During the Last Interglacial Middle Palaeolithic industries of Crvena Stijena-type rich in side-scrapers with Levallois technique of recurrent type are specific to the Balkans. These industries have analogies in Anatolia and the northern part of the Middle East (Zagros-Group), but are different from industries typical of the middle Danube basin (Taubachian) and northern Central Europe (Moustero-Levalloisian). In the period preceding and immediately following the Lower Pleniglacial the Balkans were dominated by typical Mousterian and Moustero-Levalloisian, frequently with leaf points, similar to the industries of the lower Danube and Dniester basins, but unknown in western Anatolia. During the same period Eastern Micoquian developed in the middle Danube basin and northern Central Europe. Moustero-Levalloisian with leaf points persisted until the Early Interpleniglacial, but only in exceptional cases developed some Upper Palaeolithic features, and always without typical Aurignacian forms. The Aurignacian, unless it appears as a first Upper Palaeolithic culture in the Balkans with earliest dates in Europe (>40,000 years BP), seems to be an intrusive unit without any roots in the local Middle Palaeolithic. After 30,000 years BP, parallel to the Late Aurignacian, the first industries with backed blades appear. In the early stage these developed independently from those of Central Europe. Only after 26,000/24,000 BP were they followed in the eastern Balkans by assemblages strongly linked both morphologically and by raw materials to the Gravettian of the middle Danube basin. In the western Balkans, after 20,000 years BP, assemblages with shouldered points appeared, also probably of middle Danube origin. During the Last Interglacial and Interpleniglacial the territory of Balkans played an important transitional role between Anatolia and Central Europe; in the two Pleniglacials of the Würm this territory became some kind of cul-de-sac as the refugium for population groups from the middle Danube and northern Central Europe.

Choppers and chopping tools have long been associated with the Clactonian industries of Britain. They have either been dismissed as cores, or often described as woodworking tools, but have rarely been studied from a functional perspective. The purpose of this paper is to publish the results of a series of experiments which has been carried out to investigate the functional efficiency of choppers or chopping tools as compared to other alternative tools. These results are then reviewed in the light of the archaeological information from Clactonian and other Lower and Middle Pleistocene sites.

Due to the problems of definition, for the purposes of the experiments both chopping tools and choppers have been taken to be small nodules of pebbles which have had several flakes removed bifacially along at least one edge. The morphology of the working edge is identical to that found on the edges of many of the Clactonian cores. For this reason, chopping tools, choppers and cores are regarded as artefacts with potentially an identical function. In the experiments they are termed simply as chopping tools.

An Upper Palaeolithic engraving of a female figure has recently been discovered on the base of a naturally hollowed limestone slab possibly used as a bowl or lamp, found in the cave of Courbet (Penne-Tarn, France) and preserved in the British Museum. The slab, decorated on the dished side with linear incisions, and the female figure are described in detail. The occurrence of a female engraving on a utilised slab is found to be unique to date but the relationship of the figure to the support is questioned and discussed. The female engraving is compared with others known from Germany, as well as Quercy and Périgord, France. On the basis of these comparisons and radiocarbon age estimates, the engraved slab is attributed to Magdalenian VI.

In the absence of direct evidence, archaeologists interested in the demography of stone age hunter-gatherer societies are forced to turn their attention to sites as the unit of analysis. The distribution of sites and their varying density across space and through time have been considered to be acceptable, proxy, population records (see for example Constandse-Westermann & Newell 1984,158–64). This note is a contribution to these studies, but one which is not based on the site as a unit of analysis.

Smith & Openshaw (1989) have argued that the conventional archaeological site is a poor unit of analysis in considering regional settlement patterns. It is poorly defined and the recorded distribution of sites is potentially subject to too much distortion by post-depositional factors such as geomorphological processes and landuse practices, and biases inherent in archaeological method.

Excavations were undertaken in advance of construction work at Newbury Sewage Treatment Works on the outskirts of Thatcham in Berkshire, close to the sites of previous excavations undertaken by Wymer and by Peake and Crawford. Worked flint of Mesolithic date was recovered from a sandy layer overlying river gravel in two distinct concentrations suggesting two distinct episodes. Use wear analysis of the flint suggests that the sites were used as home bases at which a wide range of activities took place, with an emphasis on the processing of plant foods. A 14C date of 9100±80 BP was obtained from a sample of hazel-nut shells from within one of the concentrations. Analysis of the soil and sedimentary sequence as well as the pollen indicates constantly changing localised environments in the early Holocene in the Thatcham area, with sporadic occupation by Mesolithic communities on the drier areas at the edge of the floodplain.

This study examines the archaeological significance of the material from a group of Neolithic chipping floors rescued during the rebuilding of the Antrim coast road, at Mad Mans Window, south of Glenarm, Co. Antrim. It shows that the lithic production strategies vary significantly between assemblages although it is presumed that they are all Neolithic in date and come from the same area of coast. It is apparent that flint axe production was of limited importance on these sites and that in spite of the abundance of flint available along the Antrim coast, relatively few polished flint axes were manufactured. Instead the numerous flint caches found in adjacent parts of the north-east of Ireland tend to produce scrapers and blades. Hoards containing arrowheads may be confined to the Bronze Age.

Around 300 polished flint axes and roughouts are known from Ireland. These are frequently small and only partially polished. A limited number of highly polished axes with ground flat side facets have been designated sub-type A. The tendency to use porcellanite rather than flint for axe manufacture may be due to its ability to withstand robust shock.

During the last 100 years, the role of flint as a key resource in the stone age of north-eastern Ireland has always been recognized but this has usually led to an uncritical assumption as to the paramount importance of flint. Work in recent years has shown that its significance in attracting and retaining Mesolithic settlement may have been over-emphasized.

The role of the flint industries in the Irish Neolithic in this region has never been properly assessed, either in relation to older Mesolithic manufacturing traditions or in the broader context of supply to the Neolithic communities of this part of Ireland.

In particular, good or even reasonable quality flint is usually only exposed in Cretaceous outcrops along a narrow strip on the edge of the basalt plateau and, therefore, has a very limited availability in parts of Co. Antrim as well as parts of Counties Down and Deny. As a contrast, erratic and beach flint is available in some quantity down the east coast of Ireland from Co. Down to Wexford. A second potential constraining factor is that unlike Britain, where flint was exploited for axe manufacture in the east and other rocks in the west, flint sources and porcellanite for axe manufacturing are both found adjacent to each other in the same corner of Co. Antrim. In particular, a number of more substantial chipping floors of Neolithic age are known, e.g. the opencast quarry sites at Ballygalley Head. The purpose of this study is to assess the role of flint production on the Antrim coast with particular reference to its significance in the Neolithic. This topic will be developed in the context of an analysis of the material found at Mad Mans Window near Glenarm.

Steven Mithen's ‘Ecological Interpretations of Palaeolithic Art’ (PPS 57, 103–14) reminded me of a Chinese meal — initially satisfying, but it doesn't stick with you for very long. While I subscribe to broadly similar paradigmatic biases at the level of the metaphysic, the ‘thoughtful forager’ model itself, proposed to relate various aspects of the art under the aegis of a particular kind of adaptationist perspective, seems to be conceptually muddled and operationally problematic. Also, Mithen's starting-point, the notion of an inherent contradiction between human creativity and an adaptationist point of view, is a red herring—wherever did he get it?! I will confine these brief remarks to three points that bear on different conceptions of adaptation and how they effect construals of pattern and the meaning of pattern in Palaeolithic art. I also respond to referees' comments.

Mithen takes me, Straus and Gamble to task for omitting the individual and individual decision-making in our conceptions of adaptation (pp. 104, 105). A conception of adaptation that is focused on the group is juxtaposed with one invoking selection operating at the level of the individual organism in a direct analogy with group vs. individual selection in biological evolution.

Recent arguments about the origins of the Predynastic in Upper Egypt have posited connections between the lithic industries of that region and those of the Western Desert. It has been further claimed that a mid-Holocene arid period may have led the inhabitants of the desert regions to migrate to the Nile Valley. This essay argues that the evidence of stone tools, as well as ceramic evidence from Dakhla Oasis, fail to support the ‘desert migration’ model, at least as it applies to Dakhla Oasis.

The excavation of an oval crop mark close to the Abingdon causewayed enclosure showed a complex sequence of development, starting with a rectangular ditched enclosure and most probably ending with an oval barrow of a type with parallels elsewhere in lowland England. The site included the grave of two individuals associated with a polished knife, a belt slider and most probably a leaf shaped arrowhead, and produced a series of radiocarbon dates extending from the Earlier to the Later Neolithic. A number of formal deposits around one end of the site are matched by similar material from the inner ditch of the causewayed enclosure, suggesting a direct link between the two monuments.

From 1946 to 1952. excavations were undertaken in advance of destruction by gravel workings of a series of Neolithic and Early Bronze Age monuments at Dorchester-on-Thames, Oxon. These included a long enclosure, a cursus, a double ditched henge, pit circles and ring ditches with primary and secondary cremation burials and a notable Beaker burial. Sites I, II, IV, V and VI, all pit circles or ring ditches, were published in 1951. This report describes sites III, VIII, IX, XI, XII, XIII and XIV. By the early 1950s much of the Neolithic complex had been quarried for gravel, and other adjacent areas subsequently were dug away. In 1981 the construction of a bypass led to the excavation of further surviving parts of the complex: site 1, a long D-shaped enclosure incorporated in the southern end of the cursus, and sites 2, 3 and 4.

Both sets of sites are presented together, largely following the chronological scheme proposed for the complex by Bradley and Chambers (1988). Site VIII and site 1 are long enclosures of Earlier Neolithic date, with human remains. The latter site has a calibrated date of 3773–3378 BC. Site III is a cursus at least 1600 m long which cuts site VIII and incorporates site 1. A date of 3360–3040 BC was obtained from the primary fill of its ditch. Site XI is a three-phase ring ditch, perhaps successive enlargements of a barrow, but its innermost ditch is cut by a pit circle with cremations. There are Ebbsfleet sherds in the outermost ditch, and dates of 3037–2788 and 3024–2908 BC from the innermost, perhaps primary, ditch. Site XIV is a ring ditch succeeded by the Big Rings henge. Site 3 is a post circle between the ditches of the southern part of the cursus, with some secondary cremations. It has dates from the outer wood of its posts of 2890–2499, 2886–2491 and 2872–2470 BC; dates on charcoal associated with secondary cremations were 2880–2470, 2870–2460 and 2123–1740 BC. Site 2 is a penannular ring ditch with a primary date of 2912–2705 BC, and secondary cremations. Site XII is a notable Beaker burial within a two-phase ring ditch. An adult man was laid crouched on a bier and was accompanied by a fine W/MR beaker, a stone bracer, a tanged copper knife and a small riveted knife with at least one rivet of tin bronze. Traces of a stretcher-like feature were found in an adjacent pit which cuts the inner ring ditch. Site XIII, the Big Rings, is a large double ditched enclosure with central bank and opposed entrances. There was Beaker pottery in the primary fill of the inner ditch, but very little material was found in either ditch, and there were minimal features in the interior, part of which was stripped. Site 4 consists of two conjoined ring ditches within the southern end of the cursus. It encloses cremation burials. One, accompanied by an awl and Collared Urn, has a date of 2290–1910 BC. Traces of a ditched field system (originally referred to as a droveway, site IX) of later Bronze Age date were found cutting sites III, VIII and XIII, and paired ditches on site 1 may be related.

The complex is related to its local and regional context, and the sequence of development is discussed. A timescale measured in generations is advocated, with phases of continuity and episodes of little activity both represented. Several aspects of monuments are considered: their differing scales including the monumental, their ability to endure and create tradition, and finally the choice of designs, from the locally customary to the exotic.

Radiocarbon dates on samples excavated earlier this century by Gray and Keiller and one more recently by Pitts, confirm that the henge at Avebury is in the same general Late Neolithic horizon as the other great Wessex enclosures of Durrington Walls, Marden and Mount Pleasant. They leave open to interpretation the details of Avebury's sequence. The monument may have been constructed in a series of stages, but further investigation is badly needed.

This paper presents a re-assessment of the date and constructional sequence of the Sanctuary, a later Neolithic timber and stone setting on Overton Hill, near Avebury. Previous interpretations involving elaborate phasing and a protracted chronology for the site are rejected. Instead, it is argued that the site represents a single or double phase monument, constructed around 2500 cal BC and associated with pottery of the Grooved Ware tradition. The character of the pre-monument activity, construction and use of the site is examined. Patterns of formal deposition, involving pottery, lithics and human bone, are recognized, and considered in the context of the site's architecture and the structured organization of space, access and movement.

Five pollen diagrams from two mires in the province of Huelva in south-west Spain are re-evaluated in the light of 15 radiocarbon dates, and multivariate numerical analyses of modern and fossil pollen samples in the region. The results of these analyses suggest that managed woodland of dehesa type can be detected in pollen diagrams, and can be interpreted within the archaeological sequence known from the area.

Detailed topographical surveys and geophysical examinations of the Penrith ‘henges’ were undertaken in order to assess the present state of the monuments for management purposes. At Mayburgh geophysical prospection was designed to assess the validity of the early reports of the presence of two concentric settings of stones. Some possible parallels in Ireland are noted. The nearby site of King Arthur's Round Table was also sampled by geophysical techniques, but was found to be too disturbed for the recovery of any further information as to its original form. The ‘cremation trench’ discovered in pre-War excavations at this site was relocated. The position of the enigmatic Little Round Table was re-established by both topographical and geophysical methods, and its north entrance firmly located. The classification of Mayburgh and of the Little Round Table is considered. The apparent disunity of these three distinctly different monuments within one complex is emphasised—all are traditionally classified as ‘henge-related’, yet the structure and orientation of each site is markedly different. Contemporary topographical considerations are now difficult to reconstruct.

Small-scale rescue excavations at Llwyn Bryn-dinas hillfort, on the Welsh Borderland, showed that the earliest fortification belonged to the late Bronze Age, with radiocarbon dates in the late 9th and 8th centuries be. A terrace, subsequently cut into the rear of the rampart to accommodate a metal-working floor, was associated with a radiocarbon date centred in the late 3rd century bc. Detailed analysis of the metal-working debris suggests that copper-alloy casting, iron forging and possibly bronze production were carried out within a single workshop. The finds include a distinctive form of handled crucible. The industry appears to have been fairly small-scale, of short duration, and probably only designed to meet the internal needs of the hillfort population. The excavation adds significantly to the local evidence for metal-working during the later prehistoric period. A distinctive zinc-impurity pattern in the copper alloy and raw copper, previously identified in material from other sites nearby, confirms the suggestion of an Iron Age bronze-working industry based on a specific metal source in the north Powys area. In addition, analysis of the iron-working debris suggests the exploitation of a distinctive local ore body. The metal-working activity appears to have come to an abrupt end, possibly with the enlargement or local repair of the rampart. Later phases of activity include a final occupation deposit with an associated radiocarbon date centred in the mid 2nd century bc.

The excavations of the cemetery groups at Balneaves, Loanleven and Park of Tongland facilitate an examination of many aspects of Bronze Age burial practices in Scotland. They are notable as much for the differences in burial ritual they imply as for the very narrow chronological period in which they were used. The three sites produced a total of seventeen 14C dates, two of which are aberrant, with means of the remaining fifteen falling within a period of 250 years (3370–3610 bp in radiocarbon years). The excavations were sponsored by Historic Scotland (formerly Historic Buildings and Monuments, Scotland).

At Balneaves, a penannular ditch enclosed sixteen features, including a group of seven pits with cremation burials, four of which were associated with a distinctive assemblage of collared urns. The cremated bone was well preserved. At least one large standing stone had been erected on the site, and this was buried in the medieval period.

At Loanleven, only a segment of the enclosing ring-ditch survived, within which were four cists, two containing inhumations and two cremations, one of the latter (Cist 2) associated with a fragment of a food vessel. A decorated slab, in so-called ‘Passage Grave Style’, was recovered from Cist 1, and the same cist produced palynological evidence for grave furnishings in the form of a mat of plant material which probably underlay the body. 14C dates give a terminus ante quem of 3620±50 bp (GU–2543) for the re-use of the decorated slab, and a terminus post quem of 3410±50 bp (GU–2542) for the food vessel grave.

Park of Tongland, regarded as a Four-Poster stone circle, was excavated after the fall of a standing stone. It was shown to be of multi-period construction, consisting of a cairn which overlay seven pits containing fragmentary cremation burials, two associated with collared urns. The standing stones may not all have been erect at the same time. A series of 14C dates fell within the range of 1480–1530 bc.

Excavations at Ewanrigg, Maryport, Cumbria (NY035353) took place in 1983 and 1985–87. The site of a Bronze Age cremation cemetery was discovered whilst fieldwalking a crop-mark enclosure site; this site had been trial excavated in 1956 and shown to be a Romano-British settlement. During the excavations 28 burials were discovered, 26 being cremations and two inhumations. Both inhumations, one a Beaker burial and the other a cist burial with a Food Vessel, had been disturbed. The Bronze Age pottery assemblage was a mixture of Collared Urns and Food Vessel Urns; the Collared Urns are mainly Secondary Series with one showing some Primary Series traits. Fragments of two Beakers were discovered, one an N/MR Beaker and the other more in the long-necked Northern series tradition. All the pottery, except the N/MR Beaker, was made from local clay. Within one of the cremation burials a clay connecting rod for a furnace was discovered and apart from one other site this is the only discovery which shows any link between metal-working and the burials within the Collared Urn tradition. Also within the cremations were a number of toggles and pins made from animal bone. The human cremated bone was sufficiently well preserved to allow analysis to show that there were six female and five male burials. Radiocarbon samples, mainly on charcoal, gave a date range for the Collared Urns of 2460–1520 BC (calibrated to two standard deviations).

About 170 round barrows and 15 flat graves from Kent are here recorded, almost all on the Chalk. In addition there are at least as many ring-ditches, mostly of ploughed-out round barrows, in Thanet, only a small selection of which are here included as a full list is being prepared by the Trust for Thanet Archaeology. Nine sites have yielded primary central inhumations and ten have contained primary central cremations. Four sites have yielded burials in their encircling ditch. Warrior equipment has come from two sites, and ‘Wessex’ interments from another two. Barrow Wouldham 1/THAN contained a central primary cremation in a four-post mortuary structure with parallels in the Toterfout-Halve-Mijl cemetery in the Netherlands, with which area relations are suggested. One site (Lydden 1/DOV) is identified as the Ellenbeorge of an Anglo-Saxon land charter. Several Bronze Age round barrows were used by the Jutes and Saxons for their intrusive burials, and there are two instances of an Anglo-Saxon cemetery being sited to adjoin a Bronze Age round barrow. More than twenty sites have recorded local names of which the most frequent general name is Mount. Some of the best surviving barrows are beside or near the North Downs Way and the Pilgrims' Way.

Long-handled combs, predominantly made of antler — although some are of bone or perhaps whalebone — and traditionally associated with weaving, have been thought to be an almost exclusively British phenomenon (Hodson 1964, 103). They are first found in middle or later Bronze Age contexts, but are more usually associated with the Iron Age and go out of use in the later 1st century AD. Although dating of these tools is imprecise, on some sites such as Glastonbury (Bullied & Gray 1911) and Maiden Castle (Wheeler 1943, 298) one can say that the earlier ones are plain and later ones decorated, often with dot and circle and linear patterns. However evidence from Danebury shows the reverse to be the case (Cunliffe & Poole 1991). The combs measure between 80 to 222 mm with an average length of 150 mm. They are mostly dentated only at one end, although some are double-ended. There are usually between 8 and 13 teeth and at the other end the shape of the butts can be plain or they may have squared or rounded ends with or without perforations.

Anna Roes (1963) drew attention to a comb from the Frisian terps in the Netherlands (fig. 1, no. 1), housed in the Fries Museum, Leeuwarden. She considered that it may well have been imported from Britain. The comb is undecorated but polished. Its length is 141 mm and width at teeth 37 mm and at the butt end 18 mm. The comb is very concave at the dentated end. There had originally been 12 teeth of which only one outer tooth remained and this is worn on its inner side. Recently two further combs have been found in the Netherlands on two domestic sites of the pre-Roman Iron Age (sites 15.04 and 16.59). Both were excavated by A. Abbink from the Institute of Prehistory of the University of Leiden in 1989 and 1990 respectively (Abbink 1989; 1991).

This paper starts by outlining the history of stone axe studies in Ireland, from their antiquarian beginnings to 1990. It then offers a critical review of the current state of knowledge concerning the numbers, distribution, findspot contexts, morphology, size, associated finds, dating and raw materials of stone axes. Having proposed an agenda for future research, the paper ends by introducing the Irish Stone Axe Project—the major programme of database creation and petrological identification, funded by the National Heritage Council, currently being undertaken by GC and EG.