Doggerland: the Cultural Dynamics of a Shifting Coastline

Downloaded from http://sp.lyellcollection.org/ at University of Edinburgh on March 10, 2013

Geological Society, London, Special Publications

Doggerland: the cultural dynamics of a shifting coastline

Bryony J. Coles

Geological Society, London, Special Publications 2000, v.175; p393-401. doi: 10.1144/GSL.SP.2000.175.01.27

Email alerting click here to receive free e-mail alerts when service new articles cite this article Permission click here to seek permission to re-use all or request part of this article Subscribe click here to subscribe to Geological Society, London, Special Publications or the Lyell Collection

Notes

BRYONY J. COLES

Department of Archaeology, School of Geography and Archaeology, The University, Exeter, UK (e-mail: [email protected])

Abstract: The landmass now covered by the North Sea, here referred to as Doggerland, has had an important but neglected influence on the course of prehistory in northwestern Europe. The physical character of Doggerland in the Late Glacial and earlier Holocene is assessed, together with its re-colonization by humans after the Last Glacial Maximum. The development of a maritime-based society along the northern coast of Doggerland is postulated, and it is argued that the coastal inhabitants, with their specialized adaptation to this zone, will have moved with the coast as relative sea-levels changed. The interactions of coastal and inland populations are considered, including the probable influence of the coastal groups in delaying the spread of farming into the region.

In northwestern Europe, from ireland to south- recently been modelled in a series of papers by ern Scandinavia, there is an absence of archae- Lambeck and colleagues (e.g. Lambeck 1993, ological evidence dated to the Last Glacial Lambeck et al. 1998) which should in due (Devensian/Weichselian) Maximum. Conditions course enable greater precision and confidence approximating those of an arctic desert per- in coastline reconstruction. The more that is tained in front of the British and Scandinavian known of the former landmass, the more its ice sheets, and it is generally accepted that the relevance to Late Glacial and Postglacial pre- lack of evidence for human occupation is due, history becomes clear (Coles 1998, 1999). In the quite simply, to the absence of humans. For present context, the focus will be predominantly 10000 years, from about 23000Be to about on Doggerland's coastal zone, following a brief 13 500BP (radiocarbon years), the region was consideration first of inland conditions at the uninhabited (Housley et al. 1997). As conditions time of the Windermere/Bolling-Allerod inter- ameliorated, people began to explore the land stadial and then of the initial character of human beyond their glacial refugia; the changing occupation of the land. character of the archaeological evidence indicates that, within a few centuries of pioneering Topography of Doggerland visits, permanent human occupation was established. For southern Britain and southern Jut- The combined effects of erosion and silta- land, the presence of settled groups of people tion, together with the physical difficulties of can be dated to about 12 400 BP, i.e. to the earlier underwater survey, make it difficult to determine part of the Windermere or Bolling interstadial. the detailed topography of Doggerland. Occa- At this time, land was continuous between the sionally, coring pin-points a former freshwater two regions, and it too will have been inhabited. lake, or commercial survey incidentally pro- The land between, subsequently submerged by vides evidence for former surface features (Firth the North Sea, is here referred to as Doggerland, 2000). Major features, resulting from Quatern- named after the Dogger Bank which has long ary glaciation and its aftermath, can be identi- been recognized as a former area of dryland and fied for some locations, and others can be fresh waters (Reid 1913). Although there has as postulated. For example, the exposed land was yet been no specifically archaeological survey of cut by deep incisions, which had been eroded out the region, something of its character can be as subglacial drainage channels under the outer gleaned from the results of geological exploration margin of ice-sheets. When the ice wasted, the and by extrapolating from the data available for channels were exposed as tunnel valleys, long, adjacent regions. The evidence for the physical narrow and steep-sided, sometimes sinuous and condition of Doggerland in successive periods is often partially filled by a freshwater lake. Tunnel discussed in Coles 1998, including that for coast- valleys 1-3 km wide, 100m deep and 25-60km line position and for the major river courses. long were present in the area between what is Sea-level change in northwestern Europe has now northeastern Scotland and what is now the

From: PYE, K. & ALLEN, J. R. L. (eds). Coastal and Estuarine Environments: sedimentology, geomorphology and geoarchaeology. Geological Society, London, Special Publications, 175, 393-401. 0305-8719/00/$15.00 6) The Geological Society of London 2000. Downloaded from http://sp.lyellcollection.org/ at University of Edinburgh on March 10, 2013

394 BRYONY J. COLES

Dogger Bank. A further group of valleys, not Doggerland had three major river networks, quite so deep, extended eastwards from the two draining to the north and one southwards present Humber estuary to the south of Dogger to the Channel River (Fig. 1) (Coles 1998, Bank, which itself is likely to have formed a p. 54-57). The northeastern network was domi- substantial upland mass. At the Weichselian nated by the Elbe which flowed to the west of maximum, land beyond the ice-sheet margins is Jutland through the Urstromtal, a vast valley still likely to have been uplifted as a glacial fore- identifiable on the sea-floor, and on across eastern bulge, an effect which only slowly dissipated Doggerland to an estuary opening into the with de-glaciation. The present Dogger Bank Norwegian Trench. The northwestern network may therefore represent former Dogger Hills had a smaller catchment, draining the region west which extended further and higher than present of the Dogger Hills together with much of what is bathymetry suggests. now eastern England, and flowing to an estuary

O" I ...

;~ IcE SHEET ' \,

BERGEJ HILLS

ICE SHEET

Rhin~ J

Doggerland: .... l~---'~" Hypothetical Maximum Extent I,--.--- of Land, t~ Later Devensian/Weichsetian l 0 200 krn /|

Fig. 1. Doggerland in the Later Devensian/Weichselian, with an indication of the major river systems. The evidence on which Figs l-4 are based is discussed in Coles 1998. Downloaded from http://sp.lyellcollection.org/ at University of Edinburgh on March 10, 2013

DOGGERLAND: THE CULTURAL DYNAMICS OF A SHIFTING COASTLINE 395 between the Dogger Hills and the Yorkshire development of open grassy parkland vegetation Wolds. South of Doggerland's main watershed, interspersed with patches of light birch wood- the Rhine and the Thames met to flow south- land and a few other trees (e.g. Kolstrup 1991; westwards into the Channel River and out to the Tipping 1991). There will have been many areas Atlantic, with an estuary well to the west. of wetland vegetation fringing shallow lakes and streams and rivers. Herds of grazing and brows- Flora and fauna ing mammals colonised the land, accompanied by their predators including humans. Direct The opening of the Windermere/B~lling-Allerod evidence for Doggerland's fauna is provided by interstadial (Fig. 2) was marked by a rapid rise bones trawled from the North Sea floor, many in temperature, followed relatively slowly by the of them from the vicinity of Brown Bank (van

0 ° I ,,,

4. ~cE SHEET ,~ ,(, SHETLAND IKING_ t ERGEN

NORTH SEA (5

54 ° Pit Lake

Rhine

Doggerland: Hypothetical Landscape with first extensive coast B~tling-Aller~d/Windermere Interstadial 0 200km ......

Fig. 2. Doggerland during the Bolling-Allerod/Windermere interstadial, showing development of northern coast and estuaries and presence of Viking-Bergen island. Downloaded from http://sp.lyellcollection.org/ at University of Edinburgh on March 10, 2013

396 BRYONY J. COLES

Kolfschoten & Laban 1995); those bones which, Colonization of western Norway was prob- themselves undated, can be attributed to the ably also by boat, given the lack of an ice-flee interstadial by reference to the faunas of corridor of land around the southwestern adjacent lands, include horse, mammoth and margin of the Scandinavian ice sheet at the red deer. The open vegetation and frequent postulated time of colonization (Anundsen expanses of fresh water coupled with warm 1996). The first human presence on the Norwe- temperatures, may have attracted many thou- gian coast is currently dated to the Allerod, sands of migrating birds; these too will have on the basis of indirect evidence in the form attracted human predators. of charcoal which is thought to derive from fires that had burned in a context where The people of Doggerland a natural origin would be unlikely (Bang- Andersen 1996). The Late Glacial hunter-gatherers who explored From these slight indications, it can be argued and then settled Doggerland came from inland that people had developed a way of life adapted Europe (Housley et al. 1997), and as such they to coastal and marine conditions before the will have been familiar with the general condi- opening of the Holocene. The abundance of tions that prevailed through the land. Even- marine mammals, sea-birds and fish known tually, however, the first exploratory groups from Scandinavian coastal sites indicates the reached, what was for them, an entirely new attractions of the sea; in cultural terms, the environment: the coast. The location for this challenge of developing new skills and equip- encounter will have been along the north coast ment in order to exploit these marine resources of Doggerland, both southern Britain and Jut- may have taken the place formerly held by land then being inland regions. land exploration, The people who looked out over the cold Ethnographic studies of coastal hunter-fisher- North Sea faced a double culture shock, not only gatherers indicate potential for the development salt water and an unfamiliar suite of plants and of complex maritime-orientated societies which, animals which might or might not be exploited when compared with contemporary inland using existing skills and equipment, but also an groups have relatively sedentary settlement environment which put a halt to the millennia- patterns and relatively high population densities long process of colonization. An engrained cul- (e.g. Rousselot eta[. 1988). It is unlikely that the tural tradition of exploring new lands became in first coastal peoples of northwestern Europe will large measure redundant when this particular have closely resembled more recent maritime northwestern frontier was reached. groups, but a relative complexity, density and The coastline of the interstadial and following settled character may have pertained from an cold phase may have been relatively stable, for early stage vis-a-vis inlanders, in so far as these although the British and Scandinavian ice-sheets aspects will have been related to the year-round were reduced, a rapid rise in relative sea-level was availability of natural resources along the coast only to come with the later wasting of the and particularly around estuaries, and to the extensive North American ice-sheets. In places technological investments related to the exploi- throughout Doggerland, there is evidence for a tation of fish and sea mammals. sea-level standstill, for example along the Norwe- Archaeological evidence for Late Glacial mar- gian Trench at some stage between 12 500 BP and itime hunter-fisher-gatherers is largely absent 10 800 BP (Johnson et al. 1993). Shetland and the from northwestern Europe, for the available Orkneys remained part of mainland Scotland, coastline was that of northern Doggerland, now and a large island existed between their position submerged by the North Sea. From the opening and Norway, referred to as Viking-Bergen or of the Holocene (Fig. 3), evidence is available, Frigg island. One small flint flake, probably from northeastern England and from southern worked by humans, has been found in a core Scandinavia, in those places where the shifting taken from Viking-Bergen (Long et al. 1986). This relationship between land and sea levels has left raises the possibility that humans reached the the coastal zone of the earlier Holocene acces- island before it was submerged, having therefore sible. It appears that from the early Mesolithic developed the boats and the skills required to onwards, a coastal and an inland way of life navigate the North Sea. It is unlikely that people can be differentiated, with the coastal peoples penetrated this far north while Viking-Bergen was exploiting their hinterland as well as the sea, and still attached to the Dogger mainland, in the light thereby establishing a broad band of occupied of Housley et al.'s (1997) calculations of the rate at territory along the coast. Inlanders may have which humans moved into the newly available exchanged goods with coastal people, and land as the ice-sheets wasted. visited the coastal zone, but their own settlement Downloaded from http://sp.lyellcollection.org/ at University of Edinburgh on March 10, 2013

DOGGERLAND: THE CULTURAL DYNAMICS OF A SHIFTING COASTLINE 397

SHETLA~

NORTH SEA

,_f-.

54 ° , \

Doggerland: Hypothetical Landscape with extensive coast and estuaries Earlier Holocene C~4AN NEL 0 20Okra

] , ,,~ ,., .#" ,, ,~,,; .... 0i o ...... l Fig. 3. Doggerland during the earlier Holocene, showing lengthening coast and major estuaries. pattern did not infringe on the coastal belt. visits to the interior of their overall territory Two strands of evidence support this hypothesis: (Andersen 1998). For Jutland the inland zone, analysis of stable carbon-isotope ratios in bones inhabited by people without access to the coast, from humans and their domestic dogs indicates extended at least as far west as the present either a predominantly marine diet or a pre- western coast, a reminder of Doggerland's former dominantly inland diet (e.g. Nordqvist 1995), presence beyond. and analysis of material culture shows the development of localized styles within a frame- Coastal changes work of contacts over long distances. By the Late Mesolithic in southern Scandinavia, three In the time period from the Late Glacial to zones can be distinguished, coastal, hinterland the Late Mesolithic, approximately 12 000 BP to and inland, with the hinterland evidence prob- 6000 cal BP (4000 cal BC) the coastline of Dogger- ably left by coastal people during seasonal land shifted in response to rising sea-levels Downloaded from http://sp.lyellcollection.org/ at University of Edinburgh on March 10, 2013

398 BRYONY J. COLES and much of the former landmass became sub- course, is an assumption, based on the reasons set merged by the North Sea. It has been argued out above. If it is accepted as plausible, it is then elsewhere (Coles 1998) that land loss may have relevant to consider what happened to these been slower than generally assumed, in part coastal people as sea-level rose. because Doggerland was not a flat and feature- It is inherently likely that, as the environment less plain. Figures 1-4 present one possible scen- which they knew shifted, the coastal people ario for land loss, and Shennan et al. (2000) moved with it. They will have had the advantage present another. A significant element in the pro- numerically over inlanders, and they had gen- cess of land loss, evident from Figs 1-4, was the erations of experience of coastal and estuarine increase in the length of coastline and ofestuarine exploitation. Moreover, their maritime habitat areas, two zones of ecological richness inhabited was expanding, and it will not have stressed by a relatively dense and settled human popula- them to absorb formerly inland groups into their tion adapted to a maritime economy. This, of midst. Throughout the millenia of sea-level rise,

,..,,,

,ine

Doggerland: Hypothetical North Sea and Islands c 5000 cal BC cHANN5~ 0 200kin

Fig. 4. Doggerland in the later stages of sea-level rise. Downloaded from http://sp.lyellcollection.org/ at University of Edinburgh on March 10, 2013

DOGGERLAND: THE CULTURAL DYNAMICS OF A SHIFTING COASTLINE 399 the people displaced were not those who saw (b) expansion of the idea of farming and their familiar beaches and living places lost to acquisition by indigenous peoples of domes- the encroaching waves, but the inlanders who tic animals and crops; yielded to the full package of advancing coast- (c) local experimentation and development of line, complete with flora, fauna and a well- plant and animal husbandry, with eventual adapted human population. acquisition of non-native species. Almost as soon as the Late Glacial people on Doggerland's coast first developed a maritime- Varying combinations of these pathways in orientated society, the long-term process of different regions and at different times are moving with the moving coast will have begun. likely. With farming comes evidence for more It must have become an entrenched cultural substantial and more permanent settlement, tradition, as was the concomitant absorbtion of for pottery and new lithic technologies, and for inlanders. To call people 'settled' is therefore in changing cultural relations with the land. This one sense misleading, but 'settled' is an apt last aspect is now seen as the crux of the description of their tenure of the coastal zone Neolithic by a number of prehistorians (see and, as suggested by Andersen (1998) for the Edmonds & Richards 1998 for examples of Late Mesolithic, they probably had tenure of a current concerns). hinterland belt as well. Neolithic settlement had spread rapidly across At about 6000-5800calBc, the watershed Europe from its roots in the Near East, for between the North Sea drainage systems and example along the Danube and its tributaries the Channel drainage system may have been and down the Rhine, with the earliest farming breached, the event being dated by reference to settlements of the Aegean and Balkans dating to conditions in the Southern Bight which changed about 6500 cal BC and the earliest of the north- from brackish to marine at about this time west to about 5400calBc (Whittle 1996). The (Eisma et al. 1981). The Storegga submarine northwestward advance of the Neolithic seems landslide off Norway and the consequent to have met with an invisible barrier some 100- tsunami are possible candidates for natural 150 km inland of today's coastline: from Ireland, catastrophic events that hastened the breach. Britain, the Netherlands, North Germany and However, sea-level rise continued for a further southern Scandinavia the archaeological evi- two millennia, until about 4000 cal BC (Shennan dence at present available indicates the passing et al. 2000; Long et al. 2000), and studies of tidal of 1200-1400 years before these regions too systems off the coast of the Netherlands indicate acquired a neolithic way of life at or shortly some impediment to open contact between before 4000 cal BC. Channel and North Sea until about 3800 calBC Prehistorians have suggested various reasons (B. van der Walk, pers. comm.). Perhaps, there- for this lengthy halt to the spread of farming, fore, the watershed zone remained as some form noting for example the farmers' preference for of barrier, a series of islands if not continuous loess and their slow adaptation to the cultiva- land, beyond 6000-5800calBc. (For a more tion of other soils, and the strength of foraging detailed discussion of these points, see Coles traditions in the lands beyond the loess. In some 1998, p. 66-69). parts of the farm-free North Sea belt, there is scant archaeological evidence for a late foraging Effects on people and human economy (Mesolithic) population in the 5th millennium cal Bc, but in southern Scandinavia in particular The cultural process postulated above, of a flourishing, culturally rich and innovative coastal populations moving with the coastal foraging population can be documented (e.g. zone, may therefore have endured until the Andersen 1998, Fischer 1995). The evidence NW European transition to farming, which comes mainly from what was the coastal zone of took place for what are now the lands border- the period, from those parts of it which are now ing the North Sea at about 4200-4000calBc. accessible to archaeological survey thanks to the Hypotheses abound as to how farming became interplay of land and sea level change in this established in Europe. The detail of conflic- particular region. In the Netherlands, the rele- ting interpretations need not concern us here, vant zone is largely blanketed by Holocene although it is relevant to note the main sedimentation, but current railroad construction possibilities: has led to targetted deep archaeological investigations which indicate a similar later Mesolithic (a) expansion of farmers together with live- potential (e.g. Hardingxveld site, L. P. Louwe stock and crops from southeastern Europe Kooijmans, pers. comm.). These flourishing via Danubian and Mediterranean routes; groups were, to take the perspective of this Downloaded from http://sp.lyellcollection.org/ at University of Edinburgh on March 10, 2013

400 BRYONY J. COLES paper, Doggerland's coastal peoples, those who lands must have been shrinking. Cultural adjust- had the advantage over inlanders. ments will have occured for both farmers Northwestern Europe in the earlier Holocene and foragers, perhaps even contributing to the can be considered as a land slowly overwhelmed LBK-R6ssen changes on the farming side and to by two advancing waves, farming from the development of cemeteries and other indica- the south and the shifting Doggerland coast tors of a strengthening hold on the land amongst from the north, Each process, farming as well as the coastal foragers. The first archaeological the coastal shift, displaced or absorbed inland signs of farming settlement in the buffer zone foragers, in the coastal zone for reasons outlined and beyond, in Jutland, south Sweden, Britain above, and along the frontier of farming for and Ireland, date to about 4200-4000 cal BC, at essentially similar reasons of disparity in popu- or more probably a little before the standstill lation densities and the expansive cultural in sea-level rise dated by Shennan et al. (2000) traditions of the people carried by the wave. and Long (2000) to around 4000 cal BC. Further Each people, coastal foragers and inland work on the chronology of both processes might farmers, will have had a forward zone of contact elucidate whether or not there is a link, whether with inland foragers, and it must have been for example marine regression drew with it the towards the late 6th millennium calBC that the coastal people, leaving a belt of relatively un- unfortunate inland foragers began to find occupied land soon exploited by farmers. The themselves pinched from both sides, between process of marine regression itself may have the two waves of advance. In this light, although sufficiently upset coastal patterns of subsistence other factors no doubt contributed, the major and cultural traditions for the system to col- force that halted the previously rapid spread of lapse. Alternatively, 1200 years of contact may farming can be seen to be Doggerland's coastal have familiarized farmers with the sea, sufficient population, supported and conditioned by long- for their final advance into Britain and Ireland term sea-level rise. It has been argued above and into southern Scandinavia. that the coastal people were more settled than Too often, the long process of sea-level rise has inlanders, their numbers were greater, their been ignored by prehistorians, and the existence economy more diversified, all of which would of Doggerland and its people has been neglected. have put them more on a par with the farmers Both were of major importance, exerting an than their inland neighbours were. And, prob- influence on regions beyond the area of the ably the most significant factor, their cultural present North Sea, just as the shifting Atlantic outlook was as expansive and tuned to absorb- coastline and its inhabitants no doubt influenced ing others, rather than themselves being over- in some way events further to the west. This whelmed, as that of the farmers. Moreover, the paper has been written to draw attention to some coastal people had no choice but to move, so of the cultural repercussions of the shifting long as the coastal zone continued to shift. North Sea coastline, partly in the hope that It was this perhaps which gave the edge to the developments in survey techniques will soon coastal foragers, and halted the advance of enable a search for evidence to support or farming 150km or so inland of the coast, refute its arguments. presumably on the outer edge of the coastal zone of exploitation. In the centuries following their arrival at the invisible barrier at about 5400calBC, farming References groups can be seen to consolidate and infill ANDERSEN, S. H. 1998. Ringkloster. Ertebolle trap- their zones of settlement. During this phase, pers and wild boar hunters in eastern Jutland. the Linear Bandkeramik or LBK, little cultural A survey. Journal of Danish Archaeology, 12, change is indicated in the archaeological record. 13-59. The R6ssen period, a time of greater cultural ANUNDSEN, K. 1996. The physical conditions for and economic adaptation, followed from about earliest settlement during the last deglaciation in 4800calBc. During the R6ssen phase in parti- Norway. In: LARSSON, L. (ed.) The Earliest cular, contact is evident with the foragers to Settlement of Scandinavia. Acta Archaeologica the north, traced archaeologically in items such Lundensia. Series in 8°, No. 24, 207-217. BANG-ANDERSEN,S. 1996. The Colonisation of South- as pottery and stone axe or adze blades of west Norway. An Ecological Approach. In: R6ssen origin, found on sites beyond the farm- LARSSON, L. (ed.) The Earliest Settlement of ing frontier (Whittle 1996; Louwe Kooijmans Scandinavia. Acta Archaeologica Lundensia, 1998; van Gijn 1998). Series in 8°, No. 24, 219 234. For so long as sea-level continued to rise, the COLES, B. J. 1998. Doggerland: a speculative survey. buffer zone between the coast and the cultivated Proceedings of the Prehistoric Society, 64, 45-81. Downloaded from http://sp.lyellcollection.org/ at University of Edinburgh on March 10, 2013

DOGGERLAND: THE CULTURAL DYNAMICS OF A SHIFTING COASTLINE 401

--1999. Doggerland's loss and the Neolithic. In: LONG, A, SCAIFE, R. & EDWARDS, R. 2000. Strati- COLES, B. J., COLES, J. M. & SCHOU JORGENSEN, graphic architecture, relative sea-level, and models M. (eds) Bog Bodies, Sacred Sites and Wetland of estuary development in southern England: new Archaeology. WARP Occasional Paper 12, 51-57. data from Southampton Water. This volume. EDMONDS, M. & RICHARDS, C. (eds) 1998. Under- LONG, D., W1CKHAM-JONES, C. R. & RUCNLEY, N. A. standing the Neolithic. Cruithne Press, Edinburgh. 1986. A flint artefact from the northern North EISMA, D., MOON, W. G. & LABAN, C. 1981 An early Sea. In: ROE, D. A. (ed.) Studies in the Upper Holocene tidal flat in the Southern Bight. In: NIO, Palaeolithic of Britain and Northwest Europe. S.-D., SCHUTTENHELM, R. T. E. & VAN WEERING, British Archaeological Report $296, 55-62. TJ. C. E. (eds) Holocene Marine Sedimentation in LOUWE KOOIJMANS, L. P. 1998 Understanding the the North Sea Basin. International Association of Mesolithic/Neolithic frontier in the Lower Rhine Sedimentologists Special Publication 5, 211 219. Basin, 5300-4300calBC. In: EDMONDS, M. & FIRTH, A. 2000. Development-led archaeology in RICHARDS, C. (eds) Understanding the Neolithic coastal environments: investigations at Queens- of Northwestern Europe. Cruithne Press, Edin- borough, Motney Hill and Gravesend in Kent, burgh, 407-427. UK. This volume. NORDQVIST, B. 1995. The Mesolithic settlements of FISCHER, A. (ed.) 1995 Man and Sea in the Mesolithic. the west coast of Sweden - with special empha- Oxbow Monograph 53, Oxford. sis on chronology and topography of coastal VAN GIJN, A. 1998 Craft activities in the Dutch settlements. In: FISCHER, A. (ed.) Man and Neolithic: a lithic viewpoint, ln: EDMONDS, M. & Sea in the Mesolithic. Oxbow Monograph 53, R~CHARDS, C. (eds) Understanding the Neolithic of 185-196. Northwestern Europe. Cruithne Press, Edinburgh, REID, C. 1913. Submerged Forests. Cambridge Uni- 328-350. versity Press. HOUSLEY, R. A., GAMBLE, C. S., STREET, M. & PETTITT, ROUSSELOT, J.-L., FITZHUGH, W. W. & CROWELL, A. P. 1997. Radiocarbon evidence for the Lateglacial 1988. Maritime Economies of the North Pacific Human Recotonisation of Northern Europe. Rim. In: FITZHUGH, W. W. 8~; CROWELL, A. (eds) Proceedings of the Prehistoric Society, 63, 25-54. Crossroads of Continents. Smithsonian Institution JOHNSON, H., RICHARDS, P. C., LONG, D. & GRAHAM, Press, Washington DC, 151-172. C. C. 1993. The Geology of the Northern North SHENNAN, I., LAMBECK, K., FLATHER, R. ETAL. 2000. Sea. Her Majesty's Stationery Office, London. Modelling western North Sea palaeogeographies VAN KOLFSCHOTEN, TH. & LABAN, G. 1995. Pleisto- and tidal changes during Holocene. In: SHENNAN, cene terrestrial mammal faunas from the North I. 8~; ANDREWS, J. E. (eds) Holocene Land - Ocean Sea. Mededelingen Rijks geologische Dienst, 52, Interaction and Environmental Change around the 135-152. North Sea. Geological Society, London, Special KOLSTRUP, E. 1991. Palaeoenvironmental develop- publications 166, 299-319. ments during the Late Glacial of the Weichselian. TIPPING, R. 1991. Climate change in Scotland during In: BARTON, N., ROBERTS, A. J. & ROE, D. (eds) the Devensian Late Glacial: the palynological The Late Glacial in North-west Europe. Council record. In: BARTON, N., ROBERTS, A. J. & ROE, O. for British Archaeology Research Report 77, 1-6. The Late Glacial in North-west Europe. Council LAMBECK, K. 1993. Glacial rebound of the British for British Archaeology Research Report 77, Isles- 1. Preliminary model results. Geophysical 7-21. Journal International, 115, 941-959. WHITTLE, A. 1996. Europe in the Neolithic." the creation --, SMITHER, C. & JOHNSTON, P. 1998. Sea-level of new worlds. Cambridge University Press, change, glacial rebound and mantle viscosity for Cambridge. northern Europe. Geophysical Journal Interna- tional, 134, 102-144.