 sondage fill Red-brown silty clay 1.35 0.3 508261 508219 sondage fill Dark orange-brown silty clay with

stones 0.85 0.2

508262 Context not used 0.55 0.4


22


APPENDIX B: THE FINDS

Roman pottery (Timby 2013)

The site produced an assemblage of 14 sherds (52g) of pottery, representing a minimum of four vessels. In

addition there are two fragments (57.5g) of fired clay of indeterminate date. The assemblage is essentially a

‘native’ one but apparently largely, or completely, composed of vessels imported from outside the region.

Catalogue

Context 508130: four sherds from a handmade grog-tempered jar. The exterior was burnished and the interior

was sooted. Provisionally this is dated to the 1st century AD. Late Iron Age/Early Roman grog-tempered pottery

is quite rare in Wales.

Context 508153: seven joining sherds from a small bowl with a burnished lattice decoration. The fabric appears

to contain an igneous or metamorphic rock temper suggesting a Malvernian source. Provisionally this vessel is

dated to the 1st or 2nd centuries AD.

Context 508255: one black handmade ware sherd (BW1) containing quartz sand and slate inclusions which may

be Late Iron Age or Early Roman and could represent a Malvernian fabric variant.

Unstratified: One black sandy ware bodysherd, which may be Late Iron Age or early Roman.

Post-Roman pottery (Courtney and McSloy 2013)

A single sherd of medieval pottery (12g) was recorded from the subsoil. The fabric is identifiable as type DGTU

(Dyfed Gravel-tempered unglazed ware), the most common medieval coarseware recorded from the pipeline

excavations, and are broadly dateable across the late 12th to 15th centuries.

Burnt clay (Hall 2008)

Fired clay fragments were recovered from contexts 508045 (331 fragments; 3318g) and 508157 (48 fragments;

2228g). All were featureless

Stone objects (Vince and Steane 2008) Context 508067

A pebble with deliberate notching on one side. The notches have a very similar appearance to wooden tally sticks

and it is possible that the object had a similar function.


23


Context 508106

A spindle whorl made from shale; 35mm in diameter and between 5 and 6mm thick. It has a central cylindrical

hole 4mm in diameter.


24


Context 508230

The top stone of a rotary quern. The stone is made from Millstone Grit and is 500mm in diameter, 130mm high,

with a central conical hopper narrowing from 130mm at the top to 90mm at the bottom.


25


Metal objects (Leahy 2008)

Context Description L (mm)

W (mm)

D (mm)

Mass (g)

Condition I.D

508004 10 small pieces of iron. No features - - - - Corroded and exfoliating

Not known

508056 Fragment of a slag block, apparently dish-shaped, about one quarter now present. Vitreous, black, vesicular, magnetic

105.7 92.2 49.5 604.6 Good Slag block from base of smelting furnace

Lump of slag, hard vesicular and containing some pieces of haematite ore

- - - - Good Slag

Four small lumps of slag, hard vesicular

- - - 7.9 Good Slag

508134 9 iron objects, all covered with corrosion and showing no features

44.6 28.2 16.3 25.2 Good Slag 38.5 24.6 14.4 17.7 19.4 19.4 13.5 7.9 28.5 20.4 16.9 9.5 14.2 17.4 10.5 3.1 17.2 12.2 7.9 2.1 14.7 11.3 9.5 1.5 14.2 13.9 7.4 1.5 14.5 7.8 9.0 1.7

508147 Two pieces of slag, hard vesicular - - - 89.5 Good Slag Piece of iron concretion 23.8 18.6 11.2 3.5 Corroded

and exfoliating

Not known

Piece of corroded iron 15.7 8.9 7.9 1.4 Corroded and exfoliating

Not known

508163 Small pieces of iron corrosion, once part of a larger object

15.9 12.2 10.7 0.8 Corroded and exfoliating

Not known

508165 Find Number 508164: fragment of corroded iron, no detail can be seen

29.3 21.9 9.0 8.3 Corroded and exfoliating

Not known

508175 Find Number 508174: six pieces of slag, hard vesicular

- - - 1.3 Good Slag

508212 Tiny specks of iron corrosion products

- - -


26


APPENDIX C: THE PALAEOENVIRONMENTAL EVIDENCE BY JAMES RACKHAM

Bone

Bone was recovered from five contexts during the hand excavation of the deposits, and a few further fragments

were recovered from three of the environmental samples. A single piece of burnt bone from a sheep sized animal

(0.8g), subsequently broken into four, was recovered during hand excavation from context 249006, the fill of pit

249007. It could not be identified further. A small fragment of indeterminate burnt bone (0.5g) was recovered by

hand from context 508194. A tiny fragment of burnt bone (


27


Table 1. Bulk environmental samples from Site 508

sample

context

Feature

description Processed wt kg Processed vol l

5083001 508014 508013 posthole fill 15

5083002 508016 508015 posthole fill 15

5083003 508018 508017 posthole fill 15

5083004 508020 508019 posthole fill


28


sample

context

Feature

description Processed wt kg Processed vol l

5083035 508092 508091 ditch B upper fill 22

5083036 508080 508079 posthole fill 4 7.5

5083037 508106 508105 ditch re-cut of 508148 12 15

5083038 508109 508062 pit lower fill – corn dryer? 10 30

5083039 508108 508107 ditch A upper fill 2 3

5083040 508111 508110 natural feature 2

5083041 508113 508112 natural feature 2

5083042 508115 508114 pit fill 10 10

5083043 508130 508129 ditch A 2nd fill 10 15

5083044 508132 508131 oven fill – oven 508160 12 30

5083045 508167 508166 pit fill 6 22

5083046 508165 508164 posthole fill 6.5 7

5083047 508135/1 508136 posthole fill 10 10

5083048 508175 508174 posthole fill 3

5083049 508152 508151 posthole fill 15

5083050 508177 508176 posthole fill 4 7.5

5083051 508163 508162 oven 15

5083152 508205 508162 posthole fill 4.5 10

5083152 508161 508160 Oven fill 4.5

5083153 508194 508193 posthole fill 9.5 15

5083154 508210 508209 posthole fill 2 2

5083155 508211 508209 posthole fill 1.5? 1

5083156 508212 508209 posthole fill 2

5083157 508212 508209 posthole fill 2

5083158 508215 508214 posthole fill 2

5083159 508216 508213 pit? fill 12 7

5083160 508224 508223 posthole fill 1 3

5083161 508239 508238 posthole fill 10 15

5083162 508229 508228 posthole fill 2

5083163 508232 508231 Possible pit/posthole fill 1

5083164 508235 508231 Possible pit/posthole fill 1 2

5083165 508234 508233 possible pit/posthole


29


A number of the features have been dated by ceramic evidence and radiocarbon dates to the late Iron Age/early

Roman period (LIA/ER), and a number of undated features are also assigned to this phase (marked ‘LIA/ER?’ in

Table 1) on the basis of their plan and associations. Four radiocarbon dates indicate the presence of early

medieval activity, although no dateable finds were recovered. These indicate a large spread of material in hollow

508052 is probably early medieval and on the basis of the dates from pit 508075, and postholes 508025 and

508193, stratigraphic relationships and cereal assemblages several other features have been assigned to the

early medieval period (Table 1- marked ‘emed?’). Sample 5083021 derives from a ditch A fill but the charred

cereal assemblage is dominated by oats, and a barley grain is dated to the 8-10th C AD, and it fits better with the

assemblages from pit 508075 and other upper ditch fills which are of medieval date.

In fact the two upper fills of ditches A and B, 508108 and 508092, may also include medieval material since both

these samples have more oats than other cereals. A number of other features, particularly postholes and pits, are

undated, and although these are probably related to the LIA/ER and early Medieval phases (particularly the oat

rich samples), they could relate to other episodes of activity. If the cereal assemblages are a reasonable guide to

date, then the majority of the sampled deposits are post-Roman (see Tables 1 and 2), although one or two

deposits (eg sample 508021) have very high concentrations of charred oat grain and this material could have

contaminated earlier deposits as a result of soil processes and worm action.

The samples were processed in the manner described in the assessment reports (Carruthers 2008) with the

additional refloating of the dried sample residues that were located whose flot volumes are indicated in Table 2

as ‘2nd flot’. The whole of the residues for the excavation samples was refloated. The unsorted fraction of the 1st

flot and the 2nd flots were then sorted for charred macrofossils and the residues re-dried and checked with a

magnet to recover any magnetic material, burnt stone, etc. Very few of the residues were located so only a small

number (10 samples) were refloated, and the original processing records for thirty of the samples were not

available so there is no information as to any finds from these samples. In fact only those samples whose

residues were located and refloated can be assumed to have an adequate record. In the absence of finds data

from the samples interpretation is constrained, and limited to the environmental evidence.

The records are poor but a little burnt stone has been recorded, a magnetic fraction and occasional burnt bone

and fired earth, but many samples may have contained finds, but in the absence of the residues it has not been

possible to check. Of the fourteen samples whose residues were located and produced a magnetic fraction two

included hammerscale flakes, with sample 508027 producing 17 flakes from 117 kg of deposit. These represent

quite low densities but do indicate iron smithing being undertaken at the site, probably in the early medieval

period. It is likely that if the residues had been retained from the other samples several more samples might have

produced hammerscale.


30


Fig. 1.

Plan

of

Site

Site 508 with the location of the environmental samples marked.


31


The detailed analysis of the plant macrofossils has focussed on the richer samples (18 samples), but the data

from the assessment (Carruthers 2008) has been checked and included on Table 2 and is utilised in the

discussion. Thirteen charcoal assemblages were assessed (Schmidl et al 2009; data is summarised in Table 2),

four probably of late Iron Age/early Roman date and the remainder probably early medieval (8-10th C AD). The

four phased to the LIA/ERB period produced alder/hazel, Maloideae, poplar/willow, elm and oak with the latter

the most abundant. The nine samples assigned to the early medieval phase produced Maloideae, alder/hazel,

poplar/willow, ash/oak and oak. Stemwood of alder/hazel and oak were frequent in most of the samples (Schmidl

et al 2009). For the post-excavation charcoal analysis only samples from the LIA/ERB oven fill, context 508132,

and the radiocarbon dated 9-10th century AD roundwood assemblage from hollow fill 508045, were studied (see

below).


32


Table 2. Data for the environmental samples from Site 508 – archaeological finds recorded from the samples

Samples marked with a ‘y’ under ‘rec.’ have a processing record, processing records for the others have not been found; Plant and wood data taken from assessment reports.

Volumes listed under ‘processed vol’ were taken from the site sample sheet and are therefore unlikely to be accurate (not all the sample was always processed and tubs were

noted as 15 litres where generally less than this was sampled); residue weights were not generally accurately recorded – only those marked with an asterisk are

accurate.

sample no

context no

processed wt kg

processed vol l

1st flot

2nd

flot

residue

Fired earth

burnt stone

flint Mag netic wt

Hammer -scale

Burnt bone

comments rec.

LIA/ERB with residual EBA?

5083030 508067 15 15 6 2 5797* 0.6 Barleyx3; oatx2; indet grain x3 y

Late Iron Age/early Roman

5083001 508014 15 8 nr Barleyx3; oatx1; indet cerealx1

5083002 508016 15 2 nr poor oat x2; barley frag x 2

5083003 508018 15


33


sample no

context no

processed wt kg

processed vol l

1st flot

2nd

flot

residue

Fired earth

burnt stone

flint Mag netic wt

Hammer -scale

Burnt bone

comments rec.

5083043 508130 10 15 1 3 5500 515


34


sample no

context no

processed wt kg

processed vol l

1st flot

2nd

flot

residue

Fired earth

burnt stone

flint Mag netic wt

Hammer -scale

Burnt bone

comments rec.

roundwood++

5083025 508061 11 15 15 nr 3000 +in flot See Table 3; alder/hazel rw, oak sw y

5083027 508045 117.5 130 1500 26 46250* + 55 17 fl Slag; see Table 3; HNSx2 y

5083028 508065 0.5 2 nr Oatx1; hulled barleyx1

5083036 508080 4 7.5 3 1 1788 27 hulled barleyx3; cf. bread wheatx1; NFI

cerealx1

y

5083039 508108 2 3 16 1 514 7


35


sample no

context no

processed wt kg

processed vol l

1st flot

2nd

flot

residue

Fired earth

burnt stone

flint Mag netic wt

Hammer -scale

Burnt bone

comments rec.

5083158 508215 2


36


sample no

context no

processed wt kg

processed vol l

1st flot

2nd

flot

residue

Fired earth

burnt stone

flint Mag netic wt

Hammer -scale

Burnt bone

comments rec.

5083026 508063 18 25 255 nr nd See Table 3; cf peax1 y

5083038 508109/06

3

?

10 30 51 5 3500


37


Charred Plant remains (Wendy Carruthers)

Eighteen samples from thirteen features have been fully analysed for this report. These included four fills of ditch

A (LIA/ERB and early medieval), one of ditch B (probably LIA/ERB), fills of four ?early medieval and undated pits,

an early medieval hollow, an early medieval burnt stone area, an undated oven fill and four undated postholes.

Methods

Sorting was carried out using an Olympus SZX 7 stereoscopic microscope. Flots were first separated into three

fractions to facilitate sorting (sieve sizes were 3mm, 1mm and 250 micron meshes). Modern material was

removed prior to measuring the flot volume, although for this site only small quantities of modern rootlets and the

occasional uncharred seed were present. Two flots were extremely large and productive (samples 5083021

(second bag) and 5083027) so these were sub-sampled into manageable quantities using a riffle box, depending

on the richness of the flot. The subsample sizes used were 20% and 12.5% respectively. For sample 5083021 it

was noticed that the two bags of flot recovered from this one context (508046) were very different in

concentration, and to some extent different in character (see sample description below). There seem to have

been no problems with labelling, and they did not appear to have been first and second flots (not all samples from

this site had been double floated). It is likely that the two bags demonstrate the natural variability within the

deposit, i.e. context 508046 had consisted of several slightly different dumps of burnt waste. In order to

investigate this further both bags were analysed separately, described as bag 1 and bag 2. Table 3 presents the

results of the analysis. Stace (2010) and Zohary and Hopf (2000) were used for nomenclature, the latter being

used for cereal taxonomy. It should be noted that some changes to the identification and interpretation may be

made once all of the Milford Haven site data has been gathered together.

Results

In order to compare the samples in more detail and to see if the oat/barley maslin ‘dredge’ had been grown, oat :

barley ratios were calculated for the richer samples. In addition, where well-preserved oat grains were found to

display the characteristics of either common oat (Avena sativa) or bristle oat (Avena strigosa) grain length

measurements were taken, up to a maximum of 100 grains wherever possible (see Figure 2). When coupled with

identifiable features such as grain shape and the presence of at least some identifiable chaff fragments it is

sometimes possible to confirm identifications to species level. Unfortunately very little chaff was recovered from

Site 508 (7 complete florets of Avena sativa from sample 5083021, ditch A) so it was only possible to confirm the

presence of common oat. The frequency of small, slender oats in some samples, e.g. sample 5083026 (emed?

pit), suggested that wild oats (Avena fatua), bristle oats (A. strigosa) or under-developed secondary florets of

common oat were likely to have been present, but this could not be confirmed due to the lack of chaff. The

absence of oat awn fragments from many of the Milford Haven samples is surprising, considering the high

frequency of oat. Bristle oat in particular is heavily awned, with both primary and secondary florets possessing

thick, geniculate (with ‘knees’) awns. Some varieties of common oat are awnless and the secondary florets of

common oat are always awnless, so quantities of awns in archaeobotanical samples can be indicative of the

species present. However, there are other possible reasons for the lack of charred awn fragments; it could be

due to total combustion under oxidised burning conditions, preservation factors such as mineral encrustation, or

processing methods (perhaps washing these small items through the fine sieves after floatation with a strong jet

of water).


38


Discussion

a) Frequency of charred plant remains and state of preservation – Some of the samples from the site

were extremely productive in terms of cereal grain concentrations, in particular those containing oat and barley

grains. However, the grains were often very poorly preserved and, like many of the Milford Haven pipeline sites,

silt encrustation made identification difficult. As with other sites along this pipeline, chaff fragments and weed

seeds were scarce. Whilst this is common for post-Roman domestic waste deposits (because free-threshing

cereals were usually the dominant grain for human consumption, so processing would have been carried out en-

mass in covered barns, away from settlement features) it is less so for Late Iron Age/Early Romano-British

contexts. In the LIA/ERB period hulled wheats (emmer and spelt) dominate assemblages across Britain, so low

levels of day-to-day de-husking waste (comprising mainly of emmer/spelt chaff and weed seeds) are commonly

found in ditch fills, pits and postholes. As discussed below, concentrations of dehusking waste appear to be much

lower on Welsh sites of this period, and this needs to be investigated further. Because less is known about

IA/native RB sites in Wales it is possible that the large quantities of oats and barley in some of the Milford Haven

pipeline samples (e.g. sample 5083026) represent the main cereals being consumed on LIA/native RB sites. This

would explain why so little emmer/spelt grain and chaff has been recovered from the probable late Iron Age/early

Roman samples (Table 3a) and undated samples possibly of this age (Table 3b). In total, twelve possible and

definite emmer/spelt (Triticum dicoccum/spelta) grains and seven emmer/spelt and spelt (T. spelta ; sample

5083033, 3 good glume bases) chaff fragments were recovered from eight samples, four of which were undated,

one of which was thought to be LIA/ERB (sample 5083044), two of which had oat (Avena sp.) and barley

(Hordeum sp.) grains dated to 8-10th centuries AD (samples 5083013 and 5083021) and one of which has been

placed in the early medieval phase although it is undated (sample 5083033). However, there are other

possibilities that need to be ruled out before this proposition can be accepted, particularly since the site has been

occupied over several periods so contamination is a distinct possibility.

b) Sample descriptions

i) LIA/ERB Ditch B fill (sample 5083035, context 508092, ditch 508091) – The 22 litre soil sample came

from an upper charcoal-rich fill of a secondary recut of ditch B. The small assemblage (Table 3a: 15 items, 0.7

fragments per litre (fpl)) consisted of three barley grains (including a twisted grain confirming the cultivation of

hulled six-row barley Hordeum vulgare subsp. vulgare), nine poorly preserved oat grains (Avena sp.), a possible

poppy seed (cf. Papaver sp.) and a crowberry seed (Empetrum nigrum). Crowberry, a member of the heather

family (Ericaceae), grows in similar places to heathers, i.e. peaty and rocky moors, bogs and mountain tops

(Stace 2010). Being a dwarf, evergreen shrub that bears berries it has many uses, including for food, dye,

medicinal (berries, leaves, stems and roots), fuel, fodder and packaging. The presence of this distinctive seed

suggests that heathland/moors/bogs existed in the area during the LIA/ERB (presuming that the small charred

seed is not intrusive). There are no obvious indications that this assemblage is LIA/ERB in date in that no hulled

wheat remains were present. In view of the presence of oats and barley and the deposit is an upper fill it is

possible that the charred remains date to the early medieval phase of occupation in the 7th – 10th centuries AD.

ii) C1st-C2nd BC / LIA/ERB / early medieval Ditch A fills (508107, 508009, 508075):

Ditch A is thought to have been a LIA/ERB ditch (samples 5083039 and 5083021) with an early medieval feature

(samples 5083006, 5083013 and 5083015) cut into the top. Four of the samples came from a section towards the

southern margin of the trench whilst sample 5083039 came from a section towards the northern side. During the

assessment phase an oat grain (Avena sp.) produced a radiocarbon date of 690-950 cal AD (sample 5083013)

and an oat from sample 5083006 produced a date of 690-900 cal AD. An additional radiocarbon date on barley


39


grain from sample 5083021 from an upper fill (508046) of the ditch has produced a 770-980 cal AD date

contrasting with the 1st-2nd century AD date indicated by pottery from a lower secondary fill of the ditch (Hart

2014). This suggests that the upper fill (508046) is broadly contemporary with pit 508075, the single grain of

hulled wheat in this sample (Table 3c) perhaps being derived from the earlier phase of activity, and that the upper

fill in the northern section (508108) may equally include early medieval material.

Sample 5083039, context 508108, ditch section 508107 – Although the ditch is dated to the late Iron Age/early

Roman period the sampled deposit is an upper fill and may be equated with upper fill 508046 in the southern

section which has produced an early medieval radiocarbon date. This sample was fairly unproductive (Table 3a),

although only 3 litres of soil was processed (16 fpl). The only remains present were 18 barley grains (Hordeum

vulgare), 15 oat grains (Avena sp.), some indeterminate poorly preserved cereal fragments and a Persicaria

maculosa/lapathifolia seed.

770-980 cal AD, sample 5083021, context 508046, ditch section 508009 – This sample was the most productive

one from the site (267.9 fpl averaged for 2 bags), particularly the second bag of flot which was over 17 times as

productive as bag 1. It is uncertain if the two bags had been taken from different areas of the deposit as no

differentiating information was recorded. Oats dominated the assemblages with common oat confirmed (Avena

sativa) due to the presence of 7 complete oat florets in bag 2. There were no obvious signs of awns on the dorsal

side of the oat lemmas, so it is possible that this was an awnless oat variety, or all of the florets were secondary

ones. This would be advantageous where oats were being used primarily for human consumption, as small

fragments of awn can be irritants. The ratios of oats to barley were 8:1 and 21:1 for bags 1 and 2, averaging out

at 18:1. A single hulled wheat grain (Triticum dicoccum/spelta) was the only item that suggested that the context

may have been prehistoric in date contrasting with the early medieval radiocarbon date. However, it should be

noted that the presence of occasional hulled cereal remains in some later samples demonstrates that residuality

does occur on the site. Two rye grains identified during the assessment (Carruthers 2008) help to confirm the

medieval date.

Weed seeds were more diverse and frequent in Bag 2 than in other samples from this site, but they still only

amounted to about 5% of the items. The ratio of 276:1:15 grain to chaff to weed seeds suggests that processed

grain with traces of contaminants had been deposited, or small amounts of processing waste were mixed in with

clean grain. As with two other ditch A samples, wild radish (Raphanus raphanistrum) capsule segments or

‘mericarps’ were present. Two undated samples that also contained this species were both from posthole 508228

in area 508052, an early medieval spread of heated stone and charcoal. It is likely that the burnt waste deposited

in the top of ditch A had derived from activities taking place in area 508052. The small number of possible gorse

spine tips (Ulex sp.) in this sample and also in undated pit fill sample 5083026 indicate that rough grazing or

heathy areas probably existed nearby, providing a source of fuel. This fits in with the recovery of a crowberry

seed from ditch B.

690-900 cal AD - Sample 5083006, context 508011, pit 508075 – This oat-rich sample from a pit cutting the

upper fills of ditch A produced a concentration of 114 fpl of identifiable charred plant remains (Table 3c). The

cereal remains were dominated by oat grains at a ratio of roughly 13:1 oat to barley grains. Although no chaff

fragments were present at least 8 different taxa of weeds were represented in small numbers, including fat hen

(Chenopodium album), sheep’s sorrel (Rumex acetosella – acid indicator) and scentless mayweed

(Tripleurospermum inodorum).


40


690-950 cal AD - Sample 5083013, context 508011, pit 508075 – From the same context as sample 5083006

reasonable quantities of charred plant remains were recovered (27.3 fpl) comprising oats with a few barley grains

(13:1 oats to barley – Table 3c). A single possible spelt-type grain (identified as Triticum dicoccum/spelta) was

present, probably representing residual Iron Age remains. However, there is also the possibility that small

amounts of hulled wheats continued in cultivation into the earlier medieval period as has been found on some

other sites (see Pelling and Robinson 2000; Pelling 2003). Weed seeds were scarce.

Table 3a. Charred plant remains from the pit samples possibly associated with the Late Iron Age/Early

Roman curvilinear ditches and ditches A and B.

Flot 1st 2nd 1st 1st 2nd 1st 1st

Period LIA/ERB LIA/ERB LIA/ERB? undated undated

Feature Oven 508160 Ditch B Ditch A Pit Pit

Context type oven fill up. fill up. fill clay pit fill

Feature no. 508131 508091 508107 508077 508114

Context no. 508132 508092 508108 508083 508115

Sample no. 3044 3035 3039 3039 3033 3042

Proc. Vol (l) 30 22 3 ? 10

Flot vol (ml)

includes

charcoal

235 15 3 29 5 3 34

Cereal grains % analysed 100 100 100 100 100 100 100

Triticum

aestivum/t

urgidum-

type

free-threshing

wheat grain cf.1

Triticum

dicoccum/

spelta

hulled wheat cf.1

Hordeum vulgare L. hulled barley 7 10 10

Hordeum vulgare L. twisted hulled

barley 2 1

Hordeum vulgare L. barley indet 7 2 2 7 1 2

Avena sp. indet oat 3 1 9 15 4

Avena/Bromus sp. oat/brome 3

Cerialia indet Indeterminate

grains 4 1 14 5

Chaff frags oat : barley

grain ratio

Triticum spelta L. spelt glume

base 3

Triticum

dicoccum/

spelta

emmer/spelt

glume base 1


41


Triticum

dicoccum/

spelta

emmer/spelt

spikelet fork 1

Hordeum vulgare L. Barley rachis

frag 1

Other plant remains

Papaver sp. Poppy seed

CD cf. 1

Corylus avellana L.

hazel nut

shell

fragments

HSW

1

Polygonum aviculare

L.

knotgrass

achene Cdo 2

Persicaria

maculosa/l

apathifolia

redshank/pale

persicaria

CD

1

south wales gas pipeline project land at conkland hill wiston...south wales gas pipeline project...

Documents