During the Easter and Summer of 1958 a programme of investigations into British Upper Palaeolithic cave deposits was initiated on behalf of the Prehistoric Society, with the aid of a grant from the Research Fund. The work was further supported by the Crowther Beynon Fund of the Museum of Archaeology and Ethnology of Cambridge. Labour in the field was provided by students in the Department of Archaeology and Anthropology of the University, with notable assistance from several members of the Society in different areas.

The prime objectives of the work, which is still in progress, are to define more precisely the character of the different stages in the British Upper Palaeolithic, and to study them against their chronological and environmental background. In this way it is hoped to throw light on wider problems of the relation of British finds to the rapidly emerging picture of the Late Glacial hunting communities of Central Europe and the Low Countries.

In 1951 Dr Werner Kissling, who for many years has been studying and collecting specimens of the old folk culture of the Outer Hebrides, asked me to undertake the excavation of some site on South Uist (fig. 1) to encourage the local people to take an interest in their antiquities. The site chosen was a large midden among the dunes in the Kilpheder machair (sandy coastal plain). Five hundred yards to the northward a midden with a few stones is all that remains of a wheel-house, known as Bruthach an Tigh Tallan (the brae of the buried house), which was completely removed some years ago. Half a mile to the north of this another wheel-house, Sithean a Phiobaire (the piper's fairy hill), was removed to build a wall round a nearby grave-yard. The middens of both these wheel-houses have produced ring-headed bronze or iron pins similar to ones found in the North Uist wheel-houses and approximately dated at Traprain Law in Haddingtonshire to between A.D. 100–200. To the south-west of our site, one about 150 yards distant, the other some 200 yards further on, are two more middens with traces of building stones. One of these produced a specimen of the well-known long-handled weaving comb. Yet another midden of the same period, two hundred yards to the south east, is now completely buried. It will be seen that this was once a highly populated area. I have not enumerated all the sites which are known to exist.

Some recent literature on the problem of hoards is reviewed, dealing particularly with the distinction between votive deposits and stores of objects. An alternative approach is developed, considering the long-term variations in the nature of intentional deposits in European prehistory. Three stages are suggested. In the initial stage, a unitary system of deposits prevailed, especially of food and selected artefacts, including metal when available. In the developed stage, characteristic of the Bronze Age, a dual system allowed both votive deposition and accumulation and recycling, especially of non-local metal. The final stage, in the later Iron Age, saw renewed emphasis on votive deposits, especially on the borders of emerging polities.

The excavations to be described were carried out in the autumn and winter of 1941–42, and a total of fifteen barrows was examined. Of these, ten were on Beaulieu Heath, to the south of the Brockenhurst–Hatchet Gate road, and the remaining five near Stoney Cross to the north-west of the New Forest. All the barrows were about to be destroyed in the course of building operations, and it was due to the urgency of the situation that none of the barrows could be more fully excavated. The time of year, and, in the case of Beaulieu Heath, the lowness of the site, made the undertaking in wet weather at least, an extremely difficult one. It is therefore with full knowledge of its incompleteness that I must present this report.

Six men were employed at both sites, with Mr Rogers as foreman. I would like to express my gratitude to him for his very willing co-operation. Other helpers included Messrs Richard Atkinson, Philip Suggett, L. V. Grinsell and Aubrey Parke, Miss Jocelyn Morris and Miss B. de Cardi, and Mr and Mrs Maxwell-Hyslop. I am most grateful to all of them, and more especially to Miss Jocelyn Morris and Mr Atkinson, to whose skill and patience is largely due the recovery of rhe remains of the mortuary-house in Barrow II. Mr B. H. St. J. O'Neil visited both sites several times. Thanks are also due to Dr F. E. Zeuner for his remarks on the soil specimens, to my husband for a note on the Barrow II mortuary-house and to Sir Cyril Fox for his pottery report.

For many years a programme of examining stone implements has been organized by the Council for British Archaeology's Sub-Committee for Implement Petrology. The East Anglian aspect of this programme, covering the counties of Norfolk, Suffolk, Cambridgeshire and Essex, has long been active, being administered from Norwich Castle Museum under the auspices of the CBA, first by the late R. Rainbird Clarke and more recently by the authors of this report. The petrological work has been very largely undertaken by Dr F. S. Wallis and Mr E. D. Evens in co-operation with the Geological Department of Bristol University. The results embodied in this report, with those for Yorkshire recently published (Keen & Radley, 1971) and other identifications noted in the CBA's Annual Reports, and with the Fifth Report on south-western implements which appears elsewhere in these Proceedings (Evens, Smith & Wallis, 1972), may be taken together as a measure of the debt which prehistorians owe to Dr. Wallis and Mr Evens—the ‘Evens and self’ of much correspondence.

This paper is dedicated to John Evans, environmental archaeologist extraordinaire, who died 14 June 2005, while this paper was in press. He continually reminded us that environmental data should address questions of people and landscape and be relevant to the understanding of Prehistory by our archaeological colleagues.

The Beaker period in north-west Europe is abound with objects, burials, and monuments, but evidence of settlement and domestic life is often absent or less easily found, and England is no exception. Despite the thousands of barrows with rich artefacts assemblages (eg, Amesbury Archer) and the numerous pits with non-domestic assemblages of placed items, evidence for houses and settlement are sparse despite the indication of increased agriculture and sedentism. This paper explores this problem on the chalklands of southern England that are rich in Beaker finds, and which are generally recognised as one of the best studied and well understood landscapes in Europe. From this study it is suggested that Beaker domestic sites are present, but are often in low lying positions on the chalk downs and have subsequently been buried by variable depths of hillwash, making them invisible to normal archaeological survey and reconnaissance.

The Vale of Glamorgan (fig. 1, 1) is the middle part of the coastal plain of South Wales. An undulating belt of country much dissected by river systems, its highest parts consisting of knolls and ridges which rarely achieve a height of more than 400 ft. above sea-level, it was occupied fairly intensively throughout prehistoric times. Round barrows are found scattered throughout its length. But they are particularly concentrated in one area in the middle portion of the Vale, where, between the ancient towns of Llantwit Major on the coast, and Cowbridge, about six miles inland, between 20 and 30 have been recognised. None of these barrows has hitherto been scientifically examined, nor do they appear to have suffered damage apart from ploughing. The excavation here described was undertaken by the Archaeological Section of the Cardiff Naturalists' Society in the hope that from it might be derived definite information bearing upon the intensive Bronze Age occupation of this part of the coastal plain which the barrows clearly attest.

This paper attempts at reconstructing the life cycle of flint bifacial tools in two major technological trajectories, each aimed at a different end. It is focused on discarded polished flint bifacial tools (axe, adze, chisel) and bifacial polished debitage items found in Neolithic and Chalcolithic sites (9th–6th millennium BP) from the Southern Levant.

Recent studies of Holocene flint assemblages revealed special types of debitage items such as blades, flakes, cores and core trimming elements, all bearing traces of polish. The discovery of these polished debitage items in Levantine sites enabled a study of bifacial tool resharpening and recycling techniques, emphasising the exceptional attention paid to this tool category. Since polished bifacial tools appear in many old world Holocene archaeological contexts, the data and interpretation presented in this study have implications far beyond the Southern Levant and could relate to universal technological properties of flint axes and other bifacial tools.

For many years archaeologists interested in the study of the Palaeolithic in North Africa and Eurasia have been using cumulative percentage frequency graphs for the comparison of prehistoric artefact assemblages. For examples we refer to the references at the foot of this text. However, it is perhaps time to carefully review this technique and its future utility.

Statistical techniques and mathematical models are slowly infiltrating and reshaping the discipline of archaeology—increasing the power and depth of both analysis and synthesis. It is perhaps already possible to distinguish the cumulative advance of these techniques in archaeology from the initial role of demonstrative aids and methods of data display towards an increasingly powerful analytical role with a capacity for predictive inference. In the ranks of the first generation we have the early use of graphs, frequency polygons and histograms for mapping severely limited numbers of percentages or attribute ratios. In the second generation of statistical techniques the impact of the computer is felt for the first time and archaeology is developing an array of exploratory attempts to integrate the probing capacity of such methods as factor analysis, matrix analysis and principal component analysis.

Plumpton plain is situated on the top of the South Downs, roughly 600 feet above sea level, six miles north-east of Brighton and four miles north-west of Lewes (fig. 1). From its western end a broad spur slopes gently southwards from the northern escarpment of the Downs, lying between Moustone Valley on the south-east and Faulkners Bottom on the west. Most of this Downland is covered with a dense scrub of gorse, thorn and bramble, and with large patches of bracken and heather. A series of broad paths running roughly at right angles with one another has been cut through this vegetation to facilitate the preservation of game. Along the main ridge of the spur running north and south is a broad gallop which, at a height of 600 feet O.D., passes through a group of earthworks situated 1500 feet from the north edge of the Downs and 2300 feet east of Streathill Farm. This group was the primary object of bur investigations and will be referred to as Site A (fig. 2).

Site B (fig. 3) lies a quarter of a mile to the south-east of Site A on a small lateral spur jutting between the twin heads of Moustone Bottom. The only visible evidence of prehistoric occupation was a quantity of coarse gritty sherds and calcined flints on the surface to the south-east of a low bank and ditch which runs across the spur.

Several groups of lynchets enclosing square Celtic fields are to be seen in the neighbourhood of these two sites. They lie principally to the south-east of Site A and to the south of Site B.

A wealth of cave sites makes southern Belgium the most important area for understanding the north-western European Early Upper Palaeolithic. However, despite their abundance, the interpretation of many assemblages remains problematic. Here we present a new study of lithic material from layer B of Trou du Renard (Furfooz, Namur Province) and consider its place in the Belgian Aurignacian. The assemblage is typical of Late Aurignacian assemblages found across western Europe, underscoring the contrast between the Aurignacian and the periods that pre- and post-date it, when we instead see profound differences between north and south. The assemblage is apparently unmixed, distinguishing Trou du Renard from other key Belgian Aurignacian cave sites. A large proportion of the site’s lithic assemblage documents the production of small bladelets from carinated/busqué burin cores, suggesting that Trou du Renard served as a short-term hunting camp. Radiocarbon dating cannot pinpoint the assemblage’s age, though here it is argued to be c. 32–33,000 bp (c. 36–37,000 cal bp) on the basis of its similarity to the well-dated Aurignacian assemblage from Maisières Canal (Atelier de Taille de la Berge Nord-Est area). For the same reason a third assemblage – Trou Walou layer CI-1 – is also argued to be contemporaneous. Trou du Renard, Maisières Canal and Trou Walou may represent three points in the same Late Aurignacian landscape. Differences between their lithic assemblages can be explained by the acquisition and transport of flint, and by a desire to produce small bladelets of highly standardised form irrespective of the size and shape of available blanks.

Excavations at a deeply alluviated site near St Ives, Cambridgeshire, revealed a complex sequence of boundary works of later prehistoric and Roman date running along the edge of a former course of the river Great Ouse. The most significant of these were two successive pit alignments constructed in the early-mid 1st millennium BC. One alignment ran along the very edge of the channel and was waterlogged over much of its length; upon excavation producing a rich assemblage of worked wood, including hedging debris. Broader discussion on the social context of pit alignments as boundary systems is offered.

Skomer is an island off the west coast of Dyfed, the archaeology of which, consisting of fields, small enclosures, cairns, huts and dams, is well preserved. The broad outline and much detail was recorded by W. F. Grimes in 1950 and it was demonstrated that about two-thirds of the island is covered in a more or less undisturbed prehistoric landscape, preservation being due to the absence of Medieval and later settlement. The fieldwork described in the present paper was designed to record in detail small enclosures, or pounds, and habitation sites, mainly at a scale of 1:100 which was not done by Grimes. In addition two sets of air photographs were used to make a map of the fields. The work was done as a student exercise and is accordingly largely descriptive. However, analysis at two spatial levels has been attempted, one for huts and pounds and one for settlement complexes as a whole. But beyond the island itself there has been no comparative survey. The remains could be of any age from the Neolithic to early Roman period.

It is widely accepted that between the beginning of the Early Neolithic period and the end of the Early Bronze Age different regions of Britain were connected to one another by sea, but little is known about the nature of maritime contacts before plank-built boats developed during the 2nd millennium bc. This paper considers a series of coastal sites, some of which were first settled from Mesolithic times. From the early 4th millennium they were also associated with artefact production and the use of imported objects and raw materials. Their distribution focuses on the region of isostatic uplift in northern Britain where the ancient shoreline still survives. It is considered in relation to a new model of coastal change which suggests that these locations were characterised by natural havens sheltered behind islands or bars. The sites can be compared with the ‘landing places’ and ‘beach markets’ discussed by historical archaeologists in recent years.

The barrow, the excavation of which is described below, is one of the two situated close to the Cranborne-Wimborne road, some 150 yards north of the Avenue Lodge entrance to St Giles' Park, in the parish of Edmondsham, Dorset.—Nat. Grid. Ref. SU 04611155 and SU 04631160, and map (fig. 1, no. 3).

The attention of the Ministry of Works was drawn early in 1959 to the denudation by ploughing of Barrow 2. Permission to excavate was given by Lord Shaftesbury, the owner, and by Mr Lawes, the tenant farmer. Excavation on behalf of the Ministry of Works was planned and carried out by the writer, with the assistance of two workmen and some volunteers, during July and August 1959.

The Edmondsham barrows are situated at the foot of the long, gentle slope down from St Giles' Crossroads, at the edge of the Park, on a slight rise in the chalk. The land in this area is fairly heavily wooded, and is not typical downland, although the sub-soil is chalk. The junction of the chalk and the clay-with-flints lies only half a mile to the east of the barrow under discussion, at the village of Edmondsham.

It is twenty years since Dr Alex. Curle published his conclusions on the problem of the brochs and it does not diminish our regard for the great services which he has given to the study of the Scottish Iron Age that his article did not end the discussion. About the same time Dr Watson and Professor Brögger expressed their views, and the Scottish Royal Commission on Ancient Monuments summarised the results of their survey, a summary which they have brought up to date in their recently published volumes. In the intervening years Professor Gordon Childe has made important contributions, and one valuable broch excavation has been reported. And yet it cannot be said that the central difficulty has been resolved.

Ethnographic data has been used to show that ‘most differences between hunters and farmers are a matter of degree, and there is no sharp dividing line between the two means of subsistence’ (Orme 1977, 46). The early Neolithic of southern England, as defined from the archaeological record, is likely to reflect the adaptations of a mature techno-economic system to an environment already much altered by the activities of late Mesolithic groups and immigrant pioneer ‘agriculturalists’. Case has shown the need for a long period of pioneer activity during which contact with indigenous groups would have taken place and hunting and gathering provided an essential supplementary means of subsistence (1969). By the later Mesolithic, internal population growth and the spread of closed forest, which affected the availability of food resources, may have created internal pressures on the mobile hunter-gatherer subsistence base (Jacobi 1978, Evans 1975). Zones of particularly rich resources, such as the Kennet and Thames valley, may have provided conditions which allowed a reduction in group mobility. Attempts may also have been made in other areas to increase the resources available by limited forest clearance, more controlled use of plant resources, and an increase in the exploitation of coastal environments (Jacobi 1973 and 1978, Mellars 1976, Clarke 1976). Thus, during the crucial period of contact both systems were under stress, and it is no longer valid to attempt to identify a Late Mesolithic with a purely hunter-gatherer economy which was rapidly replaced by a homogeneous group of sedentary agriculturalists with a distinct artifact assemblage.

The rock carving site of Bro Utmark in Bohuslän is one of the largest in Sweden. Its hundreds of carved images reflect a variety of Bronze Age concerns, and the boats, warriors and animals depicted demonstrate the work of master carvers in the period 1600-300 BC, but concentrated in the later Bronze Age (1100–600 BC). The site is central to a concentration of about 200 carved sites, burial mounds, and other monuments. During the Bronze Age, Bro Utmark lay near the shoreline of a sea which was 15 m higher in relation to the land. The evolving cultural landscape of the Bronze Age suggests incremental settlement in the emergent land, with shorelines and wetlands providing focal points for activity.