B.A.R. British Series 287 December 1999 252 pages (Price £32)
© S.Palmer
Summary of the excavation report:Culverwell Mesolithic Habitation Site

The Culverwell Mesolithic site is situated near the southern tip of the Isle of Portland, Dorset at NGR SY685694 on fields 2154 and 2153, towards the bottom of the slope of a south-facing hill. It is about 274m from the present-day cliff tops, with wide open views to the sea. It is also about the same distance from the Culver Well stream, after which the site takes its name. (To avoid confusion the site name will be written as one word). There were other springs, till recently, in the neighbourhood.

Athough Portland is known as an island, it is in reality a peninsula which is attached to the mainland by a huge pebble bank known as Chesil Beach or Bank. During Mesolithic times this would have been further out to sea than at present and was probably brought into existence by material washed from a Late Pleistocene Raised Beach which can now best be seen at the southern tip of the Isle, not far from the Culverwell site. Between the Beach and the mainland at Weymouth is a saltwater lagoon known as the Fleet with its exit to the sea at a place known as Smallmouth, at the northern end of Portland. Artefacts of the Mesolithic period are scattered all over the entire length of the landward shore of the Fleet. In the north of the Isle, the terrain rapidly slopes up to the heights of the Verne at about 122m. Finds of Late Pleistocene aspect have been found here (as well as on the Culverwell site and elsewhere on the Isle).

The slope of the hill near the southern end of the Isle on which the site of Culverwell is, is covered with a series of medieval field systems, forming step-like features known as lynchets. These have served to protect the prehistoric remains particularly well in the areas of the so-called positive lynchets, where the deposits above the Mesolithic remains are just over a metre thick. Immediately above the ancient archaeological level is the dark clay deposit of a buried relict soil which everywhere sealed the earlier remains. Despite the fact that the overburden in the areas of the negative lynchets is very shallow, sometimes less than a third of a metre, the archaeological remains are everywhere well stratified and relatively little disturbed.

A series of radiocarbon and AMS dates are in good agreement with each other and have dated the site to c7,500 years ago (uncalibrated). Thermoluminescence dates also support this date.The site was discovered in 1967 and excavated over the next 30 years with the help of volunteers.

The main features of the site comprise: a natural gully, running approximately in a north to south direction down the slope of the hill; smaller gullies in the south; a large shell midden; a floor of limestone slabs on top of the midden; a smaller paved area; a low wall forming a windbreak; a series of hearths; a pit and a number of features with a ritual aspect.

The first settlement on the site appears to have been in the northern part of the terrain on the natural ancient landsurface of rubbly loam while the big gully was utilised as a dumping ground for living debris. It apparently gradually filled up over most of its length (as far as excavated) and was then levelled off on the eastern side. Rubbish continued to be thrown into the deeper parts of the gully in the west.

A floor of flat limestone slabs was placed directly on top of the midden. It measures approximately 31 .09m long and has an average width of 3.6m. As the site could only be excavated up to the edge of the Bill road in the south, the exact limits of the feature are not certain. The floor's edges are not straight and it appears to have been laid down in stages.

At each point where a new stage seems to have started, a larger than average stone was inserted into the floor. Some of these stones were lifted and revealed that under them a small stone-lined hole had been made into the midden. Nothing of significance were found in these holes, except in one case. A roughly triangular stone towards the northern part of the floor also had a large, smooth oval beach cobble placed into the floor next to it. In the stone-lined hole underneath it a pierced scallop shell, an axe of Jurassic chert and a small round pebble, placed on its edge into the midden, were found. This may have been an offering in the nature of a 'foundation deposit' or as a n expression of some or other religious thought to benefit the site occupants. In most of the excavated areas of the gully, concentrations of round pebbles were also found which had been thrown into the bottom of the gully, which at times would have been quite muddy. They also may have had ritual connotations.

In the southern area of the floor a number of larger-than-average stones marked the beginning of a narrow wall of limestone slabs which was constructed at a right-angle to the floor. This too was constructed in stages and partly overlapped the floor, so that it formed an L-shape with the floor. This wall has an average width of 1.5m and was, as far as could be ascertained, about 18m long. It could have acted as a wind-break to the site. At one point underneath it, appeared to be the remains of a small semi-circular stone feature.

On the eastern side of the floor are a number of hearths: three definite separate ones have been excavated but a smaller, fourth one, may be part of the bigger Hearth 1. Hearth 1 is roughly central to the floor, just outside its edge. It has a diameter of about 1.5m and is contaned within a shallow depression of 20-40cms. It was full of charcoal which provided one of the C14 dates. A large number of choppers/chopping tools and other coretools were found lying round the edge of this hearth.

Hearth 4 adjacent the norhern end of the floor, had a diameter of about a metre and was contained within a hollowed-out area on the edge of the gully, which here dips under the floor. Round the south-eastern perimeter of this hearth there was a small area paved with flat limestone slabs. Next to this was a pile of 11 'Portland' picks of Jurassic limestone and chert in various stages of completion. This indicates an area where these tools were made next to the fire.

About 6.5m east of Hearth 1 is a round pit excavated into the original rubbly loam landsurface, about a metre in diameter and about the same depth. Round its edge was a slightly raised band of clay, burnt to a very deep red colour. At the bottom of the pit several large flat limestone slabs had been placed. It was full of midden material when found. The most likely usage of this pit was either a storage tank for fresh rainwater or else for cooking by indirect heat, e.g. by warming up the water with hot stones and hanging a skin bag of food into the water or else by wrapping food up in leaves and placing it within layers of hot midden material.

A great deal of environmental work carried out on the site and in the neighbourhood indicated that at least parts of the site adjacent the gully was more shady than today. The climate would probably have been one or two degrees warmer than today and the sea would have been about 300m further out than today. The beach below the site would have gradually sloped down to the water's edge. There are many caves here which could have been utilised by the Mesolithic food-gatherers.

The main food remains on the site are molluscs, mostly species of limpets, winkles and dog-whelks. It is certain that the diet would have been supplemented by edible plants and possibly some fruits or berries of which charcoal pieces have been found. Some pieces of bone have been found, mostly of pig and one small vole; other very fragmented pieces could include bird. A few crab claws also came from the midden. There would, almost certainly, have been enough food available to sustain the site occupants all the year round.

Preliminary results of oxygen isotope analysis have indicated that at least the majority of the molluscs were harvested in autumn and winter. There are no indications why the site could not, also, have been occupied in the summer as all the necessary resources of food and artefact material would have been accessible all the year round. The complexity of the site arrangements suggests strongly that the site was intended as a long-stay habitation site, and possibly also for most, if not all, the year round. This is also suggested by the rubbish disposal strategies and the fact that the floor on top of the midden would have been well-drained of rain-water and water coming down the hill.

There are no indications of why the site was eventually deserted and many hypotheses for this can be put forward, e.g. a threatened attack by another tribe or some or other natural catastrophe. Various studies suggests that between 15-30 lived on the site in 3 or 4 small huts over a period of 10-20 years (if there were no lengthy breaks in occupation).

Vast quantities of artefacts of fine grey or black Portland chert have been found on the site. The main lithic finds include microliths, particularly scalene triangles, and irregularly shaped picks made mostly of cherty limestone. Experiments on site have shown that these tools were, probably, used for digging up plants or vegetables. Scrapers and variously retouched blades and flakes are numerous and also present are burins, awls, knives, choppers and chopping tools and a number of countersunk pebbles. A special study has indicated that unknapped beach pebbles and other stones were often utilised for various purposes.

Non-lithic archaeological material include shell beads of Littorina and Patella species, ochre of which some appear to have been shaped to form crayons and a few pieces of bone which may have been shaped to form points.

Most of the material for the lithic artefacts are derived either from chert exposures in the cliffs round the Isle of Portland or from pebbles on the beaches, the nearby Chesil Bank or the late Pleistocene Raised Beach near the site. The limestone slabs for building also came from the Holocene deposits ('Head') above the Raised Beach.

Although the Culverwell site compares well in most aspects with other Mesolithic sites in England and Europe of c5,200 BC, some of its aspects are quite unique: the extensive use made of Jurassic limestone on the habitation site, the intensive production of picks, the evidence for ritual (other than burials) and the evidence it provides for site organisation and, at least, a semi-sedentary lifestyle.

The excavation Report contains several contributions by specialist researchers.

Management of the site:

Most of the trenches opened on site during the excavations, have been backfilled by placing geosynthetic material over the features, part-filling with grit and then completing the back-filling with earth previously from the trenches. With the support of two grants ("anson Environment Fund and EU Konver Fund), a part of the floor, midden, one hearth and the ritual feature and the pit have been preserved under a low-level structure of split logs. An extensive management and heritage plan is in place for the whole site and will provide for study and special interest visits.

The site is open during the summer months at published times and otherwise by contacting 01305-861576.

After all work on the archaeological material has been finished, the artefacts will be placed with the archives in the British Museum.

Close this window when you have finished with it.