The distriďution of early PalaeolithiĐ sites in Britain Nick Ashton 1 , Simon G. Lewis 2 & Claire R.E. Harris 1 1 Department of Britain, Europe & Prehistory, British Museum, Franks House, 56 Orsman Road, London, N1 5QJ, UK. nashton@thebritishmuseum.ac.uk 2 School of Geography, Queen Mary University of London, Mile End Road, London, E1 4NS, UK. Abstract This paper provides an initial examination of the distribution of handaxe sites compared to Levallois sites in Britain based on the recently digitised Palaeolithic site records at the British Museum. The pattern suggests a wide distribution of handaxe sites across southern and eastern England in contrast to a more limited distribution of Levallois sites in London and Kent. Collector bias and differential access to geological deposits are examined as factors in determining the pattern, but the conclusion drawn is that there is still an underlying pattern that relates to Lower and Early Middle Palaeolithic land-use. It is suggested that this may reflect the changing palaeogeography of Britain from MIS 13 through to MIS 7. Introduction The early human occupation of northern Europe has been primarily controlled by the cyclical changes in climate with repeated human colonisations punctuated by either retreats or possibly local extinctions as climate cooled (Stringer 2005). It would seem that through time the development of technology and to some extent physical adaptation led to increasingly successful and more sustained settlement of these areas; the earliest forays from c. 800 ka into northern Europe such as at Happisburgh (Norfolk, UK; Parfitt et al. 2010) and Pakefield (Suffolk, UK; Parfitt et al. 2005) seem to have been sporadic and short-lived, whereas by the last glaciation, humans had adapted to all but the cold extremes of Marine Isotope Stage (MIS) 4 and the Last Glacial Maximum (Roebroeks et al. 2011). Recent research has shown that the record for Britain is somewhat different to that of mainland Europe with a suggested decline in population in Britain during each warm stage from MIS 11 and a probable absence during MIS 6-4 (Ashton & Lewis 2002; Ashton & Hosfield 2010; Ashton et al. 2011; Lewis et al. 2011; Davis 2013). This has been primarily based on the numbers of Lower Palaeolithic and Middle Palaeolithic sites, but also on regional studies in the Middle Thames and Solent river basins. The regional studies used the varying densities of Lower Palaeolithic handaxes and Early Middle Palaeolithic Levallois artefacts in the flights of river terrace gravels as a proxy for relative human population. The suggested decline in population is argued to have been due to the changing palaeogeography with the creation of the Strait of Dover at the end of MIS 12 (Smith 1985; Gupta et al. 2007; Toucanne et al. 2009) and the subsidence of the floor of the North Sea Basin (Ashton et al. 2011). The effect of the breach was to make Britain an island during high sea-levels, while the down-warping of the North Sea Basin meant that larger drops in sea-level were required in each successive cold phase to reconnect Britain to the continent. One of the criticisms levelled at the Middle Thames and Solent regional studies was the use of handaxes and Levallois pieces as equivalent artefact types (White et al. 2006; Pettitt & White 2012). It was argued that they probably had different functions and would therefore have had different life histories. This would have resulted in significant variation in the numbers produced and where they might have been discarded in the landscape. A second criticism was that handaxes were more likely to have been picked up by gravel quarry workers and were more prized by many collectors, therefore producing an inevitable bias in the numbers of handaxes now held by museums (McNabb 2007). The final main criticism of this research was the dominance of soŵe sites ;oƌ supeƌ-sitesͿ iŶ the ƌeĐoƌd pƌoǀidiŶg a skew to the data (White et al. 2006; McNabb 2007). It was argued that once locations became known for producing handaxes, positive feedback would encourage disproportionate collecting at those sites. Some of these criticisms have been dealt with through examination of just the handaxe records for the Solent and Middle Thames, by studying the collecting habits of some of the collectors and by the removal of supeƌ-sites fƌoŵ the aŶalysis ;AshtoŶ & Hosfield ϮϬϭϬ; Ashton et al. 2011; Davis 2013). Taking these potential problems into account, the pattern of declining artefact numbers has stood up to more detailed scrutiny and overall it can still be suggested that there is an underlying pattern of population decline through the late Middle Pleistocene in both the Solent and Middle Thames areas. This paper takes a new approach to past shifts in population by examining the changing distribution in artefact types in Britain through the Middle Pleistocene. It uses sites rather than artefact numbers to avoid the poteŶtial pƌoďleŵs of supeƌ-sites doŵiŶatiŶg the record. In particular the study focuses on the differences between the distribution of handaxe and Levallois sites. Work over the last 20 years suggests that most handaxe assemblages date to between c. 500 and 300 ka and are attributed to the Lower Palaeolithic, whereas most Levallois assemblages date from c. 300 to 200 ka and are now termed Early Middle Palaeolithic (Ashton & Lewis 2002; Ashton et al. 2003; White et al. 2006; Scott 2011;