Roman finds at Barland Quarry, Bishopston, Gower

In the summer of 1962 two local men, John Edwards and Frank Cross, made a chance discovery in Barland Quarry, Bishopston. Taking an evening stroll up a newly excavated path to the north of the quarry face, they noticed that oyster and other shells had fallen out from the sides of the path. Suspecting that these might be shellfish refuse from an earlier period, and knowing that oysters were an important addition to the Roman diet, they sought permission to return the following day between blasting operations to remove a small area of topsoil above where the shells were found.

They discovered that the area contained fragments of pottery, which were recognisable as belonging to the Roman period. Further digging yielded a large variety of domestic pottery fragments together with numerous other artefacts. The nature of the finds and the quantity of material recovered provide proof of the existence of a settlement there sometime in the second century A.D.

The size of the settlement may never be established as this area was quarried away shortly after the dig. However, a field at a slightly higher level, within a hundred metres of the excavation, shows certain surface irregularities that might, if permission was granted, justify a trial dig.

The situation would have been highly suitable for farming, being at an elevated level and having extensive flat fertile land within easy reach of a stream of fresh water. The excavation area may have been used to dispose of refuse, as all the objects lay close together in an area approximately six by three metres, as if thrown into a shallow depression in the ground and covered over in time by only a superficial layer of top soil.

During the excavation, the late Dr. Savory, keeper of archaeology at Cardiff museum, visited the site to record the details of the discovery. All the finds were then given to the Swansea Museum where they may be seen on display in Case 19 of the permanent collection.

With the exception of the Roman Villa at Oystermouth, very little is known of the whereabouts of other Roman British civilian sites in Gower. The nature of building materials of the period, particularly away from the main centres of Roman influence, would have been such that little if anything would be left on the surface to indicate where a settlement might have existed.

The geography of the area might give clues to the possible location of other dwellings that existed at this time. Easy access to water and the suitability of the land for growing cereal and other crops would have been an important factor, particularly as farming and farming-related activities would have been the principle occupation of the people. There are numerous locations on the peninsula where these might have existed. At Ilston quarry, a human skeleton, coins and forged nails were found in a similar position to that of Barland. Further evidence of occupation may come to light there sometime in the future. Suitable areas of the peninsula would have been gradually deforested by Neolithic farmers. A Neolithic stone axe of Cornish igneous rock found above Barland quarry is evidence that they were involved with land clearance in the area three thousand or so years before the Romano British farming communities.

Human settlement would have been continuous once the land had been opened up. Evidence of Bronze Age activity in the area comes from the numerous flint artefacts found in the ploughed fields above Bishopston Valley. Throughout these periods the people would have utilised the land for most of their everyday needs. During the second century A.D. when the Barland site was occupied, a large part of southern Britain would have seen towns and villages well established as the Roman influence on building and life style gradually spread. A network of rudimentary roads, would connect the local communities with each other and to the larger arteries of communication such as the Roman roads that linked the military forts and Romanised towns.

Access to these wider routes of communication for the Gower inhabitants would probably not have been easy. Even up to the early part of the 20th century the roads were unmetaled and often muddy. Access on higher ground would afford an easier movement, compared to the difficulties of travelling through the lower, often densely wooded levels.

Occupation during the preceding period of the Iron Age in Gower is much in evidence in the numerous hill forts on higher ground and around the coast, which would have accommodated sizable populations. During the period of Roman occupation the hill forts eventually gave way to unfortified settlements that left little if any visible marks on the landscape. For this reason it is difficult to give a figure as the size of the population during this period. The Barland Quarry settlement must be one of numerous other occupation sites that may eventually come to light. The following lists all the material recovered:

Pottery

Pieces of a range of black burnished pottery, some decorated with lattice pattern.

Roman Samian ware, one of which is a section of rim of a flat vessel decorated with a rhythmical leaf pattern, which helped to date the site to the late 2nd Century A.D. The presence of Samian ware suggests a degree of sophistication in living standards.

Part of a mortarium for grinding food was also discovered in the quarry rubble a few metres away from the dig, probably used to grind cereals grown on the adjoining land.

Iron slag

Numerous pieces of iron slag, one with a clay lining, an indication that iron smelting was being practiced on the site. Roman smiths together with the native inhabitants of the period were capable of smelting iron ores and working the blooms to produce tools and other objects. The iron objects found would most probably have been forged on the site. The waste produced by smelting was analysed and found to be produced using a Roman shaft furnace with provision for tapping slag. Numerous Romano British sites in the Glamorgan area have yielded waste products of smelting, suggesting that the practice was widespread. Another Iron Age site, probably older than Barlands and located in Bishopston valley, also produced evidence of iron production. Iron ores could have been found locally, where it has been extracted from a number of localities in the Gower area.

Iron objects

A variety of forged iron objects were found, the most notable being a primitive key, shaped like a sickle and used to detach a beam or bolt, evidence of the existence of a building or defensive structure. Numerous square-shank nails and parts of other unidentified forged pieces also came to light.

Bones

Bones of domesticated animals showing well defined chop marks, suggesting the rearing and slaughtering of livestock. Wild animals such as wild boar and hare would most probably have been common in the area at the time. providing sport and an additional supplement to the diet.

Fragment of tile

A single broken piece of red terra cotta tile was also found but without any recognizable features that could identify its function in a building.

Glass

A fragment of flat glass, probably part from a vessel.

Bronze brooch

Roman bronze trumpet brooch of a type used in the 2nd century.

Charcoal

The appearance of pieces of charcoal in association with iron slag suggest its use as fuel to produce the high temperature necessary for iron smelting.

Shellfish

Oysters, mussel and cockle shells. Quarrymen at Barland recalled often seeing quantities of oyster shells in the rock debris after blasting operations. Swansea Bay was until fairly recently renowned for the quality and abundance of its oyster beds, Considering the importance of oysters in the roman diet, it is not unreasonable to suggest that the so-called villa at Oystermouth may have been involved in some way with their cultivation and distribution, with possibly a wide and ready market to the Roman forts, town and villages.

Rotary Quern fragment

Section found near the digging area in the quarry debris. This would have been used for grinding corn, probably grown on the adjoining fields.

Article by John Edwards, August 2008

home | site map