Excavations between 1930 and 1932, led by V. E. Nash-Williams of the National Museum of Wales, showed that the hillfort was begun in the third century BC — about 2,300 years ago — as a smaller enclosure surrounded by a single bank and ditch. A small stretch of this earthwork can still be seen outside the main banks on the north-east side. Around 150 years later, the main enclosure was modified with the construction of additional banks. The outside of the inner bank was reinforced with roughly coursed stone and the ditches were cut down into the limestone bedrock.

Evidence of round houses within the main enclosure was scanty in the areas excavated but it is highly likely that occupation was more intense than these results suggest. Bones recovered showed that domestic animals such as sheep and pigs as well as red deer were present. Evidence of burning, copper smelting, antler carving and cooking/eating were also found indicating that a range of activities took place here.

The third and final phase of Iron Age Llanmelin began around 50BC when the entrance was remodelled and strengthened. Could this suggest an increased threat of attack at the time?

The north side of the entrance was cut back and refaced so that timber platforms could be erected either side to create a stronger gateway. Wooden palisades were also built on top of the banks. The annexes are thought to have been added at about this time, possibly for the corralling of animals. However, human bones found in this area and the lack of entrances into the smaller enclosures suggest that the purpose of the annexe may instead be something quite different — perhaps relating to funerary practices.

Pottery found at Llanmelin, dating from around AD75, suggests that the hillfort was abandoned at this time, as no later pottery was discovered.

 
Later known activity at the site was focused around the annexe, where there are the remains of two medieval hut platforms. More recently, the hillfort formed part of the MoD Caerwent training area. Today, the hillfort is in Cadw ownership and the land around forms the home for many species of flora and fauna.

In 2012 Cadw conducted excavations at the hillfort with the help of many members of the community as part of the Llanmelin Community Project. The project aims to provide opportunities for people from all backgrounds to get involved in archaeology using Llanmelin as the focus.

We wanted to learn more about how people lived at Llanmelin so that we can tell visitors the story of the occupants as well as the history of the hillfort. 

When were people living here and what activities were taking place? Were they really part of the Silures and was Llanmelin a social and political centre at the time of the Roman conquest?

Five trenches were located in and around the areas of Nash-Williams’s excavations. We wanted to check the accuracy of his findings and see whether we could find any more information without doing much further harm to the monument.

One trench focused on the inner bank of the main enclosure, one on the entrance and three focused on areas within the main enclosure. The trenches in the main enclosure revealed a series of features, including a large midden deposit or rubbish dump. There were also some post holes which may mark the sites of structures used to dry and store grain as well as a couple of gullies. Although not certain, the gullies may indicate the circular footprints of the round houses which are so typical of the Iron Age in Britain.

The trench in the main entrance contained mixed-up deposits from when Nash-Williams filled in his excavations. As a result, it was difficult to interpret the archaeology. It was clear, however, that the northern bank was made of limestone. Similarly, the main inner bank of the enclosure was also stone built but without any clear stone facing, contradicting the results of Nash-Williams’s excavation.

The ditch was cut into the limestone bedrock and the removed rock had been used to build the bank. The cutting of the relatively deep ditch really was quite a substantial feat.

Analysis of the excavation itself and the substantial quantities of pottery and animal bone that were discovered is in hand. Information will be updated as soon as it is received from the various specialists involved.

The initial assessment of the animal bone shows that cattle was the most common species recovered with some sheep or goat and pig as well as a few fragments of dog, horse and red deer. Further analysis has the potential to tell us much more about husbandry practices. We hope to learn whether the cattle were being killed at a young age for meat production or allowed to mature as dairy cows. We may be able to tell whether they were being grazed down on the coastal plain or on the high ground near the hillfort.

Initial studies of the pottery suggest that the areas excavated were in use during the middle Iron Age — a period between 400 and 100BC. There was also relatively intense activity in the main enclosure during the first century AD. This could mean that Llanmelin was occupied at the time the Roman army arrived in Wales and made their base at Caerleon in 79AD.

The discovery of a small Roman copper-alloy nail cleaner dating perhaps from the end of the first century AD or beginning of the second century suggests some Roman contact with Llanmelin. 

Both excavations have given us a taster of the archaeology at Llanmelin and how well it is preserved. The dates of the pottery show that this site was used throughout the Iron Age and helps us to learn more about what people were doing here.