A Welsh Landscape through Time: Excavations at Parc Cybi, Holy Island, Anglesey

By Jane Kenney, Frances Lynch and Andrew Davidson

Holy Island is a small island just off the west coast of Anglesey, North Wales, which is rich in archaeology of all periods. Between 2006 and 2010, archaeological excavations in advance of a major Welsh Government development site, Parc Cybi, enabled extensive study of the island’s past. Over 20 hectares were investigated, revealing a busy and complex archaeological landscape, which could be seen evolving from the Mesolithic period through to the present day. Major sites discovered include an Early Neolithic timber hall aligned on an adjacent chambered tomb and an Iron Age settlement, the development of which is traced by extensive dating and Bayesian analysis. A Bronze Age ceremonial complex, along with the Neolithic tomb, defined the cultural landscape for subsequent periods. A long cist cemetery of a type common on Anglesey proved, uncommonly, to be late Roman in date, while elusive Early Medieval settlement was indicated by corn dryers. This wealth of new information has revolutionised our understanding of how people have lived in, and transformed, the landscape of Holy Island. Many of the sites are also significant in a broader Welsh context and inform the understanding of similar sites across Britain and Ireland.

• Language: English
• Publisher: Oxbow Books
• Release date: Aug 4, 2021
• ISBN: 9781789256901

Jane Kenney

Jane Kenney is a senior archaeologist at Gwynedd Archaeological Trust. She obtained her PhD from Edinburgh University in 1993, and excavated on sites of all periods in Britain and abroad before settling in north-west Wales. Since then she has been fortunate enough to run two large-scale excavations, including Parc Cybi, on both of which Early Neolithic buildings were found.

Book preview

1

Introduction

Introduction

Holy Island, known in Welsh as Ynys Gybi, is a small island off the west coast of Anglesey (Ynys Mȏn), a large island off the north coast of Wales (Figure 1). Holy Island is best known for its principal town of Holyhead (Caergybi), from which runs one of the main ferry routes to Ireland. Archaeologically Anglesey is known for its high density of Neolithic chambered tombs, and one such tomb, the Trefignath Chambered Tomb, lies just south of Holyhead. The Welsh Government proposed a mixed-use development, to include a business park and housing, on land adjacent to the tomb. The development aimed to help regenerate the area and was located in this position, immediately south of Holyhead, for its proximity to the A55 trunk road and easy access to the port. The Trefignath Chambered Tomb indicated that significant archaeological remains were likely to be found, so a staged programme of archaeological works was funded by Welsh Government as part of the planning process, to identify and record any archaeological remains that could not be avoided by the development. While archaeology was expected, the extent and range of the discoveries was beyond expectations for this farmed landscape.

The excavated features range from the Mesolithic to the 19th century, with particularly significant finds from the Neolithic, Bronze Age, Iron Age and Roman periods. Early Neolithic settlement was represented by a rectangular timber building and occupation of a more ephemeral character. Temporary settlement continued into the Middle and Late Neolithic period, and a burnt mound was used repeatedly throughout the Beaker period. In the Bronze Age, a complex of ceremonial monuments was constructed, consisting of a multiple-cist barrow, the ring-ditch of a barrow, a deep-ditched enclosure, and, nearby, a standing stone. A timber roundhouse and other activity probably indicate Bronze Age settlement.

A settlement of stone-built roundhouses with complex stratigraphy was accompanied by several outlying roundhouses and other structures, as well as possibly contemporary field boundaries. The settlement was preceded by boundary ditches and a stone platform. A large fire occurred on the platform in the Early Iron Age, possibly the burning of a timber building. This was replaced in the Middle Iron Age by a settlement with up to four main houses at any one time, and additional buildings including granaries. Two outlying clay-walled roundhouses were roughly contemporary with the main settlement and other small-scale activity occurred on the site at this time.

Later Iron Age activity was represented by a small structure and associated pits, and possibly by four- and six-post granaries. Early Roman activity was slight, but by the late 3rd or 4th century AD, a trackway ran through the site. Next to the trackway was a Late Roman farmstead, with industrial and storage functions. This included a clay-walled building with hearths and troughs inside, a square stone building and numerous timber storage structures. Late Roman smithing activity took place within a pre-existing long cist cemetery on top of a hill, dating the cemetery to the Roman period. Evidence for Early Medieval activity included several corn dryers, all dating to the 5th or 6th centuries AD. There was evidence for the development of later Medieval and post-medieval field systems, cottages and farmhouses, and this is supplemented from the 18th century by estate maps.

The business park has been named Parc Cybi after Saint Cybi, who was said to have established a monastery on Holy Island, and who gives his name to the island in Welsh (Ynys Gybi (Cybi’s Island)) and to the Welsh name for Holyhead (Caergybi (Cybi’s Fort)). The name of the archaeological site is therefore Parc Cybi, but the area had other names prior to this latest chapter of its history. The names of the various farms that covered the area are discussed below but one name was particularly intriguing. Tegwyn F. Jones, now an artist living in Bodedern, grew up in Holyhead. He remembers in the 1950s that one local gentleman referred to land next to a marsh, within what is now Parc Cybi, as ‘pant yr hen bobl’; a literal translation would be ‘hollow of the old people’, but in this instance it is better translated as ‘the hollow of the former, or ancient, people’. Young Tegwyn was confused by this, but that hollow was the site of the Middle Iron Age roundhouses, and the name shows that local people, if not archaeologists, knew of their existence. This knowledge may have come from taking stone from the roundhouses to build the surrounding field walls, but local people were aware that the stones they were using came from an ancient settlement. Now we have discovered how and when Yr Hen Bobl (The Ancient People) lived in this corner of Holy Island.

Figure 1. Location of Parc Cybi showing current road layout and location of Holy Island (base map © Ordnance Survey 100021874)

Project background

(For further details, including methodology, see Appendix II)

Parc Cybi covers over 41 hectares of pastureland to the south of Holyhead (centred on SH 2555 8075) (Figure 1), lying between the A55 trunk road and Kingsland Road (B4545) (Figure 2). A narrow lane, known as Lôn Trefignath, runs through its eastern side. This lane was open to traffic prior to the development, but is now a footpath and cycleway.

Figure 2. Ploughsoil stripping under way at Parc Cybi. View looking south with the A55 on the left and Kingsland Road on the right (photograph by David Longley)

Gwynedd Archaeological Trust carried out a staged programme of archaeological work, starting in 2000 with an archaeological desk-based assessment, followed in 2001 by initial field evaluation which revealed a high density of archaeological sites requiring further evaluation. A geophysical survey in 2004 was followed by further trial excavations from 2004 to 2006.

The main excavation phase started on 6 November 2006 and was completed on 30 June 2008, with a second phase between 7 September 2009 and 26 February 2010. The aim of these phases was to mitigate the impact of the development by identifying and recording all significant archaeological features and deposits within areas to be affected by construction activities. In order to understand the archaeology fully, the site needed to be approached as a complete landscape; this was achieved by controlled removal of the ploughsoil by machine under archaeological supervision. The stripped surfaces were examined to identify archaeological remains, which were evaluated and mapped. Where significant features were identified, these were fully investigated and recorded in detail.

The results from both phases of excavation were assessed and an Assessment of Potential Report was submitted in June 2011, along with a project design for post-excavation work and publication. This final essential phase was delayed for some years; when the work was tendered by Welsh Government in December 2017, the contract was won by Gwynedd Archaeological Trust and work commenced in June 2018. This resulted in the submission of a full grey literature report, including all specialist reports (Kenney et al. 2020), in December 2020. This report is available online from the Royal Commission on the Ancient and Historical Monuments of Wales Coflein website (https://www.coflein.gov.uk/) (search for Parc Cybi) and the Gwynedd Archaeological Trust website (http://www.heneb.co.uk/).

Dissemination of the results of the project during 2019 to 2021 included the creation of a website (http://www.heneb.co.uk/parccybi/index.html), a summary report available digitally, and public talks. A day school was held in association with the Anglesey Antiquarian Society in Holyhead, and public exhibitions at the Ucheldre Centre, Holyhead and Oriel Môn, Llangefni, created with the assistance of pupils from Ysgol Cybi, Holyhead.

On site, roads and other infrastructure had been built during 2008, alongside the excavations. Since then development has taken place on areas that had been archaeologically excavated and recorded. In 2020, substantial parts of the site remain uninvestigated, so future development in these areas will need to be preceded by fieldwork.

Report guide

This report provides a summary of the most significant features found at Parc Cybi. The fine details of stratigraphy and descriptions of features can be found in the grey literature report (Kenney et al. 2020). This publication contains summaries of the results of the specialist reports, but the full reports are to be found in the grey literature report, available online from the Royal Commission on the Ancient and Historical Monuments of Wales Coflein website (https://www.coflein.gov.uk/) (search for Parc Cybi) and the Gwynedd Archaeological Trust website (http://www.heneb.co.uk/). The current work contains some slight changes in interpretation compared with the grey literature report and must be taken as the final statement on the site by the excavator.

Certain conventions and abbreviations have been used in this monograph. PRN refers to the Gwynedd Historic Environment Record (HER) Primary Record Number and identifies a site recorded in the Gwynedd HER. There are many sub-sites across the development area and their PRN has been used as an identification number for these, superseding other reference terms used during the excavation.

Where a reference is given to a certain PRN FI File, this refers to Further Information files held by the HER for certain sites and filed by PRN. The abbreviation ‘SF’ indicates ‘Small Find’ and is used to distinguish numbers identifying artefacts from other record numbers, such as Context Numbers. The latter are numbers used to identify deposits and cuts and these are presented in the text without brackets, unless these are grammatically necessary. The Context Number for the cut of a feature is used to identify that feature. Numbers, part of the Context Number sequence, were allocated to groups of features, especially structures, so that these could be referenced as a whole. Some of these Structure or Group Numbers have been used in the text but in some cases it was convenient on site to refer to structures, such as the main stone-built roundhouses, by letters and this practice has been continued in the text.

All dates are based on calibrated radiocarbon dates or, for later periods, historical dates. For generalised dates, BC and AD are used, with cal BC and cal AD for specific dates quoted. All individual radiocarbon dates are presented at 95% probability, unless otherwise stated, and rounded out to the nearest 10 years. Where dates from other sites have been used for comparison, those produced some years ago have been calibrated or recalibrated using Oxcal v4.3 with the IntCal 13 curve (Bronk Ramsey 2009). Where calibrated dates have been quoted in other publications without rounding out they have been rounded out to the nearest 10 years for consistency. Details of all dates from Parc Cybi are included in Appendix I.

Where places or topographic features have both Welsh and English names, the English name is used in this report, with the Welsh name in brackets when the name is first used.

Archive

A database was created in Microsoft Access containing all site information, allowing its efficient interrogation and output. The database includes the drawing, photographic, finds and samples registers, and selected information from the context sheets. All field drawings, context sheets and object record sheets have been scanned to provide a backup digital copy. The digital archive also includes photographs, surveys and specialist data.

The paper record is held at Anglesey Archives, Llangefni; the digital archive is held by the Royal Commission on the Ancient and Historical Monuments of Wales in Aberystwyth; and the artefactual archive is held by Oriel Mȏn, Llangefni (main collection: accession number 45/2019; gold ring: accession number 9/2016). A copy of the grey literature report is held in the Gwynedd Historic Environment Record (HER) and is available on-line through the Archwilio website. The report is also held by Royal Commission on the Ancient and Historical Monuments of Wales and is available on their Coflein website.

Acknowledgements

The excavations and post-excavation analysis were funded by Welsh Government and thanks are due to Clwyd Roberts for working with us on the post-excavation phase, assisted by consultant Rob Early of WSP Parsons Brinckerhoff. On site, our work was greatly assisted by Jones Bros. Ruthin Co. Ltd., with particular thanks to Hefin Lloyd-Davies, and thanks also to Atkins Engineering Consultancy, especially Rob Sutton, who were the archaeological advisors for the Welsh Government. Thanks to the staff of Gwynedd Archaeological Planning Service, Emily La Trobe Bateman, Ashley Batten and Jenny Emmett, for their support and advice through both the fieldwork and post-excavation work. The Project Manager for Gwynedd Archaeological Trust throughout the course of the project was Andrew Davidson. The Site Directors were Roland Flook for the initial months, and Jane Kenney for the remainder of the project. Up to 40 archaeologists at any one time worked on the site during the two phases of fieldwork, and their hard work through weather of all kinds is much appreciated.

Thanks for advice and discussions to Ray Karl, Kevin Blockley, Steve Burrow, Frances Lynch, George Smith, Helen Flook, Sue and Dave Chapman and the Gwynedd branch of the Guild of Weavers, Spinners and Dyers. Roy Loveday is thanked for drawing my attention to his work on solar alignments, which clarified my ideas about the chambered tomb. Thanks are also extended to the following people for permission to use figures or unpublished information: Christopher Tolan-Smith, Jacky Nowakowski, Mark Gardiner, Stephen Thompson and Ian Grant. The author extends a very big thank you to Tegwyn F. Jones for permission to use his early paintings and for information about Pant yr Hen Bobl.

All the specialists who have contributed to the project are thanked for their contributions and willingness to discuss relevant elements of the site and presentation of the results. Frances Lynch, George Smith and Roz McKenna in particular had large tasks and put in extra work. The specialists involved in this project are as follows: Nora Bermingham, freelance (animal bone); Evan Chapman, National Museum Wales (metal and shale objects); Hilary Cool, Barbican Research Associates (Roman glass); Jolene Debert, University of Lethbridge, Canada (lithic use-wear analysis); Mary Davis and Adam Gwilt, National Museum Wales (gold penannular ring); Julie Dunne and Richard P. Evershed, University of Bristol (organic residue analysis of pottery); Ben Gearey, Tom Hill, Emma Tetlow and Emma-Jayne Hopla, Birmingham Archaeo-Environmental (pollen analysis); Jonathan Goodwin, Stoke-on-Trent Archaeology (post-medieval pottery); Pam Grinter, freelance (charcoal and charred plant remains); Derek Hamilton, SUERC (radiocarbon dating); Jana Horák, National Museum Wales (petrological description of selected objects); Emma Kitchen and Ben Gearey, Birmingham Archaeo-Environmental (insect remains and plant macrofossils); Kristina Krawiec, Birmingham Archaeo-Environmental (wood identification); Helen Lewis, University College Dublin (soil micromorphology); Frances Lynch, formerly Bangor University (prehistoric pottery); Rosalind McKenna, freelance (charcoal and charred plant remains); Elaine L. Morris, University of Southampton (Cheshire Salt Containers); Andrew Moss, Birmingham Archaeo-Environmental (marine shells); Phil Parkes, Cardiff Conservation Services (X-rays and conservation); Alan Pipe, MOLA (animal bone); Patrick Sean Quinn, UCL Institute of Archaeology (pottery fabric analysis); Alison Sheridan, National Museums of Scotland (cannel coal and amber beads); George Smith, Gwynedd Archaeological Trust (knapped stone and other worked stone objects); Peter Webster, formerly Cardiff University (Roman pottery); Penelope Walton Rogers, The Anglo-Saxon Laboratory (spindle whorls); R. Alan Williams, University of Liverpool (macroscopic petrological identification and burnt stone); John Llywelyn Williams, formerly Bangor University (Group VII stone objects); John Llywelyn Williams and David Jenkins, formerly Bangor University (pottery fabric analysis); Michael Wysocki, University of Central Lancashire (human remains); Tim Young and Thérèse Kearns, GeoArch (metal-working debris and burnt clay). Their reports can be found in volume 3 of the full report (Kenney et al. 2020).

Artefact drawings are by Tanya Williams, Frances Lynch, George Smith, Marion O’Neil and Tony Daly. Bethan Jones, Carol Ryan Young and Neil McGuinness assisted with the production of figures. Neil McGuinness, Richard Cooke, and Cat Rees wrote some of the original text on which the final text is based. Jane Kenney wrote the bulk of the text, which was then edited and revised by Frances Lynch and Andrew Davidson, providing significant contributions to style as well as content. Jane is particularly grateful for their contribution. Thanks to Catrina Appleby for copyediting and Oxbow Books for their assistance in producing this volume.

2

A landscape through time

Introduction

This chapter looks at the history of human activity on Parc Cybi and its impact on the physical landscape and vegetation, within the context of Holy Island, Anglesey, and North Wales and beyond. This history has been revealed by the excavations of 2004–2011 and interpreted in the light of our understanding in 2020. In the following chapters, the evidence on which this interpretation has been based will be laid out in the expectation that the story may be modified and refined in future years and generations.

See Figures 3 and 6 for the location of sites on Holy Island and Figures 5 and 7 for sites and topography on Parc Cybi.

Topography and geology

Holy Island, or Ynys Cybi, is located off the western coast of Anglesey, itself a large island, separated from the mainland of North Wales by the Menai Strait. Holy Island is separated from Anglesey by a narrow strait that largely empties of water at low tide (Figure 3). The tides around the headlands on the northern and western coast can be fierce, with an overfall, known as Penrhyn Mawr, famous with kayakers; however, the eastern side of the island is sheltered, with several small inlets.

At the end of the last Ice Age, sea levels around North Wales were about 30 m lower than present, but from c. 11500 calendar years BP sea level started rising quickly. After 8500 BP its rise slowed, but continued until 5000 BP, when it was 2 m below the current level. Study of seismic data and sediments in the Menai Strait has shown that much of the strait had already formed by 8600 BP. However, a narrow rocky section in the middle of the strait, now known as the Swellies, remained as a causeway between Anglesey and the mainland, although often submerged at high tide. Between 5600 and 4800 BP this causeway gradually became permanently submerged and Anglesey became a true island during the Neolithic period (Roberts et al. 2011, 153).

The strait between Holy Island and Anglesey has not been studied in such detail. The strait originated in a north-east to south-west glacial channel (Steers 1964, 118–19), later used by the lower reaches of the Afon Alaw, with the mouth of the river probably entering the sea at Trearddur Bay. Sea-level rise inundated the present strait, turning the river to the south, possibly at a similar time as the flooding of the Menai Strait. Fossilised trees within the inter-tidal zone of Trearddur Bay are an indication of significant sea-level rise (Smith 2002). Such coastal woodlands were inundated elsewhere in Wales by the very start of the Neolithic period (Lillie 2015, 58–9), so it is likely that Holy Island was a true island from the Neolithic period onwards.

Holy Island is dominated by Holyhead Mountain (Mynydd Tŵr) at its northern end, which reaches 220 m OD and has heathland vegetation. The summit is rocky, with impressive sea cliffs along its north-eastern and western sides. The southern end of Holy Island is lower and gentler, with sandy beaches at Rhoscolyn, Silver Bay and Trearddur on the west and at Penrhos on the east coast, but much of the coastline is rocky and can be treacherous.

The terrain of Parc Cybi, like much of the northern part of Holy Island, is characterised by rocky ridges aligned north-east to south-west, with intervening boggy hollows (Figure 4). The bedrock is never far below the surface, and occasionally outcrops as small crags and knolls. These outcrops of bedrock are metamorphic schists of the New Harbour Formation within the Precambrian Mona Complex (Greenly 1919; Treagus 2008). The current form of the landscape is the result of at least two phases of glaciation, with ice advancing south across the Irish Sea. This caused erosion of exposed rock and the deposition of glacial till, which includes exotic rock types brought from the north-east, from the floor of the Irish Sea and the Lake District beyond, some of which were used as filler in the prehistoric pottery made in the area. These tills were modified by the action of powerful melt waters during the northward retreat of the ice margins, which deposited sorted sands and gravels (Helm and Roberts 1984). The peri-glacial conditions towards the end of the last glaciation caused the deposition of loess, a homogenous fine wind-blown silt, which survives in patches over the till.

Figure 3. Topography of Holy Island with prehistoric sites (numbers on maps are PRNs for sites discussed in the text) (base map © Ordnance Survey 100021874)

The soils over this geology are free-draining brown earths of the Gaerwen Series, becoming more rocky over the outcrops or waterlogged where there is impeded drainage (Roberts 1958; Avery 1980). These soils have a low nutrient status compared with soils on limestone elsewhere on Anglesey, but they are extensive, loamy, well-structured soils, which are agriculturally productive. Prior to the development, most of Parc Cybi was used for grazing sheep and cattle, resulting in vegetation of improved grassland, with gorse and bramble growing on the rocky ridges.

Figure 4. Valley mire with rock outcrops in the middle of Parc Cybi

The undulating terrain led to the development of marshes. The main marsh developed in a glacial basin with low crags on one side (Figure 4). This runs east–west across the western side of Parc Cybi (Figure 5) and originated as a small lake in the late glacial period, soon infilling and developing into a valley mire in the early Holocene. A smaller marsh existed on the eastern side of Parc Cybi, a remnant of a larger marsh extending to the coast. Before the building of the roads and railway across the head of the bay during the last three centuries, this marsh must have flooded at high tide and it probably originated as an inlet from Penrhos beach leading towards the heart of Parc Cybi.

The ends of this proposed inlet and the valley mire would have been only about 460 m apart, and the ridge between them was the focus for Bronze Age and Iron Age settlement. The marshes would have provided food and other resources useful in all periods, though the exploitation of the valley mire for peat in later periods truncated the pollen record and reduced the information on vegetational change in the later periods.

The rocky bones of the landscape would have remained unchanged until the Parc Cybi development, when some of the knolls and rounded hills were levelled or altered. These features would have been familiar to the inhabitants of the area from the Neolithic onwards, used for shelter, to give prominence to the chambered tomb, as look-out points, and as landmarks for those moving through the landscape.

The largest outcrop of all, Holyhead Mountain, would have looked out over the site throughout its history, sometimes of immense cultural significance and at other times largely disregarded, but always visible. Holyhead Mountain, although only 220 m high, dominates Holy Island and much of northern Anglesey, which is otherwise low-lying. It has dramatic crags on the south-west side and especially from a distance resembles a very large cairn. Holyhead Mountain would have been a vital landmark for travellers by sea as it marks the northern corner of Anglesey, when none of the rest of the island would have been visible. It can be seen from the end of the Llŷn Peninsula and Bardsey Island (Ynys Enlli), so it could be used to navigate directly across the Caernarfon Bay. It would also act as a beacon warning of the strong tidal currents at its foot. Its importance as a seamark may have enhanced the significance of this mountain, which was already prominent from the land.

Late Glacial and Mesolithic Period

Climate and human history

Following a final cold spell, known as the Loch Lomond Stadial or Younger Dryas, the climate in Britain started to warm from 9600 BC, marking the end of the last glacial period and the start of the Holocene, the geological period in which we now live (Cunliffe 2013, 50, 54, 99). The temperature rose rapidly and as the ice sheets melted the global sea level rose. During the Ice Age Britain had been part of the continental landmass but the rising sea level converted Britain into an island, though the English Channel appears not to have formed until about 5400 BC (Coles 1998, 67, 76; Lillie 2015, 49, fig. 2.1).

Figure 5. Prehistoric sites on Parc Cybi (base map © Ordnance Survey 100021874)

With the ameliorating climate came a spread of forests: juniper, hazel and birch scrub later developing into mixed oak forest with elm, alder and hazel. In Wales, lime and ash only grew in a few sheltered places and there was pine on the uplands, with alder concentrated in wetter areas (Cunliffe 2013, 58, 101; Caseldine 1990; Lillie 2015, 68–87); forest would have covered most of the country, including the mountains up to about 600 m OD, although the coasts may have been more open (Lillie 2015, 72–3, 76; Caseldine 1990). Large mammals extended their ranges into these forests, including aurochs, red deer, roe deer, wild pig, brown bear, wolf, fox, wild cat, otter and beaver, which have all been found on Mesolithic sites in Wales, as well as elsewhere in Britain (Coles 1998, 65; Aldhouse-Green 2000, 27). The forest animals lived in smaller groups than the large herds of the open plains of the Ice Age, requiring different hunting techniques. Grazing animals and beavers would have made clearings in the forest, and people probably added to and enhanced such clearings to concentrate their prey, making hunting easier, or to improve the production of hazelnuts and other plant resources, so the forests were not uniformly dense and inaccessible (Lillie 2015, 57–8, 87–93).

Archery, which had developed by the end of the Upper Palaeolithic, was particularly suited to hunting animals in the forests. New plant foods must also have been exploited, leading to the development of the typical Mesolithic tool type – the microlith. Several microliths could be used together set in a shaft to form the head of an arrow or to make knives, saws or other tools. Flint was also used to make scrapers for processing skins into leather, and for small flaked axes. Other tools, such as harpoon heads and awls, were made of antler and bone, and various organic materials that rarely survive except at waterlogged sites such as Star Carr, near Pickering (Milner et al. 2018) would have been used for baskets and nets (Aldhouse-Green 2000, 33–6; Cummings 2017, 10).

The Mesolithic period is divided into Early and Late by the style of microliths used, with the Late Mesolithic starting about 8000–7700 cal BC (Lillie 2015, 124). A much later date of 6500 BC (Cummings 2017, 10) is often given for this change, but this is largely based on uncalibrated dates. Early Mesolithic microliths were made on broad blades and shaped into blunted points and triangles; these sometimes occur with tranchet adzes, sharpened by a sideways (tranchet) blow. In the Late Mesolithic, a narrow blade technology was used, often made from neat pyramidal cores. The narrow blades were used to make smaller microliths than in the Early Mesolithic and these were usually shaped into scalene triangles (Aldhouse-Green 2000, 33–6).

Mesolithic people are assumed to have been highly mobile, though they probably had fixed winter base camps that may have been used all year by some sections of the population (Cunliffe 2013, 106). Mesolithic settlements are often found close to the coast, though this is deceptive as in most places the coastline would then have been much further out than it is today (Aldhouse-Green 2000, 28; Lillie 2015, 49, fig. 2.1). Seasonal mobility within a territory was probably the normal situation, but these small groups must have gathered on occasion to exchange marriage partners, information and resources.

Substantial Mesolithic houses have been found in some parts of Britain, such as Howick in Northumberland and East Barns, near Dunbar. Most of the more substantial houses that have been found date to the 8th millennium BC and they have a floor area of about 30 m² with a central hearth (Waddington et al. 2007, 203, 205). While none has yet been found in Wales, postholes forming windbreaks or possible structures have been found at Rhuddlan and Llyn Brenig, with other sites having hearths dated to the Mesolithic period (Aldhouse-Green 2000, 32).

Mesolithic on Anglesey

Several sites on Anglesey have produced a few flints of Mesolithic date but the most significant site to have been excavated was on the current shore at Trwyn Du, near Aberffraw (White 1979). However, despite finding several thousand flint flakes and tools, no structural evidence emerged.

On Holy Island the situation is similar. Mesolithic flint scatters have been identified on Penrhosfeilw Common on the west coast of the island (Smith and Kenney 2014); in the eroding sands of Penrhos Beach (Williams 1950), at Porth Ruffydd; and from ploughsoil near Rhoscolyn (Lynch 1991, 329) (Figure 3). The most significant collection is the 199 knapped stone artefacts found during excavations at the roundhouse settlement of Tŷ Mawr, at the foot of Holyhead Mountain (Smith 1986, 12–23). Most of these came from a single area of the site and they are considered Late Mesolithic in date, but there is no indication of structures associated with them.

The coastal location of most of the finds is partly related to the eroding conditions, which favour discovery, but it may also indicate that most activity in this period was concentrated on the coast. The west coast of Holy Island falls steeply below the current sea level so, unlike many parts of Wales, by 7500 BC the Mesolithic coast here would have been fairly close to the present coastline, and the flint scatters represent genuine coastal activity (Bell 2007, 8). The flints from Tŷ Mawr, however, show that excavation can reveal inland activity and it is probable that these small groups of people could have been found all across Anglesey. One very characteristic microlith came from beneath the Bronze Age barrow at Bedd Branwen, quite far up the Alaw valley (Lynch 1991, 46).

Mesolithic Parc Cybi

Our evidence at Parc Cybi starts in the later glacial period when the marshes described above would have been small lakes and the open landscape of the glacial period would have been developing through birch scrub into mixed deciduous woodland. By the Mesolithic period it would have been a generally wooded landscape with hazel, birch, alder and willow forming a fairly closed canopy, allowing ferns to grow in the damp, shady conditions under the trees (Gearey et al. 2020, 686).

The evidence for Mesolithic activity at Parc Cybi is slight and widely scattered (see Chapter 3) (Figure 5). It consists mainly of a small number of microliths, probably casual losses during hunting or gathering activities. A microlith and a Mesolithic radiocarbon date were obtained from a natural hollow used extensively in the Neolithic period, which may also have provided temporary shelter at this earlier date.

There was a small group of flint and chert pieces, including a core and a microlith, found in an area of plough furrows which had been cut by two post-medieval ditches. The flint and chert must have been within the medieval ploughsoil, but the fact that the pieces had remained together suggests that they had not been much disturbed and the area may have been a focus of Mesolithic activity.

A group of postholes within an extensive area of Neolithic and Bronze Age activity may have represented a Mesolithic structure, although this interpretation is tentative. The possible structure was defined by a slot containing a line of three postholes with a nearly parallel line of four more to the east, and three on a different alignment to the west. Chert debitage was recovered from two of the postholes and another posthole produced two Late Mesolithic dates, but these dates are very different and cannot be used to provide a reliable date for the structure. Another Late Mesolithic date came from a nearby feature and together they might indicate general Late Mesolithic activity in this area of the site.

Neolithic period

The origins of farming in Britain

The development of farming, which characterises the Neolithic period, was one of the most significant developments in human history, leading ultimately to increasingly complex and large-scale societies (Whittle and Cummings 2007b, 1). The farming tradition that spread over Europe developed in the Near East around 11,500 years ago, within the Fertile Crescent, running through the Levant, Turkey, Iraq and Iran. Here plants and animals that still form the basis of our agriculture, including wheat, barley, sheep, goat, cattle and pigs, were found together in the wild (Zeder 2011, S230). The knowledge and techniques of farming, as well as the domesticated species, spread out from this original zone.

Farming had reached northern France by 5500 BC, probably largely spread by the expansion of farming communities (Cunliffe 2013, 136). The practice of farming in central Europe was associated with a different range of cultural items from those used by Mesolithic groups, most particularly pottery. There was a delay of over 1000 years before farming came north to these islands. The best current estimate for the arrival of farming and associated Neolithic objects and practices in Britain is 4100 cal BC, with the Neolithic lifestyle having spread across the whole of Britain and Ireland by at least 3700 cal BC (Whittle et al. 2011).

The mechanism for the arrival of farming in Britain is much debated. It is argued that the Mesolithic population may have played an important role in adopting farming techniques (Thomas 2007, 427), while other authors see farming as entirely introduced by incoming groups (Sheridan 2007, 442). The crops and the sheep and goats were certainly Mediterranean in origin. The issue of a single entry point or several has also been discussed, but either way the spread was rapid (Garrow et al. 2017, 98–9). While no single source area on the continent has been identified, people occupying the channel coast from Brittany to the Rhine made pottery and other artefacts very similar to those found in Britain. Monuments built in Iberia and Western France can be compared with those in Ireland, Wales and Scotland (Cunliffe 2013, 139; Bradley 2007, 34–5; Sheridan 2007, 468). The perennial debate between archaeologists favouring local development as the engine of change, and those suggesting introduction from elsewhere, has, to a certain extent, been resolved by ancient DNA studies (Olalde et al. 2018, 192; Brace et al. 2019). These suggest that the arrival of people was critical, though it is argued that it does not rule out an important role for the local Mesolithic population (Ray and Thomas 2020, 84, 89–91).

Apart from farming and the new relationship that it created with land and its fertility and with settled living, there was, in the Neolithic, a new attitude to death and the continuity of families. Mesolithic cemeteries were simple, understated and are seldom found; Neolithic tombs were huge, permanent memorials that still impress today. Where the bones survive, they show that these were communal burials, not commemorating a single individual but a protective group of ancestors, whose tomb remained the focus of their community (Lynch 1997; Lynch 2000a, 63–77). This communality and monumentality are found everywhere across northern and western Europe, but the design of the burial chambers and their covering mounds or cairns vary a good deal. These variations enable us to recognise cultural connections between groups.

In the British Isles, the Neolithic period can be divided into Early, Middle and Late by artefact and monument types, reflecting social changes. The Early Neolithic, down to about 3500 BC, is defined by finely made, generally undecorated pottery and the presence of large rectangular timber buildings in some parts of Britain and Ireland. Long barrows in southern England belong to this period (Bayliss and Whittle 2007) and probably many of the megalithic tombs of Wales. The Middle Neolithic, roughly 3500 to 3000 BC, is defined by the appearance of decorated pottery, known as Peterborough Ware. Large rectangular timber buildings are no longer built, with monumental timber buildings generally being circular and settlement being represented by groups of pits, with occasional remains of small circular or square structures. The Late Neolithic, from 3000 BC, saw a social change that continued into the Bronze Age. Houses were still small and relatively temporary, but large gatherings of people are suggested in large monuments, especially circular ditched monuments, known as henges. This is associated with the spread across Britain of a new style of decorated pottery, Grooved Ware, coming possibly from Orkney along with henges and stone circles (Cunliffe 2013, 194; Sheridan 2012a, 41). Large and impressive passage graves were built under circular mounds, two of which can be seen on Anglesey at Bryn Celli Ddu and Barclodiad y Gawres (Lynch 2000b, 73–7).

The Neolithic in Anglesey

The Neolithic seems to appear in North Wales and Anglesey fully formed and complete with the full package of pottery, polished stone axes, large timber