On a spur of land above a stream near Eglwyswrw in north Pembrokeshire stands a little Iron Age fort, and inside its banks stand thatched roundhouses, rebuilt on the very places where the houses of its people stood more than two thousand years ago. This is Castell Henllys, the castle of the old court, one of the best-known Iron Age sites in Britain.Sources200 BC – Castell Henllys ↗Current Archaeology, Timeline of BritainA promontory fort near Eglwyswrw, in the Pembrokeshire Coast National Park, with a visitor centre open from Easter to October; open to the only approach from the north-west; Harold Mytum dug most of the inside and rebuilt houses on the very places where they stood; rows of stones outside the defences, the chevaux de frise, were probably meant to stop chariots, and were found under a later rampart; a Romano-British farmstead lay outside the northern defences.Castell Henllys, Pembrokeshire (WA1888) ↗Atlas of Hillforts of Britain and Ireland, University of OxfordAn oval fort of 0.51 hectares, about 88 by 74 m; its west gate began in the fifth century BC as a long stone-walled passage with a timber gate, a bridge and guard chambers; stones set as chevaux de frise on one side of the entrance; dug by Harold Mytum since 1980; eighteen radiocarbon dates put the start between about 410 and 375 BC; a Romano-British annexe lay beside it later, with signs of late Roman or post-Roman activity.
| Where | Near Eglwyswrw, Pembrokeshire Coast National Park |
|---|---|
| Begun | Between about 410 and 375 BC |
| Size | An oval of about 88 by 74 metres, half a hectare |
| Dug | By Harold Mytum, from 1980 |
| Later | A farm of the Roman period beside it |
A fort on a spur#
Castell Henllys is a promontory fort: it sits on a tongue of land with steep slopes falling away on three sides, so that its builders only needed to fortify one side, the north-west. There they raised their bank and ditch and built their gate.Source200 BC – Castell Henllys ↗Current Archaeology, Timeline of BritainA promontory fort near Eglwyswrw, in the Pembrokeshire Coast National Park, with a visitor centre open from Easter to October; open to the only approach from the north-west; Harold Mytum dug most of the inside and rebuilt houses on the very places where they stood; rows of stones outside the defences, the chevaux de frise, were probably meant to stop chariots, and were found under a later rampart; a Romano-British farmstead lay outside the northern defences.
The gate was begun in the fifth century BC. It was a long passage walled with stone, with a great gate partly of stone, a bridge, and guard chambers set into the sides. It was rebuilt over the years, and burnt at the end.SourceCastell Henllys, Pembrokeshire (WA1888) ↗Atlas of Hillforts of Britain and Ireland, University of OxfordAn oval fort of 0.51 hectares, about 88 by 74 m; its west gate began in the fifth century BC as a long stone-walled passage with a timber gate, a bridge and guard chambers; stones set as chevaux de frise on one side of the entrance; dug by Harold Mytum since 1980; eighteen radiocarbon dates put the start between about 410 and 375 BC; a Romano-British annexe lay beside it later, with signs of late Roman or post-Roman activity.
The stones against the chariots#
Outside the gate the archaeologists found something rare in Britain: chevaux de frise, rows of stones set upright in the ground to break up a charge of horses and chariots. Defences like them are known from Iron Age forts in Spain. At Castell Henllys they had been buried and kept safe under a later bank, and came to light when it was dug away.Sources200 BC – Castell Henllys ↗Current Archaeology, Timeline of BritainA promontory fort near Eglwyswrw, in the Pembrokeshire Coast National Park, with a visitor centre open from Easter to October; open to the only approach from the north-west; Harold Mytum dug most of the inside and rebuilt houses on the very places where they stood; rows of stones outside the defences, the chevaux de frise, were probably meant to stop chariots, and were found under a later rampart; a Romano-British farmstead lay outside the northern defences.Castell Henllys, Pembrokeshire (WA1888) ↗Atlas of Hillforts of Britain and Ireland, University of OxfordAn oval fort of 0.51 hectares, about 88 by 74 m; its west gate began in the fifth century BC as a long stone-walled passage with a timber gate, a bridge and guard chambers; stones set as chevaux de frise on one side of the entrance; dug by Harold Mytum since 1980; eighteen radiocarbon dates put the start between about 410 and 375 BC; a Romano-British annexe lay beside it later, with signs of late Roman or post-Roman activity.

Houses on their own foundations#
Inside the banks the ground was crowded with roundhouses. Harold Mytum of the University of York, who has dug the fort since 1980, found the grooves where their walls had stood, and he and his team rebuilt houses on the very same spots. That is what makes Castell Henllys special: you can walk into an Iron Age house standing where one stood two thousand years ago.Sources200 BC – Castell Henllys ↗Current Archaeology, Timeline of BritainA promontory fort near Eglwyswrw, in the Pembrokeshire Coast National Park, with a visitor centre open from Easter to October; open to the only approach from the north-west; Harold Mytum dug most of the inside and rebuilt houses on the very places where they stood; rows of stones outside the defences, the chevaux de frise, were probably meant to stop chariots, and were found under a later rampart; a Romano-British farmstead lay outside the northern defences.Castell Henllys, Pembrokeshire (WA1888) ↗Atlas of Hillforts of Britain and Ireland, University of OxfordAn oval fort of 0.51 hectares, about 88 by 74 m; its west gate began in the fifth century BC as a long stone-walled passage with a timber gate, a bridge and guard chambers; stones set as chevaux de frise on one side of the entrance; dug by Harold Mytum since 1980; eighteen radiocarbon dates put the start between about 410 and 375 BC; a Romano-British annexe lay beside it later, with signs of late Roman or post-Roman activity.

After the fort#
The people did not leave when the Romans came. A farm of the Roman period grew up beside the old fort, outside its northern defences, and people were still there late in the Roman centuries.Sources200 BC – Castell Henllys ↗Current Archaeology, Timeline of BritainA promontory fort near Eglwyswrw, in the Pembrokeshire Coast National Park, with a visitor centre open from Easter to October; open to the only approach from the north-west; Harold Mytum dug most of the inside and rebuilt houses on the very places where they stood; rows of stones outside the defences, the chevaux de frise, were probably meant to stop chariots, and were found under a later rampart; a Romano-British farmstead lay outside the northern defences.Castell Henllys, Pembrokeshire (WA1888) ↗Atlas of Hillforts of Britain and Ireland, University of OxfordAn oval fort of 0.51 hectares, about 88 by 74 m; its west gate began in the fifth century BC as a long stone-walled passage with a timber gate, a bridge and guard chambers; stones set as chevaux de frise on one side of the entrance; dug by Harold Mytum since 1980; eighteen radiocarbon dates put the start between about 410 and 375 BC; a Romano-British annexe lay beside it later, with signs of late Roman or post-Roman activity.
Visiting#
Castell Henllys belongs to the Pembrokeshire Coast National Park, which runs it as an Iron Age village with a visitor centre, open from Easter to October.Source200 BC – Castell Henllys ↗Current Archaeology, Timeline of BritainA promontory fort near Eglwyswrw, in the Pembrokeshire Coast National Park, with a visitor centre open from Easter to October; open to the only approach from the north-west; Harold Mytum dug most of the inside and rebuilt houses on the very places where they stood; rows of stones outside the defences, the chevaux de frise, were probably meant to stop chariots, and were found under a later rampart; a Romano-British farmstead lay outside the northern defences.
Built by master craftsmen#
The great gateway was a long passage walled with stone between pairs of huge timber posts, which held a large gate and probably a bridge or tower above it, with two pairs of guard chambers. The archaeologists thought it was built by craftsmen who came from elsewhere to do the work.SourceCastell Henllys Hillfort (NPRN 94989) ↗Royal Commission on the Ancient and Historical Monuments of WalesCoflein, the national record of Wales (RCAHMW 2003; T. Driver 2006): a defended Iron Age settlement about 88 m by 74 m on an inland promontory, with a double rampart and ditch on the open north-west side and an annexe beyond; life inside ended about the turn of the era, when a settlement grew up in the annexe; studied in scientific digs since 1980, led by Harold Mytum, whose first full report came out in 2013; Iron Age houses rebuilt on their original foundations using prehistoric materials and methods; the name implies an ‘old palace’ or early medieval court; first bought privately, later by the Pembrokeshire Coast National Park Authority, now a centre for education and training run with Cambria Archaeology and the University of York; the main gateway was begun in the fifth century BC, a long stone-walled passage between pairs of great timber posts holding a large gate and probably a bridge or tower, with two pairs of guard chambers; the excavators thought it was built by visiting craftsmen; outside was a chevaux-de-frise of sharp upright stones, like that at Carn Alw, replaced about 300 BC by a bank and ditch; a store of several thousand sling stones lay behind the rampart; the largest rebuilt round house used 30 coppiced oaks, 90 coppiced hazels, 2,000 bundles of water reed and two miles of hemp rope and twine; after the fort was left, a small, well-off Romano-British farmstead stood just outside; it lay lost and overgrown until it was named on a late 18th-century estate map. The upright stones outside the gate, like those at the fort of Carn Alw, were replaced about 300 BC by a bank and ditch, and behind the rampart lay a store of several thousand sling stones.SourceCastell Henllys Hillfort (NPRN 94989) ↗Royal Commission on the Ancient and Historical Monuments of WalesCoflein, the national record of Wales (RCAHMW 2003; T. Driver 2006): a defended Iron Age settlement about 88 m by 74 m on an inland promontory, with a double rampart and ditch on the open north-west side and an annexe beyond; life inside ended about the turn of the era, when a settlement grew up in the annexe; studied in scientific digs since 1980, led by Harold Mytum, whose first full report came out in 2013; Iron Age houses rebuilt on their original foundations using prehistoric materials and methods; the name implies an ‘old palace’ or early medieval court; first bought privately, later by the Pembrokeshire Coast National Park Authority, now a centre for education and training run with Cambria Archaeology and the University of York; the main gateway was begun in the fifth century BC, a long stone-walled passage between pairs of great timber posts holding a large gate and probably a bridge or tower, with two pairs of guard chambers; the excavators thought it was built by visiting craftsmen; outside was a chevaux-de-frise of sharp upright stones, like that at Carn Alw, replaced about 300 BC by a bank and ditch; a store of several thousand sling stones lay behind the rampart; the largest rebuilt round house used 30 coppiced oaks, 90 coppiced hazels, 2,000 bundles of water reed and two miles of hemp rope and twine; after the fort was left, a small, well-off Romano-British farmstead stood just outside; it lay lost and overgrown until it was named on a late 18th-century estate map.
Building a house the old way#
The roundhouses are rebuilt with the materials and methods of the Iron Age. The largest of them took 30 coppiced oaks, 90 coppiced hazels, 2,000 bundles of water reed and two miles of rope and twine of hemp.SourceCastell Henllys Hillfort (NPRN 94989) ↗Royal Commission on the Ancient and Historical Monuments of WalesCoflein, the national record of Wales (RCAHMW 2003; T. Driver 2006): a defended Iron Age settlement about 88 m by 74 m on an inland promontory, with a double rampart and ditch on the open north-west side and an annexe beyond; life inside ended about the turn of the era, when a settlement grew up in the annexe; studied in scientific digs since 1980, led by Harold Mytum, whose first full report came out in 2013; Iron Age houses rebuilt on their original foundations using prehistoric materials and methods; the name implies an ‘old palace’ or early medieval court; first bought privately, later by the Pembrokeshire Coast National Park Authority, now a centre for education and training run with Cambria Archaeology and the University of York; the main gateway was begun in the fifth century BC, a long stone-walled passage between pairs of great timber posts holding a large gate and probably a bridge or tower, with two pairs of guard chambers; the excavators thought it was built by visiting craftsmen; outside was a chevaux-de-frise of sharp upright stones, like that at Carn Alw, replaced about 300 BC by a bank and ditch; a store of several thousand sling stones lay behind the rampart; the largest rebuilt round house used 30 coppiced oaks, 90 coppiced hazels, 2,000 bundles of water reed and two miles of hemp rope and twine; after the fort was left, a small, well-off Romano-British farmstead stood just outside; it lay lost and overgrown until it was named on a late 18th-century estate map. Four roundhouses and a granary now stand on their original foundations, the only place in Britain where this has been done.SourceCastell Henllys ↗WikipediaCastell Henllys (‘castle of the old court’), an Iron Age hillfort near Nevern in north Pembrokeshire, between Newport and Cardigan; dug continuously and used for experiments in prehistoric farming; four roundhouses and a granary rebuilt on their original Iron Age foundations, some 2,000 years old, the only site in Britain where this has been done; on a low hill above the Duad, a stream flowing into the Nevern; a small hillfort, of a kind especially common in south-west Wales; a training ground for young archaeologists in summer; owned and run by the Pembrokeshire Coast National Park (English and Welsh Wikipedia).
Lost and found#
Life inside the fort came to an end about the turn of the era, and in time the place was lost under trees and scrub, until it was named on an estate map late in the eighteenth century.SourceCastell Henllys Hillfort (NPRN 94989) ↗Royal Commission on the Ancient and Historical Monuments of WalesCoflein, the national record of Wales (RCAHMW 2003; T. Driver 2006): a defended Iron Age settlement about 88 m by 74 m on an inland promontory, with a double rampart and ditch on the open north-west side and an annexe beyond; life inside ended about the turn of the era, when a settlement grew up in the annexe; studied in scientific digs since 1980, led by Harold Mytum, whose first full report came out in 2013; Iron Age houses rebuilt on their original foundations using prehistoric materials and methods; the name implies an ‘old palace’ or early medieval court; first bought privately, later by the Pembrokeshire Coast National Park Authority, now a centre for education and training run with Cambria Archaeology and the University of York; the main gateway was begun in the fifth century BC, a long stone-walled passage between pairs of great timber posts holding a large gate and probably a bridge or tower, with two pairs of guard chambers; the excavators thought it was built by visiting craftsmen; outside was a chevaux-de-frise of sharp upright stones, like that at Carn Alw, replaced about 300 BC by a bank and ditch; a store of several thousand sling stones lay behind the rampart; the largest rebuilt round house used 30 coppiced oaks, 90 coppiced hazels, 2,000 bundles of water reed and two miles of hemp rope and twine; after the fort was left, a small, well-off Romano-British farmstead stood just outside; it lay lost and overgrown until it was named on a late 18th-century estate map. It stands on a low hill above the Duad, a stream that runs into the Nevern, and every summer young archaeologists learn their craft there.SourceCastell Henllys ↗WikipediaCastell Henllys (‘castle of the old court’), an Iron Age hillfort near Nevern in north Pembrokeshire, between Newport and Cardigan; dug continuously and used for experiments in prehistoric farming; four roundhouses and a granary rebuilt on their original Iron Age foundations, some 2,000 years old, the only site in Britain where this has been done; on a low hill above the Duad, a stream flowing into the Nevern; a small hillfort, of a kind especially common in south-west Wales; a training ground for young archaeologists in summer; owned and run by the Pembrokeshire Coast National Park (English and Welsh Wikipedia).
Sources and further reading





