The majority of the ‘new’ findspots identified (n=49, 67%) come from the county of Devon. Many of these are located in east Devon, in districts already well known as Palaeolithic “hotspots” such as Thorncombe in the Axe valley region. Such records however remain of value despite the well-documented richness of the area, as they serve not only to confirm the areas importance, but also add a further level of understanding of spatial patterns in hominin occupation histories. Of potentially greater importance in Devon however are the new findspots identified in areas where little evidence has been previously documented, such as the discoveries around the River Otter at Gittisham, Otterton, and Sidmouth.

A significant number of the novel findspots are also located in areas of relatively low archaeological occurrence, such as locales to the far west of the region in southern Cornwall (n=5, 7%). Whilst these findspots follow the known distribution pattern for that county (i.e. they are distributed across the south-western margins of Cornwall) they remain of importance as they increase our knowledge of hominin landscape use within the region as a whole. Similarly in Somerset the 15 new findspots (21%) repeat the distribution pattern of those already recorded, with a roughly even distribution across the central, southern, northern, and western county, although again with no finds in the east.

Although all of the ‘new’ findspots are believed to have originated from an ‘open landscape’ context (there is no indication that any of the findspots are cave deposits), information regarding the specific geological and/or depositional context was scarce. Location evidence indicated that 11 (15%) of the findspots were associated with pit or quarry sites (e.g. Brickfields, Devonport, Plymouth; Gammon’s Hill Quarry, Kilmington; gravel pit and/or quarry at Thorncombe; Manor Pit, Seaton; north side of present quarry, Thorncombe; Westford Farm gravel pits, Thorncombe), while ‘gravel’, ‘river gravel’ or ‘floodplain gravel’ were indicated as the probable geological context for an additional four of the findspots. Geological information was generally rare (n=6, 8%), with ‘loam and clay’ and (rather ambiguously) ‘chert with clay content’ recorded as the contexts for two other findspots. There were also records relating to the circumstances of discovery, although unfortunately those suggesting a river gravel context (n=5, 7%) all related to findspots where a gravel pit location was already known (see above).

Data relating to the accuracy of the findspot location was rare (n=7, 10%), reflecting the nature of the records, although in all cases where it was recorded the findspots were classified as ‘accurate’ (i.e. to within 100m).

Data relating to the number of artefacts from each findspot was of variable quality, since in a significant number of cases (n=28, 39%) references were made to ‘artefacts’, ‘implements’ etc without further details being supplied. In the remaining instances however, single artefact finds (n=35, 49%) were dominant, with smaller numbers (n=9, 12%) of single figure artefact finds.

The artefacts identified by type were dominated by handaxes, which were definitely present in 25 (35%) of the findspots. Of these, 23 findspots (32%) were single handaxe finds, with one instance of a ‘handaxe with flakes’, and one instance of two handaxes found on a single findspot. This is perhaps unsurprising given the nature of the findspot discoveries described above. The other artefact categories were present in far fewer of the findspots.

In general the evidence from the resource assessment indicated a number of small artefact discoveries documented in the HER records, but which had been absent from the major syntheses. These tended to be single/single figure artefact finds, both from aggregates pit and quarries and also from non-industrial excavations and surface finds.

The second aspect of data collection within the Lower and Middle Palaeolithic resource assessment concerned the specific artefact sampling programme. Artefact data was recorded from collections housed in all of the major regional museums in the south-west (including the Royal Cornwall museum, Royal Albert Memorial museum (Exeter), Devizes museum, Dorset County museum (Dorchester), Somerset County museum (Taunton), Plymouth museum, Torquay museum and Bristol City museum), as well as those from the Cambridge Archaeology & Anthropology museum stores.

364 artefacts were recorded. However, in many cases it was difficult (and in some cases impossible: see comments below) to link individual artefacts in the collections with their specific findspot records (whether extant or ‘new’ findspots), and therefore it is not currently possible to assess what proportion of the identified findspots’ artefacts have been recorded (this issue will be addressed during the phase three project synthesis). Nonetheless, where provenancing information was available it was clear that the provenance locations of the artefacts broadly followed the distribution patterns outlined in the findspot data, with the great majority (n=202; 55%) originating from the Axe Valley region in Devon/Dorset/Somerset (and therefore suggesting that these artefacts are associated with the documented findspots), and the bulk of these coming from the gravel pits at Broom (n=166, 46%). Similar distribution patterns as to those outlined above are also found in each of the other counties studied (i.e. artefacts in Cornwall were predominantly recorded from the southern margins of the county).

In all of the museum collections visited handaxes made up the overwhelming majority of the known Palaeolithic artefacts (n=294, 81%). This is to be expected given the high visibility of these tools (reflecting their size and distinctive morphology), and their status as a diagnostic artefact (enabling them to be assigned with relative ease to either the Lower or Middle Palaeolithic periods). This is a key point particularly with regard to fluvial contexts and deposits, since artefacts recovered from these secondary contexts have been re-worked and it is therefore extremely difficult to assign un-diagnostic pieces (e.g. débitage flakes) to a particular period. Moreover, smaller/lighter artefacts (e.g. flake tools) are also more vulnerable to destruction/damage beyond the point of recognition during transportation and re-working in fluvial environments, while the larger sized artefacts such as handaxes were more prone to be spotted and recovered by gravel workers/collectors in the context of aggregates quarries and pits. These factors almost certainly explain the prevalence of handaxes in Palaeolithic museum collections, rather than any unusual hominin behaviour (e.g. the introduction of handaxes from outside the region and their sole discard, with all other lithic material culture removed from the region by the hominins).

The overwhelming majority of the raw material used is that of chert (n=296, 81%), while the rest of the artefacts are made on flint (n=64, 18%) with the exception of one handaxe from Mill Hayes, Stockland in Devon which was made on igneous rock (unfortunately weathering and abrasion of the artefacts prohibited identification of the igneous rock type during this assessment).

It was possible to assess the general physical condition of the artefacts, following the Wymer classificatory scheme based on flake scar ridge abrasion. None of the artefacts were classified as mint, 24% (n=88) were classified as sharp, 49% (n=177) as slightly rolled, 23% (n=84) as rolled, and 4% (n=15) as very rolled. These preliminary results suggest that the majority of the sampled artefacts had been subject to fluvial transportation and were extremely likely to have been recovered from fluvial gravel (probable secondary) contexts.

Go back to Module 2 Introduction | Return to top